Documente Academic
Documente Profesional
Documente Cultură
Faculty of Engineering, The University of Auckland
Commentary to
Assessment and
Improvement of
Unreinforced
Masonry Buildings
for Earthquake
Resistance
Supplement to “Assessment and Improvement of the Structural Performance of
Buildings in Earthquakes”
Draft
12/2011
Contents
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand .......... 1‐1
C1.1 Introduction ....................................................................................................................... 1‐1
C1.3 URM Performance in Past Earthquakes ............................................................................ 1‐3
C1.4 New Zealand URM Building Stock ..................................................................................... 1‐8
C1.5 New Zealand Building Codes Pertaining to URM Construction ...................................... 1‐13
C1.6 Acknowledgements ......................................................................................................... 1‐18
C1.7 References ....................................................................................................................... 1‐18
Section C2 ‐ Material Properties of Masonry Walls ...................................................................... 2‐1
C2.1 Notation ............................................................................................................................ 2‐1
C2.3 In‐situ and Extracted Sample Testing ................................................................................ 2‐1
C2.4 Brick and Mortar Field Assessment Procedure ................................................................. 2‐5
C2.5 Additional Information ...................................................................................................... 2‐6
C2.7 Masonry Stress Block Parameters ................................................................................... 2‐22
C2.8 Worked Examples ............................................................................................................ 2‐23
C2.9 References ....................................................................................................................... 2‐27
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms ........................................ 3‐1
C3.3 Timber properties .............................................................................................................. 3‐1
C3.4 Classification ...................................................................................................................... 3‐3
C3.5 Configuration and Condition Assessment ......................................................................... 3‐3
C3.6 Characteristic Values ......................................................................................................... 3‐5
C3.7 Diaphragm Stiffness ........................................................................................................ 3‐14
C3.8 Worked Example 1 .......................................................................................................... 3‐15
C3.9 Worked Example 2 .......................................................................................................... 3‐17
C3.10 Worked Example 3 ........................................................................................................ 3‐19
C3.11 References ..................................................................................................................... 3‐21
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry Walls ............................ 4‐1
C4.1 Notation ............................................................................................................................ 4‐1
C4.2 Scope ................................................................................................................................. 4‐1
C4.3 Analysis Procedures ........................................................................................................... 4‐1
C4.4 Design Example ............................................................................................................... 4‐16
i
C4.5 References ...................................................................................................................... 4‐24
Section C5 ‐ Determination of Diaphragm Design Actions ........................................................... 5‐1
C5.2 Scope ................................................................................................................................. 5‐1
C5.3 Load Distribution .............................................................................................................. 5‐1
C5.4 Design Actions ................................................................................................................... 5‐1
C5.5 Worked Example ............................................................................................................... 5‐1
C5.6 References ........................................................................................................................ 5‐3
Section C6 ‐ Design Actions on URM Walls and Parapets loaded out‐of‐plane ........................... 6‐1
C6.1 Notations .......................................................................................................................... 6‐1
C6.2 Scope ................................................................................................................................. 6‐1
C6.3 Design Lateral Force on Walls loaded out‐of‐plane .......................................................... 6‐3
C6.4 Design Example ................................................................................................................. 6‐3
C6.5 References ........................................................................................................................ 6‐7
Section C7 ‐ Design Actions on Wall Diaphragm Anchorages ....................................................... 7‐1
C7.1 Notation ............................................................................................................................ 7‐1
C7.2 Scope ................................................................................................................................. 7‐1
C7.3 Design Shear for In‐plane Loaded Connections ................................................................ 7‐1
C7.4 Design Tension for Out‐of‐plane Loaded Connections ..................................................... 7‐1
C7.5 Design Example ................................................................................................................. 7‐2
C7.6 References ........................................................................................................................ 7‐4
Section C8 ‐ Response of walls loaded in‐plane ........................................................................... 8‐1
C8.1 Notation ............................................................................................................................ 8‐1
C8.2 Scope ................................................................................................................................. 8‐1
C8.3 Properties of walls loaded in‐plane .................................................................................. 8‐3
C8.4 Nominal Shear Capacity of walls loaded in‐plane ............................................................ 8‐4
C8.5 Deformation Limits ........................................................................................................... 8‐9
C8.6 Worked Examples ........................................................................................................... 8‐10
References .............................................................................................................................. 8‐17
Section C9 ‐ Strength and Deformation Assessment of Flexible Timber Floor Diaphragms ........ 9‐1
C9.3 Lateral Deformation .......................................................................................................... 9‐1
C9.4 Strength ............................................................................................................................ 9‐1
ii
C9.5 Worked Example 1 ............................................................................................................ 9‐2
C9.6 Worked Example 2 ............................................................................................................ 9‐3
C9.7 Worked Example 3 ............................................................................................................ 9‐4
C9.8 References ......................................................................................................................... 9‐5
Section C10 ‐ Strength and Deformation Assessment of Timber Roof Diaphragms ................... 10‐1
C10.2 Scope ............................................................................................................................. 10‐1
C10.3 Material Properties ....................................................................................................... 10‐2
C10.4 Lateral Deformation ...................................................................................................... 10‐2
C10.5 Strength ......................................................................................................................... 10‐2
C10.6 References ..................................................................................................................... 10‐2
Section C11 ‐ Out‐of‐plane Wall and Parapet Response ............................................................. 11‐1
C11.2 Scope ............................................................................................................................. 11‐1
C11.3 Wall Thickness, Height, and Slenderness Ratio ............................................................. 11‐2
C11.4 Load‐bearing walls......................................................................................................... 11‐3
C11.5 Assessment .................................................................................................................... 11‐3
C11.6 Worked example ........................................................................................................... 11‐7
C11.7 References ................................................................................................................... 11‐13
Section C12 ‐ Wall‐diaphragm Anchorages ................................................................................. 12‐1
C12.1 Notation ........................................................................................................................ 12‐1
C12.2 Scope ............................................................................................................................. 12‐1
C12.3 Diaphragm Joist Seating ................................................................................................ 12‐3
C12.4 Masonry Anchorage Failures ......................................................................................... 12‐3
C12.5 Steel Anchorage Failures ............................................................................................... 12‐4
C12.6 Timber Anchorage Failures ........................................................................................... 12‐5
C12.7 Examples........................................................................................................................ 12‐7
C12.8 References ..................................................................................................................... 12‐9
Section C13 ‐ Heritage Characteristics of URM Buildings and the Impacts of Seismic
Improvements ............................................................................................................................. 13‐1
C13.1 Introduction ................................................................................................................... 13‐1
C13.2 Heritage and Conservation ............................................................................................ 13‐2
C13.3 Architectural Character ................................................................................................. 13‐7
C13.4 Initial Considerations for Seismic Improvement ......................................................... 13‐15
iii
C13.5 Stabilisation for Seismic Improvement ....................................................................... 13‐17
C13.6 Enhancing Existing Walls ............................................................................................. 13‐21
C13.7 New Building Strengthening Systems ......................................................................... 13‐25
C13.8 Reduction of Forces .................................................................................................... 13‐31
C13.9 Approaches to Seismic Improvement ......................................................................... 13‐32
C13.10 Conclusions ............................................................................................................... 13‐35
C13.11 References ................................................................................................................ 13‐36
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms ................................................................................................................................. 14‐1
C14.2 Scope ............................................................................................................................. 14‐1
C14.3 Retrofit Selection .......................................................................................................... 14‐1
C14.4 Revised Diaphragm Assessment ................................................................................... 14‐5
C14.5 Diaphragm Chord Design .............................................................................................. 14‐5
C14.6 Subdiaphragm Design ................................................................................................... 14‐8
C14.7 Diaphragm Penetrations ............................................................................................... 14‐9
C14.8 Worked Example ......................................................................................................... 14‐10
C14.9 References .................................................................................................................. 14‐10
Section C15 ‐ Methods for Improving Wall Diaphragm Connections ......................................... 15‐1
Section C16 ‐ Unbonded Post‐tensioning ................................................................................... 16‐1
C16.1 Notation ........................................................................................................................ 16‐1
C16.2 Scope ............................................................................................................................. 16‐2
C16.3 Post‐tensioning Tendon Types, Stresses and Spacing .................................................. 16‐5
C16.4 In‐plane Wall Behaviour................................................................................................ 16‐6
C16.5 Out‐of‐plane Wall Behaviour ...................................................................................... 16‐10
C16.6 Design Example ........................................................................................................... 16‐15
C16.7 References .................................................................................................................. 16‐19
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems ............................................. 17‐1
C17.2 Notation ........................................................................................................................ 17‐1
C17.2 Scope ............................................................................................................................. 17‐2
C17.3 General Design Information .......................................................................................... 17‐2
C17.4 Shear Strength Provided by Fibre Reinforced Polymer ................................................ 17‐2
iv
Relatively little data is available for shear strength enhancement using vertical FRP systems,
and hence a constant effective strain value is proposed. ..................................................... 17‐11
C17.5 Design Examples .......................................................................................................... 17‐11
C17.6 References ................................................................................................................... 17‐16
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems ....................... 18‐1
C18.1 Notation ........................................................................................................................ 18‐1
C18.2 Scope ............................................................................................................................. 18‐2
C18.3 Retrofit Parameters ....................................................................................................... 18‐3
C18.4 In‐plane Design .............................................................................................................. 18‐3
C18.5 Out‐of‐plane Design ...................................................................................................... 18‐3
C18.6 Other Considerations .................................................................................................... 18‐7
C18.7 Design examples ............................................................................................................ 18‐9
C18.8 References ................................................................................................................... 18‐13
Section C19 ‐ Fibre Reinforced Shotcrete (FRS) .......................................................................... 19‐1
C19.1 Notation ........................................................................................................................ 19‐1
C19.2 Scope ............................................................................................................................. 19‐2
C19.3 In‐plane General Design Information ............................................................................ 19‐3
C19.4 Out‐of‐plane General Design Information .................................................................... 19‐6
C19.5 References ................................................................................................................... 19‐13
Section C20 ‐ Near Surface Mounted Steel Reinforcement ........................................................ 20‐1
Section C21 ‐ Textile Reinforced Mortars .................................................................................... 21‐1
Section C22 ‐ Restraining Parapets and Chimneys ...................................................................... 22‐1
Section C23 ‐ Pounding ................................................................................................................ 23‐1
v
List of Figures
Figure C1‐1: Raupo whare (house) at Wairau Pa, ca. 1880 (Alexander Turnbull Library). ........... 1‐1
Figure C1‐2: Raupo and timber whares of a Maori village at Maketu, Kawhia, 1895 (Alexander
Turnbull Library). ........................................................................................................................... 1‐1
Figure C1‐3: New Zealand total population data, 1858‐2007. ..................................................... 1‐2
Figure C1‐4: Shops on Queen Street, Auckland, 1859 (Alexander Turnbull Library). ................... 1‐3
Figure C1‐5: Queen Street and Queen Street Wharf, Auckland, in 1882 (Alexander Turnbull
Library). ......................................................................................................................................... 1‐3
Figure C1‐6: Group photograph of the construction workers that built the Stratford Public
Hospital during 1906‐1907 (Alexander Turnbull Library). ............................................................ 1‐4
Figure C1‐7: Brick building under construction, ca 1920 (Alexander Turnbull Library). ............... 1‐4
Figure C1‐8: The 1833 Stone Store at Kerikeri was built by the Church Missionary Society. (A P
Godber Collection, Alexander Turnbull Library). .......................................................................... 1‐4
Figure C1‐9: Two Chinese miners in front of a stone cottage in central Otago, ca 1860 (Alexander
Turnbull Library). ........................................................................................................................... 1‐4
Figure C1‐10: Collapse of a new masonry auction market building, Queen Street, 1865
(Alexander Turnbull Library). ........................................................................................................ 1‐5
Figure C1‐11: Looking along a row of commercial buildings on Queen Street, Auckland, ca 1910
(Alexander Turnbull Library). ........................................................................................................ 1‐5
Figure C1‐12: Victorian Christchurch in 1885 (Coxhead 1885) ..................................................... 1‐5
Figure C1‐13: Christchurch’s first ‘skyscraper’, photo circa 1910 (Brittenden Collection 1910) .. 1‐5
Figure C1‐14: General store damaged by the 1929 Murchison earthquake (Alexander Turnbull
Library). ......................................................................................................................................... 1‐6
Figure C1‐15: Damaged business premises after the earthquake of 17 June 1929 (Alexander
Turnbull Library). ........................................................................................................................... 1‐6
Figure C1‐16: Hastings Street, Napier, circa 1914 (Alexander Turnbull Library). ......................... 1‐6
Figure C1‐17: View down Hastings Street, Napier after the 1931 earthquake (Alexander Turnbull
Library). ......................................................................................................................................... 1‐6
Figure C1‐18: Looking over Napier at the buildings ruined by the 1931 earthquake and the fires
(Alexander Turnbull Library). ........................................................................................................ 1‐7
Figure C1‐19: Ruins of the Napier Anglican Cathedral after the 1931 Napier earthquake
(Alexander Turnbull Library). ........................................................................................................ 1‐7
Figure C1‐20: Toppled parapet in the 2007 Gisborne earthquake. .............................................. 1‐8
Figure C1‐21 Out‐of‐plane failure of a gable wall in the 2007 Gisborne earthquake. .................. 1‐8
vi
Figure C1‐22: Out‐of‐plane wall failure at the corner of Worcester and Manchester streets in the
2010 Darfield earthquake. ............................................................................................................. 1‐8
Figure C1‐23: Out‐of‐plane wall failure at 118 Manchester Street in the 2010 Darfield
earthquake. ................................................................................................................................... 1‐8
Figure C1‐24: Photographic examples of New Zealand URM typologies ...................................... 1‐9
Figure C1‐25: Estimated %NBS of URM buildings in Provinces throughout New Zealand ......... 1‐12
Figure C2‐1: Steps to determine brick properties ......................................................................... 2‐2
Figure C2‐2: Determination of mortar compression strength ...................................................... 2‐3
Figure C2‐3: Preparation for prism compression test ................................................................... 2‐3
Figure C2‐4: Masonry flexural bond test ....................................................................................... 2‐4
Figure C2‐5: In‐situ and laboratory bed‐joint shear test ............................................................... 2‐4
Figure C2‐6: Difference between exposed and unexposed brick surface ..................................... 2‐6
Figure C2‐7: Colour spectrum intensity vs. compressive strength plot ........................................ 2‐8
Figure C2‐8: Brick compressive strength vs. visible colour plot .................................................... 2‐8
Figure C2‐9: Plot of brick compressive strength vs. scratch index .............................................. 2‐10
Figure C2‐10: Plot of Mortar compressive strength vs. scratch index ........................................ 2‐11
Figure C2‐11 Plot of brick Modulus of Rupture vs. brick compressive strength ......................... 2‐12
Figure C2‐12: Plot of mortar compressive strength vs. h/t ratio ................................................ 2‐13
Figure C2‐13: 3‐D plot relating the brick, mortar and masonry compressive strengths ............. 2‐15
Figure C2‐14: Plot of masonry Modulus of Elasticity vs. compressive strength plot .................. 2‐15
Figure C2‐15: Unreinforced masonry module for density approximation .................................. 2‐16
Figure C2‐16: Plot of masonry flexural bond strength vs. masonry compressive strength ........ 2‐17
Figure C2‐17: Different types of bond failure ............................................................................. 2‐18
Figure C2‐18: Plot of masonry flexural bond strength vs. mortar compressive strength ........... 2‐18
Figure C2‐19: Plot of shear stress vs. axial compression for laboratory manufactured samples2‐20
Figure C2‐20: Plot of shear stress vs. axial compression for field samples ................................. 2‐20
Figure C2‐21: Plot of cohesion vs. masonry compressive strength ............................................. 2‐21
Figure C2‐22: Plot of cohesion vs. mortar compressive strength ............................................... 2‐21
Figure C2‐23: Campbell Free Kindergarten ................................................................................. 2‐23
Figure C2‐24: Two‐storey Irish Pub ............................................................................................. 2‐25
Figure C2‐25: Two‐storey commercial building ........................................................................... 2‐26
Figure C3‐1: Test piece for embedded strength tests ................................................................... 3‐2
vii
Figure C3‐2: Test setup for embedded strength tests .................................................................. 3‐2
Figure C3‐3: Pure block test setup to determine compression strength of wood ....................... 3‐3
Figure C3‐4: Diaphragm configuration details .............................................................................. 3‐4
Figure C3‐5: Full‐scale diaphragm testing results ......................................................................... 3‐9
Figure C3‐6: Idealised bilinear response curve ........................................................................... 3‐10
Figure C3‐7: Bilinear curves from full‐scale diaphragm testing .................................................. 3‐13
Figure C4‐1: Modal testing and system identification procedure ................................................ 4‐7
Figure C4‐2: Plots of Force‐displacement response of URM assemblages ................................. 4‐11
Figure C4‐3: Seismic hazard zonation for the North Island of New Zealand for ground motion
record selection .......................................................................................................................... 4‐14
Figure C4‐4: Layout of example building .................................................................................... 4‐17
Figure C6‐1: Flowchart for Out‐of‐plane Seismic Design of URM buildings with Flexible
Diaphragms. .................................................................................................................................. 6‐2
Figure C6‐2: Definition of heights ................................................................................................. 6‐3
Figure C6‐3: Building cross‐section ............................................................................................... 6‐4
Figure C8‐1: Effective pier model (Yi et al. 2008) ......................................................................... 8‐2
Figure C8‐2: Shear stress distribution in a rectangular wall ......................................................... 8‐3
Figure C8‐3: Tributary areas for determining normal force on each wall (Russell 2010) ............. 8‐4
Figure C8‐4: Diagonal Tension failure mode ................................................................................. 8‐5
Figure C8‐5: Rocking/toe crushing failure mode .......................................................................... 8‐8
Figure C8‐6: Bed‐joint sliding failure mode .................................................................................. 8‐9
Figure C8‐7: Equivalent bilinear approximation (Magenes and Calvi 1997) .............................. 8‐10
Figure C8‐8: Plan layout of building ............................................................................................ 8‐10
Figure C10‐1: Examples of roof configurations ........................................................................... 10‐1
Figure C11‐1: Procedure for seismic assessment of out‐of‐plane loaded URM walls ................ 11‐5
Figure C11‐2: Zonation map for the North Island ....................................................................... 11‐6
Figure C11‐3: Zonation map for the South Island ....................................................................... 11‐6
Figure C11‐4: Summary of North Island parametric study results (single‐storey 2‐leaf wall ‐
Shallow soil) ................................................................................................................................ 11‐7
Figure C12‐1: Photographs of damage in the 2010 Darfield (Canterbury) earthquake due to
deficient or absent diagram anchorages .................................................................................... 12‐2
Figure C12‐2: Examples of wall‐diaphragm anchorages that were partially or fully successful
during the 2010 Darfield earthquake ......................................................................................... 12‐4
viii
Figure C12‐3: Examples of failed anchorages where steel shows no indication of failure ......... 12‐5
Figure C12‐4: Joist oriented parallel to URM wall ....................................................................... 12‐6
Figure C12‐5: Joist oriented perpendicular to URM wall ............................................................ 12‐6
Figure C12‐6: Observed row failure modes for timber connections loaded parallel to grain. ... 12‐7
Figure C12‐7: Assumed failure plain for punching shear failure. ................................................ 12‐8
Figure C13‐1: The Birdcage hotel prior to seismic improvement and being moved ................... 13‐2
Figure C13‐2: Napier School of Music before and after retrofit. (reproduced from Robinson and
Bowman 2000) ............................................................................................................................ 13‐7
Figure C13‐3: Before and after cross sections through Cranmer Square Normal School,
Christchurch (reproduced from Wilby, 1983). Showing changed interior layouts and surfaces and
removed a large amount of the historic material ....................................................................... 13‐7
Figure C13‐4: Primary form of a URM building in Auckland. At this stage the basic shape and the
broad gestures that make up the building are noted. In this case the building has a reasonably
simple form and displays a typical 3 part vertical hierarchy ....................................................... 13‐9
Figure C13‐5: A study of the building’s openings in two dimensions. This has been done firstly as
the physical openings in the URM, and secondly as the forms surrounding the openings.
Interfering with these patterns will change the way that the building reads; something as simple
as changing the colour of the window frames will bring these forward and make the openings
smaller and less distinct ............................................................................................................ 13‐10
Figure C13‐6: Shed 13 on the Wellington waterfront features a distinctive roof that is integral
with its character ....................................................................................................................... 13‐11
Figure C13‐7: A parapet that is integral with the design of the rest of the facade and which
contributes historic information ............................................................................................... 13‐11
Figure C13‐8: The Auckland Ferry Building is integral with its waterfront setting and altering this
relationship would remove historic meaning ........................................................................... 13‐11
Figure C13‐9: Details of openings. Things to take note of here are the bond pattern and
technique used for laying bricks, colouring, striking of mortar, changes in surface, recesses and
depressions, and proportions of windows and other components .......................................... 13‐12
Figure C13‐10: Arcades in Auckland with complex spatial relationships where blocking lines of
sight between spaces or impeding established pedestrian routes will have significant negative
effects ........................................................................................................................................ 13‐14
Figure C13‐11: Severely degraded brick and mortar due to moisture ...................................... 13‐18
Figure C13‐12: Parapets strengthened using near‐surface mounted FRP strips ...................... 13‐19
Figure C13‐13: An extreme case in the 2010 Darfield earthquake where inadequate connections
have resulted in wall collapse ................................................................................................... 13‐19
Figure C13‐14: An assortment of diaphragm fixings ................................................................. 13‐20
ix
Figure C13‐15: Spiral threaded rods being installed to secure an outer layer of bricks ........... 13‐21
Figure C13‐16: Struts from the floor above to improve out‐of‐plane performance ................ 13‐23
Figure C13‐17: Post‐tensioning bars used in the Birdcage hotel with concrete load spreaders cut
into the brick ............................................................................................................................. 13‐24
Figure C13‐18: Simple and visually interesting in‐plane strengthening ................................... 13‐25
Figure C13‐19: Pier damage between openings in the Darfield Earthquake ............................ 13‐26
Figure C13‐20: A concrete moment frame inside the façade of a large URM building in
Wellington (Dunning Thornton Consultants). ........................................................................... 13‐27
Figure C13‐21: Eccentric bracing in a walkway (Dunning Thornton Consultants). ................... 13‐27
Figure C13‐22: An eccentrically braced core (Dunning Thornton Consultants). ...................... 13‐27
Figure C13‐23: Shotcrete walls on the rear of the birdcage hotel ............................................ 13‐29
Figure C13‐24: FRP applied to a wall (Dunning Thornton Consultants). .................................. 13‐29
Figure C13‐25: It is possible to make diaphragm improvements visibly interesting, as with this
recently installed floor .............................................................................................................. 13‐30
Figure C13‐26: A roof bracing system which clashes with the existing structure, resulting in a
confused ceiling space .............................................................................................................. 13‐33
Figure C13‐27: The ornate façade of this building made exterior strengthening difficult ....... 13‐34
Figure C13‐28: The solution was to innovatively strengthen the interior, which allowed retention
of its most important visual elements ...................................................................................... 13‐34
Figure C13‐29: Openings in this Christchurch café allowed placement of frames which
compliments the space ............................................................................................................. 13‐34
Figure C13‐30: Steel structure installed in an intrusive way which is visually distracting ........ 13‐34
Figure C13‐31: No attempt has been made to conceal this new shear wall, rather the aesthetic is
celebrated by the rest of the architecture ................................................................................ 13‐35
Figure C13‐32: Excessive floor and wall ties somehow work with this façade ......................... 13‐35
Figure C14‐1: Diaphragm retrofit examples ............................................................................... 14‐4
Figure C14‐2: Photographs of tested retrofitted diaphragms .................................................... 14‐6
Figure C14‐3: Full‐scale retrofitted diaphragm test results ........................................................ 14‐7
Figure C14‐4: Diaphragm chord schematic ................................................................................. 14‐7
Figure C14‐5: Subdiaphragm example (from Oliver 2010) ......................................................... 14‐9
Figure C14‐6: Typical Out‐of‐Plane Wall Support Detail at Diaphragm Opening (from Oliver 2010)
.................................................................................................................................................. 14‐10
Figure C16‐1: Post‐tensioning Retrofit Technique ...................................................................... 16‐3
Figure C16‐2: Definition of Limit States ...................................................................................... 16‐3
x
Figure C16‐3: Flowchart for Post‐tensioning Seismic Retrofit Design ......................................... 16‐4
Figure C16‐4: Typical Stress‐Strain Plots ..................................................................................... 16‐5
Figure C16‐5: Post‐tensioned In‐plane Pier Definition and Stress Profile at Initial State ........... 16‐6
Figure C16‐6: Post‐tensioned In‐plane Pier Equilibrium at Nominal Strength Limit State ......... 16‐7
Figure C16‐7: Post‐tensioned In‐plane Pier Equilibrium at First Cracking Limit State ................ 16‐8
Figure C16‐8: Post‐tensioned In‐plane Pier Equilibrium at Hinge Formation Limit State ........... 16‐9
Figure C16‐9: Test Frame Dimensions ....................................................................................... 16‐10
Figure C16‐10: Calculated Vs Experimental Values ................................................................... 16‐10
Figure C16‐11: Post‐tensioned Out‐of‐plane Wall Definition and Stress Profile at Initial State ... 16‐
11
Figure C16‐12: Post‐tensioned Out‐of‐plane Wall at Nominal Strength Limit State ................ 16‐11
Figure C16‐13: Calculated Vs Experimental Values ................................................................... 16‐14
Figure C16‐14: URM Wall Details .............................................................................................. 16‐16
Figure C17‐1: Examples of FRP systems. ..................................................................................... 17‐2
Figure C17‐2: Examples of non‐vertical FRP retrofit systems. .................................................... 17‐4
Figure C17‐3: Examples of vertical FRP retrofit systems. ............................................................ 17‐4
Figure C17‐4: FRP continuous fabric system extended ld beyond the top of the doors. The vertical
FRP system was designed for the pier between the two doors. ................................................. 17‐6
Figure C17‐5: FRP design flowchart ............................................................................................. 17‐7
Figure C17‐6: Truss analogy for FRP‐retrofitted URM wall ......................................................... 17‐8
Figure C17‐7: Plot showing the relationship between fe and .............................................. 17‐10
Figure C17‐8: FRP retrofit for wall in Example 15.1 .................................................................. 17‐13
Figure C17‐9: FRP retrofit for building in Example 15.2 ............................................................ 17‐14
Figure C18‐1: NSM CFRP retrofit application .............................................................................. 18‐3
Figure C18‐2: Stress and strain profile ........................................................................................ 18‐4
Figure C18‐4: Typical wall failure mechanisms ........................................................................... 18‐7
Figure C18‐3: Out‐of‐plane design procedure ............................................................................. 18‐7
Figure C18‐5: Experimental program .......................................................................................... 18‐8
Figure C18‐6: Experimental results of NSM CFRP wall testing .................................................... 18‐9
Figure C18‐7: CFRP wall retrofit for Example C18.7.2 ............................................................... 18‐13
Figure C19‐1: Sequence of FRS strengthening procedures. ........................................................ 19‐3
Figure C19‐2: In‐plane FRS design procedure ............................................................................. 19‐4
xi
Figure C19‐3: Distances of xt and xc with respect to axial load ................................................... 19‐6
Figure C19‐4: Out‐of‐plane FRS design procedure ..................................................................... 19‐7
Figure C19‐5: NSM reinforcement .............................................................................................. 19‐8
Figure C19‐6: Illustration of FRS effective depth ........................................................................ 19‐8
xii
List of Tables
Table C1‐1: New Zealand URM typologies .................................................................................. 1‐10
Table C1‐2: Number of URM buildings from QV according to construction decade .................. 1‐11
Table C1‐3: Estimated number of potentially earthquake prone and earthquake risk URM
buildings ...................................................................................................................................... 1‐13
Table C2‐1: Height to thickness ratio correction factor for mortar compressive strength ........... 2‐2
Table C2‐2: Parameters used in the NZSEE (2006) guideline ........................................................ 2‐7
Table C2‐3: Mohs hardness scale .................................................................................................. 2‐9
Table C2‐4: Mortar dividing factors based on h/t ratio ............................................................... 2‐14
Table C2‐5: List of αf'm values from publications ........................................................................ 2‐23
Table C3‐1: Full‐scale diaphragm testing summary ...................................................................... 3‐8
Table C3‐2: Example backbone force‐displacement data for time‐history analysis ................... 3‐10
Table C4‐1: Modal parameter extraction methods ....................................................................... 4‐8
Table C4‐2: Recommended suite profile for each seismic zone ................................................. 4‐15
Table C8‐1: Wall Schematics ......................................................................................................... 8‐6
Table C8‐2: Wall Specifications (Russell et al. 2012) ..................................................................... 8‐6
Table C8‐3: Comparison of predicted strength accounting for and neglecting flanges (Russell et
al. 2012) ......................................................................................................................................... 8‐6
Table C11‐1: Estimated wall dimensions ..................................................................................... 11‐8
Table C11‐2: %NBS.Z for slenderness ratio of 17.1, lower storey ............................................... 11‐9
Table C11‐3: %NBS.Z for slenderness ratio of 19.6, top‐storey .................................................. 11‐9
Table C11‐4: Estimated parapet dimensions ............................................................................. 11‐10
Table C11‐5: Calculating %NBS for regions excluding near fault effect .................................... 11‐10
Table C11‐6: Calculating %NBS for regions including near fault effect ..................................... 11‐10
Table C11‐7: Calculating shear demand .................................................................................... 11‐11
Table C11‐8: Calculating shear capacity .................................................................................... 11‐11
Table C11‐9: One‐leaf wall assessment ..................................................................................... 11‐12
Table C11‐10: One‐leaf wall assessment ................................................................................... 11‐13
Table C16‐1: k Value for Moment Capacity Evaluation at Hinge Formation Limit State ............ 16‐9
Table C16‐2: Post‐tensioned Wall Dimensions and Post‐tensioning Details ............................ 16‐13
Table C16‐3: Masonry Materials ............................................................................................... 16‐15
xiii
Table C16‐4: Input Parameters ................................................................................................. 16‐15
Table C17‐1: Typical values of FRP thickness, design rupture strain and modulus of elasticity. .. 17‐
11
Table 18‐1: Typical values of material properties for CFRP ........................................................ 18‐3
xiv
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C1 ‐ History and Prevalence of Unreinforced
Masonry Buildings in New Zealand
C1.1 Introduction
New Zealand’s masonry construction heritage is comparatively young, spanning from 1833 until
the present time – a period of less than 200 years. Furthermore, the majority of URM
construction in New Zealand occurred between 1880 and 1935. Consequently, a study of New
Zealand’s URM building stock has a narrow scope in comparison with international norms (see
for instance Binda and Saisi (2005), Lourenço (2006) and Magenes (2006)). This comparatively
narrow time period has the advantage of facilitating the documentation and reporting of New
Zealand URM construction practice with a greater degree of accuracy than is often possible in
countries with an older and more diverse history of URM construction (Binda 2006).
C1.2.1 Early Settlement
The first inhabitants of Aotearoa New Zealand were groups of Polynesian explorers who
discovered and settled the islands in the period A.D. 800‐1000 (King 2003). These people did not
develop a tradition of building in masonry, but instead built using timber, earth and most
commonly raupo (bulrush). There are however, numerous stone‐related archaeological sites in
New Zealand attributed to Maori society, the majority of which are garden walls or associated
with fortifications. Examples of Maori construction are shown in Figure C1‐1 and Figure C1‐2.
Figure C1‐1: Raupo whare (house) at Wairau Figure C1‐2: Raupo and timber whares of a
Pa, ca. 1880 (Alexander Turnbull Library). Maori village at Maketu, Kawhia, 1895
(Alexander Turnbull Library).
Captain James Cook anchored off the coast of New Zealand on 9 October 1769. This event was
followed by a gradual haphazard increase in the population of Europeans in New Zealand over
the next 70 years, initially primarily associated with whaling, but also involving kauri timber
extraction and gold mining. Jacobs (1985) reports that the European population of New Zealand
in 1830 was probably a little more than 300. By 1839 the number had risen to possibly 2000, and
C1‐1
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
at the beginning of the 1850s there was 26,000 Europeans in New Zealand (see Figure C1‐3 for
recorded population data). These first European settlers found themselves without their familiar
building materials, so initially emulated the style and construction of Maori dwellings (Shaw
2003). For the most significant early buildings, such as churches and assembly buildings,
architects from Australia or England were commissioned.
5,000,000
4,000,000
Population
3,000,000
2,000,000
1,000,000
0
1840 1860 1880 1900 1920 1940 1960 1980 2000 2020
Year
Figure C1‐3: New Zealand total population data, 1858‐2007.
Captain William Hobson’s arrival in 1840 as the First Governor General of New Zealand marked
the beginning of New Zealand as a British colony. Construction during the 1840s to 1860s was
primarily of timber for residential and small commercial buildings (see Figure C1‐4), but masonry
buildings did begin to appear close to harbours (see Figure C1‐5) and in city centres. In the late
1850’s Christchurch prospered from the wool trade and this allowed the transition from wood to
stone and clay brick masonry for public buildings. In Christchurch the town hall was built in stone
in 1862‐1863, the first stone building of Christ’s College was constructed in 1863, and the stone
Provincial Council Chambers was completed in 1864 (Wilson 1984). Oliver (2006) reports that
clay bricks were first manufactured in Auckland in 1852, with production of about 5,000 bricks
per day. In Auckland central city the construction of timber buildings was not restricted until the
City of Auckland Building Act of 1856, with a fire in central Auckland in 1858 provided further
impetus for the transition from timber to clay brick masonry construction. Similar fires in other
city centres resulted in this transition from timber to masonry construction being mirrored
throughout New Zealand, such as fires in inner city Christchurch in 1861, 1864, and 1866. The
centre of Lyttelton was also destroyed in a fire on October 24th, 1870 (Christchurch City Libraries
2006; Wilson 1984). The combustibility of timber structures prompted the move to URM
construction due to its fire resistant properties, with the high level of seismic activity in New
Zealand not influencing the decision, as the poor lateral force resisting properties of URM were
largely unknown at this time of mass URM construction. The fire‐proof nature of masonry led to
it being readily adopted as the appropriate building material for high importance structures such
as government buildings, schools, and churches.
C1‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C1‐4: Shops on Queen Street, Auckland, Figure C1‐5: Queen Street and Queen Street
1859 (Alexander Turnbull Library). Wharf, Auckland, in 1882 (Alexander Turnbull
Library).
By 1914 the central area of Christchurch had been largely rebuilt, resulting in a city that was
“interesting for its architectural variety, pleasing for its scale and distinctively New Zealand”
(Wilson 1984). Figure C1‐12 and Figure C1‐13 show photos of historical Christchurch from 1885
and 1910 respectively. Two of the many influential architects of Christchurch were J.C. Maddison
(1850‐1923), whose design focus was inspired by the Italianate style, and J.J. Collins (1855–
1933), who in partnership with R.D. Harman (1859‐1927) chose brick masonry as their medium
for large commercial and institutional buildings. By the 1920’s wooden structures in Christchurch
were rare, and were seen as small irregular relics of the past.
C1.3 URM Performance in Past Earthquakes
C1.3.1 Wairarapa, Canterbury and Murchison Earthquakes
The Wairarapa earthquake occurred on Tuesday 23 January 1855, had an estimated magnitude
of M8.2 (Grapes and Downes 1997) and is the largest earthquake to have occurred in New
C1‐3
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
Zealand since the time of systematic European colonisation (see Dowrick and Rhoades (1998) for
a catalogue of major New Zealand earthquakes from 1901‐1993). The shock was felt across
almost the whole country, was highly destructive in Wellington, and also caused severe damage
in Wanganui and Kaikoura. Between seven and nine people were killed in the earthquake, and
five others sustained injuries that required hospitalisation.
Figure C1‐6: Group photograph of the Figure C1‐7: Brick building under
construction workers that built the Stratford construction, ca 1920 (Alexander Turnbull
Public Hospital during 1906‐1907 (Alexander Library).
Turnbull Library).
Figure C1‐8: The 1833 Stone Store at Kerikeri Figure C1‐9: Two Chinese miners in front of
was built by the Church Missionary Society. (A a stone cottage in central Otago, ca 1860
P Godber Collection, Alexander Turnbull (Alexander Turnbull Library).
Library).
The M7.1 North Canterbury earthquake in 1888 (Stirling et al. 2008) caused severe damage to
buildings made of cob and stone masonry and caused minor damage to buildings in Christchurch
(PapersPast 2010). A later earthquake in 1901 centred in Cheviot damaged the spire on the
Canterbury Cathedral for the third time in its short life and led to reconstruction of the spire in
timber. Further details are reported in Dizhur et al. (2010).
C1‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C1‐10: Collapse of a new masonry Figure C1‐11: Looking along a row of
auction market building, Queen Street, commercial buildings on Queen Street,
1865 (Alexander Turnbull Library). Auckland, ca 1910 (Alexander Turnbull
Library).
Figure C1‐12: Victorian Christchurch in 1885 Figure C1‐13: Christchurch’s first
(Coxhead 1885) ‘skyscraper’, photo circa 1910
(Brittenden Collection 1910)
The M7.8 earthquake that struck Murchison on the 17th of June 1929 was felt throughout New
Zealand (Dowrick 1994). Fortunately, the most intense shaking occurred in a mountainous and
densely wooded area that was sparsely populated. Casualties were therefore comparatively light
and the damage was mostly confined to the surrounding landscape, where the shaking triggered
extensive landslides over thousands of square kilometres. Nonetheless, the shock impacted with
damaging intensities as far away as Greymouth, Cape Farewell and Nelson (see Figure C1‐14 and
Figure C1‐15). Fifteen people were killed in the Murchison earthquake.
C1.3.2 The 1931 Hawke’s Bay Earthquake, Napier
As reported above, it was the combustibility of timber construction that prompted the focus
towards building in clay brick unreinforced masonry, and occasionally in stone masonry. Early
earthquakes in the Wellington region resulted in a slower adoption of masonry construction.
This caution proved to be well justified. On the morning of 3 February 1931 the Hawke’s Bay
region of the eastern North Island was struck by an M7.8 earthquake that completely destroyed
much of the city of Napier (see Figure C1‐16 to Figure C1‐18). Fires swept through the wreckage,
destroying much of what was left. Eight nurses died when the reinforced concrete Napier nurses
home collapsed, and perhaps the largest brick masonry building to collapse was the Napier
Anglican Cathedral (see Figure C1‐19). The shaking resulted in damage from Taupo to
C1‐5
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
Wellington, and left 30,000 people homeless. The official death toll was 256, and the event
remains the worst disaster of any type to occur on New Zealand soil (Dalley and McLean 2005;
Dowrick 1998).
Figure C1‐14: General store damaged by the Figure C1‐15: Damaged business premises
1929 Murchison earthquake (Alexander after the earthquake of 17 June 1929
Turnbull Library). (Alexander Turnbull Library).
Figure C1‐16: Hastings Street, Napier, circa Figure C1‐17: View down Hastings Street,
1914 (Alexander Turnbull Library). Napier after the 1931 earthquake (Alexander
Turnbull Library).
C1‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C1‐18: Looking over Napier at the Figure C1‐19: Ruins of the Napier Anglican
buildings ruined by the 1931 earthquake and the Cathedral after the 1931 Napier
fires (Alexander Turnbull Library). earthquake (Alexander Turnbull Library).
Following the demonstrated poor seismic response of URM buildings during the Hawke’s Bay
earthquake there was a rapid decline in the use of unreinforced clay brick masonry construction,
which was eventually prohibited in 1965 (New Zealand Standards Institute 1965a). The
devastation of the 1931 Hawke’s Bay earthquake prompted the New Zealand Government to
ban the use of unreinforced masonry on public buildings (Oliver 2006) and also prompted the
government to develop a national building code, with the New Zealand Standards Institution
formed in 1932. This institution has survived to the present day, and is now referred to as
Standards New Zealand.
C1.3.3 Gisborne and Darfield Earthquakes
The M6.8 2007 Gisborne earthquake (Francois‐Holden et al. 2008) caused damage to numerous
unreinforced masonry buildings, including the collapse of 22 parapets (Davey and Blaikie 2010).
Examples of damage to URM buildings in the 2007 Gisborne earthquake are shown in Figure
C1‐20 and Figure C1‐21. The 2010 Darfield earthquake caused extensive damage to a number of
unreinforced masonry buildings (Dizhur et al. 2010; Ingham and Griffith 2011). Whilst this
damage to important heritage buildings was the largest natural disaster to occur in New Zealand
since the 1931 Hawke’s Bay earthquake, the damage was consistent with projections for the
scale of this earthquake, and indeed even greater damage might have been expected. In general,
the nature of damage was consistent with observations previously made on the seismic
performance of unreinforced masonry buildings in large earthquakes, with aspects such as
toppled chimneys and parapets, failure of gables and poorly secured face‐loaded walls, and in‐
plane damage to masonry frames all being extensively documented (see Figure C1‐22 and Figure
C1‐23).
C1‐7
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
Figure C1‐20: Toppled parapet in the 2007 Figure C1‐21 Out‐of‐plane failure of a gable
Gisborne earthquake. wall in the 2007 Gisborne earthquake.
Figure C1‐22: Out‐of‐plane wall failure at the Figure C1‐23: Out‐of‐plane wall failure at 118
corner of Worcester and Manchester streets in Manchester Street in the 2010 Darfield
the 2010 Darfield earthquake. earthquake.
C1.4 New Zealand URM Building Stock
In order to foster a general understanding of the architectural character of New Zealand’s URM
building stock, a characterisation study was performed that identified the seven building
typologies shown in Figure C1‐24 and reported in Table C1‐1. The intent of the typology study is
to assist in initial seismic assessment. Further details are reported in Russell and Ingham (2010)
and Russell and Ingham (2011).
C1‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Typology A building ‐ one storey isolated Typology B building ‐ one storey row
Typology C building ‐ two storey isolated Typology D building ‐ two storey row
Typology E building ‐ three+ storey isolated Typology F building ‐ three+ storey row
Typology G building ‐ religious Typology G building ‐ institutional
Figure C1‐24: Photographic examples of New Zealand URM typologies
C1‐9
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
Table C1‐1: New Zealand URM typologies
Importance
level
Type Description Details
(from
NZS 1170.0)
C1.4.1 Estimation of URM Population and Value
In order to better understand the aggregated seismic performance of the nationwide URM
building stock, two parallel exercises were performed to estimate the number and distribution
of URM buildings throughout New Zealand. The first method involved an assumption that URM
buildings were constructed approximately in proportion to the national population distribution
of the era, and the second method was based upon data purchased from Quotable Value (QV)
regarding the exterior building fabric in the QV database. Both methods are estimates only, but
C1‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
provided similar findings, with further details reported in Russell and Ingham (2010). The QV
database was analysed according to construction date, building height and financial value, with
Table C1‐2 reporting the decade in which each URM building was constructed.
Table C1‐2: Number of URM buildings from QV according to
construction decade
Decade URM Buildings
1870 – 1880 43
1880 – 1890 23
1890 – 1900 71
1900 – 1910 469
1910 – 1920 646
1920 – 1930 878
1930 – 1940 514
1940 – 1950 218
Mixed 726
Total 3590
Table C1‐2 clearly shows a trend where the number of URM buildings initially increased until the
end of the 1920s, and subsequently declined. This trend follows the increasing rate of European
immigration and associated infrastructure development in New Zealand in the early 20th
Century (see Figure C1‐3), until the 1931 M7.8 Hawke’s Bay earthquake, after which URM was
no longer considered a favourable building material.
A report prepared for the Department of Internal Affairs in 2002 (Hopkins 2002; Hopkins and
Stuart 2003) showed that the total floor area of buildings in 32 cities and towns throughout New
Zealand was approximately 27,200,000 m2. The total floor area of URM buildings extracted from
the QV database was approximately 2,100,000 m2, suggesting that URM buildings make up
approximately 8% of the total New Zealand commercial building stock in terms of floor area.
From the QV database it was also possible to establish that 86% of the nationwide URM building
stock is either a one or two storey building, and that the aggregated value of these buildings
(based on an assessment period between July 2005 and September 2008) is approximately
NZ$1.5 billion. However, it must be recognised that many buildings have a worth greater than
C1‐11
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
their financial valuation, including an architectural, historic or heritage value to the community,
which can be difficult to quantify (Goodwin 2008; Goodwin et al. 2009).
C1.4.2 Estimated Vulnerability of URM Buildings
Following determination of the number of URM buildings and their approximate regional
distribution, an analysis was performed to estimate the expected vulnerability of the URM
building population. The details of the analysis are reported in Russell and Ingham (2010), with
the results reported in Figure C1‐25 and in Table C1‐3. It is recognised that the analysis was
essentially qualitative in nature and can be expected to overestimate the number of poorly
performing URM buildings, primarily because of the conservative nature of the IEP.
Nevertheless, as an informative estimate of the nature of the vulnerability of New Zealand’s
URM building stock, this analysis is considered robust. Additionally, this analysis does not take
into account the number of buildings which have already been seismically improved, which
Thornton (2010) notes is not insignificant. Data that became available following damage to
URM buildings in the 2010 Darfield earthquake (reported in Ingham and Griffith (2011))
indicated general support for this analysis, although it was evident following the earthquake that
the analysis presented in Figure C1‐25 underestimated the total number of URM buildings in the
Canterbury region (see Dizhur et al. (2010)).
Figure C1‐25: Estimated %NBS of URM buildings in Provinces throughout New Zealand
C1‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C1‐3: Estimated number of potentially earthquake prone and earthquake risk URM
buildings
Otago and
66 5% 664 33% 126 26%
Southland
C1.5 New Zealand Building Codes Pertaining to URM Construction
The Great Depression in the 1930s and the outbreak of World War II significantly slowed
progress in the construction sector, and few large buildings of any material were constructed in
the period between 1935 and 1955 (Megget 2006; Stacpoole and Beaven 1972). Equally
important in the history of URM buildings in New Zealand was the 1931 M7.8 Hawkes Bay
earthquake, and the changes in building provisions which it precipitated. Later in 1931, in
response to that earthquake, the Building Regulations Committee presented a report to the
Parliament of New Zealand entitled “Draft General Building By‐Law” (Cull 1931), which was the
first step towards requiring seismic provisions in the design and construction of new buildings. In
1935, this report evolved into NZSS no. 95, published by the newly formed New Zealand
Standards Institute, and required a horizontal acceleration for design of 0.1g, and this
requirement applied to the whole of New Zealand (New Zealand Standards Institute 1935). NZSS
no. 95 also suggested that buildings for public gatherings should have frames constructed of
reinforced concrete or steel. The By‐Law was not enforceable, but it is understood that it was
widely used especially in the larger centres of Auckland, Napier, Wellington, Christchurch and
Dunedin (Megget 2006). The provisions of NZSS no. 95 were confined to new buildings only, but
the draft report acknowledged that strengthening of existing buildings should also be
C1‐13
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
considered, and that alterations to existing buildings were required to comply (Davenport 2004).
In 1939 and 1955 new editions of this By‐Law were published, and apart from suggesting in 1955
that the seismic coefficient vary linearly from zero at the base to 0.12 at the top of the building
(formerly the seismic coefficient was uniform up the height of the building), there were few
significant changes (Beattie et al. 2008). It was not until 1965 that much of the recent research
at the time into seismic design was incorporated into legislation. The New Zealand Standard
Model Building By‐Law NZSS 1900 Chapter 8:1965 explicitly prohibited the use of URM: (a) in
Zone A; (b) of more than one storey or 15 ft (4.6 m) eaves height in Zone B; (c) of more than two
storeys or 25 ft (7.6 m) eaves height in Zone C. These zones refer to the seismic zonation at the
time, which have subsequently changed and evolved. Zone A consisted of regions of the highest
seismic risk and Zone C consisted of regions of the lowest seismic risk (New Zealand Standards
Institute 1965b). Again, the provisions of this By‐Law did not apply automatically and had to be
adopted by local authorities.
The 1965 code required that buildings be designed and built with “adequate ductility”, although
further details were not given. The next version of the loadings code was published in 1976 as
NZS 4203 (Standards Association of New Zealand 1976), and was a major advance on the 1965
code. Most importantly, the 1976 loadings code was used in conjunction with revised material
codes: steel, reinforced concrete, timber and reinforced masonry, which all required specific
detailing for ductility. Thus after the publication of this code in 1976, unreinforced masonry was
explicitly prohibited as a building material throughout the whole of New Zealand.
The use of URM was implicitly discouraged through legislation from as early as 1935, and
although it was still allowed in some forms after 1965, observations of existing building stock
show its minimal use from 1935 onwards, especially for larger buildings. This is thought to be
significantly attributable to the exceptionally rigorous quality of design and construction by the
Ministry of Works at the time (Johnson 1963; Megget 2006). Although two storey URM buildings
were permitted in Auckland (Zone C) after 1965, only three existing URM buildings in Auckland
City constructed after 1940 have been identified. All three are single storey and they were
constructed in 1950, 1953 and 1955.
C1.5.1 Provisions for the Seismic Upgrade of Existing Buildings
As building codes were being developed for the design of new buildings, attention was also
given to the performance of existing buildings in earthquakes. The first time this was addressed
in legislation was Amendment 301A to the 1968 Municipal Corporations Act (New Zealand
Parliament 1968). This Act allowed territorial authorities, usually being boroughs, cities or
district councils, to categorise themselves as earthquake risk areas and thus to apply to the
government to take up powers to classify earthquake prone buildings and require owners to
reduce or remove the danger. Buildings (or parts thereof) of high earthquake risk were defined
as being those of unreinforced concrete or unreinforced masonry with insufficient capacity to
resist earthquake forces that were 50% of the magnitude of those forces defined by NZS 1900
Chapter 8:1965. If the building was assessed as being “potentially dangerous in an earthquake”,
the council could then require the owner of the building within the time specified in the notice
to remove the danger, either by securing the building to the satisfaction of the council, or if the
C1‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
council so required, by demolishing the building. Most major cities and towns took up the
legislation, and as an indication of the effect of this Act, between 1968 and 2003 Wellington City
Council achieved strengthening or demolition of 500 out of 700 buildings identified as
earthquake prone (Hopkins et al. 2008).
Auckland City Council, in spite of having a low seismicity, took a strong interest in the legislation
and this led to considerable activity in strengthening buildings (Boardman 1983). In Christchurch,
a moderately high seismic zone, the City Council implemented the legislation, but adopted a
more passive approach, generally waiting for significant developments to trigger the
requirements. In Dunedin, now seen to be of low seismic risk, little was done in response to the
1968 legislation although strengthening of schools, public buildings and some commercial
premises was achieved. As a result, Dunedin has a high percentage of URM buildings compared
with many other cities in New Zealand (Hopkins 2009). Megget (2006) and Thornton (2010) state
that much of the strengthening in Wellington was accomplished with extra shear walls, diagonal
bracing or buttressing and the tying of structural floors and walls together, and that many brittle
hazards such as parapets and clock towers had been removed after the two damaging 1942
South Wairarapa earthquakes (M7.0 & M7.1) which were felt strongly in Wellington. Hopkins et
al. (2008) noted that “there was criticism at the loss of many older heritage buildings and at the
use of intrusive retrofitting measures which were not harmonious with the architectural fabric of
the building (McClean 2009). At the same time, this did provide an opportunity in many cases for
the land on which the old building was situated to be better utilised with new, larger and more
efficiently designed structures.”
“A major drawback of the 1968 legislation, which endured until 2004, surviving intact with the
passage of the Building Act 1991, was that the definition of an earthquake prone building and
the required level to which such buildings should be improved remained tied to the 1965 code.
Most territorial authorities called for strengthening to one‐half or two‐thirds of the 1965 code,
and many buildings which were strengthened to these requirements were subsequently found
to fall well short of the requirements of later design standards for new buildings” (Hopkins et al.
2008). (Wellington City Council found that in January 2008, of 97 buildings which had been
previously strengthened, 61 (63%) were subsequently identified as potentially earthquake prone
(Bothara et al. 2008; Stevens and Wheeler 2008)). This situation was recognised by the New
Zealand Society for Earthquake Engineering (NZSEE), who were also concerned about the
performance of more modern buildings, particularly after the observed poor performance of
similarly aged buildings in earthquakes in Northridge, California (1994) and Kobe, Japan (1995).
NZSEE pushed for new, more up‐to‐date and wide‐ranging legislation. This push was supported
by the Building Industry Authority, later to become part of the Department of Building and
Housing, and a new Building Act came into effect in August 2004 (New Zealand Parliament
2004). This brought in new changes as to what constituted an “Earthquake Prone Building”. In
particular, the definition of an earthquake prone building was tied to the current design
standard of the time, and no longer to the design standard of any particular year. The legislation
allowed any territorial authority that is satisfied that a building is earthquake prone to require
the owner to take action to reduce or remove the danger. Each territorial authority was required
to have a policy on earthquake prone buildings, and to consult publicly on this policy before its
C1‐15
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
adoption. Policies were required to address the approach and priorities and to state what special
provisions would be made for heritage buildings. The 2004 legislation applied to all types of
buildings except small residential ones, (residential buildings were excluded unless they
comprised 2 or more storeys and contained 3 or more household units).
As soon as the 1968 legislation came into effect to attempt to mitigate the effects of earthquake
prone buildings, the New Zealand National Society for Earthquake Engineering set up a steering
committee to provide a code of practice in an effort to assist local authorities to implement the
legislation. Since the first draft code of practice published by the NZNSEE (1972), several
successive publications have been produced, each extending on the previous version. These
guidelines have been instrumental in helping engineers and territorial authorities to assess the
expected seismic performance of existing buildings consistent with the requirements of the
legislation. Guidelines for assessing and upgrading earthquake risk buildings were published as a
bulletin article in 1972 (NZNSEE 1972) and then separately published the following year, which
became colloquially known as the “Brown Book” (NZNSEE 1973). This document provided
guidelines for surveying earthquake risk buildings and for the identification of particularly
hazardous buildings and features, and was found to be helpful in many respects. It did not
establish or recommend strength levels to which earthquake prone buildings should be
upgraded, and thus standards varied from one area to another. It was implicit that strengthening
be to more than half the standard required in Chapter 8 of the 1965 NZSS Model Building By‐
Law.
In 1982, NZSEE established a study group to examine and rationalise the use of these guidelines
and to produce further guidelines and recommendations. This activity culminated in the
publication in 1985 of what became known as the “1985 Red Book” (NZNSEE 1985). Again, this
document was primarily of a technical nature and the responsibilities of what to do with
buildings still rested with local authorities. The publication was intended to promote a consistent
approach throughout New Zealand for the strengthening of earthquake risk buildings and
included a recommended level to which buildings should be strengthened plus the time scale to
complete the requirements. The basic objective was to establish a reasonably consistent
reduction of the overall risk to life which the country’s stock of earthquake risk buildings
represented. Based on overseas experiences, particularly in Los Angeles in Southern California, a
philosophy was accepted of providing owners of earthquake risk buildings with the option of
interim securing to gain limited extension of useful life, after which the building should be
strengthened to provide indefinite future life. The design of interim securing systems was to be
based on minimum seismic coefficients which represented two‐thirds of those specified in NZSS
1900, Chapter 8 (New Zealand Standards Institute 1965b). For “permanent” strengthening
measures, it was recommended that the building be strengthened to the standard of a new
building, but with the design lateral forces reduced depending on the occupancy classification
and type of strengthening system. This publication was widely used by territorial authorities and
designers.
In 1992 the NZNSEE again set up a study group to review the 1985 publication, and this resulted
in another publication, which similarly became colloquially known as the “1995 Red Book”
(NZNSEE 1995). This document extended the approach and content of its predecessor and took
C1‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
into account the changing circumstances, technical developments and improved knowledge of
the behaviour of URM buildings in earthquakes. In particular, earthquake risk buildings in that
document were taken to include all unreinforced masonry buildings, and not just those which
were defined as “earthquake prone” in terms of the Building Act of the time, which still referred
back to the 1965 code. Another key difference from the 1985 Red Book was that a single stage
approach to strengthening was suggested, in contrast to the two stage securing and
strengthening procedure of the 1985 document. The guidelines also highlighted the differences
in analysis for unsecured buildings in comparison to a building which has positive connections
between floor, roof and wall elements, and cantilever elements secured or removed. Greater
emphasis was placed on the assessment of the likely performance of URM buildings in their
original form and with interim securing only in place, as distinct from the performance of the
building with any strengthening work which was subsequently found to be necessary.
Furthermore, material strengths were given in ultimate limit state format. Historic or heritage
buildings were not given any specific or separate treatment, and the guidelines stated that “the
issues of risk versus the practicalities of strengthening associated with historic buildings require
evaluation on a case‐by‐case basis. The principal problem with such buildings is that the greater
the level of lateral forces that is specified for strengthening, the greater the risk of damaging the
fabric that is to be preserved” (NZNSEE 1995).
After the introduction of a new Building Act in 2004 (New Zealand Parliament 2004) the
Department of Building and Housing supported NZSEE in producing a set of guidelines,
“Assessment and Improvement of the Structural Performance of Buildings in Earthquakes”
(NZSEE 2006). This was a major review and extension of previous guidelines, to account for the
wider scope of the proposed new legislation. Prior to enacting The Building Act 2004, the term
‘earthquake risk building’ related only to URM buildings, but now an earthquake prone building
could be of any material; steel, concrete, timber or masonry. The level of risk posed by buildings
constructed as recently as the 1970s was more widely appreciated, in particular the inadequate
performance of reinforced concrete structures due to deficient detailing. Definitions of
“earthquake prone” and “earthquake risk” also changed. Essentially, earthquake prone buildings
were defined as those with one‐third or less of the capacity of a new building. While The
Building Act itself still focussed on buildings of high risk (earthquake prone buildings), NZSEE
considered earthquake risk buildings to be any building which is not capable of meeting the
performance objectives and requirements set out in its guidelines, and earthquake prone
buildings formed a subset of this. Moreover, NZSEE expressed a philosophical change, in
acknowledgment of the wide range of options for improving the performance of structures that
are found to have high earthquake risk. Some of these options involve only the removal or
separation of components, and others affect a relatively small number of members. In line with
performance‐based design thinking, the term “strengthening” was replaced with “improving the
structural performance of”, highlighting the fact that such solutions as base isolation were not
“strengthening” but were an effective way of improving structural performance.
The 2006 guidelines (NZSEE 2006) provided both an initial evaluation procedure (IEP) and a
detailed analysis procedure. The IEP can be used for a quick and preliminary evaluation of
existing buildings, and takes into account the building form, natural period of vibration, critical
C1‐17
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
C1.6 Acknowledgements
The assistance of David Hopkins, Patrick Cummuskey, Richard Deakin, Claire Stevens, Katherine
Wheeler, Neil McLeod and Bruce Mutton in providing information is gratefully acknowledged.
C1.7 References
Beattie, G. J., Megget, L. M. and Andrews, A. L. (2008). "The historic development of earthquake
engineering in New Zealand" 14th World Conference on Earthquake Engineering, Beijing,
China, 12 ‐ 17 Oct.
Binda, L. (2006). "The difficult choice of materials used for the repair of brick and stone masonry
walls" The First International Conference on Restoration of Heritage Masonry Structures,
Cairo, Egypt, 24 ‐ 27 April.
Binda, L. and Saisi, A. (2005). "Research on historic structures in seismic areas in Italy." Progress
in Structural Engineering and Materials, 7(2), 71‐85.
Boardman, P. R. (1983). "Case studies: earthquake risk buildings. Restoration of Old Auckland
Customhouse." Bulletin of the New Zealand National Society for Earthquake Engineering,
16(1), 73 ‐ 79.
Bothara, J. K., Jury, R. D., Wheeler, K. and Stevens, C. (2008). "Seismic assessment of buildings in
Wellington: Experiences and challenges" 14th World Conference on Earthquake
Engineering, Beijing, China, 12 ‐ 17 Oct.
Brittenden Collection (1910). "View south‐east from the Cathedral tower" CHAC/CM 1199,
Christchurch.
Christchurch City Libraries (2006). "Early Christchurch."
<http://christchurchcitylibraries.com/heritage/earlychristchurch>. (18 October, 2010).
Coxhead, F. A. (1885). "Victorian Christchurch" CM 182, Christchurch.
C1‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Cull, J. E. L. (1931). Report of the Building Regulations Committee, Report to the New Zealand
House of Representatives, H‐21.
Dalley, B. and McLean, G. (2005). Frontier of Dreams; The Story of New Zealand, Hodder Moa,
Auckland.
Davenport, P. N. (2004). "Value of assessing seismic vulnerability." New Zealand Society for
Earthquake Engineering Conference, Rotorua, New Zealand, 19 ‐ 21 March.
Davey, R. A. and Blaikie, E. L. (2010). "Predicted and observed performance of masonry parapets
in the 2007 Gisborne earthquake" 2010 Annual Conference of the New Zealand Society for
Earthquake Engineering, Wellington, New Zealand, 26 ‐ 28 March.
Dizhur, D., Ismail, N., Knox, C., Lumantarna, R. and Ingham, J. M. (2010). "Performance of
unreinforced and retrofitted masonry buildings during the 2010 Darfield earthquake."
Bulletin of the New Zealand Society for Earthquake Engineering, 43(4), 321‐339.
Dowrick, D. J. (1994). "Damage and intensities in the magnitude 7.8 1929 Murchison, New
Zealand, earthquake." Bulletin of the New Zealand National Society for Earthquake
Engineering, 27(3), 190‐204.
Dowrick, D. J. (1998). "Damage and intensities in the magnitude 7.8 1931 Hawke’s Bay, New
Zealand, earthquake." Bulletin of the New Zealand National Society for Earthquake
Engineering, 31(3), 139‐163.
Dowrick, D. J. and Rhoades, D. A. (1998). "Magnitudes of New Zealand earthquakes, 1901‐1993."
Bulletin of the New Zealand National Society for Earthquake Engineering, 31(4), 260‐280.
Francois‐Holden, C., Bannister, S., Beaven, J., Cousins, J. F., B., McCaffrey, R., McVerry, G.,
Reyners, M., Ristau, J., Samsonov, S. and Wallace, L. (2008). "The Mw 6.6 Gisborne
earthquake of 2007: Preliminary records and general source characterisation." New Zealand
Society for Earthquake Engineering Bulletin, 41(4), 266‐277.
Goodwin, C. O. (2008). "Architectural considerations in the seismic retrofit of unreinforced
masonry heritage buildings in New Zealand." Master of Architecture, M. Arch Thesis,
Department of Architecture and Planning, The University of Auckland, New Zealand.
Goodwin, C. O., Tonks, G. and Ingham, J. M. (2009). "Identifying heritage value in URM
buildings." Journal of the Structural Engineering Society New Zealand, 22(2), 16‐28.
Grapes, R. and Downes, G. (1997). "The 1855 Wairarapa, New Zealand, earthquake: Analysis of
historical data." Bulletin of the New Zealand National Society for Earthquake Engineering,
30(4), 271‐368.
Haarhoff, E. (2003). Architecture of Central Auckland, Balasoglou Books.
Hodgson, T. (1992). The Heart of Colonial Auckland 1865‐1910, Random Century.
Hopkins, D. C. (2009). "Earthquakes and Existing Buildings: New Zealand Experience 1968 to
2008" ATC & SEI Conference on Improving the Seismic Performance of Existing Buildings and
Other Structures, San Francisco, California, 9 ‐ 11 Dec.
Hopkins, D. C. (2002). "Report on cost benefit of improving the performance of buildings in
earthquake." Department of Internal Affairs, Wellington.
Hopkins, D. C., Stannard, M., Lawrance, G. and Brewer, I. (2008). "Strengthening buildings for
earthquake implementing New Zealand legislation" 14th World Conference on Earthquake
Engineering, Beijing, China, 12 ‐ 17 Oct.
C1‐19
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
Hopkins, D. C. and Stuart, G. (2003). "Strengthening existing New Zealand buildings for
earthquake, An analysis of cost benefit using annual Probabilities" 2003 Pacific Conference
on Earthquake Engineering, Christchurch, New Zealand, 13 ‐ 15 February.
Ingham, J. M. and Griffith, M. (2011). "Performance of unreinforced masonry buildings during
the 2010 Darfield (Christchurch, NZ) earthquake." Australian Journal of Structural
Engineering, 11(3), February 2011.
Jacobs, W. (1985). The Birth of New Zealand, A Nation’s Heritage, Kowhai Publishing Ltd.
Johnson, J. A. R. (1963). "Earthquake engineering in New Zealand." New Zealand Engineering,
18(9), 305.
King, P. (2003). The Penguin History of New Zealand, Auckland.
Lourenço, P. B. (2006). "Recommendations for restoration of ancient buildings and the survival
of a masonry chimney." Construction and Building Materials, 20(4), 239‐251.
Magenes, G. (2006). "Masonry building design in seismic areas: recent experiences and
prospects from a european standpoint." 1st ECEES, Geneva, Switzerland, 3 ‐ 8 September.
McClean, R. (2009). Toward improved national and local action on earthquake‐prone heritage
buildings, New Zealand Historic Places Trust Pouhere Taonga, Wellington, New Zealand.
Megget, L. (2006). "From Brittle to Ductile: 75 Years of Seismic Design in New Zealand" New
Zealand Society for Earthquake Engineering Conference, Napier, New Zealand, 10 ‐ 12
March.
New Zealand Parliament (2004). Building Act 2004, Department of Building and Housing ‐ Te Tari
Kaupapa Whare, Ministry of Economic Development, New Zealand Government,
Wellington, New Zealand, Date of assent: 24 August 2004.
New Zealand Parliament (1968). Municipal Corporations Act, incorporating Amendment 301A,
The Department of Internal Affairs ‐ Te Tari Taiwhenua, New Zealand Government,
Wellington, New Zealand, Date of assent: 1968.
New Zealand Standards Institute (1935). NZSS No. 95:1935, Model Building By‐Law, New Zealand
Standrds Institute, Wellington, New Zealand.
New Zealand Standards Institute (1965a). NZS 1900:1965, Model Building Bylaw, New Zealand
Standards Institute, Wellington, New Zealand.
New Zealand Standards Institute (1965b). NZSS 1900:1965, Model Building By‐Law. Chapter 8:
Basic Design Loads, New Zealand Standrds Institute, Wellington, New Zealand.
NZNSEE (1972). "Classification of high earthquake risk buildings." Bulletin of the New Zealand
National Society for Earthquake Engineering, 5(2).
NZNSEE (1973). Recommendations for the classification of high earthquake risk buildings, New
Zealand National Society for Earthquake Engineering, Wellington, New Zealand.
NZNSEE (1985). Recommendations and guidelines for classifying, interim securing and
strengthening earthquake risk buildings, [1985 Red Book], New Zealand National Society for
Earthquake Engineering.
NZNSEE (1995). Draft guidelines for assessing and strengthening earthquake risk buildings, [1995
Red Book], New Zealand National Society for Earthquake Engineering.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
C1‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Oliver, J. (2006). John Oliver's Brick Book, Lifetime Books, Auckland.
PapersPast (2010). "Official report on public buildings." Star, Issue 6333, 3 September,
<http://paperspast.natlib.govt.nz/cgi‐bin/paperspast?a=d&d=TS18880903.2.25&cl=CL2.188
8.09.03>. (20th October, 2010).
Russell, A. P. and Ingham, J. M. (2010). "Prevalence of New Zealand's unreinforced masonry
buildings." Bulletin of the New Zealand Society for Earthquake Engineering, 43(3), 192‐207.
Russell, A. P. and Ingham, J. M. (2011). "Characterisation of New Zealand’s unreinforced
masonry building stock." Earthquake Spectra, xx(xx), xxxx, xx‐yy.
Shaw, P. (2003). A History of New Zealand Architecture, Hodder Moa Beckett Publishers Ltd.
Stacpoole, J. and Beaven, P. (1972). New Zealand Art; Architecture 1820‐1970, A. H. & A. W.
Reed, Wellington, New Zealand.
Standards Association of New Zealand (1976). NZS 4203: Code of Practice for General Structural
Design and Design Loadings for Buildings, Standards Association of New Zealand,
Wellington, New Zealand.
Stevens, C. M. and Wheeler, K. E. (2008). "Implementing earthquake prone building policy under
the Building Act 2004 ‐ Wellington City's Approach" New Zealand Society for Earthquake
Engineering Conference,Wairakei, New Zealand, 11 ‐ 13 April.
Stirling, M., Gerstenberger, M., Litchfield, N., McVerry, G. H., Smith, W., Pettinga, J. and Barnes,
P. (2008). "Seismic hazard of the Canterbury region, New Zealand: New earthquake source
model and methodology." Bulletin of the New Zealand National Society for Earthquake
Engineering, 41(2), 51‐67.
Thornton, A. W. (2010). "Twenty‐five years of strengthening Wellington." New Zealand Society
for Earthquake Engineering Conference, Te Papa, Wellington, New Zealand, 26 ‐ 28 March.
Wilson, J. (1984). Lost Christchurch, Te Waihora Press, Springston, New Zealand.
C1‐21
Section C1 ‐ History and Prevalence of Unreinforced Masonry Buildings in New Zealand
C1‐22
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C2 ‐ Material Properties of Masonry Walls
C2.1 Notation
c Bed‐joint cohesion, MPa
Em Masonry modulus of Elasticity, MPa
f’b Average brick compressive strength, MPa
f’bl Lower characteristic brick compressive strength, MPa
f’fb Flexural bond strength, MPa
f’j Average mortar compressive strength, MPa
f’jl Lower characteristic mortar compressive strength, MPa
f’m Average masonry compressive strength, MPa
MoR Brick modulus of rupture strength, MPa
mb Mass of brick units, kg
m j Mass of mortar, kg
N Axial compression stress, MPa
νb Volume of brick units, m3
νj Volume of mortar, m3
µf Bed‐joint coefficient of friction
ρm Masonry density, kg/m3
τ Bed‐joint shear stress, MPa
C2.3 In‐situ and Extracted Sample Testing
C2.3.1 Introduction
Whilst many aspects associated with the seismic response of URM buildings are significantly
influenced by characteristics such as boundary conditions and the behaviour of connections, the
accurate determination of material properties will clearly lead to improved seismic assessments,
less conservative retrofit designs and superior numerical models describing the seismic
behaviour of URM buildings. However, the non‐homogenous nature of masonry, combined with
the age of URM buildings, make it difficult to accurately predict the material properties of
masonry walls.
It is recommended that, whenever a project permits, field sampling or field testing is conducted.
Field sampling refers to the extraction of samples from an existing building and testing them in a
laboratory, while field testing refers to testing for material properties in‐situ. A set of techniques
are described below that can be used to determine masonry material properties.
Individual brick units and mortar samples shall be tested as per C2.3.2 and C2.3.3 when sampling
of large building sections is not permitted. Masonry properties shall then be predicted using the
obtained brick and mortar properties as per Section 2.5. When field sampling or testing is not
possible, the recommendations detailed in Section 2.6 shall be used to predict the masonry
constituent material properties.
C2‐1
Section C2 ‐ Material Properties of Masonry Walls
C2.3.2 Brick Properties
Individual brick units shall be extracted for the half brick compression test ASTM C 67‐03a (ASTM
2003c). The extracted brick units shall be cut into halves and capped using gypsum plaster prior
to compression testing. Figure C2‐1 shows brick sample preparation and the brick compression
testing.
(a) Brick cutting process (b) Gypsum plaster capping (c) Half brick comp. test
Figure C2‐1: Steps to determine brick properties
C2.3.3 Mortar Properties
Irregular mortar samples shall be extracted for laboratory testing. The method to determine
mortar compressive strength is detailed in ASTM C 109‐08 (ASTM 2008). This method involves
testing of 50 mm cube mortar samples, which generally are not attainable in existing buildings as
most mortar joints are only 12 to 18 mm thick. Consequently, irregular mortar samples shall be
cut into approximately cubical sizes with two parallel sides (top and bottom). The height of the
mortar samples shall exceed 15 mm in order to satisfactorily maintain the proportion between
sample size and the maximum aggregate size. The prepared samples shall be capped using
gypsum plaster to ensure a uniform stress distribution and tested in compression (Valek and
Veiga 2005).
The height to least lateral dimension (h/t) ratio of the mortar samples shall be measured and
used to determine the mortar compressive strength correction factors as in Table C2‐1. The
compression test result shall be divided by the corresponding correction factors listed in Table
C2‐1, and the average corrected strength is equal to the average mortar compressive strength,
f’j. Figure C2‐2 shows examples of irregular mortar samples and the mortar compression test.
See Section C2.5.3 for further details regarding the correction factors and the influence of height
to thickness ratio on mortar compressive strength. When conducting tests on laboratory
manufactured samples, 50 mm mortar cubes shall be made, left to cure under room
temperature (±20 °C) for 28 days and tested in compression following the mortar cube
compression test ASTM C 109‐08 (ASTM 2008).
Table C2‐1: Height to thickness ratio correction factor for mortar compressive strength
C2‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(a) Irregular mortar samples (b) Irregular mortar compression
test
Figure C2‐2: Determination of mortar compression strength
C2.3.4 Masonry Properties
C2.3.4.1 Masonry Compressive Strength and Stiffness
Masonry assemblage (prism) samples shall be extracted for laboratory testing. This extraction
must be conducted using a masonry chainsaw, operated by experienced personnel or by a hired
cutting contractor. The prism samples shall be single leaf and at least three bricks high. Samples
that are two leaf thick or more are to be cut into single leaf masonry prism samples. Rendering
plaster shall be removed from both sides of the sample, if present. The prepared samples shall
be capped using gypsum plaster to ensure uniform stress distribution and tested in compression
following ASTM C 1314‐03b (ASTM 2003a).
Laboratory calibrated displacement measurement devices may be attached on the masonry
prisms during the compression tests. At least two measurement devices shall be incorporated to
record displacements at opposing faces, and their gauge lengths shall cover the distance from
the middle of the top brick to the middle of the bottom brick, which furthermore shall be used
to derive the masonry stress‐strain relationship and Modulus of Elasticity, Em. The stress and
strain values considered in the calculation of Em shall be those between 0.05 and 0.70 times the
masonry compressive strength (f’m). Figure C2‐3 shows examples of sample extraction,
arrangement of displacement gauges and the prism compression test.
(a) Sample extraction (b) Prism with displacement (c) Prism compression test
gauges
Figure C2‐3: Preparation for prism compression test
C2‐3
Section C2 ‐ Material Properties of Masonry Walls
C2.3.4.2 Masonry Flexural Bond Strength
Two brick high masonry prisms shall be extracted and subjected to the flexural bond test
AS 3700‐2001 (Australian Standards 2001). The removal of rendering plaster from the sides of
the sample is necessary prior to performing the flexural bond test. Samples that are two leafs
thick or more shall be cut into single leaf masonry prism samples. Alternatively, the flexural bond
test may be conducted in‐situ if deemed to be more practical (refer to Figure C2‐4 (b)). The
flexural bond test is shown in Figure C2‐4.
(a) Laboratory flexural bond test (b) In‐situ flexural bond
test
Figure C2‐4: Masonry flexural bond test
C2.3.4.3 Masonry Bed‐joint Shear Strength
The in‐situ bed‐joint shear test ASTM C 1531‐03 (ASTM 2003b) shall be conducted to determine
the masonry bed‐joint properties. Alternatively, three brick high masonry prisms shall be
extracted for laboratory testing following the triplet shear test BS EN 1052‐3 (BSI 2002). The
triplet shear test shall be conducted whilst applying axial compression loads of approximately
0.2 MPa, 0.4 MPa and 0.6 MPa. At least three masonry triplet samples shall be tested at each
level of axial compression. Rendering plaster shall be removed from both sides of the sample, if
present. Masonry samples that are two leafs thick or more shall be cut into single leaf masonry
prism samples. Bed‐joint shear tests in‐situ and in the laboratory are shown in Figure C2‐5.
(a) Laboratory shear triplet (b) In‐situ shear test without (c) In‐situ shear test with flat
test flat jacks jacks
Figure C2‐5: In‐situ and laboratory bed‐joint shear test
C2‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C2.4 Brick and Mortar Field Assessment Procedure
The scratch test to determine brick and mortar properties shall be conducted as follows:
(1) Prepare the appropriate scratch test tools.
(2) Use 400 grit sand paper to clean and level the surface of the brick unit or mortar sample.
For brick units in particular, ensure that the degraded exterior of the sample is suitably
removed and the interior of the brick is exposed. Refer to Figure C2‐6 for an example of
exposed and unexposed brick surfaces.
(3) Begin by using a scratch tool having the lowest scratch number and apply just enough
pressure to ensure a firm contact on the brick or mortar surface at a 70 degree angle from
the test material surface.
(4) With uniform pressure, draw the scratch test tool along the test material surface for at least
20 mm.
(5) If no scratch is formed use the next scratch tool specified in Table 2‐1 and Table 2‐3.
(6) Repeat step (5) until a scratch is formed.
(7) Repeat step (1) to (7) at least six times at different locations within a wall section.
(8) Use Table 2‐1 or Table 2‐3 to determine the brick or mortar scratch index.
(9) The average scratch index of the material shall be equal to the average of the six scratch
tests.
(10) The average and lower characteristic brick compressive strength shall be obtained based on
the average scratch index using Table 2‐2.
(11) The mean and lower characteristic mortar compressive strength shall be obtained based on
the average scratch index using Table 2‐4.
(12) Use linear interpolation when required.
Note that for a scratch test on brick samples, the scratch test shall create visible indentation on
the brick surface.
Note that for a scratch test on mortar samples, the definition “Easily scratches using fingernail”
(scratch index of 1.5 in Table 2‐3) applies when the mortar can be scratched using minimal effort
and the particles become loose as the fingernail is drawn along the sample. A scratch index of 2
corresponds to the scenario when a fingernail leaves an indented mark whilst the mortar is not
heavily raked. A scratch index of 2 is commonly indicated by a worn fingernail although the
mortar surface is scratched. See Section C2.3.1 for more details on the scratch test.
Most Mortars have a scratch index below 2.5 (aluminium). Mortars that cannot be scratched
using aluminium are likely to be cement based mortars, and mechanical tests according to
Section C2.3.3 are strongly recommended when this type of mortar is present, because these
mortars were not included within the scope of the study reported here.
C2‐5
Section C2 ‐ Material Properties of Masonry Walls
Exposed Unexposed
interior interior
surface surface
Figure C2‐6: Difference between exposed and unexposed brick surface
C2.5 Additional Information
C2.5.1 Scratch Test Procedure
The material assessment technique detailed in Section 2.2 is a departure from the technique
detailed in the 2006 NZSEE guideline (NZSEE 2006). The reasons for the departure and the
background research conducted in order to support the introduction of the new technique are
discussed below.
C2.5.1.1 Recommendations Provided in the 2006 NZSEE Guideline
The NZSEE (2006) guideline provided the recommended masonry material strength parameters
for preliminary assessments as in Table C2‐2.
C2‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C2‐2: Parameters used in the NZSEE (2006) guideline
Material Visual Characteristic and hand tests
Mortar
Stiff High Portland cement content. Punch test < 10 mm
Firm Lime based. Punch test < 20 mm
Soft Lime binding, possibly mildly leached, can be raked out of joint, but
stays bound. Punch test < 30 mm
Non‐cohesive Lime‐based mortar that is heavily leached and weathered. Sand like,
easily raked out by hand, aggregate is unbound. Not suitable for
earthquake resistance
Bricks
Hard Dense, hard surface, well fired, dark reddish brown
Stiff Common brick, can be scored with a knife, red. Lower figures if split
Soft Weathered, pitted, distinct colour variation with depth, probably
under fired, bright orange
As detailed in Table C2‐2, the field assessment procedure for mortar characterisation measured
the penetration depth of a standard carpenter 3 mm diameter nail driven with a standard
carpenter hammer for 6 blows. This test was difficult to quantify as the nail penetration depth is
highly dependent on the force applied by the hammer blow.
The brick unit characterisation technique detailed in the NZSEE (2006) guideline was heavily
reliant on the brick physical attributes and visual appearance. Lumantarna (2011) concluded that
there is no direct relationship between brick unit colour and compressive strength. It was found
that brick colour is influenced by the presence of minerals such as iron oxides and calcium oxides
(Cultrone et al. 2005; Kreimeyer 1987), while brick compressive strength is highly dependent on
the firing temperature and level of porosity (Elert et al. 2003; Karaman et al. 2006; Tite and
Maniatis 1975). Therefore, the use of brick colour as a classification parameter is invalid.
Figure C2‐7 shows a plot of brick unit compressive strength against the Red Green Blue (RGB)
colour model, which commonly is used to represent a wide array of colours. Figure C2‐7 shows
that there is no relationship between each of the three different colour spectrums and the brick
unit compressive strength. In addition, Figure C2‐8 shows the relationship between brick unit
compressive strength and the visible brick colour. Although the visible brick colour is subjective,
Figure C2‐8 shows that there is no discernable relationship between brick visible colour and brick
compressive strength.
C2‐7
Section C2 ‐ Material Properties of Masonry Walls
350
300
250
Intensity
200
150
100
Red
50 Green
Blue
0
0 10 20 30 40 50
Compressive Strength (MPa)
Figure C2‐7: Colour spectrum intensity vs. compressive strength plot
50
45
Compressive Strength (MPa)
40
35
30
25
20
15
10
5
0
Figure C2‐8: Brick compressive strength vs. visible colour plot
C2.5.1.2 Developed Assessment Technique
Alternative assessment techniques were investigated due to the inadequacy of the
recommendations in the NZSEE (2006) guideline. For brick unit characterisation, non‐destructive
testing techniques including ultrasonic pulse velocity, Schmidt hammer and porosity tests were
studied (Lumantarna 2011). Although some of these techniques to a certain extent enabled the
characterisation of brick units, conducting these tests was not practical. The ultrasonic pulse
velocity test involved a set of tools such as a pundit, a transmitter and a receiver, while the
Schmidt hammer test required a Schmidt hammer, which was not widely available. Extraction of
brick units was necessary to perform an accurate ultrasonic pulse velocity test, as an in‐situ
ultrasonic pulse velocity test would encounter interference due to discontinuity in the brickwork
between the wall leafs. The porosity test ASTM C 20‐00 (ASTM 2000) also required brick unit
extraction. For the mortar, the characterisation technique detailed in the NZSEE (2006) guideline
was deemed to be insufficient.
C2‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Hardness has often been used to measure the mechanical strength of metals, rocks and minerals
(Shalabi et al. 2007; Tabor 1956), and therefore the scratch test based on the Mohs hardness
scale was nominated as an alternative assessment technique. Principally, the scratch test based
on the Mohs hardness scale is similar to the abrasion test described by Shalabi et al. (2007),
where the resistance of one material against another material of different hardness is tested.
Shalabi et al. (2007) also reported that there is a linear relationship between the compressive
strength of rocks and their abrasion hardness.
The Mohs hardness scale uses 10 minerals with different hardness values, where Mohs numbers
of 1 to 10 are used to rank the hardness of the different minerals (Tabor 1956). Although the
Mohs minerals in Table C2‐3 are not commonly available, there are materials (denoted as
‘Equivalent Materials’ in Table C2‐3) which have similar hardness values to the Mohs minerals
(MineralTown 2010; Nicholson and Shaw 2000). Due to their accessibility for use by
practitioners, these equivalent materials were investigated, and the scratch index assigned to
the tested sample was equal to the Moh’s number of the tool that scratched it.
Table C2‐3: Mohs hardness scale
1 Talc Baby powder
2 Gypsum Fingernail
3 Calcite Copper coin
4 Flourspar Iron nail
5 Apatite Glass
6 Feldspar Folding pocket knife
7 Quartz Kitchen steel knife
8 Topaz Sandpaper
9 Corundum Ruby
Individual brick units were collected from numerous historic buildings throughout New Zealand
and were cut into halves for laboratory testing. These half bricks were capped using gypsum
plaster, subjected to the Mohs scratch test and were subsequently tested in compression in
accordance with C2.3.2 to determine the direct relationships between the scratch test results
and the brick unit compressive strengths. It was found that the strongest brick units used in the
experimental programme were scratched using an iron nail. However, the increment from
fingernail (Mohs number 2) to copper coin (Mohs number 3) was found to be too wide, as most
brick units were scratched using a copper coin, whilst none were scratched using a fingernail. Shi
and Atkinson (1990) reported that aluminium has a lower hardness value than copper, and
C2‐9
Section C2 ‐ Material Properties of Masonry Walls
therefore it was decided that an aluminium pick (given a Mohs number of 2.5) could be used as
an intermediate material between fingernail and copper coin. Figure C2‐9 shows the brick
compressive strength‐scratch index relationship. The box and whisker plots (see Figure C2‐9)
were calculated according to Peck et al. (2009).
2.5: aluminium pick
Median = 14.4 MPa
50
Lower Quartile = 10.8 MPa
Compressive Strength (MPa)
Sample Size = 100
40
3.0: 10 cent copper coin
30
Median = 26.9 MPa
Lower Quartile = 21.7 MPa
20
Sample Size = 45
4.0: iron nail
10
Median = 35.6 MPa
Lower Quartile = 28.6 MPa
2.5 3 4
Sample Size = 30
Scratch Index
Figure C2‐9: Plot of brick compressive strength vs. scratch index
Mortar samples were extracted from a number of historic buildings and were cut into
approximately cubical shapes having two parallel sides (top and bottom). These mortar samples
were capped using gypsum plaster, subjected to the Mohs scratch test and were subsequently
tested in compression in accordance with C2.3.3. Although most mortars had equivalent
hardness values corresponding to being scratched by a fingernail, extremely weak mortars could
be easily scratched using a fingernail, where the mortar was scraped and the particles became
loose as the fingernail was drawn along the sample. When this scraping of mortar occurred, a
Mohs number of 1.5 was assigned. A Mohs number of 2 was assigned when a fingernail left an
indented scratch mark on the mortar joint while the mortar was not heavily raked. Figure C2‐10
shows box and whisker plots of the mortar compressive strength‐scratch index relationship.
C2‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
1.5: fingernail (easy)
Median = 1.4 MPa
12
10
Lower Quartile = 1.0 MPa
Compressive Strength (MPa)
Sample Size = 76
2.0: fingernail
8
Median = 5.5 MPa
6
Lower Quartile = 3.2 MPa
Sample Size = 54
4
3.0: 10 cent copper coin
2
Median = 7.4 MPa
0
Lower Quartile = 5.8 MPa
1.5 2 2.5
Sample Size = 43
Scratch Index
Figure C2‐10: Plot of Mortar compressive strength vs. scratch index
One‐way analysis of variance and Tukey multiple comparison tests revealed that there are
significant statistical differences between the means of the sample groups shown in Figure C2‐9
and Figure C2‐10 (Bewick et al. 2004; Chrichton 1999), which implies that for both brick unit and
mortar, each scratch index represents a different group of samples. This statistical analysis
emphasised that the scratch test based on the Mohs hardness scale is an adequate tool to
characterise brick unit and mortar joint properties. Refer to C2.4 for scratch test methodology.
C2.5.2 Brick Unit Modulus of Rupture and Modulus of Elasticity
Brick units that were extracted from different buildings were subjected to the Modulus of
Rupture test ASTM C 67‐03a (2003c). The remainder of the bricks from the Modulus of Rupture
test were then subjected to the half brick compression test ASTM C 67‐03a (2003c) in order to
obtain a direct relationship between the brick Modulus of Rupture, MoR, and compressive
strength, f’b. Figure C2‐11 shows that the brick unit MoR can be approximated to be equal to
0.13f’b.
C2‐11
Section C2 ‐ Material Properties of Masonry Walls
6
Modulus of Rupture (MPa)
5
4
y = 0.1296x
3 R² = 0.6249
1
Sample Size = 75
0
0 10 20 30 40 50 60
Compressive Strength (MPa)
Figure C2‐11 Plot of brick Modulus of Rupture vs. brick compressive strength
C2.5.3 Height to Least Lateral Dimension Ratio of Mortar
The characterisation of mortars in existing buildings has long been known to be difficult. Several
overseas studies (Moropoulou et al. 2005a; Moropoulou et al. 2000; Papayianni 2006) have
previously focused mainly on the chemical and microstructural analyses of mortars, which
enabled in‐depth characterisation of the mineralogical and chemical constituents of the mortar,
but provide insufficient information on mechanical properties. This focus on chemical and
microstructural analyses was due to the difficulty of extracting samples of a sufficiently large
volume from existing buildings (Drdacky et al. 2008; Magalhaes and Veiga 2009). Whilst the
number of samples allowed for extraction is usually restricted, most mortar joints in existing
URM buildings are between 12 mm and 18 mm thick, and therefore the size of the attainable
samples is small in comparison to the standardised size of 50mm x 50mm x 50mm. (ASTM 2008).
The small sample dimensions combined with restrictions on the number of samples allowed for
extraction makes mechanical tests unreliable as knowledge on testing techniques for non‐
standard mortar samples is currently insufficient. Investigations on non‐standard mortar testing
techniques have been conducted by several researchers (Drdacky et al. 2008; Magalhaes and
Veiga 2009; Valek and Veiga 2005). Valek and Veiga (2005) and Magalhaes and Veiga (2009)
performed compression tests on non‐standard mortar samples using two different methods. The
first method involved cutting of two mortar sides to produce flat and parallel loading surfaces
prior to compression testing, while the second method used strong confinement mortars to
transfer the compression load to the mortar sample.
Drdacky et al. (2008) conducted experiments on mortar samples with varying dimensions and
found that the mortar compressive strength, f’j, is primarily influenced by the sample
slenderness ratio and quality. An experimental programme involving three different mortar
mixes and ten different sample dimensions with varying slenderness ratio was conducted to
investigate the relationship between mortar compressive strength and dimensions. Five mortar
C2‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
samples were tested for each sample group. The materials used in the mortar mixes were
ordinary Portland cement, hydrated lime and river sand. The water/cement ratio during the time
of mixing was monitored to ensure consistency within each mortar mix. The mortar samples
were cast using timber moulds, left to cure under room temperature (±20 °C) for 28 days, and
were then capped using gypsum plaster and tested in compression. Figure C2‐12 shows the
relationship between average mortar compressive strength and slenderness ratio (height to
least lateral dimension ratio) of the mortar samples.
6
Mortar compressive strength (MPa)
4 Mortar 1:2:9
Mortar 1:3:12
3
Mortar 0:3:9
2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
h/t ratio
Figure C2‐12: Plot of mortar compressive strength vs. h/t ratio
Figure C2‐12 reveals that there was an overall increase in compressive strength with increasing
cement proportion. The mortar compressive strength dividing factors for samples having h/t
ratios other than 1.0 are shown in Table C2‐4, which shows that the dividing factors at the same
h/t ratio increments for the three different mortar grades were similar. Therefore, the mortar
correction factor for each h/t ratio increment (see Table C2‐4) was determined as the average
dividing factor of the three different mortar grades. The laboratory experiments have covered
different types of lime based mortars, and therefore it is assumed that the correction factors in
Table C2‐4 are valid for heritage mortars, which rarely have a compression strength exceeding
7.5 MPa (Lumantarna 2011).
C2‐13
Section C2 ‐ Material Properties of Masonry Walls
Table C2‐4: Mortar dividing factors based on h/t ratio
Mortar dividing factor
h/t ratio Average mortar
0:3:9 Mortar 1:3:12 Mortar 1:2:9 Mortar correction factor
1.0 1 1 1 1
C2.5.4 Masonry Compressive Strength, Modulus of Elasticity and Density
The importance of masonry assemblage properties such as compressive strength, stress‐strain
relationship and Modulus of Elasticity in the analyses of masonry structures has been
emphasised by many authors (Drysdale et al. 1999; Kaushik et al. 2007b). As the masonry
assemblage properties are related to the constituent material properties (Kaushik et al. 2007b),
an experimental programme to investigate the relationship between brick, mortar and masonry
compressive strengths was conducted. Laboratory manufactured samples with 18 different
brick‐mortar combinations and samples extracted from 9 historic buildings in New Zealand were
tested in compression. Displacement measurement devices were attached during the
compression tests to derive the masonry stress‐strain relationship and Modulus of Elasticity. The
density of masonry is an essential property for in‐plane and out‐of‐plane assessments of
masonry walls (Derakhshan 2011; Russell 2010), and therefore was investigated as reported in
Section C2.3.4.3.
C2.5.4.1 Masonry Compressive Strength
The predictive equation relating the brick, mortar and masonry compressive strengths was
developed based on both laboratory manufactured and field extracted masonry prisms. The
brick, mortar and masonry prism compressive strengths were determined following the
procedures detailed in Section C2.3.2, C2.3.3 and C2.3.4.1 respectively. The mortar properties of
some of the laboratory manufactured and field prism types were not determined, and therefore
only 14 different laboratory manufactured brick‐mortar combinations and 8 different field
sample types were included in the analysis. Nonlinear regression analysis was performed and a
three dimensional plot relating the average brick, average mortar and average masonry prism
compressive strengths is shown in Figure C2‐13. The surface plot in Figure C2‐13 represents the
prediction on the compressive strength of masonry with different brick and mortar properties.
The predictive expression relating the brick, mortar and masonry compressive strengths is
shown in Equation C2‐1.
C2‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C2‐13: 3‐D plot relating the brick, mortar and masonry compressive strengths
′ . .
0.7 (C2‐1)
C2.5.4.2 Masonry Modulus of Elasticity
The masonry Modulus of Elasticity was obtained following the procedure outlined in Section
C2.3.4.1. The masonry Modulus of Elasticity‐compressive strength relationships for both the
laboratory manufactured and the field extracted samples are presented in Figure C2‐14, which
shows that the Em/f’m relationship of the field samples was generally consistent with the results
from laboratory manufactured samples. The combination of data from laboratory manufactured
and field extracted samples suggests that the masonry Modulus of Elasticity, Em, can be
satisfactorily equated to 300f’m (see Figure C2‐14).
14000
12000
10000
y = 296.94x
R² = 0.732
Em (MPa)
8000
6000
4000
Field Samples
2000
Lab Samples
0
0 5 10 15 20 25 30 35 40
f'm (MPa)
Figure C2‐14: Plot of masonry Modulus of Elasticity vs. compressive strength plot
C2‐15
Section C2 ‐ Material Properties of Masonry Walls
C2.5.4.3 Masonry Density
The density of masonry was determined by creating a masonry module which consisted of a
single brick unit, two‐half mortar bed‐joints and two half mortar head joints (see Figure C2‐15).
Using the brick‐mortar volumetric proportion and a relationship shown in Equation C2‐2, the
density of masonry, ρm, was approximated.
URM
Module
Figure C2‐15: Unreinforced masonry module for density approximation
(C2‐2)
The mechanical strengths of brick units and mortar are related to the level of porosity (Elert et
al. 2003; Papayianni and Stefanidou 2006; Tite and Maniatis 1975), which furthermore is
inversely proportional to density. Therefore, it was thought that having an expression to relate
masonry density to the brick and mortar compressive strengths would be useful.
The brick and mortar density‐compressive strength relationships were investigated in the
laboratory. Brick units and mortar samples that were extracted from existing buildings were
subjected to the bulk density test ASTM C 20‐00 (2000b). These samples were then subjected to
the half brick compression test ASTM C 67‐03a (2003c) and irregular mortar compression tests
following the procedure detailed by Valek and Veiga (2005). The density‐compressive strength
relationships of the brick unit and mortar were incorporated into Equation C2‐2, and the
masonry density was furthermore expressed in terms of the brick and mortar compressive
strengths (see Equation C2‐3).
1578 5 8 (C2‐3)
C2.5.5 Masonry Flexural Bond Strength and Mortar Bed‐joint Shear Strength
The importance of masonry flexural bond strength and bed‐joint shear strength properties in the
out‐of‐plane and in‐plane analyses of unreinforced masonry walls has been emphasised by
Derakhshan (2011) and Russell (2010). Flexural bond tests (Australian Standards 2001), triplet
shear tests (BSI 2002) and prism compression tests (ASTM 2003a) were conducted on laboratory
manufactured samples with 6 different brick‐mortar combinations and on samples extracted
from 6 historic buildings, in order to investigate the relationships of masonry flexural bond
strength and mortar bed‐joint shear strength with other properties such as masonry and mortar
compressive strengths. The laboratory manufactured samples were left to cure under room
C2‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
temperature (±20 °C) for 28 days prior to testing. The brick unit and mortar compressive
strengths were determined following the procedures detailed in Section C2.3.2 and C2.3.3.
C2.5.5.1 Masonry Flexural Bond Strength
A number of researchers have investigated the relationship between masonry flexural bond
strength and masonry compressive strength (Sarangapani et al. 2005; Venkatarama Reddy and
Gupta 2006). However, these investigations were performed on masonry samples constructed
using soft brick units and stiff mortar, which are commonly used in India (Gumaste et al. 2006;
Kaushik et al. 2007a). When a soft brick‐stiff mortar masonry prism is subjected to axial
compression forces, the brick units have a greater tendency to expand laterally than does the
mortar, due to their low stiffness. Due to the presence of the brick‐mortar bond, the stiff mortar
joints prevent the lateral expansion of brick units, thus subjecting the brick units to bilateral
compression forces, whilst the mortar joints are subjected to bilateral tension forces as the soft
bricks try to expand the mortar joints laterally (Kaushik et al. 2007a; Sarangapani et al. 2005).
The masonry prism consequently fails after the brick‐mortar interface bond failure occurs. These
occurrences explain that, for soft brick‐stiff mortar combinations, the brick‐mortar interface
bond strength is one factor that often governs the masonry compressive strength, and therefore
the masonry compressive strength can be related to the masonry flexural bond strength
(Sarangapani et al. 2005).
The existence of a relationship between masonry flexural bond strength and masonry
compressive strength for strong brick‐weak mortar combinations, which are commonly
encountered in North America and New Zealand (Aryana and Matthys 2008; Lumantarna 2011),
is currently poorly documented. In order to investigate the existence of this relationship,
laboratory manufactured and field extracted masonry prisms were subjected to compression
and flexural bond tests according to Sections C2.3.4.1 and C2.3.4.2. The masonry flexural bond
strength‐masonry compressive strength relationships of the laboratory manufactured and field
samples are shown in Figure C2‐16.
0.6
Field Samples
Flexural Bond Strength (MPa)
0.5
Lab Samples
0.4
0.3
0.2
0.1
0
0 5 10 15 20 25
Masonry Compressive Strength (MPa)
Figure C2‐16: Plot of masonry flexural bond strength vs. masonry compressive strength
C2‐17
Section C2 ‐ Material Properties of Masonry Walls
Figure C2‐16 shows the absence of a distinct relationship between masonry flexural bond
strength and masonry compressive strength. It was furthermore observed that the flexural bond
failures of most of the field samples had occurred within the mortar joint instead of at the brick‐
mortar interface (see Figure C2‐17), and therefore the failures were governed by the flexural
strength of the mortar. Several authors (Cizer et al. 2008; Moropoulou et al. 2005b) have
documented that both flexural and compressive strengths of mortar increase over time.
Therefore, relating masonry flexural bond strength to mortar compressive strength, as in Figure
C2‐18, was deemed to be more appropriate than plotting masonry flexural bond strength
against masonry compressive strength.
(a) Mortar failure in flexure (b) Failure at the brick‐mortar interface
bond
Figure C2‐17: Different types of bond failure
0.8
Field Samples
0.7 y = 0.0786x
Lab Samples R² = 0.7742
Flexural Bond Strength (MPa)
0.6
0.5
0.4
0.3 y = 0.0281x
R² = 0.7428
0.2
0.1
0
0 2 4 6 8 10
Mortar Compressive Strength (MPa)
Figure C2‐18: Plot of masonry flexural bond strength vs. mortar compressive strength
Figure C2‐18 indicates that there is a relationship between masonry flexural bond strength and
mortar compressive strength, although the relationship for the laboratory manufactured
samples is different to that of the field extracted samples. This difference exists as the flexural
C2‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
bond failures of some laboratory manufactured samples occurred at the brick‐mortar interface,
and therefore those failures were governed by the quality of the brick‐mortar interface bond.
C2.5.5.2 Masonry Mortar Bed‐joint Shear Strength
It is not always possible to perform mechanical tests to determine the masonry bed‐joint shear
strength, and therefore an expression to predict this property based on other strength
parameters was sought. Sarangapani et al. (2005) tested masonry prisms in shear and
determined the relationship between masonry compressive strength and mortar bed‐joint shear
strength by recognising that the brick‐mortar interface bond strength often governed the failure
of masonry in compression (see Section C2.3.5.1). These shear tests were performed without
applying axial compression forces, and therefore the Mohr‐Coulomb friction envelope of the
mortar joints could not be derived. It was known that the shear failure of mortar joints may be
represented by the Mohr‐Coulomb friction law as shown in Equation C2‐4 (ASTM 2003b;
Lourenco et al. 2004):
μ (C2‐4)
Where: τ = Shear stress; c = Shear stress at zero axial compression (cohesion); µf = Coefficient of
friction; and N = Axial compression stress.
Laboratory manufactured and field masonry prisms were tested in shear according to Section
C2.3.4.3, where three different levels of axial compression forces were applied. Some laboratory
manufactured and field extracted prism types were not subjected to shear triplet tests due to
the limited availability of the samples, and therefore only 5 different laboratory manufactured
brick‐mortar combinations and 5 different field sample types were included in the analysis.
C2‐19
Section C2 ‐ Material Properties of Masonry Walls
Figure C2‐19 and Figure C2‐20 show plots of the mortar bed‐joint shear strength against axial
compression load of the laboratory manufactured and field extracted samples respectively.
1.8
1.6
1.4
Shear Stress (MPa)
1.2
1
0.8
0.6
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Axial Compression (MPa)
Figure C2‐19: Plot of shear stress vs. axial compression for laboratory manufactured samples
1.4
1.2
1
Shear Stress (MPa)
0.8
0.6
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Axial Compression (MPa)
Figure C2‐20: Plot of shear stress vs. axial compression for field samples
Instead of following the procedure recommended in BS EN 1052‐3 (2002), the axial compression
loads were maintained at 0.2 MPa, 0.4 MPa and 0.6 MPa to simplify the triplet shear tests. Shear
stresses at axial compression levels below 0.2 MPa for the field extracted samples (marked in
Figure C2‐20) were obtained in‐situ, and therefore the axial compression loads were estimated
using the overburden above the test locations. Figure C2‐19 and Figure C2‐20 reveal that the
shear strength variation at each level of axial compression was wide, which was also observed by
Lourenco et al. (2004).
C2‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
The mortar bed‐joint shear strengths at zero axial compression (Cohesion), c, were calculated as
the y‐intercepts of the shear stress‐axial compression relationships of the different sample types
(see Figure C2‐19 and Figure C2‐20). Following Sarangapani et al. (2005), the bed‐joint cohesion
was plotted against masonry compressive strength as in Figure C2‐21.
0.6
Field Samples
0.5
Lab Samples
0.4
Cohesion (MPa)
0.3
0.2
0.1
0
0 5 10 15 20 25
Masonry Compressive Strength (MPa)
Figure C2‐21: Plot of cohesion vs. masonry compressive strength
Figure C2‐21 reveals that there is no discernable relationship between bed‐joint cohesion and
masonry compressive strength. However, similar to observations during the flexural bond tests
(see Section C2.3.5.1), most bed‐joint shear failures of the field samples occurred within the
mortar, and therefore these failures were governed by the mortar properties instead of by the
quality of the brick‐mortar interface bond. Figure C2‐22 shows the mortar bed‐joint cohesion‐
mortar compressive strength relationships of the laboratory manufactured and field extracted
samples.
0.6
0.5
0.4
Cohesion (MPa)
Joint interface
0.3 failure
0.2
y = 0.0479x Field Samples
0.1 R² = 0.8553
Lab Samples
0
0 2 4 6 8 10
Mortar Compressive Strength (MPa)
Figure C2‐22: Plot of cohesion vs. mortar compressive strength
C2‐21
Section C2 ‐ Material Properties of Masonry Walls
The relationship between mortar bed‐joint shear strength and mortar compressive strength is
apparent, in particular for the field extracted samples (see Figure C2‐22). This relationship is
weak for the laboratory manufactured samples as the mortar bed‐joint shear failures of some
laboratory samples had occurred at the brick‐mortar interface, and therefore those failures were
governed by the brick‐mortar bond quality.
The three laboratory data points marked in Figure C2‐22 represent the samples in which mortar
bed‐joint shear failures had occurred at the brick‐mortar interface. Shear failure within the
mortar joints had occurred for the other two types of laboratory manufactured samples. Due to
differences in the failure modes and mortar compositions, it was decided that the laboratory
manufactured samples were not representative of the field samples. The mortar bed‐joint
cohesion, c, can be approximated as 0.045f’j when considering field samples only.
The mortar bed‐joint coefficient of friction, µf, was calculated as the gradient of the shear stress‐
axial compression relationships. Russell (2010) had previously suggested that µf shall be equal to
0.65. To validate this suggestion, the average µf of the field samples was considered, and the
average field value of µf = 0.88 was deemed to be reasonably close to the value suggested by
Russell (2010), especially when considering that there were only five different field sample types
available. The value of μf = 0.65 that is recommended in Equation 2‐6 is also consistent with
values presented in Table 10‐2 of NZSEE (2006), where in that document μf was specified to have
a range of 0.0‐0.8 depending on the condition of the mortar. Therefore, it was decided that the
previously suggested value of µf = 0.65 shall be used.
C2.7 Masonry Stress Block Parameters
Table C2‐5 provides a list of αf'm values from recent publications that address either clay brick
masonry or methods for improving the earthquake resistance of unreinforced masonry. In this
guideline αf'm has been taken as 0.85f'm consistent with both NZS 4230:2004 and most other
relevant recommendations.
C2‐22
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C2‐5: List of αf'm values from publications
Designs of Structures for Earthquake Resistance Part 3:
0.86f'm CEN (2005)
Assessment and Retrofitting of Buildings, Eurocode 8
Design Guideline for the Strengthening of Unreinforced
Galati et al.
0.855f'm Masonry Structures Using Fibre Reinforced Polymers (FRP)
(2005)
Systems
Masonry: Design on the Basis of Semi‐Probabilistic Safety Din 1053‐100
0.85f'm
Concept, Din 1053‐100 (2004)
Prestandard and Commentary for the Seismic Rehabilitation of
0.7f'm BSCC (2000)
Buildings.” Building Society Safety Council
C2.8 Worked Examples
Worked examples are briefly presented here to demonstrate the correct use of the equations
presented in Section 2 and the indicative values for masonry material properties when using
these equations.
C2.8.1 Example 1
Site: 1910 Campbell Free Kindergarten, Auckland (see Figure C2‐23)
Scenario: No mechanical test performed and sample extraction was not permitted.
Figure C2‐23: Campbell Free Kindergarten
Step 1: Perform scratch tests according to Section C2.4 to assess URM material properties:
C2‐23
Section C2 ‐ Material Properties of Masonry Walls
Brick
Scratch index (6 locations) = 2.5, 2.5, 2.5, 2.5, 3.0, 3.0
Average = 2.67
Linear interpolate from Table 2‐2, f’b = 18.17 MPa
Mortar
Scratch index (6 locations) = 2.0, 2.0, 2.0, 2.0, 2.5, 2.5
Average = 2.17
Linear interpolate from Table 2‐4, f’j = 6.13 MPa
Step 2: Predict brick unit Modulus of Rupture strength:
From Equation 2‐1, MoR = 0.130 x 18.17
= 2.36 MPa
Step 3: Predict masonry compressive strength and Modulus of Elasticity:
From Equation 2‐2, f’m = 0.7 x 18.17 0.75 x 6.13 0.3
= 10.61 MPa
From Equation 2‐4, Em = 300 x 10.61
= 3183 MPa
Step 4: Predict masonry density:
From Equation 2‐3, ρm = 1578 + 5 x 18.17 + 8 x 6.13
= 1718 kg/m3
Step 5: Predict masonry flexural bond strength and bed‐joint cohesion:
From Equation 2‐5, f’fb = 0.025 x 6.13
= 0.15 MPa
From Equation 2‐6, c = 0.045 x 6.13
= 0.276 MPa
From Equation 2‐7, µ = 0.65
C2.8.2 Example 2
Site: 1930’s Two‐storey Irish Pub, Auckland (see Figure C2‐24)
Scenario: Brick and mortar samples were extracted and tested in compression.
C2‐24
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C2‐24: Two‐storey Irish Pub
Step 1: Perform brick and mortar compression tests in accordance with Section C2.3.2 and
C2.3.3
Brick
Average of six half brick samples, f’b = 22.33 MPa
Mortar
Average of 6 mortar samples subjected to correction factors, f’j = 6.5 MPa
Step 2: Predict brick unit Modulus of Rupture strength:
From Equation 2‐1, MoR = 0.130 x 22.33
= 2.90 MPa
Step 3: Predict masonry compressive strength and Modulus of Elasticity:
From Equation 2‐2, f’m = 0.7 x 22.33 0.75 x 6.5 0.3
= 12.61 MPa
From Equation 2‐4, Em = 300 x 12.61
= 3783 MPa
Step 4: Predict masonry density:
From Equation 2‐3, ρm = 1578 + 5 x 22.33 + 8 x 6.5
= 1742 kg/m3
Step 5: Predict masonry flexural bond strength and bed‐joint cohesion:
From Equation 2‐5, f’fb = 0.025 x 6.5
= 0.163 MPa
From Equation 2‐6, c = 0.045 x 6.5
= 0.293 MPa
From Equation 2‐7, µ = 0.65
C2‐25
Section C2 ‐ Material Properties of Masonry Walls
C2.8.3 Example 3
Site: 1907 two‐storey commercial building, Gisborne (see Figure C2‐25)
Scenario: Sample extraction was not possible, but moderately destructive in‐situ tests were
permitted. In‐situ shear tests were performed in accordance with Section C2.3.4.2.
Figure C2‐25: Two‐storey commercial building
Step 1: Perform scratch tests according to Section C2.4 to assess URM material properties:
Brick
Scratch index (6 locations) = 2.5, 2.5, 2.5, 2.5, 2.5, 2.5
Average = 2.5
Linear interpolate from Table 2‐2, f’b = 14.0 MPa
Mortar
Scratch index (6 locations) = 2.0, 2.0, 2.0, 2.0, 2.5, 2.5
Average = 2.17
Linear interpolate from Table 2‐4, f’j = 6.13 MPa
Step 2: Predict brick unit Modulus of Rupture strength:
From Equation 2‐1, MoR = 0.13 x 14.0
= 1.82 MPa
Step 3: Predict masonry compressive strength and Modulus of Elasticity:
From Equation 2‐2, f’m = 0.7 x 14.0 0.75 x 6.13 0.3
= 8.73 MPa
From Equation 2‐4, Em = 300 x 8.73
= 2619 MPa
Step 4: Predict masonry density:
From Equation 2‐3, ρm = 1578 + 5 x 14.0 + 8 x 6.13
C2‐26
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
= 1697 kg/m3
Step 5: Predict masonry flexural bond strength and bed‐joint cohesion:
From Equation 2‐5, f’fb = 0.025 x 6.13
= 0.153 MPa
From Equation 2‐6, c = 0.045 x 6.13
= 0.276 MPa
From Equation 2‐7, µ = 0.65
C2.9 References
Aryana, S. A., and Matthys, J. H. (2008). "Statistical Analysis of Compressive Strength of Clay
Brick Masonry: Testing." The Masonry Society Journal, 26(2), 43‐52.
ASTM (2000a). "Standard test method for measurement of masonry flexural bond strength."
C 1072 ‐ 00a, ASTM International, Pennsylvania, United States.
ASTM (2000b). "Standard test methods for apparent porosity, water absorption, apparent
specific gravity, and bulk density of burned refractory brick and shapes by boiling water."
C 20 ‐ 00, ASTM International, Pennsylvania, United States.
ASTM (2003a). "Standard Test Method for Compressive Strength of Masonry Prisms." C 1314 ‐
03b, ASTM International, Pennsylvania, United States.
ASTM (2003b). "Standard test methods for in situ measurement of masonry mortar joint shear
strength index." C 1531 ‐ 03, ASTM International, Pennsylvania, United States.
ASTM (2003c). "Standard test methods for sampling and testing brick and structural clay tile."
C 67 ‐ 03a, ASTM International, Pennsylvania, United States.
ASTM (2008). "Standard Test Method for Compressive Strength of Hydraulic Cement Mortars
(Using 2‐in. or [50‐mm] Cube Specimens)." C 109/C 109M ‐ 08, ASTM International,
Pennsylvania, United States.
Australian Standards (2001). "Appendix D: Method of test for flexural strength." AS 3700 ‐ 2001.
Bewick, V., Cheek, L. and Ball, J. (2004). "Statistics review 9: One‐way analysis of variance."
Critical Care, 8(2), 130‐136.
BSI (2002). "Methods of test for masonry. Determination of initial shear strength." BS EN
1052 ‐ 3, British Standards Institution, London, England.
Chrichton, N. (1999). "Tukey Multiple Comparison test." Journal of Clinical Nursing, 8, 304.
Cizer, O., Van Balen, K., Van Gemert, D., and Elsen, J. (2008). "Blended lime‐cement mortars for
conservation purposes: Microstructure and strength development." Structural Analysis of
Historical Constructions: Preserving Safety and Significance, Bath, UK, 2 965‐972.
Cultrone, G., Sidraba, I., and Sebastian, E. (2005). "Mineralogical and physical characterization of
the bricks used in the construction of the "Triangul Bastion", Riga (Latvia)." Applied Clay
Science, 28(1‐4), 297‐308.
Derakhshan, H. (2011). "Seismic assessment of out‐of‐plane loaded unreinforced masonry
walls." PhD thesis, Department of Civil and Environmental Engineering, The University of
Auckland, Auckland, New Zealand.
C2‐27
Section C2 ‐ Material Properties of Masonry Walls
Karaman, S., Gunal, H., and Ersahin, S. (2006). "Assesment of clay bricks compressive strength
using quantitative values of colour components." Construction and Building Materials,
20(5), 348‐354.
Kaushik, H. B., Rai, D. C., and Jain, S. K. (2007a). "Stress‐strain characteristics of clay brick
masonry under uniaxial compression." Journal of Materials in Civil Engineering, 19(9),
728‐738.
Kaushik, H. B., Rai, D. C., and Jain, S. K. (2007b). "Uniaxial compressive stress‐strain model for
clay brick masonry." Current Science, 92(4), 497‐501.
Kreimeyer, R. (1987). "Some notes on the firing colour of clay bricks." Applied Clay Science, 2(2),
175‐183.
Lourenco, P. B., Barros, J. O., and Oliveira, J. T. (2004). "Shear testing of stack bonded masonry."
Construction and Building Materials, 18(2), 125‐132.
Lumantarna, R. (2011). "Material characterisation of New Zealand unreinforced masonry
buildings." PhD thesis, Department of Civil and Environmental Engineering, The University
of Auckland, Auckland, New Zealand.
Magalhaes, A., and Veiga, R. (2009). "Physical and mechanical characterisation of historic
mortars. Application to the evaluation of the state of conservation." Materiales de
Construccion, 59(295), 61‐77.
MineralTown (2010). "Minerals identification: Mohs Scale of mineral hardness." Rocks, minerals
and fossils, <http://www.mineraltown.com/infocoleccionar/mohs_scale_of_hardness.htm>
(September 27, 2010).
Moropoulou, A., Bakolas, A., and Bisbikou, K. (2000). "Investigation of the technology of historic
mortars." Journal of Cultura Heritage, 1(1), 45‐58.
Moropoulou, A., Bakolas, A., and Anagnostopoulou, S. (2005a). "Composite materials in ancient
structures." Cement & Concrete Composites, 27(2), 295‐300.
Moropoulou, A., Bakolas, A., Moundoulas, P., Aggelakopoulou, E., and Anagnostopoulou, S.
(2005b). "Strength development and lime reaction in mortars for repairing historical
masonries." Cement & Concrete Composites, 27(2), 289‐294.
Nicholson, P. T., and Shaw, I. (2000). Ancient Egyptian materials and technology, The Press
Syndicate of The University of Cambridge, Cambridge, United Kingdom.
C2‐28
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Papayianni, I. (2006). "The longevity of old mortars." Applied Physics A: Materials Science &
Processing, 83(4), 685‐688.
Papayianni, I., and Stefanidou, M. (2006). "Strength‐porosity relationships in lime‐pozzolan
mortars." Construction and Building Materials, 20(9), 700‐705.
Russell, A. (2010). "Characterisation and seismic assessment of unreinforced masonry buildings."
PhD thesis, Department of Civil and Environmental Engineering, The University of Auckland,
Auckland, New Zealand.
Sarangapani, G., Venkatarama Reddy, B. V., and Jagadish, K. S. (2005). "Brick‐mortar bond and
masonry compressive strength." Journal of Materials in Civil Engineering, 17(2), 229‐237.
Shalabi, F. I., Cording, E. J., and Al‐Hattamleh, O. H. (2007). "Estimation of rock engineering
properties using hardness tests." Engineering Geology, 90(3‐4), 138‐147.
Shi, H., and Atkinson, M. (1990). "A friction effect in low‐load hardness testing of copper and
aluminium." Journal of Materials Science, 25(4), 2111‐2114.
Tabor, D. (1956). "The physical meaning of indentation and scratch hardness." British Journal of
Applied Physics, 7(5), 159‐166.
Tite, M. S., and Maniatis, Y. (1975). "Scanning electron microscopy of fired calcareous clays."
Transaction of the British Ceramic Society, 74(1), 19‐22.
Valek, J., and Veiga, R. (2005). "Characterisation of mechanical properties of historic mortars ‐
testing of irregular samples" Structural studies, repairs and maintenance of heritage
architecture XI: Ninth international conference on structural studies, repairs and
maintenance of heritage architecture, Malta, 22‐24 June.
Venkatarama Reddy, B. V., and Gupta, A. (2006). "Tensile Bond Strength of Soil Cement Block
Masonry Couplets Using Cement‐Soil Mortars." ASCE Journal of Materials in Civil
Engineering, 18(1), 36‐45.
C2‐29
Section C2 ‐ Material Properties of Masonry Walls
C2‐30
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C3 ‐ Material Properties of Flexible Timber Floor
Diaphragms
C3.3 Timber properties
Timber diaphragms in New Zealand URM buildings were typically constructed with native timber
species such as Kauri, Rimu and Matai. It is recommended that wherever possible, the material
properties of timber be determined from testing samples extracted from the existing diaphragm.
It is recommended that at least twelve samples be taken. If physical testing of timber is not
achievable, then the published material properties may be used as a representative basis for
material property values.
The material properties of Rimu and Matai have been established from testing timber samples
salvaged from structures that have been demolished around New Zealand. Representative Kauri
samples could not be gathered for testing and subsequently the material properties of this
species were not determined. Further information regarding the material properties of native
New Zealand timbers can be found at http://www.nzwood.co.nz.
C3.3.1 Density
The density of Matai and Rimu were determined in accordance with the testing procedure
outlined in AS/NZS 1080.3:2000. As recommended by the standard, the test pieces were taken
from the specimens that were already prepared for bolted connection testing. Thus, the density
test pieces were prepared after conducting the main bolted connection tests. Samples were
prepared with average dimensions of 50.0 mm (length) 40.0 mm (width) 10.0 mm (thickness)
measured to an accuracy of 0.1 mm. Before placing the test pieces into the oven with 105C
for drying, weight of each test piece was recorded to an accuracy of 0.01 g. After the first 24
hours of the oven dry process, the weighing and measuring operations were conducted and
continued until a consistent weight of the test piece was achieved.
C3.3.2 Embedded strength
Rectangular test pieces with dimensions 100 mm (length) 50 mm (width) 90 mm (height)
were prepared for embedded strength testing. In accordance with ISO/DIS 10984‐2: 2008, a half
13 mm diameter hole was notched into each test piece for the placement of a 12 mm diameter
fastener during testing, as shown in Figure C3‐1. The test setup shown in Figure C3‐2 was
implemented to avoid the bending of the fastener during testing.
C3‐1
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
H (height)
half 13 mm
diameter hole
W (width)
L (length)
Figure C3‐1: Test piece for embedded strength tests
Figure C3‐2: Test setup for embedded strength tests
C3.3.3 Yield compression strength
The yield compression strength perpendicular to grain of Matai and Rimu was determined by
conducting a pure block test as shown in Figure C3‐3. Dimensions of the tested timber blocks
were 100 mm (length) 110 mm (width) 50 mm (height). The yield compression strength
perpendicular to the grain of wood is determined using the yield load divided by the compressed
wood area (AS/NZS 4063.1:2007). Note that the yield load is evaluated using a 2 mm offset
method.
C3‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C3‐3: Pure block test setup to determine compression strength of wood
C3.4 Classification
Timber floor diaphragms are routinely classified as flexible using the quantitative definition
published in ASCE/SEI 41‐06 (2007):
‘Diaphragms shall be classified as flexible when the maximum horizontal
deformation of the diaphragm along its length is more than twice the average
interstorey drift of the vertical lateral‐force resisting elements of the story
immediately below the diaphragm’.
Given that almost all New Zealand URM buildings comprise timber diaphragms, and that these
diaphragms consistently satisfy the ASCE/SEI 41‐06 definition of ‘flexible’, it is appropriate to
default to a flexible diaphragm classification, unless it is proven otherwise.
C3.5 Configuration and Condition Assessment
C3.5.1 Configuration
Timber diaphragms in New Zealand URM buildings typically comprise either straight‐edge or
tongue & groove floorboards nailed perpendicular to joists that span between vertical support
elements, such as masonry walls or intermediate columns. When perimeter walls were close
enough (approximately less than 6 m) joists would often span continuously between these
elements. For larger spans, joists were typically lapped or butted, either with or without some
form of mechanical connection, over intermediate steel or timber cross‐beams supported on
columns. Diaphragm blocking and chord elements were almost never provided, and timber
cross‐bracing between joists to prevent out‐of‐plane buckling may, or may not, have been fitted.
C3‐3
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Joist ends were typically either simply supported on a brick ledge resulting from the perimeter
walls reducing in width at each storey height, or pocketed into the wall to a depth equal to one
brick width. Figure C3‐4 shows graphics that illustrate these diaphragm configuration details.
Diaphragm properties determined in Sections 3.5 and 3.6 are based on the typical diaphragm
configurations outlined above, and remain representative with only minor variations to
configuration values.
(a) Schematic of typical diaphragm (b) Underside of typical diaphragm
(c) Continuous joists. Diaphragm span L < 6.0 (d) Lapped joist connection. Diaphragm span
m L > 6.0m
(e) Joists pocketed into URM wall (f) Joists seated on ledge of URM wall
Figure C3‐4: Diaphragm configuration details
C3‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C3.5.2 Condition Assessment
Table 3‐5 has been developed from collaborative engineering judgement and focuses on
diaphragm conditions that most affect the performance of the primary nail connections, which
have been proven to govern overall diaphragm performance. If the designer is not certain which
condition category the subject diaphragm falls within, it is recommended that a conservative
condition rating be selected.
C3.6 Characteristic Values
An integrated experimental and analytical research program undertaken at the University of
Auckland has quantitatively established key structural performance characteristics of timber
floor diaphragms. This research has generated the much needed test data on typical
configurations to improve the assessment of these structures. Table C3‐1 summarises the
cornerstone of the research program that involved pseudo‐statically testing four full‐scale
diaphragms to establish the realistic behaviour of representative diaphragms. The total force‐
midspan displacement results from these tests are presented for reference in Figure C3‐5. For
use in independent nonlinear time‐history analyses, the force‐displacement data of the
backbone curves of diaphragms FS1a (loaded parallel to joists) and FS3a (loaded perpendicular
to joists) have been provided in Table C3‐2. Equivalent negative values of the data points can be
used for the reverse direction.
It is acknowledged that the characteristic values determined below have been established from
diaphragms constructed with new pine timber and new wire‐drawn nails. The completed
experimental program, however, included the testing of nailed timber connections and small‐
scale diaphragm samples extracted from heritage URM buildings that have provided the
necessary data to assess the effects of age and decay on diaphragm performance. It is the
intention to use these results to appropriately adjust the presented diaphragm performance
values to account for the effects of heritage construction materials, and the effects of age and
decay.
C3.6.1 Stiffness, Strength and Ductility
Timber diaphragms are characterised by highly nonlinear hysteretic response to lateral loading,
with no clearly defined yield point. The establishment of stiffness, strength and ductility is
therefore difficult, as these values are sensitive to their definition and no universally accepted
method to determine these values currently exists. Given the need to provide simple, yet
accurate diaphragm performance properties, a bilinear representation of the force‐displacement
response of each full‐scale diaphragm tested was developed using the principle of hysteretic
energy conservation (Mahin and Bertero 1981). The idealised curves provided the necessary
characteristics to define recognisable performance properties such as stiffness, strength and
ductility as presented in Tables 3‐6 and 3‐7 (see Figure C3‐6 and Figure C3‐7).
A distinction has been made for the presence of joist connections in the loading direction
perpendicular to joists. The classification ‘without joist connections’ refers to diaphragms that
C3‐5
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
comprise continuously spanning joists between the perimeter URM walls, as discussed in section
C3.4.1. The classification ‘with joist connections’ refers to situations where the URM walls are
too far apart for joists to span continuously between them, so that the joists must span between
walls and intermediate cross‐beams. This classification applies to any type of connection
between the joists. Typically joists are lapped or butted over the cross‐beam either with or
without a mechanical connection.
Observations from full‐scale diaphragm testing indicated that in the direction parallel to joists,
the flexural bending of floorboards effectively engaged a shear response from the nail couples to
resist the applied lateral load. However, in the direction perpendicular to joists, the nail couples
were much less engaged and diaphragm response was governed by flexural bending of the
joists. Considering this observed response mechanism, the strength and stiffness values for
diaphragms with joist connections in the direction perpendicular to joists have been slightly
reduced to account for the reduced flexural resistance a non‐continuous joist has compared to a
continuous joist. Further analysis using a calibrated finite element diaphragm model is expected
to verify this adjustment.
Methodology for bilinear curve development
(a) A monotonic backbone (or envelope) curve was fitted to each hysteretic data set.
(b) The area beneath this backbone curve was calculated using the ‘trapezoidal rule’. This area
is analogous to the amount of hysteretic energy absorbed by the structure during
monotonic loading. With no established method for characterising diaphragm response, the
conservation of this energy is the only rational method available to develop a
representative bilinear curve.
(c) The proposed bilinear curve was given the following constraints to ensure that a unique
solution could be found when equating hysteretic energy:
(1) Must pass through zero load and zero displacement.
(2) Final displacement was set to ±150 mm with corresponding load taken from the
backbone curve.
(3) Secondary stiffness, K2, was taken as the average gradient of the lines joining
displacement amplitudes 75 mm, 100 mm and 150 mm. This portion of the backbone
curve is essentially linear, therefore justifying the constraint of its gradient for the
bilinear curve.
(4) Having appropriately constrained the solution, the yield displacement, Δy, was calculated
using Equation C3‐1.
1
2
2 (C3‐1)
1
2
(5) And the corresponding yield load, Fy, was calculated using Equation C3‐2.
C3‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(C3‐2)
(6) Finally, the bilinear curve was drawn
C3‐7
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Table C3‐1: Full‐scale diaphragm testing summary
Loading
Schematic Test details
direction
Test FS1a
10.4 m x 5.535 m
Aspect ratio 1.88 : 1
Homogeneous sheathing
Parallel to joists
Test FS2a
10.4 m x 5.535 m
Aspect ratio 1.88 : 1
3.2 m x 1.08 m corner
perforation
Test FS3a
5.535 x 10.4 m
Aspect ratio 1 : 1.88
Homogeneous sheathing
Perpendicular to joists
Continuous joists
Test FS4a
5.535 x 10.4 m
Aspect ratio 1 : 1.88
Homogeneous sheathing
Joists discontinuous with
central bolted lap
connection
C3‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
50 50
40 40
30 30
20 20
10 10
Load (kN)
Load (kN)
0 0
-10 -10
-20 -20
-30 -30
-40 -40
-50 -50
-200 -150 -100 -50 0 50 100 150 200 -200 -150 -100 -50 0 50 100 150 200
Mid-span displacement (mm) Mid-span displacement (mm)
(a) Diaphragm FS1a (b) Diaphragm FS2a
120 120
100 100
80 80
60 60
40 40
20 20
Load (kN)
Load (kN)
0 0
-20 -20
-40 -40
-60 -60
-80 -80
-100 -100
-120 -120
-200 -150 -100 -50 0 50 100 150 200 -200 -150 -100 -50 0 50 100 150 200
Mid-span displacement (mm) Mid-span displacement (mm)
(c) Diaphragm FS3a (d) Diaphragm FS4a
Figure C3‐5: Full‐scale diaphragm testing results
C3‐9
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Table C3‐2: Example backbone force‐displacement data for time‐history analysis
Diaphragm FS1a Diaphragm FS3a
Midspan Midspan
Load (kN) Load (kN)
displacement (mm) displacement (mm)
0 0 0 0
0.2 1.2 0.1 0.8
0.6 2.9 0.4 2.5
1.4 3.9 1.1 4.7
1.8 4.3 1.5 5.6
2.2 4.6 1.9 6.9
4.9 6.2 2.1 7.4
9.9 8.4 5.1 11.9
14.8 10.2 10.0 17.1
24.0 13.1 14.7 22.0
49.4 19.2 24.8 29.9
74.0 23.7 49.6 45.7
98.7 27.7 73.9 59.6
147.3 35.6 98.6 73.9
157.2 36.0 141.8 102.0
Load
Fmax
K2
Fy
K1
Δy Δmax
Displacement
Figure C3‐6: Idealised bilinear response curve
(d) Steps (a) to (c) were performed for the positive and negative displacement regions
separately, and averaged to produce an equivalent bilinear curve either side of zero
displacement.
The bilinear curves developed for each full‐scale diaphragm test are presented in Figure C3‐7.
C3‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Diaphragm strength, Qn
Diaphragm strength is conventionally presented as shear strength per lineal metre depth of the
diaphragm (kN/m). The shear strength values presented in Table 3‐6 were therefore calculated
by taking the relevant bilinear yield load, which is the total applied load to the diaphragm,
halving it to find shear resistance at the diaphragm’s edge, and dividing it by the depth of the
diaphragm, as illustrated in the following equation:
(C3‐3)
2
When calculating total diaphragm shear strength, it is recommended that the effective
diaphragm depth, be, determined from Table 3‐8 be used.
It is acknowledged that the shear strength values presented in Table 3‐6 are significantly lower
than those published in NZSEE (2006). It is not fully understood why this discrepancy exists but
one theory suggests that the original authors of the NZSEE strengthening guide lines
benchmarked diaphragm strength back to ABK (1981) test values. This may have lead to
unconservative values because referring to Section A110.4 of the IEBC (2006), the strength
values, which were also based on the ABK testing, already include an allowance for overstrength
of the wood framed elements and dynamic response of the rigid masonry walls (Oliver 2010).
Nonetheless, the shear strength values in Table 3‐6 have been established directly from full‐
scale diaphragm testing, as detailed above, and provide a new basis for expected diaphragm
strength in URM buildings.
Diaphragm overstrength has not yet been thoroughly considered and is a subject for further
analysis. A basic interpretation of the bi‐linear force‐displacement plots in Error! Reference
source not found. suggests that if ultimate diaphragm strength was taken at a midspan
displacement of 150 mm, then diaphragm overstrength factor would be as high as 4, which is
considerably higher than conventional values. Until additional analysis provides appropriate
guidance on diaphragm overstrength, it is recommended that conventional methods taking μ =
1.25 be adopted in the design of non‐yielding diaphragm components.
Diaphragm shear stiffness, Gd
It is necessary to present diaphragm stiffness in Table 3‐6 as shear stiffness, Gd, to achieve
independence from diaphragm geometry. Using Equation C3‐3, the value of Gd was back
calculated by setting diaphragm stiffness, Kd, to the initial stiffness, K1, of the relevant bilinear
curve.
4
(C3‐4)
C3‐11
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Ductility capacity, µ
The diaphragm ductility capacities presented in Table 3‐7 were calculated as the ratio of plastic
displacement to yield displacement assuming elastic‐perfectly plastic behaviour, as presented in
Equation C3‐5 below:
μ (C3‐5)
It can be observed in Figure C3‐5 that failure (strength degradation) did not occur in any of the
tested diaphragms, even at midspan displacements in excess of 150 mm. Representative values
of Δmax were therefore not captured during testing, and any extrapolation of the force‐
displacement response plots would have been speculative. In order to gauge some level of
diaphragm ductility capacity, Equation C3‐5 was applied by taking the maximum recorded
displacement value of each test as Δmax and the corresponding Δy value determined from bilinear
representation. Ductility capacities using these conservative definitions were calculated to be
between 6.8 and 10.8. Given that AS/NZS 1170 (2002a) stipulates a maximum ductility capacity
of μ = 6, and that all tested diaphragms demonstrated the ability to exceed this value, it has
been recommended that diaphragm ductility capacity be defaulted to μ = 6. It is believed that
this value appropriately reflects the flexible and ductile nature of these diaphragms.
40 40
30 30
20 20
10 10
Load (kN)
Load (kN)
0 0
-10 -10
-20 -20
-30 -30
-40 -40
-150 -100 -50 0 50 100 150 -150 -100 -50 0 50 100 150
Mid-span displacement (mm) Mid-span displacement (mm)
(a) Diaphragm FS1a (b) Diaphragm FS2a
C3‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
125 125
100 100
75 75
50 50
25 25
Load (kN)
Load (kN)
0 0
-25 -25
-50 -50
-75 -75
-100 -100
-125 -125
-150 -100 -50 0 50 100 150 -150 -100 -50 0 50 100 150
Mid-span displacement (mm) Mid-span displacement (mm)
(c) Diaphragm FS3a (d) Diaphragm FS4a
Figure C3‐7: Bilinear curves from full‐scale diaphragm testing
Diaphragm condition effects
Provisions for diaphragm condition in Tables 3‐6 and 3‐7 are provided to reflect the adverse
effect that deteriorating diaphragm condition has on performance. The effect of diaphragm
condition on strength, stiffness and ductility will be identified through parametric studies of
varying nail connection conditions using a calibrated finite element model. Until this is
completed, the characteristic values for diaphragms classified as ‘Fair’ and ‘Poor’ will be 85% and
65% of ‘Good’ diaphragms, respectively.
C3.6.2 Damping
Diaphragm damping has been determined using the approximate method presented in Chopra
(2007). In this method, the equivalent viscous damping ratio(eq) can be approximated by
determining the equivalent strain energy, ES, dissipated during each cycle of the force‐
displacement response, as described in Equation C3‐6, where do is the average of the maximum
displacements in the positive and negative directions of the cycle, and k is the secant stiffness to
the corresponding maximum force. The equivalent viscous damping ratio is then determined
according to Equation C3‐7, where ED is equal to the hysteretic energy dissipated in the same
cycle (see Section C3.5.1).
1
(C3‐6)
2
1
(C3‐7)
4
While preliminary analysis demonstrates that diaphragm damping could be in excess of 0.20
(20%) using the above method, it is important to recognise that this hysteretic energy is already
accounted for in the concept of ductility and so in the absence of exhaustive nonlinear time‐
C3‐13
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
history analyses and for the purpose of linear static assessment procedures, it is most
appropriate to default to a typical damping ratio of 0.05 (5%).
C3.7 Diaphragm Stiffness
Diaphragm stiffness is the mathematical gradient of the line joining zero displacement with the
yield point, as demonstrated in Equation C3‐8:
(C3‐8)
Equation 3.1 for diaphragm stiffness, Kd, derives from Equation 8‐3 in ASCE/SEI 41‐06 (2007) for
yield displacement:
(C3‐9)
2
Where:
2
The inclusion of effective diaphragm dimensions be and Le, and coefficients αp and αc was
intended to enhance the application of the equation to a wider range of diaphragm geometries
and configurations. The basis of these values is described hereafter.
C3.7.1 Diaphragm Geometry
Diaphragm performance is a function of the number nail couples available to resist lateral
loading. The introduction of effective diaphragm span (Le) and effective diaphragm depth (be) is a
simple concept to account for the number of available nail couples while maintaining user
friendly assessment procedures.
C3.7.2 Diaphragm Penetrations
Penetrations in timber diaphragms, such as stairwells and elevator shafts, cause a reduction in
diaphragm stiffness due to reduced nail couples available to resist lateral loads. The proposed
reduction in diaphragm stiffness by the ratio of net diaphragm area to gross diaphragm area
remains under review, and will be validated through parametric studies using a calibrated finite
element model.
C3‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C3.7.3 Diaphragm Configuration
The equations to determine the effects of diaphragm configuration will be determined through
parametric studies using a calibrated finite element model. In the interim, αc shall be taken as
1.0.
C3.8 Worked Example 1
10.4 m
N
5.535 m
Step 1: Configuration Details
Dimensions See illustration above
Floorboard type Tongue & groove
Floorboard size 135 mm x 21 mm
Joist size 51 mm x 221 mm
Joist spacing 515 mm crs
Joist direction North‐south
Joist connection No ‐ joists continuous
Joist bracing 50 mm x 70 mm, X‐pattern
Joist bracing location 1/3 along length of joists
Nail schedule 2 nails per floorboard‐joist connection
Nail spacing 100 mm
Step 2: Select Condition Rating
Condition Good
Step 3: Select Characteristic Values
North South Direction
Given the orientation of the joists, loading in the north south direction would be ‘parallel to
joists’.
Shear stiffness and shear strength from Table 3‐6:
Gd = 300 kN/m
C3‐15
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Qn = 1.5 kN/m
Ductility capacity from Table 3‐7:
μ = 6
Damping for all diaphragms:
= 5%
East West Direction
Similarly it can be shown that for loading direction perpendicular to joists:
Gd = 175 kN/m
Qn = 1.2 kN/m
μ = 6
= 5%
Step 4: Diaphragm Stiffness Calculation
North South Direction
4
With reference to Table 3‐8:
L e = L = 10.4 m
Using Equation 3‐2:
αp = 1.0 (no penetrations)
And using Equation 3‐3:
αc = 1.0 (until reduction factors for diaphragm configuration variations are
determined)
Hence:
1.0 ∗ 1.0 ∗ 4 ∗ 5.535 ∗ 300
10.4
= 639 kN/m
East West Direction
Similarly it can be shown that:
L e = L = 5.535 m
αp = 1.0
C3‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
αc = 1.0
1.0 ∗ 1.0 ∗ 4 ∗ 10.4 ∗ 175
5.535
= 1315 kN/m
C3.9 Worked Example 2
N
b1 = 9.1 m
b2 = 4.9 m
L = 12.2 m
Step 1: Configuration Details
Dimensions See illustration above
Floorboard type Tongue & groove
Floorboard size 135 mm x 25 mm
Joist size 55 mm x 295 mm
Joist spacing 500 mm crs
Joist direction North‐south
Joist connection Lapped with no mechanical connection
Joist bracing 50 mm x 70 mm, X‐pattern
Joist bracing location 1/3 along length of joists
Nail schedule 2 nails per floorboard‐joist connection
Nail spacing 95 mm
Step 2: Select Condition Rating
Condition Fair
Step 3: Select Characteristic Values
North South Direction
Given the orientation of the joists, loading in the north south direction would be ‘parallel to
joists’.
C3‐17
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Shear stiffness and shear strength from Table 3‐6:
Gd = 255 kN/m
Qn = 1.3 kN/m
Ductility capacity from Table 3‐7:
μ = 5
Damping for all diaphragms:
= 5%
East West Direction
Similarly it can be shown that for loading direction perpendicular to joists:
Gd = 125 kN/m
Qn = 0.8 kN/m
μ = 5
= 5%
Step 4: Diaphragm Stiffness Calculation
North South Direction
4
With reference to Table 3‐8:
L e = L = 12.2 m
. .
be = = 7.0 m
Using Equation 3‐2:
αp = 1.0 (no penetrations)
And using Equation 3‐3:
αc = 1.0 (until reduction factors for diaphragm configuration variations are
determined)
Hence:
1.0 ∗ 1.0 ∗ 4 ∗ 7.0 ∗ 255
12.2
= 585 kN/m
East West Direction
Similarly it can be shown that:
C3‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
. .
L e = = 7.0 m
αp = 1.0
αc = 1.0
1.0 ∗ 1.0 ∗ 4 ∗ 12.2 ∗ 125
7.0
= 871 kN/m
C3.10 Worked Example 3
L1 = 7.2 m L2 = 7.2 m
b2 = 6.2m
N
L1 = 12.4 m
b1 = 6.2m
L1 = 14.4 m
Step 1: Configuration Details
Dimensions See illustration above
Floorboard type Straight edge
Floorboard size 150 mm x 25 mm
Joist size 50 mm x 350 mm
Joist spacing 400 mm crs
Joist direction East‐west
Joist connection Butted with bolted light steel plate
Joist bracing 50 mm x 70 mm, X‐pattern
Joist bracing location Approximately 3.0 m crs along joist length
Nail schedule 2 nails per floorboard‐joist connection
Nail spacing 110 mm
Step 2: Select Condition Rating
Condition Fair
C3‐19
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
Step 3: Select Characteristic Values
North South Direction
Given the orientation of the joists, loading in the north south direction would be ‘perpendicular
to joists’.
Shear stiffness and shear strength from Table 3‐6:
Gd = 125 kN/m
Qn = 0.8 kN/m
Ductility capacity from Table 3‐7:
μ = 5
Damping for all diaphragms:
= 5%
East West Direction
Similarly it can be shown that for loading direction parallel to joists:
Gd = 255 kN/m
Qn = 1.3 kN/m
μ = 5
= 5%
Step 4: Diaphragm Stiffness Calculation
North South Direction
4
With reference to Table 3‐8:
L e = 7.2 + 7.2 = 14.4 m
Using Equation 3‐2:
14.4 ∗ 12.4 7.2 ∗ 6.2
0.75
14.4 ∗ 12.4
And using Equation 3‐3:
αc = 1.0 (until reduction factors for diaphragm configuration variations are
determined)
Hence:
0.75 ∗ 1.0 ∗ 4 ∗ 12.4 ∗ 125
14.4
C3‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
= 323 kN/m
East West Direction
Similarly it can be shown that:
L e = 6.2 + 6.2 = 12.4 m
αp = 0.75
αc = 1.0
0.75 ∗ 1.0 ∗ 4 ∗ 14.4 ∗ 255
12.4
= 889 kN/m
C3.11 References
ABK. (1981). "Methodology for mitigation of seismic hazards in existing unreinforced masonry
buildings: Diaphragm testing." ABK‐TR‐03, National Science Foundation, El Segundo,
California.
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06. American Society of
Civil Engineers, Reston, VA.
Australian/New Zealand Standard (2000). AS/NZS 1080.3:2000 Timber ‐ Methods of Test ‐
Method 3: Density, Standards New Zealand (electronic copy).
Australian/New Zealand Standard (2007). AS/NZS 4063.1:2007 Structural timber ‐ Characteristic
values of strength‐graded timber ‐ Part 1: Test methods. Standards Australia (draft only).
Chopra, A. K. (2007). Dynamics of Structures: Theory and Applications to Earthquake Engineerin.,
Prentice Hall, Upper Saddle River, NJ, 3rd edition.
ICC. (2006). International Existing Building Code. Country Club Hills, Illinois, USA.
International Organization for Standardization (2008). Draft International Standard
ISO/DIS 10984‐2 Timber structures ‐ Dowel‐type fasteners ‐ Part 2: Determination of
embedding strength and foundation values. International Organization for Standardization,
Geneva, Switzerland.
Mahin, S. A., and Bertero, V. V. (1981). "An evaluation of inelastic seismic design spectra."
Journal of the Structural Division, 107(ST9), 1777‐1795.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Oliver, S. (2010). "Personal communication." A. W. Wilson, Auckland, New Zealand.
Standards Association of New Zealand. (1993). NZS 3603:1993, Timber Structures. Standards
New Zealand, Wellington, New Zealand.
Standards Association of New Zealand. (2002a). AS/NZS 1170.1:2002, Structural Design Actions
Part 0: General Principles. Standards New Zealand, Wellington, New Zealand.
C3‐21
Section C3 ‐ Material Properties of Flexible Timber Floor Diaphragms
C3‐22
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C4 ‐ Design Actions on In‐plane Loaded
Unreinforced Masonry Walls
C4.1 Notation
C1 Modification factor to relate expected maximum inelastic displacement to that
calculated for linear elastic response
C3 Modification factor to represent increased displacements due to dynamic P‐delta
effects
C(Ti) The elastic site hazard spectrum for a period of Ti in the direction being considered
Cd(Tw) Horizontal design action coefficient
Kd Diaphragm stiffness, kN/m
n Number of diaphragms
Ti The period of the diaphragm at storey i, s
Ts The characteristic period of the response spectrum, s
Tw Period of the in‐plane wall, s
Vd Shear force transferred from the diaphragm, kN
Vw Horizontal seismic shear force induced from the inertial forces of the in‐plane wall, kN
Vdi Shear force transfer from the diaphragm at level i, kN
Vsi Storey shear force for level i, kN
Wd Total seismic weight tributary to the diaphragm, kN
Ww Total dead load of the in‐plane wall, kN
Δd The maximum in‐plane deflection of the diaphragm, due to a lateral load in the
direction under consideration equal to the weight tributary to the diaphragm, m
αw Modification factor for period of diaphragms to account for out‐of‐plane wall stiffness
C4.2 Scope
This section details recommendations for the determination of design seismic forces for in‐plane
loaded walls of unreinforced masonry buildings with flexible diaphragms. The storey shear
design action defined in Section 4.3.2.9 shall be considered against the capacity of the individual
elements calculated in Sections 8–12.
C4.3 Analysis Procedures
Linear analysis procedures are appropriate when the expected level of nonlinearity is low. Static
procedures are appropriate when higher mode effects are not significant, which is generally true
for short, regular buildings. Dynamic procedures are required for tall buildings, buildings with
torsional irregularities, or non‐orthogonal systems.
The Nonlinear Static Procedure is acceptable for most buildings, but should be used in
conjunction with the Linear Dynamic Procedure if mass participation in the first mode is low.
C4‐1
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
The term “linear” in linear analysis procedures implies “linearly elastic”. The analysis procedure,
however, may include geometric nonlinearity of gravity loads acting through lateral
displacements and implicit material nonlinearity of concrete and masonry components using
properties of cracked sections. The term “nonlinear” in nonlinear analysis procedures implies
explicit material nonlinearity or inelastic material response, but geometric nonlinearity may also
be included.
C4.3.1 Modelling of the Structure
Plane frame analysis will provide reasonably accurate predictive capacity if the following
considerations are observed:
(a) Because of the importance of shear deformation in URM buildings, the approach of
NZS 3101 of incorporating shear deformation through a reduced flexural stiffness is not
recommended. Instead, shear and longitudinal strains should each be specifically allowed
for.
(b) Regions common to two or more intersecting members (joint panel zones) may require
special attention. Parametric studies indicate that the member cross‐sections may be
assumed to penetrate unmodified to the intersection point or may be assumed to have
stiffer (or rigid) end sections. Examination of stresses in the joint regions is seldom
necessary.
(c) Much of the seismic mass of most unreinforced masonry buildings (particularly those with
timber floors) lies in the walls themselves. This mass distribution, coupled with the size of
the members, suggests that large inertia effects may be involved. Rocking modes in
particular, but also other modes, will involve rotational accelerations and vertical
accelerations that are of greater importance than in modern skeletal structures. Modelling
of the mass distribution is therefore important.
(d) To be able to predict inelastic strains, some form of pushover analysis is desirable.
However, this exercise need not necessarily require sophisticated software as simple hand
procedures will often suffice.
The preferred method is to model the elements as assemblages of plane stress elements, shell
elements, or solid three‐dimensional elements (so‐called “brick” elements). Reasons for this
include the relative ease with which the structure can be accurately described and because the
strains and stresses can be monitored. However, sufficiently sophisticated software is not readily
accessible and the computational costs can be high.
Plane stress element modelling might use one of the following three levels of complexity, with
each level representing progressively greater abstraction.
(a) Modelling may treat the mortar as separate elements from the bricks. Given the relative
thinness of the mortar bed joints it seems appropriate to model these joints as line
elements, with quadrilateral elements used for the bricks. It is common to ignore the
effects of head joints, except to determine failure criteria. Separate failure criteria are used
for each of the bricks and the mortar. Gap elements can be introduced to model separation
C4‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
of layers (or laminae) at bed joints. Properties of the materials (or of the earthquake effects
at the site) are seldom known sufficiently fully to warrant this level of modelling, which
might be used for very important historic buildings or for smaller sections of larger
assemblages.
(b) The properties of the mortar joints and of the bricks can be combined using
homogenisation techniques. The weighted properties are then used to produce continuum
elements. For best results, sampling and/or integration points are aligned with the bed
joints, with or without gap elements. It is common to use the initial (elastic) stiffness
throughout, with appropriate iteration at each load step.
(c) The most abstract modelling combines the stiffness and strength properties of the bricks
and mortar as above, but otherwise treats the masonry as a continuum. It is common to
assume that the masonry is isotropic. Gap elements are seldom included. This level of
complexity is suited to large structures and produces adequate results for assessment
purposes.
Plane stress elements can be employed for modelling in three dimensions. However, because
these elements have only two degrees of freedom at each node, some contrivance is necessary
to prevent numerical instability. One means, in which spring elements are added to the
necessary degrees of freedom, is often sufficient. However, the allocation of mass to the nodes
must then be made using other techniques, usually at the expense of accurate predictive
capacity. Use of shell elements or three‐dimensional “brick” elements overcomes this difficulty,
but at some computational expense. Suitable approximations can often be employed to allow
adequate rigour using two‐dimensional analyses. These approximations include:
(a) Inclusion of return walls that might be lifted when the wall under study is deflected. The
width of return wall that is affected can be assessed from the shear stresses at the
intersection of the two walls. However, this approach should be pursued with caution, as
there may be discontinuous bonding at corners.
C4‐3
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Three‐dimensional analysis that allows for the interaction of in‐plane and out‐of‐plane effects
can be conducted, but the necessary software is not yet readily available and its use requires
good modelling and interpretative skills.
It should be appreciated that computer programs are an aid to rapid design and evaluation, and
are not essential to the assessment procedures described in Section 4. Some examples use
simple hand calculation for inelastic effects. However, computer programs are almost
indispensable for rigorous elastic analyses and for modal response spectrum analyses.
It is recommended to verify the analytical predictions of the numerical models with the actual
response measured in the structure, and improve the model accordingly. For this purpose
experimental or operational modal analysis techniques can be applied in combination with
model updating procedures.
C4.3.1.1 Numerical Modelling of URM Buildings
Although structural analysis software has existed since the 1960’s and the first non‐linear
analysis programs were developed in the mid 1970’s (Kelly 2004), numerical modelling of
unreinforced masonry structures is still a computationally demanding procedure in terms of
memory capacities and processing time. This elevated numerical cost is related to the intrinsic
complexity of masonry (arrangements of bricks connected by mortar joints) that requires a large
number of degrees of freedom for modelling (Giordano, et al. 2002) and excludes the typical
simplifications and assumptions that are suitable for other kinds of structures (homogeneous
materials, rigid diaphragms, connections with linear behaviour, etc.). Another reason for this
complexity is the material non‐linear behaviour, whose mechanical characteristics are strongly
dependant on the characteristics of local materials and, consequently, are not easy to describe
by a general constitutive model. To address these problems, different numerical procedures
have been proposed for URM structures, which can be grouped into three main categories:
macro‐modelling, micro‐modelling and homogenized finite element modelling.
C4.3.1.2 Macro‐modelling
The simplest and most economical approach is to model the structural components (e.g.
columns, beams, walls, floors) with a single modelling resort (e.g. truss, frame, panel or shell
elements). In this category are included lumped‐mass models, equivalent frame models and
macro element representations. These models are especially recommended for time‐history
analysis using ground motion records and non‐linear constitutive material models. The models
provide information about the overall performance of the structure, but are poorly
representative of the local behaviour of the elements, such as: crack propagation, failure of
individual elements and damage localization (Lourenço 2004).
C4‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
In the last decade, some software packages have been developed to analyze masonry structure
using a macro‐modelling approach (Lagomarsino et al. 2004; Magenes et al. 2003) and other
programs have incorporated modelling resorts to represent the masonry behaviour (Carr 2004).
Good examples of this kind of approach are presented in the works of Doherty et al. (2002),
Pasticier et al. (2007), Chen et al. (2008) and Augenti and Romano (2008).
C4.3.1.3 Micro‐modelling
In a micro‐modelling approach, an analysis based on the characteristics of each component
(brick and mortar) are performed separately. The model discretization of the structure follows
the actual geometry of the brick and mortar joint arrangement, and adopts different constitutive
models for each component. Special attention is given to the interfaces, as they are considered
to be the weakest member where tensile‐sliding failure can be expected. In most cases, it can be
considered that all the non‐linear behaviour of the combined material occur in the mortar joints.
A complete micro model must include all the failure mechanisms of masonry, namely: cracking
of joints, bed‐joint sliding and cracking or crushing of the masonry units (Lourenço 2004).
Micro‐modelling is preferred for simple structures (such as a single wall) and for rigorous
analysis of the local stress‐strain behaviour of masonry to predict crack propagation. The use of
these procedures is still primarily restricted to research applications, because the approach
demands a high computational effort in terms of numerical processing and memory.
In micro‐modelling, three models appear to be the most effective and widely used to represent
URM structures: the finite elements method with discrete elements (FEMDE), the discrete
elements method (DEM) and the rigid blocks limit state analysis (RBLSA).
In FEMDE, bricks are represented by deformable or rigid continuum elements, while the mortar
joints are simulated by non‐linear interface layers. The brick‐mesh and the interface‐mesh have
the same joints along the contact surface. In this group of micro‐models, remarkable works are
the studies conducted by Lourenço and Rots (1997) and Gambarotta and Lagomarsino (1997a)
and, more recently, the research of Sacco (2009).
In DEM, the structure is considered as an assembly of rigid or deformable blocks, connected
through elasto‐plastic link elements (non‐linear springs). These elements follow a degrading
hysteretic rule to represent the brittle behaviour of the masonry. The main advantages of this
method are an adequate formulation for large displacements, the chance to follow the
displacement of every block, and a better prediction of the collapse mechanism. The
disadvantage of this method is the large number of link points required to represent the
interface stresses accurately, which means time and memory consuming analysis. Interesting
studies that have applied this method are Formica et al. (2002) and Casolo and Peña (2007).
The RBLSA method has received less attention from the structural engineering community, even
though its fundamentals were established more than thirty years ago (Livesley 1978). In this
approach, the structure is considered as an assembly of non‐deformable blocks. The interactions
are represented through interfaces that are able to transfer compression and shear stresses
(using a non‐cohesive Coulomb criterion), but not tensile stress. In this method, the kinematic
problem is mathematically solved by simultaneously satisfying the equilibrium, compatibility and
C4‐5
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
the failure and flow rules at each block’s contact surface. The numerical simplicity and the
subsequent fast analysis are the main advantages of this method. However, in this method, it is
only possible to determinate the collapse load, but not the collapse displacement. Also, it is
difficult to include the load history in the analysis (Lourenço 2008). A recent example for
application of this method is presented in Orduña et al. (2007).
C4.3.1.4 Homogenized Finite Element Modelling
Homogenized finite element modelling is a meso‐modelling approach that bypasses the bi‐
material nature of masonry. URM is considered as a continuum and homogeneous material with
properties that are equivalent to the average properties of the combined material. These
properties are incorporated into the model by means of homogenization techniques (Milani et
al. 2006; Yu et al. 2008; Addessi et al. 2010; Casolo and Milani 2010). This representation of
masonry allows the use of traditional finite element (FE) modelling techniques, which do not
require modelling of the arrangement of each brick.
Homogenized FE modelling allows analysis with more complicated excitations and non‐linear
constitutive material models (non‐linear time‐history analysis). This procedure also provides
good information about the local performance of the structural elements (e.g. crack
propagation, stress and strain distribution). All these characteristics make this method especially
interesting for the analysis of real structures; however, their computational cost is high. The
main difficulty in implementing these techniques is the lack of experimental data regarding the
linear and non‐linear behaviour of masonry. However, several researchers have proposed
constitutive models that characterize the anisotropic and quasi‐brittle behaviour of masonry
(Lourenço and Rots 1997; Formica, et al. 2002; Gambarotta and Lagomarsino 1993). Interesting
works to refer to about this method are Gambarota and Lagomarsino (1997a; 1997b), Berto et
al. (2004) and Brasile et al. (2007a; 2007b; 2010)
C4.3.1.5 Verification and Tuning of Numerical Models of URM Structures
Currently, there is an increasing concern to develop techniques to determine the degree of
coherence between model prediction and the actual behaviour of buildings. This interest has
originated from the need for accurate models that predict the performance of existing masonry
structures, not only under operational conditions, but also under extreme events (strong winds,
earthquakes, imposed loads, etc.). Accurate models aid in the optimal design of retrofit
interventions.
In general, current models have been validated by studies based on structural component
behaviour (e.g. walls, piers, spandrels). However, a validation at system level (entire buildings or
sections of building) is still necessary. In the last twenty years, several techniques have been
developed for quantitatively assessing the structural health condition of existing buildings and,
also, to verify the concordance of the proposed models with actual structural response. One of
the most attractive approaches to this subject corresponds to methods based on recording the
dynamic structural response generated by small magnitude excitations. The information
extracted from this test is used later to improve the numerical models. Then, the updated
C4‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
models are employed to predict the structural response under large magnitude excitations. In
these techniques, procedures such as modal testing, system identification, model verification
and updating are applied.
Modal testing and system identification are two techniques, that combined are able to extract
data from physical dynamic tests (response time‐series records) and utilize this data to
determine the dynamic characteristics of the structure, normally expressed in terms of modal
properties (natural frequencies, mode shape and modal damping). Then, these modal properties
are used to obtain the physical properties of the system (mass, stiffness and damping matrices)
required to develop a mechanical model of the system (Figure C4‐1).
Figure C4‐1: Modal testing and system identification procedure
Traditionally, modal data has been obtained from measurements of the excitation applied to the
structure (input) and the structural response (output). This approach is known as experimental
modal analysis (EMA). Another approach is operational modal analysis (OMA). OMA utilizes only
response measurements of the structure in its operational condition when subjected to ambient
or natural excitation to identify modal characteristics. Further explanations about the
fundamentals of experimental and operational modal analysis are presented in the references
(Ewins 2000; He and Fu 2001; Heylen et al. 1997; Silva and Maia 1999; Juang 1994; Brownjohn
2007). A number of methods have been proposed to obtain the modal parameters from
vibration measurements, which can be classified as time‐domain and frequency‐domain
techniques. Some of these procedures are briefly presented in Table C4‐1. Detailed descriptions
of each method are presented in Ewins (2000), He and Fu (2001) and the references that are
presented in Table C4‐1.
C4‐7
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Table C4‐1: Modal parameter extraction methods
Peak picking method (Bendat and Piersol 1993)
Frequency‐domain
Circle‐fit method (Maia 1999)
Inverse frequency response function method (Ewins 2000)
Dobson’s method (Dobson 1987)
Frequency domain decomposition (Brincker et al. 2000)
Enhanced frequency domain decomposition (Brincker et al. 2001)
Least square time domain method (Smith 1981)
Ibrahim’s method (Ibrahim and Mikulcik 1976)
Time‐domain
Random decrement method (Ibrahim 1977)
Auto‐regression moving average (ARMA) method (Andersen et al. 1995)
Eigen‐system realization algorithm (ERA) (Juang and Pappa 1985)
Stochastic subspace identification method (van Overschee and de Moor 1996)
Model updating techniques automate the procedures of reducing the discrepancies between the
numerical model predictions and the observed behaviour measured in the real structure (Link
2001). The basic steps to validate and improve the predictions of a numerical model are: (a)
comparison and correlation of model and experimental results, (b) model conciliation and error
localization, and (c) model correction (Ewins 2001).
Recently, many experiences on modal testing and model updating techniques applied to civil
engineering structures have been published, such as: Bagchi (2005), Castillo and Boroschek
(2005), Jaishi and Ren (2005; 2007), Skolnik et al. (2006), Pavic et al. (2007), Pallares et al. (2009),
and Núñez and Boroschek (2010). However, few studies on URM structures are available, with
most considering relatively slender structures (towers, minarets, chimneys, etc): De Sortis et al.
(2005), Bayraktar et al. (2008), Gentile et al. (2009), and Ramos et al. (2010).
C4.3.2 Equivalent Static Method
The simplified linear static analysis presented herein is based on the simplified analysis
presented in FEMA 356 (2000), which was the precursor document to ASCE/SEI 41‐06 (2007).
The equivalent static method assumes that the in‐plane loaded masonry shear walls are
relatively rigid elements which do not amplify the earthquake ground motions with increasing
height. The dominant mode of response of the URM building is assumed to be the in‐plane
fundamental mode of each of the diaphragms excited by the inertial force of the out‐of‐plane
loaded URM walls. The dynamic response of unreinforced masonry buildings is dominated by
the behaviour of the flexible timber diaphragms in their fundamental mode. An assumption of
rigid diaphragm behaviour may lead to unconservative assessments of diaphragm accelerations
and inaccurate estimates of load distribution between lateral load resisting elements (Oliver,
2010).
C4‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
A URM building with flexible diaphragms does not have a defining fundamental period that can
be identified by a mass participation of greater than 80%. Instead, the dynamic response is more
closely described by a system of single degree of freedom systems. Each diaphragm responds at
its own natural frequency based on its geometry, stiffness and tributary mass, and is considered
uncoupled from the response of adjacent stories.
C4.3.2.1 Limitations of this Section
Section 4.3.2.1 sets out the limitations of the Equivalent Static Method. The simplified method
for finding the force demand only applies to the specific case of unreinforced masonry buildings
with flexible timber diaphragms at all storey levels, up to a maximum of six stories. This
restriction of the analysis method is unlikely to be limiting in its use as a study on the prevalence
and typology of New Zealand URM buildings has found that of all the URM buildings in New
Zealand only 14% exceed two stories in height, and less than 1% are over five stories in height
(Russell and Ingham 2010).
C4.3.2.2 In‐plane Loads on Shear Walls
The in‐plane loads on shear walls are calculated from two components, one based on the
diaphragm response, and the other based on the wall in‐plane response. The dynamic response
of a multi‐storey unreinforced masonry building with flexible diaphragms is concentrated in the
diaphragms. The equivalent static method assumes that the in‐plane wall is rigid in comparison
to the flexible diaphragm for the purpose of calculating the inertial force induced in the
diaphragm. Adequate connections between the diaphragm and the in‐plane loaded walls are
required in order to transfer the inertial force induced in the diaphragms and the out‐of‐plane
loaded walls into the in‐plane loaded walls. The wall in‐plane response component is based on
the shear wall behaving in its fundamental mode. For solid shear walls, calculation of the elastic
stiffness and therefore period is relatively simple, but for the case of perforated walls this
calculation is much more complicated and hence the recommendation that the in‐plane loaded
wall period is assumed to lie in the constant acceleration region of the response spectra.
The two components of shear load are based on the pseudo lateral load calculation outlined in
FEMA 356 (2000) and adopted by the NZSEE (2006) guidelines. Further explanation of the
background to the equation can be found in section 3.3.1 of FEMA 356.
C4.3.2.3 Damping
Unreinforced masonry does not respond is a classical elastic‐plastic manner, exhibiting more of a
nonlinear elastic response with pinched hysteretic loops. From analysis of full scale pseudo‐static
tests on in‐plane loaded flanged walls and spandrel/pier sub‐structures, a range of 5‐15%
equivalent hysteretic damping was found. The past recommendation of 15% of critical was
formed from a combination of approximately 5% inherent viscous damping associated with
elastic structures, 5% hysteretic damping and 5% damping from dynamic impact (NZSEE, 2006).
The hysteretic damping includes Coulomb (or “dry”) damping associated with sliding shear,
radiation damping, pinched hysteretic damping associated with flexural/diagonal shear softening
C4‐9
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
and equivalent damping due to impact on rocking. The current recommendation of 5% is a
reduction from the 15% recommended by Magenes and Calvi (1997) and advocated in the NZSEE
(2006) guidelines, and is supported by the increase in displacement ductility recommended for
use in finding demand on in‐plane loaded URM shear walls which allows for a reduction in loads
based on nonlinear behaviour. The recommendation of 5% of critical is based on the
understanding that the increase in ductility to 2, and the reduction due to the structural
performance factor, allows for the reduction in load due to nonlinear response and therefore
the reduction would be two‐fold if damping was also increased.
C4.3.2.4 Ductility and Structural Performance Factor
Unreinforced masonry is generally considered a brittle material and therefore it has been
justified in the past to use a displacement ductility factor of 1. The notion of URM as a brittle
material is inaccurate and conservative. Ductility is present not from conventional yielding of
steel reinforcing bars, but from inelastic response and energy dissipation due to shear sliding
along mortar joints and rocking of masonry block elements. When URM walls respond in‐plane,
particularly with a rocking mechanism, there is significant inelastic displacement capacity and a
strongly nonlinear behaviour. A typical shear response is shown in Figure C4‐2(a), and shows no
rapid and severe strength loss as associated with a typical brittle failure. The force‐displacement
plot associated with pier and pier/spandrel sub‐structure testing indicates that a flexural failure
mechanism, or a flexural/shear combined failure mechanism, is capable of ultimate
displacements that greatly exceed the yield displacement, as shown in Figure C4‐2(b). URM
buildings have multiple sources of ductility that develop from energy dissipation drawn from
shear mechanisms in the in‐plane loaded walls, rocking of in‐plane loaded URM elements, and
energy dissipation from deformations of the timber diaphragms. The recommendation for using
a displacement ductility of 2 recognises the nonlinear response of URM walls to in‐plane loading
and the energy dissipation that occurs during loading cycles beyond the elastic state. This value
of displacement ductility of 2 corresponds to a structural performance factor of 0.7 as per
NZS 1170.5. The structural performance factor of 0.7 allows for a decrease in elastic site hazard
spectra based on observations of strength reduction and stiffness degradation apparent in the
second and third testing cycles to a specific displacement. It is noted that these values are put
forth as a recommendation and it is expected that the designer apply these or other values as
they feel appropriate for the condition of the building under assessment. If a value of 1 is
adopted for both the displacement ductility and the structural performance factor then an
increase in damping to 15% is recommended.
C4‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(a) URM pier (b) URM spandrel substructure
Figure C4‐2: Plots of Force‐displacement response of URM assemblages
C4.3.2.5 Diaphragm Shear Load Transfer
Coefficient C1 is a modification factor used to account for the difference in maximum elastic and
inelastic displacement amplitudes in a structure exhibiting relatively stable and full hysteretic
loops. The values of the coefficient are based on analytical and experimental investigations of
the earthquake response of yielding structures (NZSEE, 2006). See ASCE/SEI 41‐06 (2007) for
further discussion.
Coefficient C3 is the modification factor for framing systems that exhibit negative post‐yield
stiffness, as dynamic P‐Δ effects may lead to significant amplification of displacements for
buildings exhibiting this characteristic. Such effects cannot be explicitly addressed with linear
procedures as no measure of the degree of negative post‐yield stiffness can be explicitly
included in a linear procedure (NZSEE, 2006). When the stability coefficient is less than 0.1 in all
stories the p‐delta effects need not be considered. If the stability factor is between 0.1 and 0.33,
the seismic actions and deformations shall be increased according to equation 4‐4. When the
stability factor exceeds 0.33, the structure shall be considered unstable and retrofit intervention
to reduce the computed lateral deflections is required (FEMA 356, 2000).
C4.3.2.6 Period of Diaphragms
From experimental research it has been established that diaphragm deformations are governed
primarily by shear deformations rather than flexural deformations. The source of the period
equation expressed in section 4.3.2.6 is based on the shear deflection of a beam. As outlined in
Chopra (2001), a shape function can be used to describe the beam deformations, which can then
be used to develop a generalised mass and generalised stiffness for shear. This information can
then be introduced to the generalised period equation to obtain a specific expression for the
period. A more detailed explanation can be found in Knox (2011).
C4‐11
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
It is important to use the correct stiffness value, as the stiffness of the diaphragm varies
significantly with direction of loading with respect to the orientation of the joists.
The reduction factor is based on modelling conducted on the effect of the out‐of‐plane wall
stiffness on the period of the diaphragm.
C4.3.2.7 In‐plane Inertial Shear Load
The in‐plane inertial shear load calculation uses the equivalent static method outlined in section
5.2 of NZS 1170.5. The damping, ductility and structural performance factor pertinent to URM
in‐plane loaded shear walls is applied through the reduction of the response spectra value and
the calculation of the horizontal design action coefficient.
C4.3.2.8 Vertical Distribution of Inertial Shear Load
The rectangular distribution of the inertial shear load vertically up the height of the building
reflects the assumption of effectively rigid wall response in comparison to the flexible diaphragm
response. A rectangular distribution implies no modification of the ground accelerations up the
height of the wall.
C4.3.2.9 Storey Shear Load
Storey shear load is the addition of the diaphragm shear load and the component of in‐plane
inertial shear load distributed to that level using the Square Root of Sum of Squares rule.
C4.3.2.10 Total Shear Load
Summation of the inertial storey shear load and the individual diaphragm shear loads using a
CQC or SRSS method allows for the individual elements to respond at different periods and
therefore not necessarily in phase.
The SRSS rule for modal combination, developed by Rosenblueth (1951) is given in Equation
C4‐1.
(C4‐1)
The complete quadratic combination (CQC) rule is given in Equation C4‐2.
(C4‐2)
C4‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
/
8
(C4‐3)
1 4 1 4
Where:
and is the damping ratio for mode i.
A more detailed explanation of the CQC method can be found in Chopra (2007).
C4.3.3 Modal Response Spectrum Method
The applicability of this method to URM buildings is currently under review. URM structures do
not satisfy several of the fundamental hypotheses required to apply modal spectral analysis,
such as rigid diaphragms, mass concentrated at diaphragm levels, or well spaced and
differentiated modes. In general, the vibration modes are associated with the particular
response of structural members (walls, diaphragms, parapets, etc.) and not to the global
building response.
Although this method might be inappropriate for determining design actions, it is adequate to
use the dynamic properties (fundamental frequency, modal damping and mode shapes)
obtained by this method to update a numerical model. In this process the numerical predictions
are contrasted with experimental response obtained from a vibration test (experimental or
operational modal test).
C4.3.4 Inelastic Time History Method
To assist in the selection of ground motion records for use in inelastic Time History analysis, five
zones of similar seismological signature have been defined for the North Island of New Zealand
(Figure C4‐3). The hazard scenarios defined for each of these zones in the period range between
0.4s and 2.0s are presented in Table C4‐2.
The Inelastic Time‐History analysis shall be guided by the recommendations stated in
NZS 1170.5:2004. However, it is suggested to consider the average response of seven or more
records for determining the design actions, instead of the most disadvantageous response of
three or more records.
C4‐13
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Figure C4‐3: Seismic hazard zonation for the North Island of New Zealand for ground motion
record selection
C4‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C4‐2: Recommended suite profile for each seismic zone
Hazard Distance
Zone Magnitude Tectonic class/Fault mechanism
scenario (km)
Shallow crustal earthquakes without
NF1 7.0 ‐ 7.5 <~10
pronounced forward‐directivity features
Records with strong forward directivity
North NF
NF2 6.5 ‐ 7.5 <~10 characteristics (2 or 3 records in the suite
of seven records)
Records from regions of extensional
North V V1 6.0 – 7.0 <20
tectonic regimes
NZS1170.5:2004 (SNZ, 2004a ‐ 2004b) requires that suites of at least three records must be
considered for time‐history analysis, and that the most disadvantageous response must be
considered for design (for example, the largest value of the maximum strength demands).
However, when following this criterion the design can be strongly influenced by the response of
a specific ground motion with particularly strong peaks. In contrast, other standards and
recommendations (CEN 2004; ASCE/SEI 2007) propose to design considering the average
response of seven or more records. Recent studies (Dhakal et al. 2007; Oyarzo‐Vera and Chouw,
2008) have demonstrated that this latter option improves the consistency of the time‐history
analysis, because it reduces the influence of specific ground motions. Also, the results of
multiple simulations considering the average response of seven time‐history analyses offer a
lower dispersion in terms of the variability in the results, than do the results of multiple
simulations taking into account the maximum response of three time‐history analyses. In
addition, in the North Island of New Zealand, there are few locations where the hazard is totally
dominated by one earthquake source. Even for sites adjacent to one of the major strike‐slip
faults, there are usually other nearby faults, or the underlying subduction zone has influence,
especially for spectral periods of about 1 second and longer. Thus seven accelerograms for a
C4‐15
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
zone generally covers a wider range of earthquake types or magnitude‐distance combinations.
Taking into account the international trend, the studies mentioned previously and the particular
seismic characteristics of the North Island, suites of seven records were defined for each zone
and soil class. (Oyarzo‐Vera et al. 2011).
As URM structures do not have well defined global modes, the target period considered for
scaling shall consider the fundamental periods associated with the local modal response of the
most important structural member (in‐plane wall response, out‐of‐plane wall response, flexible
diaphragm response, parapets, etc). For this purpose and to reduce the number of analyses
required to assess the response of the different structural members, the range of periods
considered for scaling can be extended to cover the fundamental periods associated with more
than one structural member.
The study conducted by Oyarzo‐Vera et al. (2011) suggests suites of records for each zones
presented in Figure C4‐3, to be used in time‐history analysis in the absence of site specific
studies. These suites of records have not been included in this document, as they can be
updated, by replacing or adding new records, according to the state‐of‐the‐knowledge and the
requirements or future amendments of the Standard.
C4.4 Design Example
The following is an example calculation for finding the in‐plane loads on an unreinforced
masonry shear wall.
The example building is a five storey URM building with flexible timber diaphragms at all levels.
The building has regular perforations on two parallel sides and mainly solid masonry with
minimal perforations on the other two sides. The building is located in Auckland, on shallow soil.
The floor plan is rectangular, with dimensions of 29.9 m by 17.9 m. The walls decrease in
thickness from five leafs for level one and level two, four leafs for level three and four, and down
to three leafs for the top level.
An elevation view of the building’s eastern wall and a plan view of the building dimensions are
shown in Plan view (b) Elevation of eastern wall
Figure C4‐4. Note that this is an extension of the example previously published by Oliver (2010),
associated with the seismic assessment and improvement of flexible timber diaphragms.
C4‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
3
0
3
6
0
N
joists
extg
1
7
3
6
0
extg
joists
extg
steel
beam 0
typical 0
0 0
0 2
0
4
5
joists
9
extg 7
3
0
0 0 9
0 6 5
9 3 3
9 0 5
2 3 4
6
3
joists
0
extg
0
7
3
5
5
8
3
0
1
0
joists
4
extg
5
0
2
4
0
0
4 0
4 0
joists
2
extg
2
1
4
0
0
1
4
0
0
1
4
0
0
1
4
0
0
1
4
0
0
1
4
0
0
1
4
0
0
(a) Plan view (b) Elevation of eastern wall
Figure C4‐4: Layout of example building
Find Diaphragm Periods:
Table C4‐3 – Diaphragm Weight
Diaphragm Stiffness, Kd:
4
αp = 1
αc = 1
Gd = 175 kN/m (joists perpendicular to loading direction)
be = 17.9 m
L e = 29.9 m
C4‐17
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
1 1 4 17.9 175
419.06 kN/m
29.9
Table C4‐4 – Diaphragm Periods
Note that the deflections calculated in Table C4‐4 are applicable for a 1g acceleration.
Diaphragm shear load transfer
C1 = 1
C3 = 1
Z = 0.13
R = 1
N(T,D) = 1
Ch(Ti) = spectral shape factor determined from Table 3.1 of NZS 1170.5
C(Ti) = Ch(Ti) × Z × R × N(T,D)
Cd(Ti) = C(Ti) × Sp / Kμ
Sp = 0.7
μ = 2
kμ = 2
Table C4‐5 – Diaphragm Shear Load Transfer
C4‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
In‐plane inertial shear load
Tw = Unknown, but assume that it lies on the plateau of the design spectrum
Ww = 2671 kN
Ch(Tw) = 2.36
C(Tw) = 2.36 x 0.13 x 1 x 1 =0.3068
kμ = 1.2
Cd(Tw) = 0.3068 x 0.7 / 1.2 = 0.17897
C1 = 1
C3 = 1.2
Vw = 1 x 1.2 x 0.17897 x 2671 = 573.62 kN
Storey Shear Load
2
94.8 2 4.205 2
573.6 = 131.28 kN
2 19.7
89.3 2 3.855 2
573.6 = 120.78 kN
2 19.7
52.2 2 3.645 2
573.6 = 109.28 kN
2 19.7
49.2 2 3.795 2
573.6 = 113.20 kN
2 19.7
29.5 2 2.00 2
573.6 = 60.06 kN
2 19.7
Total Base Shear Load
Β = 3.16/3.06 = 1.026 < 1.35 therefore use CQC method
Note: The CQC method requires an approximation of the in‐plane loaded wall period.
C4‐19
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Calculate table of
Calculate table of
Note: For this example is taken as 0.05 for all modes.
/
8
1 4 1 4
Calculate table of x Vi x Vn
/
For example:
for i = 3 and n = 2, = 0.073
Vd3/2 = 26.04 kN and Vd2/2 = 44.6 kN, so 0.073*26.04*44.6 = 84.49
C4‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Total base shear load using CQC method = 584.6 kN
For comparison, using the SRSS method
/
Total base shear load = (Σ47.362 + 44.602 + 26.042 + 24.582 + 14.712 + 573.62)1/2 = 587.6 Kn
Storey Shear Force
584.65 584.65
1.094
∑ 131.28 120.78 109.28 113.2 60.06
Therefore,
Shear force level 5 = 1.094*60.06 =65.7
Shear force level 4 = (1.094*60.06) + (1.094*113.20) = 189.5
Shear force level 3 = (1.094*60.06 + (1.094*113.20) + (1.094*109.28) = 309.0
Shear force level 2 = (1.094*60.06 + (1.094*113.20) + (1.094*109.28) + (1.094*120.78) = 441.1
C4‐21
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Shear Force Diagram
Now repeating the exercise assuming the diaphragms to be stiffened as reported in Oliver
(2010), we obtain,
Diaphragm Stiffness, Kd:
14730 /
Table C4‐4 – Diaphragm Periods
Diaphragm shear load transfer
C1 = 1
C3 = 1
Z = 0.13
R = 1
N(T,D) = 1
Ch(Ti) = spectral shape factor determined from Table 3.1 of NZS 1170.5
C(Ti) = Ch(Ti) × Z × R × N(T,D)
C4‐22
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Cd(Ti) = C(Ti) × Sp / Kμ
Sp = 0.7
μ = 2
kμ = 2
Table C4‐5 – Diaphragm Shear Load Transfer
In‐plane inertial shear load
Vw = 1 x 1.2 x 0.17897 x 2671 = 573.62 kN (same as in above example)
Storey Shear Load
2
220.9 2 4.205 2
573.6 = 164.9 kN
2 19.7
204.2 2 3.855 2
573.6 = 151.7 kN
2 19.7
180.3 2 3.645 2
573.6 = 139.3 kN
2 19.7
161.0 2 3.795 2
573.6 = 136.7 kN
2 19.7
129.0 2 2.00 2
573.6 = 86.9 kN
2 19.7
Note that with respect to the loads calculated for the case of the unretrofitted diaphragms,
stiffening of the diaphragms resulted in a significant increase in the storey shear loads. This
increase ranged from 21% to 45%, emphasizing the need to re‐evaluate the loads actions for an
proposed diaphragm improvements.
Total Base Shear Load
Β = 0.907/0.850 = 1.067 < 1.35 therefore use CQC method
C4‐23
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Note: The CQC method requires an approximation of the in‐plane loaded wall period.
Using the CQC method outlined above the total base shear load, Vb, is 651 kN.
C4.5 References
Addessi, D., Sacco, E., and Paolone, A. (2010). "Cosserat model for periodic masonry deduced by
nonlinear homogenization." European Journal of Mechanics ‐ A/Solids, 29(4), 724‐737.
Andersen, P., Brincker, R., and Kirkegaard, P. H. (1995) "On the uncertainty of identification of
civil engineering structures using ARMA models." Proc., 13th International Modal Analysis
Conference (XIII IMAC), Nashville, Tennesse.
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, VA.
Augenti, N., and Romano, A. "Seismic design of masonry buildings through macro‐elements."
Proc., 14th International Brick and Block Masonry Conference (14IBMAC), Sydney, Australia,
17 – 20 February.
Bagchi, A. (2005). "Updating the mathematical model of a structure using vibration data."
Journal of Vibration and Control, 11(12), 1469‐1486.
Bayraktar, A., Altunişik, A., Sevim, B., Türker, T., Akköse, M., and Çoşkun, N. (2008). "Modal
analysis, experimental validation, and calibration of a historical masonry minaret." Journal
of Testing and Evaluation, 36(6), 1‐9.
Bendat, J., and Piersol, A. (1993). Engineering applications of correlations and spectral analysis,
John Wiley & Sons, New York, USA.
Berto, L., Saetta, A., Scotta, R., and Vitaliani, R. (2004). "Shear behaviour of masonry panel:
Parametric FE analyses." International Journal of Solids and Structures, 41(16‐17),
4383‐4405.
Brasile, S., Casciaro, R., and Formica, G. (2007a). "Multilevel approach for brick masonry walls ‐
Part I: A numerical strategy for the nonlinear analysis." Computer Methods in Applied
Mechanics and Engineering, 196(49‐52), 4934‐4951.
Brasile, S., Casciaro, R., and Formica, G. (2007b). "Multilevel approach for brick masonry walls ‐
Part II: On the use of equivalent continua." Computer Methods in Applied Mechanics and
Engineering, 196(49‐52), 4801‐4810.
Brasile, S., Casciaro, R., and Formica, G. (2010). "Finite Element formulation for nonlinear
analysis of masonry walls." Computers & Structures, 88(3‐4), 135‐143.
Brincker, R., Andersen, P., and Zhang, L. "Modal identification from ambient responses using
frequency domain decomposition." Proc., 18th International Modal Analysis Conference
(XVIII IMAC), San Antonio, Texas, 625‐630.
C4‐24
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Brincker, R., Ventura, C. E., and Andersen, P. "Damping estimation by frequency domain
decomposition." Proc., 19th International Modal Analysis Conference (XIX IMAC),
Kissimmee, Florida, 698‐703.
Brownjohn, J. M. W. (2007). "Structural health monitoring of civil infrastructure." Philosophical
Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences,
365(1851), 589‐622.
Carr, A. J. (2004). "Ruaumoko – A program for inelastic time‐history analysis." Department of
Civil Engineering, University of Canterbury, New Zealand.
Castillo, A. E., and Boroschek, R. L. "Uso del método de frecuencia no paramétrica y SSI para la
identificación de propiedades modales de edificios." Proc., IX Jornadas de la Asociacion
Chilena de Sismologia e Ingenieria Antisismica, Universidad de Concepcion, Chile.
Casolo, S., and Peña, F. (2007). "Rigid element model for in‐plane dynamics of masonry walls
considering hysteretic behaviour and damage." Earthquake Engineering and Structural
Dynamics, 36(8), 1029‐1048.
Casolo, S., and Milani, G. (2010). "A simplified homogenization‐discrete element model for the
non‐linear static analysis of masonry walls out‐of‐plane loaded." Engineering Structures,
32(8), 2352‐2366.
Chen, S. Y., Moon, F. L., and Yi, T. (2008). "A macroelement for the nonlinear analysis of in‐plane
unreinforced masonry piers." Engineering Structures, 30(8), 2242‐2252.
Chopra, A.K. (2007) Dynamics of Structures ‐ theory and applications to earthquake engineering.
Pearson Prentice Hall, New Jersey, United States of America
Comité Européen de Normalisation (CEN) (2004). Eurocode 8: Design of structures for
earthquake resistance ‐ Part 1: General rules, seismic actions and rules for
buildings. Eurocode, Brussels, Belgium.
De Sortis, A., Antonacci, E., and Vestroni, F. (2005). "Dynamic identification of a masonry
building using forced vibration tests." Engineering Structures, 27(2), 155‐165.
Dobson, B. J. (1987). "A straight‐line technique for extracting modal properties from frequency
response data." Mechanical Systems and Signal Processing, 1(1), 29‐40.
Doherty, K., Griffith, M. C., Lam, N., and Wilson, J. (2002). "Displacement‐based seismic analysis
for out‐of‐plane bending of unreinforced masonry walls." Earthquake Engineering &
Structural Dynamics, 31(4), 833‐850.
Ewins, D. J. (2000). Modal Testing: Theory, Practice and Application, Research Studies Press Ltd.,
Hertfordshire, England.
Ewins, D. J. (2001). "Modal analysis, experimental. Applications." Encyclopedia of Vibration,
Elsevier, Oxford, 829‐838.
C4‐25
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
FEMA 356 (2000) 'FEMA 356' Prestandard and Commentary for the Seismic Rehabilitation of
Buildings. American Society of Civil Engineers and Federal Emergency Management Agency,
Reston, VA.
Formica, G., Sansalone, V., and Casciaro, R. (2002). "A mixed solution strategy for the nonlinear
analysis of brick masonry walls." Computer Methods in Applied Mechanics and Engineering,
191(51‐52), 5847‐5876.
Gambarotta, L., and Lagomarsino, S. (1997a). "Damage models for the seismic response of brick
masonry shear walls. Part I: The mortar joint model and its applications." Earthquake
Engineering and Structural Dynamics, 26(4), 423‐439.
Gambarotta, L., and Lagomarsino, S. (1997b). "Damage models for the seismic response of brick
masonry shear walls. Part II: The continuum model and its applications." Earthquake
Engineering and Structural Dynamics, 26(4), 441‐462.
Gambarotta, L., and Lagomarsino, S. (1993). "A microcrack damage model for brittle materials."
International Journal of Solids and Structures, 30(2), 177‐198.
Gentile, C., Saisi, A., and Gallino, N. (2009) "Operational modal analysis and F.E. modelling of a
masonry tower." Proc., 3rd International Operational Modal Analysis Conference
(IOMAC'09), Ancona, Italy, 499‐506.
Giordano, A., Mele, E., and De Luca, A. (2002). "Modelling of historical masonry structures:
Comparison of different approaches through a case study." Engineering Structures, 24(8),
1057‐1069.
He, J., and Fu, Z.‐F. (2001). Modal Analysis, Butterworth‐Heinemann.
Heylen, W., Lammens, S., and Sas., P. (1997). Modal analysis theory and testing, Katholieke
Universiteit Leuven, Belgium.
Ibrahim, S. R., and Mikulcik, E. C. (1976). "The experimental determination of vibration
parameters from time responses." Shock and Vibration Bulletin, 46(5), 187‐196.
Ibrahim, S. R. (1977). "Random decrement technique for modal identification of structures."
Journal of Spacecraft and Rockets, 14, 696‐700.
Jaishi, B., and Ren, W.‐X. (2005). "Structural finite element model updating using ambient
vibration test results." Journal of Structural Engineering, 131(4), 617‐628.
Jaishi, B., and Ren, W.‐X. (2007). "Finite element model updating based on eigenvalue and strain
energy residuals using multiobjective optimisation technique." Mechanical Systems and
Signal Processing, 21(5), 2295‐2317.
Juang, J.‐N., and Pappa, R. S. (1985). "An eigensystem realisation algorithm (ERA) for modal
parameter identification and modal reduction." Journal of Guidance, Control and Dynamics,
8(5), 620‐627.
Juang, J.‐N. (1994). Applied System Identification, Prentice Hall PTR, NJ, USA.
C4‐26
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Kelly, T. E. (2004). "2003 State‐of‐the‐art report on seismic analysis." Bulletin of the New Zealand
National Society for Earthquake Engineering, 37(1), 23‐34.
Lagomarsino, S., Penna, A., and Falasco, A. (2004). "3Muri – Advanced seismic analysis of
masonry buildings." S.T.A.Data, Italy.
Link, M. (2001). "Modal updating and validating." Encyclopedia of Vibration, Elsevier, Oxford,
844‐856.
Livesley, R. K. (1978). "Limit analysis of structures formed from rigid blocks." International
Journal for Numerical Methods in Engineering, 12(12), 1853‐1871.
Lourenço, P. B. "Structural masonry analysis: Recent developments and prospects." Proc., 14th
International Brick and Block Masonry Conference (14IBMAC), Sydney, Australia, 17 – 20
February.
Lourenço, P. B., and Rots, J. G. (1997). "Multisurface interface model for analysis of masonry
structures." Journal of Engineering Mechanics, 123(7), 660.
Lourenço, P. B. (2004) "Current experimental and numerical issues in masonry research." Proc.,
International Workshop on Masonry Walls and Earthquakes, 119‐136.
Magenes, G. and G.M. Calvi (1997) “In‐Plane seismic response of brick masonry walls”,
Earthquake Engineering and Structural Dynamics, 26(11), 1091‐1112.
Magenes, G., Manzini, C., Morandi, P., Calliari, R., and Righetti, G. (2003). ANDILWall – Software
of structural design and verification of masonry buildings. CRSoft, Italy.
Maia, N. M. M. (1999). "Modal identification methods in frequency domain." Modal Analysis and
Testing, J. M. M. Silva, and N. M. M. Maia, eds., Kluwer Academics Publishers, Netherlands,
251‐264.
Milani, G., Lourenço, P. B., and Tralli, A. (2006). "Homogenised limit analysis of masonry walls,
Part I: Failure surfaces." Computers and Structures, 84(3‐4), 166‐180.
Núñez, T., and Boroschek, R. L. "Monitoreo continuo de una estructura en proceso constructivo."
Proc., X Jornadas de la Asociacion Chilena de Sismologia e Ingenieria Antisismica, Facultad
de Ciencia Fisicas y Matematicas, Universidad de Chile.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Oliver, S. J. (2010) “A design methodology for the assessment and retrofit of flexible diaphragms
in unreinforced masonry buildings”, SESOC Journal, 23(1), 19‐49.
Orduña, A., Roeder, G. and Peña, F. (2007). "Evaluación sísmica de construcciones históricas de
mampostería: comparación de tres modelos de análisis." Revista de Ingeniería Sísmica (In
Spanish), 77, 71‐88.
C4‐27
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
Oyarzo Vera, C. and Chouw, N. (2008). “Comparison of record scaling methods proposed by
standards currently applied in different countries”. Proceedings of the 14th World
Conference on Earthquake Engineering, Beijing, China , 12 ‐ 17 October.
Oyarzo Vera, C., McVerry, G. and Ingham, J. (2011). “Seismic zonation and default suite of
ground‐motion records for time‐history analysis in the North Island of New Zealand”.
Earthquake Spectra, Under Review.
Pallares, F. J., Ivorra, S., and Adam, J. M. (2009) "Monitoring masonry chimneys with operational
modal analysis." Proc., 3rd International Operational Modal Analysis Conference
(IOMAC'09), Ancona, Italy, 401‐407.
Pasticier, L., Amadio, C. and Fragiacomo, M. (2007). "Non‐linear seismic analysis and
vulnerability evaluation of a masonry building by means of the SAP2000 V.10 code."
Earthquake Engineering & Structural Dynamics, 37(3), 467‐485.
Pavic, A., Miskovic, Z., and Reynolds, P. (2007). "Modal testing of finite‐element model updating
of a lively open‐plan composite building floor." Journal of Structural Engineering, 133(4),
550‐558.
Ramos, L. F., Marques, L., Lourenço, P. B., De Roeck, G., Campos‐Costa, A. and Roque, J. (2010).
"Monitoring historical masonry structures with operational modal analysis: Two case
studies." Mechanical Systems and Signal Processing, 24(5), 1291‐1305.
Rosenblueth, E. (1951) A Basis for a Seismic Design. University of Illinois, Urbana, Illinois
Russell, A.P. (2010) “Characterisation and Seismic Assessment of Unreinforced Masonry
Buildings.” Department of Civil Engineering, University of Auckland, Auckland.
Russell, A. P. and Ingham, J. M. (2010). "Prevalence of New Zealand's unreinforced masonry
buildings." Bulletin of the New Zealand Society for Earthquake Engineering, 43(3), 192‐207.
Sacco, E. (2009). "A nonlinear homogenization procedure for periodic masonry." European
Journal of Mechanics ‐ A/Solids, 28(2), 209‐222.
Silva, J. M. M., and Maia, N. M. M. (1999). Modal Analysis and Testing, Kluwer Academics
Publishers, Netherlands.
Skolnik, D., Lei, Y., Yu, E. J., and Wallace, J. W. (2006). "Identification, model updating, and
response prediction of an instrumented 15‐story steel‐frame building." Earthquake Spectra,
22(3), 781‐802.
Smith, W. R. (1981) "Least squares time‐domain methods for simultaneous identification of
vibration parameters from multiple free‐response records." Proc., A1AA/ASME/ASCE/AHS
22nd SDM Conference, April, 194‐201.
Standards New Zealand (SNZ) (2004a). NZS 1170.5:2004 ‐ Structural Design Actions Part 5:
Earthquake actions. Standards New Zealand, Wellington, New Zealand.
Standards New Zealand (SNZ) (2004b). NZS 1170.5:2004 ‐ Structural Design Actions. Earthquake
actions ‐ Commentary. Standards New Zealand, Wellington, New Zealand.
C4‐28
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Standards New Zealand (SNZ) (2006). NZS 3101:2006, Concrete Structures Standard Part1: The
Design of Concrete Structures. Standards New Zealand, Wellington, New Zealand.
van Overschee, P., and de Moor, B. (1996). Subspace identification for linear systems: theory,
implementation, applications Kluwer Academic Publishers, Boston.
Yu, S., Wu, C., and Griffith, M. C. "Numerical analysis of out‐of‐plane loaded masonry wall using
homogenization technique." Proc., 14th International Brick and Block Masonry Conference
(14IBMAC), Sydney, Australia, 17 – 20 February.
C4‐29
Section C4 ‐ Design Actions on In‐plane Loaded Unreinforced Masonry
C4‐30
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C5 ‐ Determination of Diaphragm Design Actions
C5.2 Scope
The recommended parabolic load distribution and corresponding design actions are logically
based on the assumption that the diaphragm has uniform width (b), and therefore strictly only
apply to rectangular geometries. The capacity assessment of diaphragms with more complex
geometries will require a rational method to account for varying diaphragm width.
C5.3 Load Distribution
The parabolic load distribution is recommended by NZSEE (2006) and ASCE 41‐06 (2007)
guidelines when assessing the capacity of flexible diaphragms. It is intended to emulate the
expected distribution of horizontal inertia forces developed in the diaphragm.
C5.4 Design Actions
Diaphragm shear force and bending moment distributions have been developed by simple
integration of the load distribution function. These equations are simplified further when
determining maximum design actions from the knowledge that maximum shear force occurs at
the side of the diaphragm and maximum bending moment occurs at midspan.
C5.5 Worked Example
Establish the design actions for the diaphragm shown below.
10.4 m
N
5.535 m
Step 1: Previously Determined Values
Vd (NS) = 33 kN (using provisions of Section 4)
Vd (EW) = 35 kN (using provisions of Section 4)
L (NS) = 10.4 m
L (EW) = 5.535 m
C5‐1
Section C5 ‐ Determination of Diaphragm Design Actions
Step 2: Diaphragm Load Distribution
North South Direction
1.5 2
1
1.5 ∗ 33 2
1
10.4 10.4
4.76 1 0.037
East West Direction
Similarly it can be shown that:
9.49 1 0.131
Step 3: Diaphragm Design actions
North South Direction
Shear force distribution:
3 2
2
33 3 2
10.4 2 10.4
4.76 0.059
Maximum shear force:
33
,
2 2
VE,max (NS) = 16.5 kN
Bending moment distribution:
3 5
4 2 32
3 ∗ 33 33 5 ∗ 33 ∗ 10.4
4 ∗ 10.4 2 ∗ 10.4 32
Maximum bending moment:
C5‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
5 5 ∗ 33 ∗ 10.4
32 32
ME,max (NS) = 53.6 kNm
East West Direction
Similarly it can be shown that:
VE (EW) = 9.49x – 0.413x2
VE,max (EW) = 17.5 kN
ME (EW) = 4.74x2 – 0.103x4 – 30.26
ME,max (EW) = 30.3 kNm
C5.6 References
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, VA.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
C5‐3
Section C5 ‐ Determination of Diaphragm Design Actions
C5‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C6 ‐ Design Actions on URM Walls and Parapets
loaded out‐of‐plane
C6.1 Notations
bw Wall thickness, m
C(0) Site hazard coefficient for Tp = 0 sec, fraction of g
CHi Floor acceleration coefficient for level i
Ci(Tp) Part spectral shape factor at level i
Cp(Tp) Horizontal acceleration coefficient, fraction of g
g Acceleration due to gravity, m/s2
he Wall height measured between diaphragm restraints, m
hi Height of the wall mid‐height hinge measured from ground, m
hn Total height of the structure measured from ground, m
N(0,D) Near‐fault factor
Nt Axial load due to upper storeys, kN
R Return period factor
Rp Part risk factor
Tp Natural period of the wall, s
Vo* Design out‐of‐plane lateral force, kN
vo* Design out‐of‐plane lateral pressure, kN/m2
Ww Self weight of the wall, kN
Z Seismic hazard factor
γ Modal participation factor
C6.2 Scope
A linear static analysis procedure, the Equivalent Static Method, is presented for assessing
strength demands that an unreinforced masonry (URM) wall loaded out‐of‐plane will be
subjected to in a design level earthquake. Although URM response is highly inelastic and
nonlinear, the outlined detailed linear static method will provide a reasonable estimate of the
seismic demands on a URM wall when loaded out‐of‐plane. The procedure is adapted from the
NZS 1170.5:2004 (SNZ 2004) loading standard and the NZSEE (2006) guidelines for the
assessment of URM walls loaded out‐of‐plane. The calculated design out‐of‐plane force shall be
uniformly distributed over the wall height for the calculation of out‐of‐plane flexural demand.
The procedure is based on the assumption that a single horizontal crack will form at mid‐height,
which may not be true in some cases when masonry materials have a high flexural strength or
when any retrofit is applied such that it may redistribute the cracks on the tension face.
Therefore, the procedure shall be used with care for retrofitted URM walls and for URM walls
constructed with stronger mortars.
C6‐1
Section C6 ‐ Design Actions on URM Walls loaded Out‐of‐plane
C6.2.1 Procedure to Calculate Design Actions on Unreinforced Masonry walls
loaded out‐of‐plane
Figure C6‐1 shows a flow chart for the determination of seismic design forces for walls of
unreinforced masonry (URM) buildings, loaded out‐of‐plane, and having flexible diaphragms.
Initial assessment Detailed
based on percentage assessment not
NBS calculations required
Calculate self weight of
the wall and axial load
due to upper storeys
Calculate horizontal
acceleration coefficient ,
Cp(Tp), using Equation 13‐2
Calculate modal
participation factor in
accordance with
Section 13.3.2
Establish part risk factor,
Rp in accordance with
NZS 1170.5:2004
Calculate design lateral
force, Vo*, using
Equation 13‐1
Calculate design uniform pressure, vo*,
by equally distributing the total design
lateral force over the tributary area of
the wall i.e., vo*= Vo*/bwhe
Figure C6‐1: Flowchart for Out‐of‐plane Seismic Design of URM buildings with Flexible
Diaphragms.
C6‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C6.3 Design Lateral Force on Walls loaded out‐of‐plane
C6.3.1.1 Floor Acceleration Coefficient
Based on the experimental work performed it was established that a single horizontal crack
forms at or near the mid‐height of walls loaded out of plane (Derakhshan et al. 2011). Therefore,
the wall is assumed to form hinges at three locations, being at the top and bottom locations
where effective horizontal restraint is applied and at mid‐height between these top and bottom
hinges. Figure C6‐2 illustrates the definition of heights used in Section 6.3.1.
h
n
h
i
Figure C6‐2: Definition of heights
C6.3.1.3 Part Spectral Shape Factor
Equations for the natural period of a URM wall (Equation 6‐8) and parapet (Equation 6‐9),
loaded out‐of‐plane, have been extracted from the NZSEE (2006) guidelines for assessment of
out‐of‐plane loaded URM walls.
C6.4 Design Example
Calculate the design out‐of‐plane lateral seismic force for two leaf thick URM walls located at
level 2 and for the parapet. The building is founded on soil type D and is located in Christchurch.
Elevations are shown in Figure C6‐3.
C6‐3
Section C6 ‐ Design Actions on URM Walls loaded Out‐of‐plane
2
5
0
0
3
2
0
0 9
5
0
0
4
1
0
0
Figure C6‐3: Building cross‐section
C6.4.1 Design Lateral Force on Walls loaded out‐of‐plane
Considering a strip of 1 m width, the self weight of the wall is calculated as
Ww = Density Volume of Masonry
and the overburden weight is
Nt = 1.9 kN
he = 3.2 m
Site Hazard Coefficient (for Tp = 0 sec)
0 C 0 ZRN 0, D
Ch(0) = 1.12 for subsoil type D
Z = 0.30 for Christchurch from NZS 1170.5:2004
R = 1.0 for 1/500 APE
N(0, D) = 1.0 for D>20 km
C(0) = 0.336
Floor Acceleration Coefficient
1
6
for hi < 12 m
C6‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
hi= 5.7 m
CHi = 1.95
Part Spectral Shape Factor
2 1.75 where 0.5 < Ci(Tp) < 2.0
The first wall period is calculated as
0.67
1 2
0.67 3.2
1.9
1 2
12.4
Tp = 1.28 sec
2 1.75 1.28
Ci(Tp) = 0.94
Horizontal Acceleration Coefficient
C 0 C C T
Modal Participation Factor
γ 1.5
Part Risk Factor
Design Lateral Force on Out‐of‐plane loaded Walls
∗
γ
Or uniform lateral pressure, vo* = Vo*/he = 3.58 kN/m2
C6.4.2 Design Lateral Force on Parapet
Ww = 9.69 kN
Nt = 0 kN
C6‐5
Section C6 ‐ Design Actions on URM Walls loaded Out‐of‐plane
Site Hazard Coefficient (for Tp = 0 sec)
C(0) = 0.336 Same as calculated for wall loaded out‐of‐plane
Floor Acceleration Coefficient
1 for hi < 12 m
hi = 7.3 m
CHi = 2.22
Part Spectral Shape Factor
2 1.75 where 0.5 < Ci(Tp) < 2.0
The parapet period is calculated as
2.67 1
0.22
T 2.67 2.5 1
2.50
Tp = 2.59 sec
Ci(Tp) = 0.5
Horizontal Acceleration Coefficient
0
Modal Participation Factor
1.5
1.38
1
Part Risk Factor
Design Lateral Force on Parapet
∗
γ
Vo* = 0.37 × 1.38 × 1.0 × 9.69 = 4.95 kN per metre width of the parapet
Or uniform lateral pressure, vo* = Vo*/he = 1.98 kN/m2
C6‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C6.5 References
Derakhshan, H., Griffith, M. C., Ingham, J. M. (2011). “Out‐of‐plane behaviour of one‐way
spanning unreinforced masonry walls” ASCE Journal of Engineering Mechanics, submitted
June 2010, revised November 2010.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Standards New Zealand (SNZ) (2004). NZS 1170.5:2004 ‐ Structural Design Actions Part 5:
Earthquake actions. Standards New Zealand, Wellington, New Zealand.
C6‐7
Section C6 ‐ Design Actions on URM Walls loaded Out‐of‐plane
C6‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C7 ‐ Design Actions on Wall Diaphragm
Anchorages
C7.1 Notation
b Diaphragm width, m
Ci(0.4) Part spectral shape factor at level i
Cp(0.4) Horizontal acceleration coefficient, fraction of g
g Acceleration due to gravity, m/sec2
hi Height of the connection location, m
hn Total height of the structure, m
Rp Part risk factor
Vdi Diaphragm shear load transfer for level i, kN
Vcsi Design shear force for In‐plane loaded connections, kN/m
Vcti Design tension force for out‐of‐plane loaded connections, kN/m
Wwt1 Self weight of the top half of the out‐of‐plane wall below the connection, kN
Wwb2 Self weight of the bottom half of the out‐of‐plane wall above the connection, kN
C7.2 Scope
This section details recommendations for the determination of design seismic forces for the
connections between diaphragms and out‐of‐plane loaded or in‐plane loaded walls. The
following recommendations for calculating connection design actions are based on the
procedures defined in Section 4 and Section 6 which apply to URM bearing wall buildings with
the following characteristics:
(a) Flexible diaphragms at all diaphragm levels above the base of the structure;
(b) A maximum of six stories above the base of the structure;
(c) A minimum of two lines of vertical elements of the lateral load resisting system in each
principal direction, except for single storey buildings with an open front on one side.
C7.3 Design Shear for In‐plane Loaded Connections
The design shear load for the connections between the diaphragm and the in‐plane wall comes
directly from the diaphragm shear load transfer calculated in Section 4.
C7.4 Design Tension for Out‐of‐plane Loaded Connections
The process presented here is consistent with the Parts and Portions procedure of NZS 1170.5,
as no specific research has been performed to more accurately define the magnitude of these
actions. Oliver (2010) has questioned the validity of using the Parts and Portions process of
NZS 1170.5 for the assessment of design loads on connections and suggests that the values
C7‐1
Section C7 ‐ Design Actions on Wall Diaphragm Anchorages
obtained from this process are too conservative. More conclusive testing and modelling is
planned but for the interim the process outlined in section 7 is endorsed.
C7.5 Design Example
This design example calculates the tension forces in the connections between the diaphragm
and the out‐of‐plane loaded walls, and the shear forces in the connections between the
diaphragm and the in‐plane loaded walls. The building is a typical two storey isolated URM
building founded on soil type D and is located in Christchurch. The in‐plane length of the
building is 8 m. The elevation is shown in Figure C7‐1. The masonry density was established to be
ρm = 1800 kg/m3.
2
5
0
0
2
3
0
3
2
0
0 9
5
0
0
3
5
0
4
1
0
0
Figure C7‐1: Building cross‐section
Considering a strip of 1 m width , the self‐weight of the wall is calculated as
9.81 4.1
1800 0.35 1 12.7 kN Weight of top half of first storey
1000 2
9.81 3.2
1800 0.23 1 6.5 kN Weight of bottom half of second storey
1000 2
9.81 3.2
1800 0.23 1 6.5 kN Weight of top half of second storey
1000 2
C7‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
9.81
1800 0.23 2.5 1 10.15 kN Weight of parapet
1000
Calculate Design Shear for In‐plane Loaded Connections
Assuming that it was previously established that the diaphragm shear load transfer is:
Vd1 (Diaphragm level one) = 156 kN
Vdr = (Diaphragm roof level) = 112 kN
and assuming equal distribution of shear load between the two in‐plane loaded walls,
156
2 9.75 kN/m
8
112
2 7 kN/m
8
Calculate Design Tension for Out‐of‐plane Loaded Connections
Determine site hazard coefficient
0 0 0,
Ch(0) = 1.12 For subsoil type D
Z = 0.22 for Christchurch from NZS 1170.5
R = 1.0 for 1/500 probability
N(0,D) = 1.0 for probability > 1/250
C(0) = 0.246
Determine Floor Acceleration Coefficient
1 for all hi < 12 m (6‐4)
6
.
Ch1 = 1 1.683
.
Chr = 1 2.217
Determine Part Spectral Shape Factor
0.4 2
Determine Horizontal Acceleration Coefficient
0.4 0 0.4
C7‐3
Section C7 ‐ Design Actions on Wall Diaphragm Anchorages
Determine the Design Tension Force
0.4
C7.6 References
Oliver, S. J. (2010) “A design methodology for the assessment and retrofit of flexible diaphragms
in unreinforced masonry buildings”, SESOC Journal, 23(1), 19‐49.
C7‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C8 ‐ Response of walls loaded in‐plane
C8.1 Notation
Af Cross‐sectional area of flange, m2
Aw Cross‐sectional area of wall, m2
af Distance to centre of flange from compression edge of wall, m
ai Distance to centre of inertia of wall from compression edge of wall, m
bf Width of flange, m
bw Width of wall, m
c Bed‐joint cohesion, kPa
fdt Diagonal tension strength of masonry, kPa
f’m Compression strength of masonry, kPa
h Height of wall, taken as between diaphragms or similar restraint, m
ler Effective length of wall in rocking failure mode, m
letc Effective length of wall in toe crushing failure mode, m
lf Effective length of flange, to be the minimum of 6 × bf or actual length of flange, m
lw Length of wall, m
Nb Normal force acting on cross‐section of wall base, kN
Nt Normal force acting on cross‐section of top of the wall, kN
V Seismic Induced In‐plane shear force applied to top of wall, kN
Vdt Nominal shear capacity in diagonal tension failure mode, kN
Vr Nominal shear capacity in rocking failure mode, kN
Vs Nominal shear capacity in bed‐joint sliding failure mode, kN
Vtc Nominal shear capacity in toe crushing failure mode, kN
Wf Weight of masonry flange, kN
Ww Weight of masonry wall, kN
Z Distance from the compression edge of the wall to the line of action of
3
γm Unit weight of masonry, kN/m
µf Coefficient of friction of masonry
ρm Masonry density, kg/m3
σave Average compressive stress, kPa
Ψ Effective height factor
C8.2 Scope
Equations 8‐8 to 8‐12 have been reworked from research undertaken by Yi et al. (2008), being
an extension of earlier research into the verification of an effective pier model (Yi et al. 2005),
with the inclusion of flange effects. The effective pier model assumes a linear stress distribution
across the base of the wall that is a product of the axial load and the moment at the base of the
wall. The moment is caused by lateral seismic loading induced by diaphragm motions and wall
inertia grouped at the location of the diaphragm (top of the wall) in accordance with the
C8‐1
Section C8 ‐ of walls loaded in‐plane
procedures of Section 4. The model is used to find an effective length, le, of the wall, where any
masonry failing in tension is outside of the effective length (the flexural cracked area in Error!
Reference source not found.). The effective length of a wall when subjected to seismic loading is
the basis for the capacity equations for each of the four failure modes. The effective pier model
is shown in Error! Reference source not found., where the effective length of the wall at the
base is denoted le, the tensile strength of masonry is denoted ft, the lateral force is V and the
axial force is not shown. σave is the average axial stress at the point of inflection of the wall and
σnmax is the maximum compression stress, occurring at the compression toe of the wall.
Figure C8‐1: Effective pier model (Yi et al. 2008)
Equations for the capacity of walls loaded in‐plane are compared in Yi et al. (2005), with those in
FEMA 356 (2000), the prestandard of ASCE/SEI 41‐06 (2007), with very close correlation in
principles and values generated. Yi et al. (2008) conducted research into the effect of flanges on
in‐plane wall response, recognising that flanges have the potential to affect both the failure
mode induced by seismic loading and the strength of a URM wall (Tomazevic et al. 1993, Costley
and Abrams 1996, Paquette and Bruneau 2003, Yi et al. 2006, Moon et al. 2007). This
consideration of flange effects on in‐plane loaded wall, pier and spandrel capacities is why
equations in Yi et al. (2008) are endorsed in this document, being representative of the most
comprehensive models of URM wall behaviour.
Equations 8‐8 to 8‐12 are based on the assumption of a cantilevered URM wall, pier or spandrel,
but Yi et al. (2008) suggest that the results obtained from this assumption can be extended to
other, less conservative, boundary conditions by utilising appropriate aspect ratios.
C8‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C8.3 Properties of walls loaded in‐plane
C8.3.5 Diagonal Tensile Strength
Equation 8‐5 is sourced from the procedure presented in FEMA 356 (2000) for determining
masonry diagonal tension strength. The FEMA 356 equation includes a 0.75 reduction factor of
the average bed‐joint shear strength, νte. In Equation 8‐5 cohesion, c, is taken as equivalent to
νte. The reduction factor is included in the FEMA 356 equation as in‐situ shear resistance tests
used to determine νte include shear resistance provided by the collar joint between leafs, but
seismic shear is not transferred across the collar joint in a multi‐leaf masonry wall. Thus the 0.75
factor is not required in this guideline, as determining c does not take collar joint shear
resistance into account and consequently it is not necessary to reduce c.
The factor of 0.75 that is included in equation 8‐5 is a reduction factor and the 1.5 is a factor to
convert peak stress to an equivalent average stress for use with walls of a rectangular cross
section. These factors of 0.75 and 1.5 are consistent with the comparable equation in FEMA 356
(2000). In a rectangular wall subjected to shear force the shear stress distribution is not linear,
see Figure C8‐2 below. The shear stress at the ends of the wall is zero, and the maximum shear
stress is at the centre of the wall. The maximum shear stress is 1.5 (3/2) times the average shear
stress. Hence the 1.5 denominator in FEMA 273 Equation 7‐1 was to convert the maximum shear
stress capacity at the centre of the wall to an average value applicable over the length of
rectangular walls. For transparency, the strength reduction factor of 0.75 is shown at the front of
the equation and the denominator of 1.5 is shown separately.
Wall in plan
Shear force, V
Figure C8‐2: Shear stress distribution in a rectangular wall
C8‐3
Section C8 ‐ of walls loaded in‐plane
C8.3.7 Average Compressive Stress
In determining the normal force acting on the cross‐section of the top of the in‐plane loaded
wall, Nt, figure C8‐3 illustrates the tributary area of floor or roof diaphragm, and thus tributary
seismic weight from diaphragms above that is attributed to the wall. The boundaries of tributary
area extending from each corner of the floor plan can be assumed to be at 45° to the URM walls
(Russell 2010). The appropriate imposed loads shall be applied in accordance with NZS 1170
(SNZ 2004) with all levels above the wall being considered, including self weight of the in‐plane
loaded wall, when determining seismic weight acting on the wall, and hence normal forces Nt
and Nb.
(a) Long walls tributary area (b) Short walls tributary area
Figure C8‐3: Tributary areas for determining normal force on each wall (Russell 2010)
C8.4 Nominal Shear Capacity of walls loaded in‐plane
When subjected to in‐plane seismic loading, URM walls have been found to fail by one of four
different failure modes (Magenes and Calvi 1997) represented by individual equations in
Section 8. Descriptions of wall behaviour when loaded in‐plane, in each of the four failure
modes, are given as follows.
C8.4.1 Capacity in Diagonal Tension Failure Mode
Diagonal tension failure, illustrated in Figure C8‐4, is a force controlled failure mode and occurs
when the maximum principal tensile stress in the pier exceeds the diagonal tensile strength of
the masonry. Diagonal tension failure leads to the formation of inclined diagonal cracks that
commonly follow the path of bed and head joints through the masonry, or, when the relative
C8‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
strength of the mortar is high, can cause cracking through the bricks themselves. Both test
results (Russell 2010) and observed performance in earthquakes (Dizhur et al. 2010) indicates
that in New Zealand, stepped diagonal cracking at mortar joints can be expected due to brick
strength typically exceeding mortar strength (see also Table 2‐2 and Table 2‐4). A URM wall
failing in diagonal tension failure suffers a rapid degradation in strength after the formation of
diagonal cracking and ultimate collapse occurs at small and unpredictable magnitudes of drift.
Figure C8‐4: Diagonal Tension failure mode
The diagonal tension capacity equations suggested by Yi et al. (2008) have been amalgamated
into Equation 8‐8a by way of introducing two flange location factors, α1 and α2. The 0.9 factor
has been added to the equation consistent with recommendations by Russell et al. (2012) that
suggests a high level of correlation between the modified equation and experimental results.
Research completed by Russell et al. (2012) with regards to experimental testing of the in‐plane
capacity of three flanged URM walls is summarised in Table C8‐1, Table C8‐2 and Table C8‐3.
Table C8‐2 provides the wall dimensions and location of attached flange/s for the three walls,
the plan diagrams of which are illustrated in Table C8‐1. The ‘+’ and ‘‐’ are used to denote the
two directions of loading, which is important due to the asymmetrical nature of wall A7. All of
the walls failed in the diagonal tension failure mode. Experimental diagonal tension capacities
(Vmax) are compared in Table C8‐3 to capacities predicted by Yi et al. (2008), the NZSEE (2006)
guideline and ASCE/SEI 41‐06 (2007). The Yi et al. (2008) model has the closest correlation with
experimental results.
C8‐5
Section C8 ‐ of walls loaded in‐plane
Table C8‐1: Wall Schematics
Table C8‐2: Wall Specifications (Russell et al. 2012)
A6 230 2000 4000 0.7 0.441 both ends both sides 230 2160
A7 230 2000 4000 0.7 0.468 one end both sides 230 2160
A8 230 2000 4000 0.7 0.576 both ends one side 230 1200
Table C8‐3: Comparison of predicted strength accounting for and neglecting
flanges (Russell et al. 2012)
When assessing the capacity of a wall loaded in‐plane, in the diagonal tension failure mode,
flanges either in tension or compression shall be accounted for separately such that the nominal
shear capacity for the diagonal tension failure mode is the average of the capacity of the wall in
C8‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
diagonal tension calculated excluding all flanges and the capacities of the wall calculated with
each flange separately. In this instance all flange variables will be calculated accounting for each
flange separately. Note: A flange located at the centre of a wall, pier or spandrel was found to
have no effect on the diagonal tension capacity of the wall being considered and thus can be
excluded when the average of the diagonal tension capacities is being calculated by simply
counting twice the capacity of the wall excluding flanges. A flange located at either side of the
centre of a wall will alter the diagonal tension capacity of the wall (Yi et al. 2008) and hence
should be included in calculations. In this instance the direction of loading does not affect the
calculated wall capacity.
C8.4.2 Capacity in Rocking and Toe Crushing Failure Modes
The rocking failure mode represents overturning of a URM wall, pier or spandrel when subjected
to in‐plane seismic loading and is characterised by large flexural cracks at the bottom and top of
the wall, pier or spandrel. As the displacement increases, the wall, pier or spandrel rotates about
the compressive toe. When force reversal occurs the flexural cracks close and then flexural
cracking from the opposite direction occurs. As a result, rocking is generally considered
deformation (drift) controlled as large displacements can be withstood with little degradation of
strength.
Unlike rocking, the toe crushing failure mode leads to loss of stiffness and strength of a URM
wall, pier or spandrel. Toe crushing is a compressive failure of masonry that occurs at the toe of
the wall, pier or spandrel and initiates when the maximum compressive stress exceeds the
maximum compressive strength of the masonry. The extent of the toe crushing that occurs is
dependent upon the ratio of applied axial stress to axial strength.
Despite being considered separate failure modes, rocking and toe crushing are often intrinsically
linked such that they represent a single failure mode. In this instance failure is defined by
overturning of the wall and simultaneous crushing of the compressed toe of the wall (Figure
C8‐5). It is self evident that even for a scenario where rocking initiates, the magnitude of lateral
displacements necessary to cause wall overturning are of a magnitude such that total
overturning is unlikely. Rocking however causes a gradual degradation of the wall capacity that
can be correlated to the magnitude of deformation (drift) that the wall is subjected to, which
gives rise to the drift limits discussed in Section C8.5. Note: It is sensible to assume that if a wall
fails the rocking force check then it will also fail the rocking drift limit check.
When calculating the capacity of a wall in the rocking failure mode, it is important to consider
that when a wall possessing a non‐centrally located flange it will have a weak direction of
loading as ai will differ in the two in‐plane directions of the wall.
In typical New Zealand URM construction it is expected that the masonry compressive strength
will not be low enough to suffer toe crushing under average axial loads except for minor and
inconsequential crushing of a limited length of the toe of the wall where the compressive force
is concentrated (this is why the toe crushing capacity will always at least slightly understate the
rocking capacity of a wall). It is important to recognise that attaining the drift required to initiate
this crushing of the toe of the wall is infeasible. This expectation justifies neglecting the
C8‐7
Section C8 ‐ of walls loaded in‐plane
assessment of toe crushing capacity except in the instance where a wall has been retrofitted
with un‐bonded post tensioning or a seismic improvement intervention such that either or both
the axial stress on the wall and the wall capacity in the shear failure modes of bed‐joint sliding or
diagonal tension is increased. When calculating the toe crushing capacity of a URM wall the
flanges shall be accounted for individually as per calculation of diagonal tension capacity.
Figure C8‐5: Rocking/toe crushing failure mode
Equation 8‐9 is the same equation found in the NZSEE guidelines (2006), re‐arranged to give the
nominal shear force at rocking failure, instead of the nominal moment at rocking failure.
Equation 8‐11a is the toe crushing failure mode capacity given by Yi et al. (2008), with no
modifications and Equation 8‐11b is the same equation with all flange parameters equal to zero.
The 1.3 factor appearing in Equation 8‐12 is a direct substitution of the factor β from both Yi et
al. (2005) and Yi et al. (2008). β is inserted to take account of the erroneous assumption of a
linear stress distribution along the length of a wall, pier or spandrel and is taken as 1.3 in keeping
with the equivalent stress block analogy as outlined in MSJC 530‐02 (2002).
Yi et al. (2008) suggest that the effective length of the wall in the toe crushing failure mode shall
be given by Equation 8‐12a for walls, piers and spandrels with an attached perpendicular flange
or flanges and Equation 8‐12b for walls, piers or spandrels where no perpendicular flange is
attached.
C8.4.3 Capacity in Bed‐joint Sliding Failure Mode
Bed‐joint sliding occurs across a continuous bed‐joint and is identifiable by the formation of
horizontal cracks along the bed‐joint. Deformation of the pier by sliding along the bed‐joint
occurs such that the upper part of the wall is displaced relative to the lower part of the wall as
shown in Figure C8‐6. Bed‐joint sliding occurs when the average shear stress acting on the
effective section of the bed‐joint exceeds the maximum bed‐joint shear strength and is likely
when the axial loads acting on the wall are low. Due to the frictional resistance a large amount
C8‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
of energy is dissipated and it is typical for large deformations to be endured without significant
strength deterioration, hence bed‐joint sliding failure is a displacement controlled failure mode.
Figure C8‐6: Bed‐joint sliding failure mode
Increasing axial load and wall area increases the bed‐joint sliding capacity expected of a wall.
Equation 8‐10 ignores any increase to the bed‐joint sliding capacity due to the self weight of the
wall being assessed or the effect of any perpendicular flanges as it is possible, and conservative
to assume, that the failure plane across which the wall bed‐joint slides is located at the top of
the wall.
C8.5 Deformation Limits
Deformation limits, taken as drift and expressed as a percentage of the clear height of the wall
between points of restraint, have been reworked from research completed by Magenes and
Calvi (1997). The use of a bilinear approximation for the cyclic envelope is suggested due to the
largely non‐linear response of a URM wall and Figure C8‐7 shows a possible definition of
parameters of an appropriate bilinear curve. The bilinear approximation has been utilised to
determine the deformation limits seen in Table 8‐4.
C8‐9
Section C8 ‐ of walls loaded in‐plane
Figure C8‐7: Equivalent bilinear approximation (Magenes and Calvi 1997)
Failure of a URM wall, pier or spandrel in drift is defined as the drift relating to a degradation of
seismic load capacity to 80% of the maximum seismic load capacity of the element, Vmax as
shown in Figure C8‐7.
C8.6 Worked Examples
8 m 10 m
10 m
Figure C8‐8: Plan layout of building
A single storey stand‐alone URM building having a rectangular plan is illustrated in Figure C8‐8.
The wall height is 2.8 m. Scratch tests showed that the bricks have a compressive strength, f’b, of
26.9 MPa and the mortar has a compressive strength, f’j, of 5.5 MPa, which are masonry
properties considered average to New Zealand URM construction.
Determine masonry density:
1578 5 8
C8‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
1756.5 kg/m3
Determine masonry unit weight:
9.81
1756.5 9.81
17.23 kN/m3
Considering the 18 m wall at the rear of the building, the length of wall, lw, is 18 m. All walls are
two leaves thick and the width of the wall, bw, is 0.25 m. The wall (and flange) height, h, is 2.8 m.
Determine weight of wall:
17.23 0.25 18 2.8
217 kN
The wall has three flanges, hereafter referred to as left flange, centre flange and right flange.
The width of all flanges, bf, is 0.25 m.
Determine flange weights:
Left flange: Flange length, lf, is equal to the minimum of the actual flange length, 10 m, and 6
times the flange width, 1.5 m, and so is 1.5 m.
18.1 kN
Centre flange: Flange length, lf, is equal to the minimum of the actual flange length, 4 m, and 6
times the flange width, 1.5 m, and so is 1.5 m.
18.1 kN
Right flange: Flange length, lf, is equal to the minimum of the actual flange length, 10 m, and 6
times the flange width, 1.5 m, and so is 1.5 m.
18.1 kN
The normal force acting on the cross‐section of the top of the wall, Nt, is 900 kN (0.20 MPa).
Determine the normal force acting on the cross‐section of the bottom of the wall:
C8‐11
Section C8 ‐ of walls loaded in‐plane
With all flanges considered separately, and having identical weights:
With all three flanges considered together:
With no flanges considered:
217 900
1117 kN
Determine cohesion and coefficient of friction (In this case it is decided that no in‐situ testing is
necessary):
0.045
0.045 5.5
0.248 MPa
0.65
Determine the cross‐sectional area of the wall:
0.25 18
4.5 m2
Determine the average compressive stress on the wall:
The wall is considered cantilevered as the building is a single storey in height.
900
0.25 18
0.2 MPa
Determine diagonal tension strength of the masonry:
0.75 0.8
1.5
0.75 0.248 0.2 0.8 0.65
1.5
C8‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
0.176 MPa
The distance to the centre of inertia of the wall from the compression edge is required to be
determined for various cases of flanges.
Σ 0.5
Σ
Centre flange only, compression edge at left end of wall:
18.1 8 0.5 217 18
18.1 217
8.9 m
Centre flange only, compression edge at right end of wall:
18.1 10 0.5 217 18
18.1 217
9.1 m
All three flanges, compression edge at left end of wall:
18.1 0.125 18.1 8 18.1 17.875 0.5 217 18
18.1 3 217
8.9 m
All three flanges, compression edge at right end of wall:
18.1 0.125 18.1 10 18.1 17.875 0.5 217 18
18.1 3 217
9.1 m
Capacity in Diagonal Tension Failure Mode
The shear stress factor, ζ, is equal to 1.0 as per Table 8‐1 (h/lw is equal to 0.16).
Consider all flanges separately, and a case with no flanges, then take the average capacity.
Left flange:
For the case where the compression edge is at the left end of the wall, the first flange location
factor, α1, is equal to ‐1.0 and the second flange location factor, α2, is equal 1.0 to as per Table 8‐
2 (af is equal to 0.125 m). This is the same case as for the right flange where the compression
edge is at the right end of the wall.
1 1
1 . . 1 2. .
0.9 6 2
1 .
. .
2
C8‐13
Section C8 ‐ of walls loaded in‐plane
For the case where the compression edge is at the right end of the wall, the first flange location
factor, α1, is equal to 1.0 and the second flange location factor, α2, is equal 0 to as per Table 8‐2
(af is equal to 17.875 m). This is the same case as for the right flange where the compression
edge is at the left end of the wall.
Centre flange:
For the case where the compression edge is at the left end of the wall, the first flange location
factor, α1, is equal to ‐1.0 and the second flange location factor, α2, is equal 1.0 to as per Table 8‐
2 (af is equal to 8 m).
For the case where the compression edge is at the right end of the wall, the first flange location
factor, α1, is equal to 1.0 and the second flange location factor, α2, is equal 0 to as per Table 8‐2
(af is equal to 10 m).
C8‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
No flanges:
For the case where the compression edge is at the left end of the wall (same as for where the
compression edge is at the right end of the wall).
0.9
1
1042 kN
, , , , ,
, , , , , , ,
kN
Note that the direction of loading makes no difference to the wall capacity and so consideration
of the location of the compression edge could have been excluded from the calculations, but
was included here for completeness.
Capacity in Rocking Failure Mode
1
2 0.85
Determine the average masonry compressive strength:
. .
0.7
. .
0.7 26.9 5.5
13.8 MPa
Determine the average normal force acting on the cross‐section:
0.5
0.5 900 1171
1035.5 kN
C8‐15
Section C8 ‐ of walls loaded in‐plane
Determine the effective height:
Because this building is a single story in height, can be taken as 1.0.
1.0 2.8
2.8 m
Determine the distance from the compression edge of the wall to the line of action of :
Assuming acts uniformly over the cross‐section, we can find Z by taking moments of area
about the left end of the wall.
0.25 1.5 0.125 8 17.875 18 0.25 9
3 0.25 1.5 18 0.25
8.9 m from the left end of the wall
9.1 m from the right end of the wall
Determine the nominal shear capacity corresponding to the rocking failure mode:
It has previously been determined; that the compression block lies within the flange area
regardless of which direction of loading we consider, but in order to take a conservative and
simplified approach we will simply take as 0.25 m. It is also clear that the Z value of 8.9 m will
be critical.
1
2 0.85
1035.5 1 1035.5
8.9
2.8 2 0.85 13800 0.25
3226 kN
Alternatively if is taken as 1.75 m, i.e the effective flange area is considered in the
compression block, will equal 3282 kN.
Capacity in Bed‐joint Sliding Failure Mode
Neither the brick nor mortar compressive strengths are classified as “soft” in Table 2‐2 or Table
2‐4, hence the bed‐joint sliding failure mode need not be considered for this example. However
for completeness of these notes, the calculations for determining bed‐joint sliding capacity
follow.
0.75 0.8
1.5
0.75 0.25 18 248 0.8 0.65 1171
1.5
862.5 kN
C8‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Capacity in Toe Crushing Failure Mode
The wall in this example has not been retrofitted with either unbonded post‐tensioning or a
seismic improvement intervention that inhibits the formation of the diagonal tension failure
mode, hence the toe crushing failure mode need not be considered for this example. However
for completeness the calculations for determining toe crushing capacity of an un‐flanged wall of
identical dimensions have been included.
Determine effective length of the wall in toe crushing, with no flanges present (conservative):
2.
1.3. .
2 1117
1.3 13800 0.25
0.498 m
Determine toe crushing wall capacity with no flanges present (conservative):
1 1
. . .
2 3
1117 1 1
18 0.498
2.8 2 3
kN
Wall Capacity
min , , ,
min 799, 3282, 862.5, 3524
799 kN
The diagonal tension failure mode is critical for this wall.
References
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, VA.
Costley, A. C. and Abrams, D. P. (1996) “Dynamic response of unreinforced masonry buildings
with flexible diaphragms." NCEER‐96‐0001, University of Buffalo, Buffalo, New York.
Dizhur, D., Lumantarna, R., Derakhshan, H. and Ingham, J. M. (2010). “Testing of an earthquake‐
damaged unreinforced masonry building.” 7th International Conference on Urban
Earthquake Engineering and 5th International Conference on Earthquake Engineering,
Tokyo, Japan, 3 ‐ 5 March.
FEMA (2000). "Prestandard and commentary for the seismic rehabilitation of buildings." FEMA
356, American Society of Civil Engineers and Federal Emergency Management Agency,
Reston, VA.
Magenes, G., and Calvi, G. M. (1997). "In‐plane seismic response of brick masonry walls."
Earthquake Engineering and Structural Dynamics, 26(11), 1091‐1112.
C8‐17
Section C8 ‐ of walls loaded in‐plane
Masonry Standards Joint Committee (MSJC) (2002). Building code requirements for masonry
structures, ACI 530‐02/ASCE 5‐02/TMS 402‐02. America Concrete Institute, Structural
Engineering Institute of the American Society of Civil Engineers, The Masonry Society,
Detroit, United States.
Moon, F. L., Yi, T., Leon, R. T. and Kahn, L. F. (2007). "Testing of a full‐scale unreinforced masonry
building following seismic strengthening." Journal of Structural Engineering, 133(9), 1215‐
1226.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Paquette, J. and Bruneau, M. (2003) “Pseudo‐dynamic testing of unreinforced masonry buildings
with flexible diaphragm.” Journal of Structural Engineering. 129(6), 708‐716.
Russell, A. P. (2010). "In‐plane seismic assessment of unreinforced masonry buildings." PhD
thesis, Department of Civil and Environmental Engineering, The University of Auckland,
Auckland, New Zealand.
Russell, A. P., Elwood, K. J. and Ingham, J. M. (2012). "Force‐displacement response of
unreinforced masonry walls with flanges." Journal of Structural Engineering, Under review.
Standards New Zealand (SNZ) (2004). NZS 1170:2004 ‐ Structural Design Actions. Standards New
Zealand, Wellington, New Zealand.
Tomazevic, M., Lutman, M. and Weiss, P. (1993). “The seismic resistance of historical urban
buildings and the interventions in their floor systems: An experimental study.”The Masonry
Society Journal, 12(1), 77‐86.
Yi, T., Moon, F. L., Leon, R. T. and Kahn, L. F. (2005). "Effective pier model for the nonlinear in‐
plane analysis of individual URM piers." The Masonry Society Journal, 23(1), 21‐35.
Yi, T., Moon, F. L., Leon, R.T. and Kahn, L. F. (2006) “Quasi‐static tests of a two‐storey
unreinforced masonry building with flexible diaphragms.” Journal of Structural Engineering,
132(5), 643‐652.
Yi, T., Moon, F. L., Leon, R. T. and Kahn, L. F. (2008). "Flange effects on the nonlinear behaviour
of URM piers." The Masonry Society Journal, 26(2), 31‐42.
C8‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C8‐19
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C9 ‐ Strength and Deformation Assessment of
Flexible Timber Floor Diaphragms
C9.3 Lateral Deformation
Further guidance on the strength and deformation assessment of flexible timber floor
diaphragms can be found in Oliver (2010).
C9.3.1 Deformation Limits
While it has been proven experimentally that diaphragms of this nature can withstand
deformations in excess of 150 mm (see Figure C3‐5), this limit has been prescribed in accordance
with ASCE 41‐06 (2007) Clause 3.3.1.3.5 to ensure that the assumptions made for linear static
procedures are not violated. As illustrated in Figure C3‐5 and C3‐6 there is considerable
conservation in this assumption as experimental testing has confirmed that flexible timber
diaphragms typically have a substantial post‐yield stiffness as denoted k2 in Figure C3‐6.
The criterion based on wall thickness has been offered to reduce the risk of out‐of‐plane wall
collapse. Although this criterion has not been thoroughly scrutinised, preliminary analysis of out‐
of‐plane walls responding in the critical cantilever mode has indicated that diaphragm
displacements below half wall thickness are within the displacement capacity of the supporting
URM wall. Diaphragms exceeding this deformation limit might jeopardise the load carrying
capacity of the out‐of‐plane loaded walls and potentially induce a cantilever‐type wall failure.
Further analysis of diaphragm displacement limits is recommended as a subject for future
research.
C9.4 Strength
Diaphragm characteristic shear strength, Qn, is independent of diaphragm geometry. In order to
determine the total shear strength of the diaphragm, Rn, the characteristic shear strength must
be multiplied by the effective depth of the diaphragm, be.
The inclusion of a strength reduction factor, , preserves established design procedures while
coefficients αo and αc appropriately reduce diaphragm shear strength for penetrations and
configuration variations, respectively, as discussed in Sections C3.6.2 and C3.6.3.
C9.4.1 %NBS
It is recommended that %NBS be calculated to assess the performance of the existing diaphragm
against the standard required for a new diaphragm. Vd can be used in the absence of new
diaphragm performance requirements, as new diaphragm construction would be designed to at
least this value.
C9‐1
Section C9 ‐ Strength and Deformation Assessment of Flexible Timber Floor Diaphragms
C9.5 Worked Example 1
Determine the deformation and strength of the diaphragm shown below.
10.4 m
5.535 m
Step 1: Previously Determined Values
Vd (N‐S) = 33 kN
Vd (E‐W) = 35 kN
Kd (N‐S) = 639 kN/m
Kd (E‐W) = 1315 kN/m
Qn (N‐S) = 1.5 kN/m
Qn (E‐W) = 1.2 kN/m
be (N‐S) = 5.535 m
be (E‐W) = 10.4 m
μ = 6
αp = 1.0
αc = 1.0
Step 2: Diaphragm Deformation Calculation
North South Direction
33
μ 6
639
Δd(NS) = 0.3098 m
= 310 mm
East West Direction
35
μ 6
1315
Δd(EW) = 0.1597 m
= 160 mm
Step 3: Diaphragm Shear Strength Calculation
C9‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
North South Direction
= 8.3 kN
East West Direction
= 12.5 kN
C9.6 Worked Example 2
Determine the deformation and strength of the diaphragm shown below.
N
b1 = 9.1 m
b2 = 4.9 m
L = 12.2 m
Step 1: Previously Determined Values
Vd (N‐S) = 40 kN
Vd (E‐W) = 44 kN
Kd (N‐S) = 585 kN/m
Kd (E‐W) = 871 kN/m
Qn (N‐S) = 1.3 kN/m
Qn (E‐W) = 0.8 kN/m
be (N‐S) = 7.0 m
be (E‐W) = 12.2 m
μ = 5
αp = 1.0
αc = 1.0
Step 2: Diaphragm Deformation Calculation
North South Direction
C9‐3
Section C9 ‐ Strength and Deformation Assessment of Flexible Timber Floor Diaphragms
40
μ 5
585
Δd(NS) = 0.342 m
= 342 mm
East West Direction
44
μ 5
871
Δd(EW) = 0.253 m
= 253 mm
Step 3: Diaphragm Shear Strength Calculation
North South Direction
= 10.5 kN
East West Direction
= 14.6 kN
C9.7 Worked Example 3
Determine the deformation and strength of the diaphragm shown below.
L1 = 7.2 m L2 = 7.2 m
b2 = 6.2m
L1 = 12.4 m
b1 = 6.2m
L1 = 14.4 m
Step 1: Previously Determined Values
Vd (N‐S) = 46 kN
Vd (E‐W) = 52 kN
C9‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Kd (N‐S) = 323 kN/m
Kd (E‐W) = 889 kN/m
Qn (N‐S) = 0.8 kN/m
Qn (E‐W) = 1.3 kN/m
be (N‐S) = 12.4 m
be (E‐W) = 14.4 m
μ = 5
αp = 0.75
αc = 1.0
Step 2: Diaphragm Deformation Calculation
North South Direction
46
μ 5
323
Δd(NS) = 0.712 m
= 712 mm
East West Direction
52
μ 5
889
Δd(EW) = 0.292 m
= 292 mm
Step 3: Diaphragm Shear Strength Calculation
North South Direction
= 14.0 kN
East West Direction
= 13.0 kN
C9.8 References
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, VA.
Oliver, S. J. (2010) “A design methodology for the assessment and retrofit of flexible diaphragms
in unreinforced masonry buildings”, SESOC Journal, 23(1), 19‐49.
C9‐5
Section C9 ‐ Strength and Deformation Assessment of Flexible Timber Floor Diaphragms
C9‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C10 ‐ Strength and Deformation Assessment of
Timber Roof Diaphragms
C10.2 Scope
Roofing in URM buildings typically consists of corrugated light gauge steel sheathing fastened to
timber purlins that are supported by timber trusses spanning onto URM walls or beams (see
Figure C10‐1(a)). Where supported on the URM walls, the ends of the purlins and trusses are
typically seated into pockets in the walls. Figure C10‐1(b) provides an example of structural
sheathing fastened to the underside of a roof truss that would effectively behave analogous to a
floor diaphragm; however ceilings can often comprise light panels that are intended for
aesthetic purposes only and would not contribute to overall seismic response.
No research to date has quantitatively established the in‐plane performance of roofs in URM
buildings. As a result, the development of seismic assessment procedures explicitly for timber
roofs remains a subject for future research, and has not been included in this document.
However, due to their similarities, the material properties and the procedures to determine
lateral deformation and strength of timber floor diaphragms have been suggested as an
appropriate set of assessment procedures for timber roofs during the interim. Because obvious
differences exist between the structural configurations of timber floors and timber roofs, it is
recommended that the suggested guidelines be applied with consideration and diligence.
(a) Roof trusses (b) Ceiling sheathing
Figure C10‐1: Examples of roof configurations
C10‐1
Section C10 ‐ Strength and Deformation Assessment of Timber Roof Diaphragms
C10.3 Material Properties
It is recommended that the designer exercise conservatism during the selection of material
properties in order to account for the uncertainty involved in applying timber floor diaphragm
properties to timber roofs.
C10.4 Lateral Deformation
In the absence of research validation, it has been assumed that given the appropriate selection
of Kd and μ, lateral deformation can be determined using the same rationale as for timber floor
diaphragms. Future research will provide accurate values of Kd and μ for timber roofs. Refer to
Section C9.3 for further information.
C10.4.1 Deformation Limits
A deformation limit of 150 mm has been prescribed in accordance with ASCE/SEI 41‐06 (2007)
Clause 3.3.1.3.5 to ensure that the assumptions made for linear static procedures are not
violated.
The criterion based on wall thickness has been offered to reduce the risk of out‐of‐plane wall
collapse. Although this criterion has not been thoroughly scrutinised, preliminary analysis of out‐
of‐plane walls responding in the critical cantilever mode has indicated that diaphragm
displacements below half wall thickness are within the displacement capacity of the supporting
URM wall. Diaphragms exceeding this deformation limit might jeopardise the load carrying
capacity of the out‐of‐plane loaded walls and potentially induce a cantilever‐type wall failure.
Further analysis of diaphragm displacement limits is recommended as a subject for future
research.
C10.5 Strength
In the absence of research validation, it has been assumed that roof strength can be determined
using the same principles as for timber floor diaphragms. Accurate values of αp, αc, and Qn for
roofs will be established from future research. Refer to Section C9.4 for further information.
C10.5.1 %NBS
It is recommended that %NBS be calculated to assess the performance of the existing diaphragm
against the standard required for a new diaphragm. Vd can be used in the absence of new
diaphragm performance requirements, as new diaphragm construction would be designed to at
least this value.
C10.6 References
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, VA.
C10‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
C11.2 Scope
This section is based on laboratory and field testing and subsequent numerical analysis
conducted by Derakhshan (2011), and the assessment technique is suitable for walls and
parapets of single‐storey and two‐storey URM buildings in most areas of New Zealand. It is
estimated that this section covers an overall 86% of URM buildings in New Zealand (Russell and
Ingham 2010).
The NZSEE (2006) guidelines propose a detailed procedure for seismic evaluation of one‐way
spanning out‐of‐plane loaded URM walls, and the procedure required substantial calculations to
be conducted. Derakhshan et al. (2009) studied the guideline in detail and showed by
comparison with ASCE/SEI 41‐06 (2007) that the results were unreasonably conservative. For
instance, NZSEE (2006) indicated that walls having a slenderness ratio as low as 3 or 4 may be
categorised as earthquake‐risk depending on the soil type and seismic zonation. Derakhshan et
al. (2009) and Derakhshan (2011) further showed that a combination of parameters used in the
existing procedure led to erroneous results over a range of practical wall slenderness ratios.
The unreasonably conservative results obtained using the complicated method proposed in
NZSEE (2006) encouraged a simpler effective assessment method to be sought. Direct time‐
history analysis was used to investigate the behaviour of out‐of‐plane loaded URM walls when
subjected to various representative ground motion records. The obtained wall displacement
demand was compared with the wall displacement capacity assuming a displacement capacity
reduction factor of 0.6. The numerical analyses were conducted for both single‐storey and two‐
storey walls, recognising that Russell and Ingham (2010) found that approximately 86% of all
URM buildings in New Zealand are either one or two‐storeys tall. Further description pertaining
to the preparation of the assessment charts is provided in Section C11.4.
Four different diaphragm stiffness values of 100 kN/m, 500 kN/m, 1000 kN/m, and 2000 kN/m
were assumed in the numerical analyses conducted in Derakhshan (2011). The results of the
analyses for single‐storey walls were found to be insensitive to diaphragm in‐plane stiffness
values greater than 500 kN/m, and a diaphragm stiffness of 100 kN/m resulted in a decrease in
%NBS.Z of approximately 10% when compared to that obtained when assuming a diaphragm
stiffness value of 500 kN/m. The displacement demand on top‐storey walls dramatically
increased when the wall was connected to a diaphragm having a stiffness value of equal to
100 kN/m, when compared to the same data when a diaphragm stiffness of 500 kN/m or higher
was assumed. Based on these observations, it is recommended that horizontal diaphragms of
both single‐storey and two‐storey URM buildings be stiffened to a level of diaphragm in‐plane
stiffness equal to at least 500 kN/m. For the purpose of this section the diaphragm in‐plane
stiffness is calculated using Equation 3‐1.
Walls tests reported in Derakhshan (2011) and having a top support with vertical stiffness
demonstrated significant resistance to cracking that is a pre‐requisite to stable out‐of‐plane
rocking. Such walls may meet the out‐of‐plane seismic load demand in the elastic range, and
C11‐1
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
evaluation of the wall dynamic stability as conducted here may not be necessary. URM walls
connected to timber diaphragms or having diaphragm connections that were strengthened by
means of vertically inserted grouted steel bars were found to have a negligible effect in
developing arching action in the wall plane or in restraining wall out‐of‐plane displacements.
These observations suggested that the procedures recommended in this section are suitable for
these categories of walls.
Experimental research conducted in Australia during the last decade (Griffith et al. 2007) and in‐
field tests reported in Derakhshan (2011) showed that two‐way spanning URM walls have
significantly higher crack resistance than that expected from one‐way spanning URM walls. The
latter study showed that a few tested two‐way spanning one‐leaf URM walls located in
Wellington satisfied the requirements of the New Zealand seismic loading standard in the elastic
range. It is recommended that URM walls that are expected to behave in a two‐way condition be
evaluated by testing a few representative walls. Alternatively, analytical methods may be used
to estimate wall behaviour based on the material properties.
C11.3 Wall Thickness, Height, and Slenderness Ratio
Russell and Ingham (2010) suggested that a regular construction form reflected from the
relatively short span of masonry construction in New Zealand is observed in the existing URM
building stock, and that this regularity facilitates a typology study of such buildings. The most
frequently observed configuration of wall thickness in two‐storey URM buildings is that
containing a three‐leaf wall in the lower‐storey and a two‐leaf wall in the top‐storey. This
assumption has been made in various steps of the development of the assessment charts,
including in the calculation of overburden for lower‐storey walls and in the construction of the
numerical models.
In‐situ testing reported in Derakhshan (2011) showed that severe degradation of the weak lime‐
based plaster of URM walls occurred when the walls were subjected to out‐of‐plane cyclic static
uniform loading. Stronger cement‐based plaster layers underwent spalling and separated from
the original brickwork, suggesting that plaster layers should not be included in the wall thickness
when estimating the rocking behaviour of walls.
Seismic evaluation of parapets is conducted assuming an effective slenderness ratio, with the
value being calculated as four times the height of the parapet divided by its thickness. Research
by Lam et al. (1995), Doherty (2000), and Griffith et al. (2007) has shown that the seismic
behaviour of free‐standing URM walls is analogous to that of a four times more slender URM
wall responding in a simply‐supported condition. Detailed investigation of the NZSEE (2006)
guidelines show a similar trend, with the calculated %NBS obtained for parapets being virtually
the same as that obtained for a four times more slender simply‐supported wall. When using the
assessment charts presented in this section, a chart that corresponds to the actual thickness of
parapet shall be used.
For walls having the same slenderness ratio but different thicknesses, the taller and thicker wall
possesses a higher fundamental period, and thus larger displacement demand would be
expected for thicker walls. This effect of increased period is further accompanied by an increase
C11‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
in wall mass for thicker walls, resulting in higher inertial forces being associated with thicker
walls. However, the displacement capacity of out‐of‐plane loaded URM walls increases with
increasing thickness. Investigation of dynamic time‐history analysis results show that the
obtained %NBS for two‐leaf walls is on average 0.75 times the %NBS for a three‐leaf wall having
the same slenderness ratio. As the thickness ratio is 0.5 for a one‐leaf wall vs. a two‐leaf wall and
the thickness ratio is 0.65 for a two‐leaf wall vs. a three‐leaf wall, a larger reduction in %NBS is
expected for the former when compared to the latter. Based on this observation, it is
recommended that the %NBS for a one‐leaf wall is assumed as 0.6 times the %NBS for a two‐leaf
wall having the same slenderness ratio.
C11.4 Load‐bearing walls
The applied overburden for load‐bearing (LB) walls was assumed to be equal to 3.8 kN/m. The
weight of a diaphragm having dimensions of 8000 mm × 16000 mm was assumed in the
calculation of the overburden for LB walls, with the diaphragm seismic weight being calculated
assuming 0.35 kPa permanent (dead) load, 2 kPa imposed (live) load, and a live load reduction
factor of 30%. Diaphragm weight was assumed to be transferred entirely through two parallel
URM walls.
Walls that have an overburden load of less than 3.8 kN/m can be evaluated using this section by
interpolating results from the provided assessment charts. Extrapolation of results shall not be
used for overburden loads in excess of 3.8 kN/m. Research has shown that applied overburden
improves wall behaviour, but the increase in %NBS has a nonlinear relationship with the applied
overburden.
Estimation of the overburden for lower storey walls should be conducted excluding the weight
of the top‐storey walls as this weight has been previously accounted for in the development of
the assessment charts. Derakhshan (2011) used geometrical stiffness models that accounted for
increased wall strength due to the weight of the top‐storey walls, and the stabilising effect of the
top‐storey wall weight was taken into consideration when developing the assessment charts
included herein.
C11.5 Assessment
To produce the graphs presented in this section, the conclusions of the numerical analyses
conducted in Derakhshan (2011) were used. %NBS was averaged from time‐history analyses
conducted using a suite of 7 ground motion records recommended by Oyarzo Vera (2011) and
Oyarzo Vera et al. (2011). The suites of ground motion corresponded to 5 different zones for the
North Island of New Zealand (see Figure C11‐2) and had different characteristics that reflected
the seismological signature of the site being investigated. The time‐history records
corresponded to 5% damping ratio and an annual probability of exceedance of 1/500. Although
it had been intended to complete zonation maps for all of New Zealand, the study for the South
Island is currently in a preliminary stage. The numerical analysis conducted in Derakhshan (2011)
was limited to the North Island, but the results suggested that the study also addressed most of
the URM building stock in the South Island. The results of the time‐history analysis proved
C11‐3
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
insensitive to the seismic zones North A‐C and V (see Figure C11‐4), suggesting that any wall
located in either of zones A, B, C, or V can be evaluated using a single average curve as labelled
“Elsewhere” in Figures 11‐1 through 11‐4. Walls located in the near‐fault (NF) zone in the North
Island demonstrated a significantly different behaviour from that of walls located in any of zones
A, B, C and V.
All walls located in any region of the South Island except those located in areas that according to
NZS 1170.5:2004 have near fault characteristics are evaluated using the curve labelled as
“Elsewhere” in Figures 11‐1 through 11‐4. The assessment charts should not be used for regions
of South Island that have NF characteristics, but this limitation affects only a small percentage of
all URM buildings in New Zealand (Russell and Ingham 2010).
C11.5.1 Assessment of Parapets for Shear Capacity
Stable rocking of cantilever walls is subject to adequate shear capacity at their support. While
simply‐supported walls are not usually prone to out‐of‐plane shear failure due to sufficient
friction and axial loads, shorter cantilever parapets are likely to fail in shear especially if a damp
proof course (DPC) membrane has been used at the parapet base. URM parapets should
therefore be assessed for their shear capacity, and the assessment shall account for the effects
of DPC membrane that is often used at the parapet base. Doherty (2000) conducted tests on
URM walls with three forms of DPC construction, namely standard, centred, and slip joint. Static
testing suggested that the coefficient of friction varied from 0.1 to 0.6, with the average being
0.5. Shake table tests showed that the coefficient of dynamic friction was up to 20% less than
the coefficient of static friction. It is considered reasonably conservative to assume a static
coefficient of friction equal to 0.45 between DPC and masonry.
C11‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C11.5.2 Flowchart for using the Procedure for Seismic Assessment of Out‐of‐plane
Loaded URM Walls
Is diaphragm in‐plane
stiffness calculated using
Section 3.6 greater than
500 kN/m?
Yes No
Are there structural elements, e.g. beams, above Diaphragm should be
the top wall edge that restrict wall’s vertical stiffened, then go back
movements? to start of procedure.
Yes No
The procedure may produce Measure the wall clear height and
highly conservative results. the wall thickness excluding plaster
and renderings, calculate the
height‐to‐thickness ratio, and
estimate the applied overburden
from the dead and live load of the
diaphragm.
Establish seismic zone, hazard
factor, and soil type and use one of
the assessment charts provided as
appropriate. Interpolate the results
to account for the correct amount
of the applied overburden.
Figure C11‐1: Procedure for seismic assessment of out‐of‐plane loaded URM walls
C11‐5
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
Figure C11‐2: Zonation map for the North Island
Figure C11‐3: Zonation map for the South Island
C11‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C11‐4: Summary of North Island parametric study results (single‐storey 2‐leaf wall ‐
Shallow soil)
C11.6 Worked example
C11.6.1 Assessment of Walls in a Two‐storey Building in Gisborne
First‐storey height 6000 mm
Second‐storey height 4500 mm
Plan dimension 6000 × 12000 mm
Live load (both stories) 2 kPa
Dead load (both stories) 0.35 kPa
Live load reduction factor 0.3
Soil type Deep
Wall diaphragm connection Inspections suggests that no structural element with significant
vertical stiffness has been used on top of URM walls, but
grouted steel bars have been used to secure the top wall edge
to the timber diaphragm
Diaphragm joist direction Parallel to shorter building dimension
C11‐7
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
Diaphragm stiffness Calculated from Section 3.6 as 300 kN/m parallel to joists and
3000 kN/m perpendicular to joists
Diaphragm load transfer The timber diaphragm loads are transferred to longitudinal walls
Wall thickness The wall total thickness is 380 mm in lower storey and 260 mm
in top‐storey
Wall finishing A thin (15 mm) layer of cement plaster on all wall surfaces
(a) Check if the recommended procedure can be used for this particular buildings:
Diaphragm stiffness parallel to joists is less than 500 kN/m, and therefore stiffening of the
timber diaphragm should be undertaken in order for the procedure to be used. The existence of
steel anchorage on wall top edge is an acceptable condition that is within the scope of the
recommended procedure. Assuming that the timber diaphragms are stiffened to a level of in‐
plane stiffness greater than 500 kN/m and that walls act in a one‐way bending condition, the
procedure recommended in this section can be used to evaluate the walls.
(b) Estimating wall dimensions:
Wall dimensions may be estimated by use of Table C11‐1
Table C11‐1: Estimated wall dimensions
(c) Estimating overburden:
Estimation of the overburden is conducted excluding the weight of the top‐storey walls as this
weight has been previously accounted for in the development of the assessment charts.
Longitudinal walls: 0.5 x 6 x (0.3 x 2 + 0.35) = 2.85 kN/m
Short direction walls: 0.0 kN/m
(d) Seismic zonation and hazard factor (Z): Seismic zone is C and Z = 0.36
(e) Assessment of lower‐storey walls:
Assuming a slenderness ratio of 17.1, the following %NBS.Z values are obtained from
Figure 11‐4(b), and an interpolation of the results is used to determine the %NBS.Z value for
lower storey load‐bearing walls. The %NBS.Z value for non‐load‐bearing walls (parallel to short
dimension of the building) is directly obtained from Figure 11‐4(b). See Table C11‐2.
C11‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C11‐2: %NBS.Z for slenderness ratio of 17.1, lower storey
Wall slenderness Overburden
ratio
Interpolation for
0.0 kN/mm 3.8 kN/mm
2.85 kN/mm
The %NBS can be calculated for all walls by dividing the %NBS.Z by the hazard factor, thus:
Lower‐storey longitudinal walls: %NBS = 33.33/0.36 = 92.5%
Lower‐storey shorter walls: %NBS = 26.5/0.36 = 73.6%
(f) Assessment of top‐storey wall:
Repeating the procedure in the previous step but assuming a slenderness ratio of 19.6 and use
of Figure 11‐3(b), the values presented in Table C11‐3 are obtained.
Table C11‐3: %NBS.Z for slenderness ratio of 19.6, top‐storey
Wall slenderness Overburden
ratio
Interpolation for
0.0 kN/mm 3.8 kN/mm
2.85 kN/mm
The %NBS can be calculated for all walls by dividing the %NBS.Z by the hazard factor, thus:
Top‐storey longitudinal walls: %NBS = 17.38/0.36 = 48.3%
Top‐storey shorter walls: %NBS = 11.9/0.36 = 33.1%
C11.6.2 Assessment of Free‐standing Parapets in Various Seismic Zones for
Bending
Storey height 5000 mm
Parapet height 450 mm and 1000 mm
Parapet thickness 230 + 15 mm plaster
Axial load on top of parapet Zero (Non load bearing)
Soil type Deep
Cohesion for masonry 25 kPa
Masonry coefficient of friction 0.6
DPC membrane Present at parapet base
Weight density 18 kN/m3
C11‐9
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
(a) Calculating effective slenderness ratio
Table C11‐4: Estimated parapet dimensions
(b) Assuming slenderness ratios of 7.8 and 17.4 and using Figure 11‐1(b) (Deep soil), the values
of %NBS.Z are obtained for “Elsewhere”. Table C11‐5 is next formed to complete
assessment for regions without near fault effects. For comparison, the results when using
the NZSEE (2006) procedure for evaluation of parapets have also been included.
Table C11‐5: Calculating %NBS for regions excluding near fault effect
%NBS
(c) Assuming the same slenderness ratios and using Figure 11‐1(b) (Deep soil), the values of
%NBS.Z are obtained for “Near Fault” regions.
Table C11‐6: Calculating %NBS for regions including near fault effect
%NBS
C11‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(d) Assessment for shear capacity
Shear demand:
Using section 6:
Table C11‐7: Calculating shear demand
Shear capacity:
0.8
As DPC has been used at the base of parapet, the coefficient of friction shall be taken as 0.45,
and the cohesion is 0. Therefore:
Vn = 0.75*0.23*(0+0.8*18*0.45*h) = 0.1725*(6.48*h)
Table C11‐8: Calculating shear capacity
%NBS
Capacity,
Parapet kN Auckland Christchurch Gisborne Wellington
(e) Final assessment results are governed by shear capacity for short parapets, and the
assessment results of tall parapets are governed by bending capacity.
C11‐11
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
C11.6.3 Assessment of single‐storey and top‐storey one‐leaf walls in Christchurch
One‐leaf wall properties
Slenderness ratio is calculated, and the %NBS.Z value are obtained from assessment charts of
two‐leaf walls having the same slenderness ratio and in the same storey category:
Table C11‐9: One‐leaf wall assessment
C11.6.4 Evaluation of a Cavity Wall
A single‐storey cavity wall located in Christchurch is to be evaluated for out‐of‐plane wall
response. Site investigation shows that the wall leaves are bonded together with steel anchors,
which have been deteriorated.
Cavity wall properties
Property
Wall height 3500 mm
Wall thickness 115 x 2 + 50 (gap)
Axial load Zero
Soil type Deep
As there are no header courses, and there is almost 100% void ratio in the gap, the walls need to
be evaluated as single‐leaf walls. %NBS is calculated as that for the previous example, and the
value is not satisfactory (%NBS=27).
A good approach to strengthening is to increase the composite action between the inner and
outer wall leaves by using new anchors. Supplementary ties specifically designed for this
purpose are available from a number of proprietors, and advice on the size and spacing of ties in
C11‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C11‐10: One‐leaf wall assessment
C11.7 References
ASCE. (2007). Seismic rehabilitation of existing buildings, ASCE/SEI 41‐06, American Society of
Civil Engineers, Reston, Va.
Derakhshan, H., Ingham, J.M., and Griffith, M.C (2009). “Out‐of‐Plane Assessment of an
Unreinforced Masonry Wall: Comparison with NZSEE Recommendations.” Proc. 2009 NZSEE
Conference, Christchurch, New Zealand, April 3 ‐ 5.
Derakhshan, H. (2011). "Seismic assessment of out‐of‐plane loaded unreinforced masonry
walls." PhD thesis, Department of Civil and Environmental Engineering, The University of
Auckland, Auckland, New Zealand.
Griffith, M. C., Vaculik, J., Lam, N. T. K., Wilson, J., and Lumantarna, E. (2007). “Cyclic testing of
unreinforced masonry walls in two‐way bending.” Earthquake Engineering and Structural
Dynamics, 36(6), 801–821.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Oyarzo Vera, C. (2011). “FE‐based damage identification of unreinforced masonry structures.”
PhD thesis, Department of Civil and Environmental Engineering, The University of Auckland,
Auckland, New Zealand.
Oyarzo‐Vera, C., McVerry, G., and Ingham, J. M. (2011). “Seismic zonation and default suite of
ground motion records for time‐history analysis in the North Island of New Zealand.”
Earthquake Spectra, Submitted (081510EQS105M) August 2010.
Russell, A. P. and Ingham, J. M. (2010). "Prevalence of New Zealand's unreinforced masonry
buildings." Bulletin of the New Zealand Society for Earthquake Engineering, 43(3), 192‐207.
C11‐13
Section C11 ‐ Out‐of‐plane Wall and Parapet Response
C11‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C12 ‐ Wall‐diaphragm Anchorages
C12.1 Notation
acr i Minimum of et and sb for row “i”, mm
CF calibration factor
d fastener diameter, mm
fc,90y,5th% 5th percentile compression strength
fc,90y,avg Average compression strength
fh,1 Embedding strength corresponding to t1, MPa
fh,2 Embedding strength corresponding to t2, MPa
fh,5th% 5th percentile embedding strength
fh,avg Average embedding strength
fv member shear strength, MPa
G relative density of timber for the oven dry condition
k1 duration of load factor for strength
k12 factor for the design of bolted connections in green timber
Kls factor for member loaded surfaces
My fastener yield moment, Nmm
nfi number of fasteners in row “i”
nr number of rows in the joint as per load component
RSi min minimum (RS1, RS2, …, RSnr), N
RSi shear capacity along two shear planes of fastener row “i”, N
t member thickness, mm
t1, t2 timber thickness or fastener penetration of member 1 and 2, mm
Ratio of the embedding strengths
Strength reduction factor
C12.2 Scope
It is widely recognised that the single most significant feature responsible for the poor seismic
performance of unreinforced masonry buildings is the absence or inadequacy of connections
between elements, and primarily the connections between diaphragms and out‐of‐plane loaded
masonry walls and parapets. As illustrated in Figure C12‐1, deficient or absent anchorages were
a principal reason for the damage to unreinforced masonry buildings during the 2010 Darfield
(Canterbury) earthquake, along with inadequate restraint of parapets. Currently a
comprehensive study on bolt tear‐out has been completed, as reported in Section C12.6.
However, insufficient information currently exists to make comprehensive experimentally‐
validated recommendations on other aspects of connection assessment, and instead guidance in
the form of observed and probable failure modes is presented in the interim.
C12‐1
Section C12 ‐ Wall‐diaphragm Anchorages
(a) Out‐of‐plane wall failure due to (b) Remains of gable anchorage
connection failure
(c) No anchorage of diaphragms to wall (d) Punching shear failure of canopy
anchorage
(e) Anchorage failure at roof diaphragm (f) Close up of partial anchorage failure
connection
Figure C12‐1: Photographs of damage in the 2010 Darfield (Canterbury) earthquake due to
deficient or absent diagram anchorages
C12‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C12.3 Diaphragm Joist Seating
If anchor plates (also referred to as rosettes) are visible on the exterior of the wall then it is likely
that some form of mechanical connection exists between the masonry wall and the diaphragm
joists. However, where no anchor plates are visible on the exterior of the wall it will be
necessary to establish whether anchorages are embedded within the wall (see for instance
Figure C12‐2(a)) or are indeed absent (see for instance Figure C12‐1(c)).
As illustrated in Figure C3‐4(f), it is frequently found that there is no mechanical anchorage
between the diaphragm and wall, and that the diaphragm is merely seated on a ledge that is
developed by a reduction in wall thickness for ascending stories. An alternative seating
condition is referred to as a ‘joist‐pocket connection’, where the joist in seated on a recess that
is built into the wall as illustrated in Figure C3‐4(e). For the purposes of assessment a check may
be performed regarding the potential for bearing friction to sustain some transfer force between
the masonry wall and diaphragm joist, but as a general rule it should be assumed that the
connection is ineffectual and that the wall spans vertically over a greater height, including
potentially acting as a vertical cantilever from the ground.
The coefficient of static friction between timber and masonry varies from 0.2 to 0.6, and the
coefficient of dynamic friction is between 0.8 and 0.9 times the coefficient of static friction
(Cross and Jones 1993a, Cross and Jones 1993b). The vertical component of the ground motion
can further reduce the effective friction coefficient.
Considering the above factors, it is recommended to assume a lower bound of 0.15 and an
upper bound of 0.45 for the effective coefficient of dynamic friction. Assuming an effective
coefficient of dynamic friction equal to 0.3 is reasonable. The sum of the weight of floors and the
reduced live load on the floor should be used when calculating the friction force in the
connections.
C12.4 Masonry Anchorage Failures
Where anchor plates are visible on the exterior of the wall there are two failure modes that
should be considered, based upon failure patterns observed in earthquakes. The first of these
failure modes is a punching type shear failure as clearly shown in partial formation in Figures
C12‐1(e) and (f). Although no laboratory testing has yet been performed to adequately validate
a prescribed assessment procedure, observations suggest that a mortar failure surface should be
identified based upon the brick bond pattern and the wall thickness, and that the bed joint shear
capacity of this failure surface can then be established using the values prescribed in Equation 2‐
5 and Equation 2‐6. Care should be taken to account for the influence of pointing and etched
mortar joints due to degradation. Note also that observations tend to suggest that the failure
surface can be based upon the notion of a 45° failure cone as commonly assumed for punching
shear in concrete slabs, thus allowing a larger failure surface to be assumed than might
otherwise be adopted if only considering the geometry of the anchor plate. It is recommended
that only cohesion be relied upon for the head joints of the failure surface, and that no effort be
made to address the fact that in multi‐leaf walls the bond pattern that is visible on the exterior
of the wall will be offset in other wall leafs.
C12‐3
Section C12 ‐ Wall‐diaphragm Anchorages
Note that post‐earthquake observations have shown that the deployment of closely spaced
anchor plates, as shown in Figure 12‐2(b), can be both structurally adequate and aesthetically
acceptable, and consequently the use of such anchor plates is strongly encouraged as an
improvement technique. Where such a technique is deployed thought should also be given to
inspection and probably improvement of the surrounding mortar joints.
The second failure mode that should be addressed when considering the performance of
masonry anchorages is the possibility of a flexural failure being initiated along a single bed joint
at or directly above where multiple anchorages are embedded into a single bed joint. This
failure mode is primarily associated with parapets and it is recommended that parapets
restraints be provided to address this issue.
(a) Embedded anchorage that partially (b) Successful anchorages spaced close
restrained the wall together
Figure C12‐2: Examples of wall‐diaphragm anchorages that were partially or fully successful
during the 2010 Darfield earthquake
C12.5 Steel Anchorage Failures
Observations from the 2010 Darfield earthquake suggest that failure of the steel (or other metal)
connection linking the diaphragm and masonry wall is less likely than masonry anchorage failure,
as illustrated in Figure C12‐3(a) and Figure C12‐3(b) where anchors have been pulled out from
the masonry wall without rupture. However, failure of the connection due to either bar rupture
or timber tear‐out was observed as shown in Figure C12‐3(c) and Figure C12‐3(d). Because
access to the inside of the building was not available, the specific failure mode associated with
Figure C12‐3(c) and Figure C12‐3(d) was not established.
Currently there is inadequate information regarding the material characteristics of the steels
used in these anchorage types, but a simple check of the expected axial stresses in these
connections should be sufficient to determine if steel failure merits closer scrutiny. Particular
attention should be given to the effective cross‐sectional area should the steel exhibit signs of
corrosion.
C12‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(c) Detached anchor (d) Detached anchor
Figure C12‐3: Examples of failed anchorages where steel shows no indication of failure
C12.6 Timber Anchorage Failures
This section provides information to assess the capacity of bolted timber connections for existing
indigenous New Zealand hardwoods that were typically used to construct the floor and roof
diaphragms in New Zealand unreinforced masonry buildings. Representative details of this
connection type are shown in Figure C12‐4 and Figure C12‐5.
C12‐5
Section C12 ‐ Wall‐diaphragm Anchorages
Figure C12‐4: Joist oriented parallel to URM wall
Figure C12‐5: Joist oriented perpendicular to URM wall
C12.6.1 European Yield Model
The European Yield Model, which considers ductile failure modes of bolted connections, is
associated with Johansen’s theory. This theory is based on the assumption that the materials
(i.e. timber under embedding stresses and fastener under bending action) behave as rigid‐plastic
(Blass, 2003). The embedding strength of timber, fh2, was determined experimentally as
described in Section C6. For the embedding strength of steel plate, fh1, the value is assumed to
be equal to C1 b 3.2 fup, where C1 is the reduction factor determination of the design bearing
strength of the steel plate as per Table 12.9.4.3 of NZS 3404:Part 1:1997, b is the strength
reduction factor equal to 0.9 as per Table 3.3(1) of NZS 3404:Part 1:1997, and fup is the tensile
strength of the steel plate. The fastener yield strength, fyf, is obtained from the grade of
fasteners. The first figure from the grade indicates the fastener tensile strength, fuf,. The second
C12‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
figure specifies the fastener yield strength proportional to the tensile strength in terms of
percentage.
C12.6.2 Row Shear Failure of Timber
Many recent studies (Quenneville and Mohammad 2000; Mohammad and Quenneville 2001;
Quenneville and Bickerdike 2006; Quenneville et al. 2006; Quenneville 2009) have identified the
row shear failure of connections in timber structures. The observed brittle failure modes are
illustrated in Figure 12.6. An equation to predict the ultimate strength of timber connections
based on the row shear failure mode observed during tests was developed by Quenneville
(Quenneville and Mohammad, 2000). An optimised Row Shear Model equation to predict the
strength of bolted connections for the row shear failure mode when occurring in New Zealand
hardwood was obtained by using the calibration factor of 4 (Abdul Karim, 2010).
Figure C12‐6: Observed row failure modes for timber connections loaded parallel to grain.
C12.6.3 Timber Member in Bearing
Ab is the contact bearing surface area between the timber member and the masonry units, and
the strength reduction factor, ϕc, of 0.8 can be found in Clause 2.5 of NZS 3603: 1993.
C12.6.4 Washer Bearing
Aw, is determined in Clause 4.4.1.2 of NZS 3603: 1993, nw is the number of washers, and the
strength reduction factor, ϕc, of 0.8 can be found in Clause 2.5 of NZS 3603: 1993.
C12.7 Examples
C12.7.1 Diaphragm joint seating example
From C6.4.2:
Vo* = 17.86 kN per metre width of the wall
C12‐7
Section C12 ‐ Wall‐diaphragm Anchorages
Tributary normal force from Dead Load of diaphragm on wall connections:
N = 1 (kPa) x 4 (half the width of diaphragm) = 4 kN per metre width of the wall
Tributary normal force from reduced Live Load of diaphragm on wall connections:
N = 4 (kPa) x 4 (half the width of diaphragm) x 0.3 = 4.8 kN per metre width of the wall
Coefficient of dynamic friction = 0.3
Total resisting force due to friction:
φVn = 0.75 x (4+4.8) x (0.3) = 1.98 kN
%NBS = 100 x 1.98/17.9 = 11%
C12.7.2 Masonry anchorage example
A 2 leaf masonry parapet having a height of 1600 mm and a thickness of 230 mm is located
above a line of anchorage plates. The anchor plates have a diameter of 180 mm and are spaced
at 2.0 m intervals, representing 1 anchor plate for every 5 joists. The masonry is assessed to
have a density of 1680 kg/m3. The mortar is assessed to be of lower characteristic mean
hardness (f’j = 3.2 MPa from Table 2‐4). The bricks have a length of 230 mm and a height of
90 mm, with a 10 mm mortar joint. Pointing is measured to be 10 mm on the exterior surface of
the wall.
The vertical axial stress at the bed‐joint due to the above parapet is:
σv* = 1680 x 9.81 x 1.600 / 1000 = 26.4 kPa
From section 2, C = 0.45 f’j and the dynamic coefficient of friction is 0.8 x 0.65
Bed joint shear strength τbj = 0.045 x 3.2 x 1000 + 0.8 x 0.65 x 26.4 = 157.7 kPa
Head joint shear strength τhj = 0.045 x 3.2 x 1000 = 144.0 kPa
Figure C12‐7: Assumed failure plain for punching shear failure.
From Fig. C12‐7 it is established that:
The horizontal length of the failure plane is 6 bricks @ 240 mm = 1.44 m
The vertical length of the failure plane is 8 bricks @ 100 mm = 0.80 m
C12‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Accounting for pointing the assessed strength of the joint is:
φVn = 0.75 x 0.22 x (1.44 x 157.7 + 0.80 x 144.0) = 56.5 kN
Assuming joist demand as for the previous example and neglecting ledge friction capacity:
%NBS = 100 x 56.5/(2 x 17.9) = 158%
Including the capacity of the 4 joists supported by ledge friction:
%NBS = 100 x (56.5 + 2 x 1.98)/(2 x 17.9) = 169%
C12.8 References
Abdul Karim, A. R. (2010). "Seismic Assessment of Wall‐Diaphragm Connections in New Zealand
URM Buildings." PhD thesis, Department of Civil and Environmental Engineering, The
University of Auckland, Auckland, New Zealand.
Australian/New Zealand Standard (2000). AS/NZS 1080.3: 2000 Timber ‐ Methods of Test ‐
Method 3: Density. Standards New Zealand (electronic copy).
Australian/New Zealand Standard (2007). AS/NZS 4063.1: 2007 Structural timber ‐ Characteristic
values of strength‐graded timber ‐ Part 1: Test methods. Standards Australia (draft only).
Blass, H. J. (2003). "Joints with Dowel‐type Fasteners." Timber Engineering, S. Thelandersson and
H. J. Larsen, eds., John Wiley & Sons Ltd., England, 315‐331.
Cross, B. W., and Jones, N. P. (1993a). "Seismic performance of joist‐pocket connections. I:
Modelling", Journal of Structural Engineering, 119(10), 2976‐3007.
Cross, B. W., and Jones, N. P. (1993b). "Seismic performance of joist‐pocket connections. II:
Application", Journal of Structural Engineering, 119(10), October, 1993, 3008‐3023.
Mohammad, M., and Quenneville, J. H. P. (2001). "Bolted Wood‐Steel and Wood‐Steel‐Wood
Connections: Verification of a New Design Approach." Canadian Journal of Civil Engineering,
28(2), 254‐263.
Quenneville, J. H. P., and Mohammad, M. (2000). "On the Failure Modes and Strength of Steel‐
Wood‐Steel Bolted Timber Connections Loaded Parallel‐To‐Grain." Canadian Journal of Civil
Engineering, 27(4), 761‐773.
Quenneville, J. H. P., Smith, I., Aziz, A., Snow, M., and Ing, H. C. (2006). "Generalised Canadian
Approach for Design of Connections with Dowel Fasteners." CIB‐W18 meeting Proceedings,
Florence, Italy, paper CIB‐W18/39‐7‐6.
Quenneville, P. (2009). "Design of Bolted Connections: A Comparison of a Proposal and Various
Existing Standards." Journal of the Structural Engineering Society (SESOC) New Zealand Inc.,
22(2), 57‐62.
C12‐9
Section C12 ‐ Wall‐diaphragm Anchorages
Quenneville, P., and Bickerdike, M. (2006). "Effective In‐Row Capacity of Multiple‐Fastener
Connections." CIB‐W18 meeting Proceedings, Florence, Italy, paper CIB‐W18/39‐7‐1.
Standards New Zealand (1993). NZS 3603: 1993 Timber Structures Standard. Standards New
Zealand, Wellington, New Zealand.
Standards New Zealand (1997). NZS 3404: Part 1: 1997 Steel Structures Standard. Standards New
Zealand, Wellington, New Zealand.
C12‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C13 ‐ Heritage Characteristics of URM Buildings
and the Impacts of Seismic Improvements
C13.1 Introduction
A heritage building is more than simply a collection of building materials. As buildings age, they
become more permanent parts of local culture. They become associated with notable people;
with the memories of the public; and places where important historic events occurred. They
become unique, as buildings are simply no longer made the same way, and time provides them
with characteristics that could never be designed into a new building, such as the patina of age –
how a building looks old. Old buildings can become ‘heritage’, where they become important to
the public rather than just to the owner and occupants; a recent example of this is the birdcage
hotel in Figure C13‐1. There is no line on one side of which lies ‘heritage’, and on the other ‘not
heritage’; all old buildings will have some heritage value. The difficulty is in determining the
degree of heritage value, and what is important and what is less important about a building.
New Zealand’s URM historic buildings form a rapidly diminishing aspect of the nation’s collective
heritage, which is a physical link to the past and provides valuable evidence of cultural
progression. There is a real danger that if insufficient consideration is given to the historic and
architectural qualities of these buildings, seismic retrofit interventions may cause visual damage
or remove important historic material. Indeed, this has already occurred many times in recent
history (see Christianson 1983, Evans 1983, Leuchars 1989, Taylor 1983, Toomath 1979 for
example). Some of these retrofits, while possibly resulting in safer buildings, have had an overall
detrimental effect on the buildings themselves.
The historic value of the surviving URM building stock is slowly becoming recognised (Trapeznik
2000) and some buildings have been protected from alteration by heritage legislation and
organisations. The Historic Places Trust (HPT) is required by the Historic Places Act 1993 to
maintain a register of recognised heritage buildings, and acts in the interests of their
conservation when any significant alteration is proposed. (NZ Historic Places Act 1993) However
the HPT has limited resources, so only the most historically and architecturally significant
buildings tend to be protected. Many other URM buildings also have significant historic value
and contribute greatly to the character of towns and cities, but are offered no protection from
demolition or significant alteration.
As highlighted by the 2010 Darfield earthquake, many older buildings, particularly those
constructed from unreinforced masonry, now require seismic improvement in order to
withstand earthquake forces. There are many approaches and techniques to seismic
improvement, the most common of which are outlined below and analysed in terms of the
impact of each on the heritage and architectural value of the building.
C13‐1
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐1: The Birdcage hotel prior to seismic improvement and being moved
Once it has been accepted that seismic improvement of a building is required, it is necessary to
determine the extent of required intervention and the appropriate methods for that particular
building. Generally, an analysis is undertaken in stages, beginning with assessing the
architectural character of the building, then addressing the basic known inadequacies of URM
construction (Russell & Ingham 2010), then determining what types, if any, of additional retrofit
systems are required. Additional requirements should be met by structural elements which are
sensitively designed and in keeping with the architectural character of the building, of the least
extent, and reversible if possible (Goodwin et al. 2009). Often, several or many different
techniques will be used within one building, and it is vital to have an appreciation of how each
technique will impact upon the architectural and historic aspects of the building, and the ways in
which these impacts can be minimised.
C13.2 Heritage and Conservation
In New Zealand, acts which seek to prolong the useful life of a building while respecting its
history and originality are known collectively as conservation, and seismic retrofitting falls, or by
virtue of its conservation goals should fall, under this category. A successful act of conservation
will ensure that a heritage building is treated with care in order to preserve its historic fabric and
aesthetics to the greatest extent possible. New Zealand has its own particular history of
development of heritage principles, and its own set of unique problems in conservation. This
history and the present day environment resulting from these particular issues have been
amalgamated with broader international principles and adapted specifically for use with New
Zealand buildings in a document produced by the International Council on Monuments and Sites
(ICOMOS) New Zealand Charter for the Conservation of Places of Cultural Heritage Value
(ICOMOS 2010). This Charter forms the current foundation of heritage conservation in New
Zealand.
Seismic improvement of unreinforced masonry (URM) buildings is likely to require invasive work
and additional structure which could fundamentally change the character of a building. Before
engaging in any such work, it is important to understand the underlying principles of good
conservation practice in order to ensure that buildings and sites are sensitively treated and their
C13‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
future integrity can be assured. It is also important to understand that there is more to a
building than simply its physical material. This intangible component is comprised of concepts
such as its history and social use, and spiritual significance of the place, which are essentially
what the building means to an observer or group at a point in time. When combined with the
physical elements, these associations are collectively known as its heritage value (Aplin 2002).
C13.2.1 Heritage
Heritage as a concept is the collection of things which relates people to who they are, where
they have come from, and why they are the way they are. The cultural aspects of heritage relate
to human activity, namely; historical, archaeological, architectural, aesthetic, scientific, spiritual,
social, traditional, or some other special cultural significance (ICOMOS 2010). These aspects
represent the ideals of people and help to define who they think they are, so heritage can be
said to be a collection of items which represent individual and group histories (Aplin 2002). This
knowledge is important in order to gain an appreciation of sense of place and culture, and
enables people to better understand the greater physical and cultural environments in which
they live. Heritage can also be a useful way of defining identity, as well as serving as a cultural
record, and implies a gift for future generations and for the benefit of communities, in terms of
both physical and cultural inheritance. Heritage can be said to be a mark left by a bygone era
which one can identify with, respect, and which can help people to understand their history.
The influence of cultural heritage is great; it links people to the past, and gives them a sense of
place and progression. However, its influence is also subtle and intangible, and has often not
been given the recognition which it deserves. There have been a great many theories on the
treatment and protection of cultural heritage, particularly buildings and artworks, throughout
history. Some of these have been considerate and respectful, and others destructive and
oblivious. An example of the type of strengthening thought typical of the late 1970’s can be
found in Smith (1983). Debates on the importance of cultural heritage and its deserved
treatment continue to this day.
C13.2.2 Conservation
Everything which is made has a relationship to the past, as something has always come before
which has informed it or set a precedent. As time passes, buildings of a certain type or character
become increasingly rare, and as such their social value for heritage or scientific purposes
increases. The ICOMOS New Zealand Charter explains that the purpose of conservation is to
‘…care for places of cultural heritage value, their structures, materials and cultural meaning
(ICOMOS 2010).’ In general, such places:
(a) have lasting values and can be appreciated in their own right,
(b) inform us about the past and the culture of those who came before us,
(c) provide tangible evidence of the continuity between past, present and future,
(d) underpin and reinforce community identity and relationships to ancestors and the land; and
C13‐3
Section C13 ‐ Heritage Characteristics of URM Buildings
(e) provide a measure against which the achievements of the present can be compared
(ICOMOS 2010).
Conservation encompasses all acts which prolong the life of cultural and natural heritage with a
view to ensuring that it is able to be appreciated by future generations, and maintained in a
state in which it can continue to communicate heritage value. For a building, conservation
ought to preserve (and potentially enhance) the historic narrative, allow it to have a continuing
useful life, and retain its value for its owner and the community into the future (Holman 2000).
Thus it becomes important to evaluate the building or site in order to discover just what its
heritage value is, and consider taking action to conserve it if required. There are several main
factors which contribute to overall heritage value, all of which relate in some way or another to
the relative rarity or uniqueness. Bowron and Harris (1994) defined these values below:
Social: The notion of a spiritual, traditional, political, national, or any other cultural
sentiment expressed by a group.
Aesthetic: Considers the formal qualities of the fabric and setting; the form, scale, materials,
space etc. It addresses the design and cultural aspects of the place.
Scientific: Is concerned with the importance of the place as evidence and with the physical
survival of that evidence in the fabric. Scientific value is the potential to provide
information about past human activity. This may encompass technology, archaeology,
philosophy, custom, taste and usage as well as technique or material.
Historic: Is the ability to demonstrate an association with persons, ideas, or events. It
includes the history of all the above concepts (Bowron & Harris 1994).
These values apply to parts of sites and buildings as well as to the whole, so a particularly
beautiful historic staircase will have strong heritage value, which will add to the site as a whole;
even if the building which contains the staircase is of comparatively low value.
The NZ Charter outlines conservation principles which meet the unique cultural requirements of
New Zealand and suit its particular stock of historic sites (ICOMOS 2010). To date this is the best
and most comprehensive document relating to conservation principles in New Zealand and
should be considered as the basis for all conservation work while it remains so. It is important to
note however, that the NZ Charter is a series of guidelines rather than mandates; almost all
heritage projects will require some degree of compromise in order to arrive at a practical and
practicable solution.
All conservation projects should have their cultural heritage value identified and defined in order
to establish which aspects are most important. On larger scale projects, this indentification is
usually contained within a Conservation Plan, which is a document authored by a conservation
professional which serves to also suggest strategies for protecting these important elements,
and for managing potential problems. There are a number of general principles which guide the
approach to conservation, outlining the responsibilities of the conservator when planning
conservation work; in essence describing the desirable traits of an act of conservation. The
outcome of a conservation project should be consistent with the cultural content of the place.
Thus, relevant influence and cultural factors shall be considered in the planning stage and the
C13‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
scope and scale of the result should not be greatly removed from the scope and scale of the
original structure. Conservation should:
Make use of all relevant conservation values, knowledge, disciplines, arts and crafts;
Show the greatest respect for, and involve the least possible loss of, material of cultural
heritage value;
Involve the least degree of intervention consistent with long term care and the principles of
this charter;
Take into account the needs, abilities, and resources of the particular communities; and
Be fully documented and recorded (ICOMOS 2010).
In addition to these principles, the conservator ought to respect a building’s alterations and
additions as they are part of the ongoing history of the building, and should only be removed
where they significantly inhibit its heritage value or meaning. This respect for a building’s
history also extends to evidence of age; it is desirable for an old building to look old, and as such
the marks of age should be considered as integral with the fabric of the building. Sometimes
alterations will detract from the heritage value of a place, but this cannot be said categorically
and investigation should always be done. Similarly, the contents and setting of a building will
often take on heritage value of their own and as such need to be considered as integral with the
building for purposes of conservation (ICOMOS 2010). Risks from natural processes or events
need to be assessed, and plans formulated for their mitigation (ICOMOS 2010). These plans
might range from ground subsidence to fungal attack to earthquake, all of which can have
damaging effects on historic buildings. This latter aspect is particularly pertinent to this section,
which aims to minimise the adverse effects of earthquake strengthening work upon heritage
buildings.
All new work on an historic site should be distinguished as being new work. This
recommendation ensures that future generations examining the building will have no difficulty
determining which parts of a building are original and which are later additions or changes. The
materials and style of alterations ought to be significantly different in order to best express their
separation in time. This work also should be reversible to allow for imperfect solutions (Price et
al. 1996). The ideal solution will almost never be arrived at, either due to lack of technology or
due to funding. If at some time in the future this solution is possible, it ought not to be
prejudiced by a previous solution.
The reality of conservation is that experts are only consulted some of the time; public bodies and
some developers will commission conservation plans, but many projects go ahead without the
benefit of these. While New Zealand possesses a series of principles in the ‘New Zealand
Charter’ that are well‐founded in established international theory, the sympathetic treatment of
heritage is still not guaranteed by legislation or by popular opinion. Many property owners and
developers are either ignorant of these principles or choose to ignore them, and until they are
universally adopted there will continue to be demolition and destruction of valuable cultural
heritage. Current legislation is inadequate to fully protect heritage buildings or ensure that all
changes are undertaken in the spirit of the Charter.
C13‐5
Section C13 ‐ Heritage Characteristics of URM Buildings
C13.2.3 Historical Approaches
Advances in seismic engineering in New Zealand between about 1960 and 2000 have tended to
focus primarily on maximising life preservation and minimising costs (Cattanach et al. 2008).
This neglect of the building itself has led to a mindset where the building is often considered to
be a liability, where interventions have therefore been undertaken without consideration of the
architectural impacts of the improvement measures, or preservation of the building fabric after
the event of an earthquake (Cattanach et al. 2008). Rather than working with the materials
which are present, designers have found it easier to add a modern structure around the original
building, often with inadequate consideration of composite action or stiffness compatibility
(Cattanach et al. 2008). A scarcity of any real guidance on the analysis and treatment of historic
URM buildings may well have contributed to this evolution.
A number of seismic retrofits were documented during the 1980s which tended to largely
prescribe coating the inside of the walls with sprayed concrete to form shear walls, and
upgrading diaphragms, which, while effective in increasing strength, permanently covered much
historical evidence on the inside of the building and changed internal surfaces and room
volumes. Some of the less sympathetic retrofits advocated removal of ‘inconvenient’ historic
fabric as a matter of course, altered interiors so as to make them unrecognisable, and were
almost completely irreversible. Some retrofits also resulted in changes to the technical
performance of a building, in some cases rendering them uninhabitable. This approach has
created numerous problems, as the heritage value of some of these buildings has now been
recognised and steps are being taken to restore them to some semblance of their former
aesthetic and function, but the retrofit alterations hinder every step of this process. The Napier
School of Music (Figure C13‐2) is an example of this problem; a retrofit removed most of ithe
building’s exterior decorative character defining elements, which transformed a remarkable
building into an unremarkable one. Work has been done to attempt to return the building
closer to its original form, but the historic material is irrecoverable (Robinson & Bowman 2000).
Similarly, the conversion of the Cranmer Square Normal School in Christchurch (Figure C13‐3)
into luxury apartments involved almost total loss of interior character and of the relationships
between interior spaces, as the interior surfaces were covered or replaced, and much of the
historic material was removed.
C13‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐2: Napier School of Music before and after retrofit. (reproduced from Robinson
and Bowman 2000)
Figure C13‐3: Before and after cross sections through Cranmer Square Normal School,
Christchurch (reproduced from Wilby, 1983). Showing changed interior layouts and surfaces
and removed a large amount of the historic material
C13.3 Architectural Character
Every building is unique and possesses many attributes that help to define it as being distinct
from all others, which are collectively known as its character. Architectural character is the
physical presence which gives a building this distinguished quality and is defined by the
building’s overall shape (architectural form), its materials and methods of construction, the size
and arrangement of openings and projections, the interior spaces and relationships between
these, and the setting and surroundings of the building (Nelson 1988). In an old building,
preservation of architectural character is particularly important, as its character differentiates it
from modern buildings.
C13‐7
Section C13 ‐ Heritage Characteristics of URM Buildings
There are also many facets to physical understanding, beginning with the overall definition of
the building style, material and type. The building may typify a period of construction, which
usually relates to an architectural style, or exemplify construction in a certain material, or
represent a particular type of building (Nelson 1988). Style and archetype will not necessarily
affect the specific treatment of a building, but will provide background information as to its
degree of importance and aesthetic qualities. Once these characteristics have been established,
further distinction can be imparted by examining the architectural form, site, construction
technique, material use and craftsmanship, evidence of age, furnishings, and decorations.
It is recognised that in many cases, time and resources will not allow for an extensive
investigation of architectural and historic significance (Trapeznik 2000). The technique outlined
below considers only the physical and tangible attributes of a building; the parts that can be
seen and appraised, and constitutes a basic level of analysis, but it is simple and is able to be
performed relatively quickly. This technique is borrowed from the United States Department of
the Interior Preservation Brief 17 (Nelson 1988) on Architectural Character. The methodology
used is largely the same here, although the examples specifically reference New Zealand URM
buildings and retrofits. See http://www.nps.gov/history/HPS/TPS/briefs/brief17.htm for further
reading. In some cases, where a building is architecturally unremarkable and is of little historical
interest, this will be an adequate investigation; however it is emphasised that all old buildings
will possess some degree of historical significance, and unless research is done, the extent of this
significance may not be properly identified. Recognition of character is an imprecise practice,
and as such there will always be some debate as to what constitutes an important element.
C13.3.1 Overall Visual Aspects
The first step when analysing architectural character is to view the building from a distance, and
walk around it if possible to view it from each side to ascertain its dominant features. Some
sides will be more important than others, but this is not to say that rear walls have no value;
merely that they contribute less to the overall character than a decorative street façade. The
purpose of this exercise is to identify the elements that make up the architectural form, which is
essentially the combination of shape, size, colour, texture, and organisation, that makes the
building identifiable as this particular building.
The overall visual picture is established by considering the main features of the exterior of the
building without concentrating too much on details; most of these will be immediately and easily
recognisable. The primary form is defined by the outline of the walls and roof, as well as any
projections such as balconies, porches, or chimneys, or where buildings are touching, by the
edge of the change in material or style. The purpose of this exercise is to consider the geometric
form of the building as if it were composed of simple blocks (Figure C13‐4).
C13‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐4: Primary form of a URM building in Auckland. At this stage the basic shape and
the broad gestures that make up the building are noted. In this case the building has a
reasonably simple form and displays a typical 3 part vertical hierarchy
Next the elevations should be checked for elements that define their makeup; openings are
important in this regard. The locations of the primary openings should be noted, along with the
location and nature of the main entrance, the pattern and rhythm of windows, and whether
these are regular or random, deep or surface set, or have any other distinctive features. URM
buildings often have a series of piers with windows between which can form a defining pattern
(Figure C13‐5). Projections and recesses should also be noted, particularly further balconies and
porches, arcades, and any dominant decorative features. The depth and articulation of the
facades, in particular of elements such as piers, cornices, and windows, and the ways in which
these elements are layered and relate to each other is an important contributor to the primary
form. In essence, it is important that the facades are considered in three dimensions.
C13‐9
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐5: A study of the building’s openings in two dimensions. This has been done
firstly as the physical openings in the URM, and secondly as the forms surrounding the
openings. Interfering with these patterns will change the way that the building reads;
something as simple as changing the colour of the window frames will bring these forward
and make the openings smaller and less distinct
The roof can form a particularly important element, although this is less common in URM
buildings. A visible roof with gables, dormers, chimneys, or an interesting surface material or
pattern will be a strong feature for defining character (Figure C13‐6). More commonly with URM
buildings, the parapets will obscure a plain roof behind. The shape of the parapets should be
noted, and their interesting features examined (Figure C13‐7). Attention should be paid to
materials and colours, particularly in the case of URM, as even subtle deviations from the
original materials can stand out starkly on an otherwise uniform surface. Similarly, any
ornamentation on the surface or surface features should be conserved wherever possible.
C13‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐6: Shed 13 on the Wellington Figure C13‐7: A parapet that is integral
waterfront features a distinctive roof that is with the design of the rest of the facade
integral with its character and which contributes historic information
The setting of the building influences how the building is read, and therefore affects its
character. A building in a garden setting is somewhat integral to that garden, and likewise a
building within a row of similar buildings is integral to that row. Drastically altering the setting in
which the building has come to be known will affect the character of that building (Figure
C13‐8).
Figure C13‐8: The Auckland Ferry Building is integral with its waterfront setting and altering
this relationship would remove historic meaning
Identifying the important primary elements that make up the exterior of the building will
provide guidance regarding suitable retrofit techniques. A building with an exposed character
brick exterior will generally be unsuitable for strengthening techniques involving exterior
diaphragms or shear walls. The size and spacing of primary vertical and horizontal elements will
inform the layout of potential moment frame or bracing frame solutions. Historically, façades
have been considered the most important aspect of a building. Unfortunately, this belief has
C13‐11
Section C13 ‐ Heritage Characteristics of URM Buildings
given rise to some particularly destructive treatments of buildings where the façade is preserved
and integrated into a new modern building directly behind. While marginally better than
complete demolition, this practice destroys all but a caricature of the original building and
should be avoided. While the façades are important, the rest of the building is equally so; it is
the interior that provides character for the occupants, and the arrangement and volumes of
spaces that governs the experience that the building provides.
C13.3.2 Close‐Up Character
Once an inspection from afar has been completed, an inspection should be made at arm’s length
(where possible) in order to appreciate how the building has been constructed, and the
particular characteristics that help to define its surface and details. The finer details of the
colour of the materials should be noted; in the case of URM, the colour of the brick and mortar
will hint as to its composition and origin, and the bond type will form a surface pattern. The
shape of the bricks and striking of mortar may also be distinctive and character defining (Figure
C13‐9). Weathering, alterations and demolition, and changes of size or demolition of adjoining
buildings will all leave distinctive marks that indicate age. Attention should be given to the
craftsmanship and craft details that have been employed. In some cases materials may have
been employed in interesting ways that give the building an individual character, or some other
small detail may make a significant impact to the way the overall building is read and
understood.
This process will likely have less impact on the types of seismic retrofit techniques that may be
used than when assessing the overall aspects, but will provide information on the construction
of the building. Some façades that look undistinguished from afar may contribute more to
character or have more significance than previously thought.
Figure C13‐9: Details of openings. Things to take note of here are the bond pattern and
technique used for laying bricks, colouring, striking of mortar, changes in surface, recesses
and depressions, and proportions of windows and other components
C13‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C13.3.3 Interior Spaces
Defining the distinctive character of internal spaces can be far more challenging than for the
building exterior. This exercise can also be done on two scales, as for the exterior assessment,
except that enclosed spaces are three dimensional volumes as well as a series of surfaces.
Sometimes the spaces themselves or the relationships between several spaces are important to
the character and nature of the building; inspections of interior spaces should be carried out
progressively by appraising each space in its entirety before moving to the next.
The first step is to identify primary spaces. These are essentially the spaces that are important
to the function or feel of the building. In many buildings the primary space will be the first space
you enter from the street; a lobby or circulation space is often designed with special treatment
as it provides the first impression of the building. Such spaces should be analysed and their
proportions, shape, ceiling height, and massing of elements recorded, as well as any large
features such as stairways or columns that help define them. There will often be a series of
interrelated spaces that also help to define the character and function of the interior of the
building. Such spaces might include an atrium that connects to one or more circulation spaces,
or prominent stairways that in turn connect to a mezzanine. All of these spaces are individually
defined in their own right, but the visual relationship between them is important to the extent
that to significantly alter one of these spaces would affect the others (Figure C13‐10).
Sequentially important spaces will often involve a primary circulation route, although this is not
always the case. Another way in which spaces can be related is through their relationship with
the function or type of the building. For example in a church there is often a long, tall nave that
is flanked by two shorter aisles that are separated from the nave only by a row of columns.
These columns are a sufficient barrier to signify the boundary between two distinct but related
spaces. If the spaces between the columns were filled in, or extra columns added, this
modification would not only fundamentally alter the spatial feel of the building, but also affect
the ability of the building to be read as a church.
C13‐13
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐10: Arcades in Auckland with complex spatial relationships where blocking lines
of sight between spaces or impeding established pedestrian routes will have significant
negative effects
Once spaces have been considered, interior surfaces and features need to be appraised. This
process is similar to that undertaken on the exterior. Important interior features are original or
historic elements that add to the feel of the space; these elements might include fireplaces and
their mantelpieces, decorative ceilings, mouldings, or panelling, lighting fixtures, elevator doors,
and open stairways. Many buildings will possess all of these features in some form; to
determine their importance, consider whether the space would have the same historic qualities
if they were removed.
Many heritage buildings, industrial buildings and warehouses in particular, are likely to feature
open plans with extensive exposed structure. In these cases, the exposed structure forms an
integral part of the industrial nature of the building and therefore contributes to defining of its
character. Exposed brick load bearing walls, cast iron or steel columns, timber or steel roof
trusses, posts and beams, and stone foundation walls were all designed to be seen and
expressed as part of the nature of the building and it is important to the architectural character
that they remain so.
C13.3.4 Other Considerations
The character of the place as it exists is formed by the visible material in the present; however
each historic building will have a long history, often involving renovations and alterations, wear
and tear, disaster damage, or removal of some features. For these reasons, it is important to
attempt to gain a historical knowledge of the building as well as knowledge of its present state.
This knowledge will help to further understand the character of the building, and help inform
C13‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
the design of retrofit solutions. The developmental sequence of the place can be important;
there are a series of events in the development of any building that will affect its current state.
In the late 1970s and early 1980s, a number of New Zealand buildings had character‐defining
towers and parapets removed in response to fears about their seismic safety, rather than being
strengthened by retrofit (Smith 1979). These elements could possibly be reinstated if they still
exist, or allusions made to their existence in subsequent treatments.
The functions of the building or parts of the building can influence treatment, even when these
have little physical manifestation within the building itself. An example of this is the series of
buildings that are now known as Victoria Park Market in Auckland City. This complex began life
as a waste incinerator and refuse collection depot, but fell into disuse and was converted into a
marketplace in the 1980s (Jones, 2008). Even though the buildings retain much of their
industrial character and possess little to distinguish them as markets, they are now known
collectively as Victoria Park Market, and it is impossible to separate this new use from the
original industrial use of the buildings in their history. Any proposed development or retrofit
would need to consider the full history of the buildings, paying heed to both their marketplace
and industrial uses.
C13.4 Initial Considerations for Seismic Improvement
C13.4.1 Requirements
Territorial Authorities require most buildings with strength of less than 33% of current
requirements to be upgraded under The Building Act 2004. It is recommended that building
performance be improved to at least 67% of current code requirement in NZS 1170.5:2004
(NZSEE 2006). Buildings between 33% and 67% are considered to be earthquake risks and
retrofit should be seriously considered. Buildings over 67% are generally considered to be
satisfactory, although are still up to 5 times more likely to suffer significant damage in an
earthquake than a building meeting NZS 1170.5 requirements (NZSEE 2006). Many
unstrengthened (and some previously strengthened) URM buildings provide less than 67% of the
strength of NZS 1170.5 requirements, and many fall below the 33% minimum level, particularly
in more seismically active parts of New Zealand (Russell & Ingham 2010).
C13.4.2 Objectives
The objectives of seismic improvement are twofold; to protect the occupants from injury or
death, and to protect the building itself from damage due to deficiencies in its design. The
protection of occupants is facilitated by protection of the building, so the building itself is the
focus here. The designer needs to consider the building as a functional but potentially deficient
system which can be modified by various degrees in order to improve its performance.
Assessing the architectural character will establish the important features of the building. Many
well‐intentioned retrofits with clever engineering solutions have completely ruined the
aesthetics of the building that they were supposed to be protecting. There are sufficient
strengthening techniques currently available for all retrofits to be reversible and sympathetic to
C13‐15
Section C13 ‐ Heritage Characteristics of URM Buildings
architectural and heritage value. Strengthening of heritage buildings is difficult and sometimes it
will be a case of selecting the least of several evils. Many approaches and options are available
for strengthening URM buildings, and all solutions will result in some impact on the building, so
the type of strengthening system used needs to be very carefully considered. The way it looks,
how it connects, its impact on heritage material and on spaces, and its harmony with the historic
aesthetic are all important considerations. Furthermore, strengthening technology is constantly
improving, and it is vital that current best practice is understood.
The designer may wish to consider the building as a functional but potentially deficient system
that can be modified by various degrees in order to improve its performance. There are several
primary objectives to be considered when undertaking this process (Robinson & Bowman 2000).
These are not rigid rules, but guidelines for achieving a sensitive solution, so there is always
some leeway for alternative approaches and creativity. If the spirit of these guidelines is adhered
to, generally a good outcome will be reached. Note that these guidelines are a heavily
summarised reading of the ICOMOS New Zealand Charter (ICOMOS 2010) for the Conservation
of Places of Cultural Heritage Value.
C13.4.3 Process
Once a building has been identified as a seismic risk, a more thorough investigation is required.
It makes sense at this stage to establish a detailed picture of its intrinsic seismic resistance, and
assess whether simple measures that address basic inadequacies could be undertaken to give
significant increases in strength. This will ensure that the existing fabric is retained and utilised
as much as possible and will minimise the extent of required additional measures. It is generally
considered that a progressive approach to strengthening is appropriate, calculating the
resistance of progressively more extensive configurations to establish the most effective and
sympathetic approach (Robinson & Bowman 2000).
A basic reading of the architectural character of the building will often assist in deciding on an
appropriate approach, and an appreciation of which parts of the building are visually important
will inform which strengthening techniques will allow a sensitive solution. By considering the
elements that form a building, a practical appreciation can be gained of its character, and
understanding will be gained of exactly what is important about it architecturally. This will give
cues as to the type of strengthening systems that might be appropriate. Most solutions will
result in some visual impact to the building; invisible strengthening solutions are generally the
safest option for visual impact, but these are often precluded by some other consideration. Of
visible solutions, the best option will work with the existing material if not to blend in, then to
complement the architectural character. In situations where strengthening work cannot be
hidden, an appreciation of the architectural character will assist the designer in deciding on ways
to incorporate visible strengthening that does not detract from the building.
There is a perception that façades, the public faces of buildings, are the most important aspects
when considering a retrofit, although this is arguable. This perception means that many
improvements are installed within the building’s interior. The spaces, surfaces and materials will
provide information about where and how these might be undertaken in a sensitive manner.
C13‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
For example, it would generally be poor practice to insert a large steel K‐brace in the middle of a
primary circulation space or lobby, and a fine decorative wall covering would suggest that an
internal concrete shear wall is not an ideal solution. A sound understanding of architectural
character will help to furnish the designer with relevant knowledge in order to exercise good
judgement in determining how to best deal with the problem of retrofitting.
Once it has been accepted that seismic improvement is required, it is necessary to determine
the required extent, and the methods that are appropriate to a particular building. Generally,
this analysis is undertaken in stages, beginning with assessing the architectural character of the
building, then addressing the basic known inadequacies of URM construction, then determining
what types of additional strengthening systems are appropriate. There are myriad approaches
and options for the strengthening of URM buildings. The first step that should always be taken is
to establish the building’s intrinsic resistance and utilise this to the greatest degree possible. In
the likely event that this proves to be inadequate, this resistance should be enhanced by
sensitively upgrading connections and other basic URM inadequacies using current best practice,
or by creating new techniques. Strengthening is a complex art, and balancing visual impact with
economy, reversibility, and architectural expression is a delicate act, but one that is possible in
most situations, and one that ought to be the aim of all seismic retrofits. With skill and care, a
good solution can almost always be reached.
C13.5 Stabilisation for Seismic Improvement
Once a building has been identified as a seismic risk, a more thorough investigation is required.
It makes sense at this stage to establish a detailed understanding of the building’s intrinsic
seismic resistance, and assess whether simple measures which address basic inadequacies could
be undertaken to give significant increases in strength (Russell, 2010). Maximising the existing
strength will ensure that the existing building fabric is retained and utilised as much as possible
and will minimise the extent of additional measures. Various techniques for addressing these
basic issues prior to the addition of any large supplementary structure are outlined below.
Unmodified URM buildings usually have a number of critical structural deficiencies which make
them prone to earthquake forces, but these can often be addressed without significant
alteration to the building fabric, resulting in a relatively large increase in strength (Robinson &
Bowman 2000). The overarching problem is that URM buildings were simply not designed for
earthquake loads, and while they can be made to perform adequately in an earthquake, they
lack a basic degree of connection to allow all parts of the building to act together as one.
C13.5.1 URM Material Stabilisation
URM deteriorates in the environment over time. Occasionally this deterioration will result in
local failures and cracking which affect the overall effectiveness of the building. Various external
forces can also cause cracking and damage in URM, such as dampness, subsidence, earthquakes,
and impacts. Deterioration similar to that shown in Figure C13‐11 can often be remedied by
reinstatement and repointing of mortar, but sometimes more substantial measures are
required. Lime mortars should always be re‐pointed with new lime mortars. Mixing lime and
C13‐17
Section C13 ‐ Heritage Characteristics of URM Buildings
Portland cement mortars can cause numerous problems. There are various techniques for the
repair of cracks, securing of lintels, and reinstatement of damage. Bonding agents such as grout
or epoxy can be injected into the mortar; there are also several metal‐based inserts such as
shaped dowels, or reinforcing bars, which can be used to reinstate and strengthen the brickwork
(Croci 1998). The visual impact of reinstatement and strengthening can be minimal if done
carefully, and the result is potentially far superior to a cracked and broken façade. Such
measures are often irreversible however, and care needs to be taken with colour matching and
the concealment of holes drilled for inserting rods. Lintels and arches will sometimes require
strengthening, particularly when these are constructed from URM. One of the best ways to
achieve this is by using drilled and inserted rods which are grouted or epoxied into place. These
rods provide the requisite tensile strength to the structural element while having little visual
impact.
Figure C13‐11: Severely degraded brick and mortar due to moisture
C13.5.2 Falling Hazards and Parapets
URM buildings will often feature numerous decorative elements built with brick and plaster
which are important parts of its architectural character, such as parapets, chimneys, gable walls,
and other, smaller, decorative features. In the past, some buildings had these elements removed
wholesale, rather than being strengthened or secured. Parapets and chimneys are usually the
first parts of a building to fail in an earthquake due to their low bending strength and high
imposed accelerations (FEMA 547). Parapets in particular are simple to strengthen. Generally a
steel section running along the length of the parapet which is fixed back to the roof structure
behind is a suitable technique, if a little crude. The back of a parapet is not often seen, which
makes the visual impact of this method low, and the steel section is bolted to the URM, which
also has good potential for reversibility.
Chimneys are more complex but can usually be strengthened using a combination of fixing them
to the building diaphragms at each level and either strengthening the projecting portion or
bracing it back to the roof structure with steel members in a similar way to parapets, or fixing
steel sections to the sides to provide flexural strength. The technique shown in Figure C13‐12
may be appropriate where the exterior has been plastered. Chimneys contribute to the
C13‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
architectural form of a building and often help define its roofscape, and as such should be
preserved if possible.
Other elements such as decorative plaster features on the face of a wall can be effectively fixed
with a single bolted connection. Less secure elements, such as plaster finials or balusters, can be
secured with a single epoxied bolt connected to a strand of stainless steel wire, to mitigate
falling hazards. However, more complex strengthening work may be appropriate in some cases.
Figure C13‐12: Parapets strengthened using near‐surface mounted FRP strips
C13.5.3 Floor and Roof Connection Upgrades
The most problematic deficiency in URM construction is inadequate connection of diaphragms
to walls (FEMA 547), as failure of these connections can potentially lead to global collapse of the
building. The addition of a network of small ties can substantially increase the strength of the
building by fixing the walls to the floor and roof diaphragms (Robinson & Bowman 2000). These
ties need to resist two actions; shear from the diaphragms trying to slide across the walls, and
tension from the diaphragm and wall trying to separate. If these ties are missing, the walls will
be acting in cantilever from ground level under lateral loads, and floors and roofs are far more
likely to be dislodged from their supports, which is the most common mode of failure for URM
buildings in an earthquake. This damage pattern is shown in Figure C13‐13.
Figure C13‐13: An extreme case in the 2010 Darfield earthquake where inadequate
connections have resulted in wall collapse
C13‐19
Section C13 ‐ Heritage Characteristics of URM Buildings
Walls are not only secured by floors and roofs however; lateral support is also provided by cross
walls, columns, and buttresses and the influence of these elements should be taken into
account. The thickness and length of walls both affect their performance; lower level walls in
URM buildings are usually several leaves thick and can have significant out‐of‐plane resistance.
Similarly, the shear response of these walls can be substantial, particularly for walls with few or
no penetrations (Robinson & Bowman 2000). Horizontal diaphragms formed by floors or roofs
can distribute loads to other load‐resisting elements and tie buildings together. The connections
of these systems are often inadequate or non‐existent, and require upgrading before they can
provide any positive effects. If the resistance of the building still falls below the critical level after
upgrading these connections, then additional strengthening will need to be undertaken.
The most common form of flooring in URM buildings is timber, although concrete is also
encountered (Russell 2010). Timber floors usually consist of tongue and groove planking nailed
to joists which are supported on URM walls, or on bearers which are supported on internal
columns. The direction of the joists affects the way in which these ties can be connected and
there are various strategies for fixing in each direction. These strategies usually involve adding
new timber chords and blocks and a number of steel fasteners to ensure adequate fixity
between the diaphragm material, the chord, and the URM wall, although where access is limited
this system may be substituted for a steel angle at floor or ceiling level. If done sensitively and
without excessive visible use of new timber and fasteners, this method can have minimal visible
impact. Consideration should be given to the type of bolt used, as tensile bolts often require a
plate on the outside wall of the building, which results in exterior visual impact as shown in
Figure C13‐14. Systems have now been developed which forego this requirement, but require
the fixing of bolts into the brick with epoxy. Through‐bolting has good reversibility, but visual
impact, and epoxying has no visual impact but is irreversible, so a balance must be struck.
Figure C13‐14: An assortment of diaphragm fixings
C13‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C13.5.4 Outer Layer Fixing
The outer façade layer of a cavity wall is problematic as it is particularly susceptible to failure by
peeling off outwards. The steel ties which were commonly installed to connect this layer to the
more robust wall behind are often missing or have rusted out, requiring attention during
retrofits (Russell et al. 2006). One approach to this problem has been to fill the cavity with grout
and reinforcing steel, which has the dual benefits of bonding the outer layer to the main wall
and also forming a reasonably strong shear wall which is hidden from view. However, this
approach fails to consider the purpose of the cavity. Lime mortars are porous whereas concrete
is not, which means that any moisture which finds its way inside the building for whatever
reason will remain trapped there. This has resulted in extensive mould and the subsequent
destruction of at least one heritage building in Auckland, making this practice hazardous and ill‐
advised. While a filled cavity may seem like an excellent strengthening solution, it is vital that the
drainage functionality of the cavity is fully considered beforehand.
The current preferred approach is to use a series of proprietary ties at regular centres which are
drilled through the face layer and are epoxied into the structure behind, the fixing of which is
shown in Figure C13‐15. This is effectively a retrofit of the steel ties which ought to have been
present. The visual impact of these ties is minimal, although care needs to be taken when
concealing drilled holes. These ties are irreversible, but their presence is visually negligible.
Figure C13‐15: Spiral threaded rods being installed to secure an outer layer of bricks
C13.6 Enhancing Existing Walls
Seismic improvement should be designed with an appreciation of the architectural character in
mind. The locations for strengthening elements will often present themselves; many
strengthening elements will be able to be located within hidden or secondary spaces where they
will have little architectural or historic impact. However, structure will often be required
immediately adjacent to the street façade(s), in primary circulation spaces, lobbies, stairways,
corridors, and large public spaces (Robinson & Bowman 2000). In these situations, particular
care needs to be taken to ensure that the strengthening technique is appropriate to the
architecture and heritage value of the building. Often it is imprudent to add elements in addition
C13‐21
Section C13 ‐ Heritage Characteristics of URM Buildings
C13.6.1 Out‐of‐Plane Strengthening
URM walls are weak when subjected to forces other than compression. Even when fully secured
to floors at each level, out‐of‐plane forces can cause significant bending. This bending is
governed by the ratio of the height between levels of support to the thickness of the wall
(Rutherford & Chekene 1990). Some walls are of sufficient thickness or have cross‐walls or
buttresses which enable them to withstand these out‐of‐plane forces without modification,
however many walls will require seismic improvement. There are a number of approaches to
combat this problem.
C13.6.1.1 Inter‐Floor Wall Supports
A series of vertical steel sections can be bolted to the inside face of the wall at sufficient spacing
to ensure that the width of wall between supports is capable of resisting the out‐of‐plane forces.
These sections act in bending to transfer wall loads to the adjacent floor diaphragms, essentially
breaking up a large planar wall into a number of buttressed segments. This is a simple method
which may be appropriate in, for example, an industrial building, where visible steel bolted to
the walls is in keeping with the character of the building, or in other buildings where the steel
can be made to be architecturally appealing. In some other situations it may be less appropriate
but less intrusive than other techniques. If there is existing internal framing with space behind
for these columns, and no historic material is lost during installation, then it is a perfectly
acceptable method. Steel sections generally fix to the historic material with bolts only, which
allows a high degree of reversibility.
In the past, rather than only supporting the URM walls for out‐of‐plane actions, these systems
have been conceived as a method to support the floors in the event that the walls fail and
collapse (Cattanach et al. 2006). In all URM buildings and in heritage buildings in particular, this
is a defeatist approach to design. To design for the inevitability of the building collapsing rather
than attempting to prevent its collapse is to deny the building itself of any inherent value, which
in most cases is patently untrue.
A technique in a similar vein to vertical steel is to provide a horizontal steel member at the mid‐
height of the wall and brace this with diagonal struts up to the floor or ceiling diaphragm above,
as shown in Figure C13‐16. This technique might be more suitable than vertical members if there
is a cornice part way up the wall which needs to be conserved, or which can be used to disguise
the steelwork. However care needs to be taken to ensure than the struts are visually
unobtrusive. Both of these techniques can also be undertaken with the steel substituted with
concrete, where this is more appropriate visually, or less commonly with timber. Steel struts can
also be recessed within the width of the wall. Recessing the members results in an irrecoverable
loss of material and may result in other complications such as cracking, although recesses may
be preferable if used beneath a plastered surface, as there it will not affect the interior space.
Concrete will be more massive than steel and therefore more intrusive. Also, once cast, concrete
C13‐22
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
is difficult to remove without significant damage, particularly from a porous and naturally
coloured material like clay brick. In‐situ concrete is a fairly permanent measure, so any activity
which requires concrete to be cast against brick should be given careful thought before being
undertaken. As so often happens in conservation work, there are a number of measures
available, all of which have positive and negative aspects.
Figure C13‐16: Struts from the floor above to improve out‐of‐plane performance
C13.6.1.2 Post‐Tensioning and Other Core Reinforcement
Post‐tensioning has very little visual impact, although its installation may be unsuitable in some
buildings, as access is needed to the top of the wall, and walls need to be of a certain minimum
thickness. Drilling cores involves some loss of historic material from the holes, though compared
to some methods this is a very minor impact. A partially completed example is shown in Figure
C13‐17. If the bars are fully grouted in place, post‐tensioning is essentially irreversible, although
this does not necessarily have to be done. The presence of post‐tensioning bars is not likely to
result in any negative effects to the historic material should their function no longer be required,
provided care is taken with all core reinforcement to ensure that it is adequately protected from
corrosion. This problem can be completely avoided by using plastic coated steel or FRP bars.
There are other methods of core reinforcement; the most common being non‐stressed bars set
in grout, where the steel only becomes stressed when the wall is loaded laterally. The visual
impact and reversibility of these methods are exactly the same as for fully grouted post‐
tensioning, although they are less effective structurally.
C13‐23
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐17: Post‐tensioning bars used in the Birdcage hotel with concrete load spreaders
cut into the brick
C13.6.1.3 FRP and Others
There are a number of other methods, such as strips of fibre reinforced polymer (FRP) fitted into
vertical saw cuts in URM. This is known as near surface mounting (NSM). NSM is a relatively
recent technique which involves epoxying FRP into saw cuts in the surface of the URM and
covering the cut with a grout mixed with brick dust (Dizhur et al. 2011). This technique would
have some visual impact in naked brick, but little if done within an existing grout line, and none
if installed in plastered walls being repainted. This technique can be a particularly effective and
non‐intrusive method of strengthening, although the finishing of this system is noticeable and
work needs to be done to conceal the inserts. Steel strapping as shown in Figure C13‐18 can also
be an effective technique.
C13.6.2 In‐Plane Strengthening
Most URM walls are required to transfer some degree of shear loading along their length. If a
building has insufficient shear capacity in a particular direction, the capacity of existing walls can
be increased instead of inserting additional structure. There are various methods for achieving
this strength increase which generally involve the application of an additional layer of material
bonded to the surface of URM to increase its strength, although there are some measures which
involve altering the wall itself, such as post‐tensioning, as described above. Most of these
measures involve a plane of extra independent structure being applied over the surface of the
URM, effectively forming new shear walls, which are described below.
C13‐24
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐18: Simple and visually interesting in‐plane strengthening
The presence of openings in a shear wall renders that section less stiff than the surrounding full
height walls, meaning that the wall above and below, or between closely spaced openings, will
likely be the first areas to fail in the event of an earthquake. Infilling the openings will eliminate
this problem by making the wall continuous, and has been advocated as a valid solution in the
past. Problems with altering the character of the building and matching brick and mortar colours
mean that this technique should only be used as a last resort and even then preferably not in
visible areas. Infilling openings is likely to be somewhat reversible if done with brick, but not
completely, and visual impact will depend on the location. If in‐filled with concrete, the work will
be less reversible and the ductile behaviour of the wall may be affected due to incompatible
stiffnesses. Localised steel cross bracing near openings is another technique which can prove
effective, but again this system is likely to be highly visible and should only be undertaken when
it does not detract from the character of the building.
C13.7 New Building Strengthening Systems
URM walls will often be unsuitable for strength upgrades, or are not of sufficient extent for
these upgrades to provide enough strength. Taller buildings in particular will frequently require
additional strengthening elements to supplement their inherent strength, as their facades often
consist of a series of piers with openings between slender piers, without enough planar surfaces
to form effective shear walls. As shown in Figure C13‐19, these are susceptible to earthquake
damage. This section describes the more common techniques which can be utilised.
C13‐25
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐19: Pier damage between openings in the Darfield Earthquake
C13.7.1 Moment Frames
Moment frames are a common method of gaining additional horizontal resistance which can
also be used as a local strengthening solution. The advantage of this system is that it is
comprised of beams and columns, so is fully customisable, and there is space between the
vertical and horizontal elements. Moment frames allow full visual and physical access between
each side of the frame, and minimal spatial disruption. In building façades with numerous
openings, some form of moment frame can often be fitted to the masonry piers on the inside or
outside (or both) depending on the effect on the architectural character. Moment frames can be
a particularly effective solution, especially where the frame is tailored to the character of the
building. Care needs to be taken with steel frames in particular to ensure stiffness compatibility
with the existing structure (Robinson & Bowman 2000). Steel is a ductile material, but URM is
not, meaning that under earthquake loads the added stiffness of the steel might not come into
effect until a load is reached where the URM has already been extensively cracked.
Moment frames can be an excellent strengthening technique, either to supplement an existing
wall or as a new, stand alone element. If a steel frame is erected against an existing wall where
weakness exists, the frame needs to be fixed either directly to the URM using bolted
connections into the wall or to the diaphragm. Installing concrete frames is a more complex
undertaking, as these will often be constructed by thickening existing piers, although a concrete
frame which is separate from the existing structure is possible. In both situations it is important
that architectural character is retained, and historic material conserved. Some considerate and
artful design strategies may need to be undertaken to achieve this.
Steel moment frames have a high degree of reversibility, as again they rely on mechanical
connections and relatively small ties to connect to the existing structure. Concrete frames are
generally far less reversible, but can sometimes be better concealed when this is a requirement.
Figure C13‐20 shows a large new moment frame which is expressed as a new element. Some
recent buildings have very effectively used precast concrete load‐resisting elements which are
separate from the URM walls, solving the problem of reversibility (Cattanach et al. 2006).
C13‐26
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐20: A concrete moment frame inside the façade of a large URM building in
Wellington (Dunning Thornton Consultants).
C13.7.2 Braced Frames
Braced frames are available in various configurations: concentric, tension only concentric,
eccentric, and ‘K’ bracing. The key functional difference between braced frames and moment
frames is that due to the diagonal braces, braced frames prevent physical continuity between
spaces on either side of the frame. Braced frames are also generally constructed from steel
rather than concrete, and are much more rigid than moment frames.
Braced frames are a very efficient method of transferring horizontal forces but have significant
setbacks, and their use in façade walls is usually precluded by the presence of windows, as
diagonal braces crossing window openings are generally considered to be poor design. It is also
difficult to get a braced frame to conform to an existing architectural character; however they
can be used to very good effect within secondary spaces, and can be made to fit architecturally
in some situations with careful consideration. Figure C13‐21 and Figure C13‐22 show braced
frames in use. Generally speaking, steel braced frames have a good degree of reversibility and
can provide excellent strengthening when used appropriately.
Figure C13‐21: Eccentric bracing in a Figure C13‐22: An eccentrically braced core
walkway (Dunning Thornton Consultants). (Dunning Thornton Consultants).
C13‐27
Section C13 ‐ Heritage Characteristics of URM Buildings
C13.7.3 Shear Walls
Shear walls are used to increase the strength of existing URM or are added as new elements.
Materials which resist shear loads can be added to the surface of the URM; these might include
gypsum plasterboard, particle board, plywood, or plate steel (Robinson & Bowman 2000), and
are generally fixed to the URM with bolts via a supplementary structure. This fixing of the
supplementary material means that the surface of the URM is generally covered and may
interfere with decorative elements on walls and openings, although this interference can be
worked around with stronger materials such as plate or strap steel. The supplementary material
can also increase the thickness of the wall, which is not particularly desirable as it can reduce the
scale and area of the interior. For these reasons shear walls can be visually detrimental if used
indiscriminately. Stand alone shear walls, which are independent of URM walls, can be
introduced, although these can be detrimental for similar reasons. Despite these negatives,
shear walls are a practical and efficient method for strengthening and are commonly used. All of
these materials can be easily removed in the future, which makes them good solutions for shear
walls in two to three storey buildings with moderate horizontal loads.
During the 1980s, shotcrete shear walls were a common strengthening technique. This
technique involves spraying concrete onto the surface of a URM wall to essentially cast a new
wall against the existing wall, as shown in Figure C13‐23. This technique provides plenty of
strength to the wall, both in‐plane and out‐of‐plane; however it is now largely regarded as
unacceptable unless absolutely necessary as it causes a significant increase in wall thickness and
it is very difficult to remove the concrete, and even more so to restore the wall behind to any
semblance of its character prior to concreting. Furthermore, its installation generally requires
the building to be gutted, which results in the loss of much heritage material and creates an
essentially new interior (Robinson & Bowman 2000).
Another technique for forming shear walls is the addition of fibre reinforced polymer sheets as
in Figure C13‐24. These sheets do not require the same invasive installation as shotcrete walls,
however are generally equally permanent, and have potentially limited application, although
new technology may soon change this. If it is possible to provide out‐of‐plane strength using FRP
inserts, coupled with an FRP surface layer for shear, this solution could be far superior to
shotcrete from an architectural perspective. An important consideration with the use of sheets
of FRP is that it is impermeable, which can lead to problems with water trapped within the
building, resulting in damp and mould issues, and potential de‐bonding of the epoxy.
C13‐28
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C13‐23: Shotcrete walls on the rear of Figure C13‐24: FRP applied to a wall (Dunning
the birdcage hotel Thornton Consultants).
C13.7.4 Diaphragms
Diaphragms are useful because they provide a layer through which lateral forces can be
distributed from their source to remote resisting elements, and also act to bind the whole
building together at each level. A building which acts as one rigid body rather than a number of
flexible panels is far more likely to survive an earthquake. Tying floors to the outer walls is
generally required regardless to ensure that joists are not dislodged (Robinson & Bowman 2000).
Floor diaphragms consist of three main elements; chords, sheathing material, and
supplementary structure. To form a diaphragm in a typical URM building, chords need to be
established, and mechanical fastenings added to take shear and tensile loads (Rutherford &
Chekene 1990). Several secondary fastenings between the chord and the floor or roof may also
be required depending on the technique used. Some tensile ties will penetrate to the outside of
the building and others will be drilled and epoxied in place. Existing historic sheathing may prove
inadequate and require strengthening or an additional layer of more rigid material.
The main visual impact of forming diaphragms is if new sheathing is required. Historic flooring
material is often a significant contributor to the character of a place and ought to be retained in
view whenever possible. If the existing sheathing is inadequate, a ceiling diaphragm below, or
stiffening the existing material might be preferable to covering it. Another approach is to
remove the existing sheathing and install a structural layer beneath it. This exercise requires
extreme care; firstly because existing sheathing, particularly tongue and groove, is very easily
damaged during removal; and secondly, care needs to be taken to restore the boards in the
correct order.
C13‐29
Section C13 ‐ Heritage Characteristics of URM Buildings
Figure C13‐25: It is possible to make diaphragm improvements visibly interesting, as with this
recently installed floor
Ties to the outside of walls may require metal load spreaders which visually impact the exterior.
Many New Zealand buildings display these spreaders, and they seem to have become somewhat
accepted as part of the strengthening process. Nevertheless, care needs to be taken when
considering their visual impact and invisible solutions may be preferable. Much of the additional
required work can be hidden within the floor space, but if this is exposed or the connections are
extensive, special attention will be required to preserve the visual character of the inter‐floor
space.
Diaphragms which are formed using mechanical connections have a high degree of reversibility;
where ties are epoxied into walls, there is less reversibility, but minimal visual intrusion.
Additional sheathing may damage or alter the nature of the historic timber below, making it less
desirable as a solution, though this can be mitigated in cases such as that in Figure C13‐25 by an
interesting replacement. Occasionally, pouring concrete over an existing timber floor is
considered. This solution can greatly increase the stiffness of the building, but in turn increases
its weight and therefore the forces acting upon it. Further, it completely changes the material of
the floor and is not a reversible action, because even if it can be removed, the concrete would
essentially destroy the character of the underlying timber. This procedure is therefore not
recommended except in exceptional circumstances.
Roof diaphragms where the structure is exposed are slightly different, as the inclusion of a
plywood diaphragm above timber sarking is generally acceptable if this area can be accessed, for
example if the roofing is being replaced. This installation can also help to protect the sarking
beneath. Roofs with suspended ceilings can be made to accommodate cross bracing, struts, and
more innovative solutions, as they can be hidden within the ceiling space. In instances where the
roof provides little diaphragm action, or the forming of a diaphragm is uneconomical or
impossible, a horizontal load resisting member at the level of the top of the walls can be used to
provide stability to the walls when subjected to out‐of‐plane loads. However, this member needs
to be fixed to stiff elements at regular intervals to transfer horizontal loads, and these stiff
elements may need to be introduced to the building if other structure cannot perform this task.
C13‐30
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C13.7.5 Foundations
Foundations in existing URM elements are usually adequate (FEMA 547). However, when
transferring horizontal loads, new strengthening elements also usually induce significant
concentrated vertical loads. This extra loading, combined with the added weight of additional
structure, means that often the foundations in URM buildings will need to be upgraded in order
to deal with these new loads. There are various techniques for meeting these new requirements,
which will usually include enlarging the existing foundations by casting a new concrete footing
beside the old and tying these footings together with drilled reinforcing rods. Existing
foundations may also need to be tied together with ground beams to increase stiffness.
Foundations themselves are generally considered to not have a great deal of heritage value;
however the ground level floor often will.
If the building has a basement or subfloor, strengthening foundations is made much easier, but
care needs to be taken when considering foundation extensions to reduce the damage to the
existing above ground material as much as possible. Some buildings are also located in historic
areas and excavations beneath will require the input of an archaeologist.
C13.8 Reduction of Forces
Another approach to seismic improvement of URM stems from its weight. Seismic actions are
directly proportional to the mass of the building, so if mass is reduced, so are the forces acting
upon the building. A lighter building requires less lateral strength and therefore less additional
strengthening. Reducing the mass of a building may seem at face value to be a sensible
approach; however past experience has shown this to not be so. The mass must be removed
from somewhere, and the higher up the mass is, the stronger the forces upon it and the more
difficult it is to strengthen, so the top of the building is the first place which has been looked at.
Historically this has led to the ad‐hoc removal of decorative elements such as parapets, gables,
chimneys, and occasionally whole towers (Robinson & Bowman 2000). These elements will
almost always significantly contribute to heritage value and character, and their retention is
essential to preserving this. Indeed, it is often desirable to replace these features if they have
been removed from buildings and still exist. While reduction of weight may be achieved in more
minor ways, such as removal of internal URM partitions or the removal of plant loads, the
wholesale removal of decorative elements is strongly discouraged.
C13.8.1 Base Isolation
A more recent approach to reducing forces acting on a building is to introduce a damping layer
between the building and the ground. This process is known as base isolation, and works by
separating the mass of the building from the lateral loads induced by ground movement with a
layer of flexible material. Base isolation is the most effective but also the most radical
strengthening technique, as it requires the building to be physically separated from its
foundations by a layer of flexible material, and the foundations and the ground floor will usually
need to be extensively strengthened. However, the strengthening required to the building above
this level is minimised, making base isolation an attractive option for particularly important or
C13‐31
Section C13 ‐ Heritage Characteristics of URM Buildings
valuable buildings (Robinson & Bowman 2000), or where the contents of the building are
valuable; for example, hospitals or museums. Visual impact is minimised as most of the work
takes place at ground level or below, although it is generally not reversible. As it is often below
ground and largely invisible, this work may be of little consequence. The main problem with base
isolation is that currently it is often prohibitively expensive to install in retrospect; however as
new technology emerges this may change.
There are other methods for modifying the behaviour or the building. One method which has
seldom been used in New Zealand is the installation of energy dissipaters or dampers. These
devices resemble large hydraulic cylinders and work in combination with a steel frame to allow
some lateral displacement, but provide an increasing level of resistance. The downside of
systems such as these is that they are expensive and difficult to conceal, as the dissipater works
on the diagonal in place of a brace. Damping is most relevant to flexible buildings, however, and
will not be effective in most URM buildings.
C13.8.2 Torsion
Torsion can be a significant contributor to the total force in a structural system. It is caused by
asymmetry in the mass and structural stiffness of the building. If the centre of mass and the
centre of stiffness of a building are aligned, then there is no torsional effect, however this
coincidental alignment is almost never the case as most buildings are not symmetrical. Buildings
with one or more weak walls with extensive openings, such as shop‐front façades, and one or
more stiff walls, such as party walls, are particularly susceptible to torsional loads. In an
earthquake the loads are focused through the centre of mass, as mass is the primary generator
of seismic loads. The building resists these loads through its centre of stiffness, so if the centre of
mass is offset from the centre of stiffness, then a rotational force is induced in addition to the
basic horizontal force. In extreme cases the rotational force can be greater than the actual linear
earthquake forces.
The main way to combat excessive torsional response is to make the stiffness of the building
more uniformly distributed. This exercise may involve stiffening weak walls with frames or
bracing, or in more extreme cases involve placing vertical saw cuts periodically along the length
of a stiff wall to reduce its stiffness. However, the architectural impact of each of these practices
must be considered. The stiffening of elements involves the same considerations as the
strengthening practices above, while saw‐cutting requires a consideration of cut location and
visibility within the structure of the building. Torsional forces must be considered in the retrofit
of all asymmetrical buildings, particularly those which have one or more walls that are
significantly weaker or stronger than others.
C13.9 Approaches to Seismic Improvement
There are numerous approaches to improving heritage buildings and it is very important that
proposed retrofit work is considered and reviewed prior to installation to prevent well‐
intentioned but insensitive solutions. It is generally accepted that the basic inadequacies of URM
construction should be addressed in every case, which includes upgrading connections between
C13‐32
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
the walls and the floors, fixing outer leaves of cavity walls, and securing parapets and chimneys.
Addressing these very simple problems will ensure that the building will behave as predicted,
and allows the designer to focus on any additional requirements.
The target strength of the building needs to be decided beforehand. In many cases it is accepted
that strengthening to 100% of NZS 1170.5 requirements is simply unrealistic and uneconomic,
but the NZSEE recommended 67% of current requirements should be aimed for if possible. The
approach taken should always refer back to the main principles of dealing with heritage as
outlined in Section C13.2.
There are sufficient seismic improvement techniques currently available for all retrofits to be
reversible and sympathetic to architectural and heritage value. Many approaches and options
are available, and all solutions will result in some impact on the building, so the type of
strengthening system used needs to be very carefully considered. As illustrated in Figure C13‐26,
the way a component looks, how it connects, its impact on existing material and spaces, and its
harmony with the historic aesthetic are all important considerations. Furthermore,
strengthening technology is constantly improving, and it is vital that current best practice is
understood.
Figure C13‐26: A roof bracing system which clashes with the existing structure, resulting in a
confused ceiling space
C13.9.1 Improvement Type
Generally, the architecture and layout of the building will inform the overall approach to seismic
improvement as well as individual strengthening solutions. The impacts of the proposed
improvement systems need to be considered and the simplest or most structurally efficient
system may not always be appropriate. The designer needs to ask how a strengthening element
will impact the building overall: visually and functionally; if it will detract from the heritage of the
building; involve excessive removal of historic material; and its degree of reversibility.
A wall with many openings is generally going to be unsuitable for use as a shear wall or for use
with diagonal bracing, but might suit a series of moment frames, whereas an ornate and
decorative exposed brick wall or complex forms, as in Figure C13‐27 and Figure C13‐28, might
C13‐33
Section C13 ‐ Heritage Characteristics of URM Buildings
not suit a moment frame, but post‐tensioning bars could be appropriate. An industrial building
might already have exposed steel structure, so some additional steel, if suitably differentiated
from the original structure, could be an excellent solution.
Figure C13‐27: The ornate façade of this Figure C13‐28: The solution was to innovatively
building made exterior strengthening strengthen the interior, which allowed retention of
difficult its most important visual elements
C13.9.2 Placement
Placement of new structure is also often informed by the layout of the building. It is desirable for
strengthening elements to be as close as possible to the source of the loads in order to minimise
additional work as in Figure C13‐29. It is generally undesirable to subdivide interior spaces,
particularly primary spaces, into smaller secondary spaces. The practice of placing intrusive
bracing in walkways or in the middle of rooms is discouraged as it forms a barrier, physical or
implied, between volumes.
Figure C13‐29: Openings in this Christchurch Figure C13‐30: Steel structure installed in an
café allowed placement of frames which intrusive way which is visually distracting
compliments the space
C13‐34
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C13.9.3 Visibility
It is not always desirable to conceal new structure as if it were part of the building; the new work
becomes part of the history of the building and it may be preferable to expose the structure if it
is done in an architecturally sensitive way. It is very easy for an ill‐considered yet simple
structure to negatively affect a building, as in Figure C13‐30. A particularly important heritage
building will need careful treatment overall, and it may be possible and desirable to hide all new
improvement work, allowing the building to retain as much of its character as possible. In other
cases, with industrial or less ornate or important buildings, it is sometimes appropriate to
celebrate strengthening work by exposing it and making it into a feature of the improved
building as in Figure C13‐31 and Figure C13‐32. These two schools of thought may seem to be
conflicting, or even mutually exclusive, but careful aesthetic consideration can potentially make
either approach appropriate for any given case.
Figure C13‐31: No attempt has been Figure C13‐32: Excessive floor and wall ties somehow
made to conceal this new shear wall, work with this façade
rather the aesthetic is celebrated by
the rest of the architecture
C13.10 Conclusions
There is now no question that something must be done about the seismic strength of a large
number of URM buildings. These buildings form a large proportion of New Zealand’s heritage
structures, and have far more importance to society than simply their functions; thus retrofit for
seismic strength is not merely a nice idea, but a vital action. This need is further bolstered by
recent legislation requiring action on buildings below a certain strength threshold, which
includes most URM buildings. Aside from retrofit, there are two alternative options for these
buildings: ignore their inadequacies and hope that there are no earthquakes, or demolish them
now. Neither of these options addresses the problem, and both options are likely to result in an
unacceptable loss of built heritage.
C13‐35
Section C13 ‐ Heritage Characteristics of URM Buildings
Retrofitting is potentially invasive and destructive; therefore a method should be adopted to
guide the design and implementation of any intervention. Fortunately, a series of authoritative
principles have been developed, and are prescribed in the ICOMOS New Zealand Charter for the
Conservation of Places of Cultural Heritage Value, which can be summarised into four basic
principles: know the important characteristics of the building; respect its character and integrity;
only do as much as is necessary; and make new work recognisable and reversible.
Seismic improvement is a complex art. Balancing strength requirements with visual impact,
economy, reversibility, and architectural expression is a delicate act, but one which is possible in
most situations, and one which ought to be the aim of all retrofits. Sometimes the design
process will involve choosing the lesser of many evils, but with skill and care, a good solution can
always be reached.
There are myriad approaches and options for the seismic retrofit of URM buildings. The first
step, which must always be taken, is to establish the building’s intrinsic resistance and utilise this
to the greatest degree possible. In the likely event that this intrinsic resistance proves to be
inadequate, further resistance should be added by sensitively upgrading connections and other
basic URM inadequacies using some of the methods described in this article, or by creating new
techniques.
A basic reading of the architectural character of the building will often assist in deciding on an
appropriate approach, and an appreciation of which parts of the building are visually important
will inform which improvement techniques will allow a sensitive solution. Most solutions will
result in some visual impact to the building; invisible solutions are generally the safest option for
visual impact, but these solutions are often precluded by some other consideration. Of visible
solutions, the best option will work with the existing material if not to blend in, then to
complement the architectural character.
The most important attribute that a designer can have is creativity. Conservation principles exist
to inform the type of action that will result in a good solution from the perspective of
conservation of heritage. Stringent adherence to these principles will not necessarily result in a
solution that is good from an architectural or usability perspective. Sound judgment and design
skill are required to deploy conservation principles with imagination and artfulness. The best
strengthening solution will work with the existing material to complement the architectural
character with a minimum of impact. There may never be a perfect solution or a single right
answer; however, a solution that significantly works towards achieving the above goals can be
considered to be a successful and effective intervention.
C13.11 References
Aplin, G. (2002) Heritage; Identification, Conservation & Management. Oxford University Press,
Melbourne.
Bowron, G, and J. Harris. (1994) Guidelines for Preparing Conservation Plans. New Zealand
Historic Places Trust, Wellington, New Zealand.
C13‐36
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Cattanach, A. G., Alley, G. W. and Thornton, A. W. (2008) “Appropriateness of seismic
strengthening interventions in heritage buildings: A framework for appraisal” Paper 30
Presented at the 2008 New Zealand Society for Earthquake Engineering Conference, Napier,
New Zealand, 11 – 13 April.
Charleson, A. W. (2008) Seismic Design for Architects: Outwitting the Quake.
Elsevier/Architectural Press, Amsterdam, Boston, London.
Charleson, A. W. (2005) Structure as Architecture. Elsevier, Oxford.
Charleson, A. W., Preston, J and Taylor, M., (2001) “Architectural expression of seismic
strengthening.” Earthquake Spectra, 17(3), 417‐426.
Ching, F. D. K. Architecture; form, space, and order. Hoboken, N. J., John Wiley & Sons, 2007.
Christianson, J. (1983) “State Opera House – upgrading.” New Zealand Society for Earthquake
Engineering Bulletin, 16(2), 175‐178.
Croci, G. (2008) The Conservation and Structural Restoration of Architectural Heritage.
Computational Mechanics Publications/WIT Press, Southampton, U.K..
DBH (2004). "Building Act 2004." The Department of Building and Housing, Wellington, New
Zealand.
Derakhshan, H., Dizhur, D., Lumantarna, R., Cuthbert, J., Griffith, M. C. and Ingham, J. M. (2010)
“In‐field simulated seismic testing of as‐built and retrofitted unreinforced masonry partition
walls of the William Weir House in Wellington.” Journal of the Structural Engineering Society
of New Zealand, ISSN 0114‐2879, 23(1), 51‐61.
Dizhur, D., Derakhshan, H., Griffith, M. C. and Ingham, J. M. (2011) “In‐situ testing of a low
intervention NSM seismic strengthening technique for historical URM buildings”
International Journal of Materials and Structural Integrity (IJMSI), Submitted May 2010.
Evans, D. S. (1983) “Strengthening of State Trinity Centre, Christchurch.” New Zealand Society for
Earthquake Engineering Bulletin, 16(2), 162‐164.
Feilden, B. M. (2003) Conservation of historic buildings. Architectural Press, Oxford, U.K..
FEMA. (2003) FEMA 397 Incremental Seismic Rehabilitation of Office Buildings. FEMA,
Washington D.C.
FEMA. (2004a) FEMA 398 Incremental Seismic Rehabilitation of Multifamily Apartment Buildings.
FEMA, Washington D.C.
FEMA. (2004b) FEMA 399 Incremental Seismic Rehabilitation of Retail Buildings. FEMA,
Washington D.C.
FEMA, (2006a) FEMA 454 Risk Management Series: Designing for Earthquakes, A Manual for
Architects. FEMA, Washington D.C.
FEMA, (2006b) FEMA 547 Techniques for the Seismic Rehabilitation of Existing Buildings. FEMA,
Washington D.C.
Gatley, J. (2008) “Salvation by works.” Architecture NZ, 2, 48‐58.
C13‐37
Section C13 ‐ Heritage Characteristics of URM Buildings
Goodwin, C. O. (2008). "Architectural considerations in the seismic retrofit of unreinforced
masonry heritage buildings in New Zealand." Master of Architecture, M. Arch Thesis,
Department of Architecture and Planning, The University of Auckland, New Zealand.
Goodwin, C. O., Tonks, G. and Ingham, J. M. (2009). "Identifying heritage value in URM
buildings." Journal of the Structural Engineering Society New Zealand, 22(2), 16‐28.
Holman, S. (2000) Guidelines for developing heritage buildings. New Zealand Historic Places
Trust, Wellington, NZ.
ICOMOS NZ. (2010) Charter for the Conservation of Places of Cultural Heritage Value. Revised
2010, original text 1993. International Council on Monuments and Sites New Zealand (Inc).
http://www.icomos.org.nz/nzcharters.html.
Ismail, N., Laursen, P. T. and Ingham, J. M., (2011) “Seismic performance of face loaded
unreinforced masonry walls retrofitted using post‐tensioning.” AJSE, In Press.
Jandl, H. W. (1988) Preservation Briefs 18 – Rehabilitating Interiors in Historic Buildings;
Identifying and preserving Character‐Defining Elements. U.S. Department of the Interior,
National Parks Service, Washington D.C.
Jokilehto, J. (1986) A History of Architectural Conservation: The Contribution of English, French,
German, and Italian Thought Towards an International Approach to the Conservation of
Cultural Property
<http://www.iccrom.org/pdf/ICCROM_05_HistoryofConservation00_en.pdf> (6 August,
2008).
Jones, M. (2004) “Campbell Free Kindergarten (former),” New Zealand Historic Places Trust
Register of Historic Places, no. 7537:
<http://www.historic.org.nz/Register/ListingDetail.asp?RID=7537&sm=> (26 October, 2008)
Leuchars, J. M. (1989) “The Use of Sprayed Concrete in the Strengthening of Earthquake Risk
Buildings.” New Zealand Society for Earthquake Engineering Bulletin, 22(3), 167‐171.
Look, D. W., Wong, T. and Augustus, S. R. (1997) Preservation Briefs 41 – The Seismic Retrofit of
Historic Buildings; Keeping Preservation in the Forefront. U.S. Department of the Interior,
National Parks Service, Washington D.C.
Mack, R. C., FAIA, and J. P. Speweik. (1997) Preservation Briefs 2 – Repointing Mortar Joints in
Historic Masonry Buildings. U.S. Department of the Interior, National Parks Service,
Washington D.C..
Mahmood, H. and Ingham, J. M. (2011) “Diagonal compression testing of FRP‐retrofitted
unreinforced clay brick masonry wallettes.” ASCE Journal of Composites for Construction. In
Press.
McLean, R. (2010) Seismic Strengthening – Improving the Structural Performance of
Heritage Buildings. New Zealand Historic Places Trust, Wellington, NZ.
Nelson, L. H. (1988) Preservation Briefs 17 ‐ Architectural Character. U.S. Department of the
Interior, National Parks Service, Washington D.C..
C13‐38
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
New Zealand Government. (1993). "Historic Places Act." Ministry for Culture and Heritage, ed.,
New Zealand Government, Wellington, New Zealand, Date of assent: 17 May 1993.
NZSEE (2006). Assessment and improvement of the structural performance of buildings in
earthquakes: Recommendations of a NZSEE study group on earthquake risk buildings, New
Zealand Society for Earthquake Engineering.
Park, S. C. (1997) Preservation Briefs 16 – The Use of Substitute Materials on Historic Building
Exteriors. U.S. Department of the Interior, National Parks Service, Washington D.C..
Price, N. S., Talley, M. K. and Vaccaro, A. M. (1996) Historical and philosophical issues in the
conservation of cultural heritage. Getty Conservation Institute, Los Angeles.
Robinson, L. and Bowman, I. (2000) Guidelines for Earthquake Strengthening. New Zealand
Historic Places Trust, Wellington, NZ.
Russell, A. P. (2010). "In‐plane seismic assessment of unreinforced masonry buildings." PhD
thesis, Department of Civil and Environmental Engineering, The University of Auckland,
Auckland, New Zealand.
Russell, A. P. and Ingham, J. M. (2010). "Prevalence of New Zealand's unreinforced masonry
buildings." Bulletin of the New Zealand Society for Earthquake Engineering, 43(3), 192‐207.
Russell, A. P. and Ingham, J. M. (2008). "Trends in the architectural characterisation of
unreinforced masonry in New Zealand" 14th International Brick/Block Masonry Conference,
Sydney, Australia, 17 ‐ 20 February.
Russell, A., J. Ingham, and M. Griffith. (2006) “Comparing New Zealand’s Unreinforced Masonry
details with those encountered in other seismically active countries.” 7th International
Masonry Conference, London, U.K., 30 October ‐ 1 November.
Rutherford & Chekene. (1990). “Seismic retrofit alternatives for San Francisco’s unreinforced
masonry buildings: estimates of construction cost & seismic damage.” San Francisco, CA.
City and County of San Francisco Department of City Planning (prepared by Rutherford &
Chekene, Consulting Engineers).
Schofield, H A, Ingham, J. M. and Pampanin, S. (2006) “Critical earthquake risk detailing in New
Zealand’s multi‐storey building stock: understanding and improving the current
perception.” 2008 New Zealand Society for Earthquake Engineering Conference, Napier, NZ.
Smith, I. C. (1979) “Our old buildings can be safe in earthquake (sic).” New Zealand Society for
Earthquake Engineering Bulletin, 12(4), 289‐291.
Taylor, J. M. (1983) “Cashfields Shopping Centre: Christchurch.” NZSEE Bulletin, 16(2), 172‐174.
Taylor, M., Preston, J. and Charleson, A. (2002) Moments of Resistance. Arcadia Press, Sydney,
Australia.
Toomath, W. (1979) “Some Aspects of the strengthening of earthquake risk buildings.” New
Zealand Society for Earthquake Engineering Bulletin, 12(4), 287‐288.
Trapeznik, A. (2000) Common Ground? Heritage and Public Places in New Zealand. University of
Otago Press, Dunedin, NZ.
C13‐39
Section C13 ‐ Heritage Characteristics of URM Buildings
Weeks, K. D. (1997) Preservation Briefs 14 – New Exterior Additions to Historic Buildings:
Preservation Concerns. U.S. Department of the Interior, National Parks Service, Washington
D.C.
Wilby, G. K. (1983) “Redevelopment of the normal school: Cranmer Square, Christchurch.” New
Zealand Society for Earthquake Engineering Bulletin, 16(2), 165‐171.
C13‐40
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C14 ‐ Methods for Stiffening and Strengthening
of Flexible Timber Floor and Roof Diaphragms
C14.2 Scope
Timber diaphragms in URM buildings are typically predicted to deform excessively during seismic
loading such that the long‐established principle has been to stiffen and strengthen the
diaphragm to reduce undesired deformation levels. This intervention is fundamentally based on
the retrofitted diaphragm possessing the required strength to withstand imposed loads without
yielding, and to effectively transmit lateral loads between the perimeter URM walls so that out‐
of‐plane wall failure due to excessive deformation is mitigated. This objective is particularly
important for bi‐directional ground motions where no wall is exclusively oriented in‐plane or
out‐of‐plane and a rigid diaphragm is highly desirable to adequately engage the URM walls as
load collectors.
However, no strategy for optimising diaphragm stiffness and strength currently exists and the
particular impact that diaphragm retrofits have on global or local URM building seismic response
mechanisms remain largely unknown. For this reason, the intervention techniques presented in
this section shall be applied with a degree of caution as not to adversely affect the seismic
performance of the URM building. Where possible, nonlinear time‐history analysis of complete
URM building response to applied ground motion is recommended to determine the sensitivity
of the structure to diaphragm modifications. Further reading on a proposed retrofit strategy of
timber diaphragms for the performance‐based retrofit of URM buildings can be found in
Brignola et al. (2008).
The presented recommendations have been compiled from long‐standing engineering principles
and collaborative judgement, much of which has been taken from Oliver (2010) and which is
suggested as additional reading. Until future research provides the necessary guidelines for the
control and optimisation of diaphragm stiffness and strength, the guidelines presented in this
section serve as the most appropriate to achieve desired URM seismic performance.
It is acknowledged that the recommendations outlined in this section predominantly relate to
floor diaphragms, with no explicit guidelines provided for the retrofit of timber roofs. As
suggested in Section C8, many similarities exist between the configuration of timber floors and
timber roofs, so the presented retrofitting guidelines are permitted to be appropriately applied
to roofs at the designer’s discretion. Future research on timber floor and roof retrofitting
techniques is anticipated to provide comprehensive and explicit guidelines for each of these
structural elements.
C14.3 Retrofit Selection
The selection of a diaphragm retrofitting technique needs to be a holistic decision that addresses
all of the issues surrounding a URM building upgrade. Most obviously the diaphragm needs to
meet performance requirements. In the absence of optimal strength and stiffness targets,
C14‐1
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms
timber diaphragms are conventionally designed with sufficient strength to resist lateral loading
without yielding, and adequate stiffness to limit deformation to within limiting values. The level
of performance improvement necessary may govern the retrofit material selected; for example,
a low level performance increase may only require additional light wood panelling, whereas a
large performance increase may demand the incorporation of steel members.
Many URM buildings in New Zealand are protected by the Historic Places Act (1993) that
regulates the alteration or removal of heritage structures. This restriction adds additional
complexity to the retrofit selection process, as Structural Engineers may be limited to the level
of modification that can be imposed on the diaphragm and surrounding URM structure during
retrofitting. Examples of diaphragm protection include:
Retention of all existing diaphragm materials and appearance where possible. Some
floor section removal for new elevator shafts, stairwells and amenities may be
permissible but otherwise is prohibited.
Retention of all existing diaphragm materials for architectural preservation. Appearance
(sheathing) may be permitted to be altered, but for the possibility that future
generations would show interest in historic construction, floor materials may be
required to be retained.
No restriction to diaphragm alteration. In such cases the existing diaphragm materials
may be utilised, or be completed removed for new diaphragm construction. Typically, if
the existing joists are adequately tied into the URM perimeter walls, then they can be
cost‐effectively used as part of the new diaphragm system.
Other factors such as available budget and practical issues, such as accessibility, may also affect
diaphragm retrofit selection. It is recommended that a comprehensive case study of the entire
URM structure be completed before making any retrofitting decisions. Some common
diaphragm retrofitting methods, together with appropriate applications, are summarised below.
Common Diaphragm Retrofitting Techniques
Plywood panelling is a popular and cost‐effective diaphragm retrofit. Given that existing
diaphragms are almost always constructed of timber, the implementation of plywood and other
timber members to strengthen the diaphragm is comparatively simple. In situations where the
existing sheathing must be retained for aesthetic purposes, plywood sheets can be fastened to
the underside of the floor as a ‘ceiling’ diaphragm. Plywood overlays can be used when it is
desired to retain the existing timber sheathing but appearance is not restricted. Where no
heritage conditions exist, it may be economical to completely remove the existing floorboards
and replace with new plywood sheathing.
Blocking must be considered for any plywood retrofit to effectively transfer shear flow between
panels. Conventional blocking involves nailing timber framing between joists along plywood
panel boundary lines. This method is appropriate for new plywood diaphragms but can be
problematic when the existing timber sheathing is required to remain. A recommended
alternative is stapled sheet metal blocking that can be fastened without excessive alteration to
the existing diaphragm, as illustrated in Figure C14‐1(a). In this instance, light gauge sheet metal
C14‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
strapping (75 mm x 24 gauge typical) is provided at plywood sheet edges and staples (92 Series
typical) are used to fasten the plywood to the blocking. Diaphragm strengthening using this
technique is common in North America and has been used at the Auckland Art Gallery and
Christchurch Arts Centre Old Arts School as part of seismic refurbishment works.
Another retrofit option is incorporation of steel frames. An example of this would be
incorporating a light steel frame in localised sections of a diaphragm where additional strength
and stiffness is required, such as adjacent to stairwell or elevator shaft openings (see Figure
C14‐1(b)). The primary concern with steel frame retrofits is its stiffness incompatibility with the
existing diaphragm and URM walls. It is crucial that a steel frame retrofit be adequately
connected into the existing structure so that uncoupled response is avoided.
Other retrofits seen around the world include meshing the existing diaphragm with light steel
strapping or carbon fibre‐reinforced polymer (CFRP) strips (see Figure C14‐1(c)), although the
application of these intervention types in New Zealand URM buildings is unknown and in reality
is likely to require further research in order to be a viable retrofitting option. Heavy steel frames
or reinforced concrete overlays are discouraged as retrofitting techniques because experience
has shown them to be costly and invasive, and to add considerable seismic mass to the URM
structure with detrimental effects on overall seismic performance.
C14‐3
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms
(a) Plywood overlay with stapled sheet metal (b) Local light steel frame incorporated into
blocking plywood diaphragm
(c) Meshing retrofit
Figure C14‐1: Diaphragm retrofit examples
Expected Retrofit Performance
To serve as a reference for Structural Engineers, a summary of an experimental program
involving the in‐plane cyclic loading of full‐scale timber diaphragms retrofitted with plywood
overlays is presented below to illustrate the expected diaphragm performance when using a
common retrofitting technique.
A total of four retrofitted diaphragms measuring 10.4 m x 5.535 m were tested to establish their
force‐displacement response to lateral loading. The testing schedule is analogous to that
outlined in Table C3‐1 for tested as‐built diaphragms. All retrofits comprised 2400 mm x
1200 mm x 15 mm structural grade plywood laid over the existing floorboards with 75 mm x 24
gauge sheet metal straps fastened to the plywood edges with ECKO SF‐9215 staples at 100 mm
centres to act as blocking. In the loading direction parallel to joists, full‐depth double timber
blocking was provided between the joists at the diaphragm edges for edge nailing and to act as a
compression chord.
C14‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
40 mm x 6 mm steel flat was fastened to the full‐depth blocking with 10 gauge screws at
100 mm centres to act as the tension chord member. In the direction perpendicular to joists, the
two end joists were utilised as compression and tension chords. Field nailing (approximately 300
mm centres) was applied to the plywood sheets at the locations of the joists, and all diaphragm
edge nailing was spaced at 100 mm centres. All diaphragm nails were 90 mm x 3.15 mm power
driven nails. Figure C14‐2 provides some photographs of the tested retrofitted diaphragms.
Further details will be available in journal articles and a PhD thesis that is currently being
authored.
The force‐displacement response for each retrofitted diaphragm is provided in Figure C14‐3.
C14.4 Revised Diaphragm Assessment
NZS 3603 (1993) can be used to design diaphragms strengthened with new plywood sheathing.
Good references for the design of new plywood diaphragms include Buchanan (2007) and Smith
et al. (1986). For other retrofitting techniques any rational and robust method to determine
stiffness and strength is permitted.
The necessary techniques to revaluate diaphragm period, seismic loading and performance
characteristics are provided in the suggested sections of this document.
C14.5 Diaphragm Chord Design
Tension and compression forces are generated when a diaphragm deforms laterally under
seismic loading, as illustrated in Figure C14‐4. Chord elements are structural members of
diaphragms that are oriented perpendicular to the imposed seismic loading and are used to
resist these induced tension and compression forces. Chord members are required at each
diaphragm edge and often adjacent to penetrations to resist the increased stresses at these
locations.
C14‐5
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms
(a) Retrofit FS1b overview (b) Blocking and steel chord
(c) Retrofit FS3b overview (d) Blocking construction
Figure C14‐2: Photographs of tested retrofitted diaphragms
Formal diaphragm chord elements are usually absent in existing diaphragms, and must be
designed as part of any diaphragm retrofit. Diaphragm chord elements are typically detailed as
elastically responding elements by following capacity design procedures or using nominally
ductile (μ = 1.25) design loads. This procedure ensures that the chord members remain elastic
even if the diaphragm is forced into the inelastic regime.
Diaphragm chord elements can be added as a retrofit intervention into existing diaphragms
relatively easily. Multiple bays of existing timber joists can be utilised by making them
continuous using nail plates and/or bolts. When new chord elements are required perpendicular
to existing joists, new light gauge steel straps or steel flats nailed or screwed to timber blocking
can be used. If a steel flat (often 30 – 40 mm x 4 – 6 mm) is required for large chord forces, it is
recommended that double blocking be implemented to allow the steel flat to be fastened flush
with the adjacent sheathing.
C14‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
100 100
50 50
Load (kN)
Load (kN)
0 0
-50 -50
-100 -100
-150 -150
-200 -200
-200 -150 -100 -50 0 50 100 150 200 -200 -150 -100 -50 0 50 100 150 200
Midspan displacement (mm) Midspan displacement (mm)
(a) Diaphragm FS1b (b) Diaphragm FS2b
250 250
Retrofit Retrofit
200 200
As-built As-built
150 150
100 100
50 50
Load (kN)
Load (kN)
0 0
-50 -50
-100 -100
-150 -150
-200 -200
-250 -250
-200 -150 -100 -50 0 50 100 150 200 -200 -150 -100 -50 0 50 100 150 200
Midspan displacement (mm) Midspan displacement (mm)
(c) Diaphragm FS3b (d) Diaphragm FS4b
Figure C14‐3: Full‐scale retrofitted diaphragm test results
Figure C14‐4: Diaphragm chord schematic
C14‐7
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms
C14.6 Subdiaphragm Design
Diaphragm retrofit design may sufficiently increase diaphragm stiffness and strength to required
levels but may neglect the necessary provisions to adequately engage this improved
performance. The subdiaphragm methodology is a design method whereby the main diaphragm
is broken up into a number of smaller (sub) diaphragms at the diaphragm perimeter. The smaller
subdiaphragms are designed to resist the amplified out‐of‐plane wall anchorage forces
previously described in Section 7 and span between diaphragm cross‐ties which are typically
provided at approximately 4‐5 m centres.
Figure C14‐5 illustrates the subdiaphragm concept. Consider the building subject to an east‐west
direction earthquake. In this instance it is recommended that continuous diaphragm cross ties
are provided between grid lines A and D. One design solution would be to provide closely spaced
(i.e. 500 mm centres), light‐gauge sheet metal straps with timber blocking to act as the
diaphragm cross ties/struts across the width of the diaphragm. However such a design solution
would likely be costly and intrusive to implement as part of a retrofit.
An alternative design solution using the subdiaphragm concept would be to utilise the existing
beams to act as the diaphragm cross tie/strut forces and then provide smaller subdiaphragms
that span horizontally between the diaphragm cross ties/struts. Consider the length of wall
between grid lines D5 and D6. As shown in Figure C14‐5, a smaller subdiaphragm could be
designed to span between grid lines 5 and 6. In this instance, when the timber joists are
orientated parallel to the wall, the existing joists are often found to be adequate to act as the
subdiaphragm chords. The depth of the subdiaphragm is typically increased until the chord
forces are sufficiently reduced that the existing joists are adequate. Alternatively the existing
joists can be doubled up to increase their capacity. The sheet metal straps described in Section
C11.5 can be used as subdiaphragm cross ties to transfer the amplified out‐of‐plane wall
anchorage forces to the rear of the subdiaphragm.
Once the subdiaphragm design is complete, checks should be made to ensure that the existing
beams that are utilised to act as diaphragm cross ties/struts have adequate axial load capacity,
and that their end connections are sufficient to transfer the necessary diaphragm cross tie
forces. Referring to Figure C14‐5 a similar strategy can be used for the north‐south direction
earthquake. In this direction the existing timber joists can be used as subdiaphragm cross‐ties.
Note that in this direction no beams are available to act as diaphragm cross ties/struts. As an
alternative, existing floor joists can be doubled up and made continuous to act as subdiaphragm
cross ties/struts. Existing beams can often be utilised to act as the subdiaphragm chords.
When designing subdiaphragm elements and cross diaphragm ties the out‐of‐plane wall loads
calculated in Section 7 should be used. Current design office practice is that these amplified out‐
of‐plane wall forces are considered to not act concurrently with those calculated in Section 4 for
main diaphragm design.
C14‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C14.7 Diaphragm Penetrations
Penetrations in diaphragms due to stair openings, elevators shafts and service risers require
special consideration to ensure that diaphragm performance is not compromised. Diaphragm
penetrations cause stress concentrations which can lead to poor diaphragm behaviour if the
openings are not addressed in the diaphragm design.
For retrofits utilising plywood panels, the ‘shear transfer’ method (Smith et al. 1986) is a simple
design method which can be used by Structural Engineers to determine increased nailing and
chord requirements adjacent to diaphragm openings.
When diaphragm penetrations occur immediately adjacent to URM walls, structural steel beams
can be used to provide the necessary out‐of‐plane restraint to the walls (refer Figure C14‐6). The
beams should be tied back into the diaphragm using subdiaphragm cross ties.
Figure C14‐5: Subdiaphragm example (from Oliver 2010)
C14‐9
Section C14 ‐ Methods for Stiffening and Strengthening of Flexible Timber Floor and Roof
Diaphragms
Figure C14‐6: Typical Out‐of‐Plane Wall Support Detail at Diaphragm Opening (from Oliver
2010)
C14.8 Worked Example
See Oliver (2010) for an example of how to stiffen and strengthen flexible floor diaphragms.
C14.9 References
Brignola, A., Pampanin, S. and Podesta, S. (2008). "Evaluation and control of the in‐plane
stiffness of timber floors for the performance‐based retrofit of URM buildings." Bulletin of
the New Zealand Society for Earthquake Engineering, 42(3), 204‐221.
Buchanan, A. (2007). Timber Design Guide, University of Canterbury, Canterbury, New Zealand.
New Zealand Government. (1993). "Historic Places Act." Ministry for Culture and Heritage, ed.,
New Zealand Government, Wellington, New Zealand, Date of assent: 17 May 1993.
Oliver, S. J. (2010). "A design methodology for the assessment and retrofit of flexible diaphragms
in unreinforced masonry buildings." SESOC Journal, 23(1) , 19‐49.
Smith, P. C., Dowrick, D. J. and Dean, J. A. (1986). "Horizontal timber diaphragms for wind and
earthquake resistance." Bulletin of the New Zealand National Society for Earthquake
Engineering, 19(2), 135‐141.
Standards Association of New Zealand (1993). NZS 3603:1993, Timber Structures. Standards New
Zealand, Wellington, New Zealand.
C14‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C15 ‐ Methods for Improving Wall Diaphragm
Connections
This section has been intentionally left blank.
C15‐1
Section C15 ‐ Methods for Improving Wall Diaphragm Connections
This section has been intentionally left blank.
C15‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C16 ‐ Unbonded Post‐tensioning
C16.1 Notation
a Length of equivalent rectangular stress block at nominal strength level, m
Abp Area of bearing plate or block, m2
Aps Effective cross‐sectional area of tendon, m2
bw Wall thickness, m
c Distance from the neutral axis to the extreme compression fibre, m
d Effective depth, m
D Outer diameter of tendon, mm
Em Masonry modulus of elasticity, Pa
Eps Modulus of elasticity for post‐tensioning tendon, Pa
f’m Masonry compressive strength, MPa
fps Tendon stress at nominal strength limit state, MPa
fpsi Tendon stress after all losses, MPa
fpu Ultimate tensile strength of tendon, MPa
fpy 0.1% proof tensile yield strength, MPa
fr Masonry modulus of rupture, MPa
fse Effective tendon stress after all losses, MPa
he Effective height of wall, m
k Parameter representing fraction of f’m at hinge formation limit state
kcr Masonry creep parameter, µε/MPa
kr Tendon relaxation parameter, MPa/MPa
ksh Masonry shrinkage parameter, µε
lw Wall effective length, m
M* Required flexural strength, kNm
Nb Total Axial load at the base of the wall due to Nt, Ww and Apsfse, kN
Nt Overburden weight, kN
Ppsi Initial post‐tensioning force, kN
vo* Design out‐of‐plane pressure, MPa
V* Design shear force at section, kN
Ww Self weight of post‐tensioned wall, kN
th
Xi Distance of i tendon from extreme compression fibre, m
α Parameter representing fraction of f’m at nominal strength limit state
mv Shear Strength reduction factor
mf Flexural Strength reduction factor
εmu Maximum useable masonry strain, m/m
εpu Tendon strain at breakage, m/m
εs Tendon strain at nominal strength, m/m
σn Initial compressive stress at crack location, MPa
θ End rotation or storey drift, Radian
C16‐1
Section C16 ‐ Unbonded Post‐tensioning
Units shown are representative of typical unit conventions and units of calculated values should
be established with respect to the units used for input parameters.
C16.2 Scope
One technique to improve the seismic performance of URM walls is to apply vertical unbonded
post‐tensioning and experimental data used to validate design expressions have been reported
previously in several research studies (Al‐Manaseer and Neis 1987; Ganz and Shaw 1997;
Laursen and Ingham 2004; Laursen et al. 2006; Rosenboom and Kowalsky 2004; Wight and
Ingham 2008; Wight et al. 2006; Wight et al. 2007; Bean Popehn et al. 2008; Ismail et al. 2011).
Research and codification of post‐tensioned masonry originated from Switzerland and the
United Kingdom, and over the last two decades significant research and development has led to
the inclusion of criteria for the design of post‐tensioned masonry into several design codes, such
as MSJC (2005) and NZS (2004).
The performance of post‐tensioned URM walls depends upon the initial post‐tensioning force,
tendon type and spacing, restraint conditions, and the level of confinement. Post‐tensioning can
either be bonded when tendons are fully restrained, by grouting the cavity, or left unbonded by
leaving cavities unfilled. Lateral restraint of post‐tensioning tendons is important when
considering second‐order effects. Because unbonded post‐tensioning is reversible to some
extent and has minimal impact on the architectural fabric of a building, the technique is deemed
to be a desirable retrofit solution for URM buildings having important heritage value. Only
unbonded post‐tensioning is addressed by the provisions of Section 16.
C16.2.1 Retrofit Procedure
A post‐tensioning retrofit is applied either by placing post‐tensioning tendons into cored cavities
located at the centre of the wall or by placing post‐tensioning tendons externally at discrete
locations. The first procedure involves coring a cavity from the top of the URM wall right through
to the foundations, then placing a tendon into the cored cavity and finally the application of a
post‐tensioning force to the tendon. From discussions with specialised local construction
contractors it was established that for the seismic retrofit of URM buildings, current techniques
are capable of coring a cavity up to four stories with a precision of ±10 mm.
Additioanlly, discretely located external unbonded post‐tensioning has been used in retrofit
projects that avoided the coring operation and involved post‐tensioning URM walls using pairs of
external post‐tensioning tendons typically located at wall corners. Figure C16‐1(a) shows a
photograph of the coring operation being performed in Auckland on a heritage URM building
and Figure C16‐1(b) shows discretely located post‐tensioned strands in a retrofitted URM
building located in Christchurch. Threaded mild steel bars are post‐tensioned by using a
hydraulic jack which is removed after tightening of the nut that clasps the post‐tensioning bar,
and strands are post‐tensioned using a hydraulic jack, with the taut strand clapsed by wedge
interlocking at the live end anchorage. Figure C16‐1(c) and C16‐3(d) show photographs of test
walls being post‐tensioned using a threaded bar and a strand respectively.
C16‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(a) Coring Operation (b) External Strands (c) Bar (d) Strand
Post‐tensioning Post‐tensioning
Figure C16‐1: Post‐tensioning Retrofit Technique
C16.2.2 Design Limit States
In order to interpret the seismic behaviour of post‐tensioned masonry walls, four limit states
are defined (see Figure C16‐2), where first cracking is the limit state when the first crack appears
and corresponds to wall performance in the elastic range; the hinge formation limit state occurs
when the upper and lower wall segments start to rotate and corresponds to the serviceability
range; the nominal strength limit state; and the failure limit state corresponding to tendon
yielding. The selection of a limit state for retrofit design is left to the discretion of the designer,
and design equations for the nominal limit state are reported in the guideline.
4
3
2
Total Lateral Force (kN)
1
1 - First Cracking
2 - Hinge Formation
3 - Nominal Strength
4 - Failure
C16.2.3 Retrofit Design Procedure
Figure C16‐3 shows a flow chart for the post‐tensioning seismic retrofit design procedure of
URM walls.
C16‐3
Section C16 ‐ Unbonded Post‐tensioning
Determine %NBS and
if retrofit required.
Detailed URM
assessment.
Out‐of‐plane retrofit In‐plane retrofit design
design of URM walls. of URM walls.
Calculation of Calculation of
Design Moment. Design Moment.
Establish nominal tendon
spacing and shear check.
Calculation of effective Calculation of effective
prestress, fse, and tendon prestress, fse, and tendon
stress at nominal strength stress at nominal strength
limit state, fps, and check limit state, fps, and check
within permissable limits. within permissable limits.
Select larger of out‐of‐
Select larger of the
plane and in‐plane and
calculated f ps values and
check within permissable
limits. If not, return to re‐
establish nominal tendon
spacing and shear check
spacing and shear check.
Calculation of
initial post‐
tensioning force
Check for wall stress at
wall bottom and at
anchorage. If not ok, re‐
establish nominal
tendon spacing and
shear check.
Design
complete
Figure C16‐3: Flowchart for Post‐tensioning Seismic Retrofit Design
C16‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C16.3 Post‐tensioning Tendon Types, Stresses and Spacing
Threaded steel bars are typically used for straight post‐tensioning over short distances and have
relatively lower tensile strength than strands. Figure C16‐1 shows representative experimental
stress‐strain plots for these two types of post‐tensioning tendons. The greased coating of
strands enables high corrosion resistance and lower frictional losses, which makes them an ideal
choice for unbonded post‐tensioning application. The added advantage of high corrosion
resistance facilitates the use of unbonded strands for external post‐tensioning applications by
locating strands at discrete locations, typically at corners of flanged or buttressed walls (see
Figure C16‐1(b)).
Strain (ratio)
Strain (ratio)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0 0.01 0.02 0.03 0.04
Fpu = 67.2 kN
60 532 200 2020
Fpy = 58.2 kN Fpu = 289.2 kN
Fpy = 170.6 kN
0.1% Proof Strength
Stress (MPa)
Stress (MPa)
Total Force (kN)
45 399
(a) Threaded Mild Steel Bar (D = 12 mm) (b) Sheathed Greased Strand (D = 12.7 mm)
Figure C16‐4: Typical Stress‐Strain Plots
C16.3.1 Post‐tensioning Losses
Current post‐tensioned masonry codes provide guidelines for assessing post‐tensioning losses,
with these losses typically attributed to shrinkage, creep, tendon relaxation, elastic shortening,
seating, undulation of tendon, friction and thermal effects. Of these factors, steel relaxation,
shrinkage and creep are the most important factors that will influence the design and longevity
of an adequate retrofit. Testing performed in previous studies (Krause et al. 1996; Ganz and
Shaw 1997; Laursen et al. 2006) investigated prestress losses occurring in post‐tensioned
masonry walls, and found that the post‐tensioning loss was mainly attributable to the use of
low‐strength post‐tensioning threaded steel bars, whereas modern anchorages and
low‐relaxation strands have been proven to result in much smaller losses. The experimentally
determined post‐tensioning losses in concrete masonry ranged from 20‐40% for strands and 25‐
55% for threaded steel bars (Laursen et al. 2006), whereas Ganz (1993) reported observed losses
in clay brick masonry to be 7%, with 3% loss attributed to strand relaxation and 4% attributed to
masonry shortening due to shrinkage and creep. The parameters ksh, kc, and kr reported in
Section 16 were established from experimental work performed by Ganz (1993) and Laursen et
al. (2006) and MSJC (2005) suggested values of creep and shrinkage parameters for clay brick
masonry.
C16‐5
Section C16 ‐ Unbonded Post‐tensioning
C16.4 In‐plane Wall Behaviour
The failure mode of URM walls when subjected to in‐plane seismic excitations depends upon
wall aspect ratio and the characteristics of the constituent materials, and are characterised as
either flexural controlled or shear controlled (see Section C8 for details). Sliding and rocking
failure modes may allow considerable displacement capacity during seismic shaking, and subject
to ensuring that collapse is avoided, these deformation modes may be tolerable to some extent.
Therefore, a flexural/rocking controlled mode is instigated in this retrofit technique by applying
a pre‐compression to the masonry wall. Figure C16‐5 shows a URM pier retrofitted using post‐
tensioning that is subject to a lateral force V* acting at a height he. The pier has a length lw and
Nb is the axial load at the base of pier due to overburden weight, Nt, self weight of the pier, Ww,
and effective post‐tensioning force after losses, Pe.
N t+Pe
V*
h
n
l
w
Figure C16‐5: Post‐tensioned In‐plane Pier Definition and Stress Profile at Initial State
C16.4.1 In‐plane Capacity at Nominal Strength Limit State
At nominal strength the compression stress distribution becomes non‐uniform, which is typically
approximated by an equivalent rectangular compression stress block idealisation (see Figure
C16‐6). The equivalent stress block has a depth of αf’m and width a, where α is a constant and
typically has a value of 0.85 (see Equation 2‐8) and f’m is the compression strength of masonry.
The effective depth is d and the distance from the neutral axis to the extreme compression fibre
is denoted by c. Predicting the seismic response of a post‐tensioned URM wall at the nominal
strength limit state is a diligent task and requires accurate estimation of maximum useable
masonry strain at the extreme compression fibre, εmu, and increased tendon stress, fps. It was
found in an experimental investigation on URM materials (Lumantarna 2011) that due to the
higher deformability of prevalent weak lime mortar used in historic URM construction, the strain
values observed at nominal strength were inflated and much higher than that typically defined
for URM i.e., 0.0035 (MSJC 2005). But to keep the wall displacement to a safe limit, the nominal
strength for clay brick post‐tensioned masonry walls is defined herein as the point when the wall
C16‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
in‐plane drift reaches 1.5% i.e., θ = 0.015 (refer Figure C16‐6) and the corresponding masonry
strain value is termed as the maximum useable masonry strain.
e P
Nt +P+
V*
h
e
f
'
m
a
c
l
w
Figure C16‐6: Post‐tensioned In‐plane Pier Equilibrium at Nominal Strength Limit State
C16.4.1.1 Tendon Stress at Nominal Strength Limit State
Figure C16‐6 shows a post‐tensioned URM pier that is assumed to rock about the neutral axis,
where the post‐tensioning tendon has an effective tendon stress fse and a modulus of elasticity
Eps. The tendon stress at the nominal strength increases due to tendon elongation and can be
presented as.
(C16‐1)
If the unbonded length of tendon is equal to the height of the pier and the rotation value is
small, then εs can be estimated as,
(C16‐2)
Equation C16‐6 can be rewritten as,
0.015 0.7
1 (C16‐3)
Where: a maximum drift of 1.5% is stipulated. The maximum tendon stress should be taken
smaller of 0.85fpy or 0.7fpu and as 0.7fpu is generally the smaller of these two calculated
values, a maximum tendon stress of 0.7fpu is stipulated herein. By substituting Equation
C16‐3 and values for constants α and β of 0.85 for both (typical in flexural design) into
Equation C16‐1 yields,
C16‐7
Section C16 ‐ Unbonded Post‐tensioning
C16.4.2 In‐plane Capacity at First Cracking Limit State
At first cracking limit state one end of the pier reaches zero stress and the other end reaches a
stress of twice the initial masonry stress at the base of the pier (see Figure C16‐5). At the first
cracking limit state the moment capacity, Mc, is evaluated by considering the equilibrium of
forces (refer Equation C16‐5). It should be noted that the masonry modulus of rupture is
conservatively assumed to be zero. Figure C16‐7 shows the stress profile at the first cracking
limit state.
N t+Pe
V*
h
e
2
n
l
w
Figure C16‐7: Post‐tensioned In‐plane Pier Equilibrium at First Cracking Limit State
where (C16‐5)
6
C16.4.3 In‐plane Capacity at Hinge Formation Limit State
The hinge formation limit state, also referred to as the serviceability limit state, is defined as the
point at which the stress in the extreme compression fibre reaches a limiting stress of kf’m (see
Figure C16‐8), where k is a parameter and has a recommended value in NZS 4230:2004 (see
Table C16‐1) ranging from 0.45 to 0.55.
C16‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C16‐1: k Value for Moment Capacity Evaluation at Hinge Formation Limit State
Stress Case Load Category
By considering force equilibrium and assuming that 2fm < kf’m, Equation C16‐6 was developed to
evaluate the moment capacity at the hinge formation limit state.
4
3 where (C16‐6)
6
N t+Pe
V*
h
f
'
l
/
2
-
2
c
/
3
m
c
w
l
w
Figure C16‐8: Post‐tensioned In‐plane Pier Equilibrium at Hinge Formation Limit State
C16.4.4 In‐plane Test Results
A portal frame with post‐tensioned piers (refer to Figure C16‐9) was structurally tested by
applying pseudo‐static cyclic in‐plane loading, and the observed performance of this unbonded
post‐tensioned URM frame (Ismail 2011) was compared to the predicted response (refer to
Figure C16‐10(a) and C16‐10(b)).
C16‐9
Section C16 ‐ Unbonded Post‐tensioning
4424
380 1190 1228 1190 426
942
Average Spandrel
1710
Pier A Pier B
812
Figure C16‐9: Test Frame Dimensions
105 1,064
Lateral Force (kN)
50 133
f ps
0 0 70 709
-100 -265
0 0
-50 -25 0 25 50
-50 -25 0 25 50
Displacement (mm)
Displacement (mm)
(a) Force‐Displacement Plot (b) Tendon Stress‐Displacement Plot
Figure C16‐10: Calculated Vs Experimental Values
As the testing was performed immediately after the application of the post‐tensioning force,
post‐tensioning losses were minimal and were not considered in the analysis. Predicted flexural
capacity was directly compared with the experimental results without applying a strength
reduction factor.
C16.5 Out‐of‐plane Wall Behaviour
The out‐of‐plane rocking failure mode of post‐tensioned URM walls enables utilisation of the
same concepts of flexural controlled limit states as defined in Section C16.2.2. Figure C16‐11
shows a URM wall retrofitted using post‐tensioning, subject to an out‐of‐plane lateral uniformly
distributed seismic force vo*. The wall has a pin‐to‐pin height he, Nt is the axial load due to
overburden weight, Ww is the self weight of the URM wall, and Pe is the effective post‐tensioning
force.
C16‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
v
o
h
/
2
h
/
2
e e
Pe +N t+0.5Ww Pe +P+Nt +0.5Ww
h
e
d
f
'
n m
a
b
w w
b
b
w w
(a) Wall Loads (b) Wall Deformations
Figure C16‐11: Post‐tensioned Out‐of‐plane Wall Definition and Stress Profile at Initial State
Typically, additional axial load exacerbates bending stresses in URM walls as the wall displace
due to P‐Δ effects, whereas ensuring that tendons are laterally restrained eliminates additional
P‐Δ effects because the line of action of the post‐tensioning force does not change with respect
to the neutral axis of the wall (Ganz and Shaw 1997).
C16.5.1 Out‐of‐plane Capacity at Nominal Strength Limit State
Figure C16‐12(a) and C16‐12(b) show the strain and stress profiles at mid‐height of an
out‐of‐plane loaded post‐tensioned URM wall. Maximum allowable masonry strain for out‐of‐
plane loaded post‐tensioned URM walls is twice the value for in‐plane loaded walls. Therefore,
the nominal strength for out‐of‐plane loaded post‐tensioned URM walls is defined herein as the
point when wall the out‐of‐plane drift reaches 3% i.e., θ = 0.03 (refer Figure C16‐11(b)) and the
corresponding masonry strain value is termed the maximum useable masonry strain.
d
0.5he mu f'm
c
a
b
w w
(a) Strain Profile (b) Stress Profile
Figure C16‐12: Post‐tensioned Out‐of‐plane Wall at Nominal Strength Limit State
C16‐11
Section C16 ‐ Unbonded Post‐tensioning
C16.5.2 Out‐of‐plane Capacity at First Cracking and Hinge Formation Limit States
Based on the same definitions of first cracking and hinge formation limit states as discussed in
previous sections, Equation C16‐9 and Equation C16‐10 were adapted for out‐of‐plane flexural
capacities at the first cracking and the hinge formation limit states. It was established from the
experimental results (refer results presented in Section C16.5.3) that the first crack appeared in
out‐of‐plane loaded post‐tensioned walls when the resisting stress exceeded the sum of stress
due to axial loads and the flexural strength of the masonry.
(C16‐9)
6
4
3 (C16‐10)
6
0.5
where:
C16.5.3 Out‐of‐plane Test Results
The observed performance of unbonded post‐tensioned URM walls in out‐of‐plane structural
testing performed (Ismail et al. 2011), was compared to the response predicted using the
equations described in these guidelines and were plotted in Figure C16‐13. As the testing was
performed immediately after the application of a post‐tensioning force, post‐tensioning losses
were minimal and were not considered in the analysis. Predicted flexural capacity was directly
compared with the experimental results without applying any strength reduction factor. Table
C16‐2 gives the dimensions and post‐tensioning details of test walls that were used to validate
the equations.
C16‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C16‐2: Post‐tensioned Wall Dimensions and Post‐tensioning Details
Wall he lw bw Ww f’m Tendon Post‐ fma/f’m
(mm) (mm) (mm) (kN) (MPa) type tensioning (ratio)
Ppsi fpsi
(kN) (MPa)
ABO‐01 3760 1200 220 18.2 6.5 ‐ ‐ ‐ ‐
20 10 120 1212
Mn
Lateral Force (kN)
10 5
Mc 100 1010
0 0
80 808
Predicted Tendon Stress
-10 -5
60 606
-20 -10
Wall Height = 3760 mm ABO-01
PTO-02
-30 Post-tensioning Force = 50 kN -15
40
-80 -60 -40 -20 0 20 40 60
404
80
-80 -60 -40 -20 0 20 40 60 80 Displacement (mm)
Displacement (mm)
(a) Force‐Displacement Response for PTO‐2 (b) Tendon Stress History for PTO‐2
Drift (%) Drift (%)
-4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4
15 140 1414
30 Wall Height = 3760 mm
Mn Post-tensioning Force = 70 kN
Analogous Moment (kNm)
20 10 120 1212
Tendon Stress (MPa)
Tendon Force (kN)
Lateral Force (kN)
10 5
Mc 100 Predicted Tendon Stress 1010
0 0
80 808
-10 -5
60 606
-20
Wall Height = 3760 mm ABO-01 -10
PTO-03 40 404
-30 Post-tensioning Force = 70 kN -15 -80 -60 -40 -20 0 20 40 60 80
-80 -60 -40 -20 0 20 40 60 80 Displacement (mm)
Displacement (mm)
(c) Force‐Displacement Response for PTO‐3 (d) Tendon Stress History for PTO‐3
C16‐13
Section C16 ‐ Unbonded Post‐tensioning
Mc
10 5
100 1010
0 0
80 808
-10 -5
60 606
-20
Wall Height = 3760 mm ABO-01 -10
PTO-04
-30 Post-tensioning Force = 100 kN -15 40 404
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Displacement (mm) Displacement (mm)
(e) Force‐Displacement Response for PTO‐4 (f) Tendon Stress History for PTO‐4
80 808
12 6
40 404
6 3
Wall Height = 3360 mm Wall Height = 3360 mm
Post-tensioning Force = 90.5 kN Post-tensioning Force = 90 kN
0 0 0 0
0 30 60 90 120 150 180 210 0 30 60 90 120 150 180 210
Displacement (mm)
Displacement (mm)
(g) Force‐Displacement Response for PSO‐5 (h) Tendon Stress History for PSO‐5
Drift (%) Drift (%)
0 1.5 3 4.5 6 7.5 9 0 1.5 3 4.5 6 7.5 9
28 15 140 1414
Predicted Nominal Strength Predicted Tendon Stress
Analogous Moment (kNm)
12 120 1212
21
Lateral Force (kN)
9 100 1010
14
6 80 808
7 60 606
3
Wall Height = 4100 mm Wall Height = 4100 mm
Post-tensioning Force = 100 kN Post-tensioning Force = 100 kN
0 40 404
0 0 30 60 90 120 150 180 210
0 30 60 90 120 150 180 210
Displacement (mm)
Displacement (mm)
(i) Force‐Displacement Response for PSO‐6 (j) Tendon Stress History for PSO‐6
Drift (%) Drift (%)
0 1.5 3 4.5 6 7.5 9 0 1.5 3 4.5 6 7.5 9
28 15 100 884
Analogous Moment (kNm)
12 80 707
Tendon Stress (MPa)
Tendon Force (kN)
21
Lateral Force (kN)
(k) Force‐Displacement Response for PBO‐7 (l) Tendon Stress History for PBO‐7
Figure C16‐13: Calculated Vs Experimental Values
C16‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C16.6 Design Example
Design a post‐tensioned seismic retrofit for a URM wall that is two leafs thick 5.0 m long, and 3.2
m high. Design action calculations were performed and an in‐plane shear force of V* = 88.6 kN
and an out‐of‐plane uniform pressure of vo* = 5.8 kN/m2 were calculated for a maximum credible
earthquake. Masonry material properties were investigated and the results of this testing are
summarised in Table C16‐3. It was decided by stakeholders that post‐tensioning tendons will be
placed into centrally located cored cavities. The overburden axial stress due to supported roof
was calculated to be 0.5 kN/m.
Table C16‐3: Masonry Materials
C16.6.1 Establishing Tendon Spacing
Maximum permissible tendon spacing = 0.4he = 1.28 m. So, use a spacing of 1.25 m and 4 equally
spaced tendons. Table C16‐4 summarises the input parameters and Figure C16‐13 shows details
of the URM wall that is being retrofitted.
Table C16‐4: Input Parameters
C16‐15
Section C16 ‐ Unbonded Post‐tensioning
Po
Pe Pe Pe Pe P+P
o e
v*o
V*
0 0
0
2 0
3 2
3
6
2
5
1
2
5
0
15
20
00
5
1
2
5
0
6
2
5
2
2
0
0
Figure C16‐14: URM Wall Details
C16.6.2 Calculating Required Capacity
V* = 88.6 kN In‐plane lateral force
vo* = 5.8 kPa Out‐of‐plane lateral pressure
C16.6.2.1 Required In‐plane Capacity
∗ ∗
M* = 88.6 x 3.2 = 283.5 kNm
C16.6.2.2 Required Out‐of‐plane Capacity
l v∗ h
M∗
8
C16.6.3 Calculating Tendon Stress at Nominal Strength
It was established from the preliminary assessment that the out‐of‐plane strength of wall is
more critical than in‐plane strength, so is designed first in this instance.
C16.6.3.1 Tendon Stress at Nominal Out‐of‐plane Strength
a
M∗ ∅ N W nf A d
2
W 0.5γ b l h
Where: γ Density of Masonry = 18.0 kN/m3
C16‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
N W 29.7 2.5 = 32.2 kN
Using a permissible maximum tendon stress ate nominal strength = 0.7fpu, the nominal out‐of‐
plane flexural strength of the wall is established as,
d 0.5b d 0.11 m
N W
65.9 4 129.6
0.012 m
0.85 10700 5.0
Substituting the values of a and d, equation for nominal strength yields,
.
∅ M 0.85 65.9 4 129.6 0.11 = 51.65 kNm
As ∅ M M ∗ , fps assumed is sufficient to provide the wall required out‐of‐plane strength.
C16.6.4 Calculating Effective Tendon Stress
As the governing mode of failure is out‐of‐plane, the calculation of post‐tensioning force
magnitude will be based on out‐of‐plane required tendon stress.
d 0.11 m
d f A
f f 0.03E 1 0.97
h f′ d
0.11 1875 98.7
1176 f 5400 1 0.97
3.2 10.7 5000 110
f 1176 185.6 1 0.030 = 996.0 MPa
C16.6.5 Checking for In‐plane Strength
First checking for cracking moment capacity of the wall,
where
W γ b l h
Where: γ Density of Masonry = 18.0 kN/m3
C16‐17
Section C16 ‐ Unbonded Post‐tensioning
65.9 4 98.3
220 5
0.417 MPa
6
0.471 0.22 5 10
6
382.58 kNm
∅ M 325.19 kNm M* = 283.5 kNm
As ∅ M M ∗ , fps assumed is sufficient to provide the wall required out‐of‐plane strength.
C16.6.6 Calculating Post‐tensioning Losses
f f 3.7 0.55
f f 0.18f
f 0.82f
996.0 0.82f
f 1214.6 MPa < 0.7fpu = 1313 MPa
So, the initial post‐tensioning force is within permissible limits.
C16.6.7 Calculating Initial Post‐tensioning Force
1214.6 98.7
P f A 119.9 kN
1000
C16.6.8 Checking for Bearing Stress
Let a bearing plate of 200 mm × 200 mm is used at anchorage.
A 0.04m
f, 10.7 MPa
0.6f , 6.4 MPa
As 0.6f , , so bearing stresses are in permissible limit.
C16‐18
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C16.7 References
Al‐Manaseer, A. A. and Neis, V. V. (1987). “Load tests on post‐tensioned masonry wall panels.”
ACI Structural Journal, 84(3), 467‐472.
Bean Popehn, J. R., Schultz, A. E., Lu, M., Stolarski, H. K. and Ojard, N.J. (2008). "Influence of
transverse loading on the stability of slender unreinforced masonry walls," Engineering
Structures, 30(10), 2830‐2839.
Ganz, H. R. (1993) “Posttensioning masonry structures,” VSL International Limited, VSL Report
Series, Berne, Switzerland.
Ganz, H. R. and Shaw, G. (1997) “Stressing masonry’s future,” Civil Engineering (New York).
67(1), 42‐45.
Ismail, N., Lazzarini, D. L., Laursen, P. T., and Ingham J. M. (2011). “Seismic performance of face
loaded unreinforced masonry walls retrofitted using posttensioning,” Australian Journal of
Structural Engineering, In Press.
Ismail, N. (2011). “Selected strengthening techniques for the seismic retrofit of unreinforced
masonry walls.” PhD thesis, Department of Civil and Environmental Engineering, The
University of Auckland, Auckland, New Zealand.
Krause, G. L., Devalapura, R., and Tadros, M. K. (1996). “Testing of Prestressed Clay‐Brick Walls.”
1996 CCMS of the ASCE Symposium in Conjunction with Structures Congress XIV, Chicago, IL,
United States, 15 ‐ 18 April.
Laursen, P. T, and Ingham, J. M. (2004). "Structural testing of large‐scale posttensioned concrete
masonry walls." Journal of Structural Engineering, 130(10), 1497‐1505.
MSJC. (2005). ACI 530‐05/ASCE 5‐05/TMS 402‐05 Building code requirements for masonry
structures and specification for masonry structures. The Masonry Standards Joint
Committee, United States, 2005.
Rosenboom,O. A., and Kowalsky,M. J. (2004). "Reversed in‐plane cyclic behavior of
posttensioned clay brick masonry walls." Journal of Structural Engineering, 130(5), 787‐798.
Standards New Zealand. (2004). NZS 4230: 2004 Design of reinforced concrete masonry
structures. Standards New Zealand, Wellington, New Zealand.
Wight, G. D., and Ingham, J. M. (2008). "Tendon stress in unbonded posttensioned masonry
walls at nominal in‐plane strength." Journal of Structural Engineering, 134(6), 938‐946.
Wight, G. D., Ingham, J. M., and Kowalsky, M. J. (2006). "Shaketable testing of rectangular post‐
tensioned concrete masonry walls." ACI Structural Journal, 103(4), 587‐595.
C16‐19
Section C16 ‐ Unbonded Post‐tensioning
C16‐20
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer
Systems
C17.2 Notation
Afrp Total FRP area intersected by the tension crack (mm2)
bfrp Total thickness of FRP system (mm)
bw Wall thickness (mm)
d Effective depth of masonry wall (mm)
Efrp FRP modulus of elasticity in the primary fibre direction (kN/mm2)
f’m Masonry compressive strength (N/mm2)
f’tm Masonry tensile strength (N/mm2)
ffrpe Effective stress in FRP system just before the nominal shear strength of the section is
reached (N/mm2)
hw Wall height (mm)
lw Wall length (mm)
nc Number of FRP strips intersected by diagonal shear crack
nt Total number of strips in the FRP system
s Vertical spacing of FRP strips (mm)
V* Design shear force at section (kN)
Vdt Shear strength corresponding to diagonal tension failure mode (kN)
Vfrp Nominal shear strength contribution of FRP (kN)
Vm Nominal shear strength contribution of masonry (kN)
Vn Nominal shear strength of section (kN)
Vs Shear strength corresponding to bed‐joint sliding failure mode (kN)
Vtc Shear strength corresponding to toe crushing failure mode (kN)
wfrp Width of each FRP strip (mm)
α Angle between the FRP strip or primary fibre of continuous fabric and horizontal axis
(degrees)
frpe Effective strain in FRP system just before the nominal shear strength of the section is
reached
frpu Design FRP tensile rupture strain
ϕ Strength reduction factor for shear as in NZS 4230:2004
frp FRP reinforcement index
ρfrp Area reinforcement ratio
ρfrph Horizontal area reinforcement ratio
ρfrpv Vertical area reinforcement ratio
ρfrpi Inclined area reinforcement ratio
C17‐1
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
C17.2 Scope
The scope of this guideline is limited to shear retrofit design for in‐plane loaded walls. For out‐
of‐plane loaded walls and in‐plane loaded walls deemed to fail in flexure, the reader is referred
to other design documents (ACI 2010; CNR 2004; Galati et al. 2005; Garbin et al. 2007).
Most data used for the preparation of this design document was obtained from tests on
unreinforced clay brick masonry walls being either one leaf thick or two leaves thick. A limited
number of previously reported tests (Haroun and Mosallam 2002) suggest the suitability of
extrapolating the recommendations of this design document for applications to thicker walls,
however appropriate strength reduction factors are recommended for walls thicker than two
masonry leaves.
C17.3 General Design Information
The strength reduction factor when using the nominal shear strength of FRP retrofitted systems
shall be in accordance with Standards New Zealand (2004) and shall be taken as follows.
C17.4 Shear Strength Provided by Fibre Reinforced Polymer
FRP materials are formed by the impregnation of high‐strength fibres (generally glass fibres,
carbon fibres or aramid fibres) in epoxy resin. For retrofit applications, surface bonded FRP
materials are available in two principal forms (see Figure C17‐1):
(a) Fabric (glass) (b) Plate (carbon)
Figure C17‐1: Examples of FRP systems.
1. Fabrics: A fabric is an assembly of high‐strength fibres, which is generally impregnated at site
with epoxy to form FRP, and the impregnated fabric is then bonded to the wall, which is
precured with epoxy. The fabric systems are flexible, and available in the form of long sheets,
C17‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
which are cut to continuous fabrics covering the whole substrate, or cut to long narrow strips.
Both glass and carbon FRP fabrics are widely used in New Zealand.
2. Plates: FRP plates are produced by pultrusion, which is an automated continuous system (and
takes place in an industrial unit) involving pulling of fibres through resin system into a heated
die. FRP plates are applied to the substrate using epoxy. Carbon FRP plates are more common
than glass FRP plates.
In this design guide, the term ‘strip’ refers to long narrow elements of FRP, which could be either
plates or strips cut out from fabrics. The term ‘continuous fabric’ refers to FRP fabrics covering
the whole wall surface.
FRP behaves linearly until its sudden rupture. However for URM‐retrofitted walls, the FRP
system rarely reaches its rupture (ultimate or maximum) strength, as the failure usually takes
place due to debonding of FRP from the clay brick masonry (i.e. the loss of bond at the FRP‐wall
interface). Consequently, the peak stress assumed in retrofit is associated with the product of
the FRP modulus of elasticity and the calculated effective strain.
FRP systems can be applied in a number of orientations (see Error! Not a valid bookmark self‐
reference. and Figure C17‐3). Inclined FRP continuous fabrics are not common and are, hence,
not included in this document. The following factors should be considered when designing FRP
systems.
1. Architectural/heritage concerns:
2. Construction issues
(a) The application of FRP plates is more time consuming than the application of FRP fabrics
and strips because FRP plates are less flexible than fabrics and strips and require an
almost plane substrate for their application. FRP fabrics and strips, being quite flexible,
can better accommodate any substrate surface imperfections.
(b) The preparation of continuous fabrics requires a larger floor space at site than is
required for FRP strips.
C17‐3
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
Horizontal FRP systems
Combined horizontal‐vertical FRP systems
(a) Uniaxial strips (b) Biaxial strips
Inclined FRP systems
Figure C17‐2: Examples of non‐vertical FRP retrofit systems.
(a) Strips (b) Continuous fabric
Figure C17‐3: Examples of vertical FRP retrofit systems.
C17‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
3. Cost
(a) Carbon FRP is more expensive than glass FRP.
(b) Cost of application of strips is higher than the cost of application of continuous fabrics.
New Zealand FRP applicators generally prefer continuous fabrics rather than strips, as
the application of continuous fabrics involves minimal fabric cutting and less surface
preparation. Also, the installation of plates is more cumbersome (see note (a) under
‘construction issues’).
4. Strength
(a) Carbon FRP systems are stiffer and stronger than glass FRP systems.
(b) Inclined plates are greatly effective in increasing shear strength of URM walls. Such
systems also result in high stiffness, and are suitable when high system strength and
low displacements are desired.
(c) Use of a number of FRP strips is preferred rather than only a single strip, as multiple
strips add to redundancy of the system.
(d) Combined horizontal‐vertical systems are recommended as compared to only
horizontal systems.
(e) If inclined FRP systems are to be used, it is better to use FRP along both the wall
diagonals. Such an arrangement is also useful for cyclic earthquake loads.
(f) Where possible, FRP systems should be extended a distance, ld, beyond the wall (see
Figure C17‐4) to the adjacent structural members (floor, spandrels, window openings).
If such extension is not possible, FRP systems should be mechanically anchored along
the perimeter. Mechanical anchors could be in the form of (i) fibre anchors, which are
inserted into the wall through the fabrics and left to dry, or (ii) reinforcing bars, which
are inserted into precut wall grooves after the application of fabrics. Other systems are
also available. Further information is available from manufacturers. The distance ld can
be calculated as follows:
(CNR 2004) (C17‐2)
2
Where: f’tm is the average tensile strength of masonry, and may be assumed equal to 0.1f’m in
the absence of test data.
The design of FRP systems has been presented in this guideline for an increase in shear strength
only. Seismic retrofit design involves other parameters in conjunction with shear strength, such
as ductility, stiffness, energy dissipation and toughness. Guidance for these parameters can be
obtained from experimental research, where required.
Recommendations herein have been developed from basic principles and experimental results,
with guidelines obtained from procedures presented in ACI (2010) and Garbin et al. (2007).
Experimental results reported in the literature for surface bonded FRP‐retrofitted clay brick URM
elements (Mahmood and Ingham 2011; Alcaino and Santa‐Maria 2008; Elgawady et al. 2007;
C17‐5
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
Wei, et al. 2007; Stratford et al. 2004; Wang et al. 2006; Chuang 2004; Santa‐Maria et al. 2004;
Haroun and Mosallam 2002; Valluzzi, et al. 2002) have been included for the formulation of
these design expressions. Other reported tests were not included due to missing data (To‐Nan,
et al. 2006; Seki, et al. 2008). A flowchart presenting the design procedure for Surface Bonded
FRP systems is given in Figure C17‐5.
Figure C17‐4: FRP continuous fabric system extended ld beyond the top of the doors. The
vertical FRP system was designed for the pier between the two doors.
C17.4.1 Limitations
FRP is considered ineffective in increasing shear strength of a URM wall that is likely to fail by toe
crushing failure (ACI 2010). Hence, the provisions of this design guide apply only to walls that are
likely to fail in a shear‐controlled mode.
FRP horizontal strips and continuous fabrics may be of limited use in walls which are prone to
horizontal sliding due to weak mortar and low axial load, as horizontal crack may not intersect
any FRP strip.
C17.4.2 General Expression for Non‐Vertical FRP Retrofit Systems
An expression for Vfrp is developed from the truss analogy (Figure C17‐6) by considering a
number of parallel FRP strips, oriented at an angle to the wall base, and crossing a 45o
diagonal tension crack in a URM wall loaded by a horizontal in‐plane force as shown. The wall
has a length, lw, greater than the wall height, h. The shear force resisted by FRP strips is equal to
the sum of the horizontal components of the tensile forces developed in the strips in accordance
with Equation C17.3.
cos (C17‐3)
Where: Afrpi is the cross‐sectional area of each FRP strip with a vertical spacing, s, ffrpe is the
effective tensile stress in each strip, and nc is the number of strips intersecting the crack
and can be calculated from the geometry (see Figure C17‐6):
C17‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Compute V*
Compute Vm = min (Vs, Vdt, Vtc)
V*>Vm V*≤Vm
Check if FRP retrofit can be Retrofit not
designed required
Does a shear‐controlled failure mode control?
i.e Vs or Vdt < Vtc
If yes, design If no, explore a different
FRP retrofit retrofit technique
(Vfrp)required = V*/ Vm
Assume a retrofit system and compute Vfrp in accordance with section 15.5. Use
horizontal strips only if diagonal tension failure mode controls. If (Vfrp)provided. ≥
(Vfrp)required, OK, else redesign.
Check Vn ≥ V*
If yes, design is If no, redesign or explore a different retrofit
complete. technique.
Figure C17‐5: FRP design flowchart
C17‐7
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
Figure C17‐6: Truss analogy for FRP‐retrofitted URM wall
tan (C17‐4)
Or
1 tan
(C17‐5)
Where: s is the vertical spacing of FRP strips.
It can be shown that for a wall section with lw < hw,
1 tan
(C17‐6)
Hence, the general expression for nc is
1 tan
(C17‐7)
Where: d is the effective wall depth, and is computed using Equation C17‐8.
C17‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
min , (C17‐8)
nc must not exceed the total number of FRP strips, nt, comprising the FRP system. Hence,
1 tan
, (C17‐9)
The area of each strip, Afrpi, is computed as in Equation C17‐10.
(C17‐10)
Where: bfrp is the thickness of each FRP strip, and wfrp is the width of each FRP strip.
Substituting Afrpi in Equation C17‐3 gives Equations C17‐11.
cos (C17‐11)
Equation C17.11 is similar to the expression presented in ACI (2010). However ACI (2010)
presents expressions only for horizontal or combined horizontal‐vertical strip systems.
For horizontal continuous fabrics, the total width of the strips intersected by the diagonal crack
(ncwfrp) can be replaced in Equation C17.11 with the length of the continuous fabric intersected
by the diagonal crack √2 . For horizontal FRP retrofit systems, =0. Hence, for horizontal FRP
continuous fabric systems, Equation 17.11 can be written as:
√2 (C17‐12)
Or
1.4 (C17‐13)
Effective FRP stress has been related in the available literature to the FRP area reinforcement
ratio, FRP thickness, FRP modulus of elasticity, masonry compression strength and masonry
tensile strength. A comparison of major FRP design guidelines was performed, and it was found
that the ACI (2010) expression resulted in the most consistent predicted values. The ACI (2010)
model relates effective stress to FRP area reinforcement ratio, FRP modulus of elasticity and
masonry compression strength, with the product expressed as the FRP reinforcement index. The
effective stress has also been indirectly related to FRP thickness by limiting the product of FRP
thickness and effective FRP stress to 260 N/mm. The relation of effective stress to FRP thickness
has also been reported in CNR (2004), where the effective FRP stress is considered to be
inversely proportional to the square root of FRP thickness. Thus, for the same area of
C17‐9
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
reinforcement, thinner (and greater number of) FRP strips are likely to be more efficient than
thicker (and hence fewer) FRP strips.
The ACI (2010) model was modified for various retrofit systems using basic principles, and
effective strain values were obtained from reported tests involving FRP and clay brick masonry
walls (Figure C17‐7). Combined horizontal‐vertical systems generally provide higher shear
strength than exclusively horizontal systems, however only limited data was available for
combined horizontal‐vertical system, and effective strain values for horizontal systems were
adopted for combined horizontal‐vertical systems. For simplicity, only one conservative effective
strain relationship was presented for various non‐vertical FRP systems.
2.5
Experimental values
Horizontal
2.0 Horizontal-vertical
Inclined
1.5
fe (%)
1.0
Trendline
0.5
0.0
0.0 0.5 1.0 1.5 2.0
Figure C17‐7: Plot showing the relationship between fe and
Typical thickness, bfrp, design rupture strain, frpu, and modulus of elasticity, Efrp, values for
different FRP systems are given in Table C17‐1.
C17‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Table C17‐1: Typical values of FRP thickness, design rupture strain and modulus of
elasticity.
C17.4.3 Vertical FRP Retrofit System
Equations for non‐vertical FRP systems cannot be applied to vertical strips as s = ∞. Hence the
expression proposed by Garbin et al. (2007) has been adopted as follows:
(a) For FRP strips:
(C17‐14)
(b) For continuous FRP fabrics:
(C17‐15)
Relatively little data is available for shear strength enhancement using vertical FRP systems, and
hence a constant effective strain value is proposed.
C17.5 Design Examples
Examples are presented here to illustrate use of the design procedure that is detailed in Section
17.
C17.5.1 Example 15.1
Design an FRP retrofit system for a URM wall, measuring 5000 mm long x 2800 mm high x 225
mm thick. The factored shear force, V* = 180 kN. Any FRP reinforcement can be used, however
the reinforcement must be provided on one wall face only. It has been determined that the wall
is likely to fail by the formation of diagonal shear cracks.
f’m = 10 N/mm2
C17‐11
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
Vs = 225 kN; Vdt = 174 kN; Vtc = 253 kN
Vm = min (Vs, Vdt, Vtc) = min (225, 174, 273) = 174 kN
As the wall is likely to fail in a shear‐controlled (diagonal tension) mode, FRP shear
reinforcement can be designed.
V∗ ϕV 180 150.5
V 66 kN
ϕ 0.75
Wall properties:
lw = 5000 mm
hw = 2800 mm
bw = 225 mm
FRP reinforcement properties:
Use 50 mm wide, 1.2 mm thick CFRP strips oriented in the horizontal direction. Horizontal strips
can be used because diagonal tension failure is the controlling failure mode.
frp = 0.017 = 1.7%
Efrp = 165 kN/mm2
ffrpu = 2800 N/mm2
tfrp = 1.2 mm
s = 500 mm (trial)
wfrp = 50 mm
Effective stress, ffrpe, is computed as;
where
1 0.50 0.15
min 0.9 , 0.3
100 0.08 for 0.15
C17‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
frpe = minimum of [(0.08(0.30)‐0.9)/100, 0.3 x 0.017)
= minimum of [0.0024, 0.0051] = 0.0024
ffrpe = (0.0024) (165000) = 390 N/mm2
bfrp ffrpe = (1.2) (390) = 468 N/mm > 250 N/mm = use 250 N/mm
1 tan ,
d = min (lw, hw) = 2800 mm
= 0
nt = 6
2800
1 tan 0 , 6
440
6.4, 6
= 6
cos
Putting values
= 75 kN > (Vfrp)required (66 kN) OK
Vn = Vm + Vfrp
= 150.5 + (0.75 × 75) = 187 kN < Vtc (0.75 x 273 = 205 kN) OK
Figure C17‐8: FRP retrofit for wall in Example 15.1
C17‐13
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
C17.5.2 Example 15.2
Design the following piers in the front wall of the two‐storey high URM building shown in Figure
C17‐9.
(1) A typical pier between any two window openings on the upper floor. The pier measures
1000 mm long x 2000 mm high x 200 mm thick. V* = 20 kN
(2) The pier between the door openings on the ground floor. The pier measures 2000 mm long
x 2200 mm high x 300 mm thick. V* = 80 kN
FRP reinforcement can be applied on either face of the piers. f’m = 14 N/mm2.
(1) A typical pier between two upper piers
Vs = 30 kN; Vdt = 19.4 kN; Vtc = 9.9 kN
Vm = min (Vs, Vdt, Vtc) = min (30, 19.4, 9.9) = 9.9 kN
Vm = 0.75 × 9.9 = 7.4 kN
V* (20 kN) > Vm (7.4kN), shear reinforcement is required. However, as it was determined that
the wall will fail by toe crushing, FRP reinforcement cannot be designed, and an alternate retrofit
system is required.
Figure C17‐9: FRP retrofit for building in Example 15.2
C17‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(2) The pier between the doors on the ground floor
Vs = 90 kN; Vdt = 155 kN; Vtc = 158 kN
Vm = min (Vs, Vdt, Vtc) = min (90, 155, 158) = 90 kN
Vm = (0.75) (90) = 67.5 kN
V* (80 kN) > Vm (67.5 kN) Design shear reinforcement.
As the wall is likely to fail in a shear‐controlled (bed‐joint sliding) mode, FRP shear reinforcement
can be designed. Provide vertical reinforcement on one face only to inhibit joint sliding.
(Vfrp)required = (V* ‐ Vm)/ = (80 – 67.5)/0.75 = 16.7 kN
Pier properties:
lw = 2000 mm
hw = 2200 mm
bw = 300 mm
FRP reinforcement properties:
As a relatively smaller increase in shear strength is required, low‐strength GFRP continuous
fabrics can be used.
frp = 0.022 = 2.2%
Efrp = 26.1 kN/mm2
ffrpu = 575 N/mm2
bfrp = 0.35 mm
For continuous FRP fabrics:
min 0.001,0.05
frpe = minimum of [0.001, 0.05 x 0.022] = minimum [0.001, 0.0011]
= 0.001
ffrpe = (0.001) (26.1) (1000) = 26.1 N/mm2
bfrp ffrpe = (0.35) (26.1) = 9.1 N/mm < 80 N/mm, OK
C17‐15
Section C17 ‐ Surface Bonded Fibre Reinforced Polymer Systems
= 18.2 kN > (Vfrp)required (16.7 kN) OK
Vn = Vm + Vfrp
= 67.5 + (0.75 × 18.2) = 81.2 kN < Vtc (0.75 x 158 = 103.5 kN) OK
C17.6 References
ACI (2010). Guide for the design and construction of externally bonded fiber‐reinforced polymer
systems for strengthening unreinforced masonry structures (ACI440.7R‐10), American
Concrete Institute, Farmington Hills, MI.
Alcaino, P., and Santa‐Maria, H. (2008). "Experimental response of externally retrofitted
masonry walls subjected to shear loading." Journal of Composites for Construction, 12(5),
489‐498.
CNR. (2004). “Guide for the design and construction of externally bonded FRP systems for
strengthening existing structures (CNR‐DT 200/2004)”. Rome, Italy: Advisory Committee on
Technical Recommendations for Construction.
Chuang, S. (2004). “Seismic retrofitting techniques for existing unreinforced masonry
structures.” PhD Thesis. School of Geoscience Minerals and Civil Engineering, University of
South Australia. Australia
Elgawady, M. A., Lestuzzi, P., and Badoux, M. (2007). "Static cyclic response of masonry walls
retrofitted with fiber‐reinforced polymers." Journal of Composites for Construction, 11(1),
50‐61.
Galati, N., Garbin, E., Tumialam, G., Nanni, A. (2005). “Design guidelines for masonry structures:
out of plane loads.” Proc., 7th International Symposium on Fiber‐Reinforced Polymer (FRP)
Reinforcement for Concrete Structures (FRPRCS‐7), Kansas City, USA, 269‐288.
Garbin, E., Galati, E., Nanni, A., Modena, C., and Valluzzi, M. R. (2007). "Provisional design
guidelines for strengthening of masonry structures subject to in‐plane loading." Tenth North
American Masonry Conference (10NAMC) St. Louis, MO., 3‐6 June 2007
Grando, S., Valluzzi, M. R., Tumialan, J. G., and Nanni, A. "Shear strengthening of URM clay walls
with FRP systems." Proc., 6th International Symposium on Fibre‐Reinforced Polymer (FRP)
Reinforcement for Concrete Structures (FRPRCS‐6), World Scientific Pub Co Inc, Singapore,
1229‐1238.
Haroun, M. A., & Mosallam, A. S. (2002). “Cyclic shear test of multi‐wythe existing brick wall:
retrofitted by a single layer of TYFO SEH‐51A applied on one side.” San Diego, California:
University of California, Irvine and California State University, Fresno.
Mahmood, H., and Ingham, J. M. (2011). "Diagonal compression testing of FRP‐retrofitted clay
brick masonry wallettes." ASCE Journal of Composites for Construction, In Press.
Russo, S., Gottardo, R. and Codato, D. "Effect of FRP mesh reinforcement on shear capacity and
deformability of masonry walls." Proc., 6th International Symposium on Fibre‐Reinforced
Polymer (FRP) Reinforcement for Concrete Structures (FRPRCS‐6), World Scientific Pub Co
Inc, Singapore, 1239‐1248.
Santa‐Maria, H., Duarte, G., and Garib, A. (2004). "Experimental investigation of masonry panels
externally strengthened with CFRP strips and fabric subjected to in‐plane shear load." Proc.,
Thirteenth World Conference on Earthquake Engineering (15WCEE), Canadian Association
for Earthquake Engineering and International Association for Earthquake Engineering (IAEE),
Vancouver, Canada, Paper no. 1627.
C17‐16
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Seki, M., Vacareanu, R., Saito, T., Cotofana, D., Lozinca, E., Popa, V. and Chesca, A. B. (2008).
"Cyclic shear tests on plain and FRP retrofitted masonry walls." Proc., 14th World
Conference on Earthquake Engineering (14WCEE), International Association for Earthquake
Engineering, Beijing, China.
Standards New Zealand (2004). NZS4230:2004 ‐ Design of reinforced concrete masonry
structures, Stanards New Zealand, Wellington, New Zealand.
Stratford, T., Pascale, G., Manfroni, O., and Bonfiglioli, B. (2004). "Shear strengthening masonry
panels with sheet glass‐fiber reinforced polymer." Journal of Composites for Construction,
8(5), 434‐443.
To‐Nan, C., Ja‐Shian, C., Yen‐Chih, L., and Yu‐Chun, L. (2006). "Performance of URM walls
strengthened with CFRP under horizontal cyclic loading." First European Conference on
Earthquake Engineering and Seismology (1st ECEES), European Association of Earthquake
Engineering and European Seismological Commission, Geneva, Switzerland, 3‐8 September
2006, Paper no. 299.
Valluzzi, M. R., Tinazzi, D., and Modena, C. (2002). "Shear behavior of masonry panels
strengthened by FRP strips." Construction and Building Materials, 16(7), 409‐416.
Wang, Q., Chai, Z., Huang, Y., Yang, Y. and Zhang, Y. (2006). "Seismic shear capacity of brick
masonry wall reinforced by GFRP." Asian Journal of Civil Engineering, 7(6), 563‐580.
Wei, C. Q., Zhou, X. G., & Ye, L. P. (2007). “Experimental study of masonry walls strengthened
with CFRP.” Structural Engineering and Mechanics, 25(6), 675‐690.
C17‐17
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C18 ‐ Near Surface Mounted (NSM) Fibre
Reinforced Polymer Systems
C18.1 Notation
ad Demand acceleration (units of ‘g’, acceleration normalised by acceleration of gravity)
Ap Cross‐sectional area of FRP, mm2
bf Width of the groove cut, mm
bp Width of FRP strip, mm
c Depth of compression block (neutral axis location), mm
Cm Masonry compressive force, kN
df Depth of the groove cut, mm
(EA)p Axial stiffness of the FRP, MPamm2
Ep Modulus of elasticity of FRP, MPa
Em Masonry modulus of elasticity, MPa
f’b Average brick compressive strength, MPa
f’bl Lower characteristic brick compressive strength, MPa
fd The (minimum) design compressive stress on the bed joint at the cross‐section under
consideration, MPa
f’fb Characteristic flexural bond strength of masonry, MPa
f’m Masonry compressive stress, MPa
f’up FRP tensile strength, MPa
fut Brick rupture strength, MPa
Lper Perimeter of the debonding failure plane, mm
PIC Axial force in the FRP required to cause the onset of intermediate crack (IC) debonding
in clay masonry, kN
Prupture Tensile rupture capacity of FRP, kN
Mch Horizontal bending capacity of the URM wall, kNm/m
M* Design moment at section, kNm
Mn Nominal moment resistance of wall section, kNm
Nt Overburden weight, kN
Nn Nominal axial compression force, kN
V* Design shear force at section, kN
Vdt Shear resistance corresponding to diagonal tension failure mode, kN
Vjs Shear resistance corresponding to joint sliding deformation mode, kN
Vm Nominal shear resistance contribution of masonry, kN
Vn Nominal shear resistance of section, kN
Vnsm Effective shear resistance of NSM FRP, kN
Vtc Shear resistance corresponding to toe compression failure mode, kN
s Design strip spacing, m
tb Distance from edge of strip to edge of groove cut, mm
td Distance from edge of strip to edge of groove cut, mm
C18‐1
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
tm Thickness of URM wall, m
Tp FRP tensile force, kN
tp Thickness of FRP strip, mm
wd Uniformly distributed load, kN/m
z Lever arm, mm
Zd Section modulus of the bedded area , mm3/m
Zu Lateral section modulus of the masonry units, in cubic mm per m length based on the
gross cross‐section
γ Specific weight of masonry, kN/m3
db FRP debonding strain
m Masonry strain
φ Strength reduction factor for flexure
φch Strength reduction factor for horizontal bending of masonry in accordance with
AS 3700
f Failure plane aspect ratio
C18.2 Scope
Using fibre reinforced polymer (FRP) material to retrofit URM walls is an established technique
for strengthening and increasing the ductility capacity of URM walls subjected to in‐plane and
out‐of‐plane earthquake loading. Externally bonded (EB) FRP sheets or plates (Section 17) and
near‐surface mounted (NSM) FRP bars or strips, are the two application techniques that are
commonly used. Using the NSM technique provides several advantages over EB reinforcement
which include: significantly higher axial strain at debonding, protection from fire and the
environment, minimal impact upon the aesthetics of the structure (Goodwin et al. 2009), and
reduced construction time (Seracino et al. 2007; Petersen et al. 2009), thus providing a cost
effective and minimally‐invasive option for seismically strengthening URM buildings.
A sound understanding of the bond behaviour between concrete and FRP has been achieved
following extensive research (Stone et al. 2002; Hassan and Rizkalla 2003), and reliable analytical
models have been established (Seracino et al. 2007). Initial experimental validation of NSM FRP
as a technique for seismic improvement of URM buildings involved laboratory testing that
focused on strengthening single leaf modern masonry walls, and based on these experimental
results an analytical model that predicts the masonry to FRP bond behaviour was developed by
Willis et al. (2009b). The model was further verified by extensive experimental testing of multi‐
leaf historic clay brick masonry full scale laboratory built walls and on walls tested in‐situ (Dizhur
et al., 2010a, Dizhur et al. 2010b, Dizhur et al. 2010c, Dizhur et al. 2010d).
The NSM technique involves insertion of FRP strips. Typically a 15 mm wide × 1.2 mm thick strip
is positioned in a vertical groove that is cut into the wall so that it extends over the entire height
of the wall. Epoxy is used to bond the FRP strip into the masonry substrate. To ensure maximum
bond area the groove is typically entirely filled with epoxy prior to insertion of the CFRP strip
(typical installation procedure is illustrated in Figure C18‐1).
C18‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C18.3 Retrofit Parameters
Table 18‐1 indicates the key material properties used for the design of FRP reinforcement in
unreinforced masonry walls when subjected to out‐of‐plane bending, for the most commonly
used CFRP, unless otherwise provided by the FRP supplier.
Table 18‐1: Typical values of material properties for CFRP
Parameter Carbon FRP
Modulus of elasticity of CFRP, Ep 165 x 103 MPa
Tensile strength of CFRP, fup 2700 MPa
C18.4 In‐plane Design
C18.4.1 Shear Resistance Provided by NSM CFRP
This section has been intentionally left blank
C18.5 Out‐of‐plane Design
C18.5.1 General Design Information
The design process for a URM wall strengthened with vertical CFRP strips is based on the aim
that the preferred wall behaviour will be governed by the onset of intermediate crack (IC)
debonding in clay brick masonry of the FRP strip, rather than the development of alternative
more brittle failure modes, such as CFRP rupture, horizontal bending failure of masonry between
vertical FRP strips, or masonry crushing. The methodology presented is generic for either FRP
material (Carbon CFRP or Glass GFRP) and therefore final selection of the FRP material is at the
C18‐3
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
discretion of the design engineer. The usual flexural theory design assumptions for the cross‐
sectional analysis of a FRP reinforced section are:
(a) Plane sections remains plane after bending;
(b) Full composite action exists between the CFRP strip and the masonry interface (i.e. no slip
or opening‐up at the interface); and,
(c) The tensile resistance of the masonry is neglected (i.e. the FRP does not contribute to
strength until the masonry has cracked).
The design process for earthquake (or wind) loading depends on the horizontal acceleration (or
wind pressure) that the wall is subject to in the out‐of‐plane direction. The usual flexural theory
design for the cross‐sectional analysis of a FRP reinforced section applies, as illustrated in Figure
C18‐2.
tm tm
Nt
εm≤0.001
fmc Tp=ApEpεdb
c z
εdb
Cm
(a) Strain profile (b) Cross‐sectional forces
Figure C18‐2: Stress and strain profile
C18.5.2 Resistance of NSM FRP on Tension Surface of Loading
C18.5.2.1 Horizontal Spacing
The horizontal spacing, s, of the vertical FRP reinforcement is selected so that the strengthened
URM wall does not fail in horizontal bending between strips due to inertial load. The horizontal
bending capacity of the URM wall per metre of wall height, Mch, is calculated using Equations
18‐5 to 18‐7 as adopted from the Australian Standard on Masonry Structures (AS 3700:2010). AS
3700:2010 refers to a perpend spacing factor, kp, and three section moduli: the section modulus
of the bedded area, Zd, the lateral section modulus of masonry units, Zu, and the lateral section
modulus based on the contact area of the perpend joints, Zp. The equations are presented in a
simplified form where Zp is approximated to equal Zd and dimensions of a typical clay brick found
in New Zealand were used to calculate kp = 0.95.
C18.5.2.2 Analytical Prediction Model
Prediction of the flexural strength of NSM FRP retrofitted URM walls is based on an existing
generic analytical model that was initially developed by Seracino et al. (2007) for concrete and
C18‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
later modified by Willis et al. (2009b) for use with modern clay brick masonry. The axial force in
the FRP strip required to cause the onset of intermediate crack (IC) debonding in clay masonry,
PIC, is given by Equation C18‐1 (Willis et al. 2009b). The accuracy of this model has been verified
against 29 pull tests by Yang (2006), 34 pull tests by Dizhur (2011) and full scale laboratory‐built
and in‐situ wall tests (Willis et al. 2009a, Dizhur et al. 2010a). Due to the widespread use of
compressive strength of clay brick, f’b, by design engineers the equation is expressed in terms of
f’b by back substituting into Equation C18‐1 the MacGregor (1988) relationship given in Equation
C18‐2. From experimental observations td and tb are taken as 1 mm.
.
.
0. 988 (C18‐1)
0.53
0.53 (C18‐2)
Substituting Equation C18‐2 into C18‐1,
. .
0. 988 (C18‐3)
Simplifying Equation C18‐3,
. .
(C18‐4)
C18.5.2.3 Neutral Axis Location
The neutral axis location, c, is solved in accordance with Equation C18‐5, using Equation C18‐6 to
satisfy axial force equilibrium and Equation C18‐7 corresponding to plane sections remaining
plane. Masonry strains are assumed to be elastic such that an elastic masonry compression
stress distribution is appropriate.
(C18‐5)
N
1 (C18‐6)
2
2 (C18‐7)
3
Where: the lever arm, (see Figure C18‐2)
For simplicity, it is assumed for the NSM strip that the level of embedment is small compared to
tp. Therefore the neutral axis location, c, may be approximated by substituting Equation C18‐7
into C18‐6 to give:
C18‐5
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
4 (C18‐8)
2
Use ≅ to give
3
2
Where and
See Figure C18‐3 for the out‐of‐plane design procedure.
Determine M*
Select strip
spacing, s
No Calculate Mch
and check
Select strip
cross‐section,
calculate PIC
Check if PIC is less No
than tensile
rupture capacity
Calculate neutral
axis location, c
No Check masonry
compressive
stress
Calculate vertical
bending
capacity, Mn
No
ΦMn > M*
C18‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Figure C18‐3: Out‐of‐plane design procedure
C18.6 Other Considerations
Vertically oriented NSM FRP strips can be placed either through the brick units or the perpend
joints. Due to aesthetics and ease of placement it is likely that in most applications the strip
would be run through the perpend joints. However, it should be noted that positioning vertical
NSM FRP strips through the perpend joints can cause a reduction in bond strength in the order
of 10%. However, this level of strength reduction may be deemed insignificant given the
beneficial effects of the relative ease of placement (i.e. cutting through half the number of brick
units) and the reduced aesthetic impact.
C18.6.1 Failure Modes
There are three typical modes of failure: gradual debonding of the CFRP strip from the substrate,
a sudden CFRP strip pull‐out, and shear failure of masonry as illustrated in Figure C18‐4.
(a) Debonding (b) Sudden pull‐out
Figure C18‐4: Typical wall failure mechanisms
C18.6.2 Experimental Results
In addition to experimental tests described in Section C18.5.2, NSM CFRP retrofitted walls were
tested by applying lateral face pressure using a system of inflatable airbags (Dizhur et al. 2010a),
with the typical test setup is illustrated in Figure C18‐5. The pressure in the airbag(s) was
increased until visible cracking within the vicinity of the CFRP strip occurred, with a typical crack
pattern that resulted being shown in Figure C18‐4(a). All walls were loaded and unloaded several
times to examine if repeated loading resulted in stiffness degradation. The total lateral load –
mid‐height displacement response for laboratory‐built and in‐situ tested walls are shown in
Figure C18‐6. The effective wall face pressure was calculated as the total measured lateral force
sensed by the load cells, divided by the wall total area. All walls exhibited ductile behavior with
the exception of AH2, where a sudden debonding of the CFRP strip occurred (Dizhur et al.
2010c).
C18‐7
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
(a) Two vertical CFRP (b) Single CFRP strip
strip retrofit (wall R1) retrofit (wall R2)
(c) Typical test setup for (d) Failure at maximum
cyclic out‐of‐plane loading (wall R4)
testing (wall R3)
Figure C18‐5: Experimental program
C18‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
30 20
R4-R 6
R4-A
10 AH-R
2 2 0.5
AH-A
Prediction
0 0 0 0
0 60 120 0 40 80
Mid-height displacement - Δ (mm) Mid-height displacement - Δ (mm)
(c) Wintec Block F (d) Avon House, AH1
18
Total lateral force (kN)
AH2-R
4
Face pressure (kPa)
AH2-A
12 Prediction
6 2
0 0
10 60 110
Mid-height displacement - Δ (mm)
(e) Avon House,
AH2
Figure C18‐6: Experimental results of NSM CFRP wall testing
C18.7 Design examples
C18.7.1 Example 18.1
In‐plane design example
This section has been intentionally left blank
C18‐9
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
C18.7.2 Example 18.2
Design an out‐of‐plane NSM FRP retrofit system for a URM wall, measuring 5000 mm long x
2800 mm high x 225 mm thick (2 leaf). The seismic acceleration is 1.0g with no axial forces acting
on the wall. The wall is to be strengthened using CFRP strips. Determine the required size and
spacing of the strips.
Wall properties and material properties obtained from in‐situ testing:
Compressive strength of masonry, f’m = 8 MPa
Compressive strength of brick, f’b = 25 MPa
Flexural bond strength of masonry, f’fb = 0.1 MPa
Masonry modulus of elasticity, Em = 2.5 GPa
Length of wall, L = 5000 mm
Height of wall, H = 2800 mm
Width of wall, t = 225 mm
CFRP reinforcement properties:
CFRP Modulus of Elasticity, Ep = 165 GPa
CFRP tensile strength, fup = 2700 MPa
Weight of wall = 16 kN/m3 × 2.8 m × 5 m × 225 mm = 50.4 kN
Horizontal seismic force = 1.0 × 50.4 kN = 50.4 kN or 18 kN/m
Design moment, M*
∗
18 2.8
17.6
8 8
Calculate horizontal bending strength of wall per metre height of wall:
1000 225
8437500 /
6
as there is no mortar pointing in the masonry wall)
Using equation 18‐5:
1.9∅ 1
0
1.9 0.6 √0.1 1 8437500 10 3.0 /
0.1
C18‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Using equation 18‐6:
3.8∅
Using equation 18‐7:
∅ 0.44 0.56
Therefore horizontal bending strength per metre height of wall is:
8 8 3.0
2500
1.0 0.225 16
Try 15 mm × 1.2 mm CFRP strips embedded into 6 mm × 25 mm grooves
. .
Where: Lper is calculated as:
≅4 2 4 2 15 1.2 35.2
and φf is calculated as:
1 1 15
≅ 5
2 2 1.2
. .
5 25 35.2 165 10 1.2 15 10 35.2
Check that PIC is less than the tensile rupture capacity of the CFRP strip, Prupture (=fup Ap)
Assume strip spacing of 1250 mm centre to centre.
C18‐11
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
.
= 0.125 16 0.225 6.3
Solve for the neutral axis location, c:
35.2 10
0.012
165 10 1.2 15
2 2 35.2 6.3 10
2.21
0.012 2.5 10 1250
Check that the masonry compressive stress is less than the masonry strength capacity, f’m
21.2
0.012 0.0012
225 21.2
0.0012 2.5 10 2.86 ′ 8 ∴
Calculate the vertical bending capacity of the FRP reinforced wall, Mn.
21.2
≅ 225 217.8
3 3
.
Therefore number of strips required 3.26 ∴ 4
.
Spacing of strips = 5000/4 = 1250 mm < 2500 mm ∴ OK
Adopt 15 mm × 1.2 mm CFRP strips embedded into 6 mm × 25 mm grooves spaced at
1250 mm centres on both sides of the wall.
C18‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
5000 225
Figure C18‐7: CFRP wall retrofit for Example C18.7.2
C18.8 References
Dizhur, D. (2011). “NSM CFRP retrofit of URM walls.” PhD thesis, Department of Civil and
Environmental Engineering, The University of Auckland, Auckland, New Zealand.
Dizhur, D., Derakhshan, H., Lumantarna, R., Ingham, J.M. and Griffith, M. C (2010a) “Out of plane
strengthening of unreinforced masonry walls using near surface mounted fibre reinforced
polymer strips.” Structural Engineering Society New Zealand (SESOC) Journal , 23(2), 91‐103
Dizhur, D., Derakhshan, H., Lumantarna, R. and Ingham, J. M. (2010b). “In‐situ out‐of‐plane
testing of unreinforced masonry wall segment in Wintec Block F building.” 2010 New
Zealand Society for Earthquake Engineering Technical Conference, Wellington, New Zealand,
26 ‐ 28 March.
Dizhur, D., Lumantarna, R., Derakhshan, H. and Ingham, J. M. (2010c). “In‐situ testing of a
residential unreinforced masonry building in Wellington, New Zealand.” 8th International
Masonry Conference, Dresden, Germany, 4 ‐ 7 July 2010.
Dizhur, D., Lumantarna, R., Derakhshan, H. and Ingham, J. M. (2010d). “Testing of an
earthquake‐damaged unreinforced masonry building.” 7th International Conference on
Urban Earthquake Engineering and 5th International Conference on Earthquake
Engineering, Tokyo, Japan, 3 ‐ 5 March.
Goodwin, C. O. (2008). "Architectural considerations in the seismic retrofit of unreinforced
masonry heritage buildings in New Zealand." Master of Architecture, M. Arch Thesis,
Department of Architecture and Planning, The University of Auckland, New Zealand.
Goodwin, C. O., Tonks, G. and Ingham, J. M. (2009). “Identifying heritage value in URM
buildings.” Journal of the Structural Engineering Society of New Zealand, 22(2), 16‐28.
C18‐13
Section C18 ‐ Near Surface Mounted (NSM) Fibre Reinforced Polymer Systems
Griffith, M. C., Magenes, G., Melis, G. and Picchi, L. (2003). “Evaluation of out‐of‐plane stability
of unreinforced masonry walls subjected to seismic excitation.” Journal of Earthquake
Engineering, 7(SPEC. 1), 141‐169.
Hassan, T. and Rizkalla, S. (2003). “Investigation of bond in concrete structures strengthened
with near surface mounted carbon fiber reinforced polymer strips.” Journal of Composites
for Construction, 7(3), 248‐257.
Korany, Y. and Drysdale, R. (2006). “Rehabilitation of masonry walls using unobtrusive FRP
techniques for enhanced out‐of‐plane seismic resistance.” Journal of Composites for
Construction, 10(3), 213‐222.
MacGregor, J. G. (1988). Reinforeced concrete: Mechanics and design. New Jersey: Prentice Hall,
Englewood Cliffs.
Mosallam, A. S. (2007). “Out‐of‐plane flexural behavior of unreinforced red brick walls
strengthened with FRP composites.” Composites Part B: Engineering, 38(5‐6), 559‐574.
Petersen, R. B., Masia, M. J. and Seracino, R. (2009). “Bond behavior of near‐surface mounted
FRP strips bonded to modern clay brick masonry prisms: Influence of strip orientation and
compression perpendicular to the strip.” Journal of Composites for Construction, 13(3),
169‐178.
Seracino, R., Jones, N. M., Ali, M. S. M., Page, M. W. and Oehlers, D. J. (2007). “Bond strength of
near‐surface mounted FRP strip‐to‐concrete joints.” Journal of Composites for Construction,
11(4), 401‐409.
Standards Australia (2001) AS 3700‐2001: Masonry structures. Sydney, Australia
Stone, D., Tumialan, G., Nanni, A. and Perretti, R. (2002). “Near surface mounted FRP
reinforcement: Application of an emerging technology.” Concrete International, 35(5),
42‐44.
Willis, C. R., Kashyap, J. and Griffith, M. C. (2009a). “Flexural behaviour of NSM CFRP retrofitted
masonry wallettes under static and cyclic loading.” FRPRCS‐9, Sydney, Australia, 13‐15 July
2009.
Willis, C. R., Yang, Q., Seracino, R. and Griffith, C. M. (2007). “Analysis of damaged masonry walls
retrofitted with vertical FRP strips subjected to out‐of‐plane loading.” 8th International
Symposium on Fibre Reinforced Polymer Reinforcement for Concrete Structures, FRPRCS‐8,
Patras, Greece, 16 – 18 July.
Willis, C. R., Yang, Q., Seracino, R. and Griffith, M. C. (2009b). “Bond behaviour of FRP‐to‐clay
brick masonry joints.” Engineering Structures, 31(11), 2580‐2587.
Yang, Q. (2006). “Seismic retrofit of URM walls with externally bonded FRP composites.” Masters
Thesis, School of Civil and Environmental Engineering, The University of Adelaide, Adelaide.
C18‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
C18‐15
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
C19.1 Notation
Afrs Total cross‐sectional area of FRS reinforcement, mm2
An Cross‐sectional area of wall, mm2
As Area of longitudinal steel reinforcement, mm2
Av Total cross‐sectional area of shear reinforcing steel, mm2
Be Effective flange width, mm
d Effective depth of masonry, mm
ds Depth of steel reinforcement embedment, mm
Efrs FRS modulus of elasticity, MPa
f’b Brick compressive strength, MPa
f’m Masonry compressive strength, MPa
f’frs FRS compressive strength, MPa
f’tfrs FRS tensile strength, MPa
fvy Design yield strength of shear reinforcing steel, MPa
fvcd Shear strength of FRS, MPa
H Wall height, m
kwt Reinforcement efficiency factor due to wall thickness
L Wall length, m
Lfrs Length of FRS applied to wall, m
M* Design moment, kNm
Mn Nominal moment capacity, kNm
N* Design axial compression force, kN
S Spacing of shear reinforcing steel, mm
SNSM Spacing of the near surface mounted steel reinforcement, mm
t Wall thickness prior to applying FRS, mm
tfrs Thickness of FRS applied to wall, m
V* Design shear force at section, kN
Vcd Shear strength provided by cement composite in FRS, kN
Vdt Shear resistance corresponding to diagonal tension failure mode Vf, kN
Vfd Shear strength provided by reinforcing fibre, kN
Vfrs Nominal shear resistance of FRS, kN
Vτfrs Nominal shear bond strength of FRS, kN
Vfrse Effective shear resistance of FRS, kN
Vjs Shear resistance corresponding to joint sliding deformation mode, kN
Vm Nominal shear resistance contribution of masonry, kN
Vn Nominal shear resistance of section, kN
Vs Shear strength provided by steel reinforcement, kN
Vtc Shear resistance corresponding to toe compression failure mode, kN
ws Width of NSM groove cut into masonry, mm
C19‐1
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
z Distance from centroid of tension reinforcement to centroid of compression
reinforcement, mm
αf’m 0.85f’m
αs Angle of shear reinforcing steel to horizontal member axis
βd Effective depth factor on shear strength of cement composite
βn Axial load factor on shear strength of cement composite
βp Ratio of axial reinforcement factor on shear strength of cement composite
Strength reduction factor for FRS, adopted as for concrete in NZS 3101:2006
C19.2 Scope
The FRS strengthening methodologies for unreinforced masonry walls are applicable for any FRS
regardless of the nature of constituent materials and reinforcing fibres, and therefore the final
selection of FRS is at the discretion of the designer. In the experimental campaign conducted to
form the methodologies, only Engineered Cementitious Composite (ECC) Shotcrete was used
due to its strain‐hardening characteristic and the capability to bond to masonry adequately
without the use of chemical bonding agents or steel connectors. The use of ECC as a
strengthening material was first verified by Li et al. (2000) and Kim et al. (2004), who applied ECC
to concrete slabs and measured an increased flexural capacity. The in‐plane strengthening effect
of ECC on masonry in‐fills was confirmed from testing conducted by Kesner & Billington (2005)
and Kyriakides & Billington (2008). Bruedern (2008) also tested masonry prisms strengthened
with ECC and verified the effectiveness.
When selecting other FRS types, the capability to adhere to a URM surface should be verified
unless physical connections are used to connect the FRS to URM. Currently there is no NZ or
ASTM standard testing method for determining shotcrete bond to substrate, but ACI (2008)
accepts the use of ASTM C1583 as a method of determining the direct tensile pull off bond
strength.
A design chart (see Figure C19‐1) is provided to indicate the sequence of strengthening
procedures that is advised when strengthening URM walls using FRS.
C19‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Does the URM wall
require strengthening
in both in‐plane and
out‐of‐plane direction?
Yes No
Go to flowchart in
Design for out‐of‐plane figure 19.2 for in‐plane
forces first, then in‐ design and flowchart in
plane forces figure 19.3 for out‐of‐
plane design
Figure C19‐1: Sequence of FRS strengthening procedures.
C19.3 In‐plane General Design Information
The in‐plane design procedure is shown in Figure C19‐2. Note that flexural strength of masonry
is not considered here as shear failures tend to be the dominant failure mode.
C19‐3
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
Determine V*
Determine Vm = min (Vjs, Vdt, Vtc)
фVm ≥ V*
фVm < V*
Retrofit not required
(Vfrse)required = (V*‐фVm)/ф
Is Vτfrs>Vfrs ?
Yes No
in‐plane design design shear
complete connectors
Figure C19‐2: In‐plane FRS design procedure
C19.3.2 Shear Resistance Provided by FRS
The shear strength that FRS can provide to a URM wall is dependent on the shear strength of the
FRS itself, and the amount of shear force that the masonry wall can transfer (shear bond
strength) to the FRS reinforcement before delamination occurs. The shear strength of FRS itself
is critical when loads are expected to be transferred into the FRS layer directly, such as loads
from diaphragms. In the case where the loads are from the URM wall that the FRS layer is
bonded to, the shear bond strength is the governing criteria.
C19.3.2.1 Shear Resistance of FRS
Equation 19‐5 is adopted from JSCE (2008) which provides a comprehensive evaluation on
various strength properties of high performance fibre reinforced composite. The design shear
capacity of FRS is the sum of the contribution from the cement composite, reinforcing fibre and
shear reinforcement.
Equation 19‐6 determines the shear strength provided by the cement composite in FRS,
considering the influence of member effective depth, the presence of axial reinforcement, and
C19‐4
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
the influence of bending moment as a result of axial force using the factors βd, βp and βn
respectively.
Equation 19‐13 determines the shear strength provided by the reinforcing fibre and is influenced
by the tensile strength of FRS. Typically the shear strength provided by the FRS is the dominant
component to the overall shear capacity unless additional steel reinforcing bars are used. If the
tensile strength of FRS is less than 1.5 MPa, then the contribution of the reinforcing fibre to
shear strength shall be neglected. Note that the original equation was also divided by tan βu,
where βu is the angle of orientation of a crack measured with respect to the horizontal and is
typically assumed to be 45°, but as tan 45° = 1.0 this term was removed from the equation.
Equation 19‐14 determines the shear strength provided by the shear reinforcement and is
equivalent to methodologies used in reinforced concrete design. The value of z = 0.72L is derived
from the two commonly adopted assumptions for reinforced masonry that the effective length is
0.8L and that z being 0.9Le.
C19.3.2.2 Shear Bond Resistance of FRS
Equation 19‐15 is based on JCI (2003). The original equation was used to predict FRP bond to
concrete and has been modified for FRS bond to URM by quantifying the discrepancies between
the predicted strength from the original equation and experimentally measured values. It should
be noted that debonding does not occur between FRS and masonry, but rather occurs within the
masonry itself near the FRS and masonry interface. As a result of the spalling failure of the
masonry surface, chemical bond agents applied between FRS and masonry do not contribute to
the shear bond strength and physical connectors are required to increase the bond of FRS to
masonry. Other debonding equations, such as developed by Willis et al. (2009), were also
compared with experimental values obtained but the discrepancy between the predicted values
and measured values was higher than predictions obtained from the use of the equation
suggested by JCI (2003).
kwt is a surface bonded reinforcement efficiency factor dependent on the thickness of the URM
wall, in leafs. For a two leaf URM wall with common bond (a header course every 4‐6 courses of
bricks), kwt is taken as 1.0. As the masonry wall thickness increases, the header courses between
each leaf are insufficient to connect the wall together to ensure a uniform deformation. The leaf
of the wall furthest from the FRS reinforcement then exhibits significantly more damage when
compared with the leaf that the FRS is bonded to.
C19.3.2.3 Shear Connector Design of FRS
Shear connectors may be used for two purposes. Firstly shear connectors may be used to
increase the bond capacity between FRS and URM and thus increase the value of Vτfrs and
secondly shear connectors may be used to connect different leafs of the URM wall together so
that they behave uniformly when subjected to loading, which increases the value of kwt, the
reinforcement efficiency factor.
C19‐5
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
C19.4 Out‐of‐plane General Design Information
The out‐of‐plane design philosophy for FRS reinforced URM walls is analogous to the flexural
design of reinforced concrete. In the experimental campaign, the FRS used (ECC) exhibited no
debonding with the URM wall regardless of the direction of out‐of‐plane loading. For other types
of FRS the bond capacity will need to be verified.
C19.4.1 Out‐of‐plane Resistance of Masonry Walls with FRS on Tension Surface of
Loading
When FRS is acting on the tension surface of loading, FRS is treated as the tension resisting
element and URM as the compression element.
An illustration of the distance of xt and xc when moment capacity is calculated about the axial
load is shown in Figure C19‐3. The out‐of‐plane design procedure is shown in Figure C19‐4.
Figure C19‐3: Distances of xt and xc with respect to axial load
C19‐6
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Is FRS going to be applied on both
surfaces of the wall?
Yes, design with FRS No, design with ECC on the compression
on the tension surface of loading first with NSM steel
suface reinforcement
Determine M* and Nominate the depth and spacing that the
N* steel reinforcment will be placed in
Assume Resolve forces and determine area of steel
thickness of FRS reinforcement required
Determine Determine the groove width and the
flexural strength effective flange width
Yes No Yes No
Figure C19‐4: Out‐of‐plane FRS design procedure
C19.4.2 Out‐of‐plane Resistance of Masonry Walls with FRS and Near Surface
Mounted (NSM) Steel Reinforcement Embedded on Compression Surface of
Loading
When FRS is acting on the compression surface of loading, the tensile capacity of the URM is
ignored and only FRS is assumed to be resisting the tensile forces. However, thin layers of FRS
are generally incapable of sustaining the flexural tension forces generated due to seismic loading
and therefore the use of Near Surface Mounted (NSM) steel reinforcement is recommended. An
illustration of NSM reinforcement is shown in Figure C19‐5, while Figure C19‐6 illustrates
Equation 19‐23 on the relationship between d, ds and tfrs.
C19‐7
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
Figure C19‐5: NSM reinforcement
Figure C19‐6: Illustration of FRS effective depth
C19.4.3 Design Examples
C19.4.4 Example 19.1
Design a FRS retrofit system for a URM wall, measuring 5000 mm long x 2800 mm high x 225 mm
thick (2 leafs). The factored in‐plane shear force, V*, is 180.0 kN and there is no axial force acting
on the wall. The FRS reinforcement must be provided on one wall surface only on the entire wall
length and no additional longitudinal steel reinforcement can be used. It has been determined
that the wall is likely to fail by the formation of diagonal shear cracks.
Compressive strength of masonry, f’m = 10 MPa
Compressive strength of brick, f’b = 20 MPa
Bed‐joint sliding resistance, Vjs = 225 kN
Diagonal tension resistance, Vdt = 174 kN
Toe crushing resistance, Vtc = 253 kN
Nominal shear resistance contribution of masonry, Vm = min (Vjs, Vdt, Vtc)
min (225, 174, 273) = 174 kN
Reduced nominal shear resistance,Vm = (0.75) (174) = 130.5 kN
V* (180 kN) > Vm (130.5)
Therefore design shear reinforcement.
C19‐8
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
(Vfrse)required = (V* ‐ Vm)/ = (180 – 130.5)/0.75 = 66 kN
Wall properties:
Length of wall, L = 5 m
Height of wall, H = 2.8 m
Width of wall, t = 0.225 m
FRS reinforcement properties:
FRS Modulus of Elasticity, Efrs = 16000 MPa
FRS compressive strength, f’frs = 40 MPa
FRS tensile strength, f’tfrs = 4 MPa
Vfrse = min (Vfrs, Vτfrs)
Evaluate Vfrs
Vfrs = Vcd + Vfd + Vs
Try 10 mm of FRS
Shear strength provided by the cement composite, Vcd
0.67 1.5
Where: pw = 0 (no longitudinal reinforcement)
1 ∗
1 (as there is no axial load)
Shear strength provided by reinforcing fibre, Vfd
Shear strength provided by shear reinforcement, Vs = 0
Therefore:
Nominal shear resistance of FRS, Vfrs = 144.0 kN
Evaluate nominal shear bond strength of FRS, Vτfrs
C19‐9
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
.
0.097 10
.
0.097 1.0 20 0.01 16000 5 2.8 395.6
Therefore,
Adopt 10 mm of FRS on a single surface
Note: while a lower thickness of FRS would be able to sustain the shear force, in practical
applications the minimum thickness should not be less than 10 mm as difficulties arise in
ensuring that the design thickness of FRS can be applied on a URM surface when the design
thickness is less than 10 mm.
C19.4.5 Example 19.2
Design a FRS retrofit system for a URM wall, measuring 5000 mm long x 2800 mm high x 225 mm
thick (2 leaf). The out‐of‐plane seismic acceleration is 1.0g and there are no axial forces acting on
the wall excepting wall selfweight. First determine the FRS thickness that would be required if
reinforcement can be applied on both surfaces, then determine the reinforcement that would
be required on the compression surface if only a single surface can be strengthened.
(1) FRS can be applied on both surfaces
Weight of wall = 16 kN/m3 × 2.8 m × 5 m × 225 mm = 50.4 kN
Horizontal seismic force = 1.0 × 50.4 kN = 50.4 kN or 18 kN/m
Design moment, M*, and required moment capacity, Mn
∗
18 2.8
17.6
8 8
∗
17.6
20.8
∅ 0.85
Self‐weight of upper half of wall, N*
∗
16 / 1.4 5 225 25.2
∗
25.2
29.6
∅ 0.85
Try a FRS thickness of 10 mm on a single surface
FRS force at failure is:
C19‐10
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
5 10 4 200
Taking moment about axial load
225 10 225 5.4
200 229.6 48.7 23.3
2 2
10 mm of FRS is sufficient if FRS is applied on both surfaces
(2) Now assuming that FRS can only be applied on a single surface, determine the
reinforcement required if FRS is on the compression surface of loading.
Nominate the spacing of the groove, in this example this has been chosen as 1 metre.
Determine the horizontal seismic force that the wall will be subjected to.
Given that the groove has a spacing of 1 metre, the width of the section analysed is also 1 metre.
The weight of the wall is then:
Using the seismic acceleration of 1.0 g, the horizontal seismic force then becomes:
Design moment,
∗
3.6 2.8
3.5
8 8
∗
3.5
4.2
∅ 0.85
∗
Self weight of upper wall, 16 / 1.4 1.0 225 5.0
∗
5.0
5.9
∅ 0.85
With the design nominal moment, Mn, being 4.2 kNm, nominate the groove depth and calculate
the reinforcement area required and consequently the groove width. It is recommended that
the depth of the groove does not reach the outer leaf of the URM wall as this may damage the
outer leaf (and therefore the appearance of the building exterior) during the construction
process.
Depth of reinforcement selected = 80 + 10 mm cover
0.0325
4.2 5.9 0.0325
C19‐11
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
4.0
Effective depth d is = 80 + 10 = 90 mm
Note that high performance FRS (such as ECC used in this experimental campaign) typically have
a significantly lower permeability when compared with normal structural concrete, therefore a
reduction in cover thickness can be used, consult the supplier regarding the recommended cover
thickness.
The effective flange width be on both side is the lesser of
(a) 16 50 16 10 210
(b) 2 50 2 80 210
(c) 50 750
(d) 1000
Try z = 0.9d
0.9 90 81
4.0
49.4
81
49.4
7.1
′ 0.83 40 210
7.1
90 86.5
2
Using the new z, C = 46.3 kN and a = 6.6 mm
Then C = 46.1 kN
Resolving forces vertically T = C‐Nn
Note that a D12 bar would be sufficient, but here a D14 bar with AS of 153.9 mm2 is selected to
show the calculation of the compression area centroid.
Note that as FRS applied on wall surface could be of very thin sections, and the effective flange
width could also be short depending on the geometries of the groove selected, it is possible that
the compression force will be outside the flange area and consequently the centroid of the
compression must be calculated.
C19‐12
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
82.9
Area of compression block 2497
′ 0.83 40
The compression area is comprised of
Adopt 100 mm groove depth
Adopt 10 mm of FRS thickness on masonry wall with grade 500 D14 reinforcing bar Near
Surface Mounted at 1 m spacing with a groove width of 50 mm and a groove depth of 100 mm.
C19.5 References
ACI‐Committee‐506 (2008). Guide to Fiber‐Reinforced Shotcrete. American Concrete Institute,
1‐18.
ASTM (2004). “Test Method for Tensile Strength of Concrete surfaces and the Bond Strength or
Tensile Strength of Concrete Repair and Overlay Materials by Direct Tension (Pull‐off
Method).” American Society for Testing and Materials. Philadelphia.
Bruedern, A‐E., Abecasis, D. & Mechtcerine, V. (2008). “Strengthening of Masonry Using Sprayed
Strain Hardening Cement‐Based Composites (SHCC).” Seventh International RILEM
Symposium on Fibre Reinforced Concrete: Design and Applications, Chennai (Madras), India,
451‐460, 17‐19 September.
BSCC (2000). “Prestandard and Commentary for the Seismic Rehabilitation of Buildings.” FEMA
356. Building Society Safety Council. Washington, USA.
CEN (2005). “Designs of Structures for Earthquake Resistance Part 3: Assessment and
Retrofitting of Buildings” Eurocode 8. Brussels, Belgium.
DIN 1053‐100 (2004). Masonry: Design on the Basis of Semi‐Probabilistic Safety Concept. NABau,
Berlin.
Galati, N., Garbin, E. & Nanni, A. (2005). Design Guideline for the Strengthening of Unreinforced
Masonry Structures Using Fibre Reinforced Polymers (FRP) Systems. University of
Missouri‐Rolla.
JCI‐SF6 (1990). “Method of Test for Shear Strength of Fibre Reinforced Concrete.” Japan
Concrete Institute. Mubanchi, Yotsuya 1‐Chrome, Shinjuku‐ku, Tokyo, Japan.
C19‐13
Section C19 ‐ Fibre Reinforced Shotcrete (FRS)
JCI (2003). “Technical report of technical committee on retrofit technology.” International
Symposium on Latest Achievement of Technology and Research on Retrofitting Concrete
Structures, Kyoto, Japan, 1‐8, 14‐15 July.
JSCE (2008). “Recommendations for design and construction of high performance riber
reinforced cement composites with multiple fine cracks (HPFRCC).” Japan Society of Civil
Engineers, 1‐88.
Kesner, K. & Billington, S. L. (2005). “Investigation of infill panels made from engineered
cementitious composites for seismic strengthening and retrofit.” Journal of Structural
Engineering ASCE, 131(11), 1712‐1720.
Kim, Y. Y., Fischer, G., Lim, Y. M. & Li, V. C. (2004). “Mechanical performance of sprayed
engineered cementitous composite using wet‐mix shotcreting process for repair
applications.” ACI Materials Journal, 101(1), 42‐49.
Kyriakides, M. A. & Billington, S. L. (2008). “Seismic retrofit of masonry‐infilled non‐ductile
reinforced concrete frames using sprayable ductile fiber‐reinforced cementitous
composites.” The 14th World Conference on Earthquake Engineering, Beijing, China, 1‐7,
12 ‐ 17 October.
Li, V. C., Horii, H., Kabele, P., Kanda, T. & Lim, Y. M. (2000). “Repair and retrofit with engineered
cementitious composites.” Engineering Fracture Mechanics, 65(2), 317‐334.
MSJC (2002). “Building code requirement for masonry structures.” ACI 530‐04/ASCE 5‐04/TMS
402‐04. Masonry Standards Joint Committee, Boulder, USA.
Standards New Zealand. (2006). NZS 3101:2006, Concrete Structures Standard Part1: The Design
of Concrete Structures. Standards New Zealand, Wellington, New Zealand.
Willis, C. R. & Yang, Q., Seracino, R. & Griffith M. C. (2009). “Bond behaviour of FRP‐to‐clap brick
masonry joints.” Engineering Structures, 31, 2580‐2587.
C19‐14
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C20 ‐ Near Surface Mounted Steel
Reinforcement
This section has been intentionally left blank.
C20‐1
Section C20 ‐ Near Surface Mounted Steel Reinforcement
This section has been intentionally left blank.
C20‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C21 ‐ Textile Reinforced Mortars
This section has been intentionally left blank.
C21‐1
Section C21 ‐ Textile Reinforced Mortars
This section has been intentionally left blank.
C21‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C22 ‐ Restraining Parapets and Chimneys
This section has been intentionally left blank.
C22‐1
Section C22 ‐ Restraining Parapets and Chimneys
This section has been intentionally left blank.
C22‐2
Commentary to Assessment and Improvement of Unreinforced Masonry Buildings for 12/2011
Earthquake Resistance
Section C23 ‐ Pounding
This section has been intentionally left blank.
C23‐1