Construction and Building Materials 182 (2018) 52–65

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Mechanical properties of high tensile steel cables at elevated

temperatures
Yong Du a,b,c,⇑, Jing-zhan Peng a, J.Y. Richard Liew a,c, Gou-qiang Li b
a
School of Civil Engineering, Nanjing Tech University, No. 30 Puzhu Road(S), Nanjing 211816, China
b
College of Civil Engineering, Tongji University, No. 1239 Siping Road, Shanghai 200092, China
c
Department of Civil and Environmental Engineering, National University of Singapore, 21 Lower Kent Ridge Road, Singapore 119077, Singapore

h i g h l i g h t s

A full range of temperature dependent stress-strain relationship of high tensile steel cable is proposed, which considers the stress hardening path.

The reduction factors of proportional limit, elastic modulus, yield strength and ultimate strength of high tensile steel cables are proposed.

The coefficient of thermal elongation of high tensile steel cable as a function of temperature is proposed for design implementation.

a r t i c l e i n f o a b s t r a c t

Article history: This study is motivated by increasingly prevalent use of cable-tensioned spatial steel structures and sus-
Received 13 January 2018 pension bridges. Fire is one of the extreme conditions that need to be taken into consideration in the
Received in revised form 30 May 2018 design of such structures. In this paper, steady-state tests have been conducted on steel cables with ten-
Accepted 2 June 2018
sile strength of 1860 MPa, which consist of a group of 7-wire twisted strands, to study their full range of
Available online 18 June 2018
stress strain relationships at elevated temperature. Thermal elongation test of steel cables has also been
conducted. A charge-coupled device camera (CCDC) system is used to capture the full range of stress-
Keywords:
strain relationship of high tensile strength steel cables till rapture at elevated temperature. The reduction
Cable tensioned structures
Fire safety
factors of proportional limit, elastic modules, effective yield strength and rupture strength at different
High tensile steel cable temperature were obtained from the steady state tests and compared with that proposed by EN 1992-
Thermal mechanical properties 1-2. The experimental work discovered that EN 1992-1-2 overestimated effective strain up to 2% and
ignored the stress-hardening phase for high tensile strength cables within the full temperature range.
The effective yield strength with 1.25% strain and a full range of stress-stain model considering stress-
hardening phase are proposed. Finally, several sets of reduction factors and thermal elongation coefficient
as a function of temperature have been proposed by fitting the test results. The reduction factors of pre-
stressing strands proposed by EN 1992-1-2 for pre-stressing concrete is found not suitable for steel cables
which are widely used for pre-tensioned steel structures. The reduction factors proposed in the present
paper are found to be reasonable for steel cables. The experimental work also shows that the mechanical
properties of steel cables at elevated temperature depends on whether the cable is made from straight
wires or twisted wires.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction structures, suspension domes and suspension bridges etc. [1] as

With the similar mechanical behaviors as the pre-stressed ten-
dons used in the pre-stressing of concrete structures, high tensile
steel cables have been widely used in tensile steel construction
for long span structures such as cable-net structures, beam string
⇑ Corresponding author at: School of Civil Engineering, Nanjing Tech University,
No. 30 Puzhu Road(S), Nanjing 211816, China.
E-mail address: yongdu_mail@njtech.edu.cn (Y. Du).
shown in Fig. 1. Fire is one of the extreme conditions that need
to be taken into account in the design of such structures. Therefore,
the mechanical properties of high tensile steel cables at elevated
temperature should be accurately determined to study the fire
safety of tensile steel structures. The mechanicalal properties at
elevated temperature, such as the effective yield strength, elastic
modulus, ultimate strength and strain etc., are the important
parameters for non-linear analysis of tensile structures study their
response behavior in various fire scenarios.

https://doi.org/10.1016/j.conbuildmat.2018.06.012
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
(a) The first tensile steel cable (b) A suspension bridge in
net roof in the world China
established in 1896, Russia
Fig. 1. Tensile cables involving in tensile structures.
At present, the mechanical properties of cold worked wires and
strands are specified in the concrete design guidelines, such as EN
1992-1-1 [2], BS 5896:2012 [3] and ASTM A416/A416M-02 [4].
Abrams et al. [5], investigated the behavior of steel strands for
pre-stressing concrete at high temperature. ACI 216.1-14 [6] and
PCI [7] provide some guidance for the fire resistance of pre-
stressing concrete. Accordingly, the high temperature mechanical
properties of wires and strands are given in EN 1992-1-2 [8].
Abrams et al. [5] reported in their tests that the ultimate
strength of 7 wire strands at 204 °C is reduced by 10%, while it
loses 50% of the ultimate strength at 427 °C. The test was also con-
ducted on the strands of 1/4 and 7/16 in. diameter at elevated tem-
perature; it was found that within the small strand size, the
percent loss of ultimate strength is not significant. In their tests,
the pre-stressed force has been applied on the specimen to conduct
the transient-state test at elevated temperature to capture the rup-
ture temperature. Thus, the thermal creep strain has been included
in the total strain. Abrams’s test also discovered that the rupture
temperature is almost the same from transient-state test or
steady-state test. The test heating rate is 2.8, 5.5 and 8.3 °C/min,
which was not correspond with the heating rate proposed by EN
10002-5, to simulate the heating rate of pre-stressed strands
imbedded in concrete. Thermal expansion elongation also cannot
be eliminated from the steady-state test. Therefore, heating rate
and thermal expansion elongation should be considered properly
in the experimental study of steel cables.
In 1970, Harmathy et al. [9] investigated the mechanical prop-
erties of A421 pre-stressing steel wires at elevated temperature.
The steady-state tests captured the full-range stress-strain curves
and it was observed that there is no significant loss in ultimate
strength before 217 °C, but a sharp decrease beyond 400 °C. The
decreasing tendency of yield stress and elastic modulus also have
been investigated. Harmathy et al. were the first to report that
the high temperature mechanical properties of strands were
dependent on the microstructure of the materials, which changed
at elevated temperature. The mechanical properties of high tensile
steel cables made from a group of twisted wires would be differ-
ence from that of A421 pre-stressing steel wire at elevated
temperature.
In 1982, Holmes et al. [10] conducted three series of steady-
state test for strands to determine the yield strength, ultimate
strength and elastic modulus at elevated temperature. In the
transient-state test, the specimen was pre-stressed to 70% mini-
mum ultimate tensile strength and then heated until rupture tem-
perature. Similar to Abrams et al. work [5], the rupture
temperature of the strands under pre-stressed load includes the
53
(c) Beam string structure roof, Pudong International
Airport, Shanghai, China
thermal creep effect. However, the specimens were manufactured
according to British Standard specifications BS 2691 and BS 3617
for steel wire and steel strand. Due to the different in material,
the steady-state test conducted by Holmes can only be taken as
reference in the present study.
Fan et al. [11] and Zhou et al. [12] carried out steady-state test
on 1860 MPa strand at elevated temperature. The necking of spec-
imen and ductile failure behavior weer observed for temperature
heated beyond 500 °C. However, the full stress-strain relationship
cannot be captured in their experiments.
Hertz et al. [13] not only proposed a simple mathematical
model to predict the reduction in mechanical properties for several
steel types at elevated temperature based on the test data resulted
from previous studies by other researchers, but also described the
crystalline structure for cold worked bars loosens above 300 °C.
The residual properties of cold worked steel suffer irrecoverable
losses if subject to temperature over 400 °C. However, this simple
mathematical model cannot predict the mechanical properties of
high tensile steel cables at elevated temperature.
Due to the complex surface of strands and the difficulty of grip-
ping the a group of strands in a tensile test, Zhang et al. [14] chose
the single wire of 1860 MPa strand to conduct tensile test at ele-
vated temperature. Comparison of the results from testing a group
of strands made of straight wires and twisted wires; it was found
that the twisting configuration of the strands affected the thermal
mechanical properties of strands made from a group of wires.
In recent years, Conor [15] and Shakyo et al. [16] studied the
mechanical properties of seven-wire pre-stressing strands by fire
test. Their studies indicated that the mechanical model of cold
worked strands specified by EN 1992-1-2 could not accurately
describe the stress-strain relation at elevated temperature as it
ignored the stress hardening with a continuous increase in strain
till failure even for temperatures in the range between 200 °C
and 500 °C.
Xiong et al. [17] tested heated-treated high tensile structural
steel at elevated temperature to study the mechanicalal properties
and the microstructure of high tensile steels at elevated tempera-
ture. They discovered from the testing that the thermal creep
becomes noticeable when the steel temperature exceeds approxi-
mately one third of the steel melting temperature, thus, the
steady-state test method is merely suitable to measure the elastic
modulus of high tensile steel up to 400°. It also noted that there -
was no significant discrepancies between the steady-state and
transient-state test for effective yield strength when the tempera-
ture is below 500 °C. In the present study, the steady-state test
method is employed to investigate the mechanical properties of
54 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
steel cables at elevated temperature. The thermal creep of cables
shall be studied in future research.
Previous studies also have shown that the extensometer only
can capture the partial range of the stress-strain curve before ulti-
mate strength at elevated temperature. The external mechanical
device with linear variable displacement transducer can trace the
full range of stress-strain curve at elevated temperature, but the
strain resulted from this kind of measurement device is the aver-
age strain, which is different from the engineering strain.
Recently, a charge-couple device camera system (CCDC) has
been developed to measure the strain in the test coupon subjected
to tension. A computer program based on this correlation method
was developed for the calculation of the displacement components
and the deformation gradients of an object surface due to deforma-
tion. Several experiments were performed to demonstrate the via-
bility and accurate of the correlation method [18]. Meanwhile the
experimental study also investigated the ability of the computer-
vision technique of CCDC to measure in-plane deformations at ele-
vated temperature up to 650 °C [19]. Test results indicate that the
computer vision is capable of measuring strains at elevated tem-
peratures with the same level of accuracy that is obtainable under
ambient temperature.
In present study, the CCDC system was conducted to capture
the full-range of the stress-strain curves of high tensile cables till
rupture at elevated temperature. Elastic modulus, thermal, strain
effective yield strength and ultimate tensile strength correspond-
ing to various levels are reported from the test data. Furthermore,
a stress-strain curve model, which accurately describes the
mechanical properties of high tensile strength cables at elevated
temperature, is proposed. Additional tests were conducted to cali-
brate the digital-image camera system to capture the displacement
of cables with more complex geometrical surface than individual
wires and plate specimens.
This paper extends the studies on high temperature mechanical
properties of cold worked strands, which are available in the pre-
pressing concrete codes, to high tensile strength cables used
mainly in tensile long span steel structures. The thermal mechan-
ical properties of steel cable with a tensile strength of 1860 MPa
subject to elevated temperature are obtained by using the CCDC
system.
(a) Servo-hydraulic tension test system
Fig. 2. Test set-up.
2. Experimental investigation
2.1. Test equipment and instrumentation
The test setup and charge-coupled device camera (CCDC) sys-
tem is shown in Fig. 2. The standard axial tensile tests were con-
ducted on a servo-hydraulic testing machine, with a maximum
stroke displacement of 75 mm and a capacity of 500 kN. The heat
apparatus was a split-tube furnace with three-zone configuration
and a side view window. A type K thermocouple was mounted at
center of each zone to measure the heating temperature (Fig. 3).
Calibration test is carried out on a S690 high strength steel cou-
pon heated to 400 °C and subject to tension. Both the extensometer
and CCDC system were used to measure the strain in the test spec-
imen under the applied tension force. Because the ceramic arm of
the extensometer has to be removed away from the test specimen
before the rupture of the test specimen due to its limited travel dis-
tance, the extensometer reading can only be used to calibrate the
CCDC system within a partial range of the stress-strain curve. A
good fit between the extensometer and CCDC readings is generally
observed as shown in Fig. 4. It is worth noting that the CCDC sys-
tem can accurately capture the full-range of stress-strain curve and
the strain is measured between two paint markers, which are close
enough to form the gauge length, on the surface of a specimen
shown in Fig. 7.
2.2. Test specimens
A summary of the test specimens for standard tension test and
thermal elongation at elevated temperature is listed in Table 1.
Each end of steel cables was connected to anchor heads as
shown in Fig. 5(a). The dimensions of the tensile test specimens
are shown in Fig. 5(b). For thermal elongation tests, specimens
with only one anchor head at one end of a cable were prepared
as shown in Fig. 6(a) and the dimensions of the test specimens
are shown in Fig. 6(b).
Markers with white spots painting on a black colour back-
ground were prepared within the middle part of a cable specimen
surface of about 90 mm length, as shown in Fig. 7. Through the side
view window of the furnace, the CCDC can capture any two spots
(b) Furnace
(c) Charge-coupled device camera
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65 55

Fig. 3. Measurement and data acquisition system.

The steady state (isothermal) tests described in BS EN 10002-5

Fig. 4. Comparison of stress-strain curves between extensometer and CCDC for
S690 steel at 400 °C.
within the marker area to form a certain gauge length smaller than
20 mm, and then, trace the relative displacement between the two
spots due to axial tension force and capture the stress strain rela-
tion of a specimen at elevated temperature.
2.3. Standard tension test procedure
In the present study, the steady-state method is adopted to per-
form tensile test of steel cables at elevated temperature. One end of
the cable specimen was tighten to the loading cell while the other
end was bolted to a rigid end as shown in Fig. 8.
[20], which is a standard test to obtained the mechanical proper-
ties of steel material at elevated temperature, is adopted in the pre-
sent study. However, this method ignored the thermal creep effect
of the steel material at elevated temperature. However, thermal
creep effect is only significant at temperature more than 600 °C.
The high tensile cables without fire protection normally failure at
temperature below 600 °C, and therefore the thermal creep effect
can be ignored and steady state test is sufficient for this purpose.
In real practice, the cable is under tension and fire will occur after
the cable is loaded. Therefore, the cables will be heated after the
load is applied. The results from the steady state test are still appli-
cable for the analysis of practical cable structure since the thermal
creep is not expected to be significant.
In the steady state (isothermal) tests, the specimen was lightly
tensioned to about 1 kN to eliminate any mechanicalal relaxation
of the cable. Then, the specimens were heated to the target tem-
peratures at a rate of 10 °C/min and subsequently loaded to failure.
Same test procedure was repeated for all test specimens with dif-
ferent target temperature, which were set as a range of 100 °C, 200
°C, 300 °C, 400 °C, 410 °C, 450 °C, 460 °C, 550 °C, 600 °C, 700 °C, and
800 °C. In addition to the ambient temperature of 20 °C, each target
temperature were held for 30 min to obtain uniform temperature
distribution in both the furnace and the specimens. During the
heating phase, the testing machine was set to load control so that
the upper load cell of the machine could displace to accommodate
the thermal expansion of the specimen, thus ensuring no load was
induced. Then, the tensile specimens were tested till failure under
displacement control at a straining rate of 0.003/min, in keeping
with the strain rates set out in the reference [20]. Five readings
have been collected per second during full range of test. Three
specimens were tested for each target temperature. Thirty-six
specimens were tested to complete the steady-state tests.

Table 1
Summary of steel cable specimens.

Test type fpk (MPa) Wire number Diameter of wire (mm) Diameter of cable (mm) Length (mm) Total amount of specimen
Standard tension test 1860 7 5 15 600 36
Thermal elongation test 1860 7 5 15 600 3
56 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65

(a) Specimens for tensile test at elevated temperature

(b) Dimension of a specimen (units in mm)

Fig. 5. Specimens for tensile test at elevated temperature.

(a) Specimens for tensile test at elevated temperature

(b) Dimension of a specimen (units in mm)

Fig. 6. Specimens for thermal elongation test at elevated temperature.

2.4. Thermal elongation test procedure
The upper end of a cable specimen was tighten to the loading
cell and the lower end was freely mounted into a sleeve, which
is bolted to the lower load cell as shown in Fig. 9. At elevated tem-
perature, the cable specimen was not loaded and free to expand
along the longitudinal axis due to thermal elongation. The heating
rate was applied at 10 °C/min, and the target temperatures were
set from 100 °C to 800 °C in increment of 50 °C. The target temper-
ature were held for 30 min to obtain uniform temperature distribu-
tion in both the furnace and the specimens. Record the value of the
thermal strain when the read displayed by CCDC system kept in
constant. Then heat to the next target temperature and repeat
the test procedure again till 800 °C in increment of 50 °C. Each
specimen was continuously tested for each target temperature to
obtain the thermal strain of the cable specimens. There were 3
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
Fig. 7. High temperature paint markers on a cable surface.
Connecting bolts
Upper anchor connecting
Fig. 8. Set-up for tensile test at elevated temperature.
Bolt hole
Lower connection for a cable
Fig. 9. Set-up for thermal elongation test at elevated temperature.
specimens repeatedly tested in thermal elongation test to get the
average value of thermal elongation strain.
3. Test observations
In general, cable failure, color change to the surface of speci-
men, reduction of cross section area have been observed and will
be reported in the following sub-sections.
As shown in Fig. 10, the specimens experienced either a brittle
failure mode, ductile failure mode or a necking failure mode
dependent on the target heating temperature. All the heated cable
specimens broke within the heated zone in the furnace. However,
when tested under the ambient temperature of 20 °C, failure
occurred closer to the upper anchor.
From 20 °C to 200 °C, the broken cross section of the cable dis-
played a cap shape and the failure was very brittle. The surface of
the cables remained as the original metal color after heated, but
the surface of the broken cross section of wires turned into black
colour.
57
At 300 °C, the broken cross section of the cable also displayed
the cap shape but somewhat shrank close to the broke cross sec-
tion. The failure was rather brittle. The surface of the cable tended
to be dark metal color, but the broken cross section of wires
appeared to be in deep blue color.
From 400 °C to 800 °C, the broken cross section of the cable
formed a cone-shape with cross section area gradually reduced
towards the broken surface. The cable failed in a more ductile man-
ner. A more ductile necking failure was observed when the cable
was subjected to high temperature in the range of 700–800 °C.
The surface of the cable displayed earthy yellow color, and the bro-
ken surface changed from dark grey to black color as temperature
increased.
For temperature 300 °C and below, the specimen ruptured with
a loud ‘bang’ sound and the wires recoiled violently, i.e., the wires
loosened at the failure load. Beyond 400 °C, the specimens rup-
tured with a soft ‘ping’ sound and the wires tend to remain in a
group.
4. Test results and recommendations
4.1. Thermal elongation of steel cables at elevated temperature
Corresponding to the thermal elongation test procedure
descripted in Section 2.4, the thermal elongation strain of cables
was measured by CCDC at different temperatures and there were
a total 45 readings consisting of 15 target temperatures with three
specimens at each temperature listed in Table 2. The average value
of thermal elongation strain also has been listed in Table 2.
The relationship between thermal strain and temperature was
graphically described in Fig. 11 corresponding to the average
value of thermal elongation strain. The thermal strain increased
nonlinearly with the increase of temperature. It is worth noting
that the thermal strain sharply dropped from 0.09737 to
0.08819 at about 750°C and then jumped to 1.0516 at about
800 °C due to the phase transition. The microstructure of cold
worked steel cables is lamellar pearlite structure below 700 °C.
When temperature is approaching to 750 °C, the microstructure
of cold worked cables turned to austenite. This phase transition
process absorb plenty of heat and lead to material shrinkage.
Because the material shrinkage due to phase transition is greater
than the thermal expansion, the thermal expansion was reduced.
As the thermal coefficient of austenite is about two times
greater than that of ferrite or cementite, the thermal expansion
jumped up suddenly. As the phase transition experienced a short
temperature history, this period of unstable thermal expansion
can be ignored.
The equation of thermal strain, eth , depended on temperature is
determined from regression analysis on the test data recorded in
Table 1 as follow:
eth ¼ 9:294  1012 T 3 þ 1:210  108 T 2 þ 9:062  106 T
 1:116  104 ð1Þ
The thermal expansion coefficient, aT , of a steel cable, which
describes the fractional change in size per degree change in tem-
perature, can be determined as
deth
aT ¼ ¼ 2:788  1011 T 2 þ 2:420  108 T þ 9:062  106
dT
ð2Þ
The thermal expansion coefficient of high tensile steel cable is
dependent on the heated temperature and their relation is nonlin-
ear in nature.
58 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65

(a) 20 (b) 100 (c) 200 (d) 300 (e) 400

(f) 450 (g) 500 (h) 600 (i) 700 (j) 800
Fig. 10. View of tensile failure of steel cables at different target temperature.

Table 2
Thermal strain of cable specimens at different temperature under thermal elongation tests.

Temperature (°C) 100 150 200 250 300 350 400 450
Specimen 1 0.0964 0.1614 0.2304 0.2855 0.3497 0.3950 0.4604 0.5442
Specimen 2 0.0980 0.1572 0.2310 0.3016 0.3632 0.4126 0.4742 0.5463
Specimen 3 0.0880 0.1449 0.2047 0.2649 0.3272 0.3866 0.4430 0.5276
Average value eth (%) 0.0941 0.1545 0.2220 0.2840 0.3467 0.3981 0.4592 0.5394
Temperature (°C) 500 550 600 650 700 750 800
Specimen 1 0.6345 0.7290 0.7896 0.8416 0.8889 0.8524 1.0213
Specimen 2 0.6320 0.7218 0.7816 0.8348 0.8862 0.8496 1.0141
Specimen 3 0.6279 0.7336 0.8158 0.8435 0.9339 0.9436 1.1193
Average value eth (%) 0.6315 0.7281 0.7957 0.8399 0.9030 0.8819 1.0516

4.2. Effective yield strength and elastic modulus of steel cables
The effective yield strengths, f0.5%,h, f1.0%,h, f1.2%,h, f1.25%,h, f1.5%,h and
f2.0%,h, at strain levels of 0.5%, 1.0%, 1.2%, 1.25%, 1.5% and 2.0% are
graphically presented in Fig. 12 and the corresponding values are
shown in Table 3 for the purpose of comparison. The 0.2% yield
strength, f0.2%-offset,h, is the intersection point of the stress–strain
curve and the proportional line off-set by 0.2% strain. the other
yield strengths are those values corresponding to the intersection
points of the stress–strain curve and the non-proportional vertical
line specified at a given strain level, as shown in Fig. 12. Both of the
rapture strain, epu,h, and the ultimate strength, fpt,h, dependent on
the limiting strain, ept,h, were also displayed in full range of
stress–strain curve shown as Fig. 12.
Fig. 13 shows that the high tensile strength cable does not exhi-
bit distinctive yield plateau at elevated temperature. Therefore,
effective yield strengths, which correspond to different strains,
should be determined for fire resistant design of steel cables. It
should be noted that the ultimate tensile strain, ept,h, and rupture
strain, epu,h, were measured in the present tests which is made pos-
sible by using the CCDC. The test data reported by most of
researchers cannot provide the full range of stress strain relation-
ship up to rather larger strain values.
Comparing the yield strength of steel cable at different strain
level as shown in Table 2, it is observed that the yield strength at
1.5% and 2% strain are not meaningful since 1.5% and 2% strain
are larger than the ultimate tensile strain, ept,h, when the tempera-
ture is at 600 °C. However, 1.25% strain is closer and lower to the
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65 59

Fig. 11. Comparison of thermal expansion strain from tests with prediction models
and EN 1992-1-2.
Fig. 13. Stress-strain curves for high tensile steel cable from isothermal tests.

Thus, the effective yield strength with 1.25% strain, f1.25%,h, is

Fig. 12. Effective yield strengths at different strain levels for steel cable at elevated
temperature.
1.355% limit tensile strain at 600 °C when phase change occurs.
Furthermore, there is a deviation of 13.7% between yield strength
at 1.25% strain and ultimate tensile strength, fpt,h, at 100 °C and
at most 9.2% deviation in a range of 200–800 °C. Especially, a devi-
ation of 0.36% can be touched between yield strength and ultimate
strength when the temperature is at 600 °C. The 1.25% yield strain
can allow the tensile deformation of steel cable large enough.
Meanwhile, the effective yield strength at 1.25% strain can safely
conserve the loading capability of steel cable at elevated
temperature.
recommended as the yield strength for high tensile strength steel
cable at elevated temperature. The effective yield strength pro-
posed by the present test displays lower ductility than that of EN
1992-1-2. The effective yield strength of high tensile steel cable
at 2% strain is the value adopted for pre-stressing tendons for con-
crete structures by EN1992-1-2.
It should be noted that the phase change not only influences on
the thermal elongation but also on the strain around 700 °C due to
microstructure of cable changed. The limiting strain for yield
strength increased as the temperature increased beyond 600 °C
also due to phase change.
Fig. 14 shows the comparison between the limiting strain for
yield strength at elevated temperature measured from the present
tests and given by EN 1992-1-2. EN 1992-1-2 ignored the stress
harden phase of the cold worked strands and assumed a constant
value of limit strain for yield strength, ept,h, from 20 °C to 200 °C.
However, in a range of 300–800 °C, the limit strain for yield
strength has been overestimated up to 60% at 700 °C by EN
1992-1-2 compared with the present test data. Fig. 14 also shows
that the measured limiting strain for yield strength decreases
within 200–600 °C range of temperature while the deviation
between the effective yield strength of f1.25%,h and the ultimate ten-
sile strength of fpt,h are becoming smaller with the increase of tem-
perature, as shown in Table 3.
Comparison of the rupture strain between the test data and
given by EN 1992-1-2, shown in Fig. 14, it is observed that the con-
stant value of 10% for rupture strain proposed by EN 1992-1-2
notably overestimates that of steel cables in the range of

Table 3
Effective yield strength with different strain level for high tensile steel cable at different temperature.

Temperature h (°C) f0.2%-offset,h f0.5%,h f1.0%,h f1.2%,h f1.25%,h f1.5%,h f2%,h fpt,h ept,h (%) epu,h (%)
20 1718.3 886.8 1692.3 1751.0 1759.0 1779.3 1828.8 2013.5 5.262 5.266
100 1666.4 939.9 1640.6 1691.2 1698.2 1728.3 1779.1 1966.9 4.987 5.054
200 1523.7 887.7 1519.0 1567.2 1584.8 1644.0 1720.3 1904.0 6.216 6.238
300 1218.2 757.0 1268.3 1307.0 1316.5 1361.7 1404.0 1432.8 4.074 10.333
400 914.7 723.0 968.3 1015.7 1020.9 1021.0 1038.1 1046.9 3.027 10.955
410 778.3 643.0 853.1 889.2 895.0 909.6 927.8 944.5 2.876 11.150
430 690.7 549.7 765.0 793.9 797.4 815.0 828.0 833.2 2.788 11.500
450 634.6 541.8 691.2 713.2 717.1 729.2 739.1 740.9 2.417 11.700
460 580.2 485.4 628.6 640.6 644.9 656.1 665.9 668.8 2.374 13.000
500 391.4 328.3 421.5 435.1 437.5 445.5 451.5 453.4 2.387 16.400
600 151.3 126.5 164.6 167.2 167.3 167.2 165.4 167.9 1.355 20.300
700 40.1 40.0 45.0 45.8 46.0 46.9 47.4 50.6 3.030 35.500
800 34.4 33.4 36.1 36.6 36.7 37.3 37.6 40.4 4.712 35.600
60 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65

tensile strength and elastic modulus, i.e. fpy, fpt and Ep at 20 °C

Fig. 14. Comparison of strain between test data and EN 1992-1-2.
20–300 °C, but underestimates up to 66% beyond 300 °C. From
200 °C to 700 °C, the rupture strain of steel cables increased
notably since the necking is more and more obvious beyond the
ultimate strain shown in Fig. 10.
The elastic modulus, Ep,h, can be determined by the tangent
modulus of the initial elastic linear phase from the full range of
stress-strain relationship and the proportional limit strain, epp,h,
is the end point of the initial elastic linear phase. Three specimens
were tested for each target temperature to get the average value of
the stress-strain relationship. Then, the average of elastic modulus,
Ep,h, and proportional limit strain, epp,h, are listed in Table 4. Finally,
proportional limit stress, fpp,h, can be determined by Ep,hepp,h,
which listed in Table 4 also.
4.3. Reduction factors
For simplified fire design methods, the reduction factors at tem-
perature, h, should be determined. Test data providing reduction
factors relative to the value of effective yield strength, ultimate
are shown in Table 5.
Cold working tensile process can increase the strength of steel
cables as microstructure with increase in the dislocation density
and taut. In order to product lower relaxing and high tensile
strength cables, the temper in the range of 350–400 °C has been
conduct to remove the residual stress due to cold working process
and tauted wires, and increase the ductility of the alloy [22]. Such
production process can influence on the mechanical properties of
cold working tensile steel cables at elevated temperature beyond
400 °C as discussed in the following sub-sections.
At first, it herein names that the characteristic value of tensile
strength of cables, f pk , is taken as the ultimate tensile strength,
i.e. f pk ¼ f pt;h , at 20 °C and the value of effective yield strength of
cables, f py , is taken as the stress at 1.25% strain, i.e. f py ¼ f 1:25%;h
at 20 °C.
The reduction factor of the effective yield strength for cold
worked strands recommended by EN 1992-1-2 is the ratio of effec-
tive yield strength, f2.0%,h, at 2.0% strain dependent on temperature
to the characteristic value of tensile strength at ambient tempera-
ture, fpk, multiplied by a parameter of b ¼ 0:9, i.e. f 2:0%;h =0:9f pk . The
present test data listed in Table 3 indicted that 0:9f pk ¼
0:9f pt;h ¼ f 2:0%;h at 20 °C. Comparing with the reduction factor of
effective yield strength, f 2:0%;h =0:9f pk , given by EN 1992-1-2 and
that of f 1:25%;h =0:9f pk resulted from present test data are shown
as Fig. 15. At 100 °C, f 1:25%;h =0:9f pk is 6.29% smaller than that pro-
posed by EN 1992-1-2; at 400 °C, is 12.7% greater; at 500 °C, is
19.5% smaller and at 700 °C is 57.7% smaller. Then, comparing with
f 1:25%;h =0:9f pk and f 1:25%;h =f py , At 100 °C, f 1:25%;h =0:9f pk is 2.98% smal-
ler than f 1:25%;h =f py ; at 400 °C, is 2.95% smaller; beyond 500 °C, only
4.27% smaller, shown as Fig. 15. It is observed that the deviation
between f 1:25%;h =0:9f pk and f 1:25%;h =f py is smaller than that of
f 1:25%;h =0:9f pk and f 2:0%;h =0:9f pk . Thus, referring to EN 1993-1-2
[21], the reduction factor derived from the present test data has
herein been proposed as kpy;h ¼ f 1:25%;h =f py which is close to the
f 1:25%;h =0:9f pk resulted from the present test data.

Table 4
Elastic modulus of steel cable at different temperature.

Temperature (°C) 20 100 200 300 400 500 600 700 800
Ep,h (105 MPa) 1.932 1.899 1.877 1.813 1.526 0.716 0.317 0.122 0.085
epp,h (%) 0.893 0.790 0.716 0.574 0.428 0.360 0.163 0.329 0.296
fpp,h (MPa) 1634.6 1449.6 1333.2 890.7 654.8 258.0 49.9 38.0 25.2

Table 5
Reduction factors for steel cables.

Temperature (°C) Reduction factor for effective Reduction factor for Reduction factor for the slop Reduction factor for ultimate
yield strength, kpy;h proportional limit, kpp;h of the linear elastic range, kE;h tensile strength, kpu;h
f py;h =f py f pp;h =f py Ep;h =Ep f pu;h =f pu

20 1.000 0.929 1.000 1.000

100 0.965 0.824 0.980 0.977
200 0.901 0.758 0.971 0.946
300 0.748 0.506 0.951 0.712
400 0.580 0.372 0.818 0.520
410 0.446 0.356 0.708 0.465
430 0.402 0.320 0.673 0.414
450 0.369 0.290 0.627 0.368
460 0.338 0.285 0.535 0.332
500 0.249 0.198 0.366 0.225
600 0.095 0.028 0.130 0.083
700 0.026 0.022 0.114 0.025
800 0.021 0.014 0.038 0.020
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
Fig. 15. Comparison of reduction factors of yield strength between test data and EN
1992-1-2.
Fig. 16. Comparison of reduction factors of yield strength between steel cable and
high strength steel.
Meanwhile, comparison of reduction factors of yield strength
between cold worked high tensile strength cable and heat-
treated high tensile structural steel given by Xiong et al. [17] is
shown in Fig. 16. It can be seen that the reduction factors of cold
worked high tensile strength cable were notably up to 90% less
than the heat-treated high tensile structural steel of S460M and
62% less than that of S690QT due to their different microstructure.
Comparison of reduction factors of yield strength between steel
cables and single wires given by Zheng et al. [14], shown as Fig. 16,
it is observed that the strength reduction factors of steel cables are
up to 10.3% higher than single wires at 400 °C because the wires
were twisted tightly. Beyond 400 °C, the reduction factors are up
to 59% lower than single wires at 600 °C because the twisted wire
were loosen.
The reduction factors of elastic modulus is herein defined as the
ratio of elastic modulus at elevated temperature, Ep,h, to that at
ambient temperature, Ep, i.e. Ep;h =Ep .
The comparisons between the reduction factors of elastic mod-
ulus from EN 1992-1-2 and test data are shown in Fig. 17. The
reduction factors of elastic modulus from the test data is the value
of 0.818 and up to 8.1% greater than that given by EN 1992-1-2
below 400 °C. When steel cables are subject to elevated tempera-
ture in test, there is no obviously change in microstructure below
400 °C.
61
Fig. 17. Comparison of reduction factors of elastic modulus between EN 1992-1-2
and test data.
Fig. 18. Comparison of reduction factors of elastic modulus between previous
studies and present test data.
The reduction factors of elastic modulus dramatically decreased
beyond 400 °C, and up to 68.3% less than that of EN 1992-1-2 at
600 °C.
Comparison of the reduction factors of elastic modulus between
heat-treated high tensile structural steel and the present test data
for steel cable is shown in Fig. 18. It can be seen that the reduction
factors from the test data are up to 19% higher than those of high
strength steel S460M given by reference [17] at the range of
200–400 °C, but dramatically down to 74% lower beyond 400 °C.
The reduction factors for steel cable are up to 8.7% greater than
those of high strength steel S690QT given by reference [17] at
the range of 200–400 °C, but dramatically down to 58.7% smaller
beyond 400 °C due to their different microstructure.
Comparison of reduction factors of elastic modulus between
steel cables and single wires given by Zheng et al. [14], shown as
Fig. 18, it displays that the reduction factors of steel cables are
up to 22.7% higher than single wires at 400 °C because a group of
wire twisted tightly. Beyond 400 °C, the reduction factors are up
to 52.3% lower than single wires at 600 °C because a group of
twisted wire loosen.
The reduction factors of the proportional limit proposed by EN
1992-1-2 and the present test, i.e. f pp;h =0:9f pk , are plotted against
the temperature as shown in Fig. 19. It can be seen that the bi-
linear model for cold-work strands proposed by EN 1992-1-1 let
f pp;h ¼ 0:9f pk at 20 °C. The present test resulted that the
62 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
Fig. 19. Comparison of reduction factors of proportion limit given by EN 1992-1-2
and test data.
proportional limit, f pp;h , is smaller than the value of 0:9f pk at 20 °C.
It indicates that the elliptical model is suitable to determine the
proportional limit for cables at 20 °C. Though the elliptical model
is employed by EN 1992-1-2 to determine the stress-strain rela-
tionship for cold-work strands at elevated temperature, the elastic
strain proposed by EN 1992-1-2 for pre-stressing tendons is lower
than that of steel cables given by present test. Thus, in the range of
100–500 °C, the reduction factor of proportional limits of steel
cable is up to 177.8% greater than those of cold worked strands rec-
ommended by EN 1992-1-2.
Referring to EN 1993-1-2 [21], the reduction factor of the pro-
portional limit of steel cable derived from the present test data
has been proposed as kpp;h ¼ f pp;h =f py , which is close to the
f pp;h =0:9f pk resulted from the present test data with the maximum
deviation of 3.04% in the full range of temperature as shown in
Fig. 19.
Comparison of reduction factors of proportional limit between
steel cable and heat-treated high tensile structural steel of
S690QT shown in Fig. 20. In a range of 20–100 °C, the reduction
factor of proportional limit of steel cables is greater than that of
S690QT. This is because the effective yield strength of steel cable
is based on 1.25% strain rather than 2.0% strain, which has been
adopted for S690QT steel. Obviously, the effective yield strength
with 1.25% strain of steel cable is closer to its proportional limit
than that of S690QT steel. For equation of kpp;h ¼ f pp;h =f py , if the
Fig. 20. Comparison of reduction factors of proportion limit between high strength
structural steel and steel cable.
effective yield strength is closer to the proportional limit, the
reduction factor will be greater. It can be seen that the reduction
of proportion limit of steel cable could be up to 81% smaller than
that of S690QT in a range of 200–800 °C. This is because the elastic
modulus decreased more rapidly than that of S690QT at elevated
temperature shown in Fig. 18.
The reduction factors of ultimate strength for cold worked
strands have not been proposed in EN 1992-1-2, but has been
employed to evaluate the reduction factors of effective yield
strength and proportional limit. In present study, the reduction
factors of ultimate strength of steel cable were compared with
those previously studied by other researchers as shown in
Fig. 21. It is shown that models proposed by other researchers pro-
vide a conservative prediction of the reduction factors of ultimate
tensile strength of cables for temperature between 300 °C and 800
°C. The maximum deviation is 73% between the values given by
Abrams et al. [5] and the present test data. The reduction factors
of ultimate strength based on the test data are closer to those pro-
posed by Zhang et al. [14] with a maximum deviation of 10.5%. It
could be due to the similar type of steel cables used by Zhang
et al. and the present test.
The reduction factors of steel cable mechanical properties as a
function of temperature are proposed based on regression analyses
and curve fitting of data resulted from standard tensile tests at ele-
vated temperature.
The reduction factor of the effective yield strength can be
obtained from the following piece wise function with a regression
factor R2 = 0.9881.
8
>
> 1:250  1011 h4  1:632  108 h3 þ 2:948  106 h2
>
<
6:008  104 h þ 1:011 20  C 6 h 6 400  C
f py;h =f py ¼
>
> 3:432  1011 h4 þ 8:304  108 h3  7:018  105 h2
>
:
þ2:304  102 h  1:955 400  C < h 6 800  C
ð3Þ
The reduction factor of proportional limit can be obtained
from the following piece wise function with regression factor
R2 = 0.9791.
8
>
> 6:693  109 h3  5:725  106 h2  1:790  104 h
>
<
þ0:925 20  C 6 h 6 400  C
f pp;h =f py ¼
> 9 3 5 2
> 9:596  10 h þ 2:225  10 h  1:709  10 h
2
>
:
þ4:365 400  C < h 6 800  C
ð4Þ
Fig. 21. Comparison of reduction factors of ultimate strength between test data and
previous studies.
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65 63

The reduction factor of elastic modulus can be obtained

4:5  1013 h4 þ 9:1  1010 h3  6:2  107 h2 þ 1:6  104 h  7:1  103
from the following piece wise function with regression factor
R2 = 0.9773.
8
>
> 3:381  1011 h4 þ 1:371  108 h3  6:173  107 h2
>
<
3:245  104 h þ 1:007 20  C 6 h 6 400  C
Ep;h =Ep ¼

1:4  107 h4 þ 3:5  104 h3  0:3h2 þ 118:4h  14637:5

>
> 1:480  108 h3 þ 3:342  105 h2  2:548  102 h
>
:
þ6:618 400  C < h 6 800  C
ð5Þ
The reduction factor of ultimate strength can be obtained from
the following function with regression factor R2 = 0.9965.

f pu;h =f pu ¼ 4:103  1014 h5 þ 8:160  1011 h4  4:936

 108 h3 þ 7:364  106 h2  5:018  104 h
þ 1:005 20  C 6 h 6 800  C ð6Þ

300  C 6 h 6 800  C
The accuracy of the regression analysis is represented by the
coefficient of R2, which is the ratio of the sum of square of devia-
tions of the response values to their predictor. The coefficient of
R2 ranges from 0.977 to 0.996, which showing reasonably good
fit. A comparison of values between predicted property response

0
and measured response for steel cables is also shown in Figs. 15,
17, 19, and 21 which indicting good fit to the test data.
Fig. 22 collects all respective reduction factors for steel cables.
The reduction of elastic modulus of steel cable is negligible before

2:3  105 h3 þ 1:0  102 h2  0:5h  116:7

300 °C, but others including limiting proportional, effective yield
strength and ultimate strength reduce more notably than elastic
modulus. It is well known that cold-working is the process of plas-
tically deforming the alloy and induces the higher dislocation den-
sity which in turn increases the stress required to deform the steel.

1:0  103 h2 þ h þ 507:4

2
h þ 3:8  10
However, this process also increases the brittleness of the steel.
The brittleness of steel cables can be reduced by subjecting it to
elevated temperature in a range of 300 °C–400 °C named as recov-
ery process. Thus, the mechanical properties of steel cables reduce epy;h < e 6 ept;h

4
less than 30% up to 300 °C, and decrease significantly beyond 300

1:0  10
°C due to lamellar pearlite structure slipping process and phase
changing at elevated temperature.

4.4. Material stress-strain relationship at elevated temperature

3

For advance numerical simulation method, accurate material

h þ 1:7  10

modelling is essential, and is sought herein. Based on the test data

collected for high tensile steel cable tests at elevated temperature,
6:0  105 h3 þ 2:1  102 h2  1:7h þ 258:8

a full range stress-strain relationship as a function of temperature

5

has been proposed as follow:

h þ 1:5  10
1:5  103 h3  0:9h2 þ 85:4h þ 14396:2
7 2
h  1:8  10
20  C 6 h < 300  C
epp;h < e 6 epy;h

10 3
5:9  10
Regression factors of a1, a2, a3.

Temperature (°C)
Table 6

a1
a2
a3

Fig. 22. Summary of the reduction factors for steel cable.

64 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
Table 7
Regression factors of b1, b2, b3.
Temperature (°C) 20  C 6 h < 300  C
b1 0 2
b2
b3
8
> Ee 0 6 e 6 epp;h
>
>
< f pt;h  ða1 eðe=a2 Þ þ a3 Þ epp;h < e 6 ept;h
r¼ ð7Þ
> b1 e þ b2 e þ b3
> ept;h < e < epu;h
2
>
: by tensile steel structures.
0 e P epu;h
where a1, a2, a3, b1, b2, b3 are regression factors given in
Tables 6 and 7.
The full range of stress-strain relationship has been divided into
three sections dependent on the strain stage at elevated tempera-
ture to improve the accurate degree of fitting with high coefficient
of determination, i.e. R2 = 0.9943–0.9999. Therefore, the mathe-
matic model of stress-strain relationship fitted the test data well.
The stress-strain relationship of steel cables proposed by Eq. (7)
accurately captures four important parts of the mechanical behav-
ior of steel cable at elevated temperature, i.e. proportional limit
strain, effective yield strain, limit strain for yield strength and rup-
ture strain in the temperature range of 20–800 °C.
The stress-strain relations predicted by Eq. (7) is also compared
with that from EN 1992-1-2 in Fig. 23. The stress-strain relation-
ship of steel cable given by Eq. (7) is different from that proposed
by EN 1992-1-2. The stress-strain relation for cold worked steel
strand proposed by EN 1992-1-2 has been expressed as a tri-
linear model. It is evident that the equation specified in EN
1992-1-2 ignores the stress hardening phase of strands from
100–200 °C, even in high temperature with notably necking, which
is not representative of the trends observed in the present test and
other previous studies. Furthermore, based on the full range of the
stress-strain relationship tested by the present study, the ultimate
tensile strength of steel cables, fpt,h, has been captured at elevated
temperature, but cannot be covered by EN 1992-1-2. However,
limiting strain for yield strength has been proposed by EN 1992-
1-2, and linearly dropped to rupture strain as assumption. Accord-
ing to the Fig. 14, not only the limiting strain but rupture strain
proposed by EN1992-1-2 was not in accordance with those of
the present test data as discussed in Section 4.2 above. Thus, the
Fig. 23. Comparison of stress-strain relations between test data and EN1992-1-2.
300  C 6 h < 500  C 500  C 6 h 6 800  C
0:6h  369:7h þ 38596 2:2h þ 1443:4
2:4  102 h2 þ 6966h þ 1778:1 4:0  103 h2 þ 7:9h  3637:9
6:2  103 h2  9:7h þ 3800:0
stress-strain relationship proposed by EN 1992-1-2 might not be
suitable for representative of the full range of stress-strain rela-
tionship for high tensile strength cables, which always employed
5. Conclusions
This paper addresses the important issue related to the mod-
elling of the mechanical behaviors of high tensile steel cable under
a fire scenario. The temperature-dependent mechanical properties
of steel cables with tensile strength 1860 MPa has been investi-
gated by steady-state tests and measured by CCDC system, which
can capture the full range of the stress-strain relationship of steel
cables at different temperature. Forty-two cable specimens were
tested at elevated temperature. From the test results, the following
conclusions are drawn:

A full range of temperature dependent stress-strain relationship
of high tensile steel cable is proposed, which considers the
stress hardening path. The value of 2.0% strain proposed by
EN 1992-1-2 for cold worked strands using by pre-stressing
concrete is not suitable to determine the effective yield strength
of steel cables for use in cable-tensioned spatial steel structures.
Instead, the effective yield strength corresponding to 1.25%
strain is proposed herein.

The rupture strain proposed by EN 1992-1-2 is too high with
maximum deviation of 75% compared to those obtained from
the tests for temperature between 20 °C and 600 °C. For tem-
perature beyond 600 °C, the cable rupture strain is higher than
that estimated by EN 1992-1-2. The rupture strain of steel cable
does not change significantly in range of 20–200 °C, but
increases rapidly from 300 °C to 800 °C due to the increase of
ductility resulting from microstructure change. The effect tem-
perature on the twisted strands on the mechanical properties of
steel cable is significantly due to wires loosening at elevated
temperature.

The reduction factors of proportional limit, elastic modulus,
yield strength and ultimate strength dependent on temperature
were proposed based on regression analysis of the test data.

The coefficient of thermal elongation of steel cable as a function
of temperature was obtained experimentally. A simplified equa-
tion based on curve fitting with test results is proposed for
design implementation.
The proposed thermal-mechanical properties of high tensile
steel cables could be used to perform advanced numerical simula-
tion of cable-tensioned steel structures and to evaluate their struc-
tural performance and responses when exposed to fire.
Conflict of interest
None declared.
Acknowledgement
The authors gratefully acknowledge the funding by means of
‘Mechanical behaviors of beam string structures under localised
 Y. Du et al. / Construction and Building Materials 182 (2018) 52–65
fire for design guidelines’ [SLDRCE14-05] support by Civil Engi-
neering Disaster Prevention National Key Laboratory and the Inter-
national Structural Fire Research Laboratory in Nanjing Tech
University, China.
References
[1] Masao Saitoh, Akira Okada, The role of string in hybrid string structure, Eng.
Struct. 21 (1999) 756–769.
[2] EN 1992-1-1, Eurocode 2: Design of Concrete Structures – Part 1-1: General
Rules and Rules for Buildings, European Committee for Standardization, 2004.
[3] BS 5896, Specification for High Tensile Steel Wire and Strand for the Pre-
Stressing of Concrete, British Standard Institution, 2012.
[4] ASTM A416/A416M-02, Standard Specification for Steel Strand, Uncoated
Seven-Wire for Pre-stressed Concrete, West Conshohocken, Pa, USA, 2015.
[5] M.S. Abrams, C.R. Cruz, Behavior at high temperature of steel strand for pre-
stressed concrete, Portland Cem. Assoc. Res. Dev. Lab. 3 (1961) 8–19.
[6] ACI 216.1-14, Code Requirements for Determining Fire Resistance of Concrete
and Masonry Construction Assemblies, American Concrete Institute,
Farmington, MI, 2014.
[7] PCI, Design for Fire Resistance of Precast/Pre-stressed Concrete, third ed.,
2011.
[8] EN 1992-1-2, Eurocode 2: Design of Concrete Structures – Part 1-2:
General rules – Structural Fire Design, European Committee for
Standardization, 2004.
[9] T.Z. Harmathy, W.W. Stanzak, in: Elevated Temperature Tensile and Creep
Properties of Some Structural and Pre-stressing Steels, ASTM STP-464: Fire
Test Performance, 1970, pp. 186–208.
65
[10] M. Holmes, R.D. Anchor, G.M.E. Cook, R.N. Crook, Effects of elevated
temperatures on the strength properties of reinforcing and pre-stressing
steels, Struct. Eng. J. Part B: R&D Q. 60 (1982) 7–13.
[11] J. Fan, Z. Lv, Experimental study on the pre-stressed steel strand at high
temperature, Build. Struct. J. 32 (2002) 50–63.
[12] H.T. Zhou, G.Q. Li, S.C. Jiang, Experimental studies on the properties of steel
strand at elevated temperatures, J. Sichuan Univ. 40 (2008) 106–110 (in
Chinese).
[13] K. Hertz, Reinforcement data for fire safety design, Mag. Concr. Res. 56 (2004)
453–459.
[14] H.Y. Zhang, W.Z. Zheng, Mechanicalal property of steel strand at high
temperature, J. Harbin Inst. Technol. China 39 (2007) 861–865.
[15] T. Conor, Mechanicalal properties of cold-drawn steel strand at elevated
temperatures (Ph.D. thesis), Lehigh University, American, 2015.
[16] A.M. Shakya, V.K.R. Kodur, Effect of temperature on the mechanicalal
properties of low relaxation seven-wire pre-stressing strand, Constr. Build.
Mater. 124 (2016) 74–84.
[17] J.Y. Ming-xiang Xiong, Richard Liew, Mechanicalal properties of heat-treated
high tensile structural steel at elevated temperature, Thin-walled Struct. 98
(2016) 169–176.
[18] T.C. Chu, W.F. Ranson, M.A. Sutton, Applications of digital-image-correlation
techniques to experimental mechanicals, Exp. Mech. 25 (1985) 232–244.
[19] J. Lyons, J. Liu, M.A. Sutton, High-temperature deformation measurements
using digital-image correlation, Exp. Mech. 36 (1996) 64–70.
[20] BS EN 10002–5, Tensile Testing of Metallic Materials – Part 5: Method of Test
at Elevated Temperatures, European Committee for Standardization, 1992.
[21] EN 1993-1-2, Eurocode 3: Design of Steel Structures – Part 1–2: General Rules,
Structural Fire, European Committee for Standardization, 2007.
[22] W. Callister, D. Rethwisch, Fundamentals of Materials Science and Engineering,
third ed., Wiley, Hoboken, N.J., 2008.