SPE 81155

Analysis of Collapse Failures on Production Tubing During Steam Injection Process

Assisted by Computer Simulations
G. Inciarte and L. Nieto, Petróleos de Venezuela PDVSA, A. Bello, SPE, Halliburton-Landmark Graphics Corporation

Copyright 2003, Society of Petroleum Engineers Inc.

• Thermal Impact on the tubing after the water steam
This paper was prepared for presentation at the SPE Latin American and Caribbean Petroleum injection – Closed Annular space with water.
Engineering Conference held in Port-of-Spain, Trinidad, West Indies, 27–30 April 2003.
• Water steam injection - Closed Annular space
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as with air.
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
• Water steam injection - Closed Annular space
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at with water.
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
Steam Injection Process
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836 U.S.A., fax 01-972-952-9435. The following lines brefly describe the evolution of the steem
inyection process for shallow wells in the Tierra Este Pesado
area. The figure 1 describes the geographic area of study.
Abstract
In the early years the steam injection process was
The alternate steam injection consist in force the reservoir to a
performed in two methods, the first one was able to select the
cyclic sequence of heating and cooling process injecting hot
production sands and the second was performed injecting all
water steam (500°-600° F) in order to improve the mobility for
the drain area (conventional), in both cases thermal packers
the extra-heavy crude in the reservoir, this is a very common
with mechanical or hydraulic setting mechanism were used to
practice for secondary recovery on the Tía Juana and
seal the productions zones.
Lagunillas fields, located in the eastern side of the
Maracaibo Lake.
In all those scenarios the annular space between the 7 in
intermediate casing and the 4 ½ in production/injection tubing,
A destructive collapse failure on the production tubing was
were always implicit the instruction to maintain this annular
observed over the last wells exposed to this injection process.
space open or vented in order to consume the temperature and
The investigation of the failure includes field visits, laboratory
the water steam generated by the heat transfer from the tubing
tests, and the computer simulations.
to fluid and to the casing.
The study was prepared using a computer program able to
Due to the packer stuck problems encountered was decided
create the different temperature profiles for the different
to change the material of the cups in the packer to nitril, which
scenarios and link them with the simulations for load cases,
are more resistant to axial stresses. At the beginning of their
the results demonstrate the deterioration effect of temperature
use some problem were observed because they have a
on the mechanical properties of the tubing, with particular
temperature limit of 400° F and the injection process arose
interest on the expansion of the fluid trapped between the
temperatures of 500 to 600° F therefore the material of the
casing and the injection tubing, leading the collapse failure
cups was changed once againg to a better thermal
(AFE Analysis).
resistance metal.
The analysis of the production tubing movement and
Because of the considerable numbers of failures was
buckling due to the thermal effect, ballooning and Hook´s law
decided to change the method to conventional injection using
is also included.
the completion equipment to inject thought the annular space,
thus the directive to maintain the annular space vented was no
Introduction
longer necessary.

Since the year 1989 the corporation decided to adopt the

The following four scenarios were studied:
practice to cement the production casing up to the surface with
the purpose of re-enforce their mechanical integrity, and
• Water steam injection - Open or vented Annular
minimize the tubing movement. In wells were the top of
space with air.
2 SPE 81155
cement did not reach surface the injection was performed
using a packer with an isolated injection tubing, therefore the
injection through the annulus was restricted only to well were
the top of cement is uncertain. A work flow of the steam
injection process can be observed in figure 2.
Actually four injection method are in use;
1. Injection throught the Annular Space.
2. Conventional Injection with Packer.
3. Conventional Injection without Packer.
4. Selective Injection.
However some collapse of injection tubing was observed
during the injection process using the second method,
therefore a study was commissioned to investigate. Differents
aspects of the failure can be seen in the digital photograps of
the figure 3.
Investigative Procedure
Field Visits
Field visits were programmed to the areas Lagunillas and
Tía Juana, in order to observe in situ the procedure of steam
injection and record the values of temperature and pressure
observed through the 2 in dump line in the well head. The
table 1 shows a resume of the field visits.
From the field visits were observed that the procedure in
some cases and for several reasons was omitted in the section
of keep the annular space vented during the injection process.
In other trend of thought, for vertical wells the injected
volume is around 3500 to 4000 Tons of water steam, for
horizontal wells is between 6000 and 8000 Tons of water
steam during a period of +/- 15 days.
Tubing Inspection
A series of laboratory evaluations of the mechanical
conditions in the collapsed tubing recovered from the wells
were performed in situ at the warehouse. The idea was to
verify if the mechanicals properties were affected by the
failure in terms of wall thickness, material hardness and the
metal micro-structure Metallographic.
Wear or Wall Thickness
The wall thickness or wear of the material directly affect
or depreciate their mechanical properties, therefore every time
a tubular reduce their wall thickness thanks to the stresses
typical of the steam injection process their mechanical
properties in terms of resistance to collapse, burst or axial,
must be re-calculated or adjusted, Those calculations must be
performed using the American Petroleum Institute (API)
standard 5CT “Bulletin on Formulas and Calculations for
Casing, Tubing, Drill pipe, and Lines Pipe Properties”.
Wall thickness measurements were performed on the
collapse injection tubing recovered form the wells LS-5542,
LS-5440 and LS-5530, a summary of the values obtained can
be seen in the Table 2.
The value showed for all the cases is in total
correspondence with the average of different measurements
done through the pipe, for all cases 8 samples of material
hardness and 8 samples of wall thickness were toke.
The wall thickness measurement equipment was a
KRAUTKRAME, model DM-2 witch uses an ultrasound
principle, transmitter-receiver. This device transmits an
ultrasound waves that penetrates through the material until
reach the end of the wall. This equipment then returns and
determines the thick of the wall in milimeters.
A mechanical properties declination is considere when the
pipe loose 12.5 % of their wall thickness.
The wall thickness value for the 4 1/2 in pipe with 12.75
ppf and an inside diameter of 3.958 in is 0.271 (6.883 mm);
the allowable wear is 0.860mm.
Well Wear
LS-5542 6.90 mm – 0.86 mm = 6.04 mm
LS-5530 7.00 mm – 0.86 mm = 6.14 mm
LS-5440 6.80 mm – 0.86 mm = 5.94 mm
(Total Wear of 1.32%)
Based on this measurements, the tubing recoved after the
steam injection process, did not experiment any mechanical
properties declination.
Hardness
Take into account the material hardness measurements on
the production tubing a primary tension resistance value can
be infer. This value is given in Rockwell B (HRB) units.
The Material hardness Measurement Equipment for
metallic alloy KRAUTKRAMER BRANSON, model MIC-10,
calibrated according with the SUN-TEC hardness standard,
68.2 hrc +/- 0.5, the principle of operation is electronic
detectors that register the penetration of the dented device in
the material and correlate this value with pre-established
hardness values.
Using the average material hardness values and taking into
account the standard established in the manual “1991 Annual
Book of ASTM Standards – Iron and steel products” an
approximated value of tension resistance of the material was
infered, the following table describes the appropiate hardness
conversion for nonaustenitic steels:
Well Average Value from Table
LS-5542 92.18 HRB 92 ksi
LS-5440 81.50 HRB 73 ksi
LS-5530 90.10 HRB 89 ksi
SPE 81155 3

In conclusion, the values found in the production tubing Combining (Eq. 1) and (Eq. 2):
recovered from the wells studied are equivalent to the nominal
values of mechanical properties of a new tubing grade J-55, 4 V0 − V0 (1 + α∆T ) α∆T
½ in, 12.75 ppf. ∆P = − = (Eq.3)
V0 B N BN
Metalographic Analysis
The equation (3) can be used to determine the upper
The metallographic analysis arose that the temperatures
boundary of the annular pressure accumulation; The actual
observed during the water steam injection process did not
pressure increase can be significantly less than this, this
produce any change in the microstructure of the material,
differential can be attributed to the dimension on the casing
which affect the mechanical properties of the tubing. However
was modify by the pressure, according to Lamé the pressure
the results obtained were compared with the properties of a
modify the casing diameter as follow:
tubing grade J-55.

Anular Fluid Expansion (AFE) x d2  D2 

u = Pi (1 − ν ) + (1 + ν )
The effects of fluid thermal expansion in confined annuli is
( 
E D2 − d 2  )  (Eq.4)
x2 
well know in the industry, the problem arose when fluid is left
in the annuli of the casings above the top of cement, when the x D2  d2 
u = − Po (1 − ν ) + (1 + ν )
temperature in the well increase the confined fluid tend to
expand the volume increasing the pressure, creating a ( 
E D2 − d 2  )  (Eq.5)
x2 
collapse-burst situation. Therefore a way to vent or release this Were:
pressure must be implemented or the casing and tubing must
be designed to withstand this additional stresses. u = Diameter Variation
x = Point of Interest
The análisis of this fenomena becomes complex when the
well geometry is complicated, the differential pressure When x is the internal diameter and all the casing
throught the casing walls creates a small deformation on it, deformation is concentrated on that as reference point,
consequently the volume of the fluid in the annuli will consequently the deformation will only depends on the
be affected. pressure applied.

Any change in the annular volume will affect the pressure, u = Pi ⋅ C 2

this will determine the deformation of the casing, and all the Where
casing, tubing and annulus in the well are together forming a
system that should be resolved simultaneous. This is simple to  d3  D2 
C2 =  ⋅ (1 − ν ) + (1 + ν ) ⋅ 2  
resolved utilizing computer algorithms although the method (
E ⋅ D − d 
2 2
) d 
can be performed directly in three steps:
u = − P0 ⋅ C 3
1. Calculus of the thermal expansion Where
2. Fluids Pressure build up (Ballooning effect) 2⋅d D2
C3 =
3. Elastic deformation of the casing or tubing
(Hooks Law) (
E D2 − d 2 )
In productions operation a well is heated, this build up in In General any temperature variation will produce a
the temperature produce the expansion of the fluids in change on the annular pressure; hence this will modify the
the annuli. casing dimensions and the annular volume, that process will
vary the pressure profile again.
V = V 0 ⋅ (1 + α ∆ T ) (Eq.1) Compressibility Coefficients and Thermal
Expansion Values
In this case the Vo represents the initial volume and α is
The coefficients of thermal expansion and compressibility
the thermal expansion coefficient. If the annuli space remains
depends directly on the pressure and temperature, however the
closed the fluid trapped must be compress a pressure P to be
compressibility in liquids don’t vary considerably with
able to fit in the original volume Vo.
pressure. The following table represents the common values of
compressibility and thermal expansion coefficients for
different types of fluids.
∆V V −V
∆P = − =− 0 (Eq.2)
V0 B N V0 B N
4 SPE 81155
Type of Fluid α (R) BN (in2 / lbs)
Water Base 2.5 x E - 04 2.8 x E - 06
Oil Base 3.9 x E - 04 5.0 x E - 06
Gas 1/T 1/P
Where:
T and P represent Temperature
R: reprent gas Pressure in psi
Anular Fluid Expansion and Cement
The top of the cement affect significantly on the pressure
accumulation on the annuli space. Only if a remedial cement
job is performed on the last casing shoe, the annular space can
be sealed and absent of any pressure accumulation. The
pressure trapped in the annuli will rise until a failure of the
casing or a formation fracture occurs, in some of the wells
eruptions of water steam were seen during the injection
process. When the cement behind the casing did not reach the
surface, the pressure in these annuli is limited by the fracture
pressure at the previous casing shoe. Any additional build up
will send the pressure directly to the formation. Assuming the
cement to be rigid therefore restricting the casing deformation
when pressure is applied. An external pressure profile is used
to reproduce the compressibility resitance of the cement.
Computer Simulations
The computer application used is able to accurate
determine temperature and pressure profiles, casing and tubing
loads and stresses tubing and well-head movements and
annular fluid expansion analysis as well.
Following lines explain in details the input parameters, the
steps follow for each simulation procedure and the
interpretation of the results obtained:
Since the object of study is focused in the collapse of the
injection tubing no drilling or completion thermal profile
needs to be defined on the other hand a series of production
operations were defined in terms of time of duration, fluids
utilized, initial temperature, casing and tubing definitions wich
were sorted by the operation order.
Therefore the thermal profile or heat transfer during the
injection process is determined for each operation in order to
draw one thermal profile for the entire production envents.
Hence the program determines if the temperatures
deteriorates the mechanical properties of the tubing, make it
more vulnerable to failures. Also the internal-external pressure
profiles and the axial loads were analized on the
triaxial envelop.
A Multi-String analysis was performed where the casing
and the tubing are acting as a single system calculating all the
pressure and stresses, the tubing and well-head movements
and annular fluid expansion.
The initial conditions for all the cases and the Casing,
Tubing and Packer configuration can be seen in the Figures
4 to 7.
Summary of Cases
The following cases were considered for the
computer simulations:
• Water steam injection - Open or vented Annular
space with air after the annuli fluid were drained
during the initial period of the steam injection (first
hours).
• Thermal Impact on the tubing after the water steam
injection – Closed Annular space with water, on the
instant moments of the steam injections
(first seconds).
• Water steam injection - Closed Annular space
with air.
• Water steam injection - Closed Annular space
with water.
Case 1
This simulation was completed following the operational
corporative procedure for Alternate Steam Injection, witch
specify to take a time with the annular space open or vented in
order to vaporize the completion residual fluid located in the
annuli. The plan was to have the data obtained from a fine
performed procedure to be able to compare it whit the
others cases.
The Figure 8 shows the thermal profile at the bottom, the
figure 9 display the Tri-axial loads limits for this case. From
this chart we can conclude that if the pre-established
procedure is followed, the stresses acting on the tubing created
by the annular fluid expansion will never overcome the
theoretical limit of the pipe and cause a collapse failure.
Case 2
This case simulates the thermal impact on the first seconds
of the injection when the temperature passes from undisturbed
(geo-temperature) to the ultra high heat of the injection.
The figures 10 and 11 display the thermal profile at the
bottom and the Tri-axial loads limits obtained for the case 2.
However the stresses acting on the tubing are higher than case
1, they are not enough to cause the collapse failure.
Case 3
This case the annular space is loaded with air because the
residual completion fluids were released on the first minutes
of the injection with the annular space open.
The figure 12 and 13 displays the thermal profile at the
bottom and the Tri-axial chart for this case. The tri-axial char
clearly shows the failure of the tubing when the tri-axial
SPE 81155 5

profile of the injection operation is out of the boundaries of the Pipes Properties” API Bulletin 5C3, USA, Sixth Edition,
tri-axial loads limits. October 1, 1994.
4. American Petroleum Institute API, “Specifications for
Case 4 Casing and Tubing” API Bulletin 5CT, USA, Seventh
Edition, October 1, 2001.
The case 4 represent the simulation of the failure, the worst 5. ASTM, “Annual Book of ASTM Standards – Section 1
case scenario when the annular space is closed and filled with Iron and Steel Products”. Volumen 01.04, USA.
residual completion fluid. The temperature profile can be seen 6. Lagoven, S.A. Producction Department Western Division,
in the figure 14 and the tri-axial chart on the figure 15. Exis R. Parra G., Alvaro Caicedo, Elgar Jimenez, Jesus
Piñero, “Estudio de Fallas, Pozos Sometidos a Inyección
As a summary for all the cases, the figures 16 shows the de Vapor”. Tía Juana, May 1996.
heat transfer, the figure 17 shows the tubing movement and 7. PDVSA INTVEP, “Manual de Consulta Rápida, Diseño
the figure 18 display the internal – external pressure profile. de Revestidores y Tuberías de Producción”. Caracas,
April 1998.
8. PDVSA INTVEP, “Informe Técnico INT-4683, 1998,
Conclusions “Procedimiento de Diseño para tubulares de
Revestimiento y Producción”. First Edition, Caracas,
There was no loss of the wall thickness of the tubing that may April 1998.
reduce the mechanical properties. 9. Samuel, G.R., Gonzales, A. “Minimum Cost Casing
Design” SPE 36448 paper presented at the SPE Annual
There was no change on the material hardness of the tubing. Technical Conference, Houston, TX, October 3-6, 1999.
10. Samuel, G.R., Gonzales, A. “Wellhead Growth Index
There is no change on the micro-structure of the steel that Aids Multistring Casing Design” Oil & Gas Journal,
implies a deterioration of the material due to October 20, 2000.
temperature effect. 11. Aeschliman, D. P. “The Effects of Annulus Water on the
Wellbore Heat Loss from a Steam Injection Well with
According to the computers simulations the pressure reached Insulated Tubing” SPE 13656 paper presented at the
after the injection process only on the cases where the annulus California Regional Meeting, Bakersfield, CA, March 27-
is closed, exceed the collapse resistance of the 29, 1985.
production tubing. 12. MacEachran, A., Adams, A.J. “Impact on Casing Design
of Thermal Expansion of Fluids in Confined Annuli”
Acknowledgments SPE/IADC 21911, presented at the SPE/IADC Drilling
Conference in Amsterdam, March 1991.
The authors wish to thank PDVSA and Halliburton-Landmark 13. Adams, A. “How to Design for Annulus Fluid Heat-Up”
Graphics Corporation managerment for permission to publish SPE 22871 presented at the SPE Annual Technical
this paper. Conference, Dallas, Texas, October 6-9, 1991.
14. G. J. Van Wylen, Richard E. Sonntag, Claus Borgnakke
Nomenclature: “Fundamental of Classical Thermodynamics”, Second
Edition, John Wiley & Sons, Inc., New York, NY, 2000.
Vo = Initial Volume
α = Thermal Expansion Coefficient SI Metric Conversion Factors
P = Pressure
u = Diameter Variation in x 2.54* E+00 = cm
x = Point of Interest ft x 3.048* E+01 = m
T & P = Temperature lbf x 4.4482222 E+00 = N
R = Gas Pressure lbm x 4.535924 E+01 = Kg
°F-32 x 5.556 E-01 = °C
ppg x 1.198264 E+02 = Kg/m3
References psi x 6.894 E+00 = kPa
1. Landmark Graphics Corporation “WellCAT – Training bbl x 1.59 E-01 = m3
Manual”, Houston, USA April 2000.
2. Landmark Graphics Corporation “Tubular Desing Course *conversion factor is exact
– Training Manual”, Houston, USA April 2000.
3. American Petroleum Institute API, “Bulletin on Formulas
and Calculations for Casing, Tubing, Drill Pipe, and Line
6 SPE 81155

Figure 1. Geographic Location of the Study Area.

Figure 2. Alternate Steam Injection Process.

SPE 81155 7

Figure 3. Failure Digital Photographs

Figure 4. Initial Conditions for the Simulations

Figure 5. Wellbore and Packer Schematic

8 SPE 81155

Figure 6. Casing Data

Figure 7. Packer Data

550

498.20 497.20
500 482.70 479.40
Tem perature (°F)

450

400

350

300
262.40

250
Injection Fluid Injection Tubing Annular Fluid 7" Casing 9 5/8" Casing
Figure 8. Heat Transfer Case 1

Figure 9. Tri-axial Chart for Case 1

550

496.30 493.50 492.40 491.60

500
Tem perature (°F)

450
418.90

400

350

300

250
Injection Fluid Injection Tubing Annular Fluid 7" Casing 9 5/8" Casing

Figure 10. Heat Transfer Case 2

Figure 11. Tri-axial Chart for Case 2
SPE 81155 9

550

495.70
500 486.40 482.10 479.00
T em p eratu re (°F )

450

400

350

300

250
Injection Fluid Injection Tubing Annular Fluid 7" Casing 9 5/8" Casing
Figure 12. Heat Transfer Case 3 Figure 13. Tri-axial Chart for Case 3

550

495.30 493.50 492.40 491.50

500
T e m p e ra tu re (°F )

450
418.90

400

350

300

250
Injection Fluid Injection Tubing Annular Fluid 7" Casing 9 5/8" Casing
Figure 14. Heat Transfer Case 4 Figure 15. Tri-axial Chart for Case 4

600

500
Case #1
Tem perature (°F)

400 Case #2

Case #3
300
Case #4
200

100

0
Injection Fluid Injection Tubing Annular Fluid 7" Casing 9 5/8" Casing

Figure 16. Heat Trasfer for all Cases

10 SPE 81155

8
6
Movement (ft)

4
Hooke Law (ft)
2
Ballooning (ft)
0
Thermal Effect (ft)
-2 1 2 3 4
-4 Total (ft)
-6
-8
Case

Figure 17. Tubing Movement

Differential Pressure

50000 48014

45000

39964
40000

Internal
35000
External
30000
Pressure (psi)

25000

20000

15000

10000

5000
1242 1600
643 593 642 643
0
1 2 3 4
Cases

Figure 18. Internal – External Pressure Profile