SPE 157977

Casing Thermal Stress and Wellhead Growth Behavior Analysis

Q. Jim Liang, IPM Schlumberger

Copyright 2012, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Asia Pacific Oil and Gas Conference and Exhibition held in Perth, Australia, 22–24 October 2012.

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 have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper 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 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
In high-temperature wells or in high-temperature gas injection wells, casing thermal stress is an important issue that should
be considered in well design. Stresses introduced by well temperature changes could be severe and critical, resulting in
catastrophic well integrity failure and surface equipment damage if not properly identified.

The problem can be more severe when a high-temperature well is constructed in a cold environment; the temperature increase
from the equipment’s as-installed condition to operation conditions would be even higher. A deepwater well or arctic area
well installed in winter needs special consideration for thermal stress.

A field example in this paper demonstrates a near miss catastrophic well integrity failure. The incident will be analyzed by
using a multistring well thermal growth model to calculate casing thermal stress and wellhead growth.

Casing thermal growth and stress is sensitive to the length of free moving casing sections. In this paper casing thermal
behavior and wellhead movement will be discussed by analyzing different casing string top of cement (TOC) depths. This
paper demonstrates how to minimize thermal force and reduce wellhead thermal growth by optimizing the casing cement
level. At given conditions, adjusting surface tension during casing installation can also help counterbalance thermal force. It
is important to understand the thermal forces and behavior of all casing strings in a well during its operational life.

Introduction
In the year 2000, an offshore gas injection well in Sakhalin, Russia had a wellhead failure during gas injection. The wellhead
“A” section locking pins sheared and the wellhead grew by approximately 2 ft (see Fig. 1), nearly reaching a platform beam
above the wellhead. Fortunately there was no surface production piping or wellhead damage. The platform was immediately
shutdown. The well eventually cooled off and the wellhead shrank back to its as-installed position.

This incident triggered an investigation into well thermal stress status conjunction with casing and wellhead design.

The well’s thermal stress was calculated using the following well structure and operation data:
• Wellhead undisturbed static temperature: 10!C
• Seabed static temperature: 0!C
• Bottomhole undisturbed static temperature: 94!C
• Surface gas injection temperature: 100!C
• Maximum surface gas injection pressure: 31,026 kPa (4,500 psi)
• Wellhead-to-seabed distance: 160 ft.

The casing setting depth and TOC of each string are listed in Table 1.

Well operation temperature distribution in each tubular string were modeled using the WEST* wellbore simulated
temperatures software. A multi-string thermal force and wellhead growth simulator was developed to calculate thermal force
and wellhead growth caused by the temperature change.
2 SPE 157977

Calculations show that the total thermal force produced by the hot gas injection reached to 610 klbs during the injection (see
Fig. 6). The original wellhead locking pins total shearing load capacity was only 47 klbs, and did not meet the operation load
conditions. The locking pins were sheared soon after the injection operation started and the wellhead grew by more than 20
in.

The capacity of wellhead locking pins was redesigned, and the wellhead was modified to meet operational requirements.

Using the Sakhalin case, this paper will make some in-depth discussions on the behavior of wellhead thermal growth and
thermal stress.

Casing thermal stress and wellhead growth calculations

There are several ways to calculate the thermal growth. The “stiffness method” was used in this investigation. It is applicable
to multi-strings that are connected together at the wellhead.

The following procedures may be used to calculate the thermal stress and wellhead growth. Tubular buckling is not
considered in this approach.

1. Calculate the stiffness of each string connected to wellhead, including the tubing:

Equation 1
Ai Ei
Ki =
Li

2. Calculate total stiffness of all casing strings:

Equation 2
Kt = ! K i

3. Calculate the fixed end actions (FEA) caused by temperature change

The FEA for single-string element is defined as:

Equation 3
FEAi = Ai Ei " i !Ti

4. Total FEA for multi-string due to thermal expansion is given as:

Equation 4
FEAthermal = ! FEAi

5. Calculate the thermal growth of wellhead assuming no buckling:

Equation 5
FEAthermal
!Lthermal =
Kt
Note that all casing strings are connected to the wellhead and !Lthermal is equal for all strings thermal growth.

6. Calculate the thermal forces in each string at the wellhead due to the temperature changes:

Equation 6
Fi !thermal = K i "Lthermal ! FEAi

7. The total force applied on the wellhead by the thermal expansion is given by:
SPE 157977 3

Equation 7
Ftotal = ! F i "thermal

During operations there will be pressure change in the tubing. Pressure changes from the as-installed condition also
contribute to wellhead movement and to the forces in casing strings. Ballooning forces are treated as FEA. An increase in
internal pressure in a string tends to shorten the string and to move the wellhead downwards, so the corresponding FEA is
negative. On the other hand, a decrease in internal pressure gives a positive FEA. Conversely an increase in external pressure
tends to lengthen a string and so the corresponding FEA is positive, while for a decrease in external pressure FEA is negative.
For each string the FEA due to ballooning may be calculated by:

Equation 8

FEAballoon = 2# ("!p I AI + !pe Ao )

The full ballooning force can only be developed for a casing fixed at each end, which means zero displacement. From
Equation 7 we have the following equation:

Equation 9

Fballoon = K i !L " FEAballoon

= 0 " 2# ("!pI AI + !pe Ao )
= 2# (!pI AI " !pe Ao )

Pressure end area forces (Fend) are treated as external loads in the analysis (positive upwards) and are added to the sum of
FEA due to thermal and ballooning effects to calculate the displacement:

Equation 10
FEAthermal + FEAballoon
!L =
Kt

The forces in individual members are then calculated using:

Equation 11

Fi = K i "L ! FEAi !thermal ! FEAi !balloon

And the total force on the wellhead is given by:

Equation 12

Ftotal = ! F i

However, ballooning would normally not have a significant effect on the force and growth of the wellhead. Furthermore
immediately after the pressure is released, the wellhead will be exposed to the maximum force. Consequently it is more
conservative to calculate the thermal force on the wellhead without considering ballooning.
During well construction, casing strings are installed in the well with an as-installed landing load at the wellhead. Any slack-
off or tension load applied to each string during installation will reside in the wellhead landing shoulder. Slack-off will
reduce the compression in the wellhead and pulling will increase the compression. During production or injection operations,
additional operation load (compression or tension) will be introduced to the string and added to the wellhead.

If there is any as-installed compression load on the wellhead, the thermal force must overcome the as-installed compression
load before applying any upward pushing force on the wellhead.
4 SPE 157977

The weight of the wellhead and production tree should also be considered in the calculation, as their weight applies
downward compression on the wellhead.

Sakhalin wellhead failure simulation

The Sakhalin wellhead failure case was analyzed using the above procedures. Fig. 7 is a snapshot of a simple multi-string
well thermal growth model that calculates the wellhead growth and thermal force. It shows that the wellhead growth was 1.53
in, and the thermal force was 610 klbs before the locking pin failure. Because the original locking pins’ total capacity in the
wellhead was only 47 klbs, they sheared soon after injection started. Fig. 8 shows the status of the wellhead growth after
locking pin failure. When the locking pins failed, restriction of the 30-in and the 185/8-in casings were removed and the
wellhead grew by 20-in to release the stress. The calculations matched well with the actual situation.

TOC effect on the wellhead movement and thermal force behavior

Looking at Equations 1, 5, and 6, one can see that the length of uncemented tubular in a well will affect the thermal force and
wellhead movement. Fig. 2, 3 show the thermal force and wellhead growth in relation to various conductor and surface
casing TOC depths respectively. To simplify these discussions, only the conductor and the surface casing TOC levels are
varied here. The intermediate and production casing TOC levels also affect the thermal stress at the wellhead, but they are
kept constant in our discussion.

When a conductor has a section that is not cemented below the wellhead (such as in offshore wells), and that the surface
casing is cemented to the wellhead inside the conductor, the cement between the surface casing and the uncemented
conductor (above seabed) would not prevent the casing from growing. This is because the tensile strength of the cement in
the annulus is weak. Casing growth would easily cause tensile failure of the cement in the annulus above the seabed, and the
cement in the annulus above seabed will grow with the both casing string as demonstrated in Fig. 6. Consequently, the dotted
line sections in Fig. 2, 3, 4 and 5 would be invalid and should not be considered in the design.

When the well temperature change is given in a well, there will be a maximum value of thermal force that the well can reach.
In the case presented in this paper, the maximum thermal force value is nearly one million pounds when the conductor casing
is cemented to surface (as shown in the case of 30-in conductor TOC at surface in Fig. 2). This is because the conductor
casing constrains the wellhead growth and takes on the total inner casing strings thermal expansion force. It demonstrates that
the thermal force could be significantly high and must be considered during casing and wellhead design.

Normally, the conductor has a large cross-section area and much shorter uncemented length, its stiffness value is large, and
most likely, it will be larger than the sum of the stiffness of all inner strings. The thermal force taken by the conductor is
equal to the sum of all inner strings thermal expansion force (see Fig. 6, thermal force in each string). If the locking capacity
of the wellhead could not withstand the thermal force (as in the Sakhalin case), the thermal force would release and the
wellhead would grow as demonstrated in Fig. 1.

For an offshore well, the conductor above the mudline is not cemented and will be allowed to grow. The conductor growth
will increase as the water depth increases. The conductor growth would release the thermal stress, and the wellhead thermal
force would not reach the maximum level until the surface casing cement top is deep enough.

When the conductor is not allowed to grow, such as in a land well where the conductor is cemented to the wellhead, the
thermal force on the wellhead will reach to the maximum level; the value of which is dependent on the temperature increment
and the inner casing string TOC level. The wellhead growth will be zero when the conductor is cemented to the wellhead.

Fig. 3 and 4 demonstrate that when the 185/8-in TOC is deep enough, the wellhead will grow linearly as water depth increases
or the conductor TOC becomes deeper.
Thermal force minimization
The thermal force can be minimized by placing the TOC at the right depth. In Fig. 2, for example, the wellhead thermal force
is 259 klbs when conductor TOC depth is at 400 ft and the 185/8-in casing TOC is at 400 ft; for the same 185/8-in casing TOC
depth, the wellhead thermal force is 930 klbs when the conductor TOC is at surface. To reduce the thermal force further, the
conductor TOC needs to be lower. However, the wellhead growth would increase as the thermal force decreases as shown in
Fig. 3. The surface hookup and pipeline must be designed to allow for the wellhead movement.

To reduce the wellhead growth, the conductor TOC needs to be at surface. Increasing the inner casing string TOC will also
reduce the wellhead growth, but only up to the same depth as the conductor TOC; any inner casing TOC higher than the
conductor TOC would not reduce wellhead growth any further.

If the inner casing TOC levels could not be placed at the designed depth, thermal stresses could be higher than the designed
SPE 157977 5

limit; in this case, casing tension could be applied during installation to compensate future thermal force. The “as-set-load at
WH” in Fig. 6 can be adjusted to minimize the operation total thermal force.

Considerations
Normally the effect of helical buckling on the thermal growth will be small in comparison to the total growth. This is because
the casing axial stiffness is large.

The contribution of the tubing thermal force at the wellhead is relatively low, mainly because its stiffness is very low.

If any casing is hung by a slip hanger at the wellhead, it is possible that the expansion of the uncemented section will result in
upward motion, unseating the slip hanger in the wellhead. This may result in dropping the casing below the hanger once it is
cooled off and losing the seal at the hanger.

When the wellhead thermal force is high, the conductor could be exposed to excessively high compression stress. In some old
offshore wells, when the conductor is corroded, the conductor would collapse due to excessive compressional load, and the
wellhead would drop as the conductor would not be able to support it [1].

The total thermal force on the conductor is always equal to the sum of the thermal force on the inner tubular strings when the
tubing or casing pressure is zero. The internal tubing or casing pressure will decrease the total wellhead thermal force and
increase the thermal growth, but not significantly.

The casing thermal stress and wellhead growth is a function of all casing string TOC levels and the temperature increment.

Wellhead growth may damage the surface lines during operation, and a hot and cold cycle may result in fatigue, which will
shorten metal element life. Internal locking or sealing elements may be damaged by high thermal stress in the wellhead. All
these risks should be considered during the casing and wellhead design phase.

Nomenclature
Ai cross sectional metal area of string i (in2)
AI internal cross-section area of casing
Ao external cross-section of casing
Ei Young’s modulus of string i (lbf/in2)
Fi-thermal force in string i due to temperature change (lbf) (negative for compression)
FEA total fixed end action (lbf)
FEAi fixed end action of string i (lbf)
FEAthermalfixed end action due to temperature change (lbf)
Ki stiffness of string i (lbf/in)
Kt total stiffness (lbf/in)
Li distance from bottom of wellhead to top of cement for string i (in)
!i the coefficient of linear thermal expansion
!Lthermal thermal growth (in) (upward growth is positive)
"L total wellhead growth (in)
"pI internal pressure change
"po external pressure change
!Ti the temperature change from the installed condition for string i (!F, positive for an increase)
" Poisson’s ratio

References
1. Aasen J. A., Aadnoy B. S.: Multistring Analysis of Well Growth, IADC/SPE 88024 presented at the IADC/SPE Pacific
Drilling Technology Conference and Exhibition, Kuala Lumpur, Malaysia, 13–16 Sept, 2004.

Type Size Depth TOC

(in) (ft) (ft)
Conductor 30 in 408 driven
Surface 185/8 in 1,893 400
Intermediate 133/8 in 5,011 1,384
Production 95/8 in 7,387 4,618
Liner 7 in 8,883
Tubing 41/2 in 8,078
6 SPE 157977

Table 1. Casing Depth and TOC

!"#$%#&'
()*+,-.'
/,-

Figure 1. Well Thermal Growth after Locking Pins Failed

SPE 157977 7

!"#$%&,*%%-*$.&!-*(?$%&'"()*&/@%123
$%##

$###

%##
!"#$%&'"()*&"+&,*%%-*$.&/0%123

#
# "## &## '## (## $### $"## $&## $'##

!%##

!$###

!$%##

!"###
4567859&:$;<+=&!>:&/2#3

)#*+,-.+/0+1234/56 )#*+,-.+/0+"# )#*+,-.+/0+&# )#*+,-.+/0+'#

)#*+,-.+/0+(# )#*+,-.+/0+$'#7 )#*+,-.+/0+"##7 )#*+,-.+/0+&##7

Figure 2
8 SPE 157977

!"##$"%&':$")=%#'()*+,$'-./0
'"#

'"!

&"#

&"!
!"##$"%&'()*+,$'-./0

%"#

%"!

$"#

$"!

!"#

!"!
! %!! '!! (!! )!! $!!! $%!! $'!! $(!!

1234526'7%8./9':;7'-<,0

&!*+,-.+/0+1234/56 &!*+,-.+/0+%! &!*+,-.+/0+'! &!*+,-.+/0+(!

&!*+,-.+/0+)! &!*+,-.+/0+$(!7+ &!*+,-.+/0+%!!7 &!*+,-.+/0+'!!7

Figure 3
SPE 157977 9

!"##$"%&'7$"):%#'()*+,$'-./0
#

'"&

'"%

'"$
!"##$"%&'()*+,$'-./0

'"#

'

!"&

!"%

!"$

!"#

!
! #! $! %! &! '!! '#! '$! '%! '&!

123'4%5./6'784'-9,0

'&()*&+,-./,01,'!2 '&()*&+,-./,01,#!2 '&()*&+,-./,01,$!2

'&()*&+,-./,01,%!2 '&()*&+,-./,01,'!!2 '&()*&+,-./,01,#!!2
'&()*&+,-./,01,)!!2 '&()*&+,-./,01,'!!!2 '&()*&+,-./,01,#!!!2

Figure 4
10 SPE 157977

!"#$%&,*%%-*$.&!-*(<$%&'"()*&/=%123
$%##

$###

%##
!"#$%&'"()*&"+&,*%%-*$.&/0%123

#
# "# &# '# (# $## $"# $&# $'# $(#

!%##

!$###

!$%##

!"###
456&7$89+:&!;7&/2#3

$(!%)(*+,-.+/0+"###1 $(!%)(*+,-.+/0+$###1 $(!%)(*+,-.+/0+%##1

$(!%)(*+,-.+/0+"##1 $(!%)(*+,-.+/0+$##1 $(!%)(*+,-.+/0+'#1
$(!%)(*+,-.+/0+&#1 $(!%)(*+,-.+/0+"#1 $(!%)(*+,-.+/0+$#1
$(!%)(*+,-.+/0+2345/67

Figure 5
SPE 157977 11

!""#$%&
'()("*&
*("+%$(&,-%$#.(&&

Figure 6. Annuli Cement Tensile Failure

12 SPE 157977

Wellhead Thermal Growth and Locking Load Modeling

Tubular OD (in) 30 18.625 13.375 9.625 4.5
Tubular ID (in) 28 17.655 12.615 8.681 3.958
2
A (in ) 91.104 27.639 15.513 13.572 3.600
2
E (lbf/in ) 3.00E+07 3.00E+07 3.00E+07 3.00E+073.00E+07
Thermal expansion (/F) 6.65E-06 6.65E-06 6.65E-06 6.65E-066.65E-06
dP (casing pressure psi) 0
dT (F) 61 116.3 113.0 111.3 126.2
Length of fixed point from WH, LOFP (in) (TOC) 1920 4800 16608 55416 96816
Stiffness K (lbf/in) 1423492 172741 28022 7347 1116
Fixed End Action (thermal force) FEA (lbf) 1117641 641105 349582 301366 90667
Fixed End Action (balloon force) FEA (lbf) 0 0 0 0 0
Fend (pressure force) (lbf) 0 0 0 0 0
Sum K = 1632719
Sum EFA = 2500361
Sum Fend = 0

Wellhead Thermal Growth, dL 1.53

Thermal force in each string F (lbf) 1062309 -376567 -306669 -290115 -88958
Check: SumFend = Sum F ===> Sum F 0

As-set-load at WH from TDAS (lbf) -430300 48200 121000 207200 53900

Wellhead equipment weight (lb) -22000
Total thermal force at WH (lbf) 1062309

WH Locking Load (lbf) 610,009

Figure 7. Thermal Force Acting on Wellhead and Constrained Wellhead Growth

Wellhead Thermal Growth and Locking Load Modeling

Tubular OD (in) 13.375 9.625 4.5
Tubular ID (in) 12.615 8.681 3.958
2
A (in ) 15.513 13.572 3.600
2
E (lbf/in ) 3.00E+07 3.00E+073.00E+07
Thermal expansion (/F) 6.65E-06 6.65E-066.65E-06
dP (casing pressure psi) 0
dT (F) 113.0 111.3 126.2
Length of fixed point from WH, LOFP (in) (TOC) 16608 55416 96816
Stiffness K (lbf/in) 28022 7347 1116
Fixed End Action (thermal force) FEA (lbf) 349582 301366 90667
Fixed End Action (balloon force) FEA (lbf) 0 0 0
Fend (pressure force) (lbf) 0 0 0
Sum K = 36485
Sum EFA = 741615
Sum Fend = 0

Wellhead Thermal Growth, dL 20.33

Thermal force in each string F (lbf) 0 0 220011 -152021 -67990
Check: SumFend = Sum F ===> Sum F 0

As-set-load at WH from TDAS (lbf) 121000 207200 53900

Wellhead equipment weight (lb)
Total thermal force at WH (lbf) 0

WH Locking Load (lbf) 0

Figure 8. Wellhead Growth after Locking Pins Sheared