Original Article

Proc IMechE Part C:

J Mechanical Engineering Science
Effect of the friction coefficient for contact 2014, Vol. 228(16) 2881–2887
! IMechE 2014

pressure of packer rubber Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0954406214525596
pic.sagepub.com

Weiguo Ma1, Baolong Qu2 and Feng Guan1

Abstract
A systematic study of the packer rubber contact pressure under a fixed-displacement load is conducted to gain further
insight into the packer seal mechanism. A Y221-114 double rubber packer is investigated using the finite element
software ANSYS, where a design of experiments method is utilized to study the effects of the friction coefficient.
The results show that the friction coefficient of the packer and the tubing had the greatest effect on contact pressure
than other factors. Decreasing the rubber friction coefficient is conducive to forming the double rubber seal and
increasing the maximum contact pressure working range. However, there is additionally a slight decrease in the value
of maximum contact pressure. The results of the study provide valuable insight into the importance of packer design
optimization.

Keywords
Packer rubber, contact pressure, design of experiments, finite element method, friction coefficient

Date received: 10 August 2013; accepted: 4 February 2014

Introduction
A packer is a necessary down-hole tool used in the
process of oil layering. Its core component is the
packer rubber. When the packer rubber supports an
axial load such as gravity or pressure from a liquid, it
experiences a relatively large deformation, which pro-
duces a larger contact pressure and forms a seal
between the rubber and the casing, resulting in sealing
of the annular gap and isolation of the production
layer. Thomas et al.1 proposed that the load capacity
of the packer rubber is a function of the pressure
(or stress) that exists during the sealing process. To
achieve and maintain the seal, the pressure (or stress)
should be greater than the working value. The max-
imum contact pressure between the rubber and the
casing will directly determine the quality of the
packer seal; therefore, research on the factors that
affect the contact pressure is extremely important to
understand the seal mechanism and to optimize
packer design.
The material, molding process, size, shape, and
load mode of the packer rubber each influence the
contact pressure between the rubber and the casing
to a different extent. There are several published solu-
tions to the contact pressure problem,1–4 but the lit-
erature does not consider frictional contact between
the rubber and the casing. Recently, scholars have
begun to focus on the effect of various frictional coef-
ficients for the contact pressure,5–8 therefore provid-
ing through numerical simulation more accurate
results than suggested by classical theory. However,
the new methods do not consider the difference
between the transverse and radial coefficient of the
rubber or the effect of the frictional coefficient
between the support ring and the tubing.
This paper reports the results of a systematic study
conducted using a Y221-114 double packer. The finite
element software ANSYS, along with a design of
experiments method, and an orthogonal array testing
procedure implemented numerically is used to inves-
tigate how and to what extent the frictional coefficient
of the component near the rubber is affected by a
contact pressure. The research introduces concrete
measures and countermeasures for improving the con-
tact pressure of packer rubber.
Material model and calculation model
Material model
Packer rubber is primarily made of rubber materials,
specifically, hyperelastic materials, which deform
1
College of Mechanical Engineering, Yangtze University, Jingzhou, China
2
College of Chemistry and Environmental Engineering, Yangtze
University, Jingzhou, China
Corresponding author:
Baolong Qu, College of Chemistry and Environmental Engineering,
Yangtze University, Jingzhou 434023, China.
Email: qubaolong_78@163.com

Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015

2882
nonlinearly when subjected to an external force.2,9
Available research on the mechanical properties of
swelling rubber is limited.10,11 With the development
of the finite element method (FEM) and computer
technology, the mechanical properties of rubber
materials have been widely researched. Currently,
the strain energy density function is commonly used
in FEM to indicate the mechanical properties
of rubber materials, as shown in the hyperelastic
constitutive models of Neo-Hookean,12 Yeoh,13 and
Mooney–Rivlin.14,15 This research adapts the
Mooney–Rivlin model, which simulate uniaxial ten-
sile stress and strain, to study the friction contact
behavior of packer rubber.
The strain energy density function W of a rubber
material is a function of the right Cauchy–Green
deformation tensor invariant I1, I2, I3, that is
W ¼ WðI1 , I2 , I3 Þ
where
8
>
> I1 ¼ 21 þ 22 þ 23
>
<I
2 ¼ 21 22 þ 22 23 þ 23 21
>
> I3 ¼ 21 22 23
>
: MPa and C10 is 2.36 MPa, then C01 is 1.18 MPa.
i ¼ 1 þ
i
i is the principal stretch ratio, and
i is the main
strain.
Based on the stress–strain relationship, Mooney14
established a constitutive relationship for the rubber
material via phenomenological theory. As the rubber
materials are isotropic and incompressible, I3  1.
Rivlin15 studied the model with respect to the
Mooney theory and developed the strain energy dens-
ity model of incompressible materials, where Cij is a
mechanical property constant.
X
1
W¼ Cij ðI1  3Þi ðI2  3Þj
i¼0,j¼0
The Mooney–Rivilin model of rubber materials
can simulate most of the mechanical behavior of the
rubber materials. From the calculation, the Mooney–
Rivlin model can be transformed into a two-, three-,
Figure 1. Packer Y221-144 structure.
1 – Upper joint, 2 – Tubing, 3 – Packer rubber, 4 – Support ring, 5 – Pressure sleeve, 6 – Pin, 7 – Cone, 8 – Anchoring part.
Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015
Proc IMechE Part C: J Mechanical Engineering Science 228(16)
five-, or nine-parameter model. The two-parameter
model is the most commonly used.
W ¼ C10 ðI1  3Þ þ C01 ðI2  3Þ ð4Þ
The model provides a good description of the
mechanical properties of rubber materials when
deformation is less than 150% and completely satisfies
the performance calculation of packer rubber
materials.
In the Mooney–Rivlin model with two parameters,
the Rivlin coefficient C10 and C01 are positive con-
stants that can be determined by a uniaxial tensile
test of the material.
When the material is incompressible, the relation-
ship between the Rivlin coefficient and Young’s
modulus is determined as follows16
ð1Þ E ¼ 6ðC10 þ C01 Þ ð5Þ
Testing shows that the ratio of C01 to C10 is a con-
stant and ranges from 0.25 to 0.5. C01 and C10 can be
calculated using equation (5) when Young’s modulus
is known. The research presented in this paper fixes
ð2Þ the ratio at 0.5. When Young’s modulus (E) is 21.2
Calculation model
The Y221-114 double rubber packer structure is
shown in Figure 1. Here, Y means compression
type, 2 means one-way slips fixed method, the
second 2 means the sealing by turning the tubing, 1
means the releasing by lifting the tubing, and 114 is
the outer diameter of the packer.
The rubber, support ring, tubing, and casing of the
Y221-114 packer investigated in this study are axially
aligned. Additionally, the loads are applied axially;
therefore, the problem can be reduced to a two-
dimensional one for each component. The model is
ð3Þ applied to a section through the axis. The component
of interest is the rubber; therefore, the geometric
model shows the rubber and the component sur-
rounding the rubber, as shown in Figure 2.
The upper and lower end of the tubing and casing
are considered fixed when the packer is working. The
lower support ring is fixed; the upper support ring has
Ma et al. 2883

seal pressure applied, causing the rubber to expand
radially. As the seal pressure increases, the rubber
comes into contact with the casing. When the pressure
reaches a certain value, a sealed state is achieved; thus,
the annulus between the tubing and the casing is
sealed. The end angle of the rubber is 45 , and the
length is 6 mm. The remaining geometric and mech-
anical parameters are shown in Table 1.
Numerical calculation and design
of experiments
Numerical calculation
The relationship between the rubber and the sur-
rounding components is best expressed as a frictional
contact problem; therefore, ANSYS can be used to
analyze the contact pressure of the rubber. The
packer rubber is a rubber material with a large
deformation capacity; therefore, the hyperelasticity
element Plane 182, which is a quadrilateral element
with four simulation nodes, is adopted. It is defined
by four nodes having two degrees of freedom at each
node: translations in the nodal x and y directions. The
tubing, casing, and support ring are metal materials
with small deformations; therefore, the element Plane
42, which is also a quadrilateral element with four
nodes, is adopted. Element size is set to 2 mm, and
the geometric model is meshed as shown in Figure 2.
Mesh results are shown in Figure 3. The total number
of elements is 4138, and the total number of nodes is
4572. Each node has two degrees of freedom in the
X and Y directions of the Plane 42 and Plane 182
elements; hence, the total degrees of freedom in the
finite element model after being meshed is 9144.

Figure 2. The schematic diagram of double rubber packer.

1 – Support ring, 2 – Casing, 3 – Upper rubber, 4 – Tubing,
5 – Lower rubber. Figure 3. Finite element model.

Table 1. The packer parameters.

Inside Outer Height Young’s Poisson’s

Component diameter (mm) diameter (mm) (mm) modulus, E (MPa) ratio, 

Tubing 51.0 70.0 – 2.1  105 0.25

Casing 121.4 139.7 – 2.1  105 0.25
Support ring 70.0 114.5 15 2.1  105 0.25
Rubber 70.0 114.0 75 5.3 0.49

Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015

2884 Proc IMechE Part C: J Mechanical Engineering Science 228(16)

Because the rubber has frictional contact with the and deformation, and it will make contact with the
corresponding position of tubing, support ring, and casing. When the contact pressure reaches a certain
casing, the support ring has contact with the tubing. value, a sealed state in the annulus between the tubing
All the contacts are face-to-face. The contact elem- and casing is achieved. In this state, only the rubber
ent uses Contact 171, and the target element uses area where the contact pressure is maximal acts as a
Target 169. seal. That is, the quality of the seal depends on the size
Based on the working state of the packer, the dis- and range of the maximum contact pressure.
placement constraint is set for the upper and lower Additionally, to form a simultaneous seal with two
end of the tubing and casing, the transverse displace- rubber packers and to avoid seal failure, the max-
ment constraint is set the for all the support rings, and imum contact pressure between two rubbers should
the radial displacement constraint is set for the lower have similar values for engineering applications.
support ring. A vertical displacement load of 25 mm is Using the design of experiments, the research in
applied to the upper support ring. A simulation of the this paper determines the maximum contact pressure
mechanical behavior of the rubber in the sealed state difference (referred to as the pressure difference in the
can now be performed. following) between the upper and lower rubber. From
the three factors and the three levels of the design of
experiments, the effects of the frictional coefficient are
Design of experiments of friction coefficient analyzed for the pressure difference in different parts
Packer rubber is a polymer material with a molecular of the double packer rubber. Experiment’s plan and
weight greater than 200,000. Rubber products are its results are shown in Table 3.
usually made by compression molding, which can According to the design of experiments, the
easily cause macromolecular chains to be locally experiments results are analyzed by the range
aligning along certain directions during the molding method, which is a simple calculation with strong
process. Additionally, because of the difference in the intuition, and the results of the analysis are shown
mold and molding shrinkage, the transverse and in Table 4.
radial surface roughness of the rubber will be differ-
ent. Therefore, the packer rubber has different fric-
tional coefficients in the transverse and radial Table 3. Design of experiments plans and results.
directions. Friction
In the simulation of the packer rubber in a sealed coefficient Results/MPa
state, the positions that have frictional contact behav- Test (pressure
ior are the rubber and the support ring, the rubber number f1 f2 f3 difference)
and the tubing, and the rubber and the casing.
1 1 (0.2) 1 (0.3) 1 (0.3) 7.81
The frictional coefficient is divided into three cases:
2 1 2 (0.5) 2 (0.5) 10.03
the rubber and the support ring (f1), the rubber and
3 1 3 (0.7) 3 (0.7) 12.59
the tubing (f2), the rubber and the casing (f3). Each
friction coefficient has three levels. A design of experi- 4 2 (0.4) 1 2 9.31
ments is carried out and produces results with three 5 2 2 3 12.22
factors and three levels. The contact behavior between 6 2 3 1 13
the rubber and the support ring, the tubing, and the 7 3 (0.6) 1 3 9.68
casing represents the rubber–metal material friction. 8 3 2 1 10.98
Specific friction coefficients for each factor and level 9 3 3 2 14.69
are shown in Table 2.

Results
Table 4. The results analysis of design of experiments.
The main mechanism of the packer is that the rubber
Friction coefficient f1 f2 f3
bears an axial pressure, resulting in radial expansion
I – Sum of the first level 30.43 26.8 31.79
test result
Table 2. Effect factors and levels in the design of experiments. II – Sum of the second 34.53 33.23 34.03
level test result
Friction Rubber and Rubber Rubber III – Sum of the third level 35.35 40.28 34.49
coefficients support and and test result
Levels ring (f1) tubing (f2) casing (f3) I/3 10.1433 8.9333 10.5967
First level 0.2 0.3 0.3 II/3 11.51 11.0767 11.3433
Second level 0.4 0.5 0.5 III/3 11.7833 13.4267 11.4967
Third level 0.6 0.7 0.7 Range R 1.6400 4.4933 0.9000

Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015

Ma et al. 2885

The following results are observed from the data: The following results are observed from the figures:

1. In the three experiments in which the factor levels 1. Decreasing the frictional coefficient will assist in
of frictional coefficient f1 between the rubber and reducing the maximum contact pressure of the
the support ring are the same, f1 ¼ 0.2 is compared upper rubber.
with f1 ¼ 0.6, and the pressure increases by 2. When the frictional coefficient f1 ¼ 0.5, f2 ¼ 0.2,
1.6400 MPa. f3 ¼ 0.7, the lower rubber has the maximum con-
2. In the three experiments in which the factor levels tact pressure.
of frictional coefficient f2 between the rubber and
the tubing are the same, f2 ¼ 0.3 is compared to
f2 ¼ 0.7, and the pressure difference increases by
4.4933 MPa.
Discussion
3. In the three experiments in which the factor levels To verify the validity of the above conclusions, two
of frictional coefficient f3 between the rubber and different working conditions were set up: working
the casing are the same, f3 ¼ 0.3 is compared to condition A ( f1 ¼ f2 ¼ f3 ¼ 0.7) and working condi-
f3 ¼ 0.7, and the pressure difference increases by tion B ( f1 ¼ 0.5, f2 ¼ 0.2, f3 ¼ 0.7). Working condition
0.9000 MPa. B adjusts the friction coefficient to decrease the pres-
sure difference.
According to the analytical results, the following con- Figure 6 show the contact pressure distribution of
clusions can be drawn: the rubber and the casing under working conditions A
and B. According to the figures, the pressure differ-
1. For the pressure difference, the frictional coeffi- ence of working condition A is higher than that of
cient f2 has the greatest effect, followed by the
frictional coefficient f1, and the frictional coeffi-
cient f3 has the least effect.
2. Decreasing the frictional coefficient will assist in
reducing the pressure difference.
3. By adjusting the friction coefficient, if the pressure
difference is reduced, the maximum contact pres-
sure of the upper and lower rubber can be reduced
to some extent.

The maximum contact pressure between the rub-

ber and the casing will directly determine the quality
of the packer seal. So, it is necessary to study the
influence of friction coefficient on the maximum con-
tact pressure of the rubber. Similar to the
above orthogonal array testing method, using the
design of experiments, the results are shown in Figure 5. The maximum contact pressure of the lower
Figures 4 and 5. rubber under the difference factor level.

Figure 4. The maximum contact pressure of the upper Figure 6. Contact pressure under working conditions A
rubber under the difference factor level. and B.

Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015

2886
working condition B. According to the data from the
numerical simulation, the maximum pressure differ-
ence of working condition A is 15.43 MPa; however,
the maximum pressure difference of working condi-
tion B is only 8.03 MPa. Therefore, adjusting the fric-
tional coefficient of the rubber reduce the pressure
difference. In addition, compared to that of working
condition A, the maximum contact pressure between
the upper rubber and the casing of working condition
B is reduced by 5.44 MPa. But the maximum contact
pressure between the lower rubber and the casing is
increased by 1.94 MPa. Furthermore, by reducing the
frictional coefficient of the rubber, the working range
of the maximum contact pressure increases, which is
beneficial to protect the rubber, instead of avoiding a
large contact pressure in the smaller area, which
causes local damage to the rubber and seal failure.
The simulation results verify the proposed
conclusions.
The conclusions can be explained. First, when the
packer is sealing, the relative contact between the
rubber and the tubing is much greater, and its friction
can be regarded as sliding friction, but relative
displacement is smaller in other positions.
Consequently, the friction coefficient between the
rubber and the tubing can be heavily affected by the
contact pressure. Second, by decreasing the frictional
coefficient of the rubber, the frictional force of the
upper rubber can also be reduced, which is beneficial
to the transmission of the setting pressure from the
upper rubber to the lower rubber through the middle
support ring, and it causes a decrease in the maximum
pressure difference. In addition, decreasing the fric-
tional coefficient of the rubber causes the upper and
lower rubbers to work simultaneously and disperses
the contact pressure; therefore, the pressure difference
of the rubber is reduced. Finally, by increasing the
friction coefficient between the rubber and the
tubing, the pressure loss of the upper rubber is
increased, resulting in the decrease of maximum con-
tact pressure of the lower rubber.
Conclusions
By focusing on the packer rubber and utilizing the
Mooney–Rivlin rubber material model, a relevant
numerical model is established to achieve a numerical
simulation of the double packer rubber in the sealing
state. Using an orthogonal array testing procedure
was implemented numerically, and this paper studies
the effect of the maximum contact pressure of the
upper and lower rubber on the frictional coefficient
of the rubber and the surrounding components. The
research results show that (1) the effect of the fric-
tional coefficient between the rubber and the tubing
is greater for the contact pressure difference; (2)
decreasing the frictional coefficient of the rubber is
Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015
Proc IMechE Part C: J Mechanical Engineering Science 228(16)
beneficial for the double packer to work simultan-
eously; (3) any measure that leads to the double
packer working simultaneously leads to a reduction
in the maximum contact pressure; and (4) reducing
the frictional coefficient of the rubber can lead to an
increase in the working range of the maximum contact
pressure, which helps protect the rubber.
Conflict of interest
None declared.
Funding
Authors acknowledge the financial support provided by
National Science and Technology Major Project of China
(2011ZX05036-006).
References
1. Institute of oil production technology of Jianghan pet-
roleum administration. Theoretic foundation and appli-
cation of packer rubber. Beijing: Petroleum Industry
Press, 1983, pp.85–94. (in Chinese).
2. Guo ZP, Wang YF, Sun B, et al. Analysis and struc-
tural improvement of the rubber part in packer in a way
of non-linearity finite element. In: 2nd international
conference on MACE, IEEE, Inner Mongolia, China,
2011, pp.73–76.
3. Liu YQ, Guo F and Xu JP. Analyzing packer’s deform-
ation of tubular for unsetting process in HTHP Wells
under variable (T, P) fields. Open Petrol Eng J 2012; 5:
109–117.
4. Guo Z, Wang Y, Li Q, et al. The FEA contact analysis
of high pressure packers. Adv Mech Electron Eng 2012;
176: 119–124.
5. Jia SP, Yan XZ and Yang L. Frictional contact analysis
of packer rubber with large deformation by finite elem-
ent method. Lubric Eng 2005; 4: 71–74. (in Chinese).
6. Yang XJ, Yan XZ and Jia SP. Frictional contact ana-
lysis of packer rubber with large deformation based on
adhesive-slip frictional contact method. J Mech
Strength 2006; 28(2): 229–234. (in Chinese).
7. Li B and Zhang SM. Contact pressure research of drill
pipe and packer of rotating blowout preventer. Appl
Mech Mater 2012; 121–126: 3200–3204.
8. Ma WG, Zhang YC, Zhang DB, et al. Sealing analysis
of packer with double rubber. China Petrol Mach 2010;
38(11): 51–53. (in Chinese).
9. Atkinson C, Desroches J, Eftaxiopoulos DA, et al.
Wellbore stresses induced by the nonlinear deformation
of an inflatable packer. J Eng Math 2001; 41(4):
305–327.
10. Evers R, Young D, Vargus G, et al. Design method-
ology for swellable elastomer packers in fracturing
operations. In: SPE annual technical conference and
exhibition, vol. 6, 2008, pp.3710–3721.
11. Al-Yami A, Nasr-El-Din H, Al-Arfaj M, et al.
Investigation of water-swelling packers. In: SPE gas
technology symposium, vol. 2, 2008, pp.468–478.
12. Treloar LRG. The elasticity of a network of long chain
molecules (III). Trans Faraday Soc 1946; 42(1): 83–93.
13. Yeoh OH. Some forms of the strain energy for rubber.
Rub Chem Technol 1993; 66(5): 754–771.
Ma et al. 2887

14. Mooney MJ. A theory of large elastic deform at ion.
J Appl Phys 1940; 11(6): 582–592.
15. Rivlin RS. Large elastic deformations of isotropic
materials, IV: Further developments of the general
theory. Philos Trans R Soc Lond Ser A: Math Phys
Sci 1948; 241(835): 379–397.
16. Wang W, Deng T and Zhao SG. Determination for
material constants of rubber Mooney-Rivlin model.
Special Purpose Rubber Products 2004; 25(4): 8–10.
(in Chinese).

Downloaded from pic.sagepub.com at Freie Universitaet Berlin on May 6, 2015