
considering a zero fluid pressure and a zero friction
coefficient. results were obtained with the FE model
presented in Figure 4 and compared to the results of the
values of equation (1) developed by Lindley [2] and those
published by Kim et al. [32]. We note a very good
agreement between the results obtained by the analytical
model and the results of the FE model. However, there is a
significant discrepancy between the two models and the
experimental test of Kim et al. for compression ratios
greater than 15%. This difference may be due to the
geometrical and physical data considered by the authors
when establishing their model.
Both the analytical and numerical models are used to
determine the contact width values and contact pressure
distribution profiles when a clamping load is applied. Figure
6 shows a comparison between the contact width values
given by equation (2) and the FE model results shown in
Figure 4. The FE model results are the average value of the
last nodes in contact and the first nodes not in contact. The
difference between the numerical and analytical curves is at
most 5% at a clamping force of 350 N, which corresponds
to a compression ratio of 27% in this case. It is important to
recall that Lindley [2] describes the relationship of contact-
displacement force, contact width, and contact pressure as a
function of O-ring radial position for a maximum O-ring
compression ratio of 25%.
Figure 7 shows the contact pressure distributions on the
surface A1 defined in Figure 1 for four values of clamping
force. This figure also shows a comparison between the
analytical and numerical models. The contact pressure
profiles have a parabolic shape that confirms the Hertzian
shape. The comparison between the values of the analytical
model and the results of the FE analysis indicates that the
maximum contact pressure increases as the clamping force
increases and the difference between the two models do not
exceed 4%. From these remarks, we can confirm that there
is a good agreement between the two approaches, analytical
and FE, when the only load applied is the clamping force.
The effect of friction was considered from the finite element
simulation of assemblies 1 and 2 shown in Figures 3 and 4.
The axial and radial deformations of the O-ring as a
function of clamping force and fluid pressure are obtained
for
values of 0, 0.1, and 0.2 (Figure 8). This figure
shows the importance of the location of this type of seal in
a groove. For a clamping force of 700 N and a fluid
pressure of 0.75 MPa, the increase in the friction
coefficient considerably reduces the axial and radial
deformation of the seal. Indeed, for friction coefficients
0, 0.1 and 0.2. The elongation decreases respectively
from 66.5, 19.5 to 2.7%. Regarding the compression ratio,
it decreases respectively from 48.9, 37.3 to 30.9. In
conclusion, the deformation of the joint as a function of
the clamping load and the pressure of the fluid depends
strongly on the value of the friction coefficient.
Figure 9 shows the variation of the maximum
contact pressure, at the surfaces B1, B2 and B3 defined in
Figure 3, for two values of friction coefficients 0.05
and 0.2,
according to the pressure of the confined fluid. It can be
seen that the influence of the friction coefficient on the
contact pressure is negligible. Furthermore, by studying
Figure 10, the increase of the friction coefficient can
decrease the maximum value of the von Mises stress inside
the O-ring installed in a groove for high fluid pressures and
the opposite for low fluid pressures.
Given the particularity of the material characteristics of the
O-ring, a parametric study to understand the influence of
Young's modulus E becomes necessary. Several simulations
were performed, for the same geometry of the FE model
presented in Figure 3, for Young's modulus ranging
from 6.96 MPa to 17.3 MPa. The numerical calculations
were carried out with a coefficient of friction of 0.2.
Figure 11 shows the evolution of the compression ratio
C, obtained for three values of Young's modulus E,
weighting the two phases of deformation of the seal:
clamping phase F, and pressurization phase P. The results
delivered by this figure serve as a basis for determining the
influence of these three parameters on the compression
ratio of the seal in the throat. It is obvious that the
compression ratio increases with increased clamping force
and decreases with increasing fluid pressure. As can be
seen in the same figure, the compression ratio is
proportional to Young's modulus. It can be observed in
the clamping phase that, the stiffer the seal material, the
lower the compression ratio will be. On the other hand,
when the fluid pressure is applied and increases
further, the value of the compression ratio decreases.
It is also observed that the curves corresponding to the
three values of Young's modulus are inverted when the
fluid pressure exceeds 4.75 MPa.
Figure 12 shows the contact pressure distribution curves at
the contact surfaces B1, B2 and B3 defined in Figure 1, for
a compression ratio of 20% and a fluid pressure of 5 MPa.
The simulations are performed for several values of E. The
curves in this figure confirm that the stresses are
proportional to the moduli of elasticity for the same
deformation. Thus, when the joint stiffness is greater, the
contact pressure is greater. Figure 13 shows the variation of
the maximum value of Von Mises stress inside the seal for
three values of E as a function of fluid pressure for a
compression ratio of 20%. Regardless of the value of E, the
maximum Von Mises stress increases with increasing rates
as the fluid pressure increases. We note that the difference
between the three seal hardnesses shows a high value for
pressures below 5.5 MPa and remains relatively low for
fluid pressures above this value. This result does not
surprise us. Indeed, whatever the value of E, when the fluid
pressure is applied and its value increases, the zone where
the stress is maximum moves towards the outside of the
joint to the side of the extrusion gap. Lastly, it is worth
mentioning that the compression ratio as a function of
clamping force - Numerical/ analytical/ experimental
comparison is presented in Figure 5.
WSEAS TRANSACTIONS on APPLIED and THEORETICAL MECHANICS
DOI: 10.37394/232011.2022.17.27
E. El Bahloul, H. Aissaoui,
M. Diany, E. Boudaia, S. Touairi