20

THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI

FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY

ELASTIC-PLASTIC CONTACT MODELING.

PART II: NUMERICAL SIMULATIONS
AND PROGRAM VALIDATION
Sergiu SPÎNU, Emanuel DIACONESCU

Department of Applied Mechanics, “Stefan cel Mare” University of Suceava, ROMANIA

sergiu.spinu@fim.usv.ro

ABSTRACT
Elastic plastic spherical contact is simulated using an algorithm proposed in a
companion paper. Materials of contacting bodies are assumed to have different
behaviours: rigid (R), elastic (E), elastic-plastic with isotropic hardening (EP) or
elastic - perfectly plastic (EPP). Algorithm improvements allow for a high resolution
in digitization problem. A modified interference equation is proposed for the EP-EP
contact with symmetry about the common plane of contact. Program predictions are
matched against numerical results already published and a good agreement is
found. Experimental validation of residual print depth in the R-EPP contact gives
further confidence in the newly proposed algorithm.

Keywords: elastic-plastic contact, power hardening law, print depth

1. INTRODUCTION when a steel ball is pressed over a lead specimen,

An efficient algorithm for elastic - plastic contact
modeling is proposed in a companion paper.
Implementation of Fotiu and Nemat-Nasser's
universal algorithm for integration of elastoplasticity
equations [1], in solving the residual part of the
contact problem, is expected to decrease dramatically
the computational effort, allowing for finer grids in
the numerical simulation.
In this paper, numerical predictions of the newly
proposed algorithm are compared with already published
results, validating the computer code. The materials of
the contacting bodies are assumed to be either rigid (R),
or elastic (E), or elastic-plastic (EP), having a behavior
described by a power hardening law (Swift), or elastic-
perfectly-plastic (EPP). Four types of contacts are
considered: R-EP, E-EP, EP-EP with symmetry about
the common plane of contact and R-EPP.
Development of plastic region and of residual
stresses with application of new loading increments is
assessed, and contribution of residual state, which
superimposes elastic state induced by contact
pressure, is suggested.
Algorithm refinements allow for a fine grid, of
120 × 120 × 80 elementary cells, to be imposed in the
computational domain.
Residual displacement predicted by the
numerical program is matched against experimental
print depth obtained in a R-EPP contact scenario,
assumed as an elastic-perfectly plastic half-space.
A good agreement with the existing results, both
numerical and experimental, is found, giving
confidence in the newly advanced program.
2. R-EP CONTACT
The contact between a rigid sphere of radius
R = 105 ⋅ 10 −6 m and an elastic-plastic half-space is
simulated, allowing for comparison with results
published by Boucly, Nélias and Green [2]. Elastic
half-space parameters are: Young modulus,
E2 = 210GPa , Poisson's ratio, ν 2 = 0.3 . The
hardening law of the elastic-plastic material is chosen
as a power law (Swift), according to El Ghazal [3]:
σ Y ( e p ) = B( C + e p )n , (1)
with σ Y ( e p ) the yield strength function, e p the
effective accumulated plastic strain, expressed in
microdeformations, and the following parameters:
B=1.280 MPa, C = 30 , n = 0.085 .
The contact is loaded incrementally up to a
maximum value of W = 0.65N , for which the purely
elastic model (Hertz) predicts a contact radius
aH = 6.053μ m and a Hertzian pressure
p H = 8.470 MPa .
Dimensionless coordinates are defined as ratios
to aH , xi = xi aH , and dimensionless pressure or

THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY
stresses as ratios to pH . The computational domain is
a rectangular cuboid of sides L1 = L2 = 3aH ,
L3 = 1.6aH , which is dicretized with the following
parameters: N1 = N 2 = 120 , N 3 = 80 elementary
grid cells. Due to the fact that problem is
axisymmetric, three dimensional distributions are
depicted in the plane x2 = 0 only.
The imposed hardening curve is depicted in
Figure 1.
σY, [MPa]
e p ,[%]
Fig. 1. Swift's hardening law
Pressure profiles predicted by the numerical
program for six loading levels corresponding to
elastic-plastic domain are depicted in Figure 2. Hertz
pressure corresponding to maximum load is also
plotted for reference.
Fig. 2. Pressure profiles in the plane x2 = 0 ,
various loading levels
Elastic-plastic pressure distributions appear
flattened compared to the purely elastic case. At the
end of the loading loop, a central plateau of uniform
pressure can be observed in the vicinity of 6.5 pH .
This limitation of contact pressure results in an
increased elastic-plastic contact radius, compared to
its elastic counterpart, aH .
The same distributions were obtained by Jacq et
al. [4] by Boucly, Nélias and Green [2], using load
driven (ld) or displacement driven (dd) formulations,
and also by Benchea and Cretu [5], using finite
element analysis (FEA).
21
Profiles of residual prints corresponding to the
same six loading levels are depicted in Figure 3.
These profiles show that residual displacement
increase contact conformity in investigated non-
conforming contact, leading to a more uniform
distribution of contact pressure.
Fig. 3. Residual print profiles in elastic-plastic
spherical contact
The variation of residual print maximum depth
with the loading level is presented in Figure 4. This
curve was also obtained experimentally by El Ghazal
[3], numerically by Jacq et al. [4] and using FEA by
Benchea and Cretu [5].
Fig. 4. Residual print depth
Initiation of plastic flow occurs on the contact
axis, where von Mises equivalent stress firstly
exceeds initial yield strength. With application of new
loading increments, plastic zone expands to a
hemispherical domain (Fig. 5), while material
hardening state is modified according to Eq. (1).
The model assumes elastic and plastic strains,
which are of the same order of magnitude,
corresponding to elastic-plastic range. As plastic
strains are small, usually less than 2% , they can be
considered small strains and can be superimposed to
their elastic counterparts. This approach cannot be
applied to larger plastic strains, corresponding to fully
plastic range, solution of this scenario requiring FEA.
22
THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY
Towards the end of the loading cycle, the plastic
core approaches peripherally the free surface,
enveloping an elastic core. The evolution of
maximum effective accumulated plastic strain with
loading level is presented in Figure 6.
a) W = 0.25N
b) W = 0.55N
Fig. 5. Development of effective accumulated plastic
strain with different loading level.
Fig. 6. Maximum effective accumulated
plastic strain versus loading level
Plastic strains induce residual stresses, namely
elastic stresses that would persist after elastic
unloading. These stresses superimpose the ones
induced by contact pressure. The resulting state
generates further plastic strain if stress intensity
exceeds yield strength. Consequently, an accurate
estimation of stress field in the elastic-plastic body is
essential to plastic strain increment prediction.
Figures 7 and 8 depict distributions of
equivalent von Mises contact stress (stress induced by
contact pressure) and total stress in the elastic-plastic
half-space. The distribution of residual stresses for
this contact scenario was discussed in [6]. Residual
stress intensity is one order of magnitude smaller than
equivalent contact stress. The comparison of
distributions depicted in Figures 7 and 8 suggests that
residual stress reduces peaks in contact stress
intensity, thus making the resulting field more
uniform. This behavior is also suggested by the curves
traced in Figure 9. The maximum intensity of contact
stress increases more rapidly than the maximum of
the total field, due to the contribution of residual
stress. Consequently, residual stresses, which
represent material response to plastic flow, act to
impede further plastic yielding.
Fig. 7. Von Mises contact stress
Fig. 8. Total (contact and residual) Von Mises
stress in the elastic-plastic body
 23
THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY

Fig. 9. Maximum intensities of stress fields

versus loading level

On the central axis of the contact, residual and Fig. 11. Contact stress tensor components
contact stress tensor components vary according to on the central axis of the contact
r
Figures 10 and 11. The normal residual stress σ 33 is
nil on the free surface, tensile immediately under the
surface and compressive at Hertzian depths. On the
pr
other hand, normal contact stress σ 33 is
compressive.

Fig. 12. Residual hydrostatic pressure

in the plane x2 = 0

Fig. 10. Residual stress tensor components

on the central axis of the contact

Von Mises equivalent stress gives a good

indication of stress intensity, however it does not
indicate whether the stresses are compressive or
tensile. In most contact scenarios, this knowledge is
important as tensile stresses are known to favor crack
propagation, while compressive ones act against it. To
this end, a new parameter, namely the residual
hydrostatic pressure, defined as σ hr = I1r 3 , with I1r
the first invariant of residual stress tensor, is
investigated. Numerical simulations (Figs. 12 and 13),
suggest that σ hr is compressive immediately in
subsurface and at Hertzian depths, and tensile in Fig. 13. Residual hydrostatic pressure
between. on the central axis of the contact
24
THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY

3. E-EP AND EP-EP CONTACT

Normal residual displacement enters interference
equation, by superimposing the deflections induced by
contact pressure. When only one of the contacting
bodies, let it be body (2), is elastic-plastic and the
other one, let it be body (1), is elastic, the following
interference equation can be written:

h( i, j ) = hi( 1+ 2 ) ( i, j ) + u3pr( 1+ 2 ) ( i, j ) +
(2)
u3r( 2 ) ( i, j ) − ω ,
with h the gap in deformed state, hi the initial gap
(or initial contact geometry), ω rigid body approach,
u3pr the elastic part of displacement, namely that
induced by contact pressure, and u3r the residual part
of displacement, namely that induced by plastic
region.
On the other hand, when contacting bodies are
both elastic-plastic, Eq. (2) encloses residual
displacements of both surfaces:
Fig.14. Pressure profiles for various
material behaviors
Variations of maximum effective plastic strain
with loading level, in the E-EP and in the EP-EP
contact respectively, are depicted in Fig. 15. Intensity
of plastic strains in the E-EP contact is up to 40%
higher than the one corresponding to the EP-EP
scenario.

h( i, j ) = hi( 1+ 2 ) ( i, j ) + u3pr( 1+ 2 ) ( i, j ) +
(3)
u3r( 1+ 2 ) ( i, j ) − ω .
If the hardening behavior or contacting bodies is
dissimilar, residual displacement should be computed
for every body separately. The model is simplified
considerably if the bodies follow the same hardening
law and have the same initial contact geometry due to
the symmetry of the problem about the common plane
of contact,

u3r( 1 ) = u3r( 2 ) . (4)

Consequently, Eq. (3) becomes: Fig. 15. Maximum effective accumulated
plastic strain versus loading level
h( i, j ) = hi( 1+ 2 ) ( i, j ) + u3pr( 1+ 2 ) ( i, j ) +
(5)
Variations of maximum pressure with the
2u3r( 2 ) ( i, j ) − ω .
loading level in the E-E, the E-EP and the EP-EP
To validate Eq. (5), the contact between two contact, are depicted in Figure 16. The curves
spheres of radius R = 0.015 m is simulated presented in Figures 15 and 16 match well the results
numerically, for two different material behaviors: of Boucly, Nélias, and Green [2].
elastic, and elastic-plastic following Swift's law, with
the following parameters: B = 945MPa , C = 20 ,
n = 0.121 .
The contact is loaded up to a level of
W = 11.179 N , corresponding to a Hertzian pressure
pH = 8 GPa and to a Hertz contact radius
aH = 817 μ m .
Pressure distributions obtained using Eqs. (2)
and (5) respectively, depicted in Figure 14, agree well
with already published results [2]. As expected, in the
EP-EP contact, pressure appears more flattened
compared to the E-EP case, due to a more pronounced
increasing in contact conformity related to the
doubling of the residual term. Fig. 16. Maximum pressure versus loading level
 25
THE ANNALS OF “DUNĂREA DE JOS“ UNIVERSITY OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY

4. EXPERIMENTAL VALIDATION Contact between a steel ball, assumed as a rigid

As Contact Mechanics uses simplifying
assumptions in order to circumvent the mathematical
complexity of the arising equations, experimental
validation is needed to verify model viability. An
extended program of experimental research was
conducted in the Contact Mechanics Laboratory of the
University of Suceava, aiming to assess residual print
parameters in rough elastic-plastic non-conforming
contacts. The stand used for the loading experiments
was originally designed by Nestor et al. [7].
Microtopography of deformed surface was scanned
with a laser profile-meter UBM14.
indenter, and a lead specimen, simulating the elastic-
plastic half-space, was loaded up to an equivalent
Hertzian pressure pH = 0.94 GPa . The contact was
also simulated using the numerical formulation. As
lead is best described as an EPP material, a linear
hardening law with a very small slope was considered
in the numerical model. As stated in [4], the plastic
strain increment is undefined when assuming an EPP
material behavior.
Residual prints at two loading levels, correspond
to Hertzian pressures of 0.48 GPa and of 0.94 GPa ,
are depicted in Figures 17 and 18.

Fig. 17. Residual print, pH = 0.48 GPa .

Fig. 18. Residual print, pH = 0.94 GPa .

 THE ANNALS OF UNIVERSITY “DUNĂREA DE JOS“ OF GALAŢI
FASCICLE VIII, 2010 (XVI), ISSN 1221-4590, Issue 1
TRIBOLOGY
Variation of print depth with loading level is
presented in Figure 19. The agreement between the
values predicted numerically and those obtained
experimentally is considered satisfactory, giving the
complexity of the phenomena involved.
Fig. 19. Residual print depth versus loading level
5. CONCLUSIONS
The newly proposed algorithm is used to simulate,
with a high resolution of 120x120x80 elementary
cells, the spherical contact between bodies with
various behaviors: R-EP, E-EP, EP-EP and R-EPP.
Elastic-plastic pressure appears flattened
compared to the elastic case, due to changes in
hardening state of the EP material, and in contact
conformity due to superposition of residual
displacement in interference equation.
The plastic zone, initially occupying a hemi-
spherical region located at Hertzian depths, advances
toward the free surface with increased loading,
enveloping an elastic core. This development is
consistent with existing models for the elastic-plastic
process, marking the passing from elastic-plastic
range to fully plastic.
Residual stress intensity is one order of
magnitude smaller than equivalent stresses induced by
contact pressure. They contribute to total elastic field
by decreasing the peaks in contact stress intensity,
thus impeding further plastic flow.
A new interference equation is proposed for
solving the EP-EP contact with similar hardening
behavior and symmetry about the common plane of
contact.
Residual displacement predicted numerically for
the R-EPP contact matches well the print depth
obtained experimentally in indentation of a lead
specimen, assumed as an EPP half-space, with a steel
ball assumed as a rigid indenter.
ACKNOWLEDGEMENT
This paper was supported by the project
“Progress and development through post-doctoral
research and innovation in engineering and applied
sciences – PRiDE - Contract no. POSDRU 89/1.5/
S/57083”, project co-funded from European Social
Fund through Sectorial Operational Program Human
Resources 2007-2013.
REFERENCES
1. Fotiu P. A., Nemat-Nasser S., 1996, A Universal Integratio
Algorithm for Rate-Dependant Elastoplasticity, Comput. Struct.,
59, pp. 1173-1184.
2. Boucly V., Nélias D., Green I., 2007, Modeling of the Rolling
and Sliding Contact Between Two Asperities, J. of Tribology
(Trans. ASME), vol. 129, pp. 235-245.
3. El Ghazal H., 1999, Etude des proprietes microstructurales et
mecaniques des aciers 16NiCrMo13 cemente et 32CrMoV13
nitrure - Application a la prevision de leur limite d’endurance en
fatigue de roulement, PhD Thesis, INSA Lyon, France.
4. Jacq C., Nélias D., Lormand G., Girodin D., 2002, Develop-
ment of a Three-Dimensional Semi-Analytical Elastic-Plastic
Contact Code, J. of Tribology (Trans. ASME), vol. 124, pp. 653-
667.
5. Benchea M., Creţu S., 2008, An Improved Incremental Model
to Analyse Elastic - Plastic Concentrated Contacts – The Finite
Element Analysis and Validation, Acta Tribologica, vol. 16.
6. Spinu S., Diaconescu E., 2009, A Fast Numerical Method to
Predict Elastic Fields Due to Eigenstrains in an Isotropic Half-
Space - Part II. Program Validation and Simulations, The Annals of
“Dunărea de Jos“ University of Galati, Fascicle VIII, Tribology,
(XV), Issue 2, , pp. 197-204.
7. Nestor T., Prodan D., Pătraş-Ciceu S., Alaci S., Pintilie D.,
1996, Stand pentru determinarea histerezisului static la solici-tarea
de contact, (in Romanian), Proceedings of VAREHD 8, Suceava.