Extended abstract

World Tribology Congress 2013
Torino, Italy, September 8 – 13, 2013
The effects of Lubricants with Additives on the Friction Force of Smooth and
Artificially Textured Piston Rings
Anastasios Zavos1), Pantelis Nikolakopoulos2)*
1)
Department of Mechanical Engineering and Aeronautics, University of Patras,
Patra, 26504 Patra, Greece
2)
Department of Mechanical Engineering and Aeronautics, University of Patras,
Patra, 26504 Patra, Greece
*
Corresponding author: [email protected]
1. Introduction
The role of lubricant oil with additives in engine
performance has been the subject of several studies in
recent years. In particular, the influence of the piston ring
pack with oil film has generated significant interest.
The evolution of new generation C.I engines has
created lubrication problems, that cannot be solved with
mineral oils. This has led to the development of synthetic
mixtures which the properties depend on from basic
components as synthesis process and additives.
The rheological behavior of low sulphated ash,
phosphorus and sulfur lubricants, using an
elastohydrodynamic tribometer and viscometer,
investigated from Meunier et al. [1]. They examined the
rheological parameters for fresh and aged lubricants, as
well as the influence of particles sizes contained in the
lubricants.
Rajalingam et al. [2], developed a theory of piston
ring lubrication considering a non-lubricant Newtonian
oil. They presented the effects of pseudoplasticity to the
minimum oil thickness, peak pressure ratio and friction
coefficient. Wakuri et al. [3], presented in their paper an
analytical and experimental study regarding the piston
rings basic tribological parameters such us, the friction
forces and oil film thickness. Their analytical model is
based on the Reynolds equation, while the experimental
results received using a four stroke diesel engine. Later
on, these results have been used by the same authors to
make predictions of the friction in the mixed lubrication
regime. Hamatake et al. in [4], shown that the piston ring
friction influenced from a kind of lubricant oil when
contact occurred in the mixed lubrication regime.
On the other hand, the first attempts for artificial
textured piston rings on internal combustion engines, are
reported in [5-7]. Etsion and Sher [5] presented an
experimental work to estimate the effects of partially
laser surface textured piston rings on the fuel
consumption and on the exhaust gas composition of the
ignition on the compression stage of an internal
combustion engine. It was obtained that the partial
surface texturing piston rings lead to more than 4% lower
fuel consumption. Furthermore, Ryk and Etsion in [6]
presented another experimental work regarding the effect
of partial laser surface texturing on the friction reduction
of the piston rings. In their results shown that, the partial
laser textured piston rings improves the friction by 25%,
than the relevant plain rings. Mezghani et al. [7]
proposed a numerical simulation for the investigation of
the effects of the groove characteristics on the lubrication
performance, and on friction at the boundary between the
piston ring and cylinder liner. In their study, the lubricant
film rapture and the cavitation are taken into account.
The present paper is divided in two parts. In the first
part the effects of both types of oils (Newtonian and with
additives) on the tribological characteristics of a smooth
piston ring are investigated whereas in the second part
the consequences on the textured ring are also examined.
The Computational fluid dynamics (CFD) analysis is
used in order to solve the Navier Stokes equations,
calculating thus the pressure field between the piston and
the ring for fully flooded lubrication. Further the
non-Newtonian behavior of the oil is simulated using
power law models. Simulation results from a diesel
engine are presented for several parameters including
friction forces and hydrodynamic pressures.
2. Geometrical analysis
2.1. Smooth and textured piston ring model
The basic dimensions of a piston ring model presented
in Figure 1a. The piston ring moving wall generates
shear forces, exerting motion to the fluid, which flows
from the piston ring inlet to the outlet. As the fluid
convects, it builds up pressure, which exerts forces on
the piston ring wall. Pressure buildup is due to the
converging geometry, as a result of the ring deformation
due to the chamber pressure, and also due to the
presence of dimples of certain geometry. Energy is
expended by the shear forces work done at the moving
wall-fluid interface.
Figure 1(a) Basic dimensions of piston ring model (b)
Spherical geometry of dimples
In Table 1, the input numerical parameters are
depicted. The parameter, named B, is the thickness of
the ring, tr is the length that is inside of the piston
groove, tring is the free ring length inside the lubricant, tg
is the distance from top of the first groove, tland is the
lands between the ring grooves, h is the oil thickness, p1
is the combustion chamber gas pressure, L is the stroke
length, rcr is the crank radius and lrod is the length of the
rod.
Table 1 Input numerical parameters
L =120mm
h=8μm
tring=1.5mm
tr=3mm
p1=6.64MPa
lrod=296mm
B=2.4mm
Ω=2400rpm
tg=5.98mm
Mass conservation equation,
V  0
(1)
Momentum equations,
V
1

(2)
 V  V   p   2V
t


where V is the velocity vector, p is the pressure
gradient,  is the fluid density, μ is the absolute
viscosity and t is time. Equations (1) and (2) are solved
with the CFD code ANSYS Multiphysics.
3.1. Viscosity of non-Newtonian fluids
λcr=0.27: control
ratio(rcr/lrod)
rcr=80mm
tland=2.4mm
For incompressible Newtonian fluids, the shear
stress is proportional [8] to the rate-of-deformation
tensor D :
  D
(3)
In Figure 1b, the geometry of a spherical micro
dimple texturing is presented. Each spherical dimple has
a base radius rp, the dimple depth Hd and the textured
zone Lc is bounded with two untextured strips of width
where D is defined by
Lut  Lud 
and μ is the viscosity, which is independent of D. For
some non-Newtonian fluids, the shear stress can
similarly be written in terms of a non-Newtonian
viscosity μα:
B  N * Lc
on each of its sides. The
2
below non dimensional parameters can be defined here:
 the number of dimples, N
 the textured portion,  
(4)

  a D D
Lc
B
 the dimensionless dimple diameter, 
 u
u 
D j  i 
 x x 
j 
 i
(5)
However, in the non-Newtonian models the shear rate

2rp
rcr
H
 the dimple depth over diameter ratio,   d
2rp
It should be noted that, for rp=40μm, Hd=12μm and
Lc=90μm, the values of the above non dimensional
parameters take the values, N=16, γ=0.0375, δ=0.001
and ε=0.15.
3. Assumptions and Governing Equations
This work is based on Navier Stokes equations
solution using computational fluid dynamics (CFD).
The main assumptions employed are given below.
1. The cylinder is stationary.
2. The flow is considered isothermal.
3. The piston ring is secured in the groove of
the piston
4. The piston moves with a linear velocity in
y axis, defined by the equation (9).
5. The oil temperature is considered constant
along the cylinder wall.
6. The friction force between the piston ring
and ring groove is also considered
negligible.
The
conservation
equations
for
unsteady
incompressible and isothermal flow, with zero gravity
and other external body forces, are:
 is defined as :  
1
D:D
2
(6)
3.2. Power law for Non-Newtonian Viscosity
Non-Newtonian flow will be modelled according to
the following power law model, taking into account the
non-Newtonian viscosity μα:
a   k o n1
where μο, k , n and
(7)
o
are input parameters. In detail,
μο is the nominal viscosity at 20 oC, k is a value of the
average viscosity of the fluid, n is a measure of the
deviation of the fluid from the Newtonian behaviour
(the power-law index) and
0
is the cutoff shear rate.
3.3. Monograde and Multigrade Lubricant properties
Three kinds of monograde lubricant and two kinds
of multigrade lubricant are used. In Table 2 the lubricant
properties are depicted.
Table 2 Lubricant properties
SAE viscosity grade
Viscosity at 80 oC(mPas)
20
9.8
30
14
50
25
5W30
12.97
10W40
20.08
2
In addition, Table 3 shows the power law numerical
parameters of multigrade lubricant oils.
Table 3 Power law parameters
Power law n
k
 (s-1)
o
5W30
10W40
0.89
0.91
0.8
0.78
2e+6
2e+6
μο(mPas)
80
95
5. Fluid Structure Interaction model-analysis
The fluid structure interaction analysis presented in
this paragraph. Figure 4, illustrates the flow chart of the
solution of the couple field problems and the
convergence criterion used. Additionally, the Arbitrary
Lagrangian Eulerian (ALE) technique is used for the
grid deformation (mesh morphing), of the fluid and
structural problem.
4. Boundary conditions
A four stroke diesel engine is considered here. The
boundary conditions and the profile of combustion gas
pressure p1=pc that defines the 2-D problem are shown
in Figure 3. In fact, a limited loss of combustion gases
because of the existence of the ring gaps through which
the gas leaks were not considered to this stage.
Figure 4 Solution flow chart
5.1. Meshing formulation
Figure 3 Boundary conditions of the piston ring system

Section 1 : Vx=0,Vy=V,Vz=0

Section 2 : Vx=0,Vy=V,Vz=0

Section 3 : Vx=0,Vy=0,Vz=0
4.1. Integral quantities
The pressure distribution over the piston ring is obtained,
after the numerical solution of the equation (2). In the
present model, the friction force on the piston ring
Regarding the fluid field solution, tetrahedral
elements are used. Their size is on the scale of 3e-6, in
the area near to the piston ring profile. The total number
of 107080 elements for smooth case and 115053
elements for artificial textured which seemed adequate
for the certain calculations are used. The particular
number of elements and nodes were obtained for each
examined case, after extensively grid sensitivity tests,
while the pressure error considered as, Errpres  1E-6.
Specifically, simulations were performed on the
computer with 8 processors (Intel core i7-3770 CPU@
3.40 GHz) and typical simulation time varied between
20 and 30 minutes for each crank angle.
calculated by integrating the stress tensor  above the
ring surface A interfaced with the fluid film, and it is
given by the below relationship:
B
Ff    dA   Dring  2B  dy

A
(8)
2
Dring is the piston ring diameter and B is the piston ring
thickness. The velocity of piston is calculated according
to the next equation:
V  rcr (sin  
cr
2
sin 2
(9)
The density-temperature relationship is:


1  
Figure 5 Detail of fluid mesh
5.2. Validation results
The present model is validated against published
work of Wakuri et al. [3]. In particular, the piston ring
friction force was calculated for fully flooded
lubrication, illustrating a very good agreement (see
figure 6).
(10)
where ρο is the fluid density at 20 oC (ρο=850 kg/m3) and
a is the thermal expansion of the material (α=10.8 10-6
o -1
C ).
3
Figure 6 Validation results with ref. [3] (see figure 7).
Friction force versus crank angle
Figure 9 Maximum friction force vs crankshaft speed
6. Numerical Results and Discussion
7. Conclusions
Figures 7a-b shows the friction force Ff in a
smooth piston ring for Ω=2400rpm. It is depicted that
the friction force increased when the viscosity increases,
either for Newtonian lubricant or for lubricant with
additives. In consequence, by using Newtonian oil the
increment of the friction force is 50.9% while in the
other case the increment is 44.09%.
Conclusions are summarized as follows:
1.
2.
3.
The friction force decreased having an artificial
textured ring and lubricant with additives.
The friction force decreased with the use of the
lubricants with additives against to Newtonians,
either for smooth ring or artificial textured ring.
Increasing the viscosity, the increment of friction
is obvious.
Acknowledgement
The Authors acknowledge the financial support by the
Greek Secretariat for Research and Technology through
the TRIBO-MARINE project, grant Nr. 2595.
8. References
[1]
Figure 7 Friction force vs crank angle for a. Newtonian
oils and b. Lubricant with additives
In Figure 8, the variation of the friction force in a
carbon steel piston ring with spherical texturing, with
SAE 30 and SAE 5W30 lubricants, is presented. The
case of, N=16, γ=0.0375, δ=0.001 and e=0.15 examined
in detail. In the point “A”, the maximum friction force
decreased while the chamber pressure is maximum. In
this case the reduction is 14.6%.
[2]
[3]
[4]
[5]
[6]
Figure 8 Friction force vs crank angle
[7]
The maximum friction force is illustrated in Figure 9,
as a function of the rotational crankshaft velocity. The
simulation results were performed for SAE 30 and SAE
5W30 oils, either for smooth ring or textured ring. From
the results, it is obvious that using lubricant with
additives and artificial textured ring, the friction
reduction is 32.08%.
[8]
Meunier, C., Mazuyer, D.,Vergne, P., Fassi, M El.,
Obiols, J., “Correlation between the Film Forming
Ability and Rheological Properties of New and
Aged Low Sulfated Ash, Phosphorus and Sulfur
(Low SAPS) Automotive Lubricants”, Tribology
Transactions, 52:4, 2009,501-510.
Rajalingam, C., Rao B.V.A., Prabhu B.S., “The
effect of a non-Newtonian lubricant on piston ring
lubrication”, Wear, 50,1978,47-57
Wakuri, Y., Hamatake, T., Soejima, M., and
Kitahara, T., “Piston ring friction in internal
combustion engines”, Trib. Int., 25, 5, 1992, 299.
Hamatake, T., Wakuri, Y., Soejima, M., and
Kitahara, T., “Effects of lubricant viscosity on the
mixed lubrication of a piston ring pack in an
internal combustion engine”, Lubrication Science,
15, 2, 2003, 101-117.
Etsion , I., Sher E., “Improving fuel efficiency
with laser surface textured piston rings’’,
Tribology International, 42:4, 2009,542-547.
Ryk, G., Etsion, I., “Testing piston rings with
partial laser surface texturing for friction
reduction”, Wear, 261, 7-8, 2006,792-796.
Mezghani, S., Demirci, I., Zahouani, H., Mansori,
M El., “The effect of groove texture patterns on
piston-ring pack friction”, Precision Engineering,
36:2, 2012, 210–217.
Gertzos,
K.P.,
Nikolakopoulos,
P.G.,
Papadopoulos, C.A, “CFD analysis of journal
bearing hydrodynamic lubrication by Bingham
lubricant”, Trib. Int., 41, 2008, 1190-1204.
4