04-12-03-0010-A Multiscale Cylinder Bore Honing Pattern Lubrication Model For Improved Engine Friction PDF

Downloaded from SAE International by Columbia Univ, Tuesday, April 14, 2020
ARTICLE INFO
Article ID: 04-12-03-0010
Copyright © 2019
SAE International
doi:10.4271/04-12-03-0010
A Multiscale Cylinder Bore Honing

Pattern Lubrication Model for
Improved Engine Friction
Osman Taha Sen and Ozgen Akalin, Istanbul Technical University, Turkey
Abstract History
Received: 05 Apr 2019
Three-dimensional patterns representing crosshatched plateau-honed cylinder bores based on Revised: 08 Jun 2019
two-dimensional Fast Fourier Transform (FFT) of measured surfaces were generated and used to Accepted: 18 Jun 2019
calculate pressure flow, shear-driven flow, and shear stress factors. Later, the flow and shear stress e-Available: 02 Jul 2019
factors obtained by numerical simulations for various surface patterns were used to calculate lubri-
cant film thickness and friction force between piston ring and cylinder bore contact in typical diesel Keywords
engine conditions using a mixed lubrication model. The effects of various crosshatch honing angles, Cylinder liner, Honing,
such as 30°, 45°, and 60°, and texture heights on engine friction losses, wear, and oil consumption Mixed lubrication,
were discussed in detail. It is observed from numerical results that lower lubricant film thickness Flow factors, Friction
values are generated with higher honing angles, particularly in mixed lubrication regime where
lubricant film thickness is close to the roughness level, mainly due to lower resistance to pressure Citation
flow. Although, shear stress values are lower for higher honing angles, significant friction force Sen, O. and Akalin, O.,
observed in the expansion stroke with high honing angles is primarily due to metal-to-metal contact “A Multiscale Cylinder Bore
and increased viscous shear as a result of lower film thicknesses. It is observed that average asperity Honing Pattern Lubrication
contact pressures double with each 15° increase in the crosshatch angle indicating high wear particu- Model for Improved Engine
larly in the ring reversal zones. The results showed that transversal plateau-honing patterns generate Friction,” SAE Int. J. Fuels
resistance to fluid flow and enhance full film hydrodynamic lubrication, reducing friction and asperity Lubr. 12(3):155–166, 2019,
contact. However, oil film transported to the combustion chamber also increases with transversal doi:10.4271/04-12-03-0010.
patterns in the complete engine stroke that may result in increased oil consumption.
ISSN: 1946-3952
e-ISSN: 1946-3960
155
156 Sen and Akalin / SAE Int. J. Fuels Lubr. / Volume 12, Issue 3, 2019
1. Introduction the honed cylinder crosshatched pattern are very scarce in the
literature. Modeling surface textures in macroscale requires
T
extensive computational effort, and it is not practical from the
oday, engines are being downsized and firing pressures engineering point of view. Michail and Barber [12] investi-
are increased using direct injection and turbocharging gated the effects of cylinder wall topology on oil film thickness
to meet future fuel consumption and emission targets. using flow factors generated by using skewed roughness distri-
However, this increases thermal and mechanical stresses on butions having different Peklenik factors and reported that
engine components in relative motion such as piston rings the use of small crosshatch angles when honing the cylinder
and cylinder bores. Friction between power cylinder compo- wall increases the hydrodynamic action. Although, Patir and
nents is responsible for a significant part of total power loss Cheng’s flow factors can be used to simulate patterns with
in internal combustion engines [1]. In a typical piston ring different orientations, plateau-honed surfaces cannot
and cylinder bore contact, smooth surfaces enhance hydro- be accounted well with Gaussian distributions. In order to
dynamic lubrication with very thin oil film thickness and overcome this difficulty, several researchers used artificially
reduce metal-to-metal friction, wear, and oil consumption. generated or measured surfaces to calculate flow factors to
However, smooth cylinder bore surfaces are prone to scuffing include the effect of non-Gaussian textures into piston ring
due to starvation that may result in complete engine failure. lubrication modeling [13, 14]. Jocsak et al. [15] artificially
Oil retaining and heat dissipating honing patterns are neces- generated textured surfaces with different honing angles and
sary to maintain proper lubrication. Also, honing grooves investigated the frictional losses for three types of engines.
retain wear debris and eliminate abrasive action, particularly They concluded that decreasing honing angle reduces friction
in the run-in stage. Usually, this is achieved by crosshatches in mixed lubrication; however, this may result in increased oil
on the cylinder bore generated by multiple plateau-honing transport to the combustion chamber since the lubricant film
processes. In the first phase, deep valleys are created on the thickness is increased in the complete stroke. Later, this
cylinder bore using coarse stones; the smooth plateaus are approach was extended for twin-land oil control rings using
generated in the following steps applying lower pressure to artificial surfaces having additional sinusoidal wave patterns
the surface with finer abrasives. Deep grooves on the cylinder on the plateaus [16]. Mezghani et al. [17] used generated three-
bore act as oil reservoirs supplying fresh lubricant to the inter- dimensional honing patterns with different crosshatch angles
face, while smooth tops reduce friction and wear eliminating to study frictional behavior of cylinder and ring contact in
the probability of metal-to-metal contact. Typically, 40°-55° hydrodynamic lubrication. They pointed out that 40°-55° and
crosshatch angles have become industry standard over the 115°-130° honing angles, which give the minimum friction
years. In a relatively new technique called slide honing, which properties, are common crosshatch angles in plateau and
is a type of plateau honing, special diamond abrasives are used helical slide honing techniques, respectively. A detailed exper-
to produce much finer and consistent surfaces having similar imental study was performed by Grabon et al. [18] to investi-
honing angles [2]. With the development of advanced piston gate the effects of honed cylinder liner surface texture on
ring coatings such as diamond-like carbon and physical vapor tribological properties of piston ring-liner assembly using an
deposition (PVD) techniques, probability of scuffing can oscillating wear tester, and it is reported that cylinder bore
be reduced with much thinner lubricant film thickness values, samples having crosshatch angles higher than 55° show
giving more emphasis on engine friction and oil consumption. increased coefficient of friction. Although negatively skewed
In the helical slide honing method proposed by Hoen et al. honed surface textures are important for fluid flow in piston
[3], crosshatch angle is increased drastically to 140°-150° and ring and cylinder bore contact, deep valleys have very little
improved wear in reversal zones is reported. Alternative contribution to load carrying. Directional roughness peaks
methods such as laser texturing has significant potential to on the plateaus are removed after the run-in period and most
reduce friction, wear, and oil consumption. However, further of the load is carried by surface asperities. Engine cylinder
research is needed to understand lubrication mechanisms in bore surface roughness and texture have usually different
textured surfaces [4, 5]. topologies, and modeling them in different scales gives
Patir and Cheng’s [6, 7] modified Reynolds equation with freedom to better characterization of the surface features.
flow factors has been widely accepted in the simulation of Power cylinder components must be designed for an
bearings with surface roughness for decades. They obtained optimum tribological solution since engine friction, wear, and
pressure and shear flow factors using numerical simulations oil consumption is closely linked. In this study, a multiscale
for rough surfaces having Gaussian distributions with approach is proposed to simulate piston ring and cylinder
different orientations comparing the flow in a rough bearing bore lubrication. A Gaussian contact model is used to calculate
to that of a smooth bearing. Here, surface roughness orienta- the load carried by the plateau asperities in roughness scale,
tion is defined as Peklenik number which is the ratio of lengths flow factors generated based on measured cylinder bore
where autocorrelation factor is reduced to its half of the initial patterns are used to model lubricant flow in texture scale, and
value in longitudinal direction with that of in transversal finally nominal piston ring and cylinder bore geometry is used
direction. This approach was widely adopted for piston ring in the macroscale to calculate hydrodynamic pressures. Since
and cylinder bore lubrication simulations along with common surface roughness parameters cannot represent the
Greenwood and Tripp’s asperity contact model [8, 9, 10, 11]. honed surface texture, the surface shear and flow factors were
Piston ring and cylinder bore friction simulations considering
© 2019 SAE International. All Rights Reserved
Sen and Akalin / SAE Int. J. Fuels Lubr. / Volume 12, Issue 3, 2019 157
developed using representative control volumes generated FIGURE 1 Piston ring force balance.
based on two-dimensional Fast Fourier Transform (FFT) of
measured cylinder bore profiles. Friction losses are calculated
for various honing geometries.
2. Theory and Modeling

2.1. Piston Ring and Cylinder
Bore Friction Model
Neglecting ring tilt, ring twist, ring end flows, and bore distor-
tions, Patir and Cheng’s [6, 7] modified averaged Reynolds

equation can be written in the form of Equation 1. The terms
on the left-hand side represent the pressure (Poiseuille) flows
in longitudinal and circumferential directions, the first term
on the right-hand side of the equation represents the shear
(Couette) flow in a rough contact, and the final term charac-
terizes the flow due to squeeze film effect:
¶ æ ¶ æ 3 ¶p ö æ ¶hT ¶f ö
3 ¶p ö ¶hT
ç fx h ÷ + ç fy h ÷ = 6 mU p çç +s s ÷÷ + 12 m
¶x è ¶x ø ¶x è ¶y ø è ¶x ¶x ø ¶t
Eq. (1)
where ϕx and ϕy are pressure flow factors and ϕs is shear flow ¶p

factor; h and hT are nominal and roughness averaged film conditions are valid, assuming = 0 at the trailing zone.
¶x
thicknesses, respectively; μ is dynamic viscosity; and p is the Starvation and film reformation in the trailing side is not
local lubricant pressure. For an axisymmetrical analysis, net considered for simplicity. With the given cylinder and second
flow in the circumferential direction would be zero. However, land pressures, hydrodynamic fluid film pressure, asperity
the effect of transversal flow in the texture scale is still contact pressure, squeeze film velocity, and rupture locations
¶h satisfying the piston ring force balance can be calculated for
retained in the flow factors. Contact factor fc = T was intro- each crank angle position using a finite difference scheme [11].
¶h
duced by Wu and Zheng [19] to include the probability that Using the calculated lubricant film thickness values and
the surfaces are really lubricated making the averaged pressure profiles for each crank position, the shear stress
Reynolds equation suitable for mixed lubrication analysis. acting on the piston ring can be calculated including shear
Any piston ring profile can be represented by a set of polyno- stress factors ϕfp and ϕfs that were obtained by a similar method
mials or measured profile data along the ring thickness. It is as for pressure and shear flow factors:
assumed that force due to gas pressure acting behind the ring,
combustion chamber or second land pressure, whichever is mU p h æ ¶p ö
higher, and the ring tension are balanced by hydrodynamic t =
h
(f f ± f fs ) ± f fp ç ÷
2 è ¶x ø
Eq. (2)
and asperity contact pressures generated between the ring and
bore contact as shown in Figure 1. Greenwood and Tripp’s [8]
contact model is used to calculate pasp, the pressure generated where the cylinder bore is stationary and the piston ring is
by the contacting surface asperities. The applied load on the moving with a speed of Up. The plus sign is used to calculate
piston ring is mainly carried by the roughness on the flat tops the shear stress acting on the moving piston ring where the
of the cylinder bore surface, and deep grooves do not have minus sign is used to calculate the shear stress acting on the
significant effect on the load-carrying capacity. Also, direc- cylinder bore. When the film thickness between roughness
tional patterns on the plateaus are removed by wear, and peaks approaches zero, Newton’s law generates infinite shear
unlike cylinder surfaces, piston rings do not have directional stress. In order to eliminate this problem, it is assumed that
patterns. Therefore, roughness pattern of the flat tops and hydrodynamic shear stress is not generated where the film
piston ring face is assumed to be Gaussian. Thus, contact thickness is as low as σ/100, and boundary friction prevails.
parameters given by Hu et al. [10] are used for the analysis. It The ϕf term is used to include this effect, and the empirical
is also assumed that sufficient lubricant is always available at function generated for Gaussian surfaces by Patir and Cheng
the inlet of the piston ring, and the lubricant film ruptures [7] is used in the simulations. The ϕfs term is used to consider
from the ring surface where Reynolds (Swift–Stieber) boundary the effect of roughness on the shear stress acting on the
moving piston ring. Since the stationary cylinder bore has a FIGURE 2 Measured plateau-honed cylinder surface.
rough honed pattern and the moving piston ring is relatively
smooth, additional shear stress acts on the piston ring due to
the cylinder bore texture. The ϕfp term in the last part of

Equation 2 represents the shear flow acting on the piston ring
due to pressure flow. Since this effect is a few orders of magni-
tude smaller than the other effects, it is neglected in the engine
friction calculations. Even for nominally flat surfaces, local
pressure generates a net force on the sides of the piston ring
asperities. Since the piston ring roughness is very small
compared to that of the cylinder bore surface, this effect is
also small compared to shear force acting on the nominal
geometry. However, this effect would be significant for
textured piston rings or piston skirts. The force acting on
piston ring segment due to pressure acting on the sides of the
asperities is given by
matrix was found using a MATLAB function, and crosshatch
2 mU p æ s p ö
2
orientation and spacing of the control domain was selected
dFrsh = ç ÷ f fs dx Eq. (3) using this information. Measured surface profile and magni-
h ès ø
tude of its two-dimensional complex FFT in spatial frequency
domain are shown in Figures 2 and 3, respectively.
where σp is piston ring (root mean square, or rms) roughness The textured control volume surface was generated using
and σ is the composite (rms) surface texture. Another signifi- a polynomial with a similar approach as in Spencer et al. and
cant force arises due to local pressure acting on the nominal Jocsak et al. [13, 15]. An equation that represents the cross-
geometry, and this can be calculated integrating hatched pattern closely was found as
the local pressure over the piston ring profile. Integrating the
viscous shear including the effect of rough surface, local
pressure acting on the sides of the asperities, local pressure hPH ( x ,y ) = 10-w ×s ( x ,y ) × s ( x ,y )
acting on the nominal piston ring geometry, and asperity where Eq. (5)
contact pressure over the piston ring and cylinder bore contact ( x ,y ) = cos éë2p ( x + ky )ùû - 1
area, total friction force acting on the piston ring can
be calculated as
Here, x and y designate the transversal and the longitu-
dinal directions, and ω is a parameter used to adjust the
mU p é ù
2
æsp ö width of the honing grooves. Parameter k is “1/tan(α/2)”
Ftotal = òò ê(ff + ffs ) + 2 ç ÷ ffs ú dxdy
A
h ê
ë ès ø úû where α is the crosshatch angle as shown in Figure 4. The
dhs control domain was generated representing a 2-mm wide
+ òp d xdy + ò mf · p asp dxdy Eq. (4) surface area (L x ×Ly) on the cylinder bore. The number of
A
dx A
where μf is metal-to-metal friction coefficient between piston FIGURE 3 FFT of the measured cylinder surface.
ring and cylinder bore asperities.
2.2. Artificial Plateau-Honed
Surface Generation
In order to calculate flow and shear stress factors, a control
domain representing the crosshatched cylinder bore surface

was generated based on the measured cylinder bore surface.
Plateau-honed cylinder bore surface of a heavy-duty diesel
engine was measured by using an optical profiler. Cylindrical
form of the area was removed and the flattened data was stored
as a 3.3 × 3.8 μm pixel size matrix. Various surface textures
with different wavelengths and orientations may have similar
surface roughness values. Therefore, common surface rough-
ness parameters alone cannot represent the actual surface
pattern. Two-dimensional FFT of the cylinder bore texture
FIGURE 4 Representative cylinder surface. FIGURE 6 Artificially generated honed surface.


nodes in the transversal direction m was selected as 1000,
and the number of elements in the axial direction n was
selected as “m·tan(α/2)” to capture the full crosshatch pattern
regardless of the honing angle.
Areal FFT spectrum of the surface profile reveals the true 2.3. Calculation of Flow and
lay of the crosshatch pattern where crosshatch angle α is
2×atan(fx/fy), where fx and fy are the coordinates of the spatial
Shear Stress Factors
frequency peaks. The spacing of the crosshatch marks is selected In order to calculate the flow and shear factors, the flow char-
such that measured and generated surfaces have similar first acteristics between textured and smooth domains under
harmonic peaks in the spatial frequency domain. The width certain boundary conditions are compared. In the current
parameter ω is selected such that measured and generated study, the textured domain consists of a honed surface with
surfaces have the same (rms) roughness values. Artificially a given honing angle and a smooth surface with no texture as
generated surface profile and magnitude of its two-dimensional shown in Figure 7. Pressure distributions over these textured
complex FFT in spatial frequency domain are shown in and smooth domains are numerically solved, and the flow
Figure 4 and Figure 5. Also, three-dimensional view of the factors required for the solution of the averaged Reynolds
generated pattern is shown in Figure 6. Various crosshatch equation are obtained.
angles are generated using different crosshatch angles “α” using As defined by Patir and Cheng, the pressure flow factor
Equation 5. It is observed that changing the crosshatch angle ϕx is evaluated from the pressure-driven flow boundary condi-
value in the generated patterns does not influence the rough- tions. These conditions are defined as zero relative sliding
ness-to-material ratio relationship (Abbott-Firestone curve). motion between the surfaces of the domain with a pressure
FIGURE 7 Control domain used to obtain flow and shear

FIGURE 5 FFT of the representative cylinder surface. stress factors.

gradient in the direction of the flow. Hence, the pressure flow FIGURE 8 Modeling steps.
factor is the ratio of the flow rates of textured and smooth
surfaces and defined as
1 Lyæ hT3 ¶p ö
Ly ò 0
ç- ÷ dy
è 12 m ¶x ø
fx = Eq. (6)
hs3 dP

-
12 m dx
where hT is the local film thickness in the textured domain

and h s is the corresponding film thickness of the smooth
domain. Here, it should be noted that hs is evaluated so that
the textured and smooth domains have the same volumes.
Mathematically, it is expressed as
Lx Ly
ò ò hT ( x ,y ) dydx
hs = 0 0
Eq. (7)
Lx L y
where Lx and Ly are the dimensions of the domains in the hs Lx æ

Ly ¶p ö
corresponding directions. Similar to pressure flow factor ϕx, L x Ly ò ò
0 0
ç -hT
è ¶x ø
÷ dydx
the pressure stress factor ϕfp is also calculated from the Ffs ,i = Eq. (12)
solution of the pressure distribution under pressure-driven 2 m (U 1 - U 2 )
flow boundary conditions:
where U1 and U2 represent the linear velocities of the mating
1 æ h ¶p ö
Ly
surfaces. A flowchart summarizing the modeling steps is
Ly ò 0 çè - 12Tm ¶x ÷ødy shown in Figure 8.
ffp = Eq. (8)
h dP
- s
12 m dx
3. Results and Discussion
For the shear flow and shear stress factors, the pressure
distributions are solved with shear-driven flow boundary
conditions, where the pressure gradient is set to zero, but there 3.1. Flow and Shear Stress
is a constant relative velocity (U) between mating surfaces of Factors for Plateau-Honed
the domain. Hence, the pressure generation of the fluid is only
possible due to texture of surfaces. Thus, the shear and stress
Surface
flow factors are defined as Obtained pressure and shear stress factors as a function of
lubricant film thickness corresponding to a smooth surface
2 2 with respect to rms averaged values for 30°, 45°, and 60°
æs ö æs ö
fs = ç 1 ÷ Fs 1 - ç 2 ÷ Fs 2 Eq. (9) crosshatch angles are shown in Figures 9, 10, 11, and 12. Flow
ès ø ès ø and shear stress factors obtained by Patir and Cheng for
rough surfaces with different Peklenik factors are also given
æs ö æs ö
2 2
for comparison. Although its effect on total friction is rela-
ffs = ç 1 ÷ Ffs 1 - ç 2 ÷ Ffs 2 Eq. (10) tively small, piston ring’s surface roughness is assumed to
ès ø ès ø
be isotropic and has Gaussian distribution. Figure 9 shows
pressure flow factors for various honing angles. Transversal
respectively, where σ1 and σ2 are the standard deviations of honing patterns restrict the pressure flow, so flow factors
the surfaces and σ is composite rms value of these surfaces. lower than unity is observed. However, calculated flow
Furthermore, the terms Φs,i and Φ fs,i are given as factors are higher than unity where the film thickness is low,
since deep grooves enable additional lubricant flow. Shear
2 Lx æ hT3 ¶p ö
Ly flow factors for different honing orientations are given in
L x Ly ò 0 ò 0 çè - 12m ¶x ÷ødydx Figure 10. If the two surfaces moving relative to each other
Fs ,i = Eq. (11) have the same roughness, the resulting net flow is the same
(U 1 - U 2 )s i as in smooth surfaces. If the rough surface is moving, this
FIGURE 9 Pressure flow factors. FIGURE 11 Pressure-driven shear stress factor.
© 2019 SAE International. All Rights Reserved 3.2. Predicted Piston Ring Oil
will result in additional flow due to roughness. However, if
the smooth surface is moving and rough surface is stationary,
as in piston ring and cylinder bore contact, rough surface Film Thickness and
generates resistance to fluid flow. It is observed that trans- Friction
versal patterns generate higher resistance to fluid flow. Figure
Neglecting ring tilt, twist, starvation, and reformation, simu-
11 shows shear stress acting on the surfaces due to pressure
lations were performed for the top compression ring of a six-
flow. All surfaces experience lower shear stress due to rough-
cylinder, turbocharged, heavy-duty diesel engine having a
ness. Transversally oriented surfaces generate less shear due
symmetrical barrel ring profile. The engine specifications and
to pressure flow, and this effect diminishes when the lubri-
simulation input parameters are given in Table 1. Cylinder
cant film thickness is high. Shear stress factor is calculated
and second land pressures as a function of crank angle position
using a similar method as in shear flow factor. Figure 12
at full load – 1680 rpm engine speed are shown in Figure 13.
shows shear stress due to speed gradient. If the rough surface
Simulations were performed for two different surface textures,
is stationary and the smooth surface is moving, the shear
Rz (peak-to-peak texture height) 1.84 μm and 2.57 μm for 30°,
stress acting on the moving surface is increased due to
45°, and 60° crosshatch angles. Only peak-to-peak texture
roughness effect.
FIGURE 10 Shear flow factor. FIGURE 12 Shear-driven shear stress factor.

TABLE 1 Engine specifications and global input parameters. FIGURE 14 Hydrodynamic fluid film pressure.
Definition Value
Bore 115 mm

Stroke 144 mm
Ring thickness (B) 3.50 mm
Effective ring thickness (b) 2.70 mm
Ring surface rms roughness 0.1 μm
Piston ring tension 50 N
Oil viscosity 15W40 (at 100°C) 0.0119 Pa⋅s

Engine speed 1680 rpm
Modulus of elasticity (composite) 113.79 GPa
Boundary friction coefficient 0.25
profile is considered here. Therefore, generated R z values are

significantly smaller than peak-to-peak roughness of a typical
cylinder bore. Generally, peak-to-peak roughness of a cylinder
bore is five to six times higher than its rms roughness value.
The effect of surface roughness is considered only in the calcu-
lation of asperity contact pressure, and the resistance to fluid
flow due to roughness on the flat plateaus is neglected. roughness values. Total viscous shear force acting on the
Hydrodynamic fluid film pressure along the ring thick- piston ring is shown in Figure 17. Although, stationary trans-
ness is shown in Figure 14 with respect to crank angle position. versal honing patterns increase shear stresses acting on the
It is observed that transversal honing patterns slightly increase relatively smooth piston ring, reduced lubricant film thickness
lubricant film thickness throughout the complete stroke in longitudinal patterns dominates the shear stress and results
which is mainly due to resistance to fluid flow; the effect of in higher viscous shear with high crosshatch angles. Friction
honing angle is only significant in the ring reversal zones, force resulting from the local pressure acting on the sides of
where the film thickness approaches to roughness scale as the asperities is also shown in Figure 18. Although this effect
shown in Figure 15. In addition, change of lubricant film is relatively small compared to viscous effects, higher net force
thickness as a function of crank angle position for a texture is observed with transversal honing marks. Net force due to
with higher Rz value is shown in Figure 16. Although the same hydrodynamic pressure acting on the nominal piston ring
order is seen, the effect of honing angle on lubricant film profile is calculated separately. This effect results in a signifi-
thickness is more prominent for textures with higher cant force particularly in the expansion stroke, where the
FIGURE 15 Predicted minimum lubricant film thickness

FIGURE 13 Cylinder and second land pressure used (Rz = 1.84 μm, Rq = 0.48 μm).
as input.

FIGURE 16 Predicted minimum lubricant film thickness FIGURE 18 Friction due to local pressure (Rz = 1.84 μm,
(Rz = 2.57 μm, Rq = 0.70 μm). Rq = 0.48 μm).

pressure difference between the combustion chamber and three different honing angles. It is observed that high honing
second land is at its highest; however, the effect of crosshatch angles increase friction losses, and this effect is more signifi-
pattern is marginal. Total net force acting on the piston ring cant if the overall surface roughness is high. Since high
is shown in Figure 19. It is observed that in most of the stroke, friction force is developed in the expansion stroke near top
the piston ring load is supported by the hydrodynamic fluid dead center, where the piston speed is low, the effect of honing
film pressure except near top dead center in the expansion angle on engine friction losses is relatively low compared to
stroke. The effect of honing patterns on total friction force is the effect on asperity contact pressures. Total indicated mean
very small, where the lubricant film thickness is high. effective pressure for the simulated case was 24.97 bar. Figure 21
However, significant friction force is observed in the expan- shows the effect of honing angle on average contact pressure.
sion stroke with high honing angles, which is mainly due to Where the piston ring load is mainly carried by the hydrody-
metal-to-metal contact and high viscous shear. Figure 20 namic fluid film pressure in the cylinder bore with 30° cross-
shows predicted first compression ring friction mean effective hatch angle, significant metal-to-metal contact is observed
pressure values for two different surface texture values and with 45° and 60° crosshatch angles that may result in top ring
FIGURE 17 Viscous shear friction (Rz = 1.84 μm, FIGURE 19 Total friction force (Rz = 1.84 μm,
Rq = 0.48 μm). Rq = 0.48 μm).

FIGURE 20 First compression ring friction mean effective FIGURE 22 Average downstroke film thickness.
pressures at full load 1680 rpm for different honing angles and
roughness (imep = 24.97 bar).

reversal zone wear. This effect is more obvious in high surface © 2019 SAE International. All Rights Reserved Figure 22 shows average downstroke oil film thicknesses for
texture values. It is observed that average asperity contact various honing angles and roughness. For both high and low
pressures nearly double with each 15° increase in the cross- surface texture values, average downstroke lubricant film
hatch angle when the crosshatch angle increased from 30° to thickness decreases with increasing honing angle. Therefore,
45°. Since wear in hydrodynamic lubrication regime is negli- lower oil consumption is expected with surfaces with low
gible, cylinder bore wear profiles can be easily calculated with surface roughness having longitudinal textures. On the other
given asperity contact pressure and material properties. hand, as a drawback, higher wear in the top ring reversal zones
Increased asperity contact pressure with high texture heights is expected with higher honing angles.
and crosshatch angles indicate high material removal near Piston rings experience a wide range of lubrication
ring reversal zones. Another important criterion in power regimes in operation. High asperity contact near the reversal
cylinder tribological optimization is oil consumption. zones results in significant cylinder bore wear. However, the
Although there are several oil consumption models in the effect of metal-to-metal friction on engine’s total power loss
literature, there is no validated oil consumption model avail- is relatively low due to low piston speed in this range.
able considering roughness and oil retention volumes of the Transversal surface textures increase lubricant film thickness
honed surfaces [20]. However, oil film thickness remaining in mixed lubrication. High lubricant film thickness in the mid
on the cylinder walls gives a good indication to oil consump- stroke increases the amount of lubricant transported to the
tion since most part of the oil consumption is caused by combustion chamber and results in increased oil consump-
evaporation. Average downstroke oil film thickness is defined tion. High oil consumption increases hydrocarbon emissions
as the film thickness at the trailing side of the piston ring and service cost, poisons catalytic converters, and influences
where lubricant ruptures (i.e., Swift-Stieber boundary condi- customers’ satisfaction.
tions are satisfied) averaged over intake and expansion strokes.
FIGURE 21 First compression ring average asperity contact

4. Conclusions
pressures for full load 1680 rpm for different honing angles The three-stage approach implemented in this study enables
and roughness. the modeling of surface roughness, surface texture, and
nominal geometry of the piston ring simultaneously and can
be used as a practical tool for tribological analysis of concept
power cylinder designs. Areal FFT of the measured cylinder
bore surface is used to generate control volumes that represent

the honing patterns. Flow and shear factors were numerically
generated for various surface patterns.
The effect of honing angle is significant in mixed lubrica-
tion regime where lubricant film thickness is close to the
roughness level. The effect of honing angle on lubricant film
thickness is mainly due to resistance to pressure flow. Low
honing angles and transversal patterns restrict the oil flow and
enhance hydrodynamic lubrication, reducing metal-to-metal
friction and wear in ring reversal zones. Since the piston speed
is very low near top dead center, this has very little contribution 2. Dimkovski, Z., Cabanettes, F., Lofgren, H. et al.,
to reduce power losses. Although, shear stress values are lower “Optimization of Cylinder Liner Surface Finish by Slide
for higher honing angles, significant friction force is observed Honing 2012,” Proceedings of the Institution of Mechanical
in the expansion stroke with high crosshatch angles which is Engineers, Part B: Engineering Manufacture 226(4):575-584,
primarily due to metal-to-metal contact and increased viscous 2012, doi:10.1177/0954405410396154.
shear as a result of lower film thicknesses. Honing angle has 3. Hoen, T., Schmid, J., and Stumpf, W., “Less Wear and Oil
very little effect in the mid stroke where lubricant film thick- Consumption through Helical Slide Honing of Engines by
ness is high. Honing the complete stroke with high surface Deutz,” MTZ 70(4):46-51, 2009, doi:10.1007/BF03226945.
roughness may result in increased oil consumption. Therefore, 4. Tomanik, E., “Modelling the Hydrodynamic Support of
partial texturing using advanced honing techniques can be a Cylinder Bore and Piston Rings with Laser Textured
viable method to reduce friction, wear, and oil consumption Surfaces,” Tribology International 59:90-96, 2013,
while reducing the probability of seizure. doi:10.1016/j.triboint.2012.01.016.
5. Ergen, O.R., Kurnaz, G., Sydemir, N.G., and Akalin,
Acknowledgment O., “Reduced Oil Consumption by Laser Surface
Texturing on Cylinders,” SAE International Journal of
This study was supported by The Scientific and Technological Commercial Vehicles 1:446-453, 2009,
Research Council of Turkey (TUBITAK) under the grant doi:10.4271/2008-01-2688.
number 214M300. 6. Patir, N. and Cheng, H.S., “An Average Flow Model for
Determining Effects of Three-Dimensional Roughness on
Nomenclature Partial Hydrodynamic Lubrication,” Journal of
Lubrication Technology 100(1):12-17, 1978,
Ftotal - Total friction force doi:10.1115/1.3453103.
h - Nominal film thickness 7. Patir, N. and Cheng, H.S., “Application of Average Flow
hs - Corresponding film thickness of the smooth domain Model to Lubrication between Rough Sliding Surfaces,”
Journal of Lubrication Technology 101(2):220-230, 1979,
hT - Local film thickness doi:10.1115/1.3453329.
hT - Roughness averaged film thickness 8. Greenwood, I.A. and Tripp, J.H., “The Contact of Two
Lx, Ly - Dimensions of the control domain Nominally Flat Surfaces,” Proceedings of the Institution of
p - Pressure Mechanical Engineers 185(1):625-633, 1971, doi:10.1243/
PIME_PROC_1970_185_069_02.
Rq - Representative cylinder surface texture (rms)
9. Rohde, S.M., “A Mixed Friction Model for Dynamically
Ra - Representative cylinder average texture Loaded Contacts with Application to Piston Ring
R z - Representative cylinder peak-to-peak texture Lubrication,” in Winter Annual Meeting of ASME, Chicago,
Up - Piston axial velocity IL.,” 1980.
U1, U2 - Sliding velocities of mating surfaces in control domain 10. Hu, Y., Cheng, H.S. et al., “Numerical Simulation of Piston
Ring in Mixed Lubrication-A Non-Axisymmetrical
α - Cross hatch angle Analysis,” Journal of Tribology 116(3):470-478, 1994,
ϕc - Contact factor defined by Wu and Zheng doi:10.1115/1.2928867.
ϕs - Shear flow factor 11. Akalin, O. and Newaz, G.M., “Piston Ring-Cylinder Bore
ϕfs - Shear stress factor Friction Modeling in Mixed Lubrication Regime: Part I-
Analytical Results,” Journal of Tribology 123(1):211-218, 2001,
ϕx - Pressure flow factor doi:10.1115/1.1286337.
ϕfp - Pressure stress factors 12. Michail, S.K. and Barber, G.C., “The Effects of Roughness on
μ - Dynamic viscosity of oil Piston Ring Lubrication. Part-2: The Relationship Between
μf - Boundary friction coefficient Cylinder Wall Surface Topography and Oil Film Thickness,”
Tribology Transactions 38(1):173-177, 1995,
σ* - Non-dimensional surface roughness doi:10.1080/10402009508983394.
σ - Composite surface roughness (rms) 13. Spencer, A., Almqvist, A., and Larsson, R., “A Semi-
t - Shear stress Deterministic Texture-Roughness Model of the Piston Ring-
Cylinder Contact,” Proceedings of the Institution of
Mechanical Engineers, Part J: Journal of Engineering
Tribology 225(6):325-333, 2011,
References doi:10.1177/1350650110396279.
14. Leighton, M., Rahmani, R., and Rahnejat, H., “Surface-
1. Richardson, D., “Review of Power Cylinder Friction for Specific Flow Factors for Prediction of Cross-Hatched
Diesel Engines,” Journal of Engineering for Gas Turbines and Surfaces,” Surface Topography: Metrology and Properties
Power 122(4):506-519, 2000, doi:10.1115/1.1290592. 4(2):025002, 2016, doi:10.1088/2051-672X/4/2/025002.
15. Jocsak, J., Li, Y., Tian, T., and Wong, V., “Modeling and 18. Grabon, W., Pawlus, P. et al., “Effects of Honed Cylinder
Optimizing Honing Texture for Reduced Friction in Internal Liner Surface Texture on Tribological Properties of Piston
Combustion Engines,” SAE Technical Paper 2006-01-0647, Ring-Liner Assembly in Short Time Tests,” Tribology
2006, doi:10.4271/2006-01-0647. International 113:137-148, 2017, doi:10.1016/j.
16. Chen, H. and Tian, T., “The Influences of Cylinder Liner triboint.2016.11.025.
Honing Patterns and Oil Control Ring Design Parameters on 19. Wu, C. and Zheng, L., “An Average Reynolds Equation
the Interaction between the Twinland Oil Control Ring and the for Partial Film Lubrication with a Contact Factor,”
Cylinder Liner in Internal Combustion Engines,” SAE Journal of Tribology 111(1):188-191, 1989,
Technical Paper 2008-01-1614, 2008, doi:10.4271/2008-01-1614. doi:10.1115/1.3261872.
17. Mezghani, S., Demirci, I., Zahouani, H., and El Mansori, M., 20. Wahiduzzaman, S., Keribar, R., and Dursunkaya, Z.A.,
“The Effect of Groove Texture Patterns on Piston-Ring Pack “Model for Evaporative Consumption of Lubricating Oil in
Friction,” Precision Engineering 36(2):210-217, 2012, Reciprocating Engines,” SAE Technical Paper 922202, 1992,
doi:10.1016/j.precisioneng.2011.09.008. doi:10.4271/922202.
© 2019 SAE International. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE International.
Positions and opinions advanced in this work are those of the author(s) and not necessarily those of SAE International. Responsibility for the content of the work lies
solely with the author(s).

04-12-03-0010-A Multiscale Cylinder Bore Honing Pattern Lubrication Model For Improved Engine Friction PDF

Uploaded by

Copyright:

Available Formats

04-12-03-0010-A Multiscale Cylinder Bore Honing Pattern Lubrication Model For Improved Engine Friction PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

04-12-03-0010-A Multiscale Cylinder Bore Honing Pattern Lubrication Model For Improved Engine Friction PDF

Uploaded by

Copyright:

Available Formats

Downloaded from SAE International by Columbia Univ, Tuesday, April 14, 2020

A Multiscale Cylinder Bore Honing

2. Theory and Modeling

© 2019 SAE International. All Rights Reserved

where ϕx and ϕy are pressure flow factors and ϕs is shear flow ¶p

© 2019 SAE International. All Rights Reserved

domain representing the crosshatched cylinder bore surface

FIGURE 4 Representative cylinder surface. FIGURE 6 Artificially generated honed surface.

© 2019 SAE International. All Rights Reserved

FIGURE 7 Control domain used to obtain flow and shear

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

where hT is the local film thickness in the textured domain

where Lx and Ly are the dimensions of the domains in the hs Lx æ

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

profile is considered here. Therefore, generated R z values are

FIGURE 15 Predicted minimum lubricant film thickness

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

© 2019 SAE International. All Rights Reserved

FIGURE 21 First compression ring average asperity contact

bore surface is used to generate control volumes that represent

You might also like

2. Theory and Modeling