- Email: [email protected]
An investigation of tribological behaviors of dynamically loaded non-grooved and micro-grooved journal bearings
Tribology International 58 (2013) 12–19
Contents lists available at SciVerse ScienceDirect
Tribology International journal homepage: www.elsevier.com/locate/triboint
An investigation of tribological behaviors of dynamically loaded non-grooved and micro-grooved journal bearings Hakan Adatepe a,1, Aydın Bıyıklıoglu b, Hasan Sofuoglu b,n a b
Department of Mechanical Engineering, Aksaray University, Aksaray, Turkey Department of Mechanical Engineering, Karadeniz Technical University, Trabzon 61080, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 December 2011 Received in revised form 13 September 2012 Accepted 13 September 2012 Available online 26 September 2012
Tribological performances of non-grooved and micro-grooved journal bearings were studied under dynamic loading. Numerous experiments were performed using purpose-built test rig and then simulated using various numerical methods. Friction force, friction coefficient, shaft center orbit, and film thickness were determined experimentally and numerically. The experimental and numerical results were in good agreement and the friction forces progressively increased on plain and circumferential, herringbone, and transversally micro-grooved bearing. The results show that it is necessary to complete detailed investigation about the tribological properties of the micro-grooved journal bearing by taking their shape, depth and operating condition into account. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Micro-grooved bearings Dynamic load Friction force Coefficient of friction
1. Introduction It is extremely important for journal bearing designers to know tribological performances of journal bearings used in a number of machines such as internal combustion engines, jet engines, compressors, piston pumps, mechanical presses, and rolling mills during the design process. The journal bearing surfaces encountered in many studies were assumed to be smooth. However, the possibility of improving bearing performance by modifying bearing surface geometry has attracted attention of many researchers and they have performed several theoretical studies on hydrodynamic lubrication field for rough journal bearing surfaces in recent years [1]. Zhang and Qiu [2] conducted a theoretical investigation on effects of geometric structure of journal bearing surfaces under dynamic loading and hydrodynamic lubrication conditions. They investigated effects of surface roughness of dynamically loaded journal bearings on longitudinal, transversal and isotropic basis. In their analysis, they used a statistical method (Stochastic Model) that was based on estimation principles developed by Christensen et al. [3,4]. They found that the maximum oil film pressure on the transversally rough bearing was higher than those of the longitudinal and isotropic rough conditions. They also concluded that
n
Corresponding author. Tel.: þ90 462 3772931; fax: þ90 462 3773336. E-mail addresses: [email protected], [email protected] (H. Adatepe), [email protected] (H. Sofuoglu). 1 Present address: Giresun University, Faculty of Engineering, Department of Energy Systems Engineering, Turkey. Tel.: þ90 454 215 02 80; mobile: þ90 505 899 23 85. 0301-679X/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.triboint.2012.09.009
oil film thickness for an isotropic rough condition was less than those of the transversal and longitudinal rough bearings. Hata et al. [5] investigated effects of frictional characteristics on surface roughness under mixed and hydrodynamic lubrication conditions by using a test device that operates on pin-disc principles. The roughness on bearing surface was cut precisely in a triangular and trapezoidal shape. It was found that the friction on transversally grooved surface was higher than the friction on plain and longitudinally grooved bearing surfaces. Nakahara et al. [6] performed a study to measure effects of surface roughness on friction characteristics after cutting regular and irregular threads on rectangular specimens spinning on a small axis. They also used a test device that operates on pin-disc principles. They found that the results obtained from the experiments correlated with theoretical results. They also came to conclude that the effects of roughness on transverse rough surface were greater than those on the longitudinal rough surface. Nakahara [7] then remarked that it was extremely difficult to achieve accurate measurements for surface roughness on hydrodynamic lubrication characteristics. This was related to the difficulties in measuring film layer thicknesses among rough surfaces. A new generation of bearings with micro-grooves has now been used particularly in automotive engines operating at extreme conditions since last two decades. The studies about performances of micro-grooved bearings have been, theoretically, shown that circumferential micro-grooves on journal bearings enhanced dynamic characteristics of bearings [8]. Moreover, it was also shown that circumferentially grooved bearings were stronger in terms of resistance against deformation and wear as compared
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Nomenclature R P h D L B C F C
bearing radius [m] oil film pressure [N/m2] oil film thickness [m] bearing diameter [m] bearing length [m] bearing width [m] radial clearance [m] bearing load [N] radial clearance [m]
to plain journal bearings. In a study conducted by Hargreaves and Armatys [9], the performances of different shapes of micro-grooved-journal bearings, transversal and longitudinal, under different static and dynamic loads were investigated. A sample of dynamic bearing load in a near sinusoidal form and on a single axis was applied on a bearing by using a cam mechanism. The variations of frictional moments within liquid friction zone under different static loads suspended on the bearing and at operating speeds of 400–2000 rpm were determined. They found that the journal bearings with circumferential grooves had lower frictional moments as compared to the other shapes at high operating speeds bearings. Watanabe et al. [10] developed the high performance micro-grooved engine bearings by cutting circumferential grooves on plain journal bearings. They also determined the performances of the microgrooved bearings by applying hydrodynamic and elastohydrodynamic lubrication theory. According to the recent studies, plain cylindrical journal bearings with grooves were used extensively in industry to distribute oil over the entire surface of the bearings and to obtain optimum performance as mentioned above [11,12]. It was shown that as the oil flow rate increased, the bearing temperature in the microgrooved bearings became less than that of the bearings without micro-grooves and as the oil remained in the grooves, the possibility for the bearings to fade was low. Moreover, it was found that effects of micro-grooves had direct impacts on journal bearings performances under dynamic loading conditions. It was also stated in the literature that the minimum oil film thickness of microgrooved bearings was thicker than that of the traditional plain bearings due to the oil being retained in the grooves. Finally, having micro-grooves on the surface of journal bearings was proven to be an effective method to enhance the tribological behavior of the journal bearings under starved lubrication conditions. However, there has been little discussion on the shape of groove and the operating conditions, such as friction zones. It became a necessary problem to discuss the shape of the groove and operating conditions of the journal bearing. The objective of this study was, therefore, to experimentally and theoretically investigate and compare the tribological behavior of purpose-made micro-grooved and non-grooved (plain) journal bearings loaded dynamically. In order to achieve this objective, the plain journal bearings were first tested at full film lubrication zone by utilizing the journal bearing test rig under dynamic bearing load. The experiments were then repeated for the circumferential, transversally and herringbone (V-shaped) micro-grooved journal bearings for the same bearing parameters to present the effects of different shapes of microgrooves on the performance of a typical engine crankshaft main bearing. Later on, a commercial software ORBIT, developed for analyzing dynamically loaded journal bearings using the mobility and the finite difference methods [13], was used to numerically simulate the experiments. Moreover, an additional computational program which operates on the Schaffrath method [14,15]
M
13
mobility vector [N-m] oil viscosity [Ns/m2] bearing rotational speed [1/s] circumferential velocity [m/s] eccentricity bearing’s width to diameter ratio, (B/D) eccentricity ratio, (e/C) eccentricity ratio in x direction eccentricity ratio in y direction axial coordinate circumferential coordinate
Z o V e
g e a b z
y
was utilized to test the tribological behavior of journal bearing. Finally, the experimental results were compared with the numerical predictions.
2. Experimental background and test procedure In this study, the purpose-built laboratory test rig was utilized to investigate tribological behavior of the dynamically loaded engine journal bearing. It uses a direct method where the friction torque of only the test bearing is measured without any interference of the shaft-supporting bearings. The test rig was designed to measure friction force under dynamic loading conditions by Bıyıklıoglu [16] and then modified to measure the orbit of the journal center under dynamic load by Bıyıklıoglu et al. [17–19]. Fig. 1 shows the cross-sectional view of the test rig along the shaft axis while the measurement system constructed specially to measure friction force is illustrated in Fig. 2. The detailed 18
17
16
15
14
13 12 11
1 10 2
3 4 5 6 7
8 1. Fixing ring 2. Journal 3. Test bearing 4. Bearing cap 5. Screwed ring 6. Bushing
7. Conic ring 8. Positioning element for bearing 9. Pulley 10. Optic sensor 11. Bearing housing 12. Loading shoe
9
13. Spherical joint 14. O- Ring 15. Piston 16. Nut- Ring 17. Servo-vane connection 18. Main structure support plate
Fig. 1. Cross-sectional view of the test rig along journal axis [16–18, 20].
14
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Tension bar (23) Bearing housing (11) F Conic ring (7) n
Bushing (6)
Lubricating oil Journal (2)
Bearing cap (4)
Squeezed oil
Journal bearing (3)
Fixing ring (1)
F
Fig. 2. Schematic representation of the developed friction force measurement system [16–18, 20].
3
3
51.880 47.750
1
4
3.5
Fig. 3. The shapes and dimensions of the journal bearing.
information about the test rig can be obtained from the above mentioned references and [20]. The journal was made of AISI 1070 hardened steel with Rockwell C hardness of 55 and supported at each end by bronze support bearings with a radial clearance of 0.025 mm. The bearings of S- 844-GS5049 M series used in the experiments were supplied by manufacturer located in Turkey. The geometry and the dimensions of the bearings were shown in Fig. 3 while Fig. 4 illustrated the photographs of plain, circumferential, herringbone, and transversally grooved journal bearings. The circumferential, transversally and herringbone grooves were fabricated onto the plain engine bearing. The circumferential grooves were made with a hard 601 angle thread cutting tool tip whereas the transversally and the herringbone grooves were made onto the plain engine bearing by pinching straight and 451 inclined grooves via knurling wheel used for general non-cutting forming purposes, respectively. The surface profiles of the journal bearings were measured prior and after the tests by using Perthometer M1, surface roughness measuring device. The average depths of all micro-grooves were taken as 30 mm in this study. For example, the journal bearing with circumferential grooves, the average roughness height (Ra) before the tests was found to be 7.815 mm while the average roughness depth was 29.7 mm. On the other hand, the average roughness height of the presumably plain and smooth journal bearing was 1.040 mm while the average roughness depth was 4.86 mm [21]. By using a belt-pulley system, the power taken from 5.5 kW motor was transmitted to the driven shaft of diameter d¼ 47.7 mm rotated at 1240 rpm. When the journal was driven, the journal bearings housing also turned in the same direction as the journal due to the friction between the test bearing and the rotating journal. The oscilloscope was included into the system to indicate the time history of net load along each axis passing through two opposing pairs of cylinders. The polar loads, two bearing
Fig. 4. Photographs of (a) plain, (b) circumferential, (c) herringbone and (d) transverse grooved journal bearings.
H. Adatepe et al. / Tribology International 58 (2013) 12–19
load components (vertical and horizontal), and friction forces were then recorded by a DSO 400 GOULD model oscilloscope. An optic sensor was used to measure the crank angle during the experiments while four transducers were utilized to measure the journal center trajectory. The frictional moment occurring between the journal bearings and bearing cap was measured by using a Wheatstone bridge circuit via the strain gauges. A fullbridge circuit was set up by using four strain gauges. The signal taken from the bridge circuit was amplified with a KWS/T-5 model amplificatory and sent to a computer via a data acquisition system. The conversion to a bending force was executed on the four measurement beams on which strain gauges attached to both sides gave a signal proportional to the frictional moment. The signal values were then converted into friction torque and friction force values, which were determined using a calibration line obtained from the strain gauges. Journal load, friction force, and displacement of journal center were measured from the record by every 151 crank angle intervals. The test bearings were run for 2 h in order to flush out wear debris during the running-in process under low speed and low loading conditions. In order to provide temperature stability, the system was run for about 30 min and the lubricant temperature reached 23 1C. The experiments were carried out with commercial base oil.
3. Numerical background ORBIT program was applied in analyses of dynamically loaded journal bearings in this study. This computer aided engineering program was first used to find solutions with the mobility method developed by Booker [22,23]. In the mobility method, the mobility vector, M, which is the dimensionless squeeze velocity with which a non-rotating shaft moves to support the load, is calculated by converting the effect of rotation to an equivalent squeeze motion under steady load. For a rotating shaft, the squeeze velocity (e´) is given by [13] e0 ¼
9F9½C=R2 M þ oe LDZ=C
15
with the following boundary conditions: z¼0
Pðz, yÞ ¼ P1,j
z¼L
Pðz, yÞ ¼ Pn,j
ð7Þ
The Schaffrath method [14,15] is the last computational program used in this study to test performances of dynamically loaded engine journal bearing. This method calculates the friction force and the coefficient of friction by using the following equations: e Z 2p 1 F sinb þ B r o pffiffiffiffiffiffiffiffiffiffiffi þ 2pZBRg_ cos2g pffiffiffiffiffiffiffiffiffiffiffi 1 Fs ¼ ð8Þ 2R c 1e2 1e2 e BZr o 2p 2pZBR 1 _ cos2g pffiffiffiffiffiffiffiffiffiffiffi 1 pffiffiffiffiffiffiffiffiffiffiffi þ m ¼ sinb þ g ð9Þ 2R F Fc 1e2 1e2
4. Results and discussion Fig. 5 showed the dynamic bearing load views taken directly from the oscilloscope screen, the polar load and the horizontal and vertical components of bearing load. It was clear from the figure that the horizontal and vertical components of applied load were completely variable (dynamic) and act on the journal bearing in a sinusoidal form. In Fig. 6, the variations of horizontal and vertical components of applied dynamic bearing load with their resultant were given with respect to bearing’s crank angle. These figures clearly indicated that the maximum bearing load was reached its maximum value at 16.68 kN which corresponds to the crank angle of 97.51 while the minimum bearing load of 172.5 N was obtained at the crank angle of 172.51.
ð1Þ
In dynamic analyses, the components of mobility vector (M) can generally be expressed as [13] M ¼ f ðe, gÞ
ð2Þ
The analytical equations for mobility components obtained from the curve fitting of mobility components are expressed in the following manners [13,24]: ( ) pffiffiffiffiffiffiffiffiffiffi f ða, gÞ pffiffiffiffiffiffi b2 2 2 Mx ¼ ð10:4 1aÞ þ a gð4=3gÞ 3p0:24g e þ 1a pg2 ð3Þ 8 9 2 3 ag 2b2 > < 54 þ 7 þ a8 ð1þ aÞ0:3g2 1 þ 2apffiffi3 þ 15ð1aÞ > = b f ða, gÞ My ¼ 1 > > ð1aÞ : 0:016 1 0:034 ; ð1:03aÞ g2 ð4Þ f ða, gÞ ¼
ð1aÞ2 1=2
pg
2
1þ
g2
2ð1aÞ
ð5Þ
The second method provided by this commercial software is based on a direct finite difference solution of the Reynolds equation [13]: ! ! 3 3 @ h @P 1 @ h @P 1 V @h o @h @h þ ð6Þ þ þ 2 ¼ @z 12Z @z R 2 @y @t 2 @z R @y 12Z @y
Fig. 5. Dynamic bearing load views taken from the oscilloscope screen: (a) Polar load, (b) horizontal and (c) vertical components.
16
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Fx Fy Fsum
Fig. 8. Variations of friction force for four different bearing types under the same dynamic bearing load.
Friction Coefficient, µ
Fig. 6. Variation of the resultant applied bearing load with vertical and horizontal components as a function of crank angle.
Fig. 9. Variations of friction coefficients for four different bearing types under the same dynamic bearing load.
Fig. 7. Variations of measured friction forces formed on oil film of journal bearing for (a) transversally, (b) herringbone, and (c) circumferential grooved and (d) plain journal bearings.
The distributions of measured friction forces taken directly from the oscilloscope screen during the test under dynamic bearing load for all bearings tested in this study were given in Fig. 7 while Fig. 8 illustrated their variations against the crank angle. As concluded from Fig. 8 that the highest friction force of 21.12 N was obtained at 127.51 crank angle for the transversally grooved bearing followed by 20.02 N at 112.51 crank angle for herringbone grooved bearing, 19.54 N at 67.51 crank angle for circumferential grooved bearing, and 11.63 N at 82.51 crank angle for plain bearing. The average increase in the friction force based on to the plain journal bearing was obtained to be approximately 1.81, 1.72, and 1.68 times for transversally, herringbone, and circumferentially grooved bearings, respectively. It was observed from Figs. 7 and 8 under the dynamic loading condition that the
effect of groove’s shape on the friction force varied considerably with the crank angle. Both figures clearly indicated that the lowest friction force was obtained from the non-grooved journal bearing while the highest friction force was measured for the transversally grooved journal bearing as opposed to enhancing performances of the micro-grooved journal bearing given in the literature. The reason for the higher frictional values occurred on the micro-grooved bearings can be attributed to a higher disruption of the oil flow on the micro-grooved bearings as compared to non-grooved bearing. Fig. 9 depicts the variation of coefficients of friction with crank angle for four types of bearings (plain and grooved) under the effect of dynamic bearing load. It was seen from this figure that the smallest value of the coefficient of friction was exhibited on the plain journal bearing while the largest one was obtained from the transversal grooved journal bearing similar to the distribution of the friction force, as expected. The maximum values of coefficient of friction obtained from the experiments were 0.0036, 0.0024, 0.0021 and 0.0015 for the transversally, the herringbone, and the circumferential grooves, and for the plain journal bearings, respectively. The increasing ratios for the coefficient of
H. Adatepe et al. / Tribology International 58 (2013) 12–19
was 10.16 N at 1001 crank angle for the finite difference method. On the other hand, the highest friction force obtained by the Schaffrath method at 1501 crank angle was 11.29 N while 11.63 N was determined for the plain bearing at 82.51 crank angle.All results for the coefficient of friction obtained from the experimental and the numerical studies were then demonstrated in Fig. 11. It was clear from the figure that the values for the coefficients of friction were very close to each other, especially for those obtained from the mobility and finite difference methods, as well as the corresponding crank angles (between 1501 and 1701). The exception was occurred for the maximum value of the friction coefficient obtained by the Schaffrath method, similar to
Friction Coefficient, µ
friction with respect to the plain journal bearing were approximately 1.4, 1.6 and 2.4 for the circumferential, the herringbone, and the transversally micro-grooved bearings, respectively. These numbers indicated that all the experiments on the journal bearings took place within the liquid friction zone. Therefore, Fig. 9 illustrated that with the same dynamic load applied, the micro-grooved journal bearings yielded a higher coefficient of friction than that of non-grooved bearing as opposed to those given in literature. This disagreement can be attributed to the fact that studies were performed in different working conditions. For example, motors work in mixed lubrication zone during startup and stopping motions in which metal-to-metal contact could be more possible to occur due to starved lubrication conditions. The micro-grooves, in that case, help oil to easily lubricate the surface of the journal bearing in mixed lubrication zone. It can, therefore, be concluded that micro-grooves on the surface of journal bearings can be used to improve the tribological behavior of the journal bearings under starved lubrication conditions. On the other hand, the experiments and the numerical studies, in our case, were pursued in the full film lubrication (the liquid friction) zone so that there was no metal-to-metal contact and the microgrooves struck the lubricating oil to flow on the surface of microgrooved bearings. It can, therefore, be deduced that the operating condition plays an important role on the performance of microgrooved journal bearing. When Figs. 6, 8, and 9 were evaluated together, it could be concluded that the friction force and the coefficient of friction reached to peak values at the crank angle in which the bearing load was also maximum (in the vicinity of 1001). Since the bearing load is dynamics, it is possible to have different crank angles for the peak values of the friction force and the coefficient of friction. As for the theoretical study, the tribological behavior of the plain journal bearing was determined by using ORBIT software which utilizes two numerical methods, the mobility method and the finite difference methods. Moreover, another computational solution was determined by using the Schaffrath method. The friction forces obtained numerically and experimentally were then plotted at the same figure for comparison purpose, shown in Fig. 10. It was found that the numerical and the experimental results obtained in the full film lubrication zone were in good agreement. It was seen that the highest friction force was 9.94 N at 1001 crank angle obtained by the mobility method while it
17
Fig. 11. Variation of friction coefficient under the same dynamic bearing load obtained from numerical and experimental analyses.
0°
1
2
270° 270°
360°
270° 360° 180° 270°
0.2
0.4 360°
0.6
0.8
1.0 30.000 N
180° 180°
90° 90° 90°
4
3 180° Mobility Method Finite Difference Method Schaffrath Method Polar Load
Fig. 10. Variation of friction force under the same dynamic bearing load obtained from numerical and experimental analyses.
Fig. 12. Polar load diagram for the plain journal bearing and journal center orbits obtained from different numerical methods.
18
H. Adatepe et al. / Tribology International 58 (2013) 12–19
the case in the friction force. It is normal to obtain the peak values of the friction force and the coefficient of friction at the different crank angles due to the nature of the dynamic bearing load. However, although the crank angles corresponding to the peak friction forces differed from each other, the variations of the friction force through one cycle of the crank angle looked similar; i.e., increasing and decreasing regions almost coincided. This was held true for the coefficient of friction more strongly. The orbits drawn by the shaft center based on three different computational methods and polar load diagram for the dynamically loaded plain journal bearings were shown in Fig. 12. The figure showed that the orbits drawn by the shaft center complied well with each other. The small discrepancies might be caused by the different computational method used. It was also seen from the figure that the maximum eccentricity ratio existed for two methods, namely; mobility and finite difference methods, was completely coincided at the position angle of 1871 with a value of 0.76 while the Schaffrath method had a very small divergence from these two methods having the value of 0.79. The graph plotted in order to determine the relationship between the experimentally obtained coefficients of friction on the plain journal bearing and the minimum oil film thickness obtained from different numerical methods was given in
Fig. 15. Variation of minimum oil film thicknesses with respect to crank and position angles.
Fig. 13. This figure illustrated that when the oil film thickness decreased the coefficient of friction increased and wherever the coefficient of friction reached to maximum value the oil film thickness was attained to minimum value. In Fig. 14, the variation of minimum oil film thickness against the bearing load was shown. It was demonstrated that the minimum oil film thicknesses were found to be 6.17 mm, 6.11 mm, and 5.04 mm by utilizing the mobility, the finite difference and Schaffrath methods, respectively. They all existed at around crank angle of 1411. Moreover, it was seen that the crank angle position of the minimum oil film thickness, was just occurred 43.51 beyond to that of the maximum bearing load (97.51). The variation of minimum oil film thickness versus the crank and the position angles was illustrated in Fig. 15. This figure showed that the smallest oil film thickness was just occurred at the position angle of 1861.
5. Conclusions
Fig. 13. Variation of coefficients of friction and minimum oil film thickness obtained from different numerical methods.
Fig. 14. Variation of minimum oil film thickness with respect to bearing load.
Before designing micro-grooved bearings, it is necessary to predict performances of the micro-grooved bearings accurately. We, therefore, investigated the effects of micro-grooves on the tribological behavior of the journal bearing within liquid friction zone using the purpose-built friction test rig in this study. The grooves in micron scale basis were made onto the plain journal bearings as circumferential, transversal and herringbone shapes and they were then subjected to dynamic load. This investigation yielded the following results: 1. The experiments showed that the lowest value of friction force was determined on the plain journal bearing. The experimentally obtained values increased progressively for the circumferential and the herringbone micro-grooved bearings and, finally, the highest frictional force distribution was exercised on the transversal micro-grooved bearing. 2. The shape of micro-groove affects the coefficient of friction, attaining the lowest value to the plain journal bearing and resulting in the maximum value for the transversally microgrooved journal bearing, similar to the trend of the friction force. 3. The friction force distributions as well as the variations in the coefficient of friction obtained from the different numerical methods for the plain journal bearing were very similar and agreed well with each other and also with those of the experimentally obtained results.
H. Adatepe et al. / Tribology International 58 (2013) 12–19
4. The minimum oil film thickness value appeared at the furthest position from the point of the maximum bearing load value existed. In conclusion, the results of this research showed that the shapes of micro-grooves played an important role on the oil film thickness, the coefficient of friction, and the friction force under the dynamic load. This study further indicated that the microgrooves developed the worse tribological properties on the journal bearings when they operated in the liquid friction zone than the non-grooved journal bearings as opposed to knowledge written in the literature. It can, therefore, be concluded that the use of generalized results regardless of shape of micro-groove and operating condition may result in erroneous estimates of the tribological properties of the micro-grooved journal bearing. It is, therefore, recommended that performing detailed investigation in conjunction with shape, depth and operating condition is necessary to obtain reliable data regarding the tribological properties of the micro-grooved journal bearing.
Acknowledgment This work was supported by the Research Fund of Karadeniz Technical University, Project no. 2002.112.003.03. References [1] Hu J. Experimental and theoretical investigation of roughness effects on thin laminar fluids films. PhD thesis. University of Toronto, Canada; 1997. [2] Zhang C, Qiu Z. Effects of surface texture on hydrodynamic lubrication of dynamically loaded journal bearings. Tribology Transactions STLE 1998;41: 43–8. [3] Christensen H, Tonder K. The hydrodynamic lubrication of rough bearing surfaces of finite width. Journal of Lubrication Technology—Transactions of the ASME 1971;93:324–30. [4] Christensen H, Tonder K. The hydrodynamic lubrication of rough journal bearings. Journal of Lubrication Technology—Transactions of the ASME 1973;95:166–72. [5] Hata H, Nakahara T, Aoki H. Measurement of friction in lightly load hydrodynamic sliders with striated roughness, The Winter Annual Meeting of the ASME, Chicago, Illinois, 75–92; 16–21 November 1980. [6] Nakahara T, Takesue M, Aoki H. Effects of surface roughness and bearing slenderness ratio on hydrodynamic lubrication. Journal of JSLE 1983;28: 543–8.
19
[7] Nakahara T. Effect of truncation in surface roughness on hydrodynamic pivoted slider and practical expression of truncation. In: Andersson P, Ronkainen H, Holmberg K, editors. 9th NORDTRIB, 3; 2000. p. 935–44. [8] Kumada Y, Hashizume K, Kimura Y. Performance of plain bearings with circumferential microgrooves. Tribology Transactions STLE 1996;39:81–6. [9] Hargreaves DJ, Armatys D. Performance of a microgrooved journal bearing under steady and dynamic loading. Tribo-Test 1999;5:277–86. [10] Watanabe K, Natsuma J, Hashizuma K, Ozasa T, Noda T, Masuda Y. Theoretical analysis of bearing performance of microgrooved bearing. JSAE Review 2000;21:29–33. [11] Roy L. Thermo-hydrodynamic performance of grooved oil journal bearing. Tribology International 2009;42:1187–98. [12] Hirayama T, Yamaguchi N, Sakai S, Hishida N, Matsuoka T, Yabe H. Optimization of groove dimensions in herringbone-grooved journal bearings for improved repeatable run-out characteristics. Tribology International 2009;42:675–81. [13] Ricardo Software, Journal Bearing Analysis-Orbit, Documentation/User Manual Version 1.2, USA; 2000. [14] Schaffrath G. The lubricating bearings with arbitrary gap shape-shifting of the shaft journal with time-varying load. PhD thesis. Technical University, Karlsruhe, Germany (in German); 1967. [15] Harbordt J, Klumpp R. Computer program for calculating the displacement train, the local pressures and the factual gap lubricating bearings in arbitrary geometry under static and dynamic load, Research Reports Internal Combustion Engine Projects, No. 98, Issue 137; 1972 (in German). [16] Bıyıklıoglu A. Surface fatigue on dynamically loaded journal bearings. PhD thesis. Karadeniz Technical University, Trabzon, Turkey; 1986 (in Turkish). [17] Durak E, Kurbanoglu C, Bıyıklıoglu A, Kaleli H. Measurement of friction force and effects of oil fortifier in engine journal bearings under dynamic loading conditions. Tribology International 2003;36:599–607. [18] Bıyıklıoglu A, Cuvalci H, Adatepe H, Bas H, Duman MS. A new measurement test apparatus and method for the friction force measurement in the journal bearings under dynamic loading Part I. Experimental Techniques 2005;29: 22–5. [19] Bıyıklıoglu A, Cuvalci H, Adatepe H, Bas H, Duman MS. A new measurement test apparatus and method for the friction force measurement in the journal bearings under dynamic loading Part II. Experimental Techniques 2005;29: 33–6. [20] Adatepe H. The investigation of the friction behavior of the microgrooved journal bearings under static and dynamic loading conditions. PhD thesis, Karadeniz Technical University, Trabzon, Turkey; 2006 (in Turkish). [21] Adatepe H, Bıyıklıoglu A, Sofuoglu H. An experimental investigation on frictional behavior of statically loaded micro-grooved journal bearing. Tribology International 2011;44:942–8. [22] Booker JF. Dynamically loaded journal bearings: mobility method of solution. Journal of Basic Engineering—Transactions of the ASME 1965;4:537–46. [23] Booker JF. Dynamically loaded journal bearings: numerical application of mobility method. Journal of Lubrication Technology—Transactions of the ASME 1971;93:168–76. [24] Goenka PK. Analytical curve fits for solution parameters of dynamically loaded journal bearings. Journal of Tribology —Transactions of the ASME 1984;106:421–8.
Contents lists available at SciVerse ScienceDirect
Tribology International journal homepage: www.elsevier.com/locate/triboint
An investigation of tribological behaviors of dynamically loaded non-grooved and micro-grooved journal bearings Hakan Adatepe a,1, Aydın Bıyıklıoglu b, Hasan Sofuoglu b,n a b
Department of Mechanical Engineering, Aksaray University, Aksaray, Turkey Department of Mechanical Engineering, Karadeniz Technical University, Trabzon 61080, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 December 2011 Received in revised form 13 September 2012 Accepted 13 September 2012 Available online 26 September 2012
Tribological performances of non-grooved and micro-grooved journal bearings were studied under dynamic loading. Numerous experiments were performed using purpose-built test rig and then simulated using various numerical methods. Friction force, friction coefficient, shaft center orbit, and film thickness were determined experimentally and numerically. The experimental and numerical results were in good agreement and the friction forces progressively increased on plain and circumferential, herringbone, and transversally micro-grooved bearing. The results show that it is necessary to complete detailed investigation about the tribological properties of the micro-grooved journal bearing by taking their shape, depth and operating condition into account. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Micro-grooved bearings Dynamic load Friction force Coefficient of friction
1. Introduction It is extremely important for journal bearing designers to know tribological performances of journal bearings used in a number of machines such as internal combustion engines, jet engines, compressors, piston pumps, mechanical presses, and rolling mills during the design process. The journal bearing surfaces encountered in many studies were assumed to be smooth. However, the possibility of improving bearing performance by modifying bearing surface geometry has attracted attention of many researchers and they have performed several theoretical studies on hydrodynamic lubrication field for rough journal bearing surfaces in recent years [1]. Zhang and Qiu [2] conducted a theoretical investigation on effects of geometric structure of journal bearing surfaces under dynamic loading and hydrodynamic lubrication conditions. They investigated effects of surface roughness of dynamically loaded journal bearings on longitudinal, transversal and isotropic basis. In their analysis, they used a statistical method (Stochastic Model) that was based on estimation principles developed by Christensen et al. [3,4]. They found that the maximum oil film pressure on the transversally rough bearing was higher than those of the longitudinal and isotropic rough conditions. They also concluded that
n
Corresponding author. Tel.: þ90 462 3772931; fax: þ90 462 3773336. E-mail addresses: [email protected], [email protected] (H. Adatepe), [email protected] (H. Sofuoglu). 1 Present address: Giresun University, Faculty of Engineering, Department of Energy Systems Engineering, Turkey. Tel.: þ90 454 215 02 80; mobile: þ90 505 899 23 85. 0301-679X/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.triboint.2012.09.009
oil film thickness for an isotropic rough condition was less than those of the transversal and longitudinal rough bearings. Hata et al. [5] investigated effects of frictional characteristics on surface roughness under mixed and hydrodynamic lubrication conditions by using a test device that operates on pin-disc principles. The roughness on bearing surface was cut precisely in a triangular and trapezoidal shape. It was found that the friction on transversally grooved surface was higher than the friction on plain and longitudinally grooved bearing surfaces. Nakahara et al. [6] performed a study to measure effects of surface roughness on friction characteristics after cutting regular and irregular threads on rectangular specimens spinning on a small axis. They also used a test device that operates on pin-disc principles. They found that the results obtained from the experiments correlated with theoretical results. They also came to conclude that the effects of roughness on transverse rough surface were greater than those on the longitudinal rough surface. Nakahara [7] then remarked that it was extremely difficult to achieve accurate measurements for surface roughness on hydrodynamic lubrication characteristics. This was related to the difficulties in measuring film layer thicknesses among rough surfaces. A new generation of bearings with micro-grooves has now been used particularly in automotive engines operating at extreme conditions since last two decades. The studies about performances of micro-grooved bearings have been, theoretically, shown that circumferential micro-grooves on journal bearings enhanced dynamic characteristics of bearings [8]. Moreover, it was also shown that circumferentially grooved bearings were stronger in terms of resistance against deformation and wear as compared
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Nomenclature R P h D L B C F C
bearing radius [m] oil film pressure [N/m2] oil film thickness [m] bearing diameter [m] bearing length [m] bearing width [m] radial clearance [m] bearing load [N] radial clearance [m]
to plain journal bearings. In a study conducted by Hargreaves and Armatys [9], the performances of different shapes of micro-grooved-journal bearings, transversal and longitudinal, under different static and dynamic loads were investigated. A sample of dynamic bearing load in a near sinusoidal form and on a single axis was applied on a bearing by using a cam mechanism. The variations of frictional moments within liquid friction zone under different static loads suspended on the bearing and at operating speeds of 400–2000 rpm were determined. They found that the journal bearings with circumferential grooves had lower frictional moments as compared to the other shapes at high operating speeds bearings. Watanabe et al. [10] developed the high performance micro-grooved engine bearings by cutting circumferential grooves on plain journal bearings. They also determined the performances of the microgrooved bearings by applying hydrodynamic and elastohydrodynamic lubrication theory. According to the recent studies, plain cylindrical journal bearings with grooves were used extensively in industry to distribute oil over the entire surface of the bearings and to obtain optimum performance as mentioned above [11,12]. It was shown that as the oil flow rate increased, the bearing temperature in the microgrooved bearings became less than that of the bearings without micro-grooves and as the oil remained in the grooves, the possibility for the bearings to fade was low. Moreover, it was found that effects of micro-grooves had direct impacts on journal bearings performances under dynamic loading conditions. It was also stated in the literature that the minimum oil film thickness of microgrooved bearings was thicker than that of the traditional plain bearings due to the oil being retained in the grooves. Finally, having micro-grooves on the surface of journal bearings was proven to be an effective method to enhance the tribological behavior of the journal bearings under starved lubrication conditions. However, there has been little discussion on the shape of groove and the operating conditions, such as friction zones. It became a necessary problem to discuss the shape of the groove and operating conditions of the journal bearing. The objective of this study was, therefore, to experimentally and theoretically investigate and compare the tribological behavior of purpose-made micro-grooved and non-grooved (plain) journal bearings loaded dynamically. In order to achieve this objective, the plain journal bearings were first tested at full film lubrication zone by utilizing the journal bearing test rig under dynamic bearing load. The experiments were then repeated for the circumferential, transversally and herringbone (V-shaped) micro-grooved journal bearings for the same bearing parameters to present the effects of different shapes of microgrooves on the performance of a typical engine crankshaft main bearing. Later on, a commercial software ORBIT, developed for analyzing dynamically loaded journal bearings using the mobility and the finite difference methods [13], was used to numerically simulate the experiments. Moreover, an additional computational program which operates on the Schaffrath method [14,15]
M
13
mobility vector [N-m] oil viscosity [Ns/m2] bearing rotational speed [1/s] circumferential velocity [m/s] eccentricity bearing’s width to diameter ratio, (B/D) eccentricity ratio, (e/C) eccentricity ratio in x direction eccentricity ratio in y direction axial coordinate circumferential coordinate
Z o V e
g e a b z
y
was utilized to test the tribological behavior of journal bearing. Finally, the experimental results were compared with the numerical predictions.
2. Experimental background and test procedure In this study, the purpose-built laboratory test rig was utilized to investigate tribological behavior of the dynamically loaded engine journal bearing. It uses a direct method where the friction torque of only the test bearing is measured without any interference of the shaft-supporting bearings. The test rig was designed to measure friction force under dynamic loading conditions by Bıyıklıoglu [16] and then modified to measure the orbit of the journal center under dynamic load by Bıyıklıoglu et al. [17–19]. Fig. 1 shows the cross-sectional view of the test rig along the shaft axis while the measurement system constructed specially to measure friction force is illustrated in Fig. 2. The detailed 18
17
16
15
14
13 12 11
1 10 2
3 4 5 6 7
8 1. Fixing ring 2. Journal 3. Test bearing 4. Bearing cap 5. Screwed ring 6. Bushing
7. Conic ring 8. Positioning element for bearing 9. Pulley 10. Optic sensor 11. Bearing housing 12. Loading shoe
9
13. Spherical joint 14. O- Ring 15. Piston 16. Nut- Ring 17. Servo-vane connection 18. Main structure support plate
Fig. 1. Cross-sectional view of the test rig along journal axis [16–18, 20].
14
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Tension bar (23) Bearing housing (11) F Conic ring (7) n
Bushing (6)
Lubricating oil Journal (2)
Bearing cap (4)
Squeezed oil
Journal bearing (3)
Fixing ring (1)
F
Fig. 2. Schematic representation of the developed friction force measurement system [16–18, 20].
3
3
51.880 47.750
1
4
3.5
Fig. 3. The shapes and dimensions of the journal bearing.
information about the test rig can be obtained from the above mentioned references and [20]. The journal was made of AISI 1070 hardened steel with Rockwell C hardness of 55 and supported at each end by bronze support bearings with a radial clearance of 0.025 mm. The bearings of S- 844-GS5049 M series used in the experiments were supplied by manufacturer located in Turkey. The geometry and the dimensions of the bearings were shown in Fig. 3 while Fig. 4 illustrated the photographs of plain, circumferential, herringbone, and transversally grooved journal bearings. The circumferential, transversally and herringbone grooves were fabricated onto the plain engine bearing. The circumferential grooves were made with a hard 601 angle thread cutting tool tip whereas the transversally and the herringbone grooves were made onto the plain engine bearing by pinching straight and 451 inclined grooves via knurling wheel used for general non-cutting forming purposes, respectively. The surface profiles of the journal bearings were measured prior and after the tests by using Perthometer M1, surface roughness measuring device. The average depths of all micro-grooves were taken as 30 mm in this study. For example, the journal bearing with circumferential grooves, the average roughness height (Ra) before the tests was found to be 7.815 mm while the average roughness depth was 29.7 mm. On the other hand, the average roughness height of the presumably plain and smooth journal bearing was 1.040 mm while the average roughness depth was 4.86 mm [21]. By using a belt-pulley system, the power taken from 5.5 kW motor was transmitted to the driven shaft of diameter d¼ 47.7 mm rotated at 1240 rpm. When the journal was driven, the journal bearings housing also turned in the same direction as the journal due to the friction between the test bearing and the rotating journal. The oscilloscope was included into the system to indicate the time history of net load along each axis passing through two opposing pairs of cylinders. The polar loads, two bearing
Fig. 4. Photographs of (a) plain, (b) circumferential, (c) herringbone and (d) transverse grooved journal bearings.
H. Adatepe et al. / Tribology International 58 (2013) 12–19
load components (vertical and horizontal), and friction forces were then recorded by a DSO 400 GOULD model oscilloscope. An optic sensor was used to measure the crank angle during the experiments while four transducers were utilized to measure the journal center trajectory. The frictional moment occurring between the journal bearings and bearing cap was measured by using a Wheatstone bridge circuit via the strain gauges. A fullbridge circuit was set up by using four strain gauges. The signal taken from the bridge circuit was amplified with a KWS/T-5 model amplificatory and sent to a computer via a data acquisition system. The conversion to a bending force was executed on the four measurement beams on which strain gauges attached to both sides gave a signal proportional to the frictional moment. The signal values were then converted into friction torque and friction force values, which were determined using a calibration line obtained from the strain gauges. Journal load, friction force, and displacement of journal center were measured from the record by every 151 crank angle intervals. The test bearings were run for 2 h in order to flush out wear debris during the running-in process under low speed and low loading conditions. In order to provide temperature stability, the system was run for about 30 min and the lubricant temperature reached 23 1C. The experiments were carried out with commercial base oil.
3. Numerical background ORBIT program was applied in analyses of dynamically loaded journal bearings in this study. This computer aided engineering program was first used to find solutions with the mobility method developed by Booker [22,23]. In the mobility method, the mobility vector, M, which is the dimensionless squeeze velocity with which a non-rotating shaft moves to support the load, is calculated by converting the effect of rotation to an equivalent squeeze motion under steady load. For a rotating shaft, the squeeze velocity (e´) is given by [13] e0 ¼
9F9½C=R2 M þ oe LDZ=C
15
with the following boundary conditions: z¼0
Pðz, yÞ ¼ P1,j
z¼L
Pðz, yÞ ¼ Pn,j
ð7Þ
The Schaffrath method [14,15] is the last computational program used in this study to test performances of dynamically loaded engine journal bearing. This method calculates the friction force and the coefficient of friction by using the following equations: e Z 2p 1 F sinb þ B r o pffiffiffiffiffiffiffiffiffiffiffi þ 2pZBRg_ cos2g pffiffiffiffiffiffiffiffiffiffiffi 1 Fs ¼ ð8Þ 2R c 1e2 1e2 e BZr o 2p 2pZBR 1 _ cos2g pffiffiffiffiffiffiffiffiffiffiffi 1 pffiffiffiffiffiffiffiffiffiffiffi þ m ¼ sinb þ g ð9Þ 2R F Fc 1e2 1e2
4. Results and discussion Fig. 5 showed the dynamic bearing load views taken directly from the oscilloscope screen, the polar load and the horizontal and vertical components of bearing load. It was clear from the figure that the horizontal and vertical components of applied load were completely variable (dynamic) and act on the journal bearing in a sinusoidal form. In Fig. 6, the variations of horizontal and vertical components of applied dynamic bearing load with their resultant were given with respect to bearing’s crank angle. These figures clearly indicated that the maximum bearing load was reached its maximum value at 16.68 kN which corresponds to the crank angle of 97.51 while the minimum bearing load of 172.5 N was obtained at the crank angle of 172.51.
ð1Þ
In dynamic analyses, the components of mobility vector (M) can generally be expressed as [13] M ¼ f ðe, gÞ
ð2Þ
The analytical equations for mobility components obtained from the curve fitting of mobility components are expressed in the following manners [13,24]: ( ) pffiffiffiffiffiffiffiffiffiffi f ða, gÞ pffiffiffiffiffiffi b2 2 2 Mx ¼ ð10:4 1aÞ þ a gð4=3gÞ 3p0:24g e þ 1a pg2 ð3Þ 8 9 2 3 ag 2b2 > < 54 þ 7 þ a8 ð1þ aÞ0:3g2 1 þ 2apffiffi3 þ 15ð1aÞ > = b f ða, gÞ My ¼ 1 > > ð1aÞ : 0:016 1 0:034 ; ð1:03aÞ g2 ð4Þ f ða, gÞ ¼
ð1aÞ2 1=2
pg
2
1þ
g2
2ð1aÞ
ð5Þ
The second method provided by this commercial software is based on a direct finite difference solution of the Reynolds equation [13]: ! ! 3 3 @ h @P 1 @ h @P 1 V @h o @h @h þ ð6Þ þ þ 2 ¼ @z 12Z @z R 2 @y @t 2 @z R @y 12Z @y
Fig. 5. Dynamic bearing load views taken from the oscilloscope screen: (a) Polar load, (b) horizontal and (c) vertical components.
16
H. Adatepe et al. / Tribology International 58 (2013) 12–19
Fx Fy Fsum
Fig. 8. Variations of friction force for four different bearing types under the same dynamic bearing load.
Friction Coefficient, µ
Fig. 6. Variation of the resultant applied bearing load with vertical and horizontal components as a function of crank angle.
Fig. 9. Variations of friction coefficients for four different bearing types under the same dynamic bearing load.
Fig. 7. Variations of measured friction forces formed on oil film of journal bearing for (a) transversally, (b) herringbone, and (c) circumferential grooved and (d) plain journal bearings.
The distributions of measured friction forces taken directly from the oscilloscope screen during the test under dynamic bearing load for all bearings tested in this study were given in Fig. 7 while Fig. 8 illustrated their variations against the crank angle. As concluded from Fig. 8 that the highest friction force of 21.12 N was obtained at 127.51 crank angle for the transversally grooved bearing followed by 20.02 N at 112.51 crank angle for herringbone grooved bearing, 19.54 N at 67.51 crank angle for circumferential grooved bearing, and 11.63 N at 82.51 crank angle for plain bearing. The average increase in the friction force based on to the plain journal bearing was obtained to be approximately 1.81, 1.72, and 1.68 times for transversally, herringbone, and circumferentially grooved bearings, respectively. It was observed from Figs. 7 and 8 under the dynamic loading condition that the
effect of groove’s shape on the friction force varied considerably with the crank angle. Both figures clearly indicated that the lowest friction force was obtained from the non-grooved journal bearing while the highest friction force was measured for the transversally grooved journal bearing as opposed to enhancing performances of the micro-grooved journal bearing given in the literature. The reason for the higher frictional values occurred on the micro-grooved bearings can be attributed to a higher disruption of the oil flow on the micro-grooved bearings as compared to non-grooved bearing. Fig. 9 depicts the variation of coefficients of friction with crank angle for four types of bearings (plain and grooved) under the effect of dynamic bearing load. It was seen from this figure that the smallest value of the coefficient of friction was exhibited on the plain journal bearing while the largest one was obtained from the transversal grooved journal bearing similar to the distribution of the friction force, as expected. The maximum values of coefficient of friction obtained from the experiments were 0.0036, 0.0024, 0.0021 and 0.0015 for the transversally, the herringbone, and the circumferential grooves, and for the plain journal bearings, respectively. The increasing ratios for the coefficient of
H. Adatepe et al. / Tribology International 58 (2013) 12–19
was 10.16 N at 1001 crank angle for the finite difference method. On the other hand, the highest friction force obtained by the Schaffrath method at 1501 crank angle was 11.29 N while 11.63 N was determined for the plain bearing at 82.51 crank angle.All results for the coefficient of friction obtained from the experimental and the numerical studies were then demonstrated in Fig. 11. It was clear from the figure that the values for the coefficients of friction were very close to each other, especially for those obtained from the mobility and finite difference methods, as well as the corresponding crank angles (between 1501 and 1701). The exception was occurred for the maximum value of the friction coefficient obtained by the Schaffrath method, similar to
Friction Coefficient, µ
friction with respect to the plain journal bearing were approximately 1.4, 1.6 and 2.4 for the circumferential, the herringbone, and the transversally micro-grooved bearings, respectively. These numbers indicated that all the experiments on the journal bearings took place within the liquid friction zone. Therefore, Fig. 9 illustrated that with the same dynamic load applied, the micro-grooved journal bearings yielded a higher coefficient of friction than that of non-grooved bearing as opposed to those given in literature. This disagreement can be attributed to the fact that studies were performed in different working conditions. For example, motors work in mixed lubrication zone during startup and stopping motions in which metal-to-metal contact could be more possible to occur due to starved lubrication conditions. The micro-grooves, in that case, help oil to easily lubricate the surface of the journal bearing in mixed lubrication zone. It can, therefore, be concluded that micro-grooves on the surface of journal bearings can be used to improve the tribological behavior of the journal bearings under starved lubrication conditions. On the other hand, the experiments and the numerical studies, in our case, were pursued in the full film lubrication (the liquid friction) zone so that there was no metal-to-metal contact and the microgrooves struck the lubricating oil to flow on the surface of microgrooved bearings. It can, therefore, be deduced that the operating condition plays an important role on the performance of microgrooved journal bearing. When Figs. 6, 8, and 9 were evaluated together, it could be concluded that the friction force and the coefficient of friction reached to peak values at the crank angle in which the bearing load was also maximum (in the vicinity of 1001). Since the bearing load is dynamics, it is possible to have different crank angles for the peak values of the friction force and the coefficient of friction. As for the theoretical study, the tribological behavior of the plain journal bearing was determined by using ORBIT software which utilizes two numerical methods, the mobility method and the finite difference methods. Moreover, another computational solution was determined by using the Schaffrath method. The friction forces obtained numerically and experimentally were then plotted at the same figure for comparison purpose, shown in Fig. 10. It was found that the numerical and the experimental results obtained in the full film lubrication zone were in good agreement. It was seen that the highest friction force was 9.94 N at 1001 crank angle obtained by the mobility method while it
17
Fig. 11. Variation of friction coefficient under the same dynamic bearing load obtained from numerical and experimental analyses.
0°
1
2
270° 270°
360°
270° 360° 180° 270°
0.2
0.4 360°
0.6
0.8
1.0 30.000 N
180° 180°
90° 90° 90°
4
3 180° Mobility Method Finite Difference Method Schaffrath Method Polar Load
Fig. 10. Variation of friction force under the same dynamic bearing load obtained from numerical and experimental analyses.
Fig. 12. Polar load diagram for the plain journal bearing and journal center orbits obtained from different numerical methods.
18
H. Adatepe et al. / Tribology International 58 (2013) 12–19
the case in the friction force. It is normal to obtain the peak values of the friction force and the coefficient of friction at the different crank angles due to the nature of the dynamic bearing load. However, although the crank angles corresponding to the peak friction forces differed from each other, the variations of the friction force through one cycle of the crank angle looked similar; i.e., increasing and decreasing regions almost coincided. This was held true for the coefficient of friction more strongly. The orbits drawn by the shaft center based on three different computational methods and polar load diagram for the dynamically loaded plain journal bearings were shown in Fig. 12. The figure showed that the orbits drawn by the shaft center complied well with each other. The small discrepancies might be caused by the different computational method used. It was also seen from the figure that the maximum eccentricity ratio existed for two methods, namely; mobility and finite difference methods, was completely coincided at the position angle of 1871 with a value of 0.76 while the Schaffrath method had a very small divergence from these two methods having the value of 0.79. The graph plotted in order to determine the relationship between the experimentally obtained coefficients of friction on the plain journal bearing and the minimum oil film thickness obtained from different numerical methods was given in
Fig. 15. Variation of minimum oil film thicknesses with respect to crank and position angles.
Fig. 13. This figure illustrated that when the oil film thickness decreased the coefficient of friction increased and wherever the coefficient of friction reached to maximum value the oil film thickness was attained to minimum value. In Fig. 14, the variation of minimum oil film thickness against the bearing load was shown. It was demonstrated that the minimum oil film thicknesses were found to be 6.17 mm, 6.11 mm, and 5.04 mm by utilizing the mobility, the finite difference and Schaffrath methods, respectively. They all existed at around crank angle of 1411. Moreover, it was seen that the crank angle position of the minimum oil film thickness, was just occurred 43.51 beyond to that of the maximum bearing load (97.51). The variation of minimum oil film thickness versus the crank and the position angles was illustrated in Fig. 15. This figure showed that the smallest oil film thickness was just occurred at the position angle of 1861.
5. Conclusions
Fig. 13. Variation of coefficients of friction and minimum oil film thickness obtained from different numerical methods.
Fig. 14. Variation of minimum oil film thickness with respect to bearing load.
Before designing micro-grooved bearings, it is necessary to predict performances of the micro-grooved bearings accurately. We, therefore, investigated the effects of micro-grooves on the tribological behavior of the journal bearing within liquid friction zone using the purpose-built friction test rig in this study. The grooves in micron scale basis were made onto the plain journal bearings as circumferential, transversal and herringbone shapes and they were then subjected to dynamic load. This investigation yielded the following results: 1. The experiments showed that the lowest value of friction force was determined on the plain journal bearing. The experimentally obtained values increased progressively for the circumferential and the herringbone micro-grooved bearings and, finally, the highest frictional force distribution was exercised on the transversal micro-grooved bearing. 2. The shape of micro-groove affects the coefficient of friction, attaining the lowest value to the plain journal bearing and resulting in the maximum value for the transversally microgrooved journal bearing, similar to the trend of the friction force. 3. The friction force distributions as well as the variations in the coefficient of friction obtained from the different numerical methods for the plain journal bearing were very similar and agreed well with each other and also with those of the experimentally obtained results.
H. Adatepe et al. / Tribology International 58 (2013) 12–19
4. The minimum oil film thickness value appeared at the furthest position from the point of the maximum bearing load value existed. In conclusion, the results of this research showed that the shapes of micro-grooves played an important role on the oil film thickness, the coefficient of friction, and the friction force under the dynamic load. This study further indicated that the microgrooves developed the worse tribological properties on the journal bearings when they operated in the liquid friction zone than the non-grooved journal bearings as opposed to knowledge written in the literature. It can, therefore, be concluded that the use of generalized results regardless of shape of micro-groove and operating condition may result in erroneous estimates of the tribological properties of the micro-grooved journal bearing. It is, therefore, recommended that performing detailed investigation in conjunction with shape, depth and operating condition is necessary to obtain reliable data regarding the tribological properties of the micro-grooved journal bearing.
Acknowledgment This work was supported by the Research Fund of Karadeniz Technical University, Project no. 2002.112.003.03. References [1] Hu J. Experimental and theoretical investigation of roughness effects on thin laminar fluids films. PhD thesis. University of Toronto, Canada; 1997. [2] Zhang C, Qiu Z. Effects of surface texture on hydrodynamic lubrication of dynamically loaded journal bearings. Tribology Transactions STLE 1998;41: 43–8. [3] Christensen H, Tonder K. The hydrodynamic lubrication of rough bearing surfaces of finite width. Journal of Lubrication Technology—Transactions of the ASME 1971;93:324–30. [4] Christensen H, Tonder K. The hydrodynamic lubrication of rough journal bearings. Journal of Lubrication Technology—Transactions of the ASME 1973;95:166–72. [5] Hata H, Nakahara T, Aoki H. Measurement of friction in lightly load hydrodynamic sliders with striated roughness, The Winter Annual Meeting of the ASME, Chicago, Illinois, 75–92; 16–21 November 1980. [6] Nakahara T, Takesue M, Aoki H. Effects of surface roughness and bearing slenderness ratio on hydrodynamic lubrication. Journal of JSLE 1983;28: 543–8.
19
[7] Nakahara T. Effect of truncation in surface roughness on hydrodynamic pivoted slider and practical expression of truncation. In: Andersson P, Ronkainen H, Holmberg K, editors. 9th NORDTRIB, 3; 2000. p. 935–44. [8] Kumada Y, Hashizume K, Kimura Y. Performance of plain bearings with circumferential microgrooves. Tribology Transactions STLE 1996;39:81–6. [9] Hargreaves DJ, Armatys D. Performance of a microgrooved journal bearing under steady and dynamic loading. Tribo-Test 1999;5:277–86. [10] Watanabe K, Natsuma J, Hashizuma K, Ozasa T, Noda T, Masuda Y. Theoretical analysis of bearing performance of microgrooved bearing. JSAE Review 2000;21:29–33. [11] Roy L. Thermo-hydrodynamic performance of grooved oil journal bearing. Tribology International 2009;42:1187–98. [12] Hirayama T, Yamaguchi N, Sakai S, Hishida N, Matsuoka T, Yabe H. Optimization of groove dimensions in herringbone-grooved journal bearings for improved repeatable run-out characteristics. Tribology International 2009;42:675–81. [13] Ricardo Software, Journal Bearing Analysis-Orbit, Documentation/User Manual Version 1.2, USA; 2000. [14] Schaffrath G. The lubricating bearings with arbitrary gap shape-shifting of the shaft journal with time-varying load. PhD thesis. Technical University, Karlsruhe, Germany (in German); 1967. [15] Harbordt J, Klumpp R. Computer program for calculating the displacement train, the local pressures and the factual gap lubricating bearings in arbitrary geometry under static and dynamic load, Research Reports Internal Combustion Engine Projects, No. 98, Issue 137; 1972 (in German). [16] Bıyıklıoglu A. Surface fatigue on dynamically loaded journal bearings. PhD thesis. Karadeniz Technical University, Trabzon, Turkey; 1986 (in Turkish). [17] Durak E, Kurbanoglu C, Bıyıklıoglu A, Kaleli H. Measurement of friction force and effects of oil fortifier in engine journal bearings under dynamic loading conditions. Tribology International 2003;36:599–607. [18] Bıyıklıoglu A, Cuvalci H, Adatepe H, Bas H, Duman MS. A new measurement test apparatus and method for the friction force measurement in the journal bearings under dynamic loading Part I. Experimental Techniques 2005;29: 22–5. [19] Bıyıklıoglu A, Cuvalci H, Adatepe H, Bas H, Duman MS. A new measurement test apparatus and method for the friction force measurement in the journal bearings under dynamic loading Part II. Experimental Techniques 2005;29: 33–6. [20] Adatepe H. The investigation of the friction behavior of the microgrooved journal bearings under static and dynamic loading conditions. PhD thesis, Karadeniz Technical University, Trabzon, Turkey; 2006 (in Turkish). [21] Adatepe H, Bıyıklıoglu A, Sofuoglu H. An experimental investigation on frictional behavior of statically loaded micro-grooved journal bearing. Tribology International 2011;44:942–8. [22] Booker JF. Dynamically loaded journal bearings: mobility method of solution. Journal of Basic Engineering—Transactions of the ASME 1965;4:537–46. [23] Booker JF. Dynamically loaded journal bearings: numerical application of mobility method. Journal of Lubrication Technology—Transactions of the ASME 1971;93:168–76. [24] Goenka PK. Analytical curve fits for solution parameters of dynamically loaded journal bearings. Journal of Tribology —Transactions of the ASME 1984;106:421–8.