1、04FTM6The Effect of a ZnDTP Anti-wearAdditive on Micropitting Resistance ofCarburised Steel Rollersby: C. Benyajati and A.V. Olver, Department of MechanicalEngineering, Imperial College LondonTECHNICAL PAPERAmerican Gear ManufacturersAssociationThe Effect of a A ZnDTP Anti-wear Additive onMicropitti
2、ng Resistance of Carburised Steel RollersC. Benyajati and A.V. Olver, Department of Mechanical Engineering, ImperialCollege LondonThe statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Associ
3、ation.AbstractZinc di-alkyl dithio-phosphate (ZnDTP) compounds are widely used in engine and transmission oils both asanti-oxidants and as anti-wear additives. However, recent work has shown that many anti-wear additivesappear to have a detrimental effect on the resistance of gears and other contact
4、ing components to varioustypes of rolling contact fatigue, including micropitting. In the present paper we examined the effect of thepresence of a secondary C6 ZnDTP in a low viscosity synthetic base oil on the resistance to micropitting andwear of carburised steel rollers, using a triple-contact di
5、sk tester.It was found that the additive caused severe micropitting and associated wear, whereas the pure base oil didnotgiverisetoanymicropitting.Itwasfurtherfoundthattheadditivewasnotdetrimentalunlessitwaspresentduring the first 100 000 cycles of the test when it was found to exert a strong effect
6、 on the development ofroughness on the counter-rollers. It is concluded that the additive is detrimental to micropitting resistancebecause it retards wear-in of the contact surfaces, favouring the development of damaging fatigue cracks.This contrasts with some earlier speculation that suggested a di
7、rect chemical effect could be responsible.Copyright 2004American Gear Manufacturers Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314October, 2004ISBN: 1-55589-829-7The effect of a ZnDTP anti-wear additive on the micropitting resistance of carburised steel rollers C Benyajati an
8、d A V Olver Tribology Section, Department of Mechanical Engineering, Imperial College London, Exhibition Road, London, SW7 2BX, UK 1. INTRODUCTION Micropitting is a microscopic form of rolling contact fatigue and wear, which is most often found in ground, hard steel surfaces such as those in case-ha
9、rdened gears. It is normally associated with concentrated rolling-sliding contacts under conditions where the lubricant film is rather thin compared to the height of the surface roughness. Generally, micropitting consists of numerous small, shallow pits on the surface, each pit often having the char
10、acteristic dimension of some tens of micrometers. Despite the small size of individual pit, these pits together can form a large total area of damage so that the surface has a grey, matte appearance. For this reason, micropitting is also sometimes known as grey staining and frosting in the gear indu
11、stry. Micropitting, like other wear mechanisms, has a detrimental effect on the durability of the component. It can cause a significant material loss from the damaged surface. In gears, this can result in the loss of profile of the teeth, which may then generate an increase in noise, vibration, and
12、dynamic loads. In more severe cases, it can even cause the total fracture of the gear teeth. Furthermore, the progression of micropitting may eventually result in large scale (macro-) pitting. It may also develop into other types of surface damage, such as scuffing. Micropitting therefore represents
13、 a potentially severe failure mode. It has caused appreciable problems in service and has received considerable attention from tribology researchers 1-4. Recently, a number of works have investigated the effect of lubricant chemistry on micropitting 5,6. Cardis and Webster 6 showed that a number of
14、additives described as anti-wear promoted micropitting. However, no details of the additives used were given and the mechanism by which the presence of the additives led to micropitting was not investigated. Here, the results of a study of the effect of lubricant composition are reported. A PAO base
15、 stock was tested on its own and with a ZnDTP anti-wear additive, which is a common constituent of engine and transmission oils. The experimental results showed that the anti-wear additive has an unfavourable effect on micropitting. A possible new mechanism that might account for the results is disc
16、ussed. Some preliminary results from the present study were presented in 7. 2. TEST RIG In the present study, a three-contact disc machine has been employed for the study of the effect of lubricants on micropitting. The test roller is a cylinder, of diameter 12.0 mm which is run against three counte
17、rface rollers (rings) which have chamfered edges forming the 3.2 mm track width. Therefore, a line contact is generated between a roller and rings. In addition, both specimens have a circumferential finish at their outer diameters. A dip lubrication system is employed as a mean to supply lubricant i
18、nto contacts. Furthermore, a loading method is provided by the use of a motorised ball-screw, acting through a loading arm. A schematic diagram of the test rig is shown in Figure 1, more details on test rig development and its features can be found in 7. Figure 1. Schematic of test rig, showing the
19、contact configuration and a principle of load application Test rollerStrain gauged loading arm Pivot Load from ball-screwCounterface rollers 1 3. EXPERIMENTAL METHOD A test roller and three counterface rings are employed for each test. The general geometries of both roller and rings are shown in fig
20、ure 2. Both specimens were manufactured from 16MnCr5 steel to DIN 17210, commonly used in gear industries. Both specimens were gas carburised and heat treated to the specified hardness of 660-700 HV and 730-770 HV for roller and ring respectively with minimum case depth of 0.9 mm for both specimens.
21、 The average roughness of each specimen was Ra= 0.35 m and 0.5 m for roller and rings respectively. 10 54.15 25.5 3.2 8Ring specimen Roller specimen 1212.8 Figure 2. General geometries of ring and roller specimen (dimension in mm, not to scale) A poly-alpha-olefin (PAO) base stock was employed in th
22、is work. The viscosity of the lubricant was 13.7 cP at 40 C and was 2.9 cP at 100 C. The base stock was tested on its own and also with anti-wear additive present. The additive use in this paper was a secondary C6zinc dialkyl-dithio-phosphate (ZnDTP) dissolved in a high-viscosity base. The treat rat
23、e was 1.3% to give 0.1% Phosphorus by mass. For all tests, the roller, the rings, and assembly components including the damper rings were ultrasonically cleaned in solvent before mounted on the shafts. After assembly, the test started by heating up the test oil to the specified temperature. Next, th
24、e load was gradually applied at a uniform rate over 5 minutes. When the load reached the required value, the electric motors applied the required rolling and sliding speeds to roller and rings. The rig would then run for a pre-determined period. Except where stated, all the tests in this paper were
25、carried out under the same operating condition, which was 70 C bulk oil temperature, under a Hertz pressure of 1.7 GPa, and a slide-roll ratio ( v /v ) of 5.2 % (roller surface slower) with v = 3.15 m/s. (Here, v is the difference in speed and v is the mean speed of the surface with respect to the c
26、ontact). Under these operating conditions, using a simple thermal model 8, the surface temperature at the inlet of contact was predicted to be 76.2 C. At this temperature, the corresponding viscosity and pressure viscosity coefficient of the base stock was 4.77 cP and 11.2 GPa-1respectively. The cor
27、responding smooth-body minimum film thickness was predicted to be 61 nm. The tests were stopped at pre-determined intervals for specimen inspection. The inspection process involved measuring the surface profile across the wear track and taking images of the surface using a Form Talysurf Series 2 sty
28、lus profiler with 0.8 mm cut-off length and a digital microscope respectively. For the surface profile measurement, four measurements were made in axial direction of the roller across the wear track. An example of a worn surface profile is shown in figure 3. It can be seen that the wear is greater n
29、ear the edges of the track. This may be due to the edge-pressure from the chamfer corners of the rings. Only the central part of the running track was used for the wear determination. 0.5 mm5 mReference linewear depthFigure 3. An example of worn surface profile, showing determination of effective we
30、ar depth The loss in diameter of the roller against test time was employed as the means to evaluate micropitting wear. From the profile in figure 3, the wear depth was estimated by fitting a reference mean line through the unworn surface profile. The distance between this reference line and the cent
31、ral part of wear track was taken as the wear depth. 2 4. RESULTS Upon closer inspection with the microscope, it was found that the surface generally had become fairly smooth with a few pits on it. There was also a newly formed layer, which appeared to be dark purplish and bluish in colour, on the su
32、rface of the worn roller. 4.1 Tests with PAO base stock Resulting wear curves from the tests with PAO on its own are shown with a dashed line in figure 4. There are two plots corresponding to each of two tests under the same conditions. The appearance of the wear track at the end of the test (4.4 mi
33、llion contact cycles) is shown in figure 5(a) and 5(b). 010203040500.01 0.1 1 10Number of contact cycles (million)Loss of diameter ( m)0PAOPAO + ZDDPPAO + ZDDP introduced after run-in with PAOZDDP removed after initial periodAdditive introducedAdditive removedFor the tests with PAO only, it was foun
34、d that micropitting wear did not occur on the roller specimen but there were a few surface cracks. However, despite the absent of micropitting, a little initial wear was presented. It can be seen from the wear curve that the roller was already worn at an early stage of the test but the progression o
35、f wear soon stopped so that there was no further loss in diameter through out the rest of the test. Figure 4. Evolution of wear as a function of contact cycles during the tests (a) (b) 200 m PAO test 200 m PAO + ZnDTP test(c) (d) Figure 5. Images of the tested roller after 4.4 millions contact cycle
36、s (5 hours), (a) general appearance from PAO test, (b) microscopic image shows no micropitting, (c) general appearance from PAO + ZnDTP test compared with untested roller, (d) microscopic image shows micropitting covers the entire running track 3 4.2 Test with PAO + ZnDTP For the test when 1.3% ZnDT
37、P were added into the PAO, extensive micropitting occurred on the roller specimen. The resulting wear curves are shown in figure 4 with a solid line. They show a fairly steady wear rate. The general appearance of the micropitted roller is shown in figure 5(c) along with a normal roller for compariso
38、n. It can be seen that the micropitted specimen has a grey and matte appearance on the wear track compared to a normal surface. A microscopic image of the micropitted surface is shown in figure 5(d); numerous micropits on the surface can be seen clearly from the image The transverse roughness of thr
39、ee counterface rings was measured using a stylus instrument. The roughness of the counterface was found to decline gradually during the test. This is shown graphically in figure 6 where each point represents the averaged value of Raroughness over three rings. The plots show a similar trend between t
40、wo sets of tests such that there was a significant drop in the roughness value at an early stage followed by a very slow decline for the rest of the test. However, the main difference between the results is that when ZnDTP was present, the roughness declined more slowly as can be seen from the plot.
41、 Apart from carrying out the tests with ZnDTP under those operating conditions explained in section 3, an additional test was carried out with the same additive in order to examine the effect of slide-roll ratio on micropitting. In this test, v /v was increased to 15% and the bulk oil temperature wa
42、s reduced to 55 C while the other parameters were kept the same as the previous tests. Under these operating conditions, the thermal model 6 predicted a surface temperature at the inlet of contact to be approximately the same as the previous tests and hence a similar smooth-body minimum film thickne
43、ss. By increasing v /v , it was found that extensive micropitting also occurred on the roller specimen. The resulting wear curve is shown in figure 7(a) together with those from lower v /v tests. The curve shows a higher wear rate compared to the lower /v v tests. Furthermore, the general appearance
44、 and microscopic image are similar to those shown in 5(c) and 5(d) respectively. For the transverse counterface roughness, the measured values are shown in figure 7(b). The plot also shows a similar trend to those of lower /v v tests. 0.20.30.40.50.60.01 0.1 1 10Number of contact cycles (million)Cou
45、nterface roughness ( m)0Additive introducedPAOPAO + ZnDTPZnDTP removed after initial periodAdditive removedPAO +ZnDTP introduced after run-in with PAO0.5 m0.2 mm0.5 m0.2 mmPAO0.5 m0.2 mm0.5 m0.2 mmPAO + ZnDTPCounterface roughness ( m)Figure 6. Variation of counterface transverse roughness (Ra) durin
46、g the test for PAO only and with ZnDTP tests. In general, the roughness of the counterface decline gradually during the test. Typical surface profiles before, during, and after the test are shown on the top and on the right hand side of the plot respectively. 4 0102030405060700.01 0.1 1 10Number of
47、contact cycles (million)Loss of diameter ( m)05.2% SRR, 70oC15% SRR, 55oCLub: 1.3% ZnDTP + PAO0.20.30.40.50.60.01 0.1 1 10Number of contact cycles (million)Counterface roughness ( m)05.2% SRR, 70oC15 % SRR, 55oCLub: 1.3% ZnDTP + PAOFigure 7. Effect of slide-roll ratio ( /v v ) on micropitting, (a) r
48、esulting wear curve, (b) variation of counterface transverse roughness (Ra) 4.3 Effect of running-in with base stock Another test was carried out in order to determine the cause of the difference between the results of base stock only and with ZnDTP. In this test, the specimen roller and counterface
49、 rings were run with PAO only, for 5 minutes (73,000 contact cycles). After the routine inspection on the specimens, ZnDTP additive was introduced into PAO and the test was then carried on in the same manner as the previous ones. The wear curve obtained from this additive introduced test is shown in figure 4. On the graph, the dashed line represents the period when only PAO was present whereas the solid line indicates the period when ZnDTP had been introduced into the PAO. The curve shows that a little wear occurred at an earl
