Sliding wear tests of chilled cast iron tappet-disc and cam-ball tribo-partners were conducted by using ball on disc testing device. Magnetron sputtering CrTiAlN hard coating and Graphite like carbon (GLC) solid lubricant coatings were applied on either one or both of the contact surfaces on tribo-partners. Ball crater device and scanning electron microscope were used to investigate surface wear of rubbed track on disc and wear scar on ball. The sliding performances of tribo-partners were evaluated in terms of coefficient of friction at sliding surfaces and specific wear rate (SWR) on both of the rubbed surfaces under specific test conditions. All test results showed that GLC solid lubricant coating on both surfaces of tappet-disc and cam-ball was the best option to allow this tribo-partners having high performance in terms of sliding under high load with low coefficient of friction and low SWR. The advantage of using GLC solid lubricant coating is to enable mechanical parts made of ordinary and sustainable materials (e.g. cast iron or plain steels) to be operated under even harsher conditions of higher load and higher wear environment with improved performance.
Chilled cast iron materials have been recognized with combined properties of surface hardness and body ductility [
Hard coatings, such as CrTiAlN [
There were number of technologies reported [
Carbon based, e.g. polycrystalline diamond, diamond like carbon (DLC) and amorphous graphite like carbon (GLC) coatings [
Magnetron sputtering carbon coating was reported to have high hardness and low coefficient of friction [
Carbon surface was recognized [
This paper presents evaluation of sliding performances of using CrTiAlN hard coating and GLC solid lubricant coating on chilled cast iron tappet-disc and cam-ball tribo-partners under identical testing conditions. Comparison of testing results was made to provide an optimum system of coating to tribo-partners from which the best sliding performance was achieved.
All test samples were made to form as ball against disc tribo-partners. Discs of Ø30 × 3 mm were cut from chilled cast iron tappets (normally used for heavy truck). The surface of disc used for the test was the same surface on tappet to contact camshaft. Ball heads (Ø5 mm cylinder in 20 mm length with one end machined as half ball surface with Ø5 mm) were cut from chilled cast iron cams in which the half ball surface was the same surface on cam to contact tappets. Hardness of chilled cast iron was about HRC 43 in average and roughness of machined surface on disc and ball was about 0.13 µm in Ra as measured by a roughness measuring device (SJ.400 Mitutoyo, Japan Manufacture) with average error of 0.02 µm.
CrTiAlN hard coatings and GLC solid lubricant coatings were deposited on surfaces of those prepared discs and ball heads using unbalanced magnetron sputtering technology with deposition method described elsewhere [
The deposition procedure of CrTiAlN hard coating was started with ion cleaning of using high bias potential (−500 V) on substrate and low power (50 W) on all 4 magnetrons in an Ar pressure of 3.0 × 10−3 Torr. Then a deposition of Cr bond layer was conducted of using low bias potential (−80 V) on substrate and high power (5.0 kW) on 2 magnetrons of Cr targets. Afterwards, N2 was introduced to deposit CrN hard coating and quantity of N2 was feedback controlled by an optical emission monitor (OEM) set at 50% of Cr optical emission intensity. And then the powers of other 2 magnetrons (Ti and Al targets) were increased from 50 W to 6.0 kW to deposit CrTiAlN multilayer hard coating.
The initial procedure of GLC coating was the same as that of CrTiAlN process but in which targets of Ti and Al were replaced by C (graphite) targets. After deposition of Cr bonding layer, 2 magnetron powers of C targets were increased from 50 W to 4.0 kW whilst magnetron power of Cr targets were reduced from 5.0 kW to 50 W to deposit GLC coating in Ar pressure of 3.0 × 10−3 Torr without using reactive gases.
The process temperature for CrTiAlN coating was about 300˚C and the temperature for GLC coating was about 200˚C as indicated from a thermal couple in the vacuum coating chamber during deposition process.
The total coating thickness of both CrTiAlN hard coating and GLC solid lubricant coating was 3.0 µm. CrTiAlN coating was formed as nano-scale multilayer with poly-crystallite of CrN based microstructure in which each individual single layer was identified with a thickness of about 6 nm [
The adhesion of coatings was characterized using a Rockwell-C indentation (150 kg) and a standard diamond tip Scratch (up to a load of 60 N) tests on the coated surface as reported by [
Ball on disc wear test device [
A ball (Ø 30 mm) crater device [
The specific wear rate (SWR) of tested disc was calculated from measured dimensions of wear track combined with testing parameters as expressed by SWR = (Wear Volume)/(Load * Sliding Distance) in which:
WearVolume = [ ( w ∗ t ) / 2 ] ∗ π ∗ d
w and t present wear track width and coating wear depth respectively whilst d is the diameter of wear track set up in test, π ∗ d is the Circumference of wear track that multiples ( w ∗ t ) / 2 (approximate section area of wear track) to get Wear Volume.
Sliding distance = Circumference * Number of Revolutions in which Number of revolutions is set up in test. So that the SWR of coating can be written as Formula (1):
SWRofcoating = ( w ∗ t ) / ( Load ∗ 2 ∗ NumberofRevolutions ) (1)
Formula (1) is based on the test situation in which wear depth is less than that of coating thickness. If the coating is completely worn off during the test, then SWR is a combined wear of both coating and substrate. In general, SWR of coating is very much less than that of chilled cast iron substrate, and therefore, SWR is normally treated as substrate wear if the coating is prematurely worn off during test.
The SWR of ball head is expressed by Formula (2).
SWRofballhead = ( CapVolume ) / ( Load × SlidingDistance ) (2)
In which Sliding Distance is π*d*Number of Revolutions, Cap volume (CV) is the worn off volume on the rubbed ball head and can be written as Formula (3).
C V = ( 2 / 3 ) π R 3 { 1 − [ 1 − ( d ′ / 2 R ) 2 ] } − ( 1 / 3 ) π ( d ′ / 2 ) 2 { [ R 2 − ( d ′ / 2 ) 2 ] } (3)
In which d ′ is diameter of the wear scar on ball and R is the radius of the ball head.
A scanning electron microscope (SEM) was used to analyse surface morphology of those rubbed tracks on discs and wear scars on ball heads. Surface composition of the rubbed track and wear scar was analysed by energy dispersive X-Ray (EDX) provided in the SEM. The energy of electron beam was adjusted by voltages in SEM so that penetration depth of electrons into surface materials to make X-ray emission would be modified by energy level of those penetrated electrons. Therefore coating composition could be selected from surface materials rather than that from body substrate. The coating had a thickness of about 3.0 micrometres whilst X-ray emission region was able adjusted at about 2 micrometres in depth. In the case of some coating was worn off, e.g. left about 1.0 micrometre on body surface, then the composition as measured by EDX was contributed from both coating and substrate materials.
SWR, test parameters and measured data are shown in
In general, all surfaces of coated parts appeared to show significant reduction of wear comparing to uncoated parts. But GLC solid lubricant coating outperformed CrTiAlN hard coating in terms of whole system performance improvement of the tribo-partners, e.g. tribo-partners sliding with very low coefficient of friction and retaining with very low SWR on both contacted surfaces.
Both disc and ball counterpart surfaces were encountered extensive wear. The wear depth of rubbed track on disc was about 12.0 µm which was combined to track width and test parameters to give the highest SWR (380.5 × 10−16 m3/N×m) among all the tests. The wear scar on ball was a flat section with diameter of about 1366.0 µm to correspond a high ball head SWR of 97.4 × 10−16 m3/N×m.
Test parameters: disc diameter 30 mm, ball head diameter 5 mm, sliding speed 100 mm/s, track diameter 12 mm, time 60 min, load as indicated | ||||||||
---|---|---|---|---|---|---|---|---|
Option of Tribo-partners | Test | Measurement data | SWR (×10−16 m3/N×m) | |||||
Chilled cast iron tappet disc | Chilled cast iron cam ball head | Load (N) | Coefficient of friction (1) | Wear track width (µm) | Wear track depth (µm) | Ball wear scar Ø (µm) | Disc | Ball |
No coating | No coating | 20 | 0.60 - 0.90 | 1220.0 | 11.9 | 1366.0 | 380.5 | 97.4 |
CrTiAlN | No coating | 20 | 0.40 | 1300.0 | <0.1 | 1375.0 | 3.4 | 100.0 |
No coating | CrTiAlN | 20 | 0.40 - 0.80 | 914.0 | 5.8 | 1000.0 | 138.9 | 27.6 |
CrTiAlN | CrTiAlN | 20 | 0.20 - 0.50 | 400.0 | <0.1 | 885.7 | 1.0 | 17.0 |
GLC | No coating | 20 | 0.04 | 1000.0 | <0.1 | 1150.0 | 2.6 | 48.6 |
No coating | GLC | 20 | 0.07 | 400 | 2.1 | 659.0 | 22.0 | 5.2 |
GLC | GLC | 20 | 0.04 - 0.06 | 100 | <0.1 | 500.0 | 0.3 | 1.7 |
GLC none | No coating | 60 | 0.05 - 0.06 | 600 | <0.1 | 1000.0 | 0.5 | 9.2 |
No coating | GLC | 60 | 0.18 - 0.75 | 1500.0 | 5.7 | 1600.0 | 74.7 | 61.7 |
GLC | GLC | 60 | 0.02 - 0.06 | 625 | <0.1 | 620.0 | 0.6 | 1.4 |
The high wear on both contacted surfaces was correlated to high coefficient of frictions (0.6 to 0.9) in cyclical sliding in which strong interaction was taken places among wear debris and sliding contact interfaces [
Item (A%) | Fe | Si | O | C |
---|---|---|---|---|
P 1 | 25.2 | 1.0 | 54.6 | 19.3 |
P 2 | 28.6 | 1.4 | 54.8 | 15.2 |
SWR of coated disc was significantly reduced for the case of tribo-partners with CrTiAlN hard coating on disc as comparing to the case of without coating. The wear depth of rubbed coating surface was hardly to measure, but 0.1 µm was given as an estimation and based on this assumption, SWR of CrTiAlN coated disc was <3.4 × 10−16 m3/N×m, about 100 times less than the case of without coating.
The coefficient of friction was moderate high at about 0.4 and CrTiAlN coating was able to tolerate this friction for one hour at almost no wear to indicate its exceptional wear resistance. Similar work was previously reported [
The ball head counterpart was, however, encountered very high wear. The wear scar was about 1400 µm in diameter with a flat section that corresponded to a SWR of 100.0 × 10−16 m3/N×m, the highest ball wear among all tests. The ball was not coated and its hardness (about 500 HV) was much lower than that (>3000 HV) of its opposite CrTiAlN coated disc. In addition those ball wear debris were abrasive that reacted to sliding interface at ball counterpart to cause both abrasive and adhesive wear. Therefore, huge wear was taken place on the ball head surface leading to sliding performance down of the tribo-partners although CrTiAlN coated disc was not encountered much wear.
sliding test although its opposite ball head was encountered extensive wear. Ball wear debris were accommodated on the rubbed coating surface to form network plateaus. Therefore, sliding interfaces were mainly chilled cast iron against chilled cast iron with little involving of the underneath coating. There should be very high stress transferred to the local coating surface during sliding but the underneath coating surface was able to support sliding interface with no plastic deformation, cracking, delamination and noticeable wear to indicate high resistance to stress induced fatigue, abrasive and adhesive wear.
EDX compositions at pointed areas in
SWR of rubbed track on disc was 138.9 × 10−16 m3/N×m and SWR of CrTiAlN coated ball head was 27.6 × 10−16 m3/N×m. The wear reduction of CrTiAlN coated ball head was not as sufficient as that of CrTiAlN coated disc. This was because the coted ball head surface was in ‘‘point’’ contact to sliding disc and continuously encountered high stress and high shear force. Coating at this area was, therefore, undergoing cracking, fracturing and spalling failures. And then, the coating on ball head counterpart surface was completely worn off resulted in sliding interfaces as chilled cast iron against chilled cast iron with high coefficient of friction and high SWR.
Types | Point | Fe | Cr | Ti | Al | Si | O | N | C |
---|---|---|---|---|---|---|---|---|---|
CrTiAlN on disc | 1 | 16.7 | 15.5 | 1.8 | 0.0 | 0.0 | 56.3 | 0.0 | 9.7 |
2 | 1.5 | 33.6 | 3.1 | 5.7 | 0.0 | 18.1 | 38.0 | 0.0 | |
3 | 14.5 | 0.0 | 0.0 | 0.0 | 0.0 | 43.8 | 0.0 | 41.7 | |
4 | 64.8 | 0.0 | 0.0 | 0.0 | 2.3 | 0.0 | 0.0 | 32.9 | |
CrTiAlN on ball | 1 | 58.5 | 0.0 | 0.0 | 0.0 | 2.0 | 0.0 | 0.0 | 39.5 |
2 | 24.8 | 0.0 | 0.0 | 0.0 | 2.0 | 48.9 | 0.0 | 24.3 | |
3 | 25.9 | 0.0 | 0.0 | 0.0 | 1.0 | 58.0 | 0.0 | 15.1 | |
4 | 59.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 40.9 | |
CrTiAlN on both disc & ball | 1 | 0.0 | 34.9 | 3.4 | 6.2 | 0.0 | 14.1 | 41.4 | 0.0 |
2 | 0.6 | 36.3 | 3.2 | 4.9 | 0.0 | 5.9 | 49.1 | 0.0 | |
3 | 42.8 | 8.1 | 2.2 | 1.4 | 0.8 | 0.0 | 44.7 | 0.0 | |
4 | 79.2 | 2.2 | 0.0 | 0.0 | 2.0 | 0.0 | 0.0 | 16.6 |
SWR of rubbed track on coated disc was about <1.0 × 10−16 m3/N×m based on an estimation of wear depth of 0.1 µm and this SWR was 380 times less than the case of tested disc without coating. SWR of wear scar on coated ball was evaluated at 17.0 × 10−16 m3/N×m, about 6 times lower than the case of ball head wear without coating.
There was little wear taken places at rubbed coating surface on disc as the surface morphology was almost the same as its original surface. There was very little adhered wear debris on rubbed track of CrTiAlN coated disc to indicate wear debris from coated ball were less adhesive comparing to Fe based wear debris.
The diameter of wear scar on coated ball was about 2 times larger than the width of rubbed track on coated disc. It was because wear debris of coated ball head counterpart were accommodated along areas close to both edges of the rubbed track on coated disc. When the ball ran over of those abrasive wear debris, rubbing scar was formed on the ball due to continuously being crashed and pressed by abrasive debris whilst the track on disc was only periodically under sliding. In addition the surface of coated disc was very hard with ability to resist abrasive. Therefore wear scar on ball became larger due to more progress surface wear. The actual wear area on ball was up to about 350 µm in diameter as shown in
A Hertz contact stress [3f/(2 × π × r2), in which f and r presents normal load and radius of the wear scar respectively, and r was also assumed as the “true” radius of contact area on the ball] was estimated of about 312 MPa. This stress is about the level of yielding stress (≈400 MPa) for chilled cast iron raw materials. Although the ball was still under protection of the exist CrTiAlN coating, there was tendency that the coating on ball head counterpart surface would be soon worn off and then it could be expected to have extensive wear on the ball leading to sliding performance down of tribo-partners as discussed in 3.2.2.
Both SWR and coefficient of friction of the tribo-partners were very low when disc was coated by GLC solid lubricant film. The coefficient of friction was about 0.04 to 0.06 throughout testing and wear depth on disc was impossible to measure. Surface morphologies of rubbed track on GLC coated disc and wear scar on ball are shown in
The rubbed track on disc was about 1000 µm in width with no measurable wear depth as the rubbed surface retained its original morphology. But 0.1 µm was given as estimation of wear depth to specify a SWR < 2.6 × 10−16 m3/N×m for the rubbed track of coated disc, about 150 times less than case of without any coatings.
There were some small wear debris accommodated at those tiny-valleys on rubbed coating surface (point 1 in
The wear scar on ball was flat and smooth with a diameter of 1150 µm that corresponded to a SWR of 48.6 × 10−16 m3/N×m, about two times lower than the case of without any coating. This low SWR was due to very low coefficient of
Types | Point | Fe | Cr | Si | O | C |
---|---|---|---|---|---|---|
GLC on disc | 1 | 4.3 | 12.7 | 0.0 | 15.3 | 67.7 |
2 | 1.0 | 13.3 | 0.0 | 0.0 | 85.7 | |
3 | 97.2 | 0.0 | 2.8 | 0.0 | 0.0 | |
4 | 28.8 | 0.9 | 0.8 | 53.6 | 15.9 | |
GLC on ball | 1 | 68.1 | 0.6 | 1.6 | 0.0 | 29.7 |
2 | 34.4 | 1.1 | 1.9 | 62.6 | 0.0 | |
3 | 0.4 | 10.2 | 0.0 | 3.8 | 85.6 | |
4 | 42.4 | 4.2 | 1.1 | 0.0 | 52.3 | |
GLC on both disc & ball | 1 | 0.7 | 11.1 | 0.0 | 0.0 | 88.2 |
2 | 1.2 | 11.9 | 0.0 | 0.0 | 86.9 | |
3 | 15.9 | 18.3 | 0.0 | 6.8 | 59.0 |
friction at sliding interface provided by its opposite GLC coated partner. Low co- efficient of friction was responsible to low shear force and therefore the sliding wear was reduced accordingly.
High carbon was identified at some specific areas on ball counterpart, e.g. point 4 with dark grey colour in
The coefficient of friction during sliding was as low as about 0.06 to 0.07 throughout the testing. SWR of rubbed discs was <22.0 × 10−16 m3/N×m based on estimation of a wear track width of 400 µm. SWR of the GLC coated ball was estimated of <5.2 × 10−16 m3/N×m according to assumption of a flat wear section with diameter of 659 µm. The SWR of both rubbed disc and GLC coated ball was around at least 20 times less than the case of without any coating.
The rubbed track on disc was about 400 µm in width but main surface area in this track remained almost the same as its original morphology and both track edges were hardly to distinguish. There was a groove with width and depth about 100 µm and 2.1 µm respectively at the track centre to indicate the possible wear caused by this test. EDX analysis inside the groove (point 1 and 2 in
It can be seen from
indicated that the sliding interface was well supported by GLC solid lubricant coating and not much wear was taken place from the ball head surface. EDX analysis of the ball counterpart surface showed carbon based material to classify that GLC coating was still covering the rubbed scar. The surface was smooth with no sign of coating fractures on the contacted area implying that the shear force at sliding interfaces was not high enough to cause cracking, chipping and flaking at the rubbed surface on GLC coated ball head.
The coefficient of friction was as low as 0.04 to 0.06 throughout the testing. SWR of GLC coated disc and ball was estimated of about <0.3 × 10−16 m3/N×m and <1.7 × 10−16 m3/N×m respectively based on assumption of 0.1 µm wear depth on rubbed disc and flat section of 500 µm diameter on ball head counterpart. Both SWR on rubbed disc and ball head counterpart surfaces were the lowest among all the specific tests. Comparing to the cases of CrTiAlN coated and without any coating, GLC solid lubricant coating was the choice to promote sliding performance of tribo-partner system in terms of low coefficient of friction and low SWR on both sliding interfaces and GLC solid lubricant coating on both surfaces of the tribo-partners was the best option for an optimized sliding performances of the ball to disc tribo-partners.
If the white grey area in
GLC coating, e.g. coated components with ability to operate at very high load bearing capacity.
EDX compositions on rubbed surfaces were shown in
60 N high load was used to evaluate the sliding performance of GLC coated chilled cast iron tappet disc and cam ball Tribo-partners. The tribo-partners with CrTiAlN hard coating or without any coating were unable to endure this testing because that high normal stress and high shear force resulted in high wear at rubbed surfaces. Then the sliding interface was prematurely failed leading to the test was stopped due to high vibration.
The coefficient of friction was still very low, about 0.05 to 0.06 throughout the testing although with a high load of 60 N. SWR of the rubbed track on GLC coated disc and wear scar on ball was about 0.5 × 10−16 m3/N×m and 9.2 × 10−16 m3/N×m respectively. It was clear that low friction force or low shear force at sliding interfaces was the direct fact to enable this tribo-partners enduring sliding under high load with very low SWR. The maximum shear force during this test was about 60 × 0.06 = 3.6 N, much lower than that (20 × 0.5 = 10.0 N) of cases under 20 N load with either CrTiAlN coating or without coating. Therefore, low coefficient of friction at the sliding interfaces can moderate the severe level when load was high, and so the tribo-partners were still able to provide good sliding performance.
The coefficient of friction was started from low (about 0.16) but it was increased gradually up to about 0.8 close to the end of test. SWR of rubbed disc and GLC coated ball head counterpart were about 74.7 × 10−16 m3/N×m and 61.7 × 10−16 m3/N×m respectively. This high SWR was due to GLC coating on ball head counterpart surface was gradually and completely worn off during test. The wear on rubbed track of uncoated disc was first taken places and then those wear debris (chilled cast iron based materials) were abrasive to the sliding interfaces resulting in accelerated wear on GLC coated ball head counterpart surface. As soon as GLC coating on the ball head counterpart surface was completely worn off, the sliding interfaces became as chilled cast iron against chilled cast iron with high SWR.
The coefficient of friction was about 0.03 to 0.07 throughout the testing. SWR of rubbed track on GLC coated disc was less than 0.6 × 10−16 m3/N×m and SWR of wear scar on GLC coated ball head was less than 1.4 × 10−16 m3/N×m. SWR of both the disc and ball was based on assumption of 0.1 µm wear depth on rubbed disc and flat wear section on ball head counterpart. It was evidenced again that GLC coating on both disc and ball head provided the best option for this tribo- partners to give optimum sliding performance under high load.
Surface morphologies of rubbed track on GLC coated disc and wear scar on GLC coated ball head are given in
Fe content at point 2 area was much higher than that at point 1. This was because for some reason (e.g. ball head slightly tilted under high load) that the inner track area was more truly in contact with ball counterpart surface and therefore more wear debris were accommodated at this area to responsible higher Fe content. Interesting thing was that no sign of abrasive wear identified on the rubbed surface. This was due to the carbon content was also high in this area
Items | Fe | Cr | Si | O | C |
---|---|---|---|---|---|
P 1 | 5.5 | 15.5 | ≈0 | ≈0 | 79.0 |
P 2 | 60.4 | ≈0 | 1.6 | ≈0 | 38.0 |
P 3 | 5.8 | 12.6 | ≈0 | ≈0 | 81.6 |
P 4 | 65.0 | ≈0 | ≈0 | ≈0 | 35.0 |
and so rubbing induced graphitization was taken place with function as more slippery rather than abrasive. It was reported [
The wear scar on ball counterpart as shown in
Thus enough evidences were given that GLC coating on both surfaces of the tribo-partners provided benefit of very low friction coefficient to the sliding interfaces. The tribo-partner system was, therefore, able to operate under very high load with very low specific wear rate.
• Sliding performance of chilled cast iron disc/ball tribo-partners was systematically evaluated in terms of coefficient of friction, specific wear rate and load bearing capacities at sliding interfaces.
• The tribo-partners without coating were conducted high coefficient of friction at sliding interfaces that incorporated to stress on the counterpart surfaces to correspond high abrasive and adhesive wear.
• CrTiAlN hard coating gave protection to coated surface with exceptional wear resistance. But high coefficient of friction and stress at sliding interface were responsible to the opposite surface with high wear and so degraded the tribo-partners’ sliding performances.
• GLC solid lubricant coating on either surface of tappet-disc or cam-ball provided very low coefficient of friction or very low shear force at sliding interfaces. This lubricant property gave moderation of harsh conditions at sliding interfaces leading to tribo-partners with exceptional sliding performances.
• The best option is to apply GLC solid lubricant coating on both surfaces of the tribo-partners, so the system can be operating with very low coefficient of friction and very low specific wear rate under ever higher load bearing capacity.
Wen, X.B., Yao, H.T. and Yang, S.C. (2018) Improvement of Sliding Performance for Ball on Disc Tribo- Partners Using Graphite Like Carbon Solid Lubricant Coating. Materials Sciences and Applications, 9, 191-209. https://doi.org/10.4236/msa.2018.91013