Test For Fretting

The surface damage and wear associated with fretting movement lies in a synergistic space of adhesive, corrosive, and abrasive wear.

Fretting and anti-friction lubricants play a key role in metal-to-metal durability and performance.

By Melissa Mushrush, MOLYKOTE Specialty Lubricants, DuPont

“Fret” has several definitions, both figurative and literal. In the tribological context, fretting manifests as wear and/or corrosion and is due to the damage induced under load by repetitive relative surface motion such as vibration. Fretting is inherently complex; the surface damage and wear associated with fretting movement lies in a synergistic space of adhesive, corrosive, and abrasive wear. A plant engineer likely sees signs of this in any machine components where vibration is present, such as gears, bearings, dovetail joints, seals, spindles, and bolts. Fretting wear or fatigue is characterized by very small features because of the small relative motion. It often coincides with corrosion at these features because the vibration can remove surface oxides that protect the metal surface.

There are several standard methods associated with fretting, from general assessment of lubricating greases (ASTM D-1470) to more specific methods that span a range of applications. Simulating fretting conditions with a tribometer is difficult, as the instrument needs an extremely small displacement (step size) as well as high frequency and high load. There are dedicated commercial fretting testers, some with extremely small displacements and high frequencies, but sometimes adding a new (and expensive) dedicated tester is not an option. How much variation in the test parameters is allowed for a test to still simulate fretting wear?

The key condition in any fretting test is the displacement. The contact surfaces that experience the “stick” versus the “slide” components of the oscillatory movement must overlap, i.e., the Hertzian contact radius must be equal to or greater than the sliding amplitude. This is because fretting happens where a surface experiences stick and slide. Many tribometers have a small enough displacement to meet this condition.

Frequency in standard methods varies from 30 to 50 Hz, also reasonable for many tribometers, and the load needs to be high, but how high? If a setup is limited in load, the question to answer is whether the load is sufficient to differentiate lubricants with good enough resolution to rank them in fretting resistance effectiveness.

The Bruker Universal Mechanical Tester block-on-ring drive is shown with a minimal arc length setting of 4.5-deg., for a total 9-deg. oscillation (left). The piece used for this testing is shown with load applied onto the block (right).

Testing for Fretting

The test system used for this work was a Bruker UMT (Universal Mechanical Tester) using a block-on-ring module. The minimum arc length for oscillation is 9 deg. and the frequencies used are 30 Hz and 50 Hz, which meet or exceed the requirements of reference standard ASTM D1470. Load was limited to 100 N (153 MPa). The setup did not have any temperature capability. Time of the test was kept to two hours.       

Of the many lubricant products available, pastes tend to provide the best protection from fretting wear. Other lubricants such as antifriction coatings typically do not. To evaluate the fretting method and its reproducibility, both types of lubricants were tested to look at a wide range of performance. Initially, test parts were weighed to measure material lost (as mass), but in all these tests, not enough material was lost to be measured reliably on a four-place balance.

Much more straightforward is to measure the size of the scar after a test of specified time. In principle, the larger the wear scar, the less effective the lubricant. If fretting is causing wear at the contact point, the contact area becomes larger as a function of test time. It would follow that a series of lubricants that offer poor fretting wear protection result in larger wear scars than those that offer better protection.

For testing, standard LFW-1 steel test blocks and lapped rings were used, with the lubricant applied to the ring. When run unlubricated, the test occasionally failed early, i.e., the instrument stops the test if forces spike and cause the coefficient of friction to exceed 0.5. For tests that lasted the full two hours, wear scars on the ring were typically >6 mm (simply measured with a micrometer).

Four commercial antifriction coatings were tested at 100 N load, 50 Hz, 9-deg. total arc oscillation, and the resulting scars were all large, between 5.2- and 5.5-mm wide, compared to the unlubricated case of >6 mm. Unsurprisingly, the coatings do not show a huge reduction in scar compared to no lubrication and they also showed frequent early failures. This type of inconsistent behavior is more common as a system approaches higher friction and is not surprising.

To see the other end of the performance range, we tested a commercial paste that is recommended for fretting applications. MOLYKOTE Paste 1 was tested 10 times under the same conditions (50 Hz, 100 N, 9-deg. total oscillation, 2 hr.) and showed quite good reproducibility. In each of 10 replicate tests, the block wear scars were measured at top, middle, and bottom locations. The average of those three measurements varied between 1.32 mm and 1.77 mm, with a standard deviation of 0.14 mm. The ring scar was evaluated in the same manner. Average measurements varied between 3.81 and 5.96, with a standard deviation of 0.66.

As a comparison, a second paste was also run with 10 replications. Paste 2 is not as effective at fretting wear protection as Paste 1. In this case, there were more early failures and more variation in scar widths, as was seen with the coatings. Also, as expected, both the block and ring scars were wider on average for Paste 2 than for Paste 1, but smaller than for the coatings. From the data, this test appears to be a reasonable assessment of fretting performance using a non-specialized tribometer.

Future work includes measuring a suite of pastes (and using microscopy to measure wear scars more accurately) to determine with what resolution various lubricants can be differentiated for their effectiveness against fretting wear. EP

Melissa Mushrush, PhD, is a chemist for MOLYKOTE Specialty Lubricants at DuPont, Midland, MI (dupont.com). In her role, Mushrush explores new materials for anti-friction coatings designed to improve friction and wear performances in coatings and pastes. Her areas of expertise include vacuum technology, thin-film deposition, film and device characterization, process data management, statistics and error analysis, and accelerated testing methods. She is a member of the Society of Tribologists and Lubrication Engineers.