The problematic "symptom" that I try to fix, is the static friction of two sliding plastic parts.

This is about a keyboard key switch that was made during the 80s. The problem/issue is that after I applied Grease #1, then when the slider is pressed under a small angle, it gets strong binding issues (i.e. it is stuck and needs a lot more force to move it) and it also exhibits a substantial "stick-slip" after it starts moving, i.e. irregular dynamic friction, which instead of a steady velocity and a smooth sliding it slides in a starting-stopping way.

When I applied Grease #2, the aforementioned issues got greatly reduced (not completely eliminated the static friction though), resulting in a much better and smoother sliding.

However, this is contrary to what I have expected. On paper, Grease #1 is much more viscous than Grease #2. And according to the consensus saying how much viscosity matters, it should be the #1 that should make it smoother and not the other way around.

These 2 different greases have a big difference in their base oil viscosity at 40C. Their Techincal Data Sheets claim that grease #1 is 9500 mm2/s viscous and grease #2 is 1500 mm2/s viscous at 40C.

Another important difference is this:

  • Grease #1 is a grease with a Polydimethylsiloxane (PDMS) "silicon oil" as Base Oil, and an "inorganic" Thickener (probably silica)
  • Grease #2 is a grease with a Polyalphaolefin (PAO) "synthetic hydrocarbon" as Base Oil, and Silica as a Thickener.

After a simple study of the basics of Tribology and by reading various articles on the subject, I have came to know that viscosity is regarded as the number one factor for lubricating two sliding surfaces and that higher viscosity results in lower static and dynamic friction and more importantly eliminates the unwanted stick-slip phenomenon*. The higher the viscosity the better.

According to the above, for my use case, grease #1 should be much better at both reducing friction and eliminating stick-slip sliding. However, when applied and tried empirically the exact opposite is true. #2 is actually lubricating where #1 is probably making the friction and the stick-slip even worse than without applying any grease at all.

  1. Why this happens instead? Is the viscosity rated in the TDS wrong? Is there another explanation?
  2. What property exactly of #2 is responsible that provides the desired lubricating properties, which #1 lacks?
  3. Is there another property of a lubricant (grease) that would be better for this use case?

Use case: Two plastic parts sliding against each other. They are parts of a keyboard switch. The plastics are Nylon (the Slider) and the second (Housing) is probably ABS. More information: https://deskthority.net/wiki/Alps_SKCL/SKCM_series (the parts are the Slider and the Shell)


Some relevant and helpful information:

The switch was made during the 1980s, therefore this means that they must have used oils/lubricants/greases that used the technology that was available at that time. The switch's Slider came pre-applied with a grease/spray/solid/whatever from the factory, which strongly suggests that the manufacturer knew that without it, the slider would not slide smoothly. And this is a fact that have been verified, if this is cleaned off, the switch gets a high static friction.

One more important thing: Another grease, Grease #3**, actually worked very well with this switch and completely eliminated any static friction, even at obtuse angles. It has Polyalphaolefin (PAO) as Base Oil and Silica as thickener, however it is extremely viscous. It is considered a heavy "dampening" grease instead of a "lubricating" one, which makes the sliding slower, but also makes it very precise and smooth. These kind of greases are used with binoculars, microscopes, rotary dials and controls that need a very precise and smooth movement, and they are regarded as highly desirable qualities for these use cases.

My guess is that the factory applied grease/solid/whatever, is probably a "dampening" one too. Switches in a good condition from that series, have that precise and smooth slide, which is a characteristic behavior that differs significantly from the behavior of other switches applied with usual lubricants.

Technical details of the greases:

Grease #1 Technical Data Sheet: https://www.oks-germany.com/product_downloads/pdf/PI/en/PI_OKS_1110_110645_EN.pdf

Grease #2 Technical Data Sheet: https://www.oks-germany.com/product_downloads/pdf/PI/en/PI_OKS_477_113696_EN.pdf


*. "The test results show that, by increasing the viscosity of the lubricant, the potential for stick slip is greatly reduced." - How to Measure, Prevent, and Eliminate Stick-Slip and Noise Generation with Lubricants

**. Grease #3 Technical Data Sheet: NyoGel 767A "Dampening" grease

  • $\begingroup$ Are you sure slider is nylon? Acetal (aka POM, Dupont trade name is Delrin) is generally preferred for that type of application and is typically not lubricated as the plastic itself has a low coefficient of friction and self-lubricates.But if it's definitely nylon, then that's irrelevant. $\endgroup$
    – Andrew
    May 7 at 12:39
  • $\begingroup$ No, I'm not sure if the slider is nylon, it is what "the consensus" of the vintage keyboard community says. However, I am certain that it is not a self-lubricating one, precisely because this particular switch is a version that was made during the 80s, which were typically factory pre-lubricated and precisely by the fact that the manufacturer actually made a newer version, in beginning of the 90s, that is different material which is self-lubricating and thus needs no lubrication. I don't know how to tell what exact kind of material it is, are there any ways that I can use for that? $\endgroup$ May 8 at 8:17

The effectiveness of lubricants depends on a number of factors. If the objects are in motion, hydrodynamic lubrication is important.

For static friction, adhesion of the lubricant to the surface is more important. For that reason, motor oil has antiwear additives to prevent metal-to-metal contact. Since the plastic parts in a keyboard are initially at rest before sliding, static friction predominates.

  • $\begingroup$ I think your answer is making a lot of sense. The greases in fact have completely different Base Oils. Grease #1 has Polydimethylsiloxane (PDMS), which AFAIK is a "silicon oil" and grease #2 has Polyalphaolefin (PAO) i.e. synethetic hydrocarbons. Do the hydrocarbons adhere that much better in the surface? Is that what is causing the different results? You are spot on about the static friction, as this is exactly what I am trying to "fix" with the lubrication. I will update my original question to be more specific. Please check it out and give your opinion. $\endgroup$ May 8 at 8:44

The time scale of viscosity measurement is so much faster than the time scale of your switch operation that there may be only a slight connection. The temperature difference is another big effect. Although your switch may move rapidly when it does move, there will be long periods of no movement, when the grease can reorganize to bad effect.

Grease #1 has an inorganic thickener (could be silica); Grease #2 has a silicate thickener (could be some clay). Grease #1 had an entry that identified a very small amount of oil separation after some period. My thought is that a viscosity measurement is run on a homogeneous mix, and oil separation implies a non-homogenization after a period of standing, generally in the direction that the more fluid ingredient separates out while the thicker part gets thicker - to the point where it could develop a yield stress, i.e., a strength resembling a solid structure, hence the stick part of the behavior. Once the structure is broken, the more fluid part comes into play. This effect might take place only on a very small thickness level, e.g., when two parts are pressed together, squeezing the lubricant to very thin dimension. This would not be a part of technical specifications, but could well be a performance issue.

Grease #2 did not have an oil separation entry, and did have a very low pour point (-20$^o$C). Grease #1 had no drop point, indicating that it was so viscous as to have a structure, not a flow.

Very high viscosity oils by themselves could be good lubricants for your switch, but a three-dimensional thickener like silica, by not preventing oil separation, might contribute to excess thickening, a yield stress that locks up to give stick. A silicate (clay is layer-like, not three-dimensional) thickener that does not exude oil could remain in place homogeneously indefinitely and be slippery.

Since no long-term effects have been noted for silicone or petroleum oils on the switch materials, I feel there is no softening or material degradation to consider. But the performance test you require is specific to your need, and you probably need to devise your own test regime that measures slipperiness after a prolonged rest time with a pressure between two objects made of the very materials of the switch - although, if my hypothesis about the oil separation and yield stress formation is correct, it should probably happen between any two smooth objects.

Technical terms like rheopexy or thixotropy are not applicable, in my mind, because I think the problem is with time-dependent non-homogeneity. But non-Newtonian would be OK - maybe most fluids are non-Newtonian at some level.

Some petroleum greases are thickened with calcium or lithium soaps. A big difference is the temperature of the dropping point. This is probably not your concern. The lithium soaps, even at low concentration, exert powerful thickening and a slipperiness without gelling. Some of these (e.g., white lithium grease) are higher cost and come in little tubes for electrical use; others come in 55 gallon drums, with varying amounts of added lithium soap. One of these might provide the adhesion and slip without the stick.


Not the answer you're looking for? Browse other questions tagged or ask your own question.