Composites as bearing surfaces

[pictsidhe]

I've been pondering this a while, but can't find much about it. Do composite make reasonable bearing surfaces? I'm not thinking of shafts at thousands of RPM, but control linkages and maybe telescopic landing gear. I believe that glass is pretty abrasive, but I'm unsure about carbon and kevlar. Anyone tried this or have any references?

 

[pictsidhe]

I'm thinking of composite control shafts that need support. So run them through a plastic self lube bearing? I asked Igus some months back, but they didn't know.

 

[MadRocketScientist]

One thing to watch for is the different rates of thermal expansion, Plastics can tend to grip the shaft way too tight in colder temps.

 

[plncraze]

2 references which mention this: Composite Facts by Englert (Caro-engineering.com) and Evans Lightplane Designer's Handbook. Englert's bio is incredible and Evans designed the VP aircraft and did stress analysis at Convair.

 

[Jay Kempf]

I've been pondering this a while, but can't find much about it. Do composite make reasonable bearing surfaces? I'm not thinking of shafts at thousands of RPM, but control linkages and maybe telescopic landing gear. I believe that glass is pretty abrasive, but I'm unsure about carbon and kevlar. Anyone tried this or have any references?

There is a whole world of industrial bearings made from reinforced plastics. Mostly these are sleeve bearings water lubricated or dry in some severe duty applications.

 

[pictsidhe]

I've used Igus bearings in the past, hence asking them.

 

[BoKu]

I'm thinking of composite control shafts that need support. So run them through a plastic self lube bearing? I asked Igus some months back, but they didn't know.

I don't know the answer, but I would bet on the side of "No, composites don't make very good bearing surfaces." Some random commentary:

* Good bearing surfaces are generally uniform in composition. By their very nature composites tend to alternate areas of resin with areas of reinforcement.

* You want the softest and most sacrificial thing in the system to be a) not wear critical and b) inexpensive to replace. With composite pushrods or (more likely) push-pull tubes, you would probably be hard pressed to find something softer than the composite element. And if it's the element that wears through, you are flirting with a control disconnection and not just the sort of slop you get when a bushing gets egged out.

* I have occasionally considered the idea of using Graphlite pultrusions as push-pull cables, but I think that in order to do it right you'd have to perfect a process of coating the carbon with a substantial PTFE or nylon sheath. And that takes industrial processing well above my pay grade.

* From what I've heard, I suspect that there are cases in the field where folks are using 5/8" OD carbon fiber push-pull tubes with short-throw linear ball bearings such as the ones commonly used in European sailplanes, and which I use in the HP-24. You can get them at Aircraft Spruce for like $12, but I can show you how to make them for about two bucks each: http://www.aircraftspruce.com/catalog/appages/pushrodguide.php?clickkey=3377706

* One good bad example of composites as bearing surfaces is that many puddle-jumping airplanes from before WWII (and consumer availability of mass quantities of nylon) have control stick torque tube saddle bearings made of phenolic. The grades used for this are usually made of cotton cloth in some sort of brown phenolic resin. The material itself is reasonably good as a wear surface, but what happens is that the bearing surface is essentially end-grain cotton that gets embedded with silica (dirt) and wear particles (steel) and becomes abrasive. Of course, for the five year or so service life originally envisioned, that's not a problem...

--Bob K.

 

[pictsidhe]

I found a wear test of 52100 on low and high modulus CFRP.

For those who can't be bothered to read it: Both high and low modulus CFRP samples were extremely wear resistant. The 52100 pin bearing on them had neglible wear on the high modulus samples, but severe wear on low modulus. The report also mentioned tests of low on low showing little wear, but no details.

 

[BoKu]

Now that I think on it, I seem to recall that Autoreply has some practical expertise on this topic. You might ping him to chip in on this discussion.

 

[wsimpso1]

First some basics:

Both the parts sliding on each other are bearing surfaces;
Both materials must stand washing and lubrication materials;
Sliding clearances need to be small but can never bind;
Neither part can get eaten away;
Neither part can grab ambient grit and eat away the other;
All materials must stand the loading with generous factors of safety;
Usually, one surface should be hard, the other soft, both for long life and for smooth running.

Composites are comprised of strong stiff fibers and somewhat softer resins that are also fairly stiff and strong. The fibers on one surface, when (not if, when) exposed will eat into the countersurface. They will not allow fine abrasive to imbed and become harmless.

Composites will comprise a hard surface, so your other surface had better be soft. Composite on steel will usually cause all sorts of trouble.

There are bearings out there. Looks at McMaster-Carr's website, then look up plain bearings on the web. Lots of them. Some may even be suitable for composite tubes.

If your composite tubes are sliding axially, I would think about BoKu's offer on his rolling element support bearings. If your composite tubes are rotating, I would think seriously about rearranging BoKu's bearing to support the shaft in the same rolling elements.

Most of us are supporting steel tubes through bearings, you may still find that your composite tubes do not like running in bearings. Inspect frequently and be ready to replace parts. No matter how you do them, you may have to change to other materials.

Are you buying tubes or making them? If you are rotating them in the bearings, how are you planning to make them sufficiently smooth and round and the right diameter for the bearings?

If the loads are small and the speeds low, think seriously about BoKu's rolling element bearings.

Billski

 

[henryk]

DLC (Diamond Like Carbon) technology can solve friction problems...

http://www.voestalpine.com/highperformancemetals/international/en/service/beschichtung/pvd-coatings/sucaslide-the-modern-friction-reducing-dlc-coating/

 

[MadRocketScientist]

If I was needing to use a carbon tube in a linear bearing, I would consider bonding a steel sleeve over the shaft for the bearing to run on.

 

[skier]

Depebds what what you consider composites. There are PTFE impregnated fabric liner systems, carbon mixed with other materials is used as a bearing liner in certain applications, there are homogenous PTFE composites that used in self-lubricating bearing systems, and there are many other options as well. Bearings are a fascinating industry and there are way more options than most people are aware of. Some are standard off the self options and some are only available in custom designs.

 

[Dan Thomas]

If I was needing to use a carbon tube in a linear bearing, I would consider bonding a steel sleeve over the shaft for the bearing to run on.

The coefficients of linear thermal expansion vary enormously. Plastics expand and contract a lot more than metals, so in cold weather that steel sleeve might become debonded from the composite shaft.

I always found it better to design machinery that had pivoting parts rather than sliding parts. Sliding stuff is difficult to seal, so if you have close tolerances, any bit of grit can cause galling and seizure. Larger clearances lead to rattling and slop. Friction is almost always higher with sliding stuff.

The Cessna 400/ttX/Corvalis uses aluminum rods running through guides that are nothing more than three small nylon wheels. Even there they have to have clearances, especially the shaft that runs immediately aft of the stick, since it both rotates and slides. There was a service bulletin on the aileron pushrod guides; they were chafing the tube. Again, vibration and grit cause trouble. Cessna also used the three-tiny-wheel thing in the 210 and 182, in the control column glides, and rattling, chafing, and wheel failure was common. The wheels were small ball bearings that had nylon "tires" molded onto them, and wear, age and vibration would break the tires, especially if they were set too tight (rotating the shaft put high side loads on the tire, breaking it) or too loose (vibration and rattling would wear or break it). They would have been better off using a simple plastic bushing like the 150/172/180/185 and so on, but since the heavier airplanes have higher control forces, and the pilot tends to pull down as well as back in the flare, for instance, friction would be higher.

Some homebuilts as well as certified airplanes used pivoting arms to support pushrods. Avid/Kitfox and the Lake Amphib are examples. Totally pivoting systems that have very little friction and are relatively free of binding. The arm is a triangular affair, with a long bushing along its base that stabilizes it sideways, and the pushrod mounted to the pointy end. The pushrods are mounted close to joints with further rods to enable directional changes. The Lake has, IIRC, three or four of those pivots in its elevator system, but you can't feel them at all. Rudder is similar.

 

[Jay Kempf]

If I was needing to use a carbon tube in a linear bearing, I would consider bonding a steel sleeve over the shaft for the bearing to run on.

Except: https://provenproductivity.com/galvanic-corrosion-carbon-fiber-materials-2/

 

[MadRocketScientist]

Except: https://provenproductivity.com/galvanic-corrosion-carbon-fiber-materials-2/

This may be able to be mitigated with the use of an insulating resin between the carbon and steel sleeve.

 

[pictsidhe]

A thin layer of glass is my favourite. Resin is insulating, but you have to space it.

 

[Jay Kempf]

Yup, except resin is notoriously porous on any sort of cloth. When we used to do ETFE powder coatings for this application we used to used a hipot to ground with a steel brush wand to find porosity. Corrosion can work right through some polymers.... at the molecular level it is about who attracts who, who repels who and how strong the charge is as well is who is big or small enough to get through who. Some acids used to walk right through thermoplastics without hurting them and just eat the bond to the base metal.

 

[autoreply]

Now that I think on it, I seem to recall that Autoreply has some practical expertise on this topic. You might ping him to chip in on this discussion.

Yep. Composite slide bearings.

The counter surface has to be considerably harder such that the bearing will wear, not both. Typical fiber is PET with MoS2, carbon and PTFE as integrated lubricants. Glass is too abrasive, Nomex and Kevlar "smear". Carbon fiber works, but is often the cause of considerable galvanic corrosion issues. MoS2 is best avoided when moisture or submersion is possible since this ruins it's lubricating properties.

Compared to engineering plastics (POM, Nylon), composite slide bearings are up to 4 orders of magnitude more wear resistant. Our best product does 10^-10 /MPa, so for every MPa in contact pressure you see 1 mm of wear per 10000 km of sliding distance.

The other big difference is in abrasive environments. We often operate our bearings in dredging environments where they're submerged in a slush of sand and water. Engineering plastics will see spectacular wear in those circumstances.

Using composites as a counter surface would concerned me; the materials we use are too soft to not start wearing and in abrasive conditions (dust), you might see considerable wear. Better to use a stainless steel counter surface and either engineering plastics or dedicated composite slide bearings.

If you don't care about the counter surface (tail skids for example), glass fiber is the best choice.