Surface finish on carbon or glass fiber filled PEEK 1

[tr1ntx]

Does anyone have definitive information on how smooth a surface can be achieved on the 20-30% carbon or glass fiber filled PEEK compounds? I need it for a sealing application where our current spec is 16Ra for plain PEEK. Also, I wonder if fibers could be pulled out by the machining and leave voids.

The width of the surface will be 0.205", an angled face on the end of a short tube.

 

[Demon3]

If you injection mold parts and get the flow right the filler is not present at the surface so you can get a very smooth finish. If you are machining things are very different.

The one person I know of who may have the answer to your question is Dave Wilkinson at Piper Plastics Chandler Arizona as he deals in such high temperature filled plastics routinely.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[tr1ntx]

Thanks, I'll see if I can get hold of him.

 

[BiPolarMoment]

I know that CFR PEEK can be machined down to at least 0.05 micrometers (2 Microinch) because there have been friction studies done with surfaces at that roughness (applications in orthopedic bearing surfaces). What I would be interested in your case is how the open ended fibers at the surface affect sealing (e.g. is there irregular absorption?).

 

[Demon3]

If the fibers are not perfectly wetted out then there will be voids exposed when you machine down the surface. Those will create roughness. You could fill in the voids with liquid epoxy although of course that's an extra step.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[prdave00]

Chris,

Can you explain what you mean by "perfectly wetted out"? I'm not familiar with this terminology in this context but I'm guessing it means fully encapsulated by the polymeric resin matrix.

BiPolarMoment,

2 micro-inches sounds really good for a CFR-PEEK. This Ra finish is pretty standard for orthopedic implants with metallic or ceramic bearing surfaces, but I haven't heard the same for filled PEEK. Could you point me to a paper?

 

[patprimmer]

Wet out means complete surface contact with no air in between. There are normally coatings on the glass fibre that are compatible with both the glass and the polymer to assist wetting and to maximise the bond.



Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Demon3]

Yes, the epoxy need to fully coat the surface of every fiber, known as wetting out. There have been studies on that because in reality you don't get perfect wet-out. That leaves voids and they are where cracks initiate weakening your composite.

I know that SiComp (Swedish Institute for Composites) have done some nice work on that. I have a thesis here somewhere in a moving box. Can't lay my hands on it right now though.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[BiPolarMoment]

prdave,

Unfortunately I do not have the formal study where this information came from; The information I have was from a document Invibio (Victrex) sent me when I was asking about frictional coefficients... in the document there is a statement:

"From analysis it was observed that the roughness followed by the sliding velocity and sliding velocity / roughness interaction influence on the coefficient of friction of PEEK. These factors have a reduced influence on the friction coefficient of CFR-PEEK. For carbon fiber reinforced PEEK the coefficient of friction varied from 0.08 when the Ra was 0.05µm, stress was 5 MPa and sliding velocity was 0.02m/s to 0.22 when the Ra was 0.45µm, stress was 5Mpa and sliding speed was 0.02m/s. For PEEK a change in roughness from 0.05 to 0.45 produced a change in coefficient of friction of 0.14 to 0.4. "

I can't seem to corroborate this information easily, so perhaps it's hogwash, but if you'd like to know the source I'd have to suggest you contact Invibio directly.

 

[Demon3]

Just wanted to mention that you can get graphite fibers. They are made the same way as carbon fibers, just at higher temperature. Being made of graphite (a well-known lubricant) I would expect they would provide a low COF and great mechanicals in your PEEK composite.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[tr1ntx]

OK, so Invibio has test results for Ra of 0.05µm (2µin) and of 0.45µm (18µin), so we know that is possible. Of course, it might be lapped or honed to get that.

Dave at Piper said you "should" be able to get 16 Ra machining a properly molded part, but it might be difficult, and that compression molded stock shapes will prove more difficult. (Drat!)

Chris at Piper said that he had recently made parts from 30% carbon fiber filled PEEK compression molded tube and they got 24 to 32 by machining.

So I guess there's my answer: if 32 won't work, it will have to be lapped. Although Demon3's thought of applying epoxy might be the ultimate solution to getting a super-smooth surface, i.e., build it up slightly and re-machine or hone.

Demon3, any suggestion as to the exact epoxy formulation/type?

 

[prdave00]

To add to Demon3's posting, Boedecker Plastics offers Ketron HPV bearing grade PEEK compounded with graphite & PTFE here

http://www.boedeker.com/peek_p.htm.

I have also found their staff to be helpful for technical questions. They also offer machining services so maybe one of their engineers / machinists can comment on the finish attainable.

For a sealing application, you may be more concerned with other roughness parameters like the maximum height of profile (Rt) or Rz instead (or in addition to) Ra. Whether or not fibers pull out of the matrix may not be picked up by looking at Ra alone.

Here's also another past post that discusses finish of PEEK:

http://www.eng-tips.com/viewthread.cfm?qid=131141&page=22.

Lastly, attached is a paper investigating CFR-PEEK for medical applications. The Ra they report for machined CFR-PEEK of 1.106 micron (44 micro-inch) is consistent with my research to date. They also report a value for injection molded surface which is better. I'll attach second, albeit older, paper taht looked at roughness parameters for PEEK with various fillers in a 2nd post.

BiPolar: I might have the paper on file that you mention. Do you know the author?

 

[prdave00]

Sorry. Don't hold your breath for that second paper. On closer review it only reported the roughness of the steel counterpart.

 

[BiPolarMoment]

Dave,

There are 3 papers mentioned in the document but I don't know that any of them are where those numbers come from:

Effect of counterface roughness and it’s evolution on the wear and friction of PEEK and PEEK bonded carbon fiber composites on stainless steel, D.M. Eliot, J. Fisher, T.D. Clark, Wear 217 (1998) 288-296

Effect of carbon fiber reinforcement in the frictional behavior of PEEK in a water lubricated environment. J. Paulo Davim, N. Marques, A. Monteiro Baptista, Wear 251 (2001) 1100-1104

Effects of sliding velocity and applied load on the tribological mechanism of amorphous poly-ether-ether-ketone (PEEK), G. Zang, C. Zang, P. Nardin, W-Y. Li, H. Liao, C. Coddet, Tribology International 41(2008) 79-86.

To corroborate your 1.1 micron number I also have a [very short] presentation that was provided to me by Invibio regading their MOTIS bearing material on CoCr in a knee application: the initial roughness was 1.6 microns but after 5M cycles it had been burnished by the CoCr to 0.16 microns. I forget the specifics of "MOTIS" but I do know it's also a CFR PEEK material--just not quite the same as their normal filled grade (different fibers?).

 

[prdave00]

BiPolarMoment,

Motis is Invibio's pitch-based CFR-PEEK with chopped fibers. They recommend this material for bearing against hard counterparts in orthopedic implants. The paper from which I think Invibio got these numbers is: Scholes & Unsworth "Pitch-based carbon-fibre-reinforced poly (ether–ether–ketone) OPTIMA assessed as a bearing material in a mobile bearing unicondylar knee joint".

It looks like tr1ntx got an answer & is satisfied with it so I'll let it be. If you want to discuss implant grade CFR-PEEK further then I suggest a new thread in the "Bioengineers other topics" forum.