Increasing a coefficient of friction between 2 surfaces 2

[elinah34]

Hello,

A colleague told me about a problem they had and how they decided to solve it, and I would like to find out what you think about it.
His crew designed a chamber (see a simplified picture below) that is expected to be flown and subjected to varied loads and vibration.

After getting all the parts and starting assembling they found out that the Safety Of Factor for side slippage (relative movement in x and/or z directions between both clamped parts of the chamber) is very small and marginal.
Since adding bolts (for increasing clamping force) was problematic, they decided they are going to increase the coefficient of friction between both clamped parts. They decided (without any experiment)to grit blast the mating surfaces like presented in the picture below.

I have several problems with their decision:
1. they damage the Hard Anodize surface treatment that should protect from corrosion, and abrasive grit blast would expose the aluminum.
2. they didn't really explored how much the coefficient of friction is increased and if it's worth it.
3. maybe there are more elegant solutions? maybe there is a specific surface treatment or coating that on one hand might increase the friction coefficient and on the other hand protect the metal from corrosion?

what is your opinion?

thanks!

 

[desertfox]

Well you haven’t told us how many bolts the parts currently have, or the bolt preload and no details regarding the external forces causing the slippage. One could increase the size of the existing bolts if extra bolts can’t be added.
If we have more information we might be able to give better answers.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[TugboatEng]

Or the material of the bolt. Typical fastener yield strengths range from 70-210ksi. You can potentially triple your clamp load without any other changes. To maximize the clamp load from a fastener develop a torque to yield procedure that will allow you to utilize the fastener at 100% of its yield strength.

 

[desertfox]

The only problem with torquing the bolt to yield is that the clamped aluminium parts might reach their yield stress before the bolt material does, saying that we don’t know the bolt material.



“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[IRstuff]

One might consider positive position features like pin/slot designs if you want little movement and don't necessarily care about absolute position.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[GregLocock]

Knurl and re-anodize. Loctite.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[Sparweb]

Press fit dowel pins. Two or three would do.

 

[SWComposites]

they found out that the Safety Of Factor for side slippage - so just how did they find this out? Test? Analysis?

And what is a “Safety of Factor”?? Maybe you mean Factor of Safety??

 

[EdStainless]

If they roughen the surface they move from friction to soil mechanics.
If the surfaces actually dig into each other then they have an entirely different mode.
I agree, keys or pins would have been a much more positive method to do this.
At least with those you could estimate the actual values.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[elinah34]

Hello, and thanks for replying.

Adding bolts, increasing bolt's size, increasing tightening torque are according to my colleague impossible.

I also recommended him using press fit dowel pins, but he claimed it would require mutual machining since over constraining is likely to happen if he wouldn't do so.

An interesting option that was raised here is knurling prior to anodizing. Does anyone have any experience with this solution?

 

[TugboatEng]

Did you see anything that I have mentioned?

 

[elinah34]

TugboatEng - yes, thanks for replying.
I asked my colleague about it, and he said none of the offered options is possible.

 

[TugboatEng]

Maybe your colleague is unaware of offered options. I assume you're using typical stainless steel bolts which have a yield strength of 70ksi. There are stainless fasteners with yield strengths of up to 180ksi. ASTM A286 and Custom 450 fasteners are options.

 

[CWB1]

I also recommended him using press fit dowel pins, but he claimed it would require mutual machining since over constraining is likely to happen if he wouldn't do so.

If machining new, matching bores for solid dowels is too expensive/difficult, you can also add a pair of counterbores to the clamped faces and pass your bolts through a hollow alignment dowel. Piloted counterbore tools makes this a much simpler task. Similarly, if the concern is side-slip due to loose bolt clearances you can tighten that up with simple bushings.

 

[elinah34]

CWB1, Thanks.
For efficient and desired effect the dowel pins and/or bushings should be pressed fit or at least have minimum gap relative to the holes?

 

[JStephen]

What they did to actually increase the friction sounds perfectly reasonable.
What doesn't make sense is assuming that it is improved without any testing. You have an intuitive feel for how that ought to work, but it won't always be right, either.
If the friction is that critical, you'd need to look at what happens if it gets wet, oily, or dusty, either during assembly, in operation, or in future maintenance disassembly.

 

[Tmoose]

GM LS V8 engines and maybe some Mazda use diamond encrusted "friction washers" reportedly to better secure torsional dampers to their crankshafts.

 

[Scuka]

I also recommended him using press fit dowel pins, but he claimed it would require mutual machining since over constraining is likely to happen if he wouldn't do so.

Hole-and-slot system for pins provides exact constraint without the need for tight positional tolerances.

 

[LittleInch]

This one isn't easy as we don't really know the details such as:

What is this being used for?
Holding internal pressure or external pressure? If its got a pressure differential, not having a gasket will make sealing very difficult.
What is causing this lateral movement?
Given there are essentially pins in it already with those bolts, how much movement or force are they trying to resist?
What's the clamping force?
Is this joint regularly made and remade or only once - in which case why don't you just weld it?

This method is fraught with issues IMO.
My usual take on friction is that it is an unreliable friend, Never there when you need it, but appears at times when you really don't want to see them.

Your concerns appear quite valid, especially if this joint is regularly made and remade.

But we only have about 25% of the information I think you do....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MintJulep]

It's so easy to test actual parts to find actual coefficient of friction. There's really no excuse not to.

thread367-327419

So what's your colleagues' excuse for not?

With a soft material like aluminum, any surface roughness feature will tend to smush when the bolts are tightened, and maybe smush more when external loads are applied, reducing the bolt tension.

An oxide film will form spontaneously. You haven't told us about then environment other than "flown", so can't tell you much about corrosion.