This is a little off topic but, I'm trying to find out if there is some way to figure how much torque it would take on a bolt in order to cause compression and fracture of a brass washer (attached pic). Any help is appreciated.

Roy,

I'm sure somebody will do the math or tell us how (I think I could muddle through it), but my guess is that in theory you would not be able to crack it. That is given conditions absolutely perfect to the design use and we all know how often that happens.

If there was a slight bevel to the inside of the spacer between the bolt head and the brass washer, it might cause enough outward force to break the washer. It looks like there is another washer behind the brass washer. It that a flat steel washer? Brass is pretty soft and would probably crack fairly easily with a metal chip or something sandwiched in there.

When you refer to force area are you talking about the surface area of the washer that is in contact with what looks like a spacer under the bolthead? What does the bolt id of 1.25" refer too?

Good Luck,

To compute bolt axial preload force from torque:

Fp = T / (K* D)
where
Fp = preload force in lbf
T = torque in inch-pounds
D = bolt diameter in inches
K = experimentally-determined nut factor in the range of 0.2 for well-lubricated threads.

Interestingly, the thread configuration doesn't show up in the above equation. It does show up when you compute bolt effective area for purposes of determining fraction of bolt yield stress which will be created by a given torque.

The other half of your problem (the more difficult half) is determining how much force or pressure required to cause that deformation. I don't know. I can't even see the picture. All I see is a red circle.

ok I see it now. I am positive I won't be able to answer your question but I have some questions anywaY:

What is that between the bolt head and the brass washer? A thick steel washer?

Is the brass washer a belleville washer?

Also is there good support behind the washer? What size hole in that support behind the washer?

Do you know if it broke during torquing or possibly some time later? Was it exposed to stress cycles (maybe thermal) which might lead to fatigue scenario?

Thanks for the replies.
Pete,
Sorry about the pic. I didn't take it and the camera we have doesn't take very good close-up's. I actually posted this to help out a very close colleauge of mine. Thats the bolt shoulder you refer to between the head and washer. The washer is not a belleville. Support behind the washer is good. The bolt hole is slightly larger than the bolt. These doors are opened and closed whenever we do a merge change (market oriented, no set frequency)or if they have troubles with spinnerettes (also no set frequency). I believe as you do this is an over time fatigue issue.
Dan,
No other washer. What you refer to is the door surface. yes force area refers to contact area. I'll have to check on the bolt ID figure.

I have not done the math, but my gut tells me that you would yield that bolt before you would squish that washer. Thus it has to be a fatigue issue. Could be wrong though....

Computational tools and analyst skills / experience have definitely come a long way in the past twenty years or so.
Nevertheless, this is a very complex situation.
If the plant were mine, and the question had to be answered, I would pay the money to make a mockup and test it.

Hello,

This may be a bit simplistic, but you may be able to used a dye penetrant on this washer and look for micro-cracks parallel to the big crack to confirm fatigue.

My only other thought would be to polish and acid etch the washer and look at it through a microscope to see what the grain structure of the washer is doing.

An easier way would be to try and recreate this event by trying to crack another washer using a very big torque wrench...

Regards,

-Kris Sonne

Try RE Snapshot, I remember a few case studies on fatigue that may give you other ideas.

I tried RE Snapshot but you can't access the archive anymore.

This message has been edited. Last edited by: Kristopher Sonne,

Thanks for all the valued responses. We have a section of tube like the one these doors are on for test purposes that we plan to use for doing exactly what ron and kris suggested.
once again thanks