Shear pin design questions 2

[Aconcagua]

Hello - I am designing a shaft coupling that needs to be able to handle 2200 Nm under normal operating conditions, but I would like it to break at around 3500 Nm. This device will have a long life, but it will only be operated up to 1000 cycles total so I'm not expecting fatigue to play a big factor.

The shaft is about 50mm in diameter, with the coupling extending out beyond that.

Ideally I would drive a pin through one side of the outer cylinder, into a hole drilled through the shaft, and out through another hole in the outer cylinder. I have a few concerns about this though:

Is it reasonable that I could assume that I could tolerance this so that the pin is in double shear? This would allow me to use a smaller pin, and distribute the load more evenly.
I am confident in my ability to multiply Sy by .577 to figure out the shear strength, but I've read different suggestions about putting a notch in the pin. Should this be radiused so as not to create a stress concentration? Would going from a 10mm pin to a 9mm notch be enough to guarantee breaking at the right point?
Finally, I have learned from reading other forums that I should use hardened bushings for the contacts within both parts. Because these are concentric cylinders, any flat bushing that I put in will not sit face to face with the other bushing. Do I need to consider bending stresses in the pin because of this? If the bushings are reasonably close together are they negligible?

Thank you!

 

[desertfox]

Hi

Try this link, I think it will help a lot.-

http://edge.rit.edu/edge/P15571/public/Detailed Design Documents/Mechanical/PreliminaryPinShear.pdf

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

 

[EdStainless]

I have found that pins with radiused notches have much more reliable load limits.
For real values build a test rig and fail some.
We used to adjust the notch depth to match a standard load value each time we made a batch of them because material properties will be slightly different.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Aconcagua]

Thank you both.

EdStainless- How close were you able to get your breaking load to your designed breaking load? I mean, if you build a jig and iterate on the radius, how close can you expect to get reliably? And if you don't test, but just rely on best design practices, how close could you expect to be?

Are there anything you did, aside from following Shigley's Kt charts, that has helped?

Thanks

 

[Aconcagua]

Oh, and any particular material that seems to break more predictably?

 

[EdStainless]

We were making them from high ductility corrosion resistant alloys, Monel K500 or 625.
We had a min notch and a max notch size, we would make test pins at 25% and 75% depth, shear three at each, and then interpolate the needed depth.
In double shear we were roughly 30,000lb load, and we could hit a values +/-500lbs.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[desertfox]

Hi

I think it's imperative that you actually do testing because depending on the material you use it could work harden, making the shear stress to failure increase.
When I have done this type of calculation I get someone to make it then machining the notch down in stages and retest till in fails at the value you require.


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