T SLOT Tear out.

[RoarkS]

Howdy,

So I've tried problems like this before in FEA and have decided to change the design so I didnt have to deal with it. Well in this case I cant change it... I have to figure out the answer. I only have solidworks simulation to play with. I've tried every trick I can think of to relive the point loads but still I cannot get it to tell me that it wont fail at the point in the red circle at a bit over 110 lbs.

So the yellow block is C36000 and the grey pin block is steel 1018. The t slot is blind, keeping the pin block from sliding out. I know from pulling on these things they tear out starting right were solidworks says (in the red circle on the brass) but its a lot more than 110 lbs.

so the t slot flange dimensions are .5"x.125"x.375" and it rides in the t slot... 2 inches. The load is hug by a pin thru the hole. I've all but given up on simulations and hope that some FEA wizards who do hand calcs, know how to hand calc this one.

Thanks!

 

[rstupplebeen]

Fillets or even better conic fillets will go a long way to relieve the stress concentration. The general rule of thumb is not to model fillets in FEA but if it's in the area of interest I typically do include them but be ready for a computational hit due to mesh size.

An alternate approach would be to include plastic yield in the material properties. This will distribute the load away from the singularity.

I hope this helps.

Rob Stupplebeen

http://www.optimaldevice.com

 

[DanStro]

I agree with Rob, to get a better idea about what is really happening you should model the fillets/chamfers in that area. That should bring the stress concentrations closer to reality. Also, since there is about a 10-15 ksi spread between the yield and ultimate strength it is possible that the required load to get the part to yield is 110 lbf but the load to actually tear the part out would be much higher.

HTH,
Dan

Han primo incensus

 

[RoarkS]

I haven't tried conic fillets... those are new to me since I did my FEA training. I had standard fillets as .005 and .015 as required to avoid interference when I ran it. I even put some mesh controls on the fillets to locally refine it. I was well past 200,000 degrees of freedom but it didn't take too long... I think it was a 2 minute run time. I also have yet to check for jacobian and aspect ratio errors, but I've rarely ever had that be an issue the way I set up my meshes. @rstupplebeen As far as plastic yield... I had set up the results so it would determine the safety factor based on Max von-mises yield strength. I'm just using the stock solidworks material profiles for both materials. I'm assuming you're talking about a setting I'm unaware of? maybe in the menu where I select the material as being an linear isotropic material... I'll have to look at that when I can get back to it later today and see whats in there. maybe with just a bit of yield I'll get a better distributed contact?

@DanStro, I looked both at yield and ultimate. This is part of a "safety critical" component. Any permanent yield within normal working load would be considered unacceptable. I've changed the part in non-critical ways so I could show it on here, but the real part is showing a yield at 110 lbs, ultimate failure if I remember, like I said I wasn't really paying attention to it was in the 150 lbs range. Anyway, the actual part my company makes, is in service and rated for continuous use up to 400lbs. That's a long way from where I'm at in FEA.

The other part of the whole thing, there is a press condition with the brass ears under the "T" on the .375 thick steel web. Is there a way to simulate a press fit on those surfaces in solidworks? I haven't used contact sets very often.

That being said any luck on hand calcs? my copy of Roarks formulas is about 100 miles away right now... anyone see anything of the sort in there?

Thanks!

 

[RoarkS]

Ah! I just found out there are two of them working together to share the load. but still... 400 lbs is cutting it close if the FEA is anywhere near correct.