Bolt Pull Out Cold Formed Section

[BadgerPE]

Hey all,

I am trying to develop a standard detail for a mezzanine railing connection to a cold formed floor system when a concrete deck is not present. Basically I am looking to provide blocking between joists which the railing base will then bolt through the flanges of the the blocking. The bolts will carry the tension load to the blocking. After that, clips and screws will transfer the tension load from the blocking into a continuous header.

My question is at the base/blocking bolted connection. I am thinking that if I use the perimeter of the bolt head x thickness of the flange, I can develope a gross area which I can then determine the capacity using block shear rupture equations for cold formed steel. Does this sound like an appropriate approach or are there other equations that deal with bolt "pull out" for cold formed sections?

 

[Ron]

CJ47...the failure mode is usually not in direct shear. With bolts pulling out of thin sections, there is usually a lot of deformation causing the hole to stretch, the material under the bolt head to stretch and the bolt to pull through the enlarged hole. Empirical testing is the better way to determine this.

 

[BadgerPE]

Ron,

Thanks for the info. Is there any way to determine the capacity without testing?

 

[BAretired]

How about FEA?

BA

 

[Ron]

Yes, FEA can do it, the only problem is getting the constraints and boundary conditions right. You could treat the flat part of the flange as a plate with a fixed boundary or you could treat it as fixed in one direction with tension springs in the other direction. Spring constant would be the yield strength of the plate. You could also use the spring condition all the way around the plate, but I don't know if that would cause the model to "blow up". I've not tried anything like this with all springs.

A more "mechanical" FEA package such as ALGOR, NASTRAN, or similar would probably be better than what we typically use for structural analysis.

 

[trainguy]

Check out the rivet pull-thru analysis video on the NEi website.

http://www.nenastran.com/newnoran/animation3.php

If going the FEA route, you need NL static capability with contact. Even then, you'd want to corroborate with physical testing.



tg