Bolted Joints and Length of Engagement

[tc7]

I am selecting bolts to secure a machine base to a 2" thick steel foundation plate. The load is primarily in tension. The bolts will be installed into blind threaded holes in the foundation plate. I am using the joint analysis method described by Shigley, to determine the joint constant. The method requires some guesswork to make a tentative bolt selection so that you can base the subsequent calculations using the tentative bolt geometry (for instance I had to guess that my prelod would be equal to the applied load so that I could identify a bolt offering a greater proofload). This resulted in a couple of iterations before I determined a bolt I was content with.

This joint constant method requires that only a small portion of the actual threaded hole depth be considered in the calculation. But when I get to the end of my calculation and am checking thread shear, (by ASME B1.1 Appendix B calculations), the formula requires "length of engagement" (L sub e). For this calculation, shall I use the full actual depth of thread engagement? or shall I use the only the portion of thread depth used in the joint constant calculation.

I hope I my question is clear, any advice on the above is welcome.
Thanks,
Tom

 

[desertfox]

Hi tc7

I would use the actual thread engagement of the actual joint and I would also ensure that this was the minimum amount of engagement by taking account of component tolerances.


regards desertfox

 

[diamondjim]

Conservatively, I would use the
full depth minus 1 pitch at the
end of the screw and subtract the
chamfer size at the face where the
internal threads begin.

 

[Screwman]

You hit it Diamondjim, take the total 'effective' thread length. For a grade 8 bolt and LCS plate you should be able to break the bolt in tension when you get to just about one fastener diameter of effective engaged thread length. This is a rough rule of thumb, but it will help serve as a reality check to your calcs.

Dick

 

[cage]

If you use engagement length at least equal to the thickness of the applicable std nut, which presumably has been designed with adequate thickness.