Plate to HSS connection

[TempStructures]

thread507-410454

My first post here...I've encountered a conflict in the field (I'm a contractor's engineer) where they are needing to weld a padeye to an HSS.

The HSS is part of a larger panel that will be tripped from horizontal to vertical (thus the padeye will experience loading perpendicular to the HSS all the way through parallel).

I have convinced myself that the loading perpendicular to the HSS is adequate using AISC K1-9 equation.

Where I am hung up is the case where the loading is parallel to the HSS. The shackle is located 3" away from the face of the HSS, so there is a moment into the HSS wall when the panel is vertical. K1-10 addresses this case only indirectly, stating that the connecting plate needs only be weaker than the HSS wall. In my case, the padeyes have already been fabricated and are NOT weaker than the walls. Nor do I feel very confident K1-10 covers the case where the load is applied eccentric to the FACE of the HSS...the wording seems to imply eccentricity to the CL axis of the HSS is covered.

I see the need to check bending in the wall of the HSS, but am reluctant to perform an F.E.A. because I want to keep it simple. I looked at the CIDECT section on longitudinal plates welded to HSS (pages 159-161 of the following:

http://212.150.245.105/shared/eBooks/DG 9 english.pdf),

but they don't seem to address the load offset from the face of the HSS.

See attached annotated photo. Any help or thoughts are appreciated.

 

[SAIL3]

a sketch of the proposed connection would help with loads and direction of lift, etc.....applying the requirement that the connection pl need only be weaker than the HSS wall is extremely dangerous in this application...

 

[TempStructures]

SAIL3: I attached a photo to the original post with some redlines showing the proposed padeye and direction of loading.

I agree K1-10 provision isn't adequate to address this load case.

 

[Lomarandil]

I've dealt with this exact situation before (well, the padeye size was constrained by something else, but same result).

We may be looking at different steel code editions -- my K1-10 (14th Ed) refers to a transverse plate T connection. But I'm pretty sure we're talking about the same limit state, for a longitudinal plate T-connection under plate shear (in 14th, that's K1-3). That case addresses the eccentric load.

Either way, if you can get a theoretical padeye which is weaker than the HSS wall to calc out, it doesn't hurt that your physical padeye is oversized. That's the approach we used to solve the issue.

If you can't get a theoretical smaller padeye to work, the usual solution is to knife the padeye through both walls. It's a bit of a pain, but less expensive than building an FEA model (presuming you don't have several hundred of these to trip).

 

[jayrod12]

Wouldn't it be cheaper to weld a doubler plate on the face of the HSS as opposed to cutting a slot to knife through?

Maybe it's a locale thing, but the steel guys here freak out if we ever try going through an HSS column.

 

[Lomarandil]

Possible Jayrod. I've always heard that as the going wisdom. Yet somehow, whenever I go observe the solution in the field, they've knifed through instead.

I might just be seeing the unicorns though.

 

[TempStructures]

Thanks Lomarandil and jayrod12.

Yes, I was looking at 13th ed., so we are talking about K1-3 in the 14th Ed.

Assuming a theoretical weak padeye approach makes sense to avoid punching shear. However, if you take an extreme example and assuming Rn is acting 10-ft away from the face of the HSS, it seems to me this large resulting moment would easily yield the wall in bending, not punching shear. My assumption is that K1-3 (14th Ed.) is only addressing shear loads and NOT moments. I am looking at moments. I found another post where JoshPlum posted a forthcoming AISC equation addressing bending in HSS walls here:

http://www.eng-tips.com/viewthread.cfm?qid=418241.

That feels more like what I need to be checking....just don't know where that equation comes from.

I am working with the field toward an option that involves choking the panel with nylon slings and forgoing the padeye altogether. Mainly because I don't feel confident in this section of AISC.

Yes, we are lifting hundreds of panels so the nylon slings are going to wear out as they roll over the HSS sections. We'll have to have several standby slings.

 

[Lomarandil]

If it's only addressing pure shear loading with no eccentricity, where would the "punching" component come from?

I was interested to see how the new spec (2016) treats it. Interestingly enough, they remove all rectangular HSS checks from section K and refer you to section J4 (with an effective length proposed by K1-1). J4 appears roughly the same.

The commentary references a punching shear check on rectangular HSS walls -- but it's in the Manual, not the spec, so we can't see that yet (gah!).

Sorry to hear that you have so many of these... those nylons are going to get pricey.

 

[TempStructures]

Lomarandil: Yes, the punching does imply moment. Good point.

I found another equation posted over here:

http://www.eng-tips.com/viewthread.cfm?qid=418241

This looks like more what I'd expect...presented in terms of moment, not shear.

 

[Lomarandil]

Yes, I saw that thread (hence my digging into the 2016 spec) -- it looks plausible on first glance, but that's all I can say about it.

If it makes you feel any better, moment is just a figment of our imaginations anyway.