HSS Column Single Plate connection subject to out of plane forces 1

[RFreund]

Is there any information on designing a single plate or gusset plate connection to an HSS column (or WF for that matter) that is subject to out of plane loads. Particularly when the plate is welded to the face of the HSS.
How would you check the HSS face for this concentrated moment?
Would check the weld using some sort of T=C=M/d force couple?

Thanks!

EIT

http://www.HowToEngineer.com

 

[jayrod12]

I'd likely be leaning towards welding a WT to the column. If space permitted, I'd make the WT wider than the column and then you could fillet-groove weld them together. Then you don't need to worry about the wall bending aspect.

 

[FoxSE14]

I have also done as jayrod suggested...try to avoid concentrated forces at the centerline of HSS tube and instead dump the loads directly into the column's sidewalls.

 

[KootK]

1) design the plate for M/Zx flexure.
2) design the wall connected to the plate for M/Zx flexure assuming that the other walls don't participate.
3) assume the effective width of the wall to be the same as the plate length.
4) use a pair of fillet welds at least 5/8 the plate thickness. There's stuff in AISC suggesting that's enough to effectively mobilize the yield strength of the plate.
5) due you due diligence in combining the out of plane design with any other significant effects present.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[RFreund]

Thanks for the responses!

EIT

http://www.HowToEngineer.com

 

[FoxSE14]

Perhaps your loads are small, but replace that plate in KootK's sketch with an appropriate WT mounted with flange against HSS and you'll likely get:
1. better deflection performance, (check WT prying in flange)
2. more economical welds (flare bevel) to the HSS due to increased moment arm between welds (the 'd' of T/C = M/d in KootK's sketch is very small)
3. no local stress concerns on the HSS wall (if WT flange is of appropriate width to dump loads directly into HSS sidewalls)

 

[HankTexas]

I like KootK's approach but you might want to consider using a full penetration weld to make sure you get the same strength as the plate.

 

[KootK]

To clarify:

From a mechanical perspective, I very much like the WT solution as well, particularly for larger loads. I simply focused my efforts on the "how do we engineer this particular configuration" question rather that on alternate strategies. Which is not to say that alternate strategies don't also constitute sage advice.

While I do like the WT system mechanically, there are some potential drawbacks to consider:

1) I get push back from some fabricators regarding the cost. I think that the shear tab fabrication is something that is more a part of standardized fabrication procedures in some shops.

2) I get push back from some fabricators on the use of WT's. Many claim that it's cheaper for them to plate fabricate the WT's instead of ordering them. And, once the tees become plate fabricated, you're back to having to worry about the WT flange/web weld again.

3) Using the WT's really just moves the HSS wall bending problem from the front face of the HSS to the back face. Although, I suppose, one could argue that the effective participating length of wall would be larger when looking at the back face.

4) I've experienced a lot of quality control issues with flare bevel welds. And I like them much better when used in longitudinal shear applications. With flare bevel welds, the couple will basically be resisted by the welds acting in tension. That's not prohibited, I just like it less.

5) With the flare bevel weld approach, you probably want the WT flange a bit wider than the columns. And that may be exacerbated by the fact that your WT web will likely be off center to allow the beam web to remain on center. Those things combined mean that, if you're trying to bury the system in a stud wall, doing so probably just got a little tougher.





I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[TLHS]

I've done what KootK is suggesting, but I've also on occasion been a little aggressive and assumed that the HSS wall has some frame action going on with partial fixity at either end.

Have a quick conceptual look at deflection, too. The rotation at the base of the cantilevered plate due to bending in the HSS wall could become a problem

I normally wouldn't use a WT if I were doing this. If I needed to reinforce I'd use fabricated plates. WTs aren't really the right proportions. You want a thick web plate and likely a not as thick flange. Basically, you end up with a weird WT shape being required, and you probably aren't doing many locations. So the shop buys a WT length that they won't use the whole piece of and you're compromising to something that isn't necessarily all that efficient. Might as well just have them build what you actually want out of standard plate stock.

 

[FoxSE14]

Thanks for sharing your experiences guys. I agree especially on your #5 KootK, great point.

Happy to report I have successfully used the WT approach, so maybe I just got lucky with my column being out in the open + friendly/passive fabricator. Not intending to hijack this thread, but my curiosity has been sparked.

I guess I really do not like the local deflections/high stress of the wall locally at the base of the plate (may have skimmed too fast, but I don't see anything in AISC Chapter K on this condition). Obviously since this was not my question, I'm speaking from the land of hypotheticals....Instead of a WT, what about:

1. Use a longer single plate and turn it into a through-plate connection. Same fillet welds on front and back of HSS or,
2. Single angle instead of plate

I like option 2 for ease of construction, material availability + once the forces are resolved into each of the welds, make use of Chapter K equations to rationally justify the calcs. (concentrated forces on wall rather than local, weak-axis moment) Thoughts?

 

[TLHS]

The load distribution in number two is going to depend heavily on the thickness of the angle legs. Because of that, I'm not really a fan.

If they're really thin compared to the HSS wall, your load is going to try to go through the first weld directly into the HSS wall as moment and never touch the leg of the angle that's parallel to the HSS wall. If it's super thick, then it'll work as you're intending. If it's somewhere in between, it'll balance load between the two cases.

To some degree, this holds for number one as well. If your through plate is thin, the stiffer load path could involve trying to jump the moment directly into the first wall it sees and nothing ever gets to the far wall. If your through plate is thick, then you're good.

 

[FoxSE14]

Yes, I generally agree TLHS. I guess it comes down to the column's B/t ratio and your particular loading. In the situation where I specified a WT, I was up near B/t ~ 40 + had strong & weak-axis moments from the beam as well. Happy Weekend Everyone

 

[KootK]

Fun.

I guess I really do not like the local deflections/high stress of the wall locally at the base of the plate

I used to worry about this with just the in-plane forces delivered to single plate shear connections. I'd worry that this would mess with the axial capacity of columns. I can't remember where I saw it but, someplace, I read that the local wall distortions are not thought to detract from overall column capacity. Not very satisfying of an answer, I know. And, obviously, it feels a lot better in the common case where you would have beams tying in in both directions and bracing the column where the distortion would occur.

1. Use a longer single plate and turn it into a through-plate connection. Same fillet welds on front and back of HSS or,

I love it mechanically and have a standard detail of this variety. I find that fabricators dislike this one the most of all options however. The slotting is costly. I reserve the through plate detail for hard working braced frame columns these days.

2. Single angle instead of plate

This solves a number of issues, particularly the spatial one if the beam web is fastened to the inside of the outstanding angle leg. You'd still have HSS wall bending going on though I think, unless I'm misunderstanding this.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.