Compression on HSS Tube Wall

[carnahanad]

Hello,

I have a condition where I have W8 outriggers bearing on top of a continuous HSS tube beam. The outriggers place a torque on the HSS tube. I want to make sure that I'm checking the wall of the HSS tube will be ok for the compression force created by the couple force. My thought is to treat the wall of the tube as a column that is 8" wide (Flange width of the W-shape) and 12" tall (height of the HSS tube). I now have a rectangular column that I can check the compression capacity. Does this make sense? I think if I use my imagination, I can use Table K3.2 of the 14th ed steel book, equations K3-9 - K3-12; H[sub]b[/sub] would be my flange width and B[sub]b[/sub] would equal B.

Thanks for the input!

 

[DaveAtkins]

I think you are being very conservative--the load will spread across a larger width than 8".

How is the wide flange fastened to the tube? Do you have web stiffeners above both walls of the tube?

DaveAtkins

 

[carnahanad]

DaveAtkins,

I'm planning to weld the Wide Flange beam to the tube across the flange and to the tube with a flare bevel weld. I was also going to add stiffeners than are in line with the tube wall. Good questions, I meant to add that info in the original post. I may also end up adding a weld that is parallel with the beam flanges across the top of the tube.

 

[KootK]

I think it's a reasonable, if somewhat conservative approach. Conservative, at least, if there are beam web stiffeners over the HSS wall. Without stiffeners, the width of load delivery at the point of application may well be less than the beam flange width.

I believe that this could also be treated like a compression load delivered to a wide flange web. You know, web yielding and crippling. That should give you a good deal more capacity and be just as simple to calculate.

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.

 

[Istructeuk]

I’m just wondering how wide the outrigger and the back-span is? Is it okie to get to site with something like a “fish-bone” if it is welded together?

You may refer to Table 9-2, AISC DG24 for out-of-plane Bending T connection.

 

[DaveAtkins]

If there is a back-span, the tube won't see any torsion.

DaveAtkins

 

[carnahanad]

Thanks for the input everyone.

A few notes from the responses:
1. There is no back span. this outrigger creates a "bench" on an elevated floor. The main building structure is the floor, the tube sits on the main building structure and then the outrigger beams sits on the tube to get the proper height for the bench. To make it even more interesting, the main building structure is already cantilevering out 20ft (with a 40ft back span). It's an inception of cantilevers!
2. I'm planning on using stiffeners in the beam over the tube walls. Thinking it through more, I may actually do away with the wide flange and just use a tube outrigger to avoid stiffeners all together.
3. Currently the length of the outrigger is 8ft, with 6 of those feet as occupiable space and the other 2 are exterior.

What I'm finding to control design is the attachment of the outrigger to the tube. I believe I can use the welds in table K4.1 equation K4-7 as long as I include stiffeners in the wide flange beam.

Thanks again!

 

[theonlynamenottaken]

Due to the lack of redundancy I'd certainly sleep better at night (if I were you) if the welds were, 1) shop performed for higher quality / consistency and since they'd most likely be done in an overhead position in the field, and 2) were specified to have some NDT beyond minimum visual inspection.

You probably have already considered this but I thought I'd mention it for the benefit of others that seek out this thread in the future.

 

[dhengr]

Carnahanad:
I’d fillet weld the tips of the W8 beam (outrigger?) flanges to the top of the 12" high tube. Then, the real design/analysis problem is the high bearing stress (reaction) right under the W8 web, and right over the corner radius of the HSS sill/base member, and its load path. Then, as a bearing seat for the W8 beam, I would diagonally cross cut a W12x26, or some such (the final height is the important thing), so that you had about 6" of length at the top (under side of the W8 bot. flg.) and about 3" at the bottom, and square cut the end at the tube web. Cut the top flange off this seat bracket, leave the bottom flange on it, so that the top surface of the bot. flg. hits the tube web about at the tangent point of the tube corner radius. These bearing seat elements get fitted and tack welded to the W8 bot. flg. Then roll the W8 so you can do down hand fillet welds seat web to W8 bot. flg. Roll the W8 back into place and fillet weld seat web to the tube web. Clip the top corner of the seat web and don’t even try to weld into the upper corner of the tube. Weld a few inches of the top surface of the seat bot. flg. to the web of the tube, down hand. Done.

You could replace the 12" high tube side sills with 15" channels, toes out, and butt the bottom 3" of the W8 to the inside of the channel web, with fillet welds on the W8 web and top of bot. flg. to the channel. You could look for some cut offs of an MC18 about 3' long, and shop weld them to the ends of the W8 as bearing members on/to the conc. slab. Over the web of the channel provide a piece of plate to match the elevation of the top of the W8 and weld that to the end of the W8 and over the channel web. These end pls. would be about 5" high with the C15" and about 2" high, or not needed, for the 18MC. Then the channels get bolted to the floor.

 

[XR250]

I would think that excessive serviceability rotation of the tube would be a limit state before you have wall buckling