HSS Col-Beams from four sides 3

[JoeBaseplate]

I want to verify the changes I want to make to this attached obsolete detail. I have a HSS col with WF beams coming in from four sides. I can get a thru plate for connection from two sides but I am not sure about the other two sides. Could I use welded plates (as marked up) which are not thru plates? Thanks.

 

[nutte]

Is this a shear-only connection? If so, and if your plate thickness and HSS wall thickness satisfy the proper ductility ratio, you don't need through-plates at all. The shear tabs could weld directly to the column face on all four sides.

 

[hawkaz]

I prefer welding a WT section over a shear tab on the face of a tube. Bolt as if this were a shear tab. Make the WT a little bigger or a little smaller than the column width to allow for welding.

 

[JoeBaseplate]

hawkaz, thanks. Use of WTs seems like a good idea, and cheaper too.

nutte: Its sort of complicated, I have moment connection (wind frame) from one side with 8k-ft moment, so not a lot. Rest are shear connections. Where would I find the ductility ratio info in the steel book?

 

[frv]

I agree with nutte.. We use shear tabs on HSS columns all the time. See AISC 13th Ed., Sec. K1.5 for the ductility check nutte mentioned. Remember, KISS.

 

[connectegr]

Depending on the HSS wall width and thickness, a face welded shear plate is the best option in most cases. As noted this detail and design considerations are shown on the 13th Ed manual. Your detail as shown with thru plates must be welded at the center intersection. The partially inserted plates provide no advantage or reinforcement of a thin wall column. In many cases it is more economical to increase the wall thickness, then to use through-plates. Intersectting plates require that the column be cut and the welded cruciform plates inserted, then the column repaired/spliced. This is obviously easiest at the top or bottom of the column. The face welded WT detail has been shown in several publications. Note however, AWS does not advise the use of fillet welds in prying. The bending forces are concentrated at the top of the welds which can cause the welds to "un-zip". The flange thickness of the WT must be adequately thick to eliminate prying.

http://www.FerrellEngineering.com

 

[ToadJones]

connect-
where is this noted in AWS?

 

[connectegr]

Toad
I will attach references when I return to my office. The issue is that prying applies concentrated forces at the top of the weld and at the root of the fillet. These concentrated forces can propagate a crack, which significantly reduces the weld strength.

http://www.FerrellEngineering.com

 

[ToadJones]

okay...i was wondering for WT hangers.

 

[connectegr]

Hangers would have the same prying concern. The flange thickness can be increased so that weld is in tension without prying.

http://www.FerrellEngineering.com

 

[connectegr]

Toad
Page 8-15 in 13th Ed manual. The detail advises against the use of one sided fillet welds in tension. In the case of a WT flange the condition is similar, if the welds are separated by a distance and material in between allows bending. Similar to the single plate, the toe of the flange rotates and the weld stress is concentrated at the root. If the flange is designed for bending, the weld stress is tension only applied to the thickness of the weld.

http://www.FerrellEngineering.com

 

[ToadJones]

connect-
thank you.
I recently had a detail passed across my desk for a multitude of hanger type connections. The hangers were fabricated from WF shapes with the top flange being welded with fillet welds along the sides of the flange to a larger WF shape above.
I didn't care for the detail at all and asked the engineer whose did the design (another company) if he investigated prying-like forces on the connection. He was irritated and the conversation did not go well. He essentially cast me off as if I were inventing a problem. I could not remember where I had seen or read that this was a bad idea but at the same time I wasn't looking either.
He told me that the material had been order and cut for the hangers.
All I could come up with for a suggestion to eliminate prying was to cut the flange width down.

 

[ToadJones]

wow...sorry for hijacking the thread

 

[connectegr]

Toad
Another fix is to provide partial depth stiffeners. Prying increases the tension forces on a bolted hanger connection. It is not a stretch to expect the same with the welds located at the extreme width.

http://www.FerrellEngineering.com

 

[nutte]

Connect, how thick do you have to go to eliminate the "unzipping" force on the weld? You can size the WT flange for bending, but no matter how thick it is, it will still deflect, however slightly. How thick is thick enough?

 

[ToadJones]

I think Connect is suggesting that the thickness to eliminate bolt prying is sufficient....but the bolt gage is narrower than the beam flange (obviously). That would lead one to believe that the thickness required for the welded condition would be larger than that for traditional prying of bolts.

 

[connectegr]

When this detail cannot be avoided, we develop the thickness for bending. Theoretically the bending check will allow some deformation. But the bending allowables are conservative, and the thickness should be relatively stiff. It should certainly provide enough stiffness to resist rotation at root of the weld. We don't use the welded flange hanger detail or the WT shear connection HSS walls. For the hanger detail we typically use a bolted connection, and design the flanges for prying. For HSS column shear connections we use shear plates.

http://www.FerrellEngineering.com

 

[connectegr]

Toad
The thickness for prying is insufficient. The design for prying allows for flexibility. But, the thickness required to eliminate (theoretical) bending is sufficient to use a welded connection.

http://www.FerrellEngineering.com

 

[hokie66]

A contrary view...although I respect connectegr's advice, I don't use face welded shear plates on HSS columns unless the loading is very light. Through plates or tees for me. Seems illogical to make the column thicker just to accommodate a simpler connection.

 

[ToadJones]

Hokie-
I agree, but sometimes one column may have many many connections on it. A heavier column may lead to significant cost savings.