ST Welded Connections - Did Anyone Know This?

[Norby_acn]

The firm I work for has always designed our own connections, we don't delegate to other firms, and as such we've got a pretty good library of connection details. On a recent project, we used a detail that we have used selectively for 20 years... an all-welded ST connection that is intended for short, shallow beam connections with loads too high to develop into bolts. In this case we were using an ST5x17.5 with 3/8" fillet welds at the supporting and supported member. Don't ask why we used this connection or suggest alternate connections, that is not the point of this post.

Where we got a surprise was when we got a call from the fabricator and he said that he couldn't fit the size of weld we had specified on the stem of the ST to the supported beam. He said that the stem of the ST5x17 actually tapers from the "k" region down to the end of the stem, where it measured between 1/4" and 5/16" thick, limiting the size of fillet weld to 1/4" thick. The AISC Manual lists the stem thickness of member as being 0.594" and doesn't say anything at all about the stem tapering.

We reached out to several special inspectors we've used in the past and asked why no one had ever brought this to our attention before. When using this detail, we specify multi-pass welds at the stem, but the stem couldn't have multi-pass welds on it because it isn't thick enough for multi-pass welds. We checked RAM Connection and RISA Connection, they both assume the stem of the ST is a consistent thickness as given in the AISC manual for the design of connections. We even reached out to AISC for comment and haven't heard anything back yet.

Is this common knowledge in the industry? Did I (we) somehow miss the boat on this information? Are the other shapes that don't actually conform with the AISC given dimension and are smaller than what is listed?

This may have been a little rant-y of a post and I apologize for that, the information has opened a can of worms that has just been very dissapointing.

Thank you,

 

[dhengr]

Norby_acn:
I have seen some specialty tees which do have a tapered web (stem), but I don’t think this is one of them. That ST 5x17.5 is cut from a std. beam, an S 10x35; and it does not have a web which changes thickness. It does have flgs. which taper to about ¼ or 5/16” at their tips and terminate in a radiused shape. The AISC flg. thk. is an average thk. about at the grip point on the flg. Those 3/8” fillets are pretty large welds for that member. If you really need that much weld you might be better off using a larger tee section, for a longer but slightly smaller weld. I wonder that your welds may check stress-wise, but that the real failure mechanism wouldn’t be the stem just above the toe of the welds. A sketch of your connection detail would help a great deal in this discussion, I can’t see it from here.

 

[hokie66]

Yes, I think the fabricator is confused. The web thickness of an S shape doesn't vary, but the flange does. For the shape in question, the web should be a constant .594", perhaps with a bit of tolerance. And even for the flange which varies in thickness, the edge will be thicker than the fabricator says. Do you know where the steel comes from? Something here does not pass the pub test.

 

[ATSE]

I would have to see this with my own eyes to believe it. 0.594" down to 5/16"? Either a communication breakdown, or serious failure of basic tolerances.

 

[dik]

The thickness of the flange varies slightly, but not within the same 'group' and the thickness of the web changes for S Shapesl. The depth changes slightly, but not withing the same 'group'. See the table, below.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik