Bracing of Gerber girders for LTB

[mb3928]

This problem came up in the context of adding solar to a large warehouse built in the 1970s. Construction consists of steel W shape columns with cap plates, W21x68 Gerber (cantilever) girders with a column-to-column span of 50' and a cantilever of 7'. There are no web stiffeners at the column connections which I know is probably deficient. Atop the W21x68s are W12x14s which are also constructed as cantilever joists and spaced at 8'4" OC with a girder-girder span of 25' and a cantilever of 4'. The deck is 8'4" x 2' x ~3" precast concrete C channels which are clipped to the W12x14s. The manufacturer's literature for the precast planks provides an allowable load of ~120 psf.

The structural engineer (I am the contractor) is saying that the W21x68s need bottom chord bracing at every location they cross a W12x14, and that the W12x14s need continuous bottom chord bracing in (2) locations per span.

I am by no means an expert, but this seems excessive to me and I am seeing that most current specifications e.g. AISC 360-16 only call for bracing at column locations. Is the specified bracing reasonable?

 

[driftLimiter]

Its reasonable sounding, wind uplift design may require bottom flange bracing. Hopefully the designer has a good reason for needing the braces, if so I see no reason to contest the design. Hard to know without the details of the building. I suspect that wind uplift is the reason why one would need bottom flange bracing similar to what you describe.

The lack of stiffeners over the columns is especially concerning. That detail can negatively (and significantly) reduce the strength of the column.

 

[KootK]

I too find your engineer's recommendations to be quite reasonable. Even with a very aggressive design approach, it has been my experience that these systems will want bracing at these locations as a minimum:

1) The supports.

2) At, or near, the cantilever tips.

However, depending on the nature of the connections involved, the existing design might work like this:

3) The W12's brace the W21's torsionally. I'm skeptical of this given the proportions involved and the fact that the beams run over top of the girders rather than into the sides.

4) The slabs brace the W12's torsionally. Again, the connections that I'm used to seeing for such a situation would not usually support this.

Whether the system was designed this way or not, this may well be what's been making it work thus far. Based on what you've told us so far, my guess is that this addition bracing is probably needed even without the addition of the solar panels.

I recommend having a chat with your structural engineer about the potential for torsional bracing action. It's a long shot but you never know.

For what it's worth, I think that it's great that you've been upfront with us here about your being the contractor in this situation and your questioning your engineer's recommendations. Many contractors in your position will post here pretending to be structural engineers and generally trying to trick us into helping them bully their structural engineers. As you can imagine, we don't like that so much.

 

[dik]

If you're thinking of Gerber, you might want to look at plastic design instead... similar issues and generally less costly.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[phamENG]

I'll be a third vote for sounding reasonable. Depending on the support configuration for the solar panels, these things can cause a considerable increase in wind uplift. As a result, we're required to review the design and check it using current code requirements. That includes stability considerations that were often...overlooked...in the past. That's what this is.

It may seem excessive at first glance, but more than likely, this building has never seen its design load. (Design load being the worst case situation the code requires us to consider in design.) So even though it's been standing for 50 years as is, 1) it probably hasn't been tested in a meaningful way and 2) the addition of the solar panels will increase the destabilizing load condition we're concerned with here.

I've seen some of these conditions fail, and it's pretty scary. I don't blame the engineer on this project for going back and upgrading connections that might "work" for strength but have glaring stability issues that could compromise the structure in other ways.

dik - plastic analysis might let them eek a bit more out of the existing structure, but if I recall correctly it doesn't change stability requirements - right?

 

[JLNJ]

Patterned loading can be a killer for these systems as well. I think you probably want stability braces at the columns and at the first joist in on the main span. Stability failures are sudden and ugly - the opposite of plastic system failures.

 

[mb3928]

Thanks all, after a more careful examination of the unbraced lengths the engineer was able to reduce the amount of bracing specified, although a good amount is still required. An interesting learning experience for me.

KootK, thanks, not my style to try and trick anyone. Be gentle with my fellow contractors though, we end up with the short end of just about every stick.

 

[jayrod12]

... we end up with the short end of just about every stick.

Until you've got a chance to charge for an extra, then it's bend over Mr. Engineer.

 

[dik]

Stability issues remain... it generally allows members to be reduced in size, and the number of pieces handled is generally reduced... and alternate loading patterns pretty much goes away.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik