A36 vs A992 in Wood Construction

[bigmig]

I know that title sounds strange. Someone told me the other day that there is no way an A992 steel beam ever comes close to developing its allowable stress capacity because of the concept that you cannot brace a steel beam's compression flange in wood construction to the extent you can brace it with say, a metal deck or some other concrete/metal composite system. The crushing and giving of the fasteners in a wood plate give just a enough that things are not as well braced.

As a result, this person designs all of their steel beams, in a wood construction setting, with an Fy of 36 ksi instead of 50, regardless of the fact that they are using A992. Has anyone else ever seen this?

 

[fegenbush]

If the span were short enough, it would not be necessary to brace the compression flange. My opinion is that this person has tables that they have been using for years and they are trying to justify why they aren't getting with the times. I suppose conservatism is all well and good so long as you're not losing bids or overrunning costs because of it.

Example:
W12x14, 20' Unbraced length, maximum uniform load: 16 kip for Fy=50 ksi, 12 kip for Fy = 36 ksi (Ref: AISC Steel Construction Manual, 9th Edition)

 

[XR250]

I have never designed a steel beam in wood construction that was controlled by strength. Deflection always seems to be the limiting state.

 

[racookpe1978]

Little bit of his assumption might be "hidden" in the his wood-to-steel joint efficiency and deflection of the adjacent wood by the locally very high fibre stresses imposed on the wood by the "rigid" steel flange.

Assume for example, you have a steel beam W12 "pushing up" into a row of wood 2x10 joists. If the floor overhead is overloaded, the local wood fibers of each joist are going to 'squish" at the small area where the 2x10 crosses the beam flange. The whole W12 and floor will sag under the total load, but that total load is transmitted to the whole through a row of very small 1.5 inch x 3 inch contact points. So, if you had an "infinitely strong" W24 or W36 even, you'd still be squishing wood fibers in the joists.

 

[SAIL3]

This is a valid point, wheather one adresses it in that manner is another question. I would tend to focus on the connection itself of any wood pl/member to the steel bm..if I can adequately transfer the buckling load into the wood deck or bm locally and then analize that wood deck/bm for that load then I would feel more comfortable with the design...

 

[dcarr82775]

Considering I ofter see the plate connected to the beam with some nails in the side of the plate hammered down around the flange I tend to assume it is unbraced anyway.

 

[JAE]

I'm not sure you can even get A36 beams anymore. Most all W shapes are A992.

 

[AELLC]

Just limit the bending R ratio to 0.6, 0.7, your choice. Design in the real world is different from University where we have to come up with the exact beam size for full credit.

 

[Jerehmy]

Why wouldn't he just design it with Fy = 50ksi but have the unbraced length be the full beam length? That seems more appropriate.

 

[AELLC]

Jerehmy - that may be very conservative if the span length is very long. I don't think there is any clear solution. Perhaps just assume Fy=36 and design as Lb compact.

 

[Splitrings]

If you knew (or could estimate) the required bracing force, it would a matter of determining the bracing member(s) and the number of fasteners. Seems as though I have seen a certain percentage of the vertical load equates to the bracing force. 3% is the number in my head. I know I have seen it in AASHTO and in the SBCA for wood trusses. It would be an estimate that would at least give an idea if there is any chnace of bracing such a member.

 

[Lion06]

I don't understand the point of designing that way. There is no time savings and it is a lower bound approach. The tables in the steel manual provide allowable moments for all beams based on unbraced length. It's a three second look-up.

 

[XR250]

There is likely enough friction between the wood members and the nailer to properly brace the beam (assuming the nailer is fastened adequately)]
This seems more like an academic discussion than a real world practical one.
Just upsize the beam so you are less than 24 ksi and be done with it. I imagine if you go over that, serviceability is going to control anyway.