Allowable or Acceptable Beam Lateral Displacement 1

[IFRs]

Is there a source for allowable lateral displacement for a steel i-beam after placement? I have several W12 x 16 beams that are not straight. I am not yet sure why they are not straight but am tasked with evaluating an acceptable tolerance. They are about 37 feet long and about 5 feet apart. They vary from straight within mill tolerance to a several inches of lateral bow at midspan. None are bowed beyond the elastic limit of the flange (about 13 inches). They are in an industrial setting, support steel roof plate over an unoccupied space with no architectural or aesthetic concerns. On both ends, their bottom flanges are welded to steel angle supports. I've looked through a few steel construction manuals but they seem to be concerned with architectural issues, affecting subsequent alignments and load paths. This all steel structure is very simple with no other connections, walls, etc. I'm only concerned with their value as rafters and their ability to resist environmental loads applied to roof plate that is laid on top of them in contact with their top flanges. They were installed around 1980, have performed without failure to date but in documenting this structure we found that a good number of rafters are not straight, some visibly with by eye.

Essentially my question is: at what lateral midspan deflection do simply supported rafters have reduced carrying capacity?

Thanks

 

[KootK]

I don't know of any standard other than the sweep tolerances in AISC. I'd say that you're a long way off from having any problems though. Analytically, it's just a beam with a bit of torsion in it. If you're feeling ambitious, you could work out what displacement would result in a warping normal stress in the flanges of 10% (or whatever) and call that your limit. This is steel plate roof deck rather than corrugated cold formed deck? How is the deck fastened to the beams?

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.

 

[IFRs]

Thanks for the response!

The 3/16" thick deck plate is simply laid on the beams, no connection other than friction. I'm considering using twice the mill tolerance for sweep as a starting point, that would be about 2". This deflection stresses the flange to about 15% of yield. The roof plates are by no means flat, they warp from welding and just because, as you walk on them they oil can a bit.

 

[hokie66]

A W12 x 16 goes off the chart, unbraced, at about 10'. So a 37' span...no load capacity. I suppose the plate friction has held it in place until now, but it will probably move further with temperature change, etc.

 

[KootK]

I was assuming the top flanges to be braced by the deck. If they're not, lateral torsional buckling capacity would be reduced by both the absence of bracing and by the lateral sweep. The sweep could even be the result of lateral torsional buckling in action.

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.

 

[IFRs]

Plate friction is used for weak axis bracing under load. For the dead load case, the top flanges are tied together with steel flat bar at the appropriate intervals.

 

[KootK]

Friction is intentionally used for bracing?!? I don't see how the flat bars help matters unless they themselves are terminated at something rigid. Are they?

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.

 

[SAIL3]

I would be very wary in relying on friction to develop bracing in this situation......

 

[IFRs]

It may be unusual, but friction is used for bracing for live loads. The flat bars tie the midpoints of the rafters together and one bay has multiple cross-braces for the full length. Lateral stability is not so much the topic of this thread, although it is worthy of discussion. I'm focusing in on how much lateral deflection can be tolerated in an as-built condition. I was hoping to find an L/1000 or similar guideline, or some acceptable multiplier of mill tolerance. We have used a laser to scan the structure and are extracting from that the rafter lateral and vertical deflections. There are 175 rafters. I hope to establish how many would need to be replaced and then decide if the structure is salvageable.

 

[Once20036]

I would likely figure out what load in the weak axis would create the deflection that you're observing and consider that load to be present. The deflection in the calcs would likely be the deflection beyond what is allowed per AISC tolerances.
If you're feeling motivated, you could look at the weak axis curvature and see if the bend more closely resembles deflection from a point load or from a distributed load, but that might be going too far.

Then check for combined strong & weak axis bending based on your real vertical load and "imaginary" lateral load.

 

[IFRs]

I'm not sure what caused the deflection ( sloppy erection, compressive loads, ponding, etc ) but in any case they only see dead loads now and friction is preventing them from returning to straight +/- mill sweep. This portion of the process is not to figure out why are they bowed but if the bow or sweep is acceptable and which ones need to be straightened or replaced. Straightening may be as simple as hitting with a dead-blow hammer with the roof plate pushed up. But there are a lot of them and if I can rule out 50% it will be appreciated by the owner and the field hands. What do you think of using twice the mill tolerance for sweep as a guideline?

 

[KootK]

I'd say that lateral stability is very much the topic of this thread if:

1) it may be the cause of the displacement.

2) it may be they be the the most likely disastrous outcome of the displacement.

Truly, if LTB can be shown to be fine, I doubt that any other effects will matter enough to justify scrapping ainy of the beams.

Can you supply a sketch of the framing, strap bracing, and cross bracing?

Do all of the beams sweep one way or present a pattern?

What is the nature of the live loading?

What kind of temperature swings are typical in this environment?

Are the beams oriented vertically to begin with?

What is the detail at the ends of the beams that provides rotational restraint there?

Is the steel plate really a bunch of smaller steel plates side by side? How are they fastened to their neighbors and to the walls or framing lines that stabilize the diaphragm?

Or are you only interested in specs for sweep tolerance at this time?



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.

 

[KootK]

Didn't see your last post before I submitted mine. I still think that it comes down to LTB though. If you're confident in the efficacy of the bracing scheme, I'm sure that all of the beams are fine as is. Induced weak axis flexural stresses can be stressed back to normal in the plastic state so long as they're not needed to resist a real weak axis load. I thinks that it's tough to separate what's acceptable from what's going on in the first place.

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.