Steel Yielding

[abusementpark]

When you are inspection an existing steel element (beam, column, plate, etc.) that has undergone some visible deformation, would you be able to tell by visual inspection if any yielding has occurred?

 

[Ron]

If the indication is from simple bending, unless the deflection was drastic, you can't tell if it has yielded or not. Other localized deformation might be telling, but unless you can unload the piece, you can't tell very easily. It will also depend on the stress mode of the piece.

 

[connectegr]

Extreme deformation is telling, but inconclusive. I agree with Ron that the loading must be removed to identify the cause.

http://www.FerrellEngineering.com

 

[ishvaaag]

Yet of the visual inspection you can appraise the deflection, then back-analyze and get some guess.

 

[civeng80]

I would agree with Ron. Deflection is one thing while stress is another. Do a deflection check to see what is the expected deflection and compare with whats existing. That will give you an idea whats going on.

 

[EngineerTex]

Beams will often receive a camber from the factory (or a third party who specializes in this) so that they will be closer to flat when in service.

Unless you know exactly how it looked before being placed into service and exactly how it looks with no load on it, I think you're going to have a hard time. Unless, like has been noted above, it is a drastic or extreme deformation.

Engineering is not the science behind building. It is the science behind not building.

 

[paddingtongreen]

I defy anyone to look at, measure, or whatever else, a loaded member and be able to assess the deflection or stress in it unless there was an exhaustive survey recorded before loading.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[B16A2]

Draw your stress strain diagram paddington. Once you hit yield the deformation will rapidly take off. If it yields in bending, it will be at a plastic hinge, this should be visually obvious. Oil canning is not yielded material. any local deformations at connections, etc is likely yielded.

 

[Lion06]

B16-

I would say that SOME yielding in bending is not going to be noticeable regardless of the visual inspection or the stress diagrams we can draw.

We assume an elastic-perfectly plastic stress-strain curve, but that's not reality. If you have some yielding in say the outer 1/8" of the flanges, I don't think that you'll see the deflection rapidly take off. You would essentially be replacing that 1/8" at each flange with a steel that has an E of wherever it happens to fall on the curve.

Maybe I'm way off base, but I don't see that making a drastic difference in deflection.

All of the calcs you could do also assume that you actually know the exact yield stress, which you likely don't.

 

[ToadJones]

I agree with Pad & EIT.
There might be nothing more theoretical than a "plastic hinge".
The name itself implies a theoretical situation.

 

[B16A2]

Actually, hinge mechanisms are more reality than uniform yielding. Same applies to reinforced conc. But what do I know, I only see members tested to failure once a month in the lab.

Dont forget about built in residual stresses and imperfections, these are not trivial in yield behavior.

 

[Lion06]

B16-

You raise an interesting point regarding the residual stresses, but I think that only complicates the matter even more in terms of doing a stress calc to determine yielding along with a visual inspection.

 

[BAretired]

If residual stresses are taken into account, members can be locally stressed to yield strength at working loads.

BA

 

[vandede427]

why don't you measure the deflection, then try your best to estimate the load on the beam, that could give you an estimate stress in the beam.

 

[Lion06]

I don't think you can realistically measure deflection unless you had a survey done before any loading as paddington notes above.

 

[civeng80]

Structural EIT

Get a laser level and take a level at the support then at midspan.

I think you would get a pretty good idea of existing deflection.