How to decide a existing plate girder steel yield stress is 36ksi or 50 ksi?

[Hialoha]

I have to check a existing plate girder capacity. How could I decide the steel plate yield stress? I know normally it is A36, but it looks like it does not work for my load and I suspect they are 50ksi plates. Is there any way to test? Thank you.

 

[dhengr]

Hialoha:
There is a difference btwn. checking that plate girder for its intended loading and its max. cap’y. Try honing in on what the original loading was, and then check it and see what works. It sounds like you’ve kinda done that. Alternatively, it is not (was not) uncommon to design the flanges with 50ksi stl. and the webs with A36 mat’l. In the old ASD methods I always felt it was o.k. that the upper/lower few inches of web mat’l. might yield a little bit and just sluff some of their load/force off onto the flgs. The web really can’t go anyplace, it is constrained by the flg. as long as the flg. doesn’t yield much. Finally, find someplace on the girder, probably out near the bearing ends, where you can have a lab take some samples of the flgs. and web for chemical and tension tests. They can then tell you what ASTM Spec. the mat’l. meets. Obviously, you still have to check girder stability, web crippling and buckling, stiffener spacings, etc.

 

[ajh1]

You don't indicate the age of the structure. In today's marketplace the majority of plate labeled A36 comes in at or near 50 ksi yield anyway.

 

[fegenbush]

If you have access to the structure, X-Ray fluorescence testing is an option. Be aware, though, that the machines which can properly analyze grades of carbon steel are not prevalent. They are becoming more common in the marketplace, but they are still quite expensive.

 

[Canuck67]

You could have small coupon samples removed from the beam and sent to a lab for testing. This is relatively common in my area and most of the larger testing and inspection companies are familiar with the various protocols, sample sizes, etc... The results of the coupon sample testing is usually pretty definitive. Alternately, there are published 'guidelines' in some jurisdictions that summarize grades of steel historically used. This can be used as a guideline, but sometimes is not definitive.

 

[UBGrad]

Positive Material Identification (PMI) is one option. The equipment is extremely expensive and requires prep work to get to bare metal, but if you can locate a contractor in your area that does this work you can probably subcontract it out for ~$1000.00.

The cheapest and likely easiest option is to contact the owner of the girders and gain access to the original drawings. Certainly, they would have called out the material specifications on the original drawings.

 

[Hialoha]

So far after talking with an couple of XRF analyzer venders, it looks like X-Ray fluorescence testing (one kind of PMI) can't distinguish two alloys of A36 and A529 because XRF is checking chemical composition and these two alloys have ovelapping chemistries. So cut a coupon and test is the option. Thanks for all response.