Load Testing Requirements

[Rabbit12]

Involved in an interesting situation.

On a few of the jobs we are working on there are existing stairway/walkway platforms that had been installed. We had been asked if we could add some load to the walkway platforms (piping/electrical conduit). So we analyzed the bracket supporting the platform and determined it didn't meet code requirements. We had an initial conversation with the manufacturer and they sent us what I'd call a "napkin calc" showing the stress in the angle of approximately 26 ksi. This exceeds the allowable stress of 24 ksi.

So, we developed a detailed calc package using AISC equations that showed an overstress of approximately 60% and sent this to them. They didn't didn't provide any calculations back, but provided some load test results.

I can't seem to find much info or requirements on load testing. Are there any standards? If a structure is required to carry 2 kip per code and the load test proves it can carry that 2 kip is that acceptable or is there a factor of safety that must be applied?

There is no doubt in my mind that the bracket doesn't meet the theoretical code requirements (AISC), but is it possible to circumvent that by doing a load test?

 

[bookowski]

ACI chapter 20 has provisions for load testing. ASCE 11 might also have some.

 

[Rabbit12]

Thanks for the response bookowski. I'm dealing with steel so am curious if AISC has anything published.

IBC section 1715 has some info on load testing and would govern if AISC doesn't have anything.

 

[Ron]

In general, if a proof load is applied to 2x the design load and there is no failure, it is acceptable. This applies to many constructed systems from foundations through the roof.

 

[wannabeSE]

To corroborate Ron's 2x the design load, see 2012 IBC section 1709

 

[dhengr]

Rabbit12:
I’d load test those brackets (or some test samples) or the brackets in a typical platform arrangement, for the bracing effects, with a load of (1.5 to 2)(2kips, or whatever the new loading is), thinking in terms of (.6 or .67)(Fy) as a design bending stress level. And, not allow any yielding, buckling or dislocation/breakage, etc. of the connections to the primary structure. The original manufacturer should have to prove his system or live with your proof, when you are the one signing off on the new design. Although, he didn’t provide his brackets assuming your new additional loading, so his brackets may not pass. As an alternative, to increase the cap’y. of the platform, could you just add a couple more of the old brackets to the platform system. It seems that so many of those types of manuf’ers. haven’t cracked a Statics or Strength of Materials text in years, maybe ever; and they really think they don’t have to know the design codes into which their product must fit, and be used. They do often know that their product has not failed yet, but neglect to think or mention that structures are not often loaded to the full design load, or to enough loading over the design load to actually cause yielding or distress. So, their design must be o.k.

 

[JStephen]

It's not clear from your post if these are new items that were recently installed, or had been there 20 years, or whether they were designed per AISC in the first place.

 

[Rabbit12]

Thanks for all the responses.

The manufacturer provided us load test data. Per the IBC load testing procedure reference above, the brackets can't support the OSHA specified concentrated load of 1,000 lbs. At a load of 2,100 lbs the bracket exhibits permanent deformation. Divide that by 2.5 per IBC and you get an allowable load of 834 lbs.

I won't even discuss the possibility of carrying the IBC live load of 60 psf for platforms and 100 psf for stairs.

Jstephan, it's a mix of both. I would venture to guess there are hundreds of these in service and they are likely manufacturing more everyday.

I agree with the post from dhengr, these manufacturers seem to be oblivious to current codes and standards. That was made evident by the fact they sent a calculation showing a stress of 26 ksi. Another interesting tidbit. The angle in question is a 2x2x1/8 with numerous holes punched to address different fit up conditions. If they use a 3x3x5/16 they would have met all code requirements. All this to save less than $5.