AISC Error or me Error?

[Guastavino]

Friends,

So in trying to analyze an existing composite beam I went back through the motions of the AISC 6th edition (ASD of course) and I noticed the divide by 2 in formulas 18 and 19 of chapter "I" regarding shear connectors (They are the 0.85f'c*Ac/2 and As*Fy/2 for those that don't have an old copy handy). Meanwhile, the new 2005 and beyond specs got rid of the divide by 2 which makes sense. However, the "technical reason" per the AISC document "COMPARISON OF ANSI/AISC 360-06 to 1989 ASC Specification" (except below) says that the reason they took it away is because the 1989 and prior equations applied to half the span and the 2005 and beyond apply to the full span, to which I humbly say POPPYCOCK! It's clear as day (to this humble but accurate member) in I3-1c of the 2005 spec that it's related to the max positive moment point to the point of zero moment.

I'm ready to eat crow if I'm wrong, but I'm thinking the old spec and new spec are literally a factor of 2 off, albeit it seems as though the allowable loads for connectors are now higher, so maybe just the explanation by AISC is indeed what is wrong.

Thoughts, comments, anything structurally heretical in my argument? I'm not trying to just "be right", I'm just genuinely confused as to the real reason behind getting rid of the factor of 2. There is an old thread on here about this, but it didn't seem to answer the question for me.

 

[Agent666]

I'm with you, their reasoning seems wrong. There never should have been a /2 factor.

 

[Hokie93]

The factor of 2 in the denominator of Eq. (I4-1) and (I4-2) from the 1989 AISC Specification is a factor of safety that reduces the flange compression to a service-level load. The reference to half-span versus full-span is erroneous as far as I can tell.

 

[Guastavino]

Hokie,

Thanks. If you're correct on the factor of 2 getting things to service level, it really goes against logic in some ways that the spec is written like that. Basically, that will ensure that the PNA will never enter the steel section. Also, I don't see how you can use a "fully composite section" if you can't transfer the full 0.85*fc*As into the concrete slab since you are dividing it by 2 to get the required number of shear connectors.

Maybe I just need to analyze the section using the old AISC spec and realize that the working stress method is what's causing the confusion here. I can't get a fully composite section out of a pre-1989 working stress designed composite beam. These old equations just don't seem to logically make sense.

Any thoughts?

 

[Hokie93]

I think you are seeing the result of ultimate strength-based research being modified to fit into an allowable stress design methodology. The result is not always logical and can be quite conservative. My best recommendation is to consult a steel design textbook that was based on an older (ASD) AISC Specification or pertinent AISC Engineering Journal papers. A good textbook example would be the second edition of "Steel Structures - Design and Behavior" by Salmon and Johnson.

 

[Guastavino]

Hokie, good idea, I found an old steel book that said "This procedure is prescribed by the AISC specification, except that the flange compression is reduced to a service-load value by dividing by a factor of safety of 2, so that the horizontal shear is given by: AsFy/2 and 0.85f'c*b*t/2"

However, again, that seems to have the factor of safety on the wrong side of the equation to me. basically, what it seems to be saying is that we are going to design based on the whole concrete section being engaged, but then we are going to ensure that only 1/2 of the studs are actually provided to get us there.

I remain confused at how they justified this.

 

[wannabeSE]

I have never tried hand calculations on composite beams per the 9th edition or earlier. Perhaps, the difference is due to the different methods used to calculate capacity. In the older specs, elastic theory is used. While the current specs use plastic stress distribution.

Are you just curious, how it was done in the past? I have used the current specifications with plastic stress distribution to determine the capacity of existing beams. It gets a little tedious when the beam has a cover plate.