AISC 13th ed - chart vs. equation disagreement

[SLTA]

Hi folks,

In doing a review problem for the PE, I have found that a chart in AISC 13th ed. gives me a different answer than does the equation, and I'm hoping someone here can help sort it out for me.

Big picture: max shear force in a plate girder without intermediate stiffeners, using ASD.

AISC Section G gives Cv values for varying h/tw ratios. My h/tw is 128 and Fy=50, so Cv is by equation G2-5. I get Cv=0.267 and thus with Aw = 48"*3/8", Vn from equation G2-1 is 144.3 k

Then, I go to Table 3-17a in the beam section, and for h/tw=128 and a/h>3, I read Vn/(omega*Aw) = 5.1 for the max shear stress in ksi. 5.1 ksi * 48" * 3/8" = 91.4 k

Can someone please clarify why the two don't agree?

cheers,
Linnea

 

[msquared48]

Seems interesting that 144.3/91.4 = about 1.6, a very familiar factor...

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[SLTA]

lol Mike - that's exactly what I just realized! omega = 1.67 and there's my answer. At least I can sleep now!

*sigh*

 

[BAretired]

So, you are talking about Limit States Design (LSD) versus Working Stress Design (WSD), right? I'm sure this is going to create a few problems in the coming years in the U.S.A.

In Canada, WSD (Working Stress Design) or, as you call it south of the border ASD (Allowable Stress Design) is a thing of the past. For about twenty years now, Canadian engineers have used LSD.

Whether or not that is an improvement over the earlier system is a matter of opinion, but one thing is certain...as long as there are two ways of conducting structural analysis, you can be sure that practicing engineers will occasionally make the mistake of using service loads instead of factored loads and vice versa.

BA

 

[SLTA]

BA, that's exactly it. I learned LRFD in school and have picked up a bit of ASD along the way, but haven't used it enough to be super familiar with it. The practice problem was in ASD so that's what I had to use to ensure I got the right answer. Luckily, on the PE, we get to pick which method to use as answers are given for both ASD and LRFD.

 

[paddingtongreen]

A question that has bothered me. If you only analyze using LRFD/LSD, how do you handle the behavior under service loads, how do you check whether ceilings and drywall will not crack? You can't, you have to perform a working stress analysis.

If you have to perform a working stress analysis, why not factor that up for ultimate loads? I know, it's because we can save a little material doing an LRFD because WSD/ASD will oversize some members.


I hated LRFD when it came out and I still hated it when I retired. I think that LSD is a very appropriate name for it.



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

 

[BAretired]

paddingtongreen,

Service loads are calculated and used for service load deflections. They are then factored up for strength calculations. Different factors are used for live and dead loads in strength calcs.

You are not alone in hating it. I know people who feel the same way in my neck of the woods. But I really can't see what all the fuss is about. If you can't be bothered keeping track of the dead and live loads separately, use the larger factor and press on. Your design will be slightly conservative but who cares?

The real danger, in my opinion is having two systems in use at the same time. This thread is a good example of the confusion this can cause.

BA

 

[connectegr]

13th Ed is a big improvement in the application of the two methods. One formulas with appropriate phi or omega. For years AISC worked toward replacing ASD with LRFD. 13th Ed represents a change in this philosophy. In our work ASD is still the primary design method for industrial project. And LRFD is now used for the majority of commercial projects. Nearly all seismic considerations are based on ultimate strength. And ultimate strength has always been the ACI method for concrete design.

We write all of our design programs in ASD and LRFD and also for multiple steel specifications.

http://www.FerrellEngineering.com

 

[dik]

BARetired... I graduated 40 years ago and all our structures classes used Limit States... the concrete book was by Fergusson (sp?) and it had both... it wasn't until I started to get into masonry stuff that WSD or ASD came into play... for warehouses, I've been using plastic design for 40 years...

Dik

 

[BAretired]

dik,

I couldn't remember when Limit States Design became mandatory in Canada. I thought it was about twenty years ago but I could be wrong. I think it came about at different times for different materials. First concrete, then steel and then wood. Is that the way you remember it?

Limit States Design was taught at the universities long before it became the mandatory method of design in Canada.

I know of one engineer who still uses Working Stress Design for all his work. The fact that it is contrary to the current codes doesn't seem to phase him at all.

I guess it's hard to teach an old dog new tricks.

BA

 

[paddingtongreen]

Even though I disliked LRFD, I handled it okay except when it came to foundations. A foundation with LL, DL and WL, vertical and shear. Say under service loads, DL - WL(vert) and WL(horiz) the resultant is just at the edge of the kern.

The pressure is P/A +- Mc/I, Mc/I equals P/A for the service load condition and the resultant at the edge of the kern.

Should I just multiply the pressure components by the LRFD factors, or re-compute the center of loading? Reducing DL and increasing WL sometimes puts you in the position where you can't get there from here, the area under pressure gets too small, but the other method doesn't make sense either.

If I designed for LRFD and then checked for service loads, the pressures always seemed ridiculously low.

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

 

[nutte]

A little nit-picking...

lol Mike - that's exactly what I just realized! omega = 1.67 and there's my answer. At least I can sleep now!

Omega here is 1.50, making your ASD capacity 96.2 kips, very close to what you pulled off of the table.

 

[SLTA]

Thanks nutte - but when I look at the table, it says omega-v=1.67 at the bottom. I wonder if they changed it for this new version?

 

[nutte]

Hmm, I stand corrected. Only for rolled I-shapes, with h/t less than blah blah, does omega equal 1.50. The equation in question is in the next section, with omega equal to 1.67.