LRFD or ASD? The saga continues... 12

[vmirat]

I have a project being designed for us by another contractor. It involves a simple support bracket system for HVAC duct work. The structural engineer did calcs using AISC ASD method.

Whenever I see ASD used on design, I usually ask why, out of professional curiosity. Here was their response:

"Designing a large structure with large quantities of steel one should use LRFD to take advantage of the cumulative weight savings. That is not the case here, so ASD was used for simplicity and speed."

I'm wondering if we are schooling new engineers in both methods for this reason? I didn't take this any further, but I wonder how they decide the break point for LRFD vs. ASD.

 

[dcarr82775]

Toad, my apologies, I didn't read carefully enough.

It must be because I was taught LRFD, but I always calc the allowable moment in ASD instead of the stress so my mind always thinks in moments

 

[nutte]

Toad, section F12 in the black book (2005 specification) addresses unsymmetrical shapes. This wasn't even addressed in the green book (old ASD, 1989 specification). How is adding information and guidance a bad thing?

 

[ToadJones]

nutte-
I guess I disagree that old and new ASD are, at the root, the same thing. New ASD is Limit State Design, right?

 

[ToadJones]

Nutte-
Look closely, did they really provide guidance? Or, did they say "we added this section and it is basically worthless so if you run into this situation, do a research project".

I assume most structural engineers have had to analyze unsymmetrical shapes. Unfortunately for me, it comes up quite often in analyzing large built-up crane runway girders. Procedures for this analysis were laid out nicely in AIST Technical Report 13 which was last updated in 2004 and still used 1989 ASD. Obviously, TR 13 needs to update procedures. Meanwhile I dont have the time to develop special software for determining the critical buckling stress of wildly unsymmetrical shapes.

I really don't want to get into an argument. I use new ASD regularly, but there are situations where applying it are nearly impossible. I don't think everyone has had to deal with this, so they figure its "all good".

 

[ToadJones]

I guess old ASD really had limit states built in as well...so you are correct.

 

[nutte]

[quote}New ASD is Limit State Design, right?[/quote]

By Limit State Design, I assume you mean that it permits moments greater than that achieved when the extreme fiber is at the allowable stress. Old ASD did the same thing; they just hid it, as frv alluded to. That's why for compact wide flange shapes, you're allowed to take Fb to 0.66*Fy, greater than the regular 0.6*Fy.

 

[ToadJones]

...the fact that they were "built in" made my life a little easier

I guess I am just sour grapes

 

[vmirat]

One of the conflicts that I've seen is in foundation design. You can use ASD load combinations all the way down through your structure, but when you get to design the concrete foundations, you're back to LRFD type load combinations. But then the soil mechanics side of foundation design doesn't include load factors.

 

[connectegr]

We work on a wide range of projects, commercial and industrial. The overwhelming majority of industrial structures are still designed using ASD. While nearly all large commercial projects are LRFD. LRFD provides the most economical design when forces are primarily dead loads. When live loads are significant, such are manufacturing structures the ASD and LRFD provide very similar results. For most small structures the difference is insignificant. This is the practical reasoning.

But, the decision can also represent a generational gap. Prior to the early 90's the only option was the AISC ASD Manual. However with the introduction of the blue LRFD manual, AISC made a commitment to develop only the one code. Universities began teaching only LRFD and as a result all research was applied only the the LRFD manual and the two subsequent editions. In addition to the experience of our aging designers many industrial firms had significant investment on their ASD software. In the worst case ASD would provide a conservative result, which was little reason to make a change. Larger architectural engineering firms were repeatedly told the of the economic advantages of using LRFD. Many designers also began to see the ASD manual as antiquated. As young engineers advanced in their careers, their LRFD roots became more influential in their firms. Also the LRFD is the primary design principal in concrete and the only design option in many of the countries. In early 2000's AISC changed their direction, it was determined that in the interest of promoting steel it was a good idea to support both design methods. The research and development of the previous 10 years was applied to ASD and the result is the 13th edition manual.

In most cases the selection of method is a matter of choice, and comfort level. The only problem we see is when designers try to combine the two. As connection designers we find projects with both service and factored loads.

http://www.FerrellEngineering.com

 

[connectegr]

A few years ago, I found designers using LRFD to design the structure, but showing service loads on the contract drawings. Their reasoning was that LRFD would potentially result in a lighter structure. But, factored loads were larger and un-experienced fabricators and detailers would raise their bids with the expectation of larger connection sizes and more complexity. I have also seen projects with factored loads on the floor plans and service loads on the elevations.

http://www.FerrellEngineering.com

 

[connect2]

Vmirat your foundation comment,
Right on the geotechnical Load Factor Resistance Design, LRFD, Limit States design, call it what you will. In Canada it has only been limit States Design for steel and concrete for 25 years, good god I need a holiday. But the geotechnical crowd was only brought in to limit states 5 years ago, so you carried factored and unfactored loads where ever you went, dabbling between the codes and standards with metric, and then the contractors with Imperial units, we also have two official languages in Canada. Go figure.

 

[271828]

IMO the Black 2005 book is a cookbook specifically made for run-of-the-mill situations encountered in basic building design. It has left out in the cold more complicated structures encountered in industrial design.

If the 2005 Manual is a cookbook for basic building design, then what in the world would you call the 1989 Manual?!??! LOL

 

[271828]

I think the heart of the argument is, and always has been, intuition or "gut feel" when designing with the "old ASD".
It is easy to relate calculated stresses to an allowable or to yield.

"hmm, this bending stress is 29.5 ksi, that's getting close to yeild"
vs.

"hmm, this beam has 245 kip-ft max moment that's getting close to...umm, ahh, ..."

Being a native LRFD guy, I hate to say it, but what you've typed there makes a TON of sense to me.

 

[hokie66]

...and a TONNE of sense to me.

 

[SAIL3]

there are a tonne/ton of theoretical concepts out there that, at first glance, look like an elegent solution but quickly break down when one attempts to apply it to a real world problem...I put the LRFD and it's load combinations in that catagory, at least, the AISC's treatment of it ....
in contrast, the ACI code(IMO) has presented the LRFD concept in a much more clear, concise and logical manner...something a practicing engineer can apply efficiently....
frv...I don't know what a "predictable reliabilty index" indicates, is it a theoretical measurement?....
it's all a mute point, anyway, since the horse is out of the barn or the money is out of my wallet(which is really the name of the game)..
my basic gripe with LRFD and it's load combinations is that it gives a ficticious load/stress level in the stucture, a state never reached in reality...so in the real world the structure remains in an elastic state(seismic excluded)at a much lower stress level than indicated by by the LRFD load combinations...
as a practicing engineer, if I know the real load/stress level in a structure I can use my engineering judgement based on that...without that, I am compelled to to go the route of the cookbook or "lego-set" engineering approach that is becoming more common nowadays...

 

[ToadJones]

271818-
Not sure what you are getting at....

By specifically providing provisions for "unnsymmetrical shapes" and then subsequently providing very little information or procedures on how to design these shapes, AISC has made a mockery of itself and completely side stepped the issue (one that they created).
The Green Book didn't do this.

It's as if the academic folks posed a problem to themselves that they couldn't solve and then rather than tackle it, they skipped class and went for an early afternoon ale.

 

[frv]

Sail3,

For some more information on LRFD and specifically the reliability index, see the commentary to B3.3, p.16.1-214 to 216 of the AISC 13th Ed. Manual.

As far as the "fictitious" state of stress, as I mentioned before, the stress level is just as fictitious using ASD (again, see Wooten's 3rd Law).

 

[steellion]

The simple answer is: it doesn't matter.

While LRFD may lead to more economical designs from a strength standpoint, it just makes it more likely that serviceability criteria like deflection or vibration will control.

While using stress design may feel more "real", the new ASD and LRFD are both based on plastic capacity. They were developed because it was determined that limiting the steel to a fraction of its yield stress was overly conservative. There is a lot of plastic capacity that can be tapped into for the controlling-case gravity and lateral loads.

 

[ToadJones]

"There is a lot of plastic capacity that can be tapped into for the controlling-case gravity and lateral loads"

This is the part I struggle with intuitively.
Where is this capacity?

 

[271828]

By specifically providing provisions for "unnsymmetrical shapes" and then subsequently providing very little information or procedures on how to design these shapes, AISC has made a mockery of itself and completely side stepped the issue (one that they created).
The Green Book didn't do this.

It's as if the academic folks posed a problem to themselves that they couldn't solve and then rather than tackle it, they skipped class and went for an early afternoon ale.

Section F12 provides a codified basis for evaluate a member other than those covered in F1 through F11. It's for rare situations, hence its location.

If F12 didn't exist, the engineer would hopefully compute Mn for an unsymmetric member exactly as currently shown in F12. However, someone such as a calc reviewer could accuse him of not providing a design per the legally adopted specification.

I don't understand "couldn't solve." F12 is for unknown sections. Just about any section that might be needed, in other words. How can the academic community, Specifications Committee, or anyone short of God preemptively supply the solution? It's assumed that the engineer will compute Fcr using whatever means he chooses (FEA, manual calcs, elastic buckling solution from a book, etc.), and then use F12 to get to Mn.