LRFD vs ADS Wood design 2

[erik1938]

I am just curious as to how many people use ASD or LRFD for wood design. I am a new graduate and I have learned ASD wood design, but I think that LRFD may be more efficient as well as less cumbersome.

Any thought?

 

[DaveAtkins]

I use ASD, but then I am an "older" engineer (just turned 43 yesterday). One advantage to ASD is that you don't have to use two different sets of loads to determine stresses and deflections, as you do with LRFD.

DaveAtkins

 

[rday]

I use ASD for wood. Having not looked at the LRFD manual for wood, I think ASD would be less cumbersome. I agree with Dave, only one set of loads in ASD.

Of course, as steel and concrete use limit states design, it would be nice to use one methodolgy all the time.

Rik

 

[SacreBleu]

LRFD makes sense for steel and concrete, but is just an additional "bookkeeping" problem in wood. When the Code-writers devised the factors for wood LRFD, all they did is calibrate back to ASD, and in most cases, you end up with the same size.

 

[archeng59]

I use ASD for wood design. have not attempted LRFD yet.

 

[WillisV]

Agree with SacreBlue...wood LRFD design is a straight conversion from ASD with no noticable efficiencies.

 

[rday]

To those who have looked at the wood LRFD,

Are the load factors appreciably different from concrete or steel?

How about the resistance factors?

Rik

 

[SacreBleu]

rday,
Me, I don't recall - I attended a Wood LRFD seminar years ago, and we decided then it was a "lost cause". I suppose it may appeal to someone needing an exact analysis of combined gravity load/wind/seismic members, such as the compression of the post at the end of a shearwall, but that is just sharpening the pencil way too much.

 

[CraigICE]

LRFD is the answer to a question nobody asked.
It became a part of life when we started designing concrete with the Whitney's stress block. It doesn't make much sense for wood because deflection governs much of the time and wood quality varies.

Where steel is concerned, you might save 5% of your framing weight but check your floors for vibration.

Where concrete is concerned, multiply the sum of your loads by about 1.55 and you'll be within 5% of the LRFD answer. 5% doesn't mean much -- I designed a house in Guam for 170 mph gusts and a few years later Guam saw a typhoon of 170 mph sustained winds. Seismic zone II stoped in a straight line at the Oregon/California border about 25 years ago. Now western Oregon is zone III per '97 UBC and parts of the coast are zone IV.

Hence, when we don't know what the loads will really be, and nothing gets built the way we design it anyway, fine tuning loads with LRFD only makes sense to people who don't do civil-structural engineering for a living.

 

[bjb]

CraigICE: I agree with you 100%.

 

[theonlynamenottaken]

No matter what we think of the discrepancies ASD is being phased out. If you've looked at the new AISC manual you know that ASD is now gone for steel, to use "ASD" you just back out to service load level after calculating nominal strengths. NCEES is going to change the SE1 exam from ASD wood to LRFD wood (although they promised they would change by 2003).

I'm 27 years old and am in the "grey" area. I have to know both methods because: 1) I work under older engineers that won't even discuss methodologies developed after 1965, 2) I work on alot of international jobs and nobody but the U.S. seems to use ASD anymore. Most universities are teaching LRFD (because those professors are the ones mainly writing the new codes), so it seems like once the now ~30 year old engineers retire we'll be totally on LRFD.

 

[bjb]

I have looked at the new combined AISC spec-it is similar to the AISI spec for cold formed steel in that you calculate the nominal capacity and then divide by a safety factor for the allowable if designing by ASD. I prefer the ASD approach because I can use the same loads for strength and deflection checks. I have no problem with the "increased accuracy" of the LRFD approach, except that for steel and wood, deflection usually governs for beams. Maybe it's a moot point if a computer is used to perform all designs, but I like to do calcs by hand except when a computer is really necessary-say for caclulating deflections for a multi-story rigid frame. Some of the engineers that I have encountered who totally rely on the computer have had their judgement suffer as a result, in my opinion. I digress.

For wood, given its inherent material variability, and the fact that we really don't know what the loads will really be, I think that an LRFD approach is just an exercise in academic rigor. The academic community is trying to ram it down our throats, yet they don't have to perform designs to a schedule (often inadequate) and a budget(usually inadequate).

 

[CraigICE]

Question -- is it possible to organize a forum or group regarding LRFD and possibly SI (System International) units?

Even publications by organizations and the publishers of textbooks usually have several serious errors regarding LRFD and no one attempts to publish a design example from start to finish because of the possible errors that will crop up.

Maybe the engineering community can put a stop to LRFD if we speak with one (more or less) voice.

Comments?

 

[UcfSE]

Why would you want to stop LRFD? I can understand resistance and all, but stopping progress? If we stopped with what has worked in the past (why change what works) then we'd all be designing using 10,000 psi working stress and rivets, or structures might require 10^6 slaves and 30 years to build.

 

[JAE]

UcfSE,

Hmmm...cheap labor, no OSHA, unlimited materials, no government restrictions, no schedule except for your own desires (as local god)....sounds like heaven for some folks, like emperors or dictators. And no lawsuits either!

I agree - LRFD is frowned upon right now because its new and requires engineers to learn new techniques that they've grown accustomed to.

But I like CraigICE's quote: "LRFD is the answer to a question nobody asked."

 

[UcfSE]

It does sound great from that perspective lol.

I like the quote too. Maybe it's my youth talking ...kids. But then, wouldn't the question be, "This is fine and all, but is there a better way to do this?"

 

[bjb]

I am all for continuing research and expanding on the state of the art. When it comes to an LRFD approach for wood I have a problem though. Given the natural variability of the material, and the fact that the quality of construction on wood projects can often be a problem, I want a design method that has conservatism built in to it. The approach that I like (so far) is being used by AISC and AISI, where you figure out the nominal strength, and then either divide by a safety factor for allowable strength, or you can use the load and resistance factor approach. Personally, I don't like dealing with factored loads, I prefer keeping track of one set for strength and deflection calcs. Is the allowable stress method that is typically used for designing wood so deficient that there is a demand or a need to come up with a new approach?

Also, I don't think that the advances in structual engineering can all be attributed to an LRFD approach. I think you can be skeptical of LRFD but still in favor of expanding the state of knowledge.

 

[DaveAtkins]

Another engineer explained it well to me once--with concrete, ultimate strength design works well BECAUSE DEFLECTION IS RARELY AN ISSUE. With steel and wood, deflection is an issue much of the time; hence using ASD makes more sense. With masonry, like concrete, deflection is rarely an issue, so ultimate strength design should be OK.

DaveAtkins

 

[aggman]

I agree with DaveAtkins 100%. There are just too many cross check load combinations using LRFD with wood and steel. This is why everyone wants to stay with ASD when it comes to steel design and wood design. The PHd guys who are pushing the LRFD (mostly) are not having to apply it to every day design life and they don't truely understand the problems in dealing with the book-keeping. Funny that AISC got all kinds of practicing engineers to sit on the board with their new code and we now have a newer ASD spec to work with! To me I don't think there is a lot of benifit in using the LRFD. Like LRFD in steel design you are supposed to generate some weight savings due to its use. I guess I don't see the point when the typical design utilizes unity factors of 90% or less. If we were so concerned about every last ounce of weight every member in the structure would utilize 100% unity factors. I would say the guys that would sign off on the design would be few and far between and god help the fabricator dealing with all the member sizes!

 

[BubbaJ]

I'm a young"er" engineer (31) and I learned ASD from the engineer I worked for before I even started college. I agree we shouldn't stop progress, but I also agree that we shouldn't be "sharpening our pencil" down to the nub either. Like bjb, I still do some calcs by hand due to the simplicity of most of the buildings I work on. Even if you are using a computer, doesn't one still need to be able to recognize or easily check the answer one gets? How can you do that if the only design methodolgy you know involves 50 steps? Just because you can design down to the in*lb, doesn't mean you should. CraigICE makes some good points.
Foremost, nobody should be telling us that we have to stop using design processes that are proven to work. As long as clients are happy (ok, moderately satisfied) and structures are safe, why does it matter whether we used ASD or LRFD to get there?