Opinions about the new ASD 3

[mrMikee]

I'm one of those engineers that has stuck with the old ASD. As I read over the new ASD (Allowable Strength Design) I realize that there is a huge difference in, well, almost everything. There are a few things that I like and am thinking about putting into existing calculation programs, but this would be a mixing of the old and new at least at the start. I don't see a problem with this except for the fact that the new spec probably hasn't been adopted in any building code yet.

If anyone has a comment about mixing the two, or comments about the new ASD in general, it might make for an interesting topic.

Thanks,
-Mike

 

[aggman]

As a newer engineer I was trained in college to use LRFD. I used LRFD for about six months after I got out of school. Then I became involved in my first big project and my supervisor wanted to use ASD. (Because he didn't know LRFD and had no plans on learning it.) So I learned ASD at that point. I really liked it mostly because we were always dealing with service load analysis. It was easy to say the stress is such and such and the allowable is such and such. It was just easy to relate to. Anyway to answer your question I think I really like the new spec. At first I was leery of it because it looked so much like the LRFD I soon forgot out of school and I didn't want to give up the simplicity of the ASD equations. After further investigation, mostly by actually reading through the sections and following articles in Modern Steel Construction and Structural Engineer I think I am really going to like it. I especially like the more advanced stability checks and single angle checks. Other things are very much still back to their core in the ASD 9th edition when you break into the equations. So I guess I am saying that I am looking forward to using the new spec. I think it will cover some of the gray areas younger engineers have between our older supervisors. Take this example. We were designing a pedestrian bridge that had a pony trusses. After completing analysis of the bridge I was concerned that the top chord was being stabilized well enough on the ends to assume k=1. My supervisor says "Just check the vertical members under their full design loads, adding 2% of the top chord load out of plain." I said yes, but what if that does not provide enough stiffness. He says just make sure your deflections are small. I ended up checking it my way checking the stability and we were both satisfied. What I am getting at though is that we younger engineers have been seeing all these derivations in college and generating these formulas in the LRFD code. We also learned why those formulas are their. I think this will help advance analysis techniques in the design office and also help newer engineers and older engineers communicate design concerns better. I will be interested to see what some of my older and wiser engineers think of the new code.

 

[14159]

The new ASD is nothing more than a modified 3rd Ed. LRFD using omegas instead of phi factors. 89 ASD folks are bound to hate this. I actually can't think of a single provision that's similar, although I'm sure some are. I developed a design procedure for one specific component using both 89 and 2005 ASD and the new one was probably 3x as complex. Didn't bother me because they were teaching LRFD by the time I was in my MS program, but I sincerely feel sorry for 89 ASDers.

There is a large positive side for some folks that I haven't heard much about yet. If you develop software, design aids, or whatever, based on the upcoming Spec., it is a snap to use either ASD or LRFD. I think ASD will be rare, though, because you have to have a large L/D ratio before they break even.

 

[mrMikee]

Thanks aggman and 14159 for your comments.

The two aspects of the new ASD that caught my attention were combined stresses and single angle struts. Actually it was 14159 in one of the previous threads that brought up the single angle changes. I'm anxious to get going on the new spec but never having studied LRFD I'm wondering how hard it's going to be to see the logic of all this on my own. I hope a good text comes out soon.

What I plan on doing is to update my ASD calculation program in parts and very carefully. I suppose it could be argued that it would be no harder to learn LRFD at this point, but I like the intuition of allowable stresses and actual loads. If anyone else is going to write their own programs and wants to pass around some ideas, then feel free to jump in. My program is currently in Mathematica.

I have to admit that I am surprised at how little has been discussed here concerning the new combined spec. I thought the ASD users would welcome a new spec as long as it wasn't LRFD.

Thanks,
-Mike

 

[dbuzz]

Without trying to sound harsh, why not just learn LRFD?

Surely the AISC is trying to phase ASD out.

 

[mrMikee]

dbuzz,

I knew someone was going to say that, and I wonder myself if maybe that wouldn't be the smart thing to do. The problem is for me to change to LRFD would take a longer period of time than to incrementally work in parts of the new ASD. I wouldn't want to mix and match ASD and LRFD (I think there was a thread about that recently) but working with both the old and new ASD might be a plan.

Thanks,
-Mike

 

[14159]

"...The problem is for me to change to LRFD would take a longer period of time than to incrementally work in parts of the new ASD...."

You will have just as much trouble learning Allowable Strength Design as LRFD. The calculations are completely identical up to the last step when Rn is either divided by omega (ASD) or multiplied by phi (LRFD). I can only think of one exception, the alpha term in the moment magnifier section. There might also be a different term in the prying action stuff, but I can't remember for sure.

You could purchase Salmon & Johnson for a good LRFD-based textbook. I have a couple others but they stink in comparison. If you're cost-sensitive and don't mind it being a little outdated, you could pick up a 3rd Edition, 1990 version that covered the 1986 LRFD Manual. It's not terribly different than the upcoming 13th Ed. Manual. Try alibris online for this. I've found several older textbooks for great prices, for example, the McGuire classic for $31. If you got the newest S&J, you'd be good to go. From a LRFD perspective, the upcoming Spec. isn't much different from the 3rd Ed. LRFD Spec, therefore, the current S&J would serve you fine.

Another thing. There will be an extremely comprehensive Examples Manual released at about the same time as the Manual. That should be in December, last I heard. The Examples Manual has both ASD and LRFD examples for 99% of the Spec. and Manual provisions. It'll be another one of AISC's awesome free downloads. Try to find something like that from another organization.

14159

 

[mrMikee]

14159,

I've been tempted several times to get the latest S&J but was thinking about waiting for the next release. I'm not even sure there will be one since their first edition was in 1971. That's over thirty years ago. I don't want to admit which edition I have but I will say that I took my first steel design course many years ago from Jim Fisher the Chairman of the Committee. If the ASD9th would last another 16 years I could retire.

It's nice to know that S&J is still a good text and I think I'll go ahead and get it. I need to do this while I'm in the mood. Thanks for the help.

Regards,
-Mike

 

[dbuzz]

mrMikee

I can understand the frustration. I was the first generation of students in Australia whose studies were based on the limit state method.

When I graduated ten or so years ago, many of my senior colleagues had only ever dealt with the allowable stress method. It was initially tough on them to change, with their rules of thumbs and 'intuition' based on allowable stress.

Now it's almost unheard of for any practitioner to use allowable stress, except on occaision in the oil and gas sector where some American standards specify it.

 

[ajh1]

The new spec is referenced by the new IBC 2006 that should come out in early 2006.

 

[chichuck]

mrMikee,

I too was raised on ASD. When the new code came out, I wanted to try it out. I developed a new mathcad template that does the calcs for the axial-bending interaction on a wide flange shape. This parallels an older spreadsheet I had that does the same calcs using the ASD provisions. The new mathcad template runs about 7.5 pages, the old ASD version ran 2 pages. Granted the new one includes slender elements while the old did not (A992 steel has some w-shapes with slender elements, A36 steel has not). I've noticed that mathcad takes up more space to do the same calculations, what with displaying all mathematical equations and such, I can do much the same thing in less space in a spreadsheet.) But the new code provisions will, I think make for a longer calcuation.

I found the new code to be more logical and "flowed" better than the ASD code did though. I've posted both the old asd spreadsheet (asdbeamstresses.xls) and the new mathcad template (AISC 2005 beam stresses.mcd) on my website at:

http://home.mn.rr.com/gbbadger/

please check them out, I'd very much like feedback from others on how you think I've done. And feel free to use any of the files on my webpage if you'd like.


ALso, I do not think it is a good idea to mix and match the two codes in one calculation. Pick one or the other.


BTW, mrMikee, I also took a course from Jim Fisher, though it was basic statics. He taught at UW-Milwaukee back then. Is that where you went to school?


regards,


chichuck

 

[mrMikee]

chichuck,

Yes, I'm a UWM grad. My BS is in Mechanical Engineering and when I wanted to go back for a MS in Structural Engineering I found that the first steel design course was never taught in the evenings. Dr. Fisher offered to let me do the homework and come in for the tests. I was doing some design at the time so I didn't need much help, but there are professors that wouldn't have bothered.

My initial reaction to both LRFD and the new combined spec was that it was too complicated. Looking back at the old specs and the tools we used at the time, I guess overall it's really no worse when compared to what we use now, mathcad, spreadsheets, etc. I can see why your 2 page program turned into 7.5 pages.

I don't plan an mixing LRFD with ASD but am thinking about using both the old and new ASDs, at least until I can fully make the change. Do you not recommend this also? My interest now is the single angle strut provisions which many engineers with ASD (and some code checking programs) don't do correctly, at least where I have worked. I am currently working on a project now that has some heavily loaded single angles and I don't want to go through the z and w-axis manipulation if I don't have to.

I will be checking out your website too. While I don't use Excel or Mathcad, it still is interesting to check out other peoples programming. I use Visual Basic or Mathematica and like them both, but have to say that sometimes Mathematica can make you crazy.

Thanks,
-Mike