Braced frame design ASD '89 <--> ASD '05 1

[darkwing88]

I am using ASD '89 and Staad '06. I have a braced frame structure to design. Is there a way to set-up the model parameters to get results nearly identical to the results using ASD '05?

 

[Lion06]

Are you using the DAM?

 

[ToadJones]

I don't think there is much you can do within STAAD.
Newer versions I do believe have an ASD 05 engine and possibly DAM.
Funny part is, '89 and '05 probably wont yield too different results.

 

[ToadJones]

EIT-
I have been wondering for sometime now, if the DAM could be implemented in older versions of STAAD with P-delta capability.
I haven't given it tons of thought but is seems possible.

 

[darkwing88]

eit: Staad '06 does not have DAM option to my knowledge.

jones: I think your right about '89 and '05 not yielding too different results. However, I thought someone here may have researched the topic.

 

[SteelPE]

Toad

I gave it tons of thought as well and came up with the conclusion that my time was much better spent designing structures and not bashing my head against the wall. Now I have the pleasure of paying a yearly maintenance fee to Bentley for their software (the trap worked).

At the time I was extremely confused with how to apply the reductions to the member stiffness's required by the DAM analysis. I suppose you could avoid the DAM all together and use the other methods of frame analysis.

Note that if you want to use the DAM method. The commentary for Appendix 7 has two test cases that you can do by hand to make sure you are getting the correct answer from your software.

 

[ToadJones]

I haven't read this for a year or so, but I remember it being a good overview.

 

[darkwing88]

It is a given that the DAM can not be done properly with '06 Staad.

Possibly a conservative way might be to model the out-of-straightness displacements and set the elastic modulus to a reduced value. However, this method will result in Staad using the reduced modulus during the AISC checks.

 

[ToadJones]

DW:
Is it the reduced axial and flexural stiffness that '06 cant handle>?

 

[darkwing88]

jones: you just reduce the E value parameter to reduce the stiffness in selected members. So to answer the question, it can handle the reduced stiffnesses for the solver, however, the allowble stesses are based on these same stiffnesses. An unintended consequence.

 

[darkwing88]

So here is what I plan to do. Am I missing anything?

1) reduce the stiffness for all beams and columns; 2) model the structure either with out-of-straightness built-in or notional loads; 3) use factored loads; 4) use the P-delta solver; 5)'05 LRFD specification for code checking; and finally compare with 6) a linear analysis of the model with service loads, no reduced stiffness, no notional loads, and ASD '89 code checking for comparison.

The results will probably be very close to each other because it is a braced frame.

 

[JoshPlumSE]

First off, the DAM method isn't required. In fact, you can usually get away with ignoring stiffness adjusments for braced frames while still using a K=1.0.

Review section C2.2a (Design by 2nd order analysis) of the 13th edition manual for more details on when you can use this type of analysis.

Now, the real difficulty is associated with clause (2) which requires that your P-Delta analysis be performed under loads which are 1.6 times you ASD level load. Not sure if you can figure out a way to do that in the older version of STAAD. Honestly, not sure if the newer versions of STAAD even do this.

 

[ToadJones]

I believe you can use "REPEAT LOAD" to accomplish the 1.6 factor.
Works in a similar fashion to a load combination

 

[ToadJones]

Also, I dont think '05 STAAD has little p-delta

 

[darkwing88]

An aside: I think bueaucrats have infiltrated the AISC 360 committee. I'm trying to get a handle on that convoluted stability provision rat maze.

 

[ToadJones]

I am reading Chapter C for a review.
On page 22 of the spec. they list the variables for a "second-order analysis by amplified first-order elastic analysis".
Mlt and Plt don't make much sense to me. They are defined as:
"First Order moment/axial force using LRFD or ASD load combinations caused by lateral translation of the frame only".

Isn't the moment or axial force due to a translation an inherently second order moment/force?

 

[darkwing88]

jones: how one figures out how much is Mnt and Pnt and how much is Mlt is Plt for a given load case is beyond me. It is probably just me, but who out there would actually use this method?

 

[SteelPE]

That probably why everyone is making the switch to the DAM. I find it to be only a little more confusing.... which is much better than I have NFC.

 

[ToadJones]

Well hearing some responses here has helped me feel a little less inept.
I hope to make the switch to the DAM soon....just in time for them to change it again.

 

[darkwing88]

I have been doing this stuff for 27 years. I suppose it is normal to get "code change fatique." Some people lap it up, but some like me just want to get the job done and go hunting or fishing.

I am going to backtrack on what I said I was going to do about this structure and do it the old fashion way --> use '89 code. I will start using the '05 code when we update Staad.

Gone fishing