Does SAP 2000 Design for Torsion ?

[Jatfuentes]

thread801-384791
This question was posted before, but nobody answered. I think no, but would like to receive coments

 

[stressed]

As you probably know, torsion is always accounted for in analysis. For design, torsional strength and allowable torsional strength is automatically determined only for closed sections like pipes and tubes.. and that design check started in the newer steel design codes as I recall..I think starting with AISC 13th. Check the SAP design manual for whatever steel code you're working with to verify.

SAP design preferences show phi and omega torsion design preference options depending whether ASD or LRFD.

 

[Jatfuentes]

Thank you very much stressed,

 

[JoshPlumSE]

I can't claim to be an expert in SAP2000, but I don't believe what stressed said is entirely correct.

My impression was that SAP2k doesn't account for torsional warping in the analysis at all. That won't affect the analysis of closed sections, where warping effects are negligible. However, it could significantly affect the stiffness (and analysis results) of models with open sections like steel wide flange or channels members.

 

[JoshPlumSE]

Note:
My impression stems from cases where I've had to compare results from my company's program (a competitor to SAP) to SAP2k. And, when we turn off the effect of warping, the results would match. It's probably been a couple of years since I looked at this, though. So, things may have changed over there that I am not aware of.

 

[stressed]

Steel member reactions under torsion can be complex since torsion and warping usually occurs with shear and bending moments. To my knowledge, the only reliable approach for analysis of warping using beam/line elements is to have a warping degree of
freedom, 7 DOF instead of 6. I believe Abaqus is the only commercial structural software which offers a beam element with warping degree of freedom... please correct me if I am wrong about Abaqus being the only software to do this.

I wish SAP did account for more torsional design, but I guess they figure the non-closed section design procedures were too much work to add to their software for now, although I read that SAP will design for torsion for international steel codes. SAP does not account for warping in analysis of beam elements. Since presumably Risa doesn't offer a warping degree of freedom which is what really needs to be done, what approach does Risa utilize to account for warping? And is that approach to calculating warping
reliable for all or most torsional loading scenarios? I understand your wanting to tout Risa's workaround approach as a sales advantage for Risa over SAP, but it would be good to know when it's applicable and when it's not.

And you can't seriously claim that SAP doesn't calculate torsion as you've done just because warping effects are not considered in beam elements. Torsion is calculated...warping is considered with SAP shell elements, not beam/line elements. It would be no different than someone claiming Risa doesn't analyze frame reactions because it can't consider frame material nonlinear behavior, geometric nonlinear effects, or unsymmetrical moments of inertia for angle sections. Risa can analyze them, it just can't consider certain effects.. Pretty much same thing in claiming that SAP doesn't analyze torsion because warping effects are not
considered with beam elements..

SAP considers torsion... to my knowledge no other commercial structural software rigorously considers warping in beam elements
besides Abaqus.

 

[JoshPlumSE]

Stressed -

Where to start?...
1) It is rare for an analysis program that is geared towards structural engineers (which Abaqus isn't, in my opinion) to include torsional warping. And, I don't know of any that do a true 7 DOF solution like you describe. I wasn't trying to badmouth SAP, just trying to point out an important engineering considering in torsion that many engineers miss.
2) I don't want to hijack this thread to talk about RISA's torsional analysis. That would be more appropriate for a thread in the RISA forum. I'll just say that it is not a true 7 degree of freedom approach. And, it's not accurate for all torsional loading situations. I do feel, however, that RISA does a better job of flagging the issue to users so they know when they might have to investigate the issue in more detail.
3) All I claimed was that your previous comment was not entirely correct. It can give the false impression that the analysis in SAP truly account for torsion in for all steel members. You point out that the code checks handle it for closed sections. But, there is an important distinction that the analysis results do NOT consider warping stiffness which dramatically increases torsional stiffness in general.

To me it is important (IMHO) to raise the issue of warping whenever the topic of torsion in steel members is considered. To encourage engineers to investigate the issue more thoroughly for projects where warping effects may be significant. I'm not saying the SAP2K results aren't useful, just imperfect. I would encourage engineers to run supplemental hand calcs in cases where they are concerned about torsion in wide flange and channel sections.

 

[JasonMcKee71]

stressed, warping is something different and it is not considered in the analysis by commercial structural apps.

Jason McKee
proud R&D Manager of
Cross Section Analysis & Design
Software for the structural design of cross sections
Moment Curvature Analysis
Interaction Diagrams
Reifnorcement Design etc.

 

[stressed]

Josh, I stated that SAP calculates torsion. That statement still remains ENTIRELY correct. You used that statement as a cheap attempt to tout a questionable sales advantage of your Risa 3D software for handling warping, an "advantage" which may be wildly conservative or unconservative depending on the scenario. The only truly correct and rigorous way to handle warping with a beam element is to consider a warping DOF, which Risa cannot do. So if Risa can't handle material NL behavior, does that mean that it's not "entirely" correct to say that Risa 3D can't calculate beam reactions?

Let me state again - SAP calculates torsion. That is an undeniable fact which is entirely true. Entirely. The SAP beam element does not consider warping (same as Risa) although SAP shell elements do, but that doesn't mean that it's not entirely correct to claim that SAP calculates and reports torsion.

I agree that Abaqus is not suitable or economical for most structural modeling applications, but that's not the point, is it? The point is, if you're going to brag about your Risa approach over SAP, especially on a SAP forum, then you should openly acknowledge that Risa's "advantageous" approach is only a "workaround" approach at best, valid for some designs but not valid for other designs. You were not simply 'raising awareness' for the benefit of the engineering community, you were taking a partisan shot at SAP to try to benefit your company on a SAP forum. Anyone here can read what you wrote and what I wrote and look into the facts for themselves.

You posted recently on another SAP forum thread that Mastan would be a "better" option than SAP for educational users. So there seems to be a trend with your posts. I suspect CSI gives away their software to universities for free or gives it for nearly free, so how do you conclude that Mastan would be the better option?

You usually post thoughtful comments from what I’ve read, but you’re a competitor to CSI, and you've crossed a line here (IMHO) and I simply called you on it based on the specifics of what you wrote.

If JasonMcKee is too unaware to know that Abaqus beam elements handle a warping DOF then he shouldn't be posting about that subject on this thread. I fail to see any meaningful contribution from his comment. He too seems to be touting his commercial endevors based on the links in his signature

 

[Shz713]

I wouldn't stand up for any comments, but:

simply reading the manual or contacting the CSI support center woul clarify everything. FYI, all CSI products like SAP2K and CSiBridge that currently we bought for our students are not free at all and we have to renew license every single year!
Another point, if you google ASCE Bridge Engineering journal, you simply that such Q1 researchers used SAP2K for torsional behavior of large scale structures. Anyhow, I reckon simply googling would give you the answer

Cheers

Shoot for the Moon, even if U miss, U still land among Stars!