Rigid Baseplate? 2

[scanf1027]

I am designing column baseplates using Risa Base (FE element software). I recently attended a webinar from Risa on this software. The presenter suggested not checking the "force baseplate to be rigid" box saying it may not be conservative. The anchor tension values you get from assuming rigid versus not assuming rigid are huge. Hilti's software uses the rigid assumption, and the results for anchor tension agree pretty well w/Risa Base rigid analysis.
Any one have any suggestions on this? I know that AISC's design guide for baseplates assumes rigid plates, but could this be unconservative? Trying to get a handle on the anchor tension value is very difficult, when I have a difference of 12 kips/anchor.

 

[JoshPlumSE]

The "traditional" calcualtions that AISC and blodgett and such are based on a rigid plate assumption. We know that is not particularly accurate. Is it good enough for design purposes. That's a matter of engineering judgement, obviously. You could certainly argue the fact based on the relative lack of publicized base plate failures. You could also argue the fact based on what you consider to be the "standard of care" in the eningeering community.

Now, the FEM analysis of RISABase should give anchor forces that are a good bit better. It will account for prying action, side bending, corner bending and biaxial effects in a more sophisticated way than hand calcs would normally do. I have accounted for these effects using hand calculations in the past (when I was developing another base plate program for my former company).... Adding those features greatly, greatly complicated the calculations. That was fine since I was working on developing a fairly comprehensive program. But, it would be way too much work for a typical project calculation.

That being said, we also know that the FEM / RISABase analysis will be based on an elastic analysis. So, the inelastic behavior you get in the post yield range (between My and Mp) isn't really accounted for in either the traditional or FEM analyses.

Lastly, I should point out that we've been pretty successful in replicating the "traditional" calculations using RISABase. We just have to make two changes 1) Swith to the "rigid plate" assumption. 2) Modify the anchor rods so that they are also rigig. I do this by changing the Young's Modulus of the anchor rod to 100,000 ksi.

 

[sdz]

Is you baseplate thick enough to act rigidly?

 

[scanf1027]

JoshPlum, thanks for you answer. It's helpful, but I'm still not sure if I should take the lower anchor tension load as being accurate from the rigid analysis. I know there have been little failed baseplates, at least reported ones. But I am using adhesive anchors, so I'm a bit more concerned.


sdz; that was basically my question. Should I treat the plate as rigid or flexible.

 

[steve1]

If you select the rigid baseplate option in RISA base and your base plate is not thick enough to be rigid then your solution should not converge. That should tip you off that you need to increase the base plate thickness.

 

[JoshPlumSE]

Ultimately, it is an engineering judgment call. The theory and analysis behind RISABase is solid. The program has been widely used. So, IMHO, you would certainly be justified in using its numbers. The real question is whether you are comfortable enough with the procedure that (as the EOR) you are willing to take responsibility for it.

 

[slickdeals]

Watch this presentation

http://www.aisc.org/content.aspx?id=19858

 

[goncho]

"Traditional" analysis is not completely rigid, because you have an inflection point at neutral axis. If you add strain compatibility, then the analysis is rigid. Check RAM Connection software and this link as alternative to FEM.

http://communities.bentley.com/prod...l-analysis-of-general-shaped-base-plates.aspx