Slender concrete column design in RISA 3D 1

[IngeIvan]

Hello,

I am designing a few slender columns in RISA 3D and I'm trying to figure out what the program is doing.

I know RISA does a second order elastic analysis (doing the P-DELTA analysis) and I can get the amplified moment directly from the results (it is a function of the cracked stiffness of the members, etc). It is my understanding that this is the final amplified moment that I should use to design my columns using the interaction diagram, meaning that if the solution did converge, the columns are stiff enough and they are not going to buckle (the Euler critical load is less that my ultimate load). I know it is recommended to multiply these moments by 1.15 to account for some other stuff if the analysis is linear elastic.

Anyway, my question is, why does RISA still give you the option to check if the frame is sway or non-sway, and according to the general reference, it will use approximate K values for the columns? I thought all of these parameters (K, Lu, Cm, Sway) are for the moment amplification method that would be used if calcs were done manually. Will Risa do moment amplification too?

Does this mean that I still need to calculate my K's with the nomographs and all that for my analysis and check the sway option?

What am I missing?

Thanks!

 

[IDS]

Doing a buckling analysis on reinforced concrete is not straightforward because reinforced concrete is highly non-linear after the cracking moment, and even more so when the steel reaches yield. I am not sure where the 1.15 factor comes from, but considering the non-linearity in the analysis it seems very small to me. I would suggest checking the results against the traditional provisions in the applicable code.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[JoshPlumSE]

Which version of the ACI code are you using? Because the most straight forward way to handle slender columns may be different depending on the design code. Also, which version of RISA-3D are you using?

 

[KootK]

Does ACI give designer's the option of magnifying their moments via second order analysis rather than the equivalent length method? Essentially, this is the concrete equivalent of the direct analysis method in steel. One would expect analogous adjustments to stiffness etc.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[IngeIvan]

Josh,

I am using ACI-318-11, and Risa 3D 12... Are you asking because of the provision change in 08?

 

[IDS]

Essentially, this is the concrete equivalent of the direct analysis method in steel. One would expect analogous adjustments to stiffness etc.

Except that steel is linear up to yield, but reinforced concrete is highly non-linear within the working load range.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[KootK]

Steel's plenty non-linear: residual stresses, initial imperfections, p-Big/p-little, panel zone... Cross section cracking is just one drop in the non-linearity bucket.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[IDS]

It's not exactly a drop in the bucket. In spite of the non-linearities you list, the overall behaviour of a steel structure is reasonably close to linear at working loads, whereas a concrete structure is nowhere near linear.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[JoshPlumSE]

Yes, because of the provision change in 2008.

If you are using the latest version of RISA-3D and are using the 2008 or newer ACI code, then you've got 3 options for accounting for slenderness.
10.10.3 (non-linear 2nd order analysis)
10.10.4 (Elastic 2nd order analysis)
10.0.4 (the "hand calc" moment magnification procedure specified in older versions of ACI)

RISA assumes 10.10.4, which is for the most part covered by running a P-Delta analysis on your model. The only caveat is the part about P-little delta (or member curvature). You might check out the P-little delta section of the RISA help file. In there we suggest adding a few joints along the height of your slender column.

You see RISA's P-Delta analysis currently only accounts for joint deflections, not joint rotations. So, the solution is to throw enough joints along the length of the column that these joints adequately pick up the member curvature between story levels. We have a couple of examples (the "benchmark" tests from the AISC commmentary) which demonstrate that 3 joints between floor levels is sufficient for most practical cases.

Oh, and make sure that you have a reasonable setting for the Icr value for the column. By default RISA uses 0.75*Igross for strength level load combinations. If you suspect this member to crack a lot then this may not be sufficient. And, that moment of inertia used in the analysis will greatly affect the moment magnification that occurs in the P-Delta anlaysis.

 

[JoshPlumSE]

Oh, one last thing. The sway flag doesn't really apply so much for the these new ACI codes. Because, of course, we're counting on the 2nd order analysis to give us the moment amplification.

Though, it is possible that this entry is used for something else still. If not, I will put in a request to hide / suppress these input fields when using the new ACI codes. The entries would still exist behind the scenes. (That way the program remembers the settings if you switch back and forth between old and new codes.) But, we'd want to hide them to avoid confusion. I'll look into that.... when I get a chance.