P-Little Delta

[DaveAtkins]

I use RISA-3D for structural analysis. In the Help documents, it says you can account for P-little delta by dividing the columns into four segments vertically. Does this mean you need not bother with the B1 multiplier from AISC Appendix 8?

DaveAtkins

 

[haynewp]

Correct, the B1 multiplier gets captured on the analysis side.

 

[JoshPlumSE]

That's correct. If you're running a P-Delta analysis WITHOUT the intermediate nodes, then you're capturing the P-Delta effect at every node location. Since, that's probably only at floor levels, you'd only be getting the Drift or Large Delta effects.

When you add additional nodes along the length of the member, you'll capture the curvature of the member as well. That's the little delta effect. Why only three nodes between floors? Well, there were some test / validation models that were done and 3 node gave basically the same results as using 10 nodes or more. So, 3 is the minimum to get good results. Obviously, you could use more. But, it's not usually necessary.

Also, keep in mind that (at least for RISA), you're usually not actually dividing your physical member in to smaller pieces. Usually, you're just adding nodes. Internally, the "physical" member is broken up into smaller pieces. So, just adding nodes along the member length is enough.

 

[EZBuilding]

I looked at this option for a project utilizing reinforced concrete columns a few years back. Columns which were substantially over the ACI little p-delta ratio of 1.4 were not experiencing substantial second order effects when running the model with the column having several nodes along it's length. Something didn't feel like it was adding up and I have not incorporated this procedure into my practice in the time following.