Help with ACI 351.3

[atrizzy]

I'm trying to use Staad.pro to calculate the frequency of a foundation I'm designing, and am using ACI 351.3 as a guide for calculating equivalent springs at the base of the footing.

Calculating the impedance at the base for the various conditions is fairly straightforward, but I have a few questions for those with more experience here:

1. Do we need a separate model for all DOF considered (rocking, vertical, torsion, etc)? Seems like, for instance, the rocking and vertical impedance may yield contradictory values for the base support spring constants. But of course the purpose of the modal calculation is to assess all of the system's potential dynamic modes... Help!
2. Presumably the base springs should be compression only, correct?
3. Is it common to neglect the effect of burial (side contact between the foundation and backfill/embedment material) for this sort of model? Presumably it's conservative to do so...

I'm still working through how to correctly input the actual dynamic loading in staad, but for the time being I'd be satisfied with a modal frequency calculation.

Any other advice would be great, and thanks in advance!

 

[canwesteng]

1. You need additional rotational springs or need to adjust the base springs, to address the rotational stiffness deficit. I sometimes rigidly link the slab to a central spring with all DOF restrained, but if the slab can't be considered a rigid body, this isn't an adequate solution on its own.
2. Yes, though STAAD won't do that. You should make sure at no point is the foundation seeing uplift, so your model works and also as general good practice for dynamics. If the foundation lifts off, you also have problems with the spring constants you calculated being valid.
3. This is usually done, because it is so complex to capture correctly. Put an isolation joint on the side of slab to better align assumptions will real world behaviour (and also reduce chance of shaking adjacent foundations.

 

[canwesteng]

Ref 3 had something on modeling the stiffness deficit but I never quite got it to work how I liked on my last project

 

[atrizzy]

Thanks Canwest.

Regarding your response 1. When you add rotational springs, would you calculate the existing rotational stiffness provided by the vertical springs and provide the deficit with your single spring? Is that what the Harden paper focuses on?

 

[canwesteng]

The harden paper puts stiffer springs on the "flanges" of the footing. There is a method in the paper to calculate the deficit as well so you could just add the rotational spring in the, i like to sanity check by applying a static twist at the center and seeing the rotation of the node to make sure it is working properly. I'm usually actually running a few models, one with rigid links to a central support (high stiffness), one neglecting the extra rotational stiffness (low stiffness)

 

[atrizzy]

Makes sense... I suppose this type of analysis is a prime candidate for a sensitivity analysis anyways.
Thanks for your help.