RISA Foundation Questions

[Canuck65]

I don't use RISA Foundation very much, I have a copy of an older version, 4.0. I am modelling a raft slab foundation for a small (120 ft x 60 ft) building on very poor soils. Piles are not an option. Soil is very sensitive to differential loading. The building is a single storey with all loads being supported around the perimeter on ICF walls, therefore heavily loaded around the perimeter and only live load and the slab self weight on the interior areas. The pre-engineered wood roof trusses span the 60 ft. direction.

The main problem I am having is that the geotech wants as even a load beneath the raft slab, however, that does not seem feasible. (the slab is 60 ft between walls). Some questions I have are:
1. Does the program vary the stiffness of the slab based on the rebar layout and slab thickness?
2. Where does the program allow you to input your own rebar layout?
3. When modifying the subgrade modulus (me doing some testing) I find that the better (larger) the modulus input, the higher bearing pressure becomes under the perimeter walls, the less it is distributing to the interior portions of the slab. I find that as the subgrade modulus increases, the stiffness of the soil attracts the load as the slab is not stiff enough to do much of anything over 60 ft span. This seems logical to me, however, the geotech's opinion is that it should be the opposite and I should be finding a more even distribution of bearing pressure.
4. I tested a 4 ft thick concrete slab, which would be way beyond what I would consider reasonable and still had a significant difference in bearing pressure between the outer edges and the interior portion of the slab. The geotech is concerned about differential settlement.
5. Even with a more reasonable 12" to 16" slab, the theoretical deflections seem quite reasonable.

Any suggestions or comments would be appreciated.

 

[JoshPlumSE]

1. Does the program vary the stiffness of the slab based on the rebar layout and slab thickness?

Not for rebar layout. The analysis is done with plate elements that represent the thickness of the slab.

2. Where does the program allow you to input your own rebar layout?

I think you would just set up a design rule for a very specific size and spacing of your bars and apply that to your design strip.

3. When modifying the subgrade modulus (me doing some testing) I find that the better (larger) the modulus input, the higher bearing pressure becomes under the perimeter walls, the less it is distributing to the interior portions of the slab. I find that as the subgrade modulus increases, the stiffness of the soil attracts the load as the slab is not stiff enough to do much of anything over 60 ft span. This seems logical to me, however, the geotech's opinion is that it should be the opposite and I should be finding a more even distribution of bearing pressure.

There are some modeling techniques recommended in the ACI report 336.2R-88, “Suggested Analysis and Design Procedures for Combined Footing and Mats”. Essentially, you artificially modify the subgrade modulus at the edge and corners of the slab in order to achieve the desired effect. Essentially, the issue is that programs like RISAFoundation use uncoupled springs for their soil springs. In particular take a look at figure 6.10 of that ACI report "Method of computing coupled k"

4. I tested a 4 ft thick concrete slab, which would be way beyond what I would consider reasonable and still had a significant difference in bearing pressure between the outer edges and the interior portion of the slab. The geotech is concerned about differential settlement.

Certainly the more "Rigid" the foundation is, the more uniform the soil bearing will be under the slab.

 

[Canuck65]

Thank you for your response JoshPlumSE, it clarifies a few things for me. I will look into ACI 336.2R for background information. Based on the information you provided it seems to me that perhaps I should be looking at alternate analysis programs or other methods that will adequately account for the stiffness of a reinforced slab.

 

[JoshPlumSE]

Canuck65 -

I don't know.... My guess is most of the other structural programs do similar things for their foundation springs. I suppose there could be some more "geotech" based programs out there that handle this better. Or, that there could be some programs that automatically do the "spring adjustments" per region based on your soil type as an approximate method for handling the coupled spring effect.

If you really know what you're doing, you can probably make a model work with a general analysis program (like RISA or SAP2000 or such) where you are manually creating the coupled springs. But, I'm not sure how much better or more accurate this analysis is.


Caveat: I used to be a vice president at RISA, but I left when Nemetscheck took over (and still hold some animosity over the way the departure occurred). Now I work for CSI / SAP. So, just realize that I'm not exactly an impartial critic of these products.