I believe sap2000 does not support ssi modeling. the soil and building have their own resonance frequency due to the fact that they are two different materials. the substructure and superstructure have to be analysis separately. the superstructure is designed first and the forces and over turning moments and base shear and uplift on the foundation and on the piles are obtained. sap2000 gives you the applied forces on the foundation and pile but not is resonance frequency. number one rule, always use pile down to bedrock for foundation unless it's a small building. then, these parameters on the foundation and piles are transferred to the soil model for further analysis.
if you have a rectangular plate, the angle of repose for soil determines the range of soil that you are working with. "There are numerous methods for measuring angle of repose and each produces slightly different results. Results are also sensitive to the exact methodology of the experimenter. As a result, data from different labs are not always comparable. One method is the triaxial shear test, another is the direct shear test.
If the coefficient of static friction is known of a material, then a good approximation of the angle of repose can be made with the following function. This function is somewhat accurate for piles where individual objects in the pile are minuscule and piled in random order.[2]
tan ( θ ) ≈ μ s {\displaystyle \tan {(\theta )}\approx \mu _{\mathrm {s} }\,} \tan{(\theta)} \approx \mu_\mathrm{s}\,
where, μs is the coefficient of static friction, and θ is the angle of repose." - Wikipedia on measurement of angle of repose of soil.
no plastic material is allowed under the foundation such as organic silt. they have to be dug out or friction or bearing piles have to be used.
the spring constant of elastic soil is k=F/x where F equals to applied load and x equals to the vertical displacement. it is very likely that you have differential settlement due to uneven distribution of soil. so you have to divide the foundation mat into small panels. if the forces on the panel from the superstructure and angle of repose are known, you can calculate it's settlement. after that, you divide the applied loads to the settlement and obtain the spring constants. each small panel has a different spring constant. since the natural angular frequency is equal to the square root of the spring constant over mass, you will have a range of natural frequencies to work with. you have to design the superstructure to avoid these range of natural frequencies. if the frequency of the superstructure coincides with the frequency of the soil, you have to redesign the superstructure.
the frequecnies of vibration are obtained through eign-vector analysis and the total mass of modal analysis is obtained by using a participation factor of 90%. that means you probably have to add up the first 10 or 12 modes of vibration to find the total mass involved in the superstructure. resonance occurs with the largest displacement. frequency of the largest displacement is the critical frequency. once the critical frequency is known and since the total mass participation is known, you can estimate the overall stiffness of the superstructure.
disclaimer: all calculations and comments must be checked by senior engineers before they are taken to be acceptable.
