I think the best way for modeling of the lateral boundaries of a shaking table is that first of all you construct the simplest model (one layer soil) in shaking table then model it by FLAC. by comparing the results, you will be able to find out how the boundaries work. I think it can depend on Vs, your specific shaking table and it's instruments.

If you don't do that you should suppose that either vertical or horizontal boundaries are fixed in X and Y directions.

Thank you eng. Shareghi. That's a great idea. We also can slave theme together.

Thanks Alan,That's a big deal. I should try it.

Best,

Nastaran