What is being described is a classic low frequency resonance problem. In as much as you are using a table and substructure that are both suspended (and more importantly, un-damped / under damped) you are likely experiencing the effects of a high Q system. With un-damped systems, you can experience mechanical resonance with amplification effects (Q greater than 1). Reconsider your approach, use only one suspension component that is properly damped (Q = .5 to .7). Kevin Halverson