Currently it appears (to me at least) that the force applied by the springs is (largely or entirely) irrespective of the acceleration they are undergoing. This favours balisters far more than catapults as balisters tend to have very little moving mass (when compared to catapults) and thus very high accelerations can be generated. I suggest that the force applied by a spring should be modelled according to a law similar to:

F_applied = F_rest - Mass_spring*Acceleration_spring  


This would have very little impact on catapults as the springs do not experience large accelerations, instead catapults rely on long arms to generate high velocities, yet the accelerations that balisters can achieve would be greatly reduced. However a small buff to the rest force may still be needed. 

Of course I could be entirely wrong in my observation and thus would withdraw my suggestion.