It depends on the situation but most of the times, a restitution factor of zero is recommended when working with CMS bodies. A restitution factor > 0 supposes instantaneous energy/velocity redistribution. While when flexible CMS body is present in a model, it experiences phases (there is a compression phase followed by an expansion phase.) which are not instantaneous due to the flexibility. Hence it is a bit contradictory to mix CMS with restitution factor > 0.