Large Differences in Residuals after RRA and CMC

[Peter Staab]

Hi,

I´m working on an obese subject. I managed to get RRA residuals within the "good" range of the recommended boundaries (Peak_max_force: <8N; Peak_max_Moment: <45 Nm). But when I use the coordinate file and the model (both from RRA) in CMC, the Residual_Mz is drifting from -24 Nm at the beginning of the trial up to 80 Nm at the end. The other residual moments after CMC look quite similar to them of RRA.
In the last weeks, I tried everything I can imagine, including reducing tracking weights, reduce optimal forces for residual moments and in parallel increasing the controls for these actuators, etc. Controls for the muscles look ok and even the reserves. Changing mass properties in the model helps, but leads to bad residuals after running RRA again.
As I understood, the residuals are related to external forces acting on the model which could not be brought in line with the kinematics. And they are expressed between pelvis COM and ground. So, same model and same kinematics and grf should lead to similar residuals?!
Does anybody have an idea, why RRA and CMC residuals behave in that different manner?

Thank you for help in advance