simtk.org

Hello all,

I have problem with my joint reaction analysis' result. To validate the result i plot the torque in all direction (x, y, z axis) and see if torque in z axis (Mz) is 0 during the elbow flexion. But the curve of Mz is not completely 0 during the movement. It starts with a high torque and goes into near 0 and never reaches 0 point, it goes only near zero. You can see the plot below. I have no idea, where the mistake is? Can someone explain it to me?
this is the model i used for the simulation. For the setting of the CMC (slow CMC) i used tracking weight of 6 for the elbow flexion and a torque actuator with optimal force of 25 Nm as reserve.
The result of the CMC itself is good enough. The RMS for kinematic tracking error is only 1.7 degree with the max error of 2.8 degree and the RMS of reserve of torque actuator is only 2.8 Nm with the max of 17.2 Nm.

Kind regards,
Sugiarto

Did you perform Joint Reaction Analysis using the Analyze Tool after simulation? Or did you attach the Joint Reaction Analysis to your CMC Tool?

It would also be good to check that the point of application of the force is at the pin joint's location.

ongcf wrote: ↑

Thu Jun 04, 2020 7:49 pm

Did you perform Joint Reaction Analysis using the Analyze Tool after simulation? Or did you attach the Joint Reaction Analysis to your CMC Tool?

It would also be good to check that the point of application of the force is at the pin joint's location.

thanks for answering my question.
I performed Joint Reaction Analysis using the Analyze Tool after simulation. Does it make a difference if the Joint Reaction Analysis is attached to the CMC Tool?

For your advice to check the point of application of the force, how to do that checking?

The output of Joint Reaction Analysis should have three sets of 3 columns for force, torque, and point. You can use the point columns to determine where the force is being applied.

Since you are using the Analyze Tool, it's possible the difference is in the forces that the tool is aware of. You should be adding any of the same external forces that you used in running CMC. Also, CMC typically uses some "reserve" actuators, which may also account for this difference if you didn't apply the same reserve actuator forces/torques to the analysis tool too. One quick way to check if that's the only differences is to compare the CMC reserve actuator torque output for the elbow against the joint reaction analysis torque that you see.