2b-t
commented
4 weeks ago That would make sense.
Some of the torque constants seem off (in particular between the 'old' models and their new counterparts, see here). I therefore updated these constants a week ago in a399c29c12f68670124a5953de07d83ab3264337 calculating the torque constants from the rated torque and the rated current (see Wikipedia and my code here). In case of the X8-Pro-H V3 you can see here that this results in 2.13 Nm/A instead of the value given by the manufacturer which is 0.29 Nm/A as you mentioned. Potentially you are still on an older commit that still uses the manufacturer's values.
The inertias are even worse: The rotor inertias are sometimes off by a factor of 10 or so. For example the X8-25 should be identical to the X8 Pro V2. The inertia is though largely different with 30.6 kg cm² vs 3.4 kg cm² while the rest of the specifications are pretty much identical. I noted that the actuators' origins in the CAD drawings are far away from the geometric center. My assumption is that they wrongly calculated the inertias to these offset coordinates rather than its center of mass (parallel axis theorem).
ioio2995
Hello,
I noticed a small inconsistency regarding the constants of the X8-Pro-H V3 motor, which seems very low to me. In the X8ProHV3 class, the torque constant of 0.29 Nm/A probably corresponds to the motor torque without a reducer.
The technical specifications also specify "motor" for this value,
while for the X8ProV2, the torque of 2.6 Nm/A is expressed without "motor".
This change in reference also seems to apply to the entire V3 range.
What do you think?
Best regards,