Current draw for motor is different based on direction of turn

I am running a 3 phase Servo Motor (J0400-304-4-000 60mm Servo Motor, B2B Applied Motion)

When I run the motor in one direction it changes the amount of current it draws vs if it goes the other direction. For example when it is holding 0 pounds, it draws 0.5 amps going one way and -0.5 going the other way, which is expected behavior.

However when I add 10 lbs it draws 2.4 A going up and 0.8 A going down, when I add 20 lbs it draws 1.5 A going up and 0.15A going down. The extra weird thing too is if im pulling the weight up and stop it, it might read something like 2.5A, but if i keep the motor stationary there and pick the weight up and set it down it will read a lower value like 1A. So something about unloading it and reloading it is causing that behavior

similar to that unloading and reloading problme, if I have 20 lbs on there and it is drawing 1A, and then take 10lbs off, it will still say its drawing 1A until i unload it and reload and it will drop to 0.6A

and those values make no sense and aren’t even consistent, so it seems like something is going on potentially electrically in the current reading sensor or maybe something with the ODrive is weird? I have tried the motor on both axis and the results are the same. The general trend though is the more weight I add the more distance there is in the “weight going up” vs “weight going down” current measured values.

I have tried running the motor in different modes like position control and velocity control and tried moving the weights sinusoidally, and the only thing that changes the current draw is when the motor is going on direction or the other.

I also changed the motor around so that the direction the motor would turn to pull the weight up changed, and the direction the motor turned to pull the weight up was always the higher current draw than the direction the motor turned to let the weight down.

I’m very lost on what could be going on so any thoughts would be much appreciated!

Sounds like normal coulomb friction to me.