Could be motor noise on the resolver signal feedback. Check the resolver cable shielding and ensure proper connection throughout. Many times have seen the shield is not properly tie wrapped at the serial measurement unit or the shield corroded away in wet/corrosive environments.
If mechanical, can the brake safely be released and the axis manually moved to feel for something abnormal?
sounds very typical when axis is not tuned
Is it a rotating or linear axis? Are the brakes releasing?
this is also known as inertia matching. motors normally have very low own inertia compared to inertia of the load which is one of the reasons for using suitable transfer ratio helps (gear ratio). and while perfect inertia matching between motor and load may not be practical or possible, it need to be reasonably close:
1:1 is perfect
1:5 ok
1:10 bad
axis tuning involves determining correct gains (PID parameters). when the gain values are incorrect, load mismatch is a result and observed axis behavior is just like what was described (assuming everything else is ok, ie. no brake issue or high friction etc.). the question is if this is a custom design or not because if it is wrong, it may not be possible to tune it correctly.
one option (normally only used as a test) is to move axis at very different speed (usually much lower) but this is rarely practical or useful for the intended application.
another option is to reduce inertia of the load but this too is normally not practical when deploying, it is something to consider at design stage of the axis.
using higher gear ratio does trade speed for torque but it also has a dramatic (positive) effect on load matching - from motor perspective load appears to have lower inertia. check with ABB guidelines what the minimum value should be used. for example KUKA recommends minimum 15:1 ratio.
I am having trouble with a 7th axis moving while in auto and in teach. I have tried changing the load mass to a few different loads and that seemed to have had very little impact. It looks like it tries to re-adjust while moving so it overcompensates and looks like it is shaking while it moves.
Shaking while it moves, or during the start/stop of motion? Generally, when an axis has not been tuned properly, the symptoms are strongest when the axis velocity is changing -- accelerating from a stop, or decelerating to a stop.
The last time I had something like this, the axis moved okay, but when starting or stopping, it would "wobble" back and forth -- it looked almost like the drive shaft was made of rubber. Once it got up to a stable speed, it moved slowly, but any increase or decrease in speed would cause the wobbling.
I "assumed" this was troubleshooting a working process. Leave it to the build guys to sniff this one out.
Thanks for posting the fix and sharing the manual.
