Although the problem is still there I think finally I am getting some more meaningful data.
I removed the belt and it turns out when I use the keypad in the LOCAL mode the vector drive does not see the magnetic spindle rotation sensor so there is no longer 3 seconds limit.
I am able to run the motor without the spindle and have all the time I need to take some voltage measurements without the spindle interference.
The rumbling is still there, and it looks like it is of purely electrical nature.
I attempted to rewire output from the vector drive to the motor to see if the difference in voltage follows the output of the vector drive or the windings on the motor.
The connections are shown in attached file CrossConnections.jpg .
It looks like the lower voltage follows the vector drive regardless which winding is energized so the conclusion would be that the motor probably is OK.
For some reason I could not run the motor on one of combinations that's why I said “probably”.
The next test I am planning to run is to find out if the bang that occurs every 180 deg of the rotation of the motor shaft will change its location in relation to the shaft while the power from vector drive is connected to different windings.
I also managed to access a Fault Log of the vector drive.
All 32 entries are attached in the file Fault_Log.pdf.
There are 3 different faults:
PWR Base FLT,
Over Current FLT,
and Current Sens FLT,
The last one suggests a problem with a current sensor and/or it's connection to the control board.
I just wish I new where that current sensor lives, is it in the vector drive enclosure?
The first and second fault points, among other things, to the encoder to motor coupling.
But why the symptoms are so predictable and regular (cyclic)?
One would think that the coupling or electrical noise would create rather random rumbling.
Any thoughts?