Kinetix Motion Project

I would agree with you, if that is the case the ship is already at the bottom of the sea.

Let's assume I do not need both motors to power the part after it starts spinning. I am looking to do exactly what you are talking about.

"You will need one motor to produce a retarding torque in order to guarantee the backlash stays out of the system."

What is the best way to accomplish this? I see there have been several answers as to how to do this. Is there any particular one that we could agree upon to try implementing first?

Thanks,
Jarryed

Ok I see some fun here

Problem

1. you have two motors connected via some sort of coupling
2. Both motors are possibly required to get up to speed
3. while running at a fixed speed you want to remove the slop in the coupling
4. Both motors are possibly required to stop

One method using only position control

1. Motor 1 is the Master - Motor 2 is the Slave
2. Both motors are identical
3. Cycle Start - Initialize - Determine the MAG Lock position and incremental distance to move to remove backlash
   1. MSO the Master so that it is locked and monitor the commanded torque
   2. MSO the Slave
   3. MAJ the Slave (slow speed) forward Monitoring the Commanded Torque
   4. When Slave Torque > NoSlopTorque MAS and Capture the current Position (FWDLock)
   5. MAJ the Slave (slow speed) reverse Monitoring the Commanded Torque
   6. When Slave Torque < -NoSlopTorque MAS and Capture the current Position (REVLock)
   7. MAM to the middle of the lock positions
      MAGPosn = (FWDLock+REVLock)/2 absolute
      FWDNoBacklash = FWDLock - MAGPosn (incremental > 0 use forward in MAM)
      REVNoBacklash = MAGPosn - REVLock (incremental > 0 use reverse in MAM)
   8. When in Position and Stopped perform the MAG "Commanded Value"
4. You now have the motors ready to run
5. MAJ the Master to test speed - The Slave will follow
6. When at test Speed MAM the SLAVE forward or reverse to remove backlash
7. Monitor the Torque when you do this - I expect the total torque to stay the same and the torque difference to be as determined above
8. Perform your test
9. MAM to reverse the backlash
10. MAJ or MAS to Stop

I have NOT used this method on directly mechanically coupled motors
I have used similar methods on coupled motors with product held in between then

Notes: Commissioning I suggest trend of
master commanded torque
slave commanded torque
Master position Error
Slave Position Error
Master Velocity
Slave Velocity

I know this is a bit late....

Some additional info on large antenna servo systems:

Counter torque between the drives is higher at lower speeds
The speed offset of the slave drive (lagging motor) is calculated based on the counter torque desired between the master and slave.

At higher speeds, the counter torque scaling factor is decreased (or capped).
When reversing direction the role of lead motor is handed off to the opposing motor (role reversal). This can be handled by a sign value from the torque bias signal (I think).

All this is from memory from about 7 years ago, but it worked on old (non- Kinetix drives like 30 years ago). Hope it is helpful to add this to the thread.

I need to try this on some Kinetix drives...sounds fun and this is a great thread. Thanks all.