Need Help!- 'Jumpy' servo motion; NUM servo and drive.

[ralph@nes]

Hello all.

I have a KO Lee Scepter C1020N grinder that is the subject of today's post. We obtained this grinder at auction about a year ago, and it's been a fight to get up and running since.

The first issue we had was two malfunctioning drives. Both NUM AC Monodrives were showing odd faults. Schneider Electric in Greensboro, NC repaired both drives and I've got no drive faults.

The machine is currently up and producing parts now, I'm happy to say.

Unfortunately, it's exhibiting some odd symptoms on the Z axis. When moving in one direction, the movement seems pretty normal. The other direction, there's a 'pulsing' or jumpy movement. It doesn't seem to be catastrophic, as I'm grinding just fine with it, but I have a feeling it's a portent of things to come.

Removing the servo and moving the axis gear manually shows no increased resistance in that direction, so I'm fairly certain the issue isn't in the ball screw itself.

Anyone have any direction I can look on this? I know I'm specifying that it's NUM equipment, but is there a general malaise that can cause this?

Thanks very much in advance,

Ralph

[Al_The_Man]

If the drives are identical you could swap them to eliminate drive problem, but it almost sounds like a commutation misalignment on the motor.
These problems cause a lot of different symptoms, and they can be different depending on motor direction.
If possible, remove the motor and run it free.
Al.

[ralph@nes]

Commutation misalignment. Interesting you mention that.

Oh, and the drives ARE identical, however I'm hesitant to swap them due to the following:

There are four wires dealing with the power aspect of these motors, three black ones that are completely unlabeled, and a ground. I was careful to replace the black ones in the same way that they came apart when I repaired the drives, but I can't be 100% certain that they went back the same way. These have screw terminals and not real permanent 'plugs'.

Would it be functioning at all if some of these wires are in the wrong place? Or just a catastrophic lack of movment?

I have had the motor off and run before. Although at this point I don't remember if it ran any differently free than encumbered.

I am so glad you responded Al, after you helped to talk me through my Haas rotary table nightmare ( http://www.cnczone.com/forums/showthread.php?t=45025 ), I was hoping you'd jump in on this.

[Al_The_Man]

The three blask and ground sound like the three for the motor winding, which you will have whether DCBL or AC sinusoidal motor, mixing up these three can produce different results and different according to direction.
What I have done when faced with unidentified motor connections is to label the wires ABC and try each of the six possible combinations in turn.
However first, if the motor is labeled ABC etc use a ohm meter to identify each end of the motor conductors and ensure the conductors on the drive end each go to the AB&C phase.
Al.

[ralph@nes]

Is it possible that one of these out of phase (or whatever you'd call it) could be causing my issues?

[Al_The_Man]

All three phases have to be connected in the correct order for correct operation in both directions.
Al.

[ralph@nes]

Al,

We've dug farther into this.

We've completely disassembled the axis and verified that the mechanical aspect of things are correct.

We've swapped the drives and servos and this exhibits the same issue regardless.

Note, there is a glass scale for positioning.

Perhaps there is a gain mismatch or something that the glass scale is reporting is being interpreted incorrectly? These drives are a pain, there's only two potentiometers visible: Speed and Speed Offset. Anything else has to be adjusted via a cable and proprietary software (Which we have).

I believe the scales read back into the servo; I don't believe they are sent to the PLC and that the computer is doing any of the 'thinking' but I could be wrong.

All I know is that NUM is near impossible to get info from. At this point I'm wondering if a wholesale retrofit with a new control might not be a better idea.

[Al_The_Man]

It is possible to have a scale issue, often these can be cleaned by denatured alcohol and lint free cloth, the other is if any backlash exists between motor and scale, it can cause oscillations.
Do you know how the PLC is communicating with the drives? analogue signal or other?
Al.

[ralph@nes]

I would imagine this has to be an Analog system. Then again, I can't swear to it. My knowledge on the subject is limited.

I believe our maintenance guy cleaned the scale; We had the entire axis and ways off of the machine and apart. I'll clarify that.