EZTRAK Y axis loosing position, "slipping"

[melamark]

Hello Fella's,

My Eztrak has recently developed an annoying and costly intermittent problem. The Y axis is loosing actual position during travel, despite the display indicating that it is correct. This seems to be happening only during rapid movements of more than several inches and only on the Y axis. The amount of deviation varies from ~.200" up to 1". There is no indication that anything is wrong, other than measuring after the fact or noticing an obvious feature in the wrong location. Going back to 0,0 after this happens reveals the extent of the error.
Going back and measuring features, I can find when this happens. I can find no correlation other than a Y axis rapid. It is a singular event, that is, all the features after this slip are in the correct location with respect to each other - except for the amount of Y axis slip.
I checked all the mechanical features of the Y axis drivetrain, and everything seems to be in tip-top shape. This seems to be an electrical problem - being intermittent, I fear this will be a tough one to find. Anyone have any experience with a problem like this?
The machine is a 1995 SX upgraded with a Pentium 200 (socket7) motherboard and hard drive (have owned for over a year without any problems whatsoever).

Thanks,
Mark

[BigDaddyG]

Mark,
Which Software are you driving it with?

[melamark]

Sorry,

I am using version 3.08, PN 31943014

[yuso]

Mark,
Try this:
Interchange the Servo Amplifier from X axes to Y axes. See if you still have the same problem.
If you still have the problem, now interchange the Servo Motor from Y to X .
This quick and simple thing could show to you if: the Amplifier card could be the problem , or the Servo Motor encoder could be the problem.

Good luck
Yuso

[NC Cams]

Mark:

We had EXACTLY the same problem.

Solution: pull the servo motor(s) and have them checked out by the FACTORY service center. If they are SEM's, Clarkson near Chicago is the place.

Ours were found to have dirty position encoders and faulty tach circuitry on the X and Y servos. Bogus tach feedback screws up servo tune/response. Motor rebuild is $250ea and tacho replacement was an additional $200ea.

Hint: If your try to mill a circular pocket, you'll also probably find that it will be eliptical or egg shaped instead of round - further proof that the servos and amps are not talking to each other (and the CNC) propely.

[machintek]

The key question is will it go back to zero. The second key question is if it is coming out short of the destination point or past it.
I have seen the taper locks on the axis get loose and slip, especially in rapid.
I have seen encoders get loose from the motor shaft. Both would cause a failure to return to start point. And probably a failure to get to the destination.
A bad encoder or loose encoder can create a situation where the machine may overshoot because it does not see enough counting pulses. A bad encoder can still return to the start point.
Just things to check.

George

[machintek]

By the way drive tuning is most noticed as an elliptical circle.

George

[machintek]

One last thing. And I admit that I have seen this mostly on V2XTs. Check the belt tension and the condition of the Y axis belt. I have seen the pulleys get rusty and the belt teeth get chewed off.

George