Acorn+ClearPath Servos - Axis Fault during Home (Resolved)

[rdsmith24]

This may or may not have been mentioned, but Centroid's ENA logic is not really setup to be used on stepper systems, especially closed loop with alarms. They work great on systems that provide encoder feedback to the controller, but as reported here dozens of times, do not enable the drives until motion is commanded. Great where the system has feedback and can determine there is no problems, but not so great in the step and direction arena where an alarm is the only feedback.

For this reason I have never used the ENA terminals. I do this because there are no advantages to use them, but multiple disadvantages when you do. I will either use a terminal connection or when there is a reason, have the ENA signal on a relay circuit that I can control.

This is the most interesting response I have seen....can you tell me how you are controlling enabling the servo's? Are you just doing it through the Estop relay or are there specific outputs you are assigning on the Acrorn?

Thanks

[martyscncgarage]

The enable state changes back to off when CNC12 sees a drive fault. Disabling the drives. They won't set again until the axis is moved.

The enable state is changed when you invert them in the advanced tab of the Wizard.

Why don't you back up and connect ONE drive/motor at a time. Test, then add the second one, test, then add the third one. Take the Axis Drive faults out as well.

Did you ever remove the axis drive fault inputs in the Wizard and get all three axis to move as expected? I can't recall....

[Gary Campbell]

This is the most interesting response I have seen....can you tell me how you are controlling enabling the servo's? Are you just doing it through the Estop relay or are there specific outputs you are assigning on the Acrorn?

1) Most often, just to terminals that are powered with the system.

2) Sometimes, Thru an auxiliary contact on the power supply contactor on the NoFault/Estop circuit

3) Rarely, if ever, Thru a separate screen button controlled relay.

The actual method depends on the machine, its components and application. In any case the servos are enabled when the system is active and disabled when not, and there are no alarm issues caused by the onboard ENA logic. Step and direction system alarms are much more rudimentary than most higher end servos use.

[rdsmith24]

All,
Thanks for all of your inputs...we have solved the issue and I will attempt to explain here....

1. Control panel was built using pre-3/5/2021 schematics provided by Centroid (using 24VDC and no pull-up resistors)
2. When checking the function of the system by attempting to move the axis using the VCP we never got the motors to move at all. We then INVERTED the EN outputs under the advanced menu. This looked like it was the issue because we immediately got the visual feedback from the Teknic servos of turning green.
3. We then proceeded to move the motors using the VCP, of course it worked now but we had another issue, the THIRD drive that we moved would cause and arbitrary drive fault (usually on axis3).
4. After trying a few other changes (STEP/DIRECTION Inversion) it still kept doing the axis failure after the third motor move.
5. Earlier this week I noticed that Centroid had new recommendations for the Teknic servos, use the C86ACCP board or re-wire using 5VDC voltage and 100 ohm pull-up resistors.
6. I re-wired the board opting to save my friends money and try getting what we had to work before resorting to buying more material.
7. After the re-wire and careful checkout, we realized that it was acting the same as before. At this time we were deeply troubled and concerned.
8. After a few more posts, most notably for me from Gary Campbell and Marty, that they EN are NOT enabled unless they are commanded to move this started to turn on a light.
9. Turns out that we wanted to see what we wanted on the Teknic, not what was true. We had WANTED to see the Teknic's turn on as soon as we pressed the reset button. BUT, what we failed to realize until today is that the drive failure that was present after the third move was actually the EN outputs being turned OFF (remember when we inverted the EN outputs in item 2). When we were moving the first axis what was actually happening is that the output transistor for that axis was being turned off. It was being masked by the other two outputs being turned on thus making the first axis continue to be moved via the VCP. When the third axis was commanded to move the other two transistors were already turned off thus causing the drives to the disabled, the Drive(x)Fault input coming in on the Ether16 and eventually the NOFaultOut being reset.
10. Moral of the story, be very careful and do only one thing at a time, carefully noting where you were and what it caused. In retrospect, it was me looking at the voltage at the EN outputs with a voltmeter and not relying on the LEDs to finally make me realize what was actually happening.

We tested our VFD, cooling fan outputs forward/reverse, fault inputs on the rest of the panel and it all seems to work, all we need now are the ball screws and adapters to convert the mill.

Thanks to everyone that chimed in on this project and I hope that someone can benefit from my mistakes here.

Cheers,
Roger Smith

[martyscncgarage]

Glad you figured it out.
Its why it is always recommended that builders start with ONE motor, get it working first, then add the second, then the third.
Sounds like this issue is resolved and you are on your way! Thank you very much for taking the time to update us on what you found!

Marty