Kinetix 3 issues

First time i've been able to set up or use a Kinetix 3 servo drive. We don't have an engineer in our department anymore so it's on the maintenance guys to install/program now. I've got a Kinetix 3 drive with a TLY style motor/encoder. it's an incremental encoder with 40000 counts per full index. I'm able to use ultraware to program and make adjustments, but can't figure out why the homing sequence doesn't work correctly. If i fault out the drive, it will not return to the home prox. It seems like it will make one continuous revolution and thinks that wherever it faulted out, is the new home spot. Any ideas?

So, when you send the home command, the motor makes one revolution, and sets the homed status bit? (I haven't used the Kinetix 3, so my terms may differ.)

How is it configured to home? In most of these drives, the homing can be configured a number of ways, and one of them is to home to the encoder maker pulse which sounds sorta like what you describe. Most likely you want to home to an external limit switch. Could it be as simple as a limit switch that has failed (shorted, on all the time)?

You say the encoder has 40000 counts per full index, is this a linear motor?

Welcome to the forum, by the way.
Paul

I have the homing sequence stating that it homes to sensor, then back to marker. So it should home until it sees the prox, then backwards until the leading edge drops out, then forward again so it just makes it. At least that's what the description says in the manual. I know i'm getting a good signal from the home prox, as i have it tied to both my PLC (SLC 5/03) and the Kinetix drive. Both are registering the input when the prox is made.

It's a small Rotary servo motor. The rep where we got the motor from came in and gave me a crash course (very small) of how to set up the software. We figured out the math and with the table we have it on, the 5:1 gearbox, it has 40000 counts on the encoder.

Okay, so my next guess is going to be that there is electrical interference (noise) on the home switch cable. This can be too fast for you to see it in RSLogix, and even too fast for trap logic in the 5/03 to detect, but not too fast for the high speed inputs of the Kinetix drive to detect.

Can you try physically isolating the sensor wiring/cable from all motor leads? (for a temporary test, perhaps run one loose away from all controls) Also, it may be possible to add a suppressor to the input point. I would have to defer to the manuals and other experts to recommend a suppressor.

This is just a guess on my part. In most cases, noise issues on a home sensor are erratic, and from your description, it sounds like it is very repeatable.

Another w.a.g.: Is the home sensor configuration set up correctly in the software (n.o. versus n.c.)?