M92 Homing Failing

[Greg82]

I was making some new Mcodes and while testing I notice my M200's that used to show a custom message would only display "Press Cycle Start to Continue". I shut the system down and started it back up and homing now fails on M92. The axis moves till it trips the switch and backs off. Right after the switch clears, the program fails ends.

I don't know if both of these issues are connected or not. I tried commenting out the Z and then Y axis to see if it was just a specific axis but it was the same on all 3.

I tried the copy of cncm folder that I had made before working on the custom Mcodes and the homing still fails.
I had updated from r7 to r10 plc earlier today also but everything was working after the new compile. I did try the r7 again after I started having problems and still no change.

[cncsnw]

Did you recently change CNC12 software versions too?

How much hysteresis do your home switches have? In other words, once the switch trips, how far does the axis have to travel back in the other direction, before the switch clears again?

Look into Parameter 180.

Also, take some time out from troubleshooting homing issues, to see if you can troubleshoot your ethernet communication between the Oak and the PC.

[Greg82]

Yes, recently updated to 5.1 from around 4.2 I think. I have used the machine since the update and everything was fine.

The limit switches have been the same since I installed the Oak almost 2 years ago and they worked just 20 min before I had this issue. I don't know the travel that they usually take to clear but it's only about a second from tripping to clearing. I think it used to travel farther after clearing the switch until it hit the index. Now it fails as soon as the switch clears. I will look into this farther.

I will also see if I can do something about the ethernet. I haven't changed or moved those wires recently either.

Shortly before (same day) I had this issue I installed the new r10 PLC and also edited it for the changes I wanted but I kept testing it as I went.

Would a fresh install of cnc12 be worth trying?

[cncsnw]

Your touch probe input may be the problem.

In theory, as long as you provide a probe-detect signal on INP771 (as specified in your Parameter 18 value), it should not matter that your probe (normally-closed, software-inverted) appears to be in a tripped state while you are trying to home.

In reality, I don't think that CNC12 correctly checks the probe-detect signal before concluding that a "tripped" probe input (INP769, per Parameter 11) should interrupt operation.

Try jumpering your probe input (e.g. from pin 3 to pin 5 of the circular bulkhead connector, if you have one; or from pin 3 to pin 8 of the Oak probe header if you have no bulkhead connector).

[Greg82]

Today when I turned the machine on it worked as normal. I had shut it down with the r7 PLC active yesterday. Reinstalling the r10 brought the problem back.

I only have a tool setter so I think you might be onto something. There isn't anything attached to 771. I will still test this with r10 to see if that solves the problem so maybe it can help someone in the future.

The main thing I was looking for with the r10 was to be able to set rapids to 50% without changing the normal feed rates but even when r10 was installed it kept modifying both together. I have not done any digging to see how its coded yet but plan to put it on a simple input. On is 50% rapids Off is 100%.

Thank you