My Z-axis stopped moving

Hi,

I am retrofitting a CNC lathe right now. In that process, I ran into a problem: my Z axis, which already worked, stopped doing so. I was messing with the encoder wiring, but I don’t think that is related. So X still works fine, but Z does not. I still get encoder feedback and see that when the machine is enabled, the Z axis drifts slowly.

I use a 7i97T and Chinese AASD motor drivers.

Troubleshooting I already did:

Driver: I confirmed that all the I/O still works. External E-stop, enable line from LinuxCNC, these are fine and work as expected. I then injected voltage into the analog line from the Mesa connector. The drive moves and stops after the following error gets too big, and the enable signal is removed by Mesa.

So I don’t think it is the driver or the wiring to it.

PWMgen: I then switched the pluggable connectors for X and Z on the Mesa and changed the HAL accordingly. Still, the X axis works fine and Z does not. In HAL Show I watched .pwmgen.nn.value, which started to grow after the machine was enabled and the Z axis started drifting. After a few seconds, it reached 60, which is the value of output-scale. With that, I would expect 10 V on that line, but I measure barely a few mV.

That really seemed like a hardware defect to me, but as I said, I switched the outputs between X and Z, and the PWMgens are confirmed to both be OK.

So in my head, only the software (e.g. HAL and INI) is left, but I did not really change anything there, at least I cannot think of anything.

With that, I am at the end of what I can think of. Please, if anyone has an idea, I am going insane trying to find this. I swear, if it is the equivalent of a missing semicolon in C…

I have the current .hal and .ini in the attachments.

At the end, I want to describe what I was doing immediately before this problem appeared. I cannot think of a reason how this would be connected, but still:

I had both axes moving and mostly tuned. I was testing the repeatability of the hall switches I use for homing. I found these to not be reliable and wanted to use the index pulses from the motor encoders. But I have linear scales on the machine, so I wired only the index pulses from the motor drivers to the Mesa inputs, while the A and B lines from the linear scales go directly to the bed and cross slide.

I changed only HOME_USE_INDEX = NO to HOME_USE_INDEX = YES and started the machine, at least that is what I remember. Since then, the INI and HAL have changed slightly as I tried to solve this. The HOME_SEQUENCE has changed as I wanted to check if the index thing worked for the X axis at least. It does.

So yes, I would be thankful for any help you can provide.

Thank you, it was the analog line of the Z axis, it was switched around. Switching it and inverting the output scale from 60 to -60 did the trick.

Now I wonder how it worked in the first place. I think the GND of that connection must have been bad, and when I added the index lines, it got connected.