Hello, all.

I have a 1986 QT10N ATC/MC with a CAM T3 controller that is exhibiting some very weird behavior.

I bought this machine used, and had some trouble getting it running at all. I previously posted this about the axis motor drive errors I was getting, and how I figured that one out. Once that was resolved, I was still having issues with the FX727 board's battery running down, and the machine losing parameters after a week or so. This was most likely a bad super capacitor on that board, but I solved the problem by replacing the FX727 with a different unit from eBay.

Now the machine is holding parameters, so I input some basic tool info and a simple program. This all worked as expected and I can Check and Simulate the part on the control and it looks correct. The machine knows where the tool tip is, and jogging it to the position that is supposed to be the Start Point makes the on-screen tool position match what the actual tool is doing. When I start the program in Auto mode, the machine (usually) spins up the spindle and goes about its business.

However, this is where it gets weird. The tool will usually do one or maybe 2 passes correctly (BAR OUT operation, doing multiple roughing OD passes), but then it starts jumping. The Z jumps in ~1" increments, sometimes in the + direction, sometimes in the - direction. Sometimes the X also jumps, but the programmed motions are quite small, so it seems like the jumps are harder to see. Each time I run this simple program, the machine does something different- sometimes it doesn't cut at all because it is all Z+ jumps.

Additionally, I get some jumpy/shaky movement when jogging any of the machine's 3 axes, though the Z is the worst by a fair bit. This seems to have gotten slightly better when I zeroed out the screw pitch compensation values (which were all over the place before due to losing parameters so many times), but it is still clearly way more than should be expected. The problem does not seem to be mechanical- Oilers are working and the gibs seem not to be sticking. Also, the fact that the C-axis is jumpy makes me think it is more of a control issue than a specific mechanical problem.

Has anyone out there seen anything like this before? I have talked to the Mazak tech I had come out previously, as well as Mazak's Northeast service office, and no one has so far been able to give me an idea of what might cause this. It is incredibly frustrating because I am SO CLOSE to having this machine working!!

Thank you in advance for any help anyone out there can give!

Here is a video of what it does:

https://youtu.be/6_0-zaqc5iQ

And here are some photos of the program, part shape, Check screen, and final part produced:

Attachment 365984

Attachment 365986

Attachment 365988

Attachment 365990