Anyone know why my z axis would change its work zero in the middle

Anyone know why my z axis would change its work zero in the middle of a job? I can only think of 2 potential reasons off hand, and they are both mechanical:

  1. the router bit got pushed up further into the collet during the job, or

  2. the screws on my motor mount moved the pulley causing the tension in the pulley to weaken and slip. (I’m trying to correct this by redoing my z-axis plate. I have a quick fix to check it out tomorrow to see if it helps (the temporary plate is HDPE) but I’ll cut another aluminum one in the near future, if that corrects the problem.

I’ve never really had too many problems with the z axis up until now, but my previous cuts were mostly 2.5 D, and I didn’t pay as close attention to the z axis depth. If it happened before, it was such a small change that I didn’t notice it.

If it’s program, electrical or controller related, it’s going to be much more difficult to diagnose. I don’t think it is though. Have any of you ever had an issue like that?

Otherwise, my machine has been working better/smoother than it ever has. I was relieved to discover that (other than the z axis issue) the accuracy and repeatability of the machine is pretty spot on. I’ve been trying to cut some of the same 3D reliefs that I made on my 3020 machine on my larger machine in greater scale and dimensions. That’s where I discovered the issue.

Stepper motors skipping a step due to load or a fast pulse train can cause issues. Seen this on a 3d printer

@Jan_Eringa hopefully the electronics in CNC machines are a bit more robust than you typically find in consumer 3D printers. When building your own CNC machine you do have to set it up to run well within its capabilities too. My machine will rapid over 200 IPM but I keep it under 90 IPM. Just to be sure it will never miss a step. This is called, “derating”.

@Paul_Frederick I would still take a peek at the signals going to the stepper with a scope to check the waveforms are to spec and nice and clean :grinning:

Oddly enough, I just had the same thing happen on my small machine! All of the sudden, the machine forgets where it is. Frustrating!!missing/deleted image from Google+

missing/deleted image from Google+

Thursday, it was my big machine:missing/deleted image from Google+

@Jan_Eringa a scope may or may not reveal signal issues. It really would depend on the nature of the problem. If it is a case of just interference the scope should be able to nail that. But if it is something else? Then one pulse stream is going to look much like the next on a scope. I bought an AM radio for probing EMI issues. It can be an enlightening instrument. With it I can tell if something is absolute crap and needs some suppression work. The radio cost me 25 cents at a yard sale. It’s a Sony too! I probably look like Jimmy Page when I am using an AM radio to probe electromagnetic noise. I am not one to worry about appearances. https://www.youtube.com/watch?v=1JtLHa3q37w

I noticed yesterday that the z axis drive’s top LED green light went off after trying to run a job. I’m getting another DM542T Driver for the Z axis, since adjusting the DIP switches on the X axis Friday, the X axis motor can drive the spindle all the way to the end of my table. Before I had the DIP switches set to a higher current, the spindle would only reach to within 8 inches of each side and then began to grind. That makes me a little more confident in that driver. It’s considerably more expensive, but not as much as a gecko driver.

@George_Allen Maybe that’s it? It sounds to me like you’re just guessing though. So here’s to hoping you’ve guessed right. Expensive does not always mean better. The motion control software I use is free and it is arguably better than the commercial offering that costs money. It doesn’t need any kind of a translation board in order to work for starters.