Z axis goes up but not down intermittently

[wfws]

I'm having a real problem with my Z axis. It will travel up without issue, but intermittently will not move down. Here's what I've looked at so far:
1) connections - my molex connections appear solid but I have not checked inside the control box yet. Grounds are connected solidly including Z axis and Y gantry to a common ground point on the controller.
2) Mach3 motor tuning- when I lower acel and or velocity, it seems I have less problem. In fact, when I test only the Z axis (no other axis running) the problem still happens, but much less frequently. When I run a program with X, Y and Z axis in motion at even low acel/vel setting, the problem happens intermittently, but every time i.e. not at the same code location, and not in the same place each time, but each time I try to run, the Z axis fails to move down. Mach3 Z axis DRO shows the Z is moving down, but there is no corresponding movement at the Z axis.

My Z axis is a bit heavy- but moves smoothly and without any sign of mechanical issue. Because the issue is related to not moving DOWN, and up motion seems to be OK, I don't suspect mechanicial issues at this point. I upgraded the Z axis motor to handle the extra weight. I'll get specs on the motor soon.

What should I be checking? What information do you need to help? I use gecko drivers- I'll be looking at this today so I'll post info. I use a controller that runs on my PC parallel port. I use Windows XP, and the machine has NEVER been connected to the internet. Its dedicated to the CNC. I don't even do any design work on it.

I'm wondering if my Z axis motor or its driver has some intermittent issue, or the power supply is not quite adequate when all motors are running.

I'm at a loss right now- I look forward to any help here.

Similar Threads:

• intermittently stepping the wrong direction in z
• Need Help!- 100w laser cutting intermittently!!
• Need Help!- Powermax 45, arc stops after pierce intermittently
• Z-axis intermittently dropping at tool change
• CNC Spindle Motor Intermittently Won't Start

[ger21]

When it doesn't move down, does it do anything? Make noise?

[awerby]

I wouldn't rule out mechanical issues so fast. For instance, if you had a broken helical coupler, the broken ends would still push in one direction, but would slip in the other.

[wfws]

Thanks for the quick reply Andrew. I'll revisit the coupler and see if maybe its slipping. That's a good thing to double check for sure.

[wfws]

Hi Gerry- when I move it down and its failing it makes noise that sounds like motor noise- like something is out of tune. But it could also be mechanical I can't quite tell. Low pitched grindy growly noise- but again, intermittent.

UPDATE: I took a look inside my Ascension 1000 control box. I have Gecko 202 drivers. I pushed the connector on the Z axis driver to verify it is well connected. Who knows. It felt fine, but after I pushed on it a bit, I ran several programs including a program with V carving with lots of Z axis rapid changes. It kept its location just fine. So right now, its working.
Also, in my investigation I found that the A axis (mirror for the X axis) was not the same motor tuning settings as the X axis. One was at 250 acel, the other was at 200. I fixed this, but I have no idea if this could have been part of the issue with the Z axis somehow.

Frustrating - I've had Z axis problems for several years, and have been working on this with various changes. Most recently, I returned to my first iteration, a heavy but very rigid Z axis that had a worn ballscrew. I was fortunate enough to find another ballscrew that was the same size and replaced my worn screw. At the same time I removed a bunch of crappy brackets, kludgepile install and so forth, and generally cleaned and tuned up the entire mechanicals. If I can get this Z ghost chased out, I'll be in good shape.

THanks for your input.

I'll try a cut tomorrow afternoon and see if the issue returns.

[ger21]

If it's actually making noise, like it's trying to move, then I'd look at mechanical issues first.