Hello Guest it is March 29, 2024, 06:31:46 AM

Author Topic: Precipitous Loss of Zero  (Read 5014 times)

0 Members and 1 Guest are viewing this topic.

Offline Ansen

*
  •  20 20
    • View Profile
Precipitous Loss of Zero
« on: December 18, 2012, 11:13:12 PM »
After pecking 70,000 small holes, my Z axis suddenly decided to lose 0.05 of throw and started dragging the bit around. Has anybody seen this before?

The photo shows the pattern going from bottom to top, and left to right in small strips. The red arrow shows where the Z it lost it and started to not be able to clear the material. Clearance ablove the material was only 0.03, but like I said, it had done plenty of holes before it went south.

Having looked at this particular line in the program, the g-code seems really solid.

Offline Tweakie.CNC

*
  • *
  •  9,196 9,196
  • Super Kitty
    • View Profile
Re: Precipitous Loss of Zero
« Reply #1 on: December 19, 2012, 02:26:30 AM »
As I see it  possible causes could include;

a) The drill bit moving in its collet.

b) The Z axis loosing steps (possibly due to the velocity / acceleration settings being set too high for that axis).

(Assuming there are no positional errors for the X and Y axis, which may be difficult to see, then false steps caused by electrical noise could be considered unlikely).

Tweakie.
PEACE

Offline Ansen

*
  •  20 20
    • View Profile
Re: Precipitous Loss of Zero
« Reply #2 on: December 19, 2012, 10:26:45 AM »
I have had a loose bit before (operator error!) and that usually results in the gradual loss of depth as the bit is pushed further and further into the collet. This was sudden and in the opposite direction, so I think we can rule that out.

How can I check for the second possibility? The motor sounds smooth, and the fact that it had done 70,000 repititions already makes me think it might be something else. Or is this kind of error cumulative over time?

Could it be a glitch in the AC power (240 volt)? Like you said, the X and Y still look good all the way through the project.

Offline Ansen

*
  •  20 20
    • View Profile
Re: Precipitous Loss of Zero
« Reply #3 on: December 21, 2012, 10:13:57 AM »
Is there a definitive test for checking stepper speed and acceleration? Mostly what you find for advice is, "Does it sound right?" Since my problem only show us after many cycles, I need to be sure about speed and acceleration.
Re: Precipitous Loss of Zero
« Reply #4 on: December 21, 2012, 10:34:37 AM »
The best way to protect against missing steps is to test the drive, and see what the maximum velocity and acceleration it can handle are, then operate well below that, like 30-50% less.

Regards,
Ray L.
Regards,
Ray L.

Offline Tweakie.CNC

*
  • *
  •  9,196 9,196
  • Super Kitty
    • View Profile
Re: Precipitous Loss of Zero
« Reply #5 on: December 21, 2012, 10:37:21 AM »
As Ray has said.

Just as a general method the Velocity and acceleration are increased, in stages, until the motor stalls (squeals) then the settings are reduced by at least 30% (sometimes by as much as 50%).

Tweakie.
PEACE

Offline Ansen

*
  •  20 20
    • View Profile
Re: Precipitous Loss of Zero
« Reply #6 on: December 21, 2012, 11:40:59 AM »
When I try to set the speed higher than 369, Mach will default back to this number. The result is that I can never get the motor to stall.
Re: Precipitous Loss of Zero
« Reply #7 on: December 21, 2012, 11:44:08 AM »
Rapid speed is not going to be the limiting factor - acceleration is.
Regards,
Ray L.
Re: Precipitous Loss of Zero
« Reply #8 on: December 21, 2012, 08:09:29 PM »
What about the lubrication of the Z Axis mechanics? After 70,000 up and down motions is it running dry?

Offline Ansen

*
  •  20 20
    • View Profile
Re: Precipitous Loss of Zero
« Reply #9 on: December 21, 2012, 09:02:04 PM »
That's a good thought. If it were sticky in the up direction, the motor might stall and loose steps.