Hello Guest it is June 15, 2019, 09:33:26 PM

Author Topic: Rogue step pulses when system idle  (Read 2821 times)

0 Members and 1 Guest are viewing this topic.

Offline beefy

*
  •  138 138
    • View Profile
Rogue step pulses when system idle
« on: January 13, 2011, 05:18:50 AM »
Hi Everyone,

Built a plasma cutting table, using Mach3, MP3000 from Candcnc with associated Mach plugin, Gecko G203V drives for the stepper motors.

I have a rather strange problem which I believe is not caused by Mach or its MP3000 plugin. Lots of people with this system and never heard of this problem. The MP3000 has opto-isolation, buffering, etc. 

Mach is online (not in reset), but is idle (no gcode running) and when I click various tabs and buttons in Mach can hear one of my stepper motors nudge. It's exaclty when I do the clicks on screen items.

However, I also discovered when I was just standing there that these rogue step pulses also happened without me clicking on the screen. I looked close up at my gantry stepper and could see the shaft turn ever so slightly when I heard the dreaded nudge. Roughly every 10 seconds.

Doesn't happen when I put Mach in reset but I've been told the MP3000 pulls up all its outputs when Mach is in reset.

So I'm guessing I've got some driver / service / program in XP causing this or it's just some hardware problem.

All in one motherboard MSI G31M3 V2, socket LGA775 , P4 3.2 Ghz, built in parallel and serial ports, (tried an independant graphics card and set this in the bios as primary graphics adapter but problem remains). Have a 2nd parallel port installed in one of the PCI slots, unused and just sitting there if I ever need it.

Mach driver test always seems to be excellent at 100 khz

Formatted hard disk, installed XP in standard PC mode, did the recommended XP optimisation for running with Mach. No other programs besides Mach3 and the MP3000 plugin.

Any suggested process of elimination, bios settings, disabling drivers, services, processes, etc. What about trying the PCI parallel port instead.

Does anyone think this is a hardware issue and maybe I'd be better off changing the motherboard out.

I'm guessing a few people will say noise but the fact it happens exactly the instant I do mouse clicks rules that out a bit, would you agree.

Thanks,

Keith.

Offline Tweakie.CNC

*
  • *
  •  7,808 7,808
  • Super Kitty
    • View Profile
    • Tweakie.CNC
Re: Rogue step pulses when system idle
« Reply #1 on: January 13, 2011, 11:49:37 AM »
Keith,

Electrical noise can be generated by (or from) the PC parallel port and on the ground return rather than any particular pin. This could easily be triggered by a mouse click as well as other things so noise should not be ruled out too easily. You could perhaps check for earth loops as an initial source of the problem.
Connecting a small 0.1uF capacitor from each input (limit switches, estop etc etc) to ground (at the breakout board) is well worth trying (inputs only).

Tweakie.
Success consists of going from failure to failure without loss of enthusiasm.  Winston Churchill.
Re: Rogue step pulses when system idle
« Reply #2 on: March 31, 2012, 08:37:59 PM »
Hi Keith,  Sounds like you may have similar problem to me, well did have.
Am building a large plotter cutter 6 M by 2 M and I had the nudges and also the creeps. I spent days and tried different ideas and re-running my control wires etc away from power lines in my electrical cabinet. Sometimes after re-applying my cable ties (trying to keep everything tidy) different steppers would move and sometimes quite violently. The X axis would scream and travel about a Meter and the the Y axis would follow.

I'm using CNC C10 breakout board and I would only have to lightly touch the enable wire and the motors would go crazy. Me included !!
I changed my Gecko drivers around and still had problem.
Tried capacitors as mentioned by Tweakie and others in previous posts.
However I purchase a small power screw driver as I was sick of changing wires and using a long handled driver in the cabinet and then used it to go over all the screws to a set torque.
YOU guessed it. some of the stepper wires torqued up a bit more and problem gone completely.

I had tested the wiring on all circuits by connecting up all the components on a bench and had no problems at all.  And when I first attached the wires in the cabinet I would screw them in one at a time and then try to tug them out.  Not good enough I found.

As Tweakie's postings and Winston Churchill said "Success consists of going from failure to failure with loss of enthusiasm"
I would add also, loss of hair, loss of sleepless nights, loss of time, reading and searching the Forum's thousands of posts. Must admit searching has helped me also.
I'm still having trouble with my second parallel PCI card working.

Hope this helps and Good luck Keith

ET   
 

Offline Chaoticone

*
  • *
  •  5,502 5,502
  • Precision Chaos
    • View Profile
Re: Rogue step pulses when system idle
« Reply #3 on: April 01, 2012, 08:03:20 AM »
Keith, I doubt my question has anything to do with your bumps and I agree with Tweakie.  Noise can be a real pain.  When you say
Quote
Mach driver test always seems to be excellent at 100 khz

Do you have your kernal speed set to 100 kHz and if so does it need to be that high?

Brett
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!

Offline RICH

*
  • *
  •  7,337 7,337
    • View Profile
Re: Rogue step pulses when system idle
« Reply #4 on: April 01, 2012, 09:28:29 PM »
Seems the stepper must be getting some kind of signal such that they will change the existing holding position.
The question is what is causing it and that could be a number of different things.

Could be something is triggering the current disable on the drive, maybe disable current limiting of the drives ( not sure about the G302V but you can do that with the G201 via a jumper). If wires from the drive to the motors are shielded then conductive / inductive noise pickup should not be a problem.

RICH