Welcome to aurora.com! Log in or Sign up to interact with the aurora community.

Unstable, then non-working Y-axis

Discussion in 'RUMBA' started by Håvard Fjær, Feb 28, 2013.

?

What might be the reason?

  1. User error

    0 vote(s)
    0.0%
  2. Wrong cabling

    0 vote(s)
    0.0%
  3. Bad driver

    0 vote(s)
    0.0%
  4. Bad RUMBA board

    1 vote(s)
    50.0%
  5. Bad stepper motor

    0 vote(s)
    0.0%
  6. Bad power

    0 vote(s)
    0.0%
  7. Firmware setup

    1 vote(s)
    50.0%
  1. Håvard Fjær

    Håvard Fjær New Member

    Joined:
    Jan 28, 2013
    Location:
    Oslo, Norway
    I got my RUMBA board a few days ago, an set it up with a few stepper motors this evening. However, during my setup I experienced this:

    1) First I connected a driver and a stepper to the X-axis, and it tests OK when I "home" the motor.
    2) Next I connected another driver and stepper to the Y-axis, and still everything is OK; I am able to "home" both motors.
    3) Then I connect a driver and stepper to the Z-axis. The X-axis is working, Z is working, but now suddenly Y stopped working.
    4) I double check all cables to the stepper motors, the drivers are of course inserted the right way; everything looks as it should.
    5) I remove the Z-driver. The Y-stepper works again.
    6) I put the Z-driver back in. Now the Y-stepper works -- but suddenly stops again.
    7) I switch drivers between Y and Z. Y is not working, X and Z still working.

    ... and I continue to test out different setups; switching drivers, testing different stepper motors on the Y-axis, etc. Gradually the Y-axis dies completely.

    Intuitively I suspected a bad solder, but I see none; all the solder points looks perfect. Next theory is that a component somewhere on the board might initially be weak, then fail completely -- but I'm a software guy, and would'n know where or how to start debugging. Or perhaps the cause something else entirely?

    That's where you guys come in. Do any of you, based on the scenario above, have any idea what might have happened, and how to fix it? Is there any chance that the board was faulty initially? I cannot see what I might have done wrong, as all the cables and drivers has been connected as specified from the start.

    My setup:
    - I'm using Repetier firmware, using a Mac as a host.
    - I have the LCD controller connected, and use this for homing the axis.
    - I have tried two different power supplies: a 12V ATX, and a 30V 5A linear adjustable PSU (used at 12V, peaking at 0.8A when all motors are on).
    - 5x NEMA 17 steppers
     
  2. BubbleRep

    BubbleRep Administrator Staff Member

    Joined:
    Nov 11, 2012
    try to change the pins.h file to use one of the extruder ports as Z axis and see if this would work. (of course only a test solution)

    BR
    BubbleRep
     
  3. Håvard Fjær

    Håvard Fjær New Member

    Joined:
    Jan 28, 2013
    Location:
    Oslo, Norway
    First, I tried as you suggested, and moved the Z-axis to E2:
    #define Z_STEP_PIN 29
    #define Z_DIR_PIN 28
    #define Z_ENABLE_PIN 39

    The Y-axis still did not work; X and Z did work (with E2 being the Z)

    Then, for fun, I moved the Y to the E2 (setting the pins for Z back to the original setup):
    #define Y_STEP_PIN 29
    #define Y_DIR_PIN 28
    #define Y_ENABLE_PIN 39

    Then all X, Y and Z did work at the same time; E2 representing Y.

    This helps, as I now can have three axis and two extruders.

    Tough I'm still left with a dead port, we now have more debug info. Thanks!
     
  4. pomalin

    pomalin Member

    Joined:
    Feb 7, 2013
    double check the solder, even if it look good, it's perhaps a cold one, I say that because I had that before on a ramps, a solder that look good, and when I plug the lcd (and it's wasnt working) unplug it and plug it again the connector was pushed and the solder appear to be bad. Just my two cents.
     
  5. Håvard Fjær

    Håvard Fjær New Member

    Joined:
    Jan 28, 2013
    Location:
    Oslo, Norway
    I'll double check when I get back from work. Meanwhile, here are a couple of pictures of the X-axis and area around.

    By the way, when you say double check the solder (other than visually inspecting) -- do you mean using a multimeter, or go over the solder with an iron?
     

    Attached Files:

  6. Håvard Fjær

    Håvard Fjær New Member

    Joined:
    Jan 28, 2013
    Location:
    Oslo, Norway
    Sorry, the pictures are of course of the Y-axis.
     
  7. pomalin

    pomalin Member

    Joined:
    Feb 7, 2013
    I would say for begining check the solder of the connectors, trying to move the pins (pushing it with a flat screwdriver to see if it move), and with an iron after, but I don't want to suggest you somethnig that burn your card of course :). But the iron solution is for me the solution that gives you real clue if it's the solder.

    A picture of the pin and the bad solder just for illustration:
     

    Attached Files:

  8. roygpa

    roygpa New Member

    Joined:
    Feb 20, 2013
    I solved most of my problems by switching from Repetier to Marlin firmware. all it takes is adding the pins table to pins.h in the right way.
     
  9. David Rosendahl

    David Rosendahl New Member

    Joined:
    Jan 30, 2013
    I second re-soldering all connector pins going throw the card. Those kind of pins automatically soldered with lead-free solder seldom works good. I actually pushed a few free from the card (from the motor connectors and the end stop connectors) when I tested how good they where soldered. After re-soldering all pins by hand it now works perfectly. Even though things might seems to work for the moment, you can get strange errors from those connectors if you don't do that.

    Another problem I had with my electrical setup (That wasn't RUMBA-related) was that the X end stop gave strange random false positives... It was the X-motor that I had wired together with the end stop that gave enough interference to give those false positives. I fixed that with a simple low-pass filter (resistor-capacitor-combo).
     

Share This Page