Loading ...
Sorry, an error occurred while loading the content.

Re:How to stop losing steps with stepper motors

Expand Messages
  • kd006
    Doug, Not that your problem is the same but I was trying to get a simple cutout with four slots to work all day, finally I reloaded the most recent version of
    Message 1 of 54 , Oct 2, 2009
    • 0 Attachment
      Doug,

      Not that your problem is the same but I was trying to get a simple cutout with four slots to work all day, finally I reloaded the most recent version of Mach, wonder of wonders it followed the path, didn't wander off etc. I was not even cutting but just watching the Mach screen and it was shortcutting corners, wandering off straight lines etc.

      Just ran it again, same dxf, same tap file and perfect? Who knows.

      Just a thought if you haven't tried it yet, don't forget the reboot after you install.

      Kristin

      --- In taigtools@yahoogroups.com, Douglas Vogt <dbvogt@...> wrote:
      >
      > That's a good idea and I ran the program with the spindle motor off, the cutter centered directly over a printout of the drawing and added a line to send the table back to 0s. The first 3 of 5 sections to be cut out were perfect. Then at the fourth, the Taig went off on its own. The cutter ended up about 3/16 off in the Y direction and 1/8 in the X. The wheel is only 1.44 inches in diameter.
      >
      > So it's much worse at cutting brass but still can't cut air. Resistors measure 2.5 ohms.
      >
      >
      >
      >
      > ________________________________
      > From: Nicholas Carter and Felice Luftschein <felice@...>
      > To: taigtools@yahoogroups.com
      > Sent: Tuesday, September 29, 2009 9:42:39 PM
      > Subject: Re: [taigtools] Re:How to stop losing steps with stepper motors
      >
      >  
      > Try running your clock wheel program (cutting air) at a vastly reduced
      > feedrate, like 1 or 2 ipm. See if it ends up in the right place at the
      > end. (you can place a reference point at some unused portion of the mill
      > table.)
      >
      > Douglas Vogt wrote:
      > > I ran the drivertest.exe at original setup and it came out OK. I just
      > > ran roadrunner.tap with backlash off but it's a bit too big for the
      > > Taig for the Y axis. It tapped both the inner and outer stops. Anyway
      > > the X and Z axes came out exactly and the Y was slightly off.
      > >
      > > I reduced the acceleration to 7.5 and ran my tap of the clock wheel
      > > and it cut a 3/16 hole perfectly. Then it started cutting out one of
      > > the 5 sections to form the spokes and each .01 in. pass was almost
      > > minus 1/16 less in the Y axis (away from the column) than the
      > > previous cut. The cutout piece is actually in the form of steps. The
      > > second cutout that should be immediately to the right of the first
      > > started a full 1/4 inch lower than the first after 7 passes for the
      > > first cutout. Somehow the Y axis has problems and it's not the motor
      > > as I switched out the Z for the Y some time ago.
      > >
      > > It's something obvious and extreme but I couldn't guess at what it
      > > is. The X axis appears OK. The same resistors are on all motors. I'll
      > > have to get my multimeter out but besides the resistance, I wouldn't
      > > know what else to test for.
      >
      > --
      > felice@casco. net is Nicholas Carter and Felice Luftschein. See our
      > homepage at www.cartertools. com/nfhome. html
      >
      >
      >
      >
      >
      >
      >
      > [Non-text portions of this message have been removed]
      >
    • Ken Cline
      ... Again, just what will this test? Not current draw from the power supply, since that is already low assuming the motors are moved slowly. If the Gecko is
      Message 54 of 54 , Oct 4, 2009
      • 0 Attachment
        On 4 Oct 2009, at 10:56 AM, Jack wrote:

        > Reducing the value of the current setting resistors temporarily as
        > already mentioned, say to 2K or so, should be sufficient as a test
        > unless the supply is actually defective.

        Again, just what will this test? Not current draw from the power
        supply, since that is already low assuming the motors are moved
        slowly. If the Gecko is suspect, I'd just switch channels and see if
        the same axes fail.

        > Personally, I wouldn't run any type of supply in this service
        > without a large storage/filter cap at the output.

        Good point! Gecko recommends 2,000 - 10,000 uF across the output of a
        regulated or switching supply [http://www.geckodrive.com/upload/Step_motor_basics.pdf
        ]. That could be the problem. Make sure the supply will start
        properly with that sort of capacitive load.

        > Is anyone else here using 20' cables without problems? That seems
        > excessive to me.

        The signals driving our steppers barely extend above audio
        frequencies. I don't see a problem with 100' cables, as long as
        resistance is kept reasonably low.



        [Non-text portions of this message have been removed]
      Your message has been successfully submitted and would be delivered to recipients shortly.