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

Re: Unwanted Axis Moves / Shorting bar etc.

Expand Messages
  • xylotex
    Hi, If you get the three axis kit which includes the IDC26-DB25 cable, you do not need to make a shorting bar. All of the parallel port GND pins on the board
    Message 1 of 23 , Sep 2, 2003
      Hi,
      If you get the three axis kit which includes the IDC26-DB25 cable,
      you do not need to make a shorting bar. All of the parallel port
      GND pins on the board are connectoed together, and all 25 lines are
      used in the IDC26-DB25 cable, so all you need externally is a good
      sheilded parallel port extension cable that does wire through all 25
      lines.
      As we have seen from previous posts, the system does work with the
      laptop and the problem arises with noise from the Desktop PC. All
      though the STEP & DIR lines are pulled HIGH with a 10K resistor so
      they will not float when no device is attached, they have no other
      active or passive filtering built on to the board. Since we know the
      system works well with the laptop, the real issue becomes reducing
      the noise from the desktop system, not masking it with filters.
      It is not the case that any spike on the STEP lines will cause a
      step to be initiated. The datasheet for the chip specifies that the
      STEP line be HIGH for at least 1 uS to be valid. I have contacted
      the semiconductor manufacturer to find out if noise with time
      durations less than this are excluded from causing steps to be
      triggered, but as of yet, no reply.
      As mentioned earlier, only one end of the sheild around the
      stepper motors wires should actually be earth grounded. Some people
      prefer having this attached at the controller case. I prefer this to
      be attached at the motor end so absorbed noise will go to the CNC
      machine frame (which should be earth grounded) instead of bringing
      over to where noise sensitive devices are located (in the case).
      Just out of curiosity (you don't want to run it this way though),
      have you tried running the desktop without an earth gnd connection?
      Is it possible to bring the PC and your box close enough so that
      the IDC26-DB25 cable, when sticking out of the box (not attached like
      normal) can reach the desktop parallel port without an extension
      cable?

      Jeff
      http://www.xylotex.com
    • Bryan Mumford
      ... I was not referring to the Xylotex board. By hardware triggered step lines I meant typical logic gates. I know the Xylotex board uses an engineered
      Message 2 of 23 , Sep 2, 2003
        At 1:21 AM +0000 09-03-03, xylotex wrote:
        > It is not the case that any spike on the STEP lines will cause a
        >step to be initiated. The datasheet for the chip specifies that the
        >STEP line be HIGH for at least 1 uS to be valid.

        I was not referring to the Xylotex board. By "hardware triggered step
        lines" I meant typical logic gates. I know the Xylotex board uses an
        engineered driver chip with higher level features.

        --

        Bryan Mumford
        Santa Barbara, California
        http://www.bmumford.com
      • xylotex
        Hi Brian, Cool! I just wanted to make sure that the other 353 members of the group understood this, as the thread concerned a Xylotex product. Thanks, Jeff
        Message 3 of 23 , Sep 2, 2003
          Hi Brian,
          Cool! I just wanted to make sure that the other 353 members of
          the group understood this, as the thread concerned a Xylotex product.

          Thanks,

          Jeff

          --- In SherlineCNC@yahoogroups.com, Bryan Mumford <bryan@b...> wrote:
          > At 1:21 AM +0000 09-03-03, xylotex wrote:
          > > It is not the case that any spike on the STEP lines will cause
          a
          > >step to be initiated. The datasheet for the chip specifies that
          the
          > >STEP line be HIGH for at least 1 uS to be valid.
          >
          > I was not referring to the Xylotex board. By "hardware triggered
          step
          > lines" I meant typical logic gates. I know the Xylotex board uses
          an
          > engineered driver chip with higher level features.
          >
          > --
          >
          > Bryan Mumford
          > Santa Barbara, California
          > http://www.bmumford.com
        • Randy Gordon-Gilmore
          ... I should point out that I m not using a Xylotex controller. I m using three individual axis drivers. In the Xylotex, all the ground pins are taken care
          Message 4 of 23 , Sep 2, 2003
            At 05:30 PM 9/2/2003 -0700, I wrote:

            >Pins 18-25 are ground. I just used a piece of bare solid wire and soldered
            >it across the pins on the back of the socket in my controller box, making
            >sure not to let it touch the metal shell. I don't have the box reassembled
            >yet, so I can't report any results.

            I should point out that I'm not using a Xylotex controller. I'm using
            three individual axis drivers. In the Xylotex, all the ground pins are
            taken care of already.

            Best regards,

            Randy

            Randy Gordon-Gilmore ,----.___________ ______________ _________________
            ProtoTrains // = = === == || == == == = || == == == = == =|
            Rio Vista, CA, USA /-O==O------------o==o------------o==o-----------o==o-'
            zephyrus@... http://www.prototrains.com
          • Randy Gordon-Gilmore
            ... I have a good parallel cable with all 25 lines connected, but am using three individual General Controls drivers (one per axis). In the box I was just
            Message 5 of 23 , Sep 2, 2003
              At 06:02 PM 9/2/2003 -0700, Dave wrote:

              >What you don't want is a single ground wire which connects to all of the
              >ground pins.

              I have a good parallel cable with all 25 lines connected, but am using
              three individual General Controls drivers (one per axis). In the box I was
              just using one line for ground for each driver, but I've just added the
              busbar to tie all the ground pins together at the driver end. Otherwise
              the other 5 ground lines in the cable are not being used.

              Best regards,

              Randy

              Randy Gordon-Gilmore ,----.___________ ______________ _________________
              ProtoTrains // = = === == || == == == = || == == == = == =|
              Rio Vista, CA, USA /-O==O------------o==o------------o==o-----------o==o-'
              zephyrus@... http://www.prototrains.com
            • Alan Marconett KM6VV
              Hi Bryan, Thanks for the rundown. Ingenious to be sampling the inputs in the PICs. Are you also using opto-isolators? And then I assume you use state
              Message 6 of 23 , Sep 4, 2003
                Hi Bryan,

                Thanks for the rundown. Ingenious to be sampling the inputs in the
                PICs. Are you also using opto-isolators? And then I assume you use
                state machine code to generate the 4 phases, and also accommodate limit
                switches? How about an E-STOP switch?

                Alan KM6VV

                Bryan Mumford wrote:
                >
                > Each axis of the Sherline driver board is controlled by a separate
                > processor (four processors per board). I used four rather than one
                > because we didn't know how fast we might want to take the step rate.
                > Each processor is dedicated to responding to only one step line and
                > driving only one motor chip. Digital filtering is accomplished by
                > requiring the step command to be active for more than one query.
                > Hardware triggered step lines will trigger on any spurious spike as
                > short as few hundred nanoseconds. I use a Schmitt trigger input, but
                > I require that the signal be active on more than one inquiry. Each
                > inquiry might take 3 microseconds. A noise spike is not likely to
                > last this long. The EMC step command is never shorter than 20
                > microseconds.
                >
                > >What do you do for digital filtering? Do you have some sort of
                > >processor on the Sherline driver? I could see some sampling techniques
                > >that could be implemented for slower signals into a microprocessor, but
                > >the step lines are received at quite a high rate. Much too fast for
                > >digital sampling! If you can digitally process them in a uP, I'm
                > >impressed! Or are you talking about hardware, like some conditioning
                > >with a Schmidt (sp) trigger?
                >
                > --
                >
                > Bryan Mumford
                > Santa Barbara, California
                > http://www.bmumford.com
              • Bryan Mumford
                There are no opto-isolators. The phase data is clocked into the 7044 with serial data and clock lines, as per the spec. Sherline did not want to implement
                Message 7 of 23 , Sep 4, 2003
                  There are no opto-isolators. The phase data is clocked into the 7044
                  with serial data and clock lines, as per the spec. Sherline did not
                  want to implement limit switches. There are two headers on the board.
                  One of them will put all controller chips into reset condition, the
                  other is a panic stop and will immediately removed power from the
                  motors regardless of what the step and direction lines are doing.


                  At 10:07 AM -0600 09-04-03, Alan Marconett KM6VV wrote:
                  >Thanks for the rundown. Ingenious to be sampling the inputs in the
                  >PICs. Are you also using opto-isolators? And then I assume you use
                  >state machine code to generate the 4 phases, and also accommodate limit
                  >switches? How about an E-STOP switch?

                  --

                  Bryan Mumford
                  Santa Barbara, California
                  http://www.bmumford.com
                Your message has been successfully submitted and would be delivered to recipients shortly.