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

Re: [taigtools] Re: DSLS 3000 axis stopping

Expand Messages
  • Lester Caine
    ... No - If you see my other post the main problem was having the wrong POLARITY. DSLS3000 needs a much smaller pulse than the 2000LE. The4 fact that Kurt does
    Message 1 of 46 , Feb 2, 2007
    • 0 Attachment
      Steve Blackmore wrote:

      >>> Most of the ITX boxes uses the crappy VIA chipset too - but Lester was
      >>> warned about that many moons ago ;)
      >> Since I use these on all of my other business activities and don't have
      >> any trouble I will ignore that. I can't build a full size PC into the
      >> back of an LCD panel so I use the hardware that does :)
      >
      >>From threads down the line it appears that the DSL 3000 needs very large
      > pule widths to operate correctly and that was the overall problem..

      No - If you see my other post the main problem was having the wrong
      POLARITY. DSLS3000 needs a much smaller pulse than the 2000LE. The4 fact
      that Kurt does not give us a download page with the setting on was the
      main problem. I hade received upgrade kits and was working with a config
      file that has been working fine - but was wrong.

      > It's easy to blame it on software, when often it's a hardware problem.

      The point here was not the software as such - but the computer! The main
      reason I started programming was that I had spent weeks paying software
      'engineers' to produce an interface for a multi channel graphic card I
      had produced. It needed low level drivers to get the best performance,
      but the software guys were saying what I wanted to do was not possible.
      Bottom line - I sent them home one Friday morning - worked though till
      Sunday - gave them the low level driver Monday morning.

      Software and hardware have to work together. The problem *I* was seeing
      was on where the computer was pushing pulses faster than the controller
      could handle them. Since windows does not allow you to KNOW what the
      hardware of the computer is actually doing it is difficult to get things
      debugged. But now that I have the CORRECT information on the DSLS3000 I
      have the pair under control !!!

      > To save yourself grief with ACPI problems, "optimising" (crippling)
      > windows, and crap VIA chipsets why don't you try an SN series Shuttle
      > motherboard?

      ACPI is one of the areas where Windows takes a higher priority than
      anything else, so for the last 0.01% of niggles killing it is a good
      move. The open loop setup runs stably at 10mm/sec on a CN13000
      motherboard with ACPI on. The closed loop system was giving me a
      problems an hour in to a long run - switching to stop Mach3 at that
      point would have allowed me to just reset and carry on, but I was not
      even aware that the error signals were wired up. It would have been nice
      to know that had I plugged the probe in it may have killed the X
      channel! So 4 pins on the aux connector are now unusable.

      Current situation - I have a setup that is giving me 15mm /sec stable
      operation, by reducing the acceleration a bit over the stable closed
      loop setting. I may well be able to run better than that without error
      stops, but running at 20mm/sec only gives a few stops in hours of job
      run and does not loose position. ( With ACPI on )

      So I am back in a situation where I am in control. The machine that we
      were switching ACPI off on was a desktop one, it is still not running
      stably and is only 1.1Ghz so I am going to be shipping a configured
      Mach-in-a-box .....

      --
      Lester Caine - G8HFL
      -----------------------------
      L.S.Caine Electronic Services - http://home.lsces.co.uk
      Model Engineers Digital Workshop -
      http://home.lsces.co.uk/ModelEngineersDigitalWorkshop/
      Treasurer - Firebird Foundation Inc. - http://www.firebirdsql.org/index.php
    • Lester Caine
      Steve Just check the *UK* website I use a 1/3HP 50Hz motor for the ER16 spindle and a self contained 600W high speed unit is available as an option. ( Which is
      Message 46 of 46 , Feb 3, 2007
      • 0 Attachment
        Steve

        Just check the *UK* website

        I use a 1/3HP 50Hz motor for the ER16 spindle and a self contained 600W
        high speed unit is available as an option. ( Which is the one Warren is
        currently running )

        We have established that 1200mm/min is practical and the corrected .xml
        file, that I'll post on the website, seems to have sorted the problem. I
        will also start to post pin out details and wiring options to use with
        the 2000LE and DSLS3000 controllers.

        Warren has managed to get his job out all be it at 360mm/min but
        completing the job without problems was more important than speed today,
        and he had not been able to run anything clean yesterday. My mistake was
        I had assumed that he HAD been running Mach3 clean into the open loop
        2000LE controller, but in fact the reason for getting the closed loop
        upgrade was BECAUSE he was having occasional problems with the open loop
        setup, loosing position, and the upgrade to DSLS3000 seemed to be the
        solution.

        --
        Lester Caine - G8HFL
        -----------------------------
        L.S.Caine Electronic Services - http://home.lsces.co.uk
        Model Engineers Digital Workshop -
        http://home.lsces.co.uk/ModelEngineersDigitalWorkshop/
        Treasurer - Firebird Foundation Inc. - http://www.firebirdsql.org/index.php
      Your message has been successfully submitted and would be delivered to recipients shortly.