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

Re: Maximum speed?

Expand Messages
  • Spook2
    ... A couple youtube vids that illustrate the problem you re experiencing and one possible solution to it. http://www.youtube.com/watch?v=xbBD7KNQqsc
    Message 1 of 17 , Dec 5 12:41 PM
      --- In SherlineCNC@yahoogroups.com, "danp" <dan@...> wrote:
      >
      > A request for info from users of unmodified Sherline lathe or mill with Sherline driver and Sherline supplied stepper motors :
      >
      > 1. do you find that the maximum reachable axis speed (without stalling or missing steps) is different for X and Y on the mill, X and Z for the lathe ?
      >
      > 2. how fast is your achiveable "safe" maximum working speed(mm/min or ipm) - not to set up a record after cleaning+tuning but for regular work with the confidence that a 2-3 hour job will run uneventfully to the end with no stalling or step loss.
      >
      > I know this depends on material and depth of cut as well as tool and condition - please consider easy cuts.
      >
      > thanks
      >
      > dan pines
      >


      A couple youtube vids that illustrate the problem you're experiencing and one possible solution to it.

      http://www.youtube.com/watch?v=xbBD7KNQqsc
    • danp
      Yes, I have seen the discussion and photos about dampers in the past. However, these do not use the standard steppers supplied by sherline. on the contrary,
      Message 2 of 17 , Dec 5 10:44 PM
        Yes, I have seen the discussion and photos about dampers in the past.

        However, these do not use the standard steppers supplied by sherline. on the contrary, larger / stronger steppers were used in order to achieve high speeds.

        I am looking for speed limits when using an out of the box sherline mill - unmodified in any way. (leadscrews,nuts,steppers,driver,power supply)

        even the o-rings I added a few days ago are a modification (with great results !!) but I am still looking for info derived from an unmodified mill

        dan
        --- In SherlineCNC@yahoogroups.com, "Spook2" <spook0022000@...> wrote:
        >
        >
        >
        > --- In SherlineCNC@yahoogroups.com, "danp" <dan@> wrote:
        > >
        > > A request for info from users of unmodified Sherline lathe or mill with Sherline driver and Sherline supplied stepper motors :
        > >
        > > 1. do you find that the maximum reachable axis speed (without stalling or missing steps) is different for X and Y on the mill, X and Z for the lathe ?
        > >
        > > 2. how fast is your achiveable "safe" maximum working speed(mm/min or ipm) - not to set up a record after cleaning+tuning but for regular work with the confidence that a 2-3 hour job will run uneventfully to the end with no stalling or step loss.
        > >
        > > I know this depends on material and depth of cut as well as tool and condition - please consider easy cuts.
        > >
        > > thanks
        > >
        > > dan pines
        > >
        >
        >
        > A couple youtube vids that illustrate the problem you're experiencing and one possible solution to it.
        >
        > http://www.youtube.com/watch?v=xbBD7KNQqsc
        >
      • Tom Wade
        Don t know if this answers your question, but if you buy the package Sherline cnc deal, it s set up for 26 ipm rapids. Have no idea if that is the max
        Message 3 of 17 , Dec 5 11:03 PM
          Don't know if this answers your question, but if you buy the package
          Sherline cnc deal, it's set up for 26 ipm rapids. Have no idea if that
          is the max practical, but it seems to work fine. Actually, I believe you
          can cut at 26 ipm also, if you've got some which can be cut at that
          speed. Perhaps some type of plastics can be cut that fast.

          Tom Wade


          On 12/6/2010 1:44 AM, danp wrote:
          >
          > Yes, I have seen the discussion and photos about dampers in the past.
          >
          > However, these do not use the standard steppers supplied by sherline.
          > on the contrary, larger / stronger steppers were used in order to
          > achieve high speeds.
          >
          > I am looking for speed limits when using an out of the box sherline
          > mill - unmodified in any way. (leadscrews,nuts,steppers,driver,power
          > supply)
          >
          > even the o-rings I added a few days ago are a modification (with great
          > results !!) but I am still looking for info derived from an unmodified
          > mill
          >
          > dan
          > --- In SherlineCNC@yahoogroups.com
          > <mailto:SherlineCNC%40yahoogroups.com>, "Spook2" <spook0022000@...> wrote:
          > >
          > >
          > >
          > > --- In SherlineCNC@yahoogroups.com
          > <mailto:SherlineCNC%40yahoogroups.com>, "danp" <dan@> wrote:
          > > >
          > > > A request for info from users of unmodified Sherline lathe or mill
          > with Sherline driver and Sherline supplied stepper motors :
          > > >
          > > > 1. do you find that the maximum reachable axis speed (without
          > stalling or missing steps) is different for X and Y on the mill, X and
          > Z for the lathe ?
          > > >
          > > > 2. how fast is your achiveable "safe" maximum working speed(mm/min
          > or ipm) - not to set up a record after cleaning+tuning but for regular
          > work with the confidence that a 2-3 hour job will run uneventfully to
          > the end with no stalling or step loss.
          > > >
          > > > I know this depends on material and depth of cut as well as tool
          > and condition - please consider easy cuts.
          > > >
          > > > thanks
          > > >
          > > > dan pines
          > > >
          > >
          > >
          > > A couple youtube vids that illustrate the problem you're
          > experiencing and one possible solution to it.
          > >
          > > http://www.youtube.com/watch?v=xbBD7KNQqsc
          > >
          >
          >
        • danp
          Tom thanks. you say it seems to work fine - do you mean that you ran the axes at this speed? each one of them (x,y,z)? how about all 3 at once? my experience
          Message 4 of 17 , Dec 6 1:53 AM
            Tom

            thanks.

            you say "it seems to work fine" - do you mean that you ran the axes at this speed? each one of them (x,y,z)? how about all 3 at once?

            my experience is that the highest setable speed is on Y, X is slower and Z (going up) is slowest. on my lathe X can be set/run faster than Z

            I wonder if others have noticed the same differences on mill X,Y and on lathe X,Z. clearly Z on mill is slower going up without adding a counterweight

            I wired the sherline steppers as bi-polar and I am trying to find out whether I would get faster operation with uni-polar wiring as supplied by sherline.

            regards
            dan


            --- In SherlineCNC@yahoogroups.com, Tom Wade <tom@...> wrote:
            >
            > Don't know if this answers your question, but if you buy the package
            > Sherline cnc deal, it's set up for 26 ipm rapids. Have no idea if that
            > is the max practical, but it seems to work fine. Actually, I believe you
            > can cut at 26 ipm also, if you've got some which can be cut at that
            > speed. Perhaps some type of plastics can be cut that fast.
            >
            > Tom Wade
            >
            >
            >
          • Tom Wade
            I have never tried running all three at once. I generally use rapid Z up by itself, then follow it by a combo rapid x y, then rapid z down close, then a slower
            Message 5 of 17 , Dec 6 6:57 AM
              I have never tried running all three at once.

              I generally use rapid Z up by itself, then follow it by a combo rapid x
              y, then rapid z down close, then a slower z the last 0.1" or so.

              I've never had any problems using the rapid z up.

              I'm using EMC2. One factor MIGHT be that EMC2 continually looks at the
              vector speed, not just the x or y or z alone speed.

              Here's a link to a few early, tentative operations on the Sherline. I
              was feeding way to slow on many of these operations, but the rapids were
              generally full bore rapids.

              http://tomwade.me/tw/machinist/slim_vise/

              Tom Wade


              On 12/6/2010 4:53 AM, danp wrote:
              >
              > Tom
              >
              > thanks.
              >
              > you say "it seems to work fine" - do you mean that you ran the axes at
              > this speed? each one of them (x,y,z)? how about all 3 at once?
              >
              > my experience is that the highest setable speed is on Y, X is slower
              > and Z (going up) is slowest. on my lathe X can be set/run faster than Z
              >
              > I wonder if others have noticed the same differences on mill X,Y and
              > on lathe X,Z. clearly Z on mill is slower going up without adding a
              > counterweight
              >
              > I wired the sherline steppers as bi-polar and I am trying to find out
              > whether I would get faster operation with uni-polar wiring as supplied
              > by sherline.
              >
              > regards
              > dan
              >
              > --- In SherlineCNC@yahoogroups.com
              > <mailto:SherlineCNC%40yahoogroups.com>, Tom Wade <tom@...> wrote:
              > >
              > > Don't know if this answers your question, but if you buy the package
              > > Sherline cnc deal, it's set up for 26 ipm rapids. Have no idea if that
              > > is the max practical, but it seems to work fine. Actually, I believe
              > you
              > > can cut at 26 ipm also, if you've got some which can be cut at that
              > > speed. Perhaps some type of plastics can be cut that fast.
              > >
              > > Tom Wade
              > >
              > >
              > >
              >
              >
            • Ed Nisley
              ... The motor specs indicate they can run at 2.5 k full step/second with no mechanical load, which is 10 k quarter-steps per second. That works out to: (10 k
              Message 6 of 17 , Dec 6 7:17 AM
                On Tue, 2010-11-23 at 06:52 +0000, danp wrote:
                > 1. do you find that the maximum reachable axis speed ...

                The motor specs indicate they can run at 2.5 k full step/second with no
                mechanical load, which is 10 k quarter-steps per second. That works out
                to:

                (10 k step/sec) / (16 k step/in) = 0.625 in/sec = 37.5 in/min

                However, the stock Sherline control box can't handle that stepping rate,
                so the top speed is somewhat lower. Alas, my notes on the measurements I
                took seem well and truly buried.

                Quite some years ago I tweaked my Sherline controller to reduce the
                electrical noise, then did a clean-room reimplementation of the firmware
                on the PIC microcontrollers. My code can handle about 8000 steps/sec for
                a theoretical top speed of 0.5 in/sec = 30 in/min.

                The mill is unusable at that speed, though, because the combination of
                mechanical load and slight timing variations causes the motors to
                stutter. Not very often, but it happens.

                So I currently use 24 in/min for X & Y, 20 in/min on Z, and it Just
                Works.

                The stock Sherline box can (probably) handle that rate.

                I hung a counterweight on the Z axis, because that overhung motor is
                just too heavy for the poor little motore:

                http://softsolder.com/2009/01/11/sherline-mill-counterweight-gantry/

                The other consideration is acceleration, which is (IMHO) the Number One
                cause of "lost steps" on stepper systems. I use 5 in/sec^2 for X and Y,
                3.0 in/sec^2 on Z, which are *much* less aggressive than you'll see
                elsewhere. If the acceleration is too high, the machine will gradually
                lose steps and eventually stall out when starting under load.

                For operations requiring many short moves, which is basically all I do,
                the acceleration sets the upper speed limit, because the machine never
                moves far enough to reach top speed.

                Bottom line: speeds in the 20-25 in/min range, coupled with gentle
                accelerations, will work fine. Above that, you're asking too much of the
                hardware.

                All this is for EMC2 on a Foxconn R20-D2 (Atom D520, one core isolated
                for the real-time kernel) with a 50 microsecond base period. I had
                slightly different settings on the Dell Dimension 4550 (at 100
                microseconds) I'd been using for a few years.

                The config files I used for the Dell are here:
                http://softsolder.com/2010/11/29/sherline-emc2-cnc-mill-configuration-files/

                It'll take a while to write up the Foxconn settings, but they're not too
                different from what you see there.

                Hope that helps a bit... keep your leadscrews clean! [grin]

                --
                Ed
              • danp
                Ed thanks for the comprehensive response. I will have to look at the acceleration setting, mine is set very high to compensate for the very slow start speed. I
                Message 7 of 17 , Dec 6 11:14 AM
                  Ed

                  thanks for the comprehensive response.
                  I will have to look at the acceleration setting, mine is set very high to compensate for the very slow start speed. I found out that at higher start speeds I get truncated moves i.e. G0 x 1 (mm) ends up moving only 0.9 mm and an 0.01 move will frequently not move at all. The axes are not tight !!

                  As you said, low acceleration on a short move would mean limiting the top speed so a job with many short moves would take much longer to finish.

                  regards
                  dan

                  --- In SherlineCNC@yahoogroups.com, Ed Nisley <ed.08.nisley@...> wrote:
                  >
                  > On Tue, 2010-11-23 at 06:52 +0000, danp wrote:
                  > > 1. do you find that the maximum reachable axis speed ...
                  >
                  > The motor specs indicate they can run at 2.5 k full step/second with no
                  > mechanical load, which is 10 k quarter-steps per second. That works out
                  > to:
                  >
                  > (10 k step/sec) / (16 k step/in) = 0.625 in/sec = 37.5 in/min
                  >
                  > However, the stock Sherline control box can't handle that stepping rate,
                  > so the top speed is somewhat lower. Alas, my notes on the measurements I
                  > took seem well and truly buried.
                  >
                  > Quite some years ago I tweaked my Sherline controller to reduce the
                  > electrical noise, then did a clean-room reimplementation of the firmware
                  > on the PIC microcontrollers. My code can handle about 8000 steps/sec for
                  > a theoretical top speed of 0.5 in/sec = 30 in/min.
                  >
                  > The mill is unusable at that speed, though, because the combination of
                  > mechanical load and slight timing variations causes the motors to
                  > stutter. Not very often, but it happens.
                  >
                  > So I currently use 24 in/min for X & Y, 20 in/min on Z, and it Just
                  > Works.
                  >
                  > The stock Sherline box can (probably) handle that rate.
                  >
                  > I hung a counterweight on the Z axis, because that overhung motor is
                  > just too heavy for the poor little motore:
                  >
                  > http://softsolder.com/2009/01/11/sherline-mill-counterweight-gantry/
                  >
                  > The other consideration is acceleration, which is (IMHO) the Number One
                  > cause of "lost steps" on stepper systems. I use 5 in/sec^2 for X and Y,
                  > 3.0 in/sec^2 on Z, which are *much* less aggressive than you'll see
                  > elsewhere. If the acceleration is too high, the machine will gradually
                  > lose steps and eventually stall out when starting under load.
                  >
                  > For operations requiring many short moves, which is basically all I do,
                  > the acceleration sets the upper speed limit, because the machine never
                  > moves far enough to reach top speed.
                  >
                  > Bottom line: speeds in the 20-25 in/min range, coupled with gentle
                  > accelerations, will work fine. Above that, you're asking too much of the
                  > hardware.
                  >
                  > All this is for EMC2 on a Foxconn R20-D2 (Atom D520, one core isolated
                  > for the real-time kernel) with a 50 microsecond base period. I had
                  > slightly different settings on the Dell Dimension 4550 (at 100
                  > microseconds) I'd been using for a few years.
                  >
                  > The config files I used for the Dell are here:
                  > http://softsolder.com/2010/11/29/sherline-emc2-cnc-mill-configuration-files/
                  >
                  > It'll take a while to write up the Foxconn settings, but they're not too
                  > different from what you see there.
                  >
                  > Hope that helps a bit... keep your leadscrews clean! [grin]
                  >
                  > --
                  > Ed
                  >
                • danp
                  Tom all clear, thanks. I was looking for real facts, not specs. regards dan
                  Message 8 of 17 , Dec 6 11:16 AM
                    Tom

                    all clear, thanks. I was looking for real facts, not specs.

                    regards
                    dan
                    --- In SherlineCNC@yahoogroups.com, Tom Wade <tom@...> wrote:
                    >
                    > I have never tried running all three at once.
                    >
                    > I generally use rapid Z up by itself, then follow it by a combo rapid x
                    > y, then rapid z down close, then a slower z the last 0.1" or so.
                    >
                    > I've never had any problems using the rapid z up.
                    >
                    > I'm using EMC2. One factor MIGHT be that EMC2 continually looks at the
                    > vector speed, not just the x or y or z alone speed.
                    >
                    > Here's a link to a few early, tentative operations on the Sherline. I
                    > was feeding way to slow on many of these operations, but the rapids were
                    > generally full bore rapids.
                    >
                    > http://tomwade.me/tw/machinist/slim_vise/
                    >
                    > Tom Wade
                    >
                    >
                    > On 12/6/2010 4:53 AM, danp wrote:
                    > >
                    > > Tom
                    > >
                    > > thanks.
                    > >
                    > > you say "it seems to work fine" - do you mean that you ran the axes at
                    > > this speed? each one of them (x,y,z)? how about all 3 at once?
                    > >
                    > > my experience is that the highest setable speed is on Y, X is slower
                    > > and Z (going up) is slowest. on my lathe X can be set/run faster than Z
                    > >
                    > > I wonder if others have noticed the same differences on mill X,Y and
                    > > on lathe X,Z. clearly Z on mill is slower going up without adding a
                    > > counterweight
                    > >
                    > > I wired the sherline steppers as bi-polar and I am trying to find out
                    > > whether I would get faster operation with uni-polar wiring as supplied
                    > > by sherline.
                    > >
                    > > regards
                    > > dan
                    > >
                    > > --- In SherlineCNC@yahoogroups.com
                    > > <mailto:SherlineCNC%40yahoogroups.com>, Tom Wade <tom@> wrote:
                    > > >
                    > > > Don't know if this answers your question, but if you buy the package
                    > > > Sherline cnc deal, it's set up for 26 ipm rapids. Have no idea if that
                    > > > is the max practical, but it seems to work fine. Actually, I believe
                    > > you
                    > > > can cut at 26 ipm also, if you've got some which can be cut at that
                    > > > speed. Perhaps some type of plastics can be cut that fast.
                    > > >
                    > > > Tom Wade
                    > > >
                    > > >
                    > > >
                    > >
                    > >
                    >
                  • Ed Nisley
                    ... Machine acceleration is not a free variable: you cannot demand more torque from the stepper than it can provide. The motor s ability to provide torque sets
                    Message 9 of 17 , Dec 6 12:53 PM
                      On Mon, 2010-12-06 at 19:14 +0000, danp wrote:
                      > acceleration setting,
                      > mine is set very high to compensate for the very slow start speed.

                      Machine acceleration is not a free variable: you cannot demand more
                      torque from the stepper than it can provide. The motor's ability to
                      provide torque sets the upper limit on the machine's acceleration.

                      As nearly as I can tell, there's no (easy) way to measure the machine's
                      starting / running torque requirement, nor verify that the motor can
                      actually produce that torque under load, so you must do careful
                      experimentation to figure out what works.

                      That boils down to discovering what *doesn't* work, then backing off a
                      judicious amount. If the resulting acceleration is too slow for your
                      requirements, then a (stock) Sherline isn't the right hammer for the
                      job. You can either pimp the Sherline or buy a bigger machine, but you
                      cannot get more out of the stock hardware.

                      > at higher start speeds I get truncated moves

                      Yup, your acceleration is too high.

                      There are two limits: acceleration and traverse speed. For the short
                      moves you describe, acceleration is the limiting factor and traverse
                      speed has no effect.

                      Back the acceleration off to an absurdly low level, say 1 in/sec^2 or
                      so. You'll be able to hear / watch the motors wind up on each move, but
                      short motions will be exact. Increase the acceleration until it stops
                      working, decide how far to back off, then make more comprehensive tests.

                      You can probably get higher than 5 in/sec^2, but it's not clear there's
                      much to be had by pushing the motors harder.

                      Then twiddle the traverse speed limit; it won't get much over 25 in/sec
                      no matter what.

                      The Z axis will always be the slowest of the three, simply because of
                      the additional mass of that big spindle motor hanging out there. A
                      counterweight helps prevent the dovetails from binding, but doesn't
                      improve the acceleration one little bit.

                      Good luck!

                      --
                      Ed
                    Your message has been successfully submitted and would be delivered to recipients shortly.