The stepper skipping steps and the encoders not working at high speed is the same issue. The bottleneck is the parallel port. The software only "looks" at the port at a finite rate. If stuff happens faster than that it is not noticed and falls into the "bucket of missing bits"
On Wed, May 6, 2020 at 5:04 AM Dan Henderson <luvtof...@gmail.com> wrote: > A lot of great input here. I’m not sure I fully understand all of the > talking points around kHz and PPR and RPM vs optocoupler speed and parallel > port timing. > > I also have 3 stepper motors in xyz that are being controlled via the same > BOB parallel port combo. They appear to be working okay with the exception > of maxing out at 1” ips jog speed before they lose steps. That seems slow > to me. They are nema 23 425 oz-in with DMQ542 drivers rated up to 80v. I’m > only supplying them 36v. > > But back to the original concern with the spindle motor rpm fluctuation > with encoder feedback. Without the encoder configured in the HAL the > spindle speed is more consistent but not 100% steady. I can drop the PPR on > the encoder from 200 down to 48. I’m not sure if this would significantly > improve the situation at the expense of lower resolution. > > I believe Gene gave me some items to look at in Halscope. I’m going to > attempt to pull this up now and see if I can draw any conclusions as to > where the fluctuation is occurring. > > Thanks for all the input! > > On Wed, May 6, 2020 at 3:44 AM John Dammeyer <jo...@autoartisans.com> > wrote: > > > 1000 PPR encoder turning 300 RPS (18000 RPM) will produce 300,000 Hz > > output. At 10000 RPM it's 167 RPS or 167Khz. > > > > I've been reading through the LinuxCNC source code for the last hour or > > two. If one was just using a parallel port and not sophisticated > external > > hardware the hal_parport.c code looks like it reads the input at the > > base_period rate; say 25kHz => 40 uS. > > > > If you figure a standard mill running max 3000 RPM (50 RPS) then a you > > probably need to look at the spindle encoder at least twice per encoder > > period. At 25kHz we're seeing a ParPort task run time of 40 uS. So I'd > > guess the max pulse size from the spindle encoder has to be at least 80 > > uS. > > > > The spindle at 50 RPS uses up 20mS. Divide that by the 80uS implies the > > largest usable encoder is 250 lines. > > > > Have I got that right? I've not found the spindle support code yet. > Only > > the HAL install code. > > > > John > > > > > > > -----Original Message----- > > > From: andrew beck [mailto:andrewbeck0...@gmail.com] > > > Sent: May-06-20 12:31 AM > > > To: Enhanced Machine Controller (EMC) > > > Subject: Re: [Emc-users] Fluctuating RPM using CUI ATM 10 encoder > > > > > > John while we are on this subject got a quick question. Just want to > > check > > > my maths. > > > > > > My encoder card on vfd max count rate is 300Khz. Max rpm is 10000rpm > > > > > > So a 1000 ppr encoder should give me 18000 rpm at 300khz. I just need > > > 10000 rpm so that should be well within spec > > > > > > Regards > > > > > > Andrew > > > > > > On Wed, May 6, 2020, 6:45 PM John Dammeyer <jo...@autoartisans.com> > > wrote: > > > > > > > > > > > > > > > > -----Original Message----- > > > > > From: Nicklas Karlsson [mailto:nicklas.karlsso...@gmail.com] > > > > > Sent: May-05-20 9:13 PM > > > > > To: Enhanced Machine Controller (EMC) > > > > > Subject: Re: [Emc-users] Fluctuating RPM using CUI ATM 10 encoder > > > > > > > > > > > Do you have a dual trace 'scope? Spin the spindle and look at > the > > > > input > > > > > > and output of the opto-isolator. See if there really is a lag. > > If > > > > you > > > > > > have a new scope would you have it capture the screen and post > it. > > > > I'm > > > > > > really skeptical that those isolators are THAT bad. > > > > > > > > > > > > Here is the datasheet for the opto. > > > > > > https://www.mouser.com/datasheet/2/143/EL817-G-26528.pdf > > > > > > They claim an 18 uS rise and fall time. A 1KHz square wave > will > > > > have a > > > > > > 1000 uS period and a 500 uS pulse width at 2 KHz the pulse width > is > > > > still > > > > > > more than 10X the raise time of the opto. But maybe the opto > is > > a > > > > fake > > > > > > counterfeit part with very poor performance. > > > > > > > > > > Opto couplers are rather slow. With a high resolution encoder > > frequency > > > > could be maybe up to a few hundred kHz, > > > > > 2000*period/revolution*3600rpm=120000period/second=120kHz, 200PPR > > will > > > > be 12kHz at 3600rpm, 1/(2*18�s) = about 27778Hz = > > > > > about a little bit below 28kHz if mad no misstake. > > > > > > > > > > > > > I think your math is out. Most of the hi res encoders are running > 2500 > > > > lines per rev. Many are less. But we're only looking at one > encoder > > > > channel and that works out to twice the number of edges but the > actual > > > > frequency is still lines per rev per second. > > > > > > > > So how many revs per second is that? You have to divide RPM by 60 > to > > > > get revs per second so 3600 RPM is 60 RPS. > > > > > > > > Now you have 60 * 2500 Hz = 150kHz. Still way too high for that > > > > opto-isolator. > > > > > > > > Now if you do like Sam did for his hex milling you update the > encoder. > > > > But initially he was milling with about 60 teeth and Linux is more > than > > > > fine for that when power tapping etc. So 60 * 60 is 3600 Hz and the > > opto > > > > would be fine for that. Even 200 lines at 60 rps (3600 RPM) is > > 12,000 > > > > hz. > > > > > > > > Don't understand how you got 27778Hz > > > > > > > > John > > > > > > > > > > > > > > _______________________________________________ > > > > > Emc-users mailing list > > > > > Emc-users@lists.sourceforge.net > > > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > > > > > > > > > > _______________________________________________ > > > > Emc-users mailing list > > > > Emc-users@lists.sourceforge.net > > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > > > > _______________________________________________ > > > Emc-users mailing list > > > Emc-users@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > -- Chris Albertson Redondo Beach, California _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
