On Thursday 01 January 2015 11:16:02 John Alexander Stewart did opine And Gene did reply: > Ok - I had an issue with the encoder scale - setting it to a fraction > seems to have helped. > > It seems like I'm having lots of noise, though. Halscope on > hm2_5i25.encoder.00.input-a gives me a nice smooth signal, but reading > a spindle setting under about 500rpm gives me a reading of approx > 500rpm or more. > > Pushing it through a limit2 seems to have smoothed the faster > velocities. > > Above, say, 800 rpm, the system tracks quite well; up to the max of > 2250rpm. I can easily tell if the machine is in low or high gear, but > only when running at a requested speed of above 800 rpm. > > I'm going to put the 6n138 on a small bit of pc board to reduce the > long leads, and see if that reduces the noise somewhat. > > I am confused, though, that Halscope gives me a perfect signal, but the > encoder output is haywire. I guess I'm used to old-school scopes, > either digital or analog screens and with real probes and stuff...
I have come to the conclusion that for such as this, the halscope isn't always telling us the truth. And I believe the timing jitters we are seeing are an artifact of a finite sampling period, which is relatively short, so a one count ambiguity represents a rather large amount of noise when the software is doing the sampling. Putting a 5i25 in my lathe setup reduced it a lot, but the ideal situation is not available in the modules directory for LCNC. IMO a shift register, at least 16 bits or more wide, and 4 shift stages would be ideal, each register summed with a gain of .25, which should balance out any cyclic/mechanical irregularities. I suppose one could coble something up and halcomp it just to prove the point, but that wouldn't drop my minimum rpm enough to make a huge difference. Currently I am good down to about 200-250 revs before it gets to hammering the backgears too badly. Before the 5i25, it was hammering the backgears at 5 or 600 revs, but with the 5i25 I can triple the PGain, making it pretty stiff even at 175 revs. > On Tue, Dec 30, 2014 at 10:38 PM, Peter C. Wallace <p...@mesanet.com> wrote: > > On Tue, 30 Dec 2014, John Alexander Stewart wrote: > > > Date: Tue, 30 Dec 2014 21:45:14 -0500 > > > From: John Alexander Stewart <ivatt...@gmail.com> > > > Reply-To: "Enhanced Machine Controller (EMC)" > > > > > > <emc-users@lists.sourceforge.net> > > > > > > To: "Enhanced Machine Controller (EMC)" > > > <emc-users@lists.sourceforge.net > > > > > > Subject: [Emc-users] Possibly dumb question - 7i76 encoder for > > > spindle > > > > speed > > > > > feedback > > > > > > Hi all; > > > > > > I'm back working on my mill with spindle tachometer, to utilize > > > Andy > > > > Pugh's > > > > > automatic spindle speed selector example code. > > > > > > I have, from the spindle sensor, input going to a 6N138 > > > optoisolator's input. Ground and voltage source tied to machine. > > > It sources enough > > > > current > > > > > to drive the input quite well. (originally sent to the input of an > > > Amtel processor in the LED spindle speed display) > > > > > > Output of 6N138, with +5v, ground, to the 7i76 ENCA+ and > > > surrounding connectors on the 7i76. > > > > > > Looking with Halscope gives me a pretty good signal on > > > hm2_5i25.0.encoder.00.input-a. Running at 1,000 rpm gives me 14 > > > "square" teeth, running at 500 rpm gives me 7, and running at > > > 300rpm gives me 4. (not that the counts mean much, except that the > > > reading tracks spindle speed linearly) > > > > > > The 7i76 jumper is set for TTL input. ONLY the ENCA+ input pin has > > > > anything > > > > > connected, the IDX, ENCB+/- and ENCA- are all left floating. > > > > > > The issue I have is that the velocity never seems to do much other > > > than just futz around between 6 and 40, no matter what the spindle > > > speed. > > > > > > Here's my ini code: > > > setp hm2_5i25.0.encoder.00.counter-mode 1 > > > setp hm2_5i25.0.encoder.00.filter 1 > > > setp hm2_5i25.0.encoder.00.index-invert 0 > > > setp hm2_5i25.0.encoder.00.index-mask 0 > > > setp hm2_5i25.0.encoder.00.index-mask-invert 0 > > > setp hm2_5i25.0.encoder.00.latch-enable 0 > > > setp hm2_5i25.0.encoder.00.scale -16 > > > setp hm2_5i25.0.encoder.00.vel-timeout 1.0 > > > > > > Should I tie the other encoder lines high/low or to somewhere? > > > > (especially > > > > > the index pulse) > > > > > > Anybody with some advice? > > > > > > Thank you - John A. Stewart. > > > > If the inputs are jumpered for TTL levels you must leave the -inputs > > unconnected. The + inputs have 2K pullups to 5V so will be in a high > > state if > > left open. I forget the 7I76 encoder polarity but you may need to > > ground the > > ENCB+ input to get the count direction correct (ENCB is the direction > > input in > > count = COUNT/DIR mode ) > > > > > > > > > > > > --------------------------------------------------------------------- > > --------- > > > > > Dive into the World of Parallel Programming! The Go Parallel > > > Website, sponsored by Intel and developed in partnership with > > > Slashdot Media, is > > > > your > > > > > hub for all things parallel software development, from weekly > > > thought leadership blogs to news, videos, case studies, tutorials > > > and more. Take > > > > a > > > > > look and join the conversation now. > > > http://goparallel.sourceforge.net > > > _______________________________________________ > > > Emc-users mailing list > > > Emc-users@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > Peter Wallace > > Mesa Electronics > > > > (\__/) > > (='.'=) This is Bunny. Copy and paste bunny into your > > (")_(") signature to help him gain world domination. > > > > > > > > --------------------------------------------------------------------- > > --------- Dive into the World of Parallel Programming! The Go > > Parallel Website, sponsored by Intel and developed in partnership > > with Slashdot Media, is your > > hub for all things parallel software development, from weekly thought > > leadership blogs to news, videos, case studies, tutorials and more. > > Take a look and join the conversation now. > > http://goparallel.sourceforge.net > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > ----------------------------------------------------------------------- > ------- Dive into the World of Parallel Programming! The Go Parallel > Website, sponsored by Intel and developed in partnership with Slashdot > Media, is your hub for all things parallel software development, from > weekly thought leadership blogs to news, videos, case studies, > tutorials and more. Take a look and join the conversation now. > http://goparallel.sourceforge.net > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS ------------------------------------------------------------------------------ Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
