The blue is encoder position.. white in encoder position interpolated.. https://electronicsam.com/images/emco/Screenshot%20from%202022-09-19%2015-28-44.png
On Mon, Sep 19, 2022 at 2:05 PM John Dammeyer <jo...@autoartisans.com> wrote: > And is correct. For RPM measurement a count of 6 is still required. > Here's my math for this. Perhaps you can explain what position > interpolated means. > > So we have a spindle turning 6000 RPM or 50 RPS. That means if we didn't > have the longer slot we'd see an edge every 0.0033333333333333333333 > seconds. > > Now assume our counter (parallel port for example) is running 20kHz. > That's a tick every 50 uS so divide that into 0.00333333 seconds and you > get 66.6666666666 counts before the next hole on the disk. > > In reality you would see 66 counts and then 67 counts and then 66 counts > etc. and then 133.333333 counts for the index slot. > > Anyway, assume we test and can tell the difference between the slot and > the holes we get 66. Multiply that by 0.000050 (50uS) and we get the time > for one hole edge to hole edge which is 0.0033 seconds or inverted 303. > Scale it by 10 to get RPM which is 3030 RPM; out by 30 RPM. > > Next time we count 67 which works out to 2985 RPM; out by 15 RPM. > > Or we could accumulate 5 readings of 66,67,66,67,66 which gives us 66.4 a > bit closer at 3012 RPM but really not the 3000 we're actually turning. And > if it's the other way around 67,66,67,66,67 it's 66.6 average and is 3003 > RPM. Closer. > > Increasing to 60 holes improves things but there will still be a certain > amount of jitter since we have repeating decimals in the math. With 20kHz > we're still able to count 5 or 6 ticks but above that encoder count the > 20kHz tick clock has to be much faster. > > Anyway, that's how I'd calculate RPM. What's the position interpolated. > And is there a better way to get more accurate RPM? > John > > > > > -----Original Message----- > > From: Sam Sokolik [mailto:samco...@gmail.com] > > Sent: September-19-22 10:20 AM > > To: Enhanced Machine Controller (EMC) > > Subject: Re: [Emc-users] missing tooth index questions.. > > > > That was my plan - use position-interpolated.. 100 counts is barely > enough > > for smooth motion. > > > > sam > > > > On Mon, Sep 19, 2022 at 11:51 AM andy pugh <bodge...@gmail.com> wrote: > > > > > On Mon, 19 Sept 2022 at 17:06, John Dammeyer <jo...@autoartisans.com> > > > wrote: > > > > > > > No. > > > > A slot is still a hole. So if you are counting, say falling edges, > you > > > > will get 5 of them. If you count rising edges you will get 5 of > them. > > > > > > > > > Actually, the missing-tooth scheme takes a different opinion. Each hole > > > corresponds to 1/6 of a revolution, so the correct scale is 6. But it > only > > > responds to actual edges seen, so does not create a synthetic count in > the > > > middle of the gap, instead it recognises the approximately double > period > > > and adds two (or more) counts when it sees the gap. > > > > > > Using position-interpolated will smooth through this jitter. > > > > > > -- > > > atp > > > "A motorcycle is a bicycle with a pandemonium attachment and is > designed > > > for the especial use of mechanical geniuses, daredevils and lunatics." > > > � George Fitch, Atlanta Constitution Newspaper, 1912 > > > > > > _______________________________________________ > > > 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
