Spindle and encoder

hm2_7i96s.0.encoder.00.velocity-rpm should be the current RPM

Note that with a 10 PPR encoder wheel, RPM will be quite noisy unless the encoder
signal quality is very good.

A perfect quadrature signal would have 50% duty cycle on A and B and exactly
90 degree phase difference between A and B. Any deviation from these values
will cause apparent noise in the velocity (RPM) readout

For RPM display use, you can low-pass filter the RPM from the encoder to smooth
the displayed values.

As far as noise goes, on spindle synchronized moves, a 10 PPR (40 counts per turn)
will have large (9 degree) steps in the spindle position which will make the spindle
synchronized axis moves very noisy as they try to follow this stepped signal.

If you have a relatively current LinuxCNC master installation, you can substitute

hm2_7i96s.0.encoder.00.position

with

hm2_7i96s.0.encoder.00.position-interpolated

Which will eliminate the large steps in spindle position

Thanks.  The hm2_7i96s.0.encoder.00.velocity-rpm isn't the current RPM yet, however, I'd like it corrected, and please advise if know.  The last RPM is 1000 commanded in the enclosed video.  How to make a low-pass filter from the encoder to smooth the RPM?  The -interpolated was added to the end of hm2_7i96s.0.encoder.00.position at the .hal file but the attempt to launch LinuxCNC errored and 2.9.0 might not be the latest version unless something missing.    

Per the video foregoing.

What RPM does hm2_7i96s.0.encoder.00.velocity-rpm report vs the real RPM

The hm2_7i96s.0.encoder.00.velocity-rpm showed shuffled numbers.  Please go to the above link and select 'Per the video foregoing' if not already. The MDI, g97 s1000 m3 during the Hal Meter shows pin hm2_7i96s.0.encoder.00.velocity-rpm but a portable tachometer read 400 RPM.

The plot show that the A,B signals are not even close to quadrature
maybe a sensor positioning or disk reflectivity issue.

Here is what I would expect to see:

 
 

The Masso encoder came with a spacer and it was used to set the distance of the encoder from the face of the code wheel (pulley). docs.masso.com.au/wiring-and-setup/setup...asso-optical-encoder . I can try to set the offset distance of the encoder a little closer. The encoder comes with a top cover (shield) to reduce some of the surrounding light. I might want to try an alternative encoder but I'd expect the Masso encoder pretty good for the price. I wonder if I could have missed something complex or not obvious setting up the profile in the PncConfig Wizard but what? Is a suggestion for a detailed site that provides clear instructions for configuring the 7i96s? I'd like to review my configuration.

An optical encoder with code wheel slits and a photo sensor (transmissive) might be more accurate but they could be difficult to maintain if they're unshielded.

Your halscope seems to me that something is not right. It appears your vertical scale is 1 volt per division and high appears to be 1 volt but I would have expected it would be 10 volt when high. Check your wiring and test with a multimeter. You did not tell us what the Hugh voltage was.

Rod, he did.

smplc wrote:

Spindle-off, manually rotate the pulley until the line intersects the channel and is lit.
-ve, Ground
+ve, 12.95VDC power supply
Output Signal Z 9.5VDC
Output Signal B 9.49VDC
Output Signal A 9.52VDC