what causes 455 Spindle encoder quadrature error and how do you make it less picky?

[rk9268vc]

trying to get my mill to work with a new spindle motor/ encoder

turns on, turns for a few seconds, displaying the rpm in CNC12, then it stops and throws the error

what causes this error?
can i make it ignore the error and let 'er buck?

spindle works fine with encoder unplugged, just displaying 0rpm, but plug encoder in, turns a few turns, and stops with the error

my spindle turns backward about an 1/8th turn before going forwards, you think the encoder dont like this?

Does it matter that I am not using the Z/M and only A A\ B B\ ? It shouldnt throw an encoder over that would it? I didnt before i mounted it in an enclosure

would it throw an error if doing an M3 move gives a negative speed instead of a positive? Does that matter?


can i just disable this error message? It seems to be working fine for a few seconds until it stops. displays correct rpm set to, fwd and rev work. start stop work. just cant run for more than a few seconds without throwing the error. really frustrating

report_E0623479566D-0617214834_2022-06-23_21-16-38.zip

(625.94 KiB) Downloaded 23 times

here is a video of the error (.mp4 in a zip)

screen cap of encoder error june23.zip

(1.82 MiB) Downloaded 21 times

[suntravel]

Your encoder/wiring is not working right, with disabling the error the encoder will not work any better.

You must fix the encoder and the wiring.

Uwe

[eng199]

Quadrature error indicates that the A/B signals had an invalid transition. This means the encoder is not working reliably.
Because your are not using Z, I assume this encoder does nothing critical. Is it just for actual speed readout? The quadrature warnings and errors can be disabled by setting p334 and p335 to 16. Disabling the errors is not recommended for applications where the control needs encoder feedback (rigid tapping). However, if it is just for visual feedback, ignoring the problem is not a major issue.

[rk9268vc]

Disabling the errors is not recommended for applications where the control needs encoder feedback (rigid tapping). However, if it is just for visual feedback, ignoring the problem is not a major issue.

what is weird is it worked fine when i wired it up to test. even machined a few parts. then i tore it apart to make a nice enclosure and add contactors and such. now it dont wanna go and im pretty sure i didnt change any wiring. Even tried a new cable

I dont use the Z bit because my encoder is not 1:1 with the spindle.
I do want to do rigid tapping tho, i just wont be able to re-enter the same hole and i plan on using a tension-compression tapping head for a little bit of float

thanks for the parameters. I will use as a last resort

[Dave_C]

I had the same issue 4 years ago and the problem turned out to be "NOISE" and the reason is my encoder cable is not shielded. I sent the encoder to Centroid, they tested it and said it was 100% fine. I just need to cut the cable short, add a plug and make a new shielded cable. Had the same issue on the Lathe conversion but I solved it with a Centroid encoder and their cable. No more noise issue!

I did the 334 and 335 thing on the mill, and my rigid taping works just fine.

Dave C.

[eng199]

i tore it apart to make a nice enclosure and add contactors and such. now it dont wanna go

Do the contactors have snubbers on the coils? Are snubbers on any inductive loads they control?

I dont use the Z bit because my encoder is not 1:1 with the spindle.
I do want to do rigid tapping tho, i just wont be able to re-enter the same hole and i plan on using a tension-compression tapping head for a little bit of float

Using a tapping head is not rigid tapping. That is good because rigid tapping will break things when you don't set up the encoder 1:1 as recommended or have a properly counting encoder.

[rk9268vc]

So here is the problem

I ignored those two errors with setting the parameters

When i command it to go a speed, it goes exactly the speed. If i do m3s3000 it goes exactly 3000 rpm, verified with two tachometers

however, CNC12 says it is only going 26xx rpm, and when i go m4s3000 it says 29xx rpm
so it is reading slightly less rpm than it is actually turning for some reason
also, the rpm it is reading is different going FWD vs REV
what?

also I am using a shielded encoder cable, and even tried grounding the shield to the PSU GND directly and it didnt change anything
What do i try next? It is reading the pulses, even with correct direction, it just is like it is missing some pulses

[rk9268vc]

Do the contactors have snubbers on the coils? Are snubbers on any inductive loads they control?

Using a tapping head is not rigid tapping. That is good because rigid tapping will break things when you don't set up the encoder 1:1 as recommended or have a properly counting encoder.

contactor has a .5MFD cap on it. Contactor only switches the spindle driver
took the contactor out and just hardwired the driver. no change.

It is not a tapping head like you would find on a bridgeport. It just allows a little bit of float.
Here is the link
https://tormach.com/modular-tension-com ... 31807.html

[tblough]

Quadrature errors means that it IS missing pulses. You've just disabled the warnings; not fixed the problem.

[Dave_C]

Quadrature errors means that it IS missing pulses. You've just disabled the warnings; not fixed the problem.

TRUE and if your RPM is that far off you must be getting lots of missed pulses OR you have the encoder count setup wrong!

IE: A 1,000 line encoder has 4,000 pulses per rev and so on.

My noise issue was not that big of a deal and my display is right on compared to my tach reading so I've not bothered to fix the cable. (I should do that someday though)

Dave C.