[Contract_Pilot]

Finally got my electronics wired to the lathe and got some movement.

When I hooked up the encoder DRC M21AASODB4D-2000 "Data Sheet For Encoder" mach3 will show somewhat correct and stable spindle speed to 360rpm but after that it is all over the place. In diagnostics I do not even see the Index pulse blink.

I have it Wired +5V, Ground, ZR+ (Index)

Tried messing with Debounce anything other than 0 make it inoperable.

I Did a Google search but only thing I came up with is maybe the pulse may be to short? And will require a stretching circuit. Monostable or "One Shot" or Pulse Extender Circuit

Would there be any other setting in mach3 I can try?

Thanks.

[Al_The_Man]

Mach itself is not going to read the index pulse, especially on a 2000p/rev encoder.
You need a pulse stretcher.
Al.

[RCaffin]

mach3 will show somewhat correct and stable spindle speed to 360rpm but after that it is all over the place. In diagnostics I do not even see the Index pulse blink.

Known problem. Typically an encoder does not have the guts to drive much at all. Very slow stuff, as you found, may be OK, but fast ... nope.
It's not that the pulse is too short; it just 'fades away' at speed when wrongly loaded. You will need an interface - often called a Break out Board or BoB. And a CRO would help.

Having just rebuilt my system to have proper encoder interfaces good to about 3,000 rpm, I sympathise.

Cheers
Roger

[Contract_Pilot]

Ok, Playing around I had to build a simple RC filter 10K & 5UF was the magic combo for the ZR+ Index Pulse. Now the RPM Displayed in mach3 is correct and stable now at all speeds!

Now that I have it all wired with flying leads I can do finial wiring and she will be ready to start making chips! Lets hope no more setbacks or delays.

[RCaffin]

a simple RC filter 10K & 5UF was the magic combo for the ZR+ Index Pulse. Now the RPM Displayed in mach3 is correct and stable now at all speeds!

Well, the general idea of an RC filter is certainly right, but 10k and 5 uF give a time constant of 50 milliseconds. If those are the correct values, you should be getting an RPM reading of zero. Do you maybe mean 5 nF, or 5 pF? I use 10k and 150 pF myself.
Or do you mean a pull-up of 10k with 5 uF in parallel?

Cheers
Roger

[lancut]

Maybe this help. While back I have downloaded a pdf created by Chris Humpris
and where I found simple encoder idea that is very accurate and works well with Mach3 'INDEX"..

Good luck.

[RCaffin]

That's the general idea, but only if you do not already have an encoder mounted on the motor.
If you already have an encoder you need to cut off the bit to the left of the Schmidt trigger, and (probably) add a pull-up resistor to the encoder output.
A 74HC14 will do nicely for the Schmidt trigger, and does not need the buffer transistors. It should be mounted near the BoB.
What the circuit does not show is any filtering, which is very often needed.

Cheers
Roger

[Contract_Pilot]

RCaffin, Yes I am Using RC as a pull-up was not able to find in my stash of parts pF or nF caps so used what I had. I will try and scavenge some from some scrap pcb's and make a proper filter. But it is working with what I have done. Only issue now is when the machine is powered down the spindle speed in mach defaults to a few hundred RPM not 0 so it is still receiving a signal.

I have a few 556 (Dual 555's) and was looking at the Schmidt trigger based on the 555 to clean it up.

[RCaffin]

OK, so the encoder is likely pulling the line down sharply enough, although that capacitor will be loading the encoder electronics something awful.
Then when the encoder output 'lets go', the pulse will rise slowly. This will act as a pulse stretcher as well as a filter.

But you really weant to reduce the value of the cap to something more reasobnable to rotect the encoder. If the cap is across the 10k, then a 0.01 uF would be very suitable imho.

> when the machine is powered down the spindle speed in mach defaults to a few hundred RPM not 0 so it is still receiving a signal.
Um. That's a bit odd. It means the input line is waggling up and down a great old rate. Puzzling.
I suggest checking all the earth connections for a start.

Cheers
Roger

[Contract_Pilot]

Thank's to a friend that had a BOB laying around for his unfinished mill project "PMDX-126" we braved the cold and wired it up today results! Amazon will be pickup my china bob in the morning for a return and should have my pmdx-126 by the end of the week I hope. No filters or anything needed hooked encoder up and bam worked as it should! Not sure if I want to run a Ethernet Smooth Stepper or just stay with 2 LPT ports!

[RCaffin]

Really good news.
I am currently (today) rewiring my CNC to clean up the mess which grew. hand-wired custom circuits are being replaced by made PCBs, and rats nest wiring by a proper loom. Yeah, it was working, but I did not have full faith in the reliability. Your non-Chinese gear might be a similar case. :-)

The case for an ESS - is strong. A PC can only do so much by itself; add a dedicated designed-for-it custom controller to handle the very high-speed stuff which neither a PC nor Windows is ever good at, and you have a far better system imho. The ESS is arguably the best of the breed. I would urge that you do not go down the USB route - there are hard techie reasons why it is not good.

Cheers
Roger

[Contract_Pilot]

I have my ESS & PMDX-126 Up and running spindle encoder is working correct.