advantages of supplying phase B to spindle encoder

Hey Zahnrad, I hooked up my stock EMCO 120 encoder today and thought I'd let you know that the encoder count ended up being 150. Not the 360 PPR on the retrofit wiki. Don't know where that number came from. Even if you did add another read head for phase B then the PPR would be 600.

Drew

Drew,

Wow. That doth rightly sucketh. Hmmm... What are your plans? I could make some discs easily enough but I'd still need to buy those reader heads like Andy pointed me to. Damn.

Acutally I'm not sure that it sucks. 150 slots / 600 PPR should be just fine for threading although I have to admit that I haven't tried it yet. Many people seem to be happy with 120 hole/slot disks. I'd only worry about it if I was trying to do something like making the spindle a "C" axis and doing something with live tooling Not likely in my case.

Drew

No, I'm CERTAINLY not going to do that to such a little machine... Has anyone ever determined what the optimal count for threading is?

Has anyone ever determined what the optimal count for threading is?

Depends entirely upon how you are counting.

I tried all sorts of encoders from 1024 down to 250ppr and the software encoder could not cope with them, above a certain speed it went haywire.

I have a 120 hole disc and optical sensor, which is the limit of holes for the disc diameter.
It is not quadrature either, so it 120 pulses, don't think quadrature will give many advantages with just an A phase and may give too many pulses to work in software

I also think it sensible for accuracy to have the number of holes ( and thereby pulses ) as an exact divisor of 360.
The computer can do the calculations based on any number, but there will always be some variation in figures depending how many places the division is done to
and keeping it simple, just makes sense.

If you use a FPGA card it will be determined by the frequency of the oscillator chip used to sample, which will be way faster than a realtime thread.
PCW would be able to advise on the theoretical maximum based upon frequency of the counter,

My lathe used to have a Fanuc control with a 1024 ppr encoder in quadrature mode, which gives an idea
regards

andypugh wrote:

ArcEye wrote:

I also think it sensible for accuracy to have the number of holes ( and thereby pulses ) as an exact divisor of 360.

I don't think this matters.

Helps keep my OCD tendencies happy anyway

ArcEye wrote:

My lathe used to have a Fanuc control with a 1024 ppr encoder in quadrature mode, which gives an idea
regards

Exactly. Mazak's, too. That's why I ask. I believe that's the default in the industry, but I have no idea _why_. Hence the question.

ArcEye wrote:

Helps keep my OCD tendencies happy anyway

Ha. And likely the biggest factor driving my decisions most days... : )