Re: [CAD_CAM_EDM_DRO] Re: simple (low cost) encoder readout software ?

On Sunday 14 November 2004 12:42 pm, caudlet wrote:

> Dave, your first challenge will be to see if they are dual output
> encoders that can be used in quadrature. There are a lot of encoders
> where direction is not a issue and they just give single pulses to be
> used for velocity or speed feedback.

Lookin' for lotsa wires... :-)

> Start by counting wires. You need at least 4. (+, - , A, B) Less
> than that and you may not have a digital encoder. Your next
> challenge is to determine + and - power leads. A lot of encoders
> will use Red and Blk as + and - respectively. YOu might have to open
> the encoder and try to determine from electrolytics or active
> components waht are power and ground leads. If its a sealed unit and
> the color codes are weird and there is no valid part number then you
> may be SOL.

I think that at least in the case of the stuff I have on hand there's a good
possibility of figuring out much of that by using a meter and seeing what's
wired how...

> If it were me I would build a simple two LED tester that had buffered LED's
> that would turn on when the input to the buffer is high. Just some simple
> transistors would work. If there is some intrest for this I will draw and
> post a schematic.

Good idea! I'd use a simple buffer chip, though. Got plenty of 7406s on
hand here...

> This won't get you the line count but it would allow you to determine the
> lead configuration. If you get far enough to make the leds flash as you
> turn the shaft then a counter hooked up to one of the signals and
> going one revolution of the shaft will give you the basic line
> count.

'nother good idea! Maybe I can use some of my junkbox parts to get this
tested out. Any thoughts on how long a counter I'd need? I have no idea how
many pulses/revolution are typical for these things. I looked at the wheel
on the ones I have on hand here and all I can tell you is there's way more
than I'd want to try and count. :-)

> Have multiple ground inputs so you can handle differential encoders.

I don't follow you here.

Otherwise, good points...