Re: [CAD_CAM_EDM_DRO] Re: Encoders on the opposite end of the shaft

jmw@... wrote:

> Jon, you keep giving me these doses of reality. Thanks, I needed
> that.
>
> If the encoder is best placed on the end of the motor, does that
> pretty much quash the idea of using a cheap, single-end non-servo
> PMDC motors?

Absolutely not! You can belt drive the leadscrew, leaving the very end
free for a shaft encoder. This is actually more accurate, anyway, as
all errors caused by belt pitch, belt slippage and eccentricity of the
sprockets will not be seen by the encoder. Yes, I know we generally
use toothed belts, but after they are run for a while, the belt teeth
wear down, and the belt can slip just a tiny bit when reversing
directions. (My old belts would make little creaking sounds when they
did this.) I bought some brand new sprockets from W. M. Berg, and
found that the center bores were more than .010" eccentric! Fortunately,
I had to bore them up anyway, but that was a shock to find out that
they are not made accurately at all!

> I suppose one could make another end bell for the back
> of the motor, thread a hole axially in the back of the armature
> shaft ... Sureley there's no problem putting a tach on the far end?

Yes, you could do this. It is a little tricky, but certainly possible.
But, I used a fine instrument-grade belt to run the tach.
See the (awful) pictures at http://www.artsci.wustl.edu/~jmelson/CNCconv.html
3rd picture down, and next to the last. The black can is the tach.
These pictures don't show the tach or belt drive too well, but
they are already on the web page.

Jon