Re: How to get a signal to controll spindle speed?

[Ray Henry <rehenry-at-up.net>]

John

Thanks for the thoughts.  I may be getting over my head in both directions 
here by I'll try to respond anyway.  Feel free to disagree.   

On Saturday 18 May 2002 19:00, you wrote:
> Hi Ray
> most obvious solution would be a serial adc like maxim's
> max1232. needs 3 wires out the parallel port.
>
> hovever this is proving that its possible to extend
> ancient tech like the parallel port to ridiculous extents.

It does, doesn't it.  More so in light of the difficulty getting pci bus 
parports to work under linux.

> Is there any obvious reason why USB could not be used
> instead? much more flexible.

I've tended to shy away from USB because it was not really well supported 
until the 2.4 kernels and I tend to run about 6 months behind in these 
things.  I am also leary because I set up some USB breakouts for my SO's 
classroom and sometimes devices crash each other and require real effort to 
fix.

I should try some USB with the BDI-TNG and see how it does.  If it is any 
good, we should hack EMC to create USB out that is compatable with Steve 
Stallings USB motion control board and see how it goes.  From there we would 
have a better idea of what is possible here.

> A more reasonable solution would be to stuff a small
> opto interruptor or hall effect onto the shaft, connect that
> to a single chip PIC,  and get serial speed messages from that.
> Serial could then go into  standard serial ports, or usb serial,
> or a multiport serial card
> for the wealthy ones. usb devices on cypress stuff is pretty
> easy to add.
> I can post code and circuits, even a couple of proto boards and
> programmers  for the PIC if people are interested.

This is great.  Anything that you want to put up could go in the 
LinuxCNC.org/dropbox.  If you wish we could put a PIC page up on the site and 
give you control of it.

> What kind of feedback comes from the laser as to power?
> And I thought the thread started with a PWM request. where did analog
> come in? must have missed something key.

I don't think that you missed anything.  I think that I took a leap from the 
spindle speed signals used to control laser power that was the initial post 
to a feedback system that might lead us to a low cost closed loop spindle 
control for stepper driven mills.  And the leap I took was toward both 
velocity and position.

To get us back onto Jacob's track, the PIC you suggest could use the existing 
second parport spindle signals.  These are spindle forward, reverse, speed 
up, speed down, and brake.  Off and on are easy for me to understand.  Up 
down is also rather easy but without feedback, how would you know where the 
power setting was on the ramp.

Ray