Re: [CAD_CAM_EDM_DRO] Re: EMC spindle speed control

Ray wrote:

> garyswindell@... wrote
>
> >Does EMC have the ability to control the spindle speed in some way?
> >I notice there are bits on the parallel port definition for spindle
> >speed decrease/increase, so I assume the answer is yes.
>
> Gary
>
> That's what assume gets you. <g> Spindle speed control is a neglected
> area. What you see in the bridgeportio and ini were written to work with an
> outboard device that cranked the speed on a bridgeport-like head. This
> system did not have any real feedback and control of speed. What you hear
> is what you get. There is some legacy C code from the initial K&T test
> still laying around but folks think it is badly broken.

I think it is in better shape than that. I tried to follow through the code, and it does
seem there is something missing between the 'heart' of EMC, and the I/O routines,
but I think it is on the order of one or two lines of code needing an update.
There is something new in the history comments that indicate that if there is a spare
DAC on a servo system, that spindle speed should come out on that DAC.

>
> IMHO - The EMC spindle problem is a motion module problem. The g-code
> interpreter reads spindle code like (M3 S3000) and issues canonical
> commands to NML there is just nothing that I know of hanging around the NML
> channel to comprehend or do anything with the S3000 part of the message at
> the moment. (Don't even think about constant surface speed)

I think there is more code past the NML that sets a static variable with the
desired spindle speed in RPM. (This is what gets displayed on the screen, too.)
I don't know whether this value is passed up from the real-time section, or only
exists at the interpreter level. (Sorry, I was looking at this over 6 months ago, and
the details are fuzzy, now.)

I think constant surface speed is in there, it just needs to come out the same way
that plain Sxxxx RPM does. I need to get in there and make some big hacks to allow
all the aux. I/O to be passed through the real-time section so that access to my
parallel port axis control boards can be serialized by somebody. While I'm in
there, I will attempt to figure out what else is needed to make spindle speed control
work.

Jon