Re: emc spindle

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

At 01:17 AM 3/20/2000 -0500, you wrote:
>
>
>
>Fred Proctor wrote:
>
>> The differences between minimilltask/io and bridgeporttask/io are in the
>> handling of the spindle and estop. In the minimill controller, spindle
>> commands are sent as DAC writes to the 4th axis of the servo to go. The
>> estop sense is simulated as being equal to the estop output.
>>
>> Ming Him Mak reports that it doesn't work, which isn't surprising since
>> the software probably rusted since the last time I tried this, about a
>> year ago. I'll look into this, but in the meantime brave souls can try
>> to figure this out. The way I'd do it is to run the Linux "emcmotsim"
>> instead of the RT Linux stgmod.o, so that you can debug it, e.g.,
>
>I spent an evening tracking through the code.  I found where the interpreter
>
>sends the speed setting to the next level, but it seems to only send it to
>
>Anyway, what I'd like, whenever someone gets time to poke at it,
>would be to have the DAC channel selected in some configurable
>manner, either a .ini file parameter, or by selecting the first unused
>DAC channel (MAX_AXIS+1).  This should also be configurable
>such that the DAC can give only positive values, and the spindle
>direction is commanded with a forward and a reverse start line,
>or that the DAC output would be bipolar, and there would only be
>one spindle start line.  This would make a very flexible system
>for controlling anything from a VFD on a mill spindle to a full
>servo spindle on a turning center.

I'll vote for this.  I need +-10v and closed loop with a quadrature 
encoder.  The interpreter handles a spindle speed command.  I'd prefer 
treating it just like another axis in the ini with a new type definition
- spindle - to make it distinct from the rolled linear definition - angular
that we use now.

Ray