On Sat, Mar 28, 2009 at 11:39:07PM -0500, Jon Elson wrote:
> Hi, all,
> 
> I'm trying to put together a jog pendant to show at NAMES.  To save 
> digital I/O points I want to multiplex the axis selection, then I can 
> select 3 axes with 2 points.  But, I need to do a demultiplexer, like 
> the reverse of the mux4 HAL component.  I don't see such a component, so 
> maybe I am missing a simple way to do this numerically, or something.  
> Any ideas, or should I write a new component?

I can imagine doing it with a sprinkling of and2 and not components,
or the logic component, or ladder especially if you are already
using ladder.

> While I'm asking questions, is there an example anywhere where a spindle 
> speed tach is set up to display on the VCP?  I have a shaft encoder on 
> the spindle of my minimill, and wired to an encoder counter on my UPC 
> board, for threading operations.  I can see that just taking encoder 
> delta (raw encoder counts) from the PPMC driver and multiplying by the 
> right factor would give RPM most of the time, but it would glitch when 
> the encoder counter is syncing up for threading.  I'd have to hard-code 
> in the servo period, too.  Also, a bit of filtering would give a more 
> stable value.  Has anyone done this?  Are there better signals on a HAL 
> pin somewhere?

http://www.linuxcnc.org/docs/devel/html/hal_general_ref.html#sec:CanonEncoder

It's up to the driver to figure out how to make a good velocity
output.  As you say, if the driver doesn't do this, you can use ddt or
a delta debug output, but it won't be index-safe.  Seb put a lot of
effort into the velocity output of the new mesa drivers lately, and
they are index-safe.  It would be nice if the ppmc encoders had this
feature too.

I notice the hal doc doesn't specify that the velocity should be
index-safe.  It goes on to mention the simplest implementation
which is NOT index-safe as described.  John K, what was the intent?

sim/lathe has a spindle speed readout you could copy.  It's pretty
simple.


------------------------------------------------------------------------------
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to