On Tuesday, 29 March 2022 09:24:51 EDT andy pugh wrote:
> On Tue, 29 Mar 2022 at 14:16, Thaddeus Waldner <[email protected]> 
wrote:
> > I have spent some time trying to speed up the probe cycle for a
> > non-contact laser probe that I want to use as a kind of CMM.
> > 
> > I cannot seem to find ANY way to get access to encoder data from
> > within a gcode file without first calling a pause and sync (G66),
> > which halts all motion for a moment. I would like to avoid this.
> G-code is converted to canonical commands and queued in the motion
> queue. This means that the G-code isn't interpreted "live". What you
> are doing is deliberately keeping the queue empty to work around this.
> 
> Depending on your requirements you might be able to use halsampler /
> sampler http://linuxcnc.org/docs/stable/html/man/man9/sampler.9.html
> 
Humm, this looks as if it might also be usefull to me, but still no real 
usage examples.

Say I have tewo globally defined vars #<_z_tmp> and #<_y_tmp> which are 
the variables in a loop that determines the z depth, and instant y 
offset, and these are invoked with a fifo depth of 1, then how does it 
acquire in the gcode stream, those two vars and save them to a file? What 
I'll want to do is after thats done is pause for a few seconds to give me 
a chance to hit the stop button if the dool is getting dull. Then, when 
the tool has been replaced, do a TLO detect and correct, then init these 
vars to zero, then reload them once before entering the loop, if the file 
exists, so that the program can re-start with a fresh, sharp tool right 
where it was stopped.

paramsaver6.zip looks to be usable too, with the added observation that 
it doesn't write trash into the fifo for the first line if it overwriting 
whats there, which sampler claims it does. Either one could be bent into 
doing what I want altho paramsaver at first glance seems simpler to use. 
Its a more one stop read var and store, then read file and restore var.

And with modern SSD storage at 600M/sec read and write speeds and 
nanosecond track change speeds, looks to me that it would not disturb the 
realtime performance.  50 byte writes after the sub doing the carving has 
returned, and 50 byte reads before the machine starts the real work at 
startup.

Advice, sample g-code syntax to trigger a paramsaver write seems to be 
the hurdle.

A job for the digital M62-65 perhaps? And maybe the reverse m66-69 to get 
the data across the gcode to hal fence into paramsaver?  Ditto for read I 
assume?

Thank you.

Cheers, Gene Heskett.
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis





_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to