Greetings all;

If I can distract someone from the current gui discussion, I am having 
drift troubles with the cds cell and knife edge approach.

Background:
I have inserted a limit3 to slow the application of offset values from a 
setp done in the hal_show_config. So I can setp a quite large jump in 
position of x w/o tripping a following error. My thinking is that Z move 
comparatively slow even at 50 ipm that this limit will not get in the 
way of actively following what comes out of the lincurve module.

And I'm writing an o100 sub that does something like this, but haven't 
wired up the buttons to do it yet.

First, drive Z to 24.5 absolute and start the spindle at 600 revs, which 
seems to do a decent job of averaging the cds cell's output as read on a 
DVM in ohms mode. Max R is around 265-270 ohms with the full beam on the 
cds. Then I have a shutter clamped to the toolpost to block part of the 
beam, presumably around half of it, shooting for 400 ohms. But my 
repeatability is equal to around .001", and may be from a long term 
non-synch between the spindle revs and the A/D in the meter in use.  Its 
something I can see, with about a 50 ohm wobble if I slow the spindle to 
500 revs.

Unforch, there is not an m-pause that enables my ability to drive the x 
to restore the 400 ohm reading, so that I can read the offset from the 
dro.

So since I still have gpio's to spare on the 7i90, I want to use 2 
buttons to drive an updown and drive the equ to a setp into the 
corresponding y-val-pin, or cobble up another temporary number display 
or even use a halmeter and a stick to record or to show me when I have 
pushed the right button to bring the ohmmeter reading back to 400 ohms.  
The updown counter will need to be multiplied by .0001 of course before 
use.

So I've set a bunch of #<_vars> at the top of the file that I can feed 
the subroutine as #1 etc to move the machine. And I can rig the buttons 
in an hour or so.

So my main problem is in makeing sure I set the correct y-val pin for the 
instant z distance. Is there a way I can indirectly address this stuff 
in gcode? Like passing it those two digits and appending it to the 
lincurve.0.y-val to get the correct y-val-## address on the lincurve?

This would be tons easier to do if the 7i90 had an A/D to read things 
directly, but there's that old saw about beggars and horses. ;-)
 
Thanks for any good ideas guys.

-- 
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)
Genes Web page <http://geneslinuxbox.net:6309/gene>


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

Reply via email to