On Saturday 30 November 2013 13:12:17 Frederic Rible did opine:

> On Wed, 27 Nov 2013 21:14:58 -0500, Gene Heskett <[email protected]>
> 
> wrote:
> > So, the next question is, what file do I put that offset pair into,
> > and what buttons do I click to tell LinuxCNC to reference those
> > offsets when
> > 
> > setting up to "etch" the top of a board?
> 
> Hello Gene,
> You need to set the offset in the tool table, tool #99.
> I have used the following procedure :
> - drill a very small hole with a V-carve tool in your spindle, 0.1mm
> depth - set X/Y zero on the DRO (this is the reference hole position) -
> now, move your table to center the camera on the hole
> - store the DRO X/Y reading in the tool #99
> - reload the tool table
> - move back to X=0 Y=0 (so your spindle is now back on the hole)
> - now, if you check the G43H99 checkbox, the table should move and the
> camera should be perfectly centered on the hole and you still read X=0
> Y=0 on the DRO
> - uncheck G43H99, now the spindle should move were the camera was
> pointing, and should should read again X=0 Y=0
> 
> Now, your setup is ok.

And it appears to be working well.  I coated a piece of sheet alu with a 
black magic marker pen, then made a just visible spot with the etching bit.  
But quickly found that the specular reflection from the ink made it 
invisible to the camera with its surround the lens illumination on, so had 
to take a flashlight, and hit it from an angle to get enough contrast to 
see the mark even if I drilled it 10 thou deep.  The video circle is a good 
pixel or more off center to the crosshair if I shrink it to about the 
marker size, using the crop to bring it up to the mark being about a mm in 
diameter on the screen.

> The procedure to set the zero for a new workpiece is:
> - check G43H99
> - move your table to point your camera to the needed zero
> - set DRO to 0/0

By home, or touch off?

> - uncheck G433H99
> - the spindle should now be at the needed zero position
> 
> Be careful with the nc file associated to G43H99: it does not move the
> spindle up, so you may experience collision if it is to low. I have
> modified the nc file to move up to top position before any X/Y motion.
> 
> > Do  also need to drill a 2nd hole and similarly record its location,
> > but
> > 
> > against the first set of data so as to facilitate the auto correction
> > of
> > 
> > any rotation of the workpiece about the Z axis?
> 
> There is a Z rotation compensation feature in LinuxCNC, but I do not
> know how to use it.

I believe that is what the align.zip kit of pyvcp buttons etc is supposed 
to do.  But it also comes with zero docs.  And from reading the code it 
doesn't reference the 8 global vars established by the "Save XYZ" function.
So in the long view, we should establish a procedure, or perhaps even use 
the diffs from global vars camviews button can save, so that it uses the 
first 3 of the *X saves, and the first 3 of the *Y saves to calculate the 
rotation AND apply the correction in a 2 button scheme, where the second 
button would be the "clear it  all back to defaults" button.  Done right, 
the alignX and alignY buttons would be superfluous, and all data entry for 
the math would come straight from the Save XYZ button.  However, the .ngc 
stuff in align.zip needs help, I just found it uses global var names that 
my own code is riddled with, bad dog, no biscuit for one of us.  And it 
also applied all those settings to the default G54 map, whereas I use it 
for location calibrations now, but apply the corrections to the G55 map for 
the top of a pcb, or G56 for the bottom of a pcb.  I save the G54 map just 
so I do have a default to fall back on.

However, re reading the code, it looks as if you would not need to use both 
alignx and aligny, you could record the align-start, and then move either 
axis to the other end of that edge, if x, then drive y to match the board, 
and click align x.  Ditto for the align y button, drive it up the y axis & 
correct x, and tap the align y button.  The applied results should closely 
match in either case.

At least that is the tangent my thinking is going off on.  Is there 
anything wrong with that thinking?

Cheers, Gene
-- 
"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>

All laws are simulations of reality.
                -- John C. Lilly
A pen in the hand of this president is far more
dangerous than 200 million guns in the hands of
         law-abiding citizens.

------------------------------------------------------------------------------
Rapidly troubleshoot problems before they affect your business. Most IT 
organizations don't have a clear picture of how application performance 
affects their revenue. With AppDynamics, you get 100% visibility into your 
Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to