On Sunday 30 December 2018 23:02:20 Gene Heskett wrote: > On Friday 28 December 2018 20:12:12 Andy Pugh wrote: > > What “align.zip” package? > > > > Where did you find it? What does it do? > > Well, I think I made some progress, maybe. But a couple problems > persist. > > One of which may be fixable in my code by not applying the g55-56 > offsets that determine where the db25 openings will be carved until > the code arrives at the usage of that offset. Based on the theory that > setting it at the top of the file doesn't include whatever history > there might be between hitting the r, and arriving at that actual code > some 30 to 50 minutes later. > > In the align-x.ngc or the align-y.ngc, there is a g92 x0 y0 before the > call to apply the -r to the p0 co-ord map. Its not at all desireable > to actual move the global home at anytime in this align code, so > commenting that out seems to make sense. And by inverting the division > in the statement so we get an above 1 result, then taking the atan of > that as the r in the G92, the angle of the displayed L for the axis > angles leans in the exact amount needed to align a crookedly mounted > workpiece, at which point running the panel code then draws the angled > stuff, not at the initial backplot & square, but at the pattern > rotated to fit the out of alignment workpiece. This is very good, > until I switch maps to do the db25's, at which point it fails, losing > the angular compensation. > > But the G92 r#<angle> call is still moving the home position to where > its at now, and its my reading that says nothing moves if you do not > give it data for any axis's? Somethings aglay. All I am giving the > G92 is the R. > > But when r for run is entered, its instantly back to its true home > position while retaining the calculated angle, however, when it > switches to g55 and g56 to cut out the db25's, the R setting > disappears, and its carving out the db25's up to 1/4" out of position > and uncorrected, and their axis's angle red/green display in the > backplot is back to zero angle. When the run is finished, the > displayed angle, since its in g54 at that point, is correct, showing > the angle. > > Either I am not doing it right, a definite possibility, or this is a > buglet. G92 is the wrong call maybe? I'm as usual, confused. Maybe I > should give a g10 l2 p2 (or p3) the r, then in a separate call set the > x offset so it would be applied after the angle R has already been > set? > > Ideas? Thanks everybody.
I found it, (RTFM until you do "get it", see page ) seems my understanding of G10 L2 P0 was defective, I understood it to mean that it actually moved the G53 home position, but its the CURRANT map that is moved regardless of the map in use at the time. So if I probe for the angle determination, I should first set that var globally to zero to create it, change it if needed, and use it in the "G10 L2 P0 X# R#" immediately after the G55 or G56 commands. I've not done that yet, but will try it. I don't do it to the G54 map because there is later in this code, a command series to cut the panel out of a larger sheet, using the G54 map, currently switched off. I am going to actually mount the boxes panels by its two holes to a spoil board, then mount the spoil board to this mills bed, then probe it to determine any angle corrections needed with G38.2's, changing the align code so as to use the G38.2 found locations of the edges of the panel. That code hasn't been written yet. No biggie though. Learning is such great fun... ;-) And TBT, I did wonder why you would allow such an easy way to gut shoot oneself by actually offsetting the global home position with a P0 argument, this makes far more sense now. Further reading seems to indicate that perhaps the G28 series is what I should be using for this??? Comments on that method please, and thanks. Have a happy and prosperous 2019 everybody. 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-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
