On Monday, January 05, 2015 05:39:32 AM andy pugh did opine And Gene did reply: > On 5 January 2015 at 03:29, Gene Heskett <[email protected]> wrote: > > Today I tried to wrap it up in a step & repeat, putting all the "bit > > in wood" into relative moves from where its at, but using absolute > > movements to move to the next place I need another such pocket > > carved. > > Rather than use relative code for this sort of thing I prefer to use > absolute code assuming that the routine entry point is (0,0) > On entry to the routine I do a G92 X0 Y0 > On Exit G92,2
I have read that several times without the light coming on as to what they might be useful for. Your suggestion to use it for this, plus the reading of it again, it now makes perfect sense. And it sounds like the perfect fix for my problem. With my jig mounting, all this stuff marches along in a straight line in the x direction, so no need to fudge y or z. Cogitating on it over my nightly beer, I think I found yet another way that would work but its not been coded yet, nor has this as I just woke up, haven't even started the coffee yet. The fact that we have around 10 co-ordinate systems available, I thought of writing a loop at the top of the program to set 5 of them to the correct offsets, and then change the co-ordinate system to use for each invocation of the subroutine. But this way sounds a lot simpler codewise. > I have a bunch of standard panel connector cutouts coded this way. > > > Unforchunately the G3 moves seem to be verboten in this setup. > > 50% error is the usual pop up. > > Depending on how you nornally do G3 moves, they are probably _already_ > in relative coordinates, so don't need any changing. I think they are, however the positioning g1 move to the 9 oclock position of the circle as its starting/ending point is absolute. Since there are either 4 or 5 of these in a row, depending on the board, ends have 4 fingers, sides have 5, each of which needs this treatment, how is the final G92.2 handled? Is one G92.2 call sufficient, or am I popping the previous values off a stack and would need to do 5 G92.2's to get back to the original? I can see arguments for doing it both ways. Flipping the board end for end, and then marching backwards across the other end comes to mind. G92.3 also would appear to be useful, but not for this. I have a few .03125 pcb mills that I'll probably use to carve the ebony buttons out of a thin sheet of it. I've made some by hand for the breadboard ends of the lid, glued then in and shaved them flush, but they really do need a contoured top shape. So I'll likely make a jig I can mount on my rotary table, with an exact fitting socket, mount and make a custom roundover bit from an old router bit by taking the guide nose off. Align the table, cut a side, rotate the table 90 degrees and repeat. Thats the plan anyway since there are 48 of them to make. Ought to be fun making the jigs. But the temps have dropped like a rock overnight, so a lot of that jig carving will be done on the sim here so mistakes won't screw up $70 worth of ebony. Thanks Andy. 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> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS ------------------------------------------------------------------------------ Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
