On Mon, 2008-04-14 at 20:22 -0500, Chris Radek wrote: > On Mon, Apr 14, 2008 at 05:44:53PM -0700, [EMAIL PROTECTED] wrote: > > > > Sounds like you are facing many of the issues we encounter with > > larger retrofits. Certainly your choice of M words to associate with > > various gears is one such set but I caution that these things are not > > standardized in any meaningful way among control vendors. > > I keep seeing this come up - seems like we should do something. > > Making some M codes that manipulate some HAL bits is not too hard. As > Ray says, picking the M codes and figuring out how they should work is > the hard part.
For my HNC I use M64/65 Px to set the one of two "gears" (clutches). Basically, I did this because it was a way to get'r done. But, after using this arrangement for a while, it seems to work okay. I use M1xx for other "missing" codes (cut-off slide, part chute, etc.), and I may assign a couple more to take over for the M64/65 gear change. But I am not particular hurry because what I have works. > For instance should they be a group so only one can be on at a time? > What turns them off? Does M2/M30/% reset them? My machine has 2 > gears, Sam's has 16; maybe the next guy's has 64? What happens if > you command a gear change while the spindle is running full speed? In > what order are the M codes run if one of our new ones shares a line > with other M codes? I bet there are other questions I haven't > thought of right away. For my setup, I know I need to pay attention to spindle motor RPM, the current gear, and appropriate status for gear changes. I prefer to handle these issues beforehand in the g-code file rather than an automated feature that tries to guess what I want. > It would be great if you guys who have experience on a variety of > machines could come up with a design that has the possibility of being > flexible enough to run most gearboxes. The wiki would be a great > place for this collaboration. > > I see M codes 10-29, 31-47, 70-99 are currently unused. There are > some others but those are the big blocks. > > Chris > I like having the core codes being standard, but for codes that are known to have variation among different machines, having features is much more important than what the code letter is. Handling code variation is an inconvenience that doesn't last long. -- Kirk Wallace (California, USA http://www.wallacecompany.com/machine_shop/ Hardinge HNC lathe, Bridgeport mill conversion, doing XY now, Zubal lathe conversion pending Craftsman AA 109 restoration Shizuoka ST-N/Bandit CNC) ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
