On Friday 26 August 2016 11:04:03 [email protected] wrote: > Yeah the probe input definitely works. I did get the G38.2 probe to > work in some controlled circumstances. > > Say that machine Zhome=0 is 7" above the bed, and ZMin= -7.0". In > most cases this probe will be near the Z min (thin materials). Z min > is not set absurdly low (or no limit) to at least guarantee people > don't rub the NUT against the bed. The nut stops like 1/2" above the > bed. > > The G38.2 is in work coords. The Z-target must be given, but that's > in work coords, and it will fail to enter the .ngc call if the > Z-target is outside machine limits. > > So, if you have no work offset, Zmachine and Zwork are the same, and > G38.2 Z-6.5 works. > > But if the user already zeroed on one material, there might be a -5" > work offset. Zwork=0 is Zmachine=-5. It won't matter where your > z-axis currently is located, the G38.2 Z-6.5 becomes a command to move > to machine coord Z= -11.5 which, as the error says, exceeds joint 2's > negative limit. > > Someone's example I found online that I was copying started by > declaring the current Z to be work Z=0. That's undesirable because if > you abort or fail the probe cycle, your prior Z offset is still gone. > In any case it won't actually do much to fix the z-limit problem. If > you're at Zmachine = -5 and make that Zwork =0, your routine can only > probe to z=-2 max in that case, but there's no way to know what that > distance is when coding the .ngc. It could be -1" or -7". > > I guess that does mean that if you have one probe cycle fail to make > contact with a -1" drop, you would just repeat until it does because > the new cycle will reset Zwork=0 and G38.2 Z-1 down again. But that's > a bit confusing. > > Danny
Its also a problem I've not frequently encountered. I've made heavy use of it in my wood carving, where I stick a piece of pcb material someplace on the jig, probe for it, then set my co-ords for z in the map in use. But thats mainly to keep me from cutting up the jig when doing outline cuts. Wood being unhappy that its been killed, will often move far enough in its drying phase that the jig 'zero' is not good enough, so where I am cutting a roundover like on Green & Green joinery, I'll reset the reference by using another piece of pcb laying directly on the wood. Even that may not be good enough when working with s4s lumber because the mills planer may not be properly adjusted, and one edge of a 1x12 is .734" thick, and the other edge is .778" thick. Takes a lot of carefull sandpapering to hide that. :( I have 30+ board feet of mahogany on the rack, waiting for small projects that won't care before I can use it. > ---- Gene Heskett <[email protected]> wrote: > > On Friday 26 August 2016 00:27:00 [email protected] wrote: > > > I'm trying the G38.2 z-probe in an ngc call. I keep getting an > > > error "Probe move on line 6 would exceed joint 2's negative > > > limit". > > > > > > Problem seems to be that G38.2 uses the current coord system. > > > I've got the Z-home working now, and machine coords knows the > > > furthest it can go, whereas the active coord system Z-value is > > > meaningless for a max probe target. > > > > > > I tried to prefix the G38.2 with G53 (machine coord operation), > > > but I get an error that G53 only works on G0/G1. > > > > > > Am I looking at the problem right? How can I make this "probe to > > > a fixed machine coord point"? > > > > > > Danny > > > > Generally, give it a probe distance that would not exceed that > > limit. It moves in the current co-ordinate system. So if your > > limit is set 1mm above the table, the z-value you give it, shouldn't > > be lower than 1.001mm at the end of the probe move from where its at > > at the start. If the probe contact is not made by then, you get a > > different error. > > > > I assume you have verified with a halmeter, watching > > motion.probe-input, that the probe circuit is working? > > > > > ------------------------------------------------------------------ > > >---- -------- _______________________________________________ > > > Emc-users mailing list > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > 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 > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/emc-users 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 [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
