#emc-devel | Logs for 2008-07-15

[02:16:59] <jepler> ouch, I think I just hit a bug
[02:17:05] <jepler> here's the sequence that I think did it:
[02:17:10] <jepler> G0 Z.2
[02:17:14] <jepler> G28 (moves X, Y, Z)
[02:17:30] <jepler> X-2 Y1
[02:17:55] <jepler> the 'XY-only' move goes to its destination, and then Z moves back down towards -2
[02:17:59] <jepler> er, .2
[02:18:05] <jepler> emc2 2.2.5
[02:18:43] <jepler> hm .. it did it twice based on an mdi sequence like the above one .. but not a third time
[02:18:44] <jmkasunich> aren't you on vacation?
[02:19:21] <jepler> jmkasunich: no, I got back yesterday evening
[02:19:28] <jmkasunich> oh
[02:36:59] <cradek> are you unwittingly issuing a command G28 X-2 Y1 on the third line?
[02:37:17] <jepler> G28 isn't a modal motion command, is it?
[02:37:35] <cradek> I don't know
[02:37:57] <jmkasunich> that would be somewhat astonishing
[02:38:20] <jmkasunich> does it show in the "active modal codes" box?
[02:39:15] <jepler> I no
[02:39:18] <jepler> no, it doesn't
[02:39:30] <jepler> whatever it was, I saw it happen twice -- but now it doesn't.
[02:39:40] <cradek> G28/G28.1/G30/G30.1 are group 0; G0 is group 1
[02:43:05] <cradek> I'm trying to figure out whether I could have written that bug when I wrote the new G28/G30
[02:44:47] <cradek> I'm not seeing this problem (with only minimal testing) in trunk
[02:45:25] <jmkasunich> did you see it on 2.2.5? (or is jeff the only one to see it at all?)
[02:45:32] <cradek> I haven't tried 2.2 yet
[02:46:56] <cradek> no I'm not getting it, argh
[02:47:04] <cradek> g0z.2
[02:47:05] <cradek> oops
[02:47:46] <jepler> I am pretty sure that these are the last mdi I did before I saw the behavior I'm talking about:
[02:47:49] <jepler> x-4.05
[02:47:52] <jepler> x-2y1
[02:47:54] <jepler> y-.05
[02:47:56] <jepler> g1z.01f1
[02:47:59] <jepler> g28
[02:48:01] <jepler> x-2y1 (I recall that it went at g0 speed?!)
[02:48:44] <jmkasunich> it sure sounds like it was doing "g28 x-2y1"
[02:49:02] <jepler> no; going to machine z0 would have been OK
[02:49:25] <jepler> g28 position was fine .. z<small> in g54 was a position for tool-breaking
[02:49:54] <jmkasunich> so when you issued G28, it rapided to machine home?
[02:50:09] <jepler> yes
[02:50:12] <jepler> G28 did what I expected
[02:50:24] <jepler> then I commanded an XY move
[02:50:38] <jmkasunich> and then when you issued x-2y1 (with an implied G1), it rapided first in XY, then in Z?
[02:50:45] <cradek> g28x-2y1 (if that got issued) would have gone to x-2y1 in the currently active coordinate system and then moved x,y only to machine origins
[02:50:47] <jepler> it went to the commanded XY then down towards g54's z<small>
[02:51:36] <cradek> sadly I'm not seeing it here
[02:52:09] <jepler> Of all the number of times I could see this problem, two is about the worst
[02:52:14] <jepler> two is the loneliest number
[02:52:23] <jmkasunich> heh
[02:52:29] <jmkasunich> one you can write off as a typo
[02:52:35] <jmkasunich> many you can debug
[02:56:58] <jepler> goodnight guys
[02:57:03] <jmkasunich> goodnight
[02:57:05] <cradek> goodnight
[02:57:12] <cradek> hope you/we can reproduce that.
[02:58:13] <jepler> ugh
[02:58:17] <jepler> I was hoping to forget it :-P
[02:58:33] <cradek> but it already cost you $7
[02:58:38] <cradek> you will remember, out of spite
[02:58:40] <jepler> heh
[02:58:45] <jmkasunich> pcb routing big?
[02:58:48] <jmkasunich> bit
[02:59:01] <jepler> no, a .0292 drill
[02:59:07] <jmkasunich> I have additional motivation to figure out how to do pc boards
[02:59:17] <jepler> what motivation is that?
[02:59:22] <jmkasunich> I found about 5 square feet of single sided board stock in the dumpster
[02:59:35] <jepler> oh cool
[02:59:39] <jepler> how big are the pieces?
[02:59:46] <cradek> nice. often it's double which is a pain
[02:59:57] <jepler> if you can stand to learn eagle and learn their programming language a tiny bit you can generate gcode for it pretty easily
[03:00:05] <jmkasunich> 3 pieces ~24" x ~8", one about 8 x 12
[03:00:15] <jepler> and you are better than me at being careful
[03:00:27] <cradek> any progress on your fast spindle yet?
[03:00:31] <jmkasunich> no
[03:01:10] <jmkasunich> why is double sided a pain?
[03:01:17] <jmkasunich> * jmkasunich likes ground planes
[03:01:22] <cradek> it's more than twice as hard to do
[03:02:10] <fenn> because you have to flip it over to cut out the lead holes even if it's single sided artwork
[03:02:29] <jmkasunich> countersink!
[03:02:57] <jmkasunich> (only solder on the traces side, clear the metal around the holes on the plane side with the c'sink)
[03:03:17] <fenn> right, still it's more effort than using single sided
[03:03:24] <jmkasunich> more effort yes
[03:03:30] <jmkasunich> better grounding is a good thing
[03:03:31] <fenn> i've heard of people peeling the copper off but not sure i believe it
[03:03:46] <jmkasunich> whether you actually need better grounding depends on the project
[03:13:32] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[03:15:11] <jmkasunich> goodnight all
[19:23:15] <alex_joni> hello
[19:40:03] <alex_joni> .. and good night
[21:46:00] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/quickstart/stepper_quickstart.lyx: Changed to a worksheet for using Stepconf
[21:47:12] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/ (Master_Integrator.lyx Submakefile): added file stepper_quickstart.lyx
[21:51:07] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/examples/gcode.lyx: short description of sample g code files
[21:52:16] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Submakefile: added file gcode.lyx
[22:09:44] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Master_User.lyx: Added g code examples
[22:14:22] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/ (Master_User.lyx Submakefile): Added gcode example descriptions
[22:14:23] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/examples/gcode.lyx: Added gcode example descriptions
[22:16:26] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/nc_files/comp-g1.ngc: Added G1 comp example
[22:21:49] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/ (Master_Integrator.lyx Submakefile): Added stepper quickstart
[22:21:50] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/quickstart/stepper_quickstart.lyx: Added stepper quickstart