Back
[00:58:28] <fenn_> fenn_ is now known as fenn
[02:53:31] <Guest264> jmkasunich: passed the 160v 20a smoke test. (20ft-lbs of torque also)
[02:53:40] <Guest264> Guest264 is now known as skunkworks_
[02:54:08] <jmkasunich> nice
[02:55:00] <skunkworks_> the 6kva transformer outputs a nice 3 phase 125v supply also. That will be the actuall servo supply
[02:55:35] <skunkworks_> once we figured out how the taps worked.
[02:57:25] <skunkworks_> and still - the igbts barely get warm.. Very happy.
[03:08:06] <jmkasunich> dangit - can't find my Machinery's Handbook
[03:09:00] <SWPadnos> make sure you're not using it as a monitor stand
[03:10:15] <jmkasunich> found it
[03:18:23] <jmkasunich> debate debate
[03:18:37] <jmkasunich> M10 x 1 thread (0.393 diameter)
[03:18:44] <jmkasunich> internal thread
[03:18:52] <jmkasunich> I want good quality threads
[03:18:57] <jmkasunich> do I tap it, or threadcut it
[03:19:11] <jmkasunich> if it was external, no brainer - threadcut
[03:19:39] <jmkasunich> guaranteed concentric and square to centerline of part
[03:20:13] <SWPadnos> do you mean turning it or threadmilling?
[03:20:15] <cradek> threadcut undersize and then tap
[03:20:43] <jmkasunich> turning
[03:20:59] <jmkasunich> the challenge with threadcutting is getting a tool small enough
[03:21:03] <cradek> same idea as bore+ream, it should end up centered pretty well I think?
[03:21:03] <SWPadnos> well, I guess you could buy both tools and decide later ;)
[03:21:16] <jmkasunich> cradek: you think tapping would be more accurate (size and threadform)?
[03:21:27] <SWPadnos> only problem is lining up the tap to the already-nearly-cut threads
[03:21:32] <cradek> PD is sure easy with a tap
[03:21:35] <SWPadnos> unless you do it by hand
[03:21:39] <cradek> SWPadnos: by hand
[03:21:43] <jmkasunich> I
[03:22:03] <cradek> inside PD I don't know how to measure well
[03:22:10] <jmkasunich> I'm thinking more "bore and then tap" with the back end of the tap wrench supported by tailstock to keep it centered
[03:22:26] <jmkasunich> PD isn't super critical, I will be threadcutting the external thread to match
[03:22:35] <cradek> what is critical?
[03:22:55] <jmkasunich> squareness of thread to face of part
[03:23:00] <cradek> I don't think a tap in a bored hole will necessarily center very well.
[03:23:21] <jmkasunich> its basically a custom nut, to tighten against a bellville to set preload
[03:23:23] <cradek> ah, I bet it will turn out pretty square
[03:23:51] <jmkasunich> I've thought about tapping the nut, then screwing it onto a threadcut mandrel and facing the end square to the threads
[03:23:56] <cradek> I'd try it a few times I guess
[03:24:05] <jmkasunich> could also turn the OD while on the mandrel
[03:24:18] <jmkasunich> make the part concentric to the threads, instead of trying to make the threads concentric to the part
[03:24:20] <cradek> a tiny taper on the mandrel thread would work nice for that
[03:24:26] <cradek> good idea
[03:24:43] <jmkasunich> going for 30-36K RPM, so I want it balanced
[03:24:53] <cradek> woo
[03:25:11] <jmkasunich> wish mcmaster had M10x0.75, but the finest they show is M10x1
[03:25:30] <cradek> I wonder how well a perfectly concentric nut balances
[03:26:16] <jmkasunich> the nut will be round, not hex
[03:26:37] <jmkasunich> probably slot the end for tightening
[03:26:43] <jmkasunich> and not much mass
[03:26:57] <jmkasunich> M10 thread, no more than say 14mm OD, 10mm long
[03:27:48] <jmkasunich> 1/4 turn past first contact should compress the bearing preload bellvilles to 30 lbs
[03:28:32] <cradek> is that 30lb for a collet on a 1/8 tool?
[03:28:43] <jmkasunich> 30 lb of bearing preload
[03:28:47] <cradek> oh, bearing preload, rtfs
[03:29:06] <jmkasunich> I'm also planning for a stack of bellvilles for collet retention
[03:29:17] <jmkasunich> that will be more like 60-100 lbs tho
[03:30:57] <jmkasunich> dang - enco has M10x1 spiral point taps for less than mcmaster wants for hand taps (mcm doesn't have spiral point)
[03:32:46] <CIA-2> EMC: 03eric-johnson 07TRUNK * 10emc2/src/emc/usr_intf/ (Submakefile emcrsh.cc): This is the initial submission of the program scheduler. While it compiles without error, nothing has yet been tested.
[03:33:04] <cradek> huh, wonder what a program scheduler is
[03:33:45] <jmkasunich> ISTR someone talking about a utility that would let you say "run 10 of this part, then 5 of that, then 3 of this, etc"
[03:33:57] <jmkasunich> for plasma, where you cut a bunch of parts out of one sheet
[03:34:31] <cradek> goodnight folks
[03:38:41] <SWPadnos> night
[03:40:48] <SWPadnos> so much for that commit message
[03:41:05] <SWPadnos> "While it compiles without error, nothing has yet been tested."
[03:49:53] <garage_seb> alex_joni: are you awake?
[03:49:59] <CIA-2> EMC: 03eric-johnson 07TRUNK * 10emc2/src/emc/usr_intf/ (emcsched.cc emcsched.hh schedrmt.cc): This is the initial submission of the program scheduler. While it compiles without error, nothing has yet been tested.
[03:50:00] <CIA-2> EMC: 03eric-johnson 07TRUNK * 10emc2/src/hal/components/feedcomp.comp: This is the initial submission of the program scheduler. While it compiles without error, nothing has yet been tested.
[06:16:57] <CIA-2> EMC: 03seb 07TRUNK * 10emc2/src/hal/drivers/mesa-hostmot2/stepgen.c: fix the stepgen dithering error reported by Alex Joni
[06:19:29] <CIA-2> EMC: 03seb 07v2_3_branch * 10emc2/src/hal/drivers/mesa-hostmot2/stepgen.c: fix the stepgen dithering error reported by Alex Joni
[07:07:42] <CIA-2> EMC: 03alex_joni 07TRUNK * 10emc2/debian/emc2.files.in: attempt to fix the deb building issue
[14:27:14] <jepler> whee, now that seb fixed that problem I have no excuse not to switch from pluto to 5i20
[14:27:23] <jepler> and every reason to; I'm sure it's better overall
[14:47:33] <cradek> so I can blame seb when pluto-servo maintenance drops off?
[14:47:50] <SWPadnos> I'm ok with that
[15:03:22] <jepler> cradek: I don't use pluto-servo anyway
[15:03:44] <jepler> you'll never use your itty bitty lathe again anyway
[15:04:11] <skunkworks> I am pushing a BA servo around with the pluto.. :)
[16:01:03] <cradek> where should I put tooledit.tcl? I would prefer it to be in the path, unlike tcl/tkemc.tcl and friends
[16:01:21] <cradek> tcl/ and src/emc/usr_intf/ seem like the candidates
[16:01:50] <SWPadnos> it doesn't really need to know anything from the ini, does it?
[16:02:00] <cradek> I don't think so
[16:02:18] <SWPadnos> are you talking about packaging or source?
[16:02:24] <cradek> both
[16:02:44] <cradek> well for packaging I think it will be installed like any of the userland binaries
[16:02:49] <SWPadnos> ok, I guess for packages it should be wherever all the other executables are
[16:02:51] <SWPadnos> yep
[16:03:12] <SWPadnos> we should switch to git or something and then move stuff around in the source tree :)
[16:03:43] <SWPadnos> is latency-test (source) in the usr_intf subtree?
[16:11:56] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/image-to-gcode.py: A user asked for a 30 degree tool option, which is easy to add.
[16:21:15] <dgarr> re tooledit.tcl, i made some changes per suggestions and include patches for axis & tkemc:
http://filebin.ca/yregps/07apr09.tooledit.tcl
[16:25:26] <cradek> hi dewey
[16:25:41] <cradek> cool, I was working on checking in the last version I had from before
[16:27:46] <jepler> * jepler wonders if dgarr is a log-watcher or if it's coincidence
[16:27:54] <dgarr> i think it's usable, but if anyone sees a problem or needed features, please let me know ;-)
[16:28:00] <dgarr> log watcher -- yes
[16:33:06] <jepler> did you set up some sort of program to watch the log, or do you just reload it in your webbrowser sometimes?
[16:33:43] <dgarr> i have a script that loads it into vi
[16:37:28] <jepler> cool
[16:45:28] <cradek> if {[file tail $::argv0] == "tooledit.tcl"} {
[16:45:28] <cradek> # configure for standalone usage:
[16:45:38] <cradek> is there a better way to do this?
[16:45:55] <cradek> I want it named tooledit.tcl in the source tree, but when installed it becomes tooledit
[16:48:00] <cradek> if not I'll just changet his line to == "tooledit", no big deal
[16:50:36] <dgarr> maybe: if {[file tail $::argv0] == "tooledit.tcl" || [file tail $::arv0] == "tooledit"} {
[16:50:46] <dgarr> would allow rename or with extension
[16:52:07] <jepler> the intent is to determine if tooledit.tcl is run as a command or 'source'd into another wish or Tkinter?
[16:53:14] <dgarr> yes
[16:55:19] <jepler> if {[info script] == $argv0} { puts "main program" } { puts "sourced" }
[16:55:48] <jepler> I'm still not sure it's the best way, but it does remove any reference to the expected name of the file..
[16:59:14] <jepler> the tcl wiki suggests using [file tail [info script]] and [file tail $argv0], though I'm not sure why
[16:59:40] <jepler> maybe it's required when the script comes from PATH
[17:00:53] <jepler> I don't immediately find a case where it's necessary, but probably best to use that formulation anyway
[17:01:07] <jepler> second item on
http://wiki.tcl.tk/1384
[17:01:17] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/bin/.cvsignore: Tool table editor, written by Dewey Garrett
[17:01:19] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: Tool table editor, written by Dewey Garrett
[17:01:19] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/ (tooledit.tcl Submakefile): Tool table editor, written by Dewey Garrett
[17:01:20] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/tcl/tkemc.tcl: Tool table editor, written by Dewey Garrett
[17:01:27] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/share/axis/tcl/axis.tcl: Tool table editor, written by Dewey Garrett
[17:05:56] <cradek> lunch, bbl
[17:06:06] <jepler> indeed, lunch
[17:07:04] <dgarr> yes, that seems to be a better way:
[17:07:15] <dgarr> -if {[file tail $::argv0] == "tooledit.tcl"} {
[17:07:15] <dgarr> +if {[file tail [info script]] == [file tail $::argv0]} {
[17:56:25] <cradek> I wonder what else I put off doing while we were stabilizing
[17:56:42] <cradek> dangit, Blamelist: cradek
[17:57:16] <cradek> ohhh
[17:57:22] <cradek> hey, that's pretty neat, actually
[17:57:31] <alex_joni> yup :)
[17:57:49] <alex_joni> it is.. and that way we will always have building debs
[17:57:59] <jepler> it builds .. ship it
[18:00:03] <alex_joni> jepler: sounds like less work for the release manager then :D
[18:01:03] <jepler> that's great, because I get lazier by the day
[18:01:18] <jepler> (on the othar hand, I really hope somebody else will do 2.4.x)
[18:02:15] <jepler> "othar"? wtf
[18:02:45] <SWPadnos> I thought you were channeling Warcraft 3 :)
[18:02:49] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/Makefile: oops, tooledit needs to be installed
[18:02:50] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/debian/emc2.files.in: oops, tooledit needs to be installed
[18:02:53] <SWPadnos> Lothar
[18:03:24] <SWPadnos> or was that Starcraft? Hmmm
[18:09:27] <alex_joni> jepler: I'll take 2.4.x if I'm still around by then :D
[18:09:45] <skunkworks> where would you be?
[18:09:53] <skunkworks> honestly..
[18:09:56] <alex_joni> skunkworks: bus-scenario :D
[18:10:31] <jepler> alex_joni: OK, I've noted that and will hold you to it
[18:10:41] <alex_joni> you do that.. :)
[18:16:06] <CIA-2> EMC: 03jepler 07v2_3_branch * 10emc2/src/emc/task/emccanon.cc: from TRUNK: fix for full circles and naive cam detector
[18:17:00] <skunkworks> I thought you had already fixed that.
[18:17:05] <jepler> skunkworks: I had, but only on TRUNK
[18:17:14] <skunkworks> oh - duh
[19:02:20] <alex_joni> yay, seb fixed the hostmot2/stepgen issue reported
[19:59:10] <Martinp23> [Server Notice] Hi folks! The server you are on (kubrick) will be going down for around one hour for maintenance on Monday 13th April at 1400 UTC. We suggest that you connect to another server in the irc.freenode.net rotation to avoid disruption, asap. Thanks for using freenode and have a nice day!