#emc | Logs for 2009-05-15

[12:25:57] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[13:08:28] <Optic> moo
[14:01:57] <Optic> moo
[14:04:42] <Dallur1> boo
[15:00:07] <alSMT> I get this running axis "NoteBook::_destroy .#BWidget.#Class#NoteBook"
[15:00:07] <alSMT> (command bound to event)
[15:00:07] <alSMT> invalid command name "138352196callit"
[15:00:07] <alSMT> while executing
[15:12:46] <skunkworks> SWPadnos: http://finance.groups.yahoo.com/group/geckodrive/message/18941
[15:13:27] <cradek> alSMT: when you do what? on what emc version?
[15:13:57] <skunkworks> cradek: taking the bus?
[15:14:38] <cradek> skunkworks: yes
[15:14:46] <skunkworks> neat
[15:14:48] <cradek> or at least that's the plan today.
[15:15:14] <skunkworks> I am down to a couple rows of shingles and a ridge capl.
[15:15:16] <skunkworks> cap.
[15:15:54] <cradek> cool
[15:16:10] <skunkworks> reggae fest this weekend.. So I don't know how much work will be done. The place will be a zoo.
[15:16:20] <alSMT> 2.3 load a program
[15:17:10] <alSMT> maybe it didn't like m2 at the end
[15:17:50] <cradek> does sim/axis do it? I'm not getting that behavior.
[15:17:54] <cradek> brb
[15:22:27] <alSMT> have not tried that
[15:27:06] <alSMT> its the m2
[15:28:41] <alSMT> m30 works
[15:30:18] <SWPadnos> so if you edit the file (with it already loaded in AXIS), then reload, you get the error?
[15:31:23] <alSMT> it just hangs
[15:31:42] <SWPadnos> ok, that's odd
[15:32:07] <alSMT> act busy I should say
[15:32:15] <SWPadnos> does ESC stop it?
[15:32:37] <alSMT> didnt try that either
[15:39:07] <alSMT> didn't do it this time after reload of axis
[15:42:48] <SWPadnos> do you have loops or other control statements in the Gcode?
[15:42:59] <SWPadnos> O words - that's the term I was looking for :)
[15:44:06] <alSMT> no G81
[15:45:34] <alSMT> it ran through once then it acted busy at the end and had to close axis out
[15:46:11] <cradek> sometimes AXIS will act busy / hang after printing an error like the one you showed. watch for that connection.
[15:46:27] <alSMT> reload and it did it again
[15:46:46] <alSMT> how long will it stay busy?
[15:47:11] <alSMT> seems fine now though
[15:47:18] <cradek> maybe until you restart it
[15:47:29] <alSMT> ok
[15:47:37] <cradek> I hope you can figure out the pattern and then tell us how to do it... it sounds important.
[15:48:20] <alSMT> ill keep trying
[15:52:20] <alSMT> it doing it again
[15:59:17] <alSMT> here is the error and program http://pastebin.ca/1423809
[15:59:24] <skunkworks> alSMT: could you http://pastebin.ca/ your gcode? (the minimum that makes it error)
[15:59:29] <skunkworks> heh
[15:59:32] <skunkworks> never mind
[16:00:35] <cradek> 2.3.0 release package?
[16:01:03] <alSMT> yes
[16:02:36] <cradek> now what do I do?
[16:03:20] <alSMT> I just reload axis then home and load program
[16:03:22] <LawrenceG> turn left, goo 30 steps forward, dig a hole
[16:06:17] <cradek> seems fine here
[16:06:26] <alSMT> ok
[16:06:48] <cradek> I've tried various orders of reloading/running/aborting/homing
[16:06:56] <cradek> running 2.3.0 on dapper
[16:06:56] <micges> here too
[16:07:27] <alSMT> does seem odd
[16:08:31] <cradek> do you have a manual quill that you use to do those first spot drills?
[16:09:13] <alSMT> no
[16:11:56] <alSMT> i don't think axis likes my computer
[16:13:22] <alSMT> it seems to be tied to the preview screen but I cant verify that it seem shifted left
[16:13:45] <cradek> you don't have "Desktop effects" turned on do you?
[16:13:56] <cradek> (one of the "effects" is to screw up opengl programs)
[16:14:02] <alSMT> yes
[16:14:10] <cradek> yes it is on?
[16:14:13] <alSMT> screen saver?
[16:14:23] <cradek> no, it's actually called "desktop effects"
[16:14:34] <alSMT> no
[16:14:39] <SWPadnos> System->Preferences->Appearance->Desktop Effects tab
[16:14:53] <cradek> thanks, I was hoping someone would know where it is
[16:15:44] <archivist> Visual Effect ?
[16:15:55] <SWPadnos> err - could be
[16:16:04] <SWPadnos> it's the last tab I think
[16:16:06] <archivist> or did name change hardy to ..
[16:16:24] <SWPadnos> dunno - that was from memory ;)
[16:16:46] <archivist> * archivist verifies off on his box
[16:17:25] <alSMT> no effects
[16:19:45] <eric_unterhausen> I'm starting to appreciate the value of a cnc lathe
[16:20:13] <archivist> only starting :)
[16:20:29] <eric_unterhausen> making things manually is killing my knees
[16:20:48] <eric_unterhausen> too old to stand by the lathe for hours
[16:21:39] <archivist> ya have to stand by the steam method to make parts to convert to new method
[16:32:14] <skunkworks> LawrenceG: how is the drive coming?
[16:35:03] <LawrenceG> skunkworks, working on the motherboard part to mount 3 drives on.... schematic is close to done
[16:35:54] <skunkworks> cool!
[16:36:06] <archivist> 3... go for 5 :)
[16:37:22] <LawrenceG> well the plan plan was to use to to go to 6 if needed
[16:37:47] <LawrenceG> flying fingers need coffee..
[18:28:06] <jboss2000> got a question, after seting a pin connection, can i force an analog +/-10v signal to my spindel drive through halcmd on the terminal screen??
[18:29:00] <jboss2000> im using hal_stg as my rt
[18:34:21] <jboss2000> my spendel refrence signal is connected physicaly to the stg board adc out "slot" 00
[18:35:03] <SWPadnos> yes. use halcmd
[18:35:26] <SWPadnos> the command would be something like setp stg.0.adc.00.out ###
[18:35:37] <SWPadnos> where ### is the volts you want on the output
[18:35:52] <SWPadnos> you can run an interactive halcmd "shell" using halcmd -kf
[18:36:19] <jboss2000> ok thanks that's what i was looking for!! :-)
[18:36:27] <SWPadnos> (-k means "keep going if there's an error", and -f means "from file", which means from the terminal if you don't specify a filename)
[18:36:28] <SWPadnos> sure
[18:36:31] <SWPadnos> man halcmd for more info
[18:36:41] <SWPadnos> or help from within the halcmd interactive session
[18:36:49] <SWPadnos> and remember, tab completion is your friend :)
[18:37:00] <jboss2000> thanks
[18:37:09] <SWPadnos> you're welcome. enjoy
[18:38:36] <skunkworks> SWPadnos: how is it going?
[18:38:59] <SWPadnos> err, depends on which part of "it' I think about at any given time :)
[18:43:00] <skunkworks> heh - I meant all of it ;)
[18:43:32] <JymmmEMC> SWPadnos: the 'DUH' part of "it"
[18:43:59] <JymmmEMC> As in... DUH, which way did he go George, which way did he go.
[18:47:04] <SWPadnos> DUH is fine
[18:47:45] <SWPadnos> a lot of the rest is fine too, I just need to get NFS and DHCP servers/clients set up, then make some mega-scripts that run stuff on remote computers and stuff
[18:48:04] <SWPadnos> and that can check to be sure that exactly the right number of files get dumped in the right place when requested
[18:48:14] <SWPadnos> and that check or errors (missing/extra files from some clients)
[18:48:23] <SWPadnos> and that check for connection/disconnection of cameras
[18:48:26] <SWPadnos> etc.
[18:50:10] <JymmmEMC> SWPadnos: Well setup a syslogd that they all report to, and SNMPd for the status that you can pole live
[18:50:28] <JymmmEMC> poll
[18:50:41] <SWPadnos> it's more realtime than that
[18:51:21] <JymmmEMC> GIT, rdiff-backup and rsync are your friends
[18:51:23] <SWPadnos> like we might get 3 or 4 events in less than a minute (or sometimes only a second or two apart), and need to be sure that the 50 files from each event are all present
[18:51:31] <SWPadnos> not for second-to-second stuff :)
[18:51:41] <JymmmEMC> I bet you can poll pretty quickly if you wanted to
[18:52:10] <JymmmEMC> or have them report an error ro snmpd
[18:52:16] <JymmmEMC> s/ro/to/
[18:53:13] <SWPadnos> if they were snmp devices, or an error could be detected like that, it would be a great idea
[18:53:36] <SWPadnos> there are 48 USB devices connected to 12 separate computers
[18:53:37] <JymmmEMC> Just create a MIB for them
[18:54:00] <SWPadnos> if an event doesn't happen on one of them, it's a little hard to tell back at the master station
[18:54:08] <JymmmEMC> USB?! Hmmmm
[18:54:15] <SWPadnos> heh
[18:54:18] <SWPadnos> DUH! :)
[18:54:55] <JymmmEMC> http://www.oidview.com/mibs/0/USB-MIB.html
[18:55:06] <JymmmEMC> Looks liek someone already wrote the MIB for you
[18:55:38] <SWPadnos> intresting. thanks
[18:55:41] <SWPadnos> interesting
[18:56:19] <SWPadnos> I'm not sure if that will tell me if soemthing *ddoesn't* happen though
[18:56:34] <SWPadnos> I don't know enough about SNMP/MIBs to tell from looking at it :)
[18:57:11] <JymmmEMC> Well, you have yout monitor station look for a missing event after n seconds
[18:57:56] <SWPadnos> hmmm. I wonder if it can do this: if (an event on any of the 48 devices) then (error if no event on any of the devices after N seconds)
[18:58:26] <JymmmEMC> sure, consider a hsot down as example
[18:58:31] <JymmmEMC> host
[18:58:47] <SWPadnos> with the potential issue that there could be another desired event before the timeout, and we want to know if there were not exactly two (or N) events on all devices
[18:59:42] <JymmmEMC> http://www.nagios.org/about/screenshots.php
[19:01:46] <SWPadnos> ok, thanks
[19:02:10] <SWPadnos> I don't know that I'll be able to figure that method out by tomorrow night, but I'll keep it in mind for later
[19:02:12] <JymmmEMC> That's just one of many that are available out there
[19:03:58] <JymmmEMC> Well, there are cmd line tools to walk the tree as well
[19:04:50] <SWPadnos> I can't tell if that USB MIB allows me to configure a specific type of event
[19:05:18] <SWPadnos> though I can probably just use a filesystem monitor anyway, there should be a file automatically created by each device for each event
[19:06:02] <JymmmEMC> http://www.oidview.com/mibbrowser.html
[19:06:26] <alex_joni> http://pastebin.ca/1423988
[19:06:59] <JymmmEMC> looks liek a weather report
[19:07:56] <alex_joni> it's actually surprisingly easy to read
[19:08:00] <alex_joni> not english though :)
[19:15:12] <JymmmEMC> alex_joni: 2009/05/15 18:53 KIWI 151853Z AUTO 18008KT 10SM CLR 21/13 A3020 RMK AO2 SLP225 T02060133
[19:18:17] <JymmmEMC> alex_joni: The above decoded... http://codepad.org/MkOr0AME
[19:23:56] <JymmmEMC> SWPadnos: Ah, it was a TRAP that I couldn't think of before that you can setup
[19:24:30] <JymmmEMC> http://www.oidview.com/snmp_trap_management.html
[19:33:04] <SWPadnos> heh.l too funny
[19:33:27] <SWPadnos> I finally connect to the NFS server, and I'm wondering why I can't see any files
[19:33:51] <SWPadnos> which is because I had just created a new directory to share, and there were no files there ...
[19:34:00] <JymmmEMC> heh
[20:12:35] <alex_joni> SWPadnos: http://www.reghardware.co.uk/2009/05/14/tosh_first_with_512gb_ssd_laptop/
[20:13:26] <SWPadnos> nice
[20:13:32] <SWPadnos> and expensive
[20:14:15] <alex_joni> in a couple of years
[20:14:35] <SWPadnos> yep
[20:14:48] <alex_joni> how's the cameras?
[20:14:54] <alex_joni> everything set?
[20:14:55] <SWPadnos> OK I guess
[20:14:58] <SWPadnos> not really
[20:15:07] <SWPadnos> but getting set, bit by bit
[20:15:14] <alex_joni> good luck ;)
[20:15:19] <SWPadnos> (bytes would be more efficient :) )
[20:55:00] <colin__> why is it when you run the puma sim, and do an xyz movement in jog the line in vis looks like an arc ?
[21:35:15] <archivist> colin__, presumably the xyz are rotary joints not linear as normal, are you expecting the kins to fix that (as I would probably)
[21:47:44] <colin__> yeah i was expecting the kins to do it
[21:48:03] <colin__> unless there is a mismatch between the vis sim and the kins model
[21:59:50] <andypugh> That puzzled me, somehow I set jog speed to zero when stopping a program. I could _not_ work out why no axes would move but the spindle would turn.
[22:59:44] <colin__> hmmm
[22:59:46] <colin__> midnight again
[23:02:53] <archivist> only just
[23:26:55] <maddash> uh guys?
[23:27:31] <maddash> 4*(3.74*10^-10)^2*6.022*10^23*101325*sqrt(pi/(8.314*298*.028))
[23:27:54] <maddash> what's the result?
[23:28:54] <maddash> bc -l keeps giving me 6749066828.73557988691317375000
[23:29:04] <maddash> but the correct number is 7265128940.978957696217764445328192
[23:29:08] <maddash> wtf?
[23:30:20] <archivist> 7265129130.73 in open office
[23:31:01] <archivist> floats are evil if used incorrectly
[23:31:16] <maddash> wtf wtf
[23:31:24] <maddash> the problem's with scale
[23:31:40] <maddash> bc -l, bydefault has scale=20
[23:31:45] <archivist> like I said floats are evil
[23:32:13] <maddash> know any replacements for bc?
[23:32:19] <maddash> cli replacements*