#emc | Logs for 2006-10-07

[01:00:31] <jmkasunich> oh joy
[01:01:01] <jmkasunich> my 74 year old mother who lives 100 miles away is having modem problems with her XP box
[01:03:09] <A-L-P-H-A> jmkasunich... set her up with bandwidth light?
[01:03:20] <jmkasunich> ?
[01:03:23] <A-L-P-H-A> broadband light
[01:03:37] <jmkasunich> she doesn't need broadband
[01:03:40] <A-L-P-H-A> sorry...
[01:03:42] <jmkasunich> she just needs a modem
[01:03:52] <jmkasunich> the description sounds like the modem hardware died
[01:04:32] <jmkasunich> that means I get to buy a modem at the local store (easy) then drive 100 miles to install it, fight with (*)%(&#()*& windoze XP to make it work, then drive 100 miles back
[01:04:54] <A-L-P-H-A> jmkasunich, skin linux for her to make it look like XP?
[01:05:15] <jmkasunich> nah
[01:06:14] <jmkasunich> when its doze, there is at least a chance that her ISP's help desk etc, can help her
[01:06:35] <jmkasunich> in this case, they took the first three calls, only after they deduced that it was at her end did she call me
[01:07:01] <A-L-P-H-A> oh... well... 100 miles... is what? almost a 2 hr drive? :(
[01:07:02] <A-L-P-H-A> shibby
[01:07:17] <A-L-P-H-A> hmm... do I want to order a pizza?
[01:07:22] <A-L-P-H-A> :}
[01:12:33] <A-L-P-H-A> PIZZA!
[01:12:37] <A-L-P-H-A> well in 15 minutes
[01:48:40] <A-L-P-H-A> ahhhhhhh. yum.
[01:48:42] <A-L-P-H-A> :D
[01:49:01] <A-L-P-H-A> jmkasunich, watch when you go there, it's a power cable was loose.
[01:49:26] <A-L-P-H-A> internal winmodem card... those are pretty faultless.
[01:49:45] <jmkasunich> she called back a little while ago - if she takes the phone off the hook while it dials, it can connect
[01:49:49] <jmkasunich> but only at around 26K
[01:49:54] <jmkasunich> weird
[01:50:17] <A-L-P-H-A> hmm... odd. but seriously? how'd she figure that out?
[01:50:28] <jmkasunich> trial and error
[01:50:29] <A-L-P-H-A> like... I wouldn't have troubled shooted with the phone being on/off hoke.
[01:50:41] <A-L-P-H-A> that's like trying everything and the kitchen sink.
[01:50:48] <jmkasunich> I think she wanted to hear if it was dialing so she picked it up to listen
[01:50:55] <jmkasunich> and was surprized when it connected
[01:51:02] <A-L-P-H-A> how old is the modem?
[01:51:12] <jmkasunich> same as the puter, about 3 years
[01:52:16] <A-L-P-H-A> odd.
[01:53:21] <A-L-P-H-A> this south park is halarious.
[01:53:27] <A-L-P-H-A> they're TRYING to loose at baseball, cause it's boring...
[01:54:00] <A-L-P-H-A> so south park vs denver... hahaha... and denver is kicking their ass at loosing...
[02:35:46] <A-L-P-H-A> how do you tell if you're climb milling, or normal milling?
[02:39:49] <A-L-P-H-A> if I have a cube mounted in a vise and wanted to mill it into a cylinder, what direction would be which?
[02:41:48] <jmkasunich> if the cutter is turning clockwise, then milling clockwise is climb milling, and milling counterclockwise is normal milling
[02:42:02] <A-L-P-H-A> jmkasunich. thanks.
[02:46:10] <A-L-P-H-A> jmkasunich that's the top down view of things, right?
[02:46:18] <A-L-P-H-A> jmkasunich, like viewing the mill from above.
[02:46:22] <jmkasunich> yeah
[02:46:28] <A-L-P-H-A> cool
[02:46:40] <jmkasunich> think of the cutter as a little wheel running on the part
[02:46:51] <jmkasunich> if its trying to "roll" forward, then you are climb milling
[02:47:06] <jmkasunich> if its turning against the direction of travel, thats normal milling
[02:47:26] <A-L-P-H-A> and normal milling is what is good for a not as ridgid machine.
[02:47:30] <A-L-P-H-A> if I remember correctly.
[02:47:39] <jmkasunich> its called climb milling because if things aren't clamped tight, the cutter can dig in and pull the work into itself
[02:47:53] <A-L-P-H-A> oh.
[02:47:57] <jmkasunich> yes, normal for less rigid machines
[02:47:59] <A-L-P-H-A> now that makes sense to me
[02:58:06] <A-L-P-H-A> logger_aj, search
[02:58:06] <A-L-P-H-A> Try this address for searching the logs:
[03:21:27] <CIA-8> 03jepler 07HEAD * 10emc2/tests/overrun/ (README test.hal): a test to make sure the overrun detection in 'runtest' works
[03:23:15] <CIA-8> 03jepler 07HEAD * 10emc2/scripts/runtests: detect overruns and re-run the test up to 10 times
[03:23:45] <jmkasunich> do you restart realtime each time?
[03:24:30] <CIA-8> 03jepler 07HEAD * 10emc2/src/rtapi/sim_rtapi_app.cc: don't write things to stdout
[03:24:34] <jepler> jmkasunich: yes
[03:24:38] <jmkasunich> ok
[03:24:42] <jmkasunich> just checkin ;-)
[03:24:56] <jepler> jmkasunich: each test is run with 'halrun' which does a realtime start & stop
[03:25:15] <jmkasunich> I see
[03:27:00] <jepler> goodnight
[03:28:00] <jmkasunich> goodnight
[03:52:24] <A-L-P-H-A> what's the command symbol in emc gcode?
[03:52:30] <A-L-P-H-A> //
[03:52:33] <A-L-P-H-A> #
[03:52:34] <A-L-P-H-A> ?
[04:13:44] <CIA-8> 03jmkasunich 07HEAD * 10emc2/src/rtapi/rtapi_math_i386.h: make HUGE_VAL be available to RT code
[05:23:13] <CIA-8> 03jmkasunich 07HEAD * 10emc2/src/emc/motion/ (command.c control.c motion.c motion.h): Realtime part of leadscrew compensation code. UNTESTED.
[05:23:24] <jmkasunich> alex_joni: have fun!
[05:23:27] <jmkasunich> bedtime for me
[05:30:47] <A-L-P-H-A> am I missing entire conversations?
[07:15:39] <K`zan> Night all
[07:15:42] <A-L-P-H-A> holy shit. http://www.break.com/index/biker_outruns_cops_on_highway.html
[07:15:45] <A-L-P-H-A> K`zan, don't go
[07:15:46] <A-L-P-H-A> :)
[07:15:47] <A-L-P-H-A> stay!
[07:15:51] <A-L-P-H-A> show more photos
[08:36:37] <Lerneaen_Hydra> 'lo
[08:38:09] <anonimasu> morning
[08:38:14] <anonimasu> command?
[08:38:19] <anonimasu> what's up?
[08:42:37] <anonimasu> bbl
[08:53:13] <A-L-P-H-A> sure.
[08:53:17] <A-L-P-H-A> say hi, and then ditch us
[10:10:47] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/task/taskintf.cc: abort start if there is even one axis busted while iniAxis. seems rather strange to continue when some axes aren't ok..
[10:29:20] <anonimasu> hehe
[10:31:20] <anonimasu> anyone here that can help me get started?
[11:09:01] <Lerneaen_Hydra> anonimasu: get started with what?
[11:52:32] <anonimasu> nml
[11:55:00] <anonimasu> :)
[11:57:06] <Lerneaen_Hydra> where have I heard that before...
[11:59:19] <Lerneaen_Hydra> nml that is
[11:59:41] <Lerneaen_Hydra> oh wasn't that some internal emc comm. protocol/bus?
[12:00:59] <anonimasu> yeah
[12:00:59] <anonimasu> :)
[12:04:49] <anonimasu> well, I'll be back later
[12:04:55] <anonimasu> going to go to work and make some parts
[12:05:51] <Lerneaen_Hydra> what is the goal of emc? is it to be a commercial-grade cnc-controller replacement?
[12:08:37] <jepler> A-L-P-H-A: I'm not sure what you mean by "command symbol". The documentation for our g-code dialect is here: http://linuxcnc.org/docs/html/gcode/main/index.html
[12:13:39] <CIA-8> 03jepler 07HEAD * 10emc2/bin/.cvsignore: ignore new programs
[12:14:45] <CIA-8> 03jepler 07HEAD * 10emc2/tests/overrun/xfail: this test is intended to fail
[12:16:31] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/halmodule.cc: fix format string warning
[12:17:42] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/hal.h: Get the desired widths for hal types, even on machines where sizeof(long)==8
[12:19:19] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/components/ (sampler_usr.c streamer.h): use hal typedefs in shmem_data_t
[12:25:54] <anonimasu> Lerneaen_Hydra: well, yeah ;)
[12:26:16] <anonimasu> bbl
[12:26:21] <Lerneaen_Hydra> ok, nice
[12:26:26] <Lerneaen_Hydra> bye
[12:28:41] <anonimasu> im really pondering about the plc code :)
[12:28:50] <anonimasu> * anonimasu is making a mount for the panel
[12:40:50] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/ini/iniaxis.cc: (log message trimmed)
[12:40:50] <CIA-8> changed the way the compensation file is interpreted:
[12:40:50] <CIA-8> - by default a file (COMP_FILE) with
[12:40:50] <CIA-8> nominal, forward and reverse triplets is assumed (old behaviour)
[12:40:50] <CIA-8> - added an ini entry called COMP_FILE_TYPE which allows to specify the type of
[12:40:50] <CIA-8> information in the file. for now type = 0 is the old behaviour
[12:40:52] <CIA-8> type != 0 means nominal, forward_trim and reverse_trim triplets are in the file
[12:40:56] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/motion/ (usrmotintf.cc usrmotintf.h): (log message trimmed)
[12:40:58] <CIA-8> changed the way the compensation file is interpreted:
[12:41:00] <CIA-8> - by default a file (COMP_FILE) with
[12:41:02] <CIA-8> nominal, forward and reverse triplets is assumed (old behaviour)
[12:41:04] <CIA-8> - added an ini entry called COMP_FILE_TYPE which allows to specify the type of
[12:41:06] <CIA-8> information in the file. for now type = 0 is the old behaviour
[12:41:10] <CIA-8> type != 0 means nominal, forward_trim and reverse_trim triplets are in the file
[12:41:12] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/nml_intf/ (emc.cc emc.hh): (log message trimmed)
[12:41:15] <CIA-8> changed the way the compensation file is interpreted:
[12:41:16] <CIA-8> - by default a file (COMP_FILE) with
[12:41:18] <CIA-8> nominal, forward and reverse triplets is assumed (old behaviour)
[12:41:20] <CIA-8> - added an ini entry called COMP_FILE_TYPE which allows to specify the type of
[12:41:24] <CIA-8> information in the file. for now type = 0 is the old behaviour
[12:41:26] <CIA-8> type != 0 means nominal, forward_trim and reverse_trim triplets are in the file
[12:41:28] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/task/ (emctaskmain.cc taskintf.cc): (log message trimmed)
[12:41:30] <CIA-8> changed the way the compensation file is interpreted:
[12:41:32] <CIA-8> - by default a file (COMP_FILE) with
[12:41:34] <CIA-8> nominal, forward and reverse triplets is assumed (old behaviour)
[12:41:38] <CIA-8> - added an ini entry called COMP_FILE_TYPE which allows to specify the type of
[12:41:40] <CIA-8> information in the file. for now type = 0 is the old behaviour
[12:41:42] <CIA-8> type != 0 means nominal, forward_trim and reverse_trim triplets are in the file
[12:41:44] <CIA-8> 03alex_joni 07HEAD * 10emc2/src/emc/usr_intf/ (emcsh.cc halui.cc usrmot.c): (log message trimmed)
[12:41:46] <CIA-8> changed the way the compensation file is interpreted:
[12:41:48] <CIA-8> - by default a file (COMP_FILE) with
[12:41:54] <CIA-8> nominal, forward and reverse triplets is assumed (old behaviour)
[12:41:56] <CIA-8> - added an ini entry called COMP_FILE_TYPE which allows to specify the type of
[12:41:58] <CIA-8> (2 lines omitted)
[12:48:01] <alex_joni> bbl..
[12:48:14] <SWPadnos> thanks - I was just catching up reading the commit messages ;)
[13:23:40] <jepler> http://www.google.com/codesearch?q=tpRunCycle
[13:25:59] <jepler> (just saying that the new google code search has some copoes of emc in it, nothing more than that)
[13:26:11] <jepler> copies
[13:37:43] <SWPadnos> what's also interesting is that there's a non-emc project that seems to be using some emc code
[13:38:09] <SWPadnos> or maybe it's just a different application of emc, since it's also from NIST
[13:46:31] <fenn> omac is very similar to emc
[13:48:59] <fenn> actually, it's more like how i would have wished emc were
[13:58:55] <jepler> whee my laptop's reinstalled with dapper
[14:04:14] <Lerneaen_Hydra> woo
[14:20:31] <skunkworks> logger_aj: bookmark
[14:20:31] <skunkworks> I'm feeling lazy .. but here's the log anyways:
[14:22:48] <jepler> 05:09:38 <CIA-8> alex_joni HEAD * emc2/src/emc/task/taskintf.cc: abort start if there is even one axis busted while iniAxis. seems rather strange to continue when some axes aren't ok..
[14:22:54] <jepler> this appears to have broken emc2 badly
[14:23:26] <jepler> axis 3..7 fail to init on sim/{axis,mini,tkemc}.ini
[14:32:47] <jmkasunich> he broke it!
[14:32:47] <jmkasunich> wasn't me, no sir, not me at all
[14:32:51] <jmkasunich> (I hope not anyway)
[14:34:31] <jepler> no, I'm pretty sure it was that change by alex
[14:34:37] <jepler> I appreciate his good intentions, though
[14:34:39] <jmkasunich> I know
[14:34:54] <jmkasunich> I'm not entirely awake yet, and I was rambling
[14:35:33] <jepler> does emcAxisInit() return 0 for failure? Or 0 for success and -1 for failure?
[14:35:46] <jmkasunich> dunno
[14:35:59] <jepler> I mean, I'm pretty sure it returns 0 for success
[14:36:33] <jepler> I'll be back in a few hours .. hope someone sorts it out by then
[14:36:57] <jepler> hum, 'man -Thtml' doesn't work on ubuntu .. I hadn't realized
[14:40:24] <jepler> er, I guess the bits for '-Thtml' aren't installed by default?
[14:46:58] <NickServ> This nickname is owned by someone else
[14:46:58] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[14:48:11] <anonimasu> ^_^
[14:48:42] <Lerneaen_Hydra> oh btw, what happened after I left yesterday wrt the troll?
[14:49:01] <jepler> I dunno, I /ignore'd him
[14:49:17] <jmkasunich> strange - it returns 0 for an illegal axis, which would seem to be an error to me
[14:49:31] <jmkasunich> he went away after a while
[14:49:46] <jmkasunich> btw, I don't think he was a troll, I think he was a kook
[14:50:48] <jmkasunich> trolls say things they know are stupid to get a reaction
[14:50:48] <jmkasunich> kooks actually believe the stupid tnings they are saying
[14:51:01] <jepler> either one is best ignored
[14:51:04] <anonimasu> he always trolls everywhere..
[14:51:14] <jepler> bbl
[14:51:17] <anonimasu> ^_^
[14:59:35] <Lerneaen_Hydra> jmkasunich: point taken
[15:00:19] <Lerneaen_Hydra> kook/evangelianist
[15:11:05] <NickServ> This nickname is owned by someone else
[15:11:05] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[15:19:09] <NickServ> This nickname is owned by someone else
[15:19:09] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[15:25:02] <skunkworks> jmkasunich: I have some 70uf 250v caps.. If I use 3 I will put them on the back of the board.. Does that make sense?
[15:25:20] <jmkasunich> why on the back?
[15:25:41] <skunkworks> they are coaxial - about 3 inches long.
[15:25:46] <skunkworks> ok 2 inches
[15:25:53] <jmkasunich> oh :-(
[15:26:05] <jmkasunich> axial lead caps are fairly high inductance
[15:26:15] <skunkworks> I just can't win :)
[15:26:25] <jmkasunich> what DC bus voltage are you planning to use?
[15:26:26] <skunkworks> I might have to buy a few huh
[15:27:17] <skunkworks> 200 would be the max I think. But I would be doing a lot of testing before I got there with some smaller servos.
[15:28:11] <jmkasunich> 200V actual voltage?
[15:28:21] <skunkworks> dc yes.
[15:28:25] <jmkasunich> rectified AC line is about 180, is that what you are thinking of?
[15:28:38] <skunkworks> yes
[15:29:10] <skunkworks> but we have a bunch of variacs for testing. so I can see what I can get to work.
[15:29:24] <jmkasunich> what is the rated motor voltage?
[15:29:34] <jmkasunich> (or don't you have motors picked out yet?)
[15:30:39] <skunkworks> Well the ones that would be really nice are 200v peak
[15:31:05] <skunkworks> but we have a selection of brushed servos also. :)
[15:31:25] <jmkasunich> with a PWM drive, DC bus caps are almost as important as the power semiconductors
[15:31:44] <jmkasunich> you don't want to leave them as an afterthought
[15:31:46] <skunkworks> (I don't know how far I want to take this... It is more of learning experience.)
[15:32:19] <jmkasunich> do you follow geckos at all.... are you aware of the "add an external 470uF cap" thing?
[15:32:20] <skunkworks> Point taken. I take minimalness to an extreme some times.
[15:33:49] <skunkworks> I have always had access to junk with 10000uf and larger can caps. I always used them for my dc power supplies.
[15:34:23] <jmkasunich> if you have large cans rated at 200V or more you can use them
[15:34:49] <jmkasunich> the important part is to have low inductance between the caps and the totem-poles
[15:35:40] <jmkasunich> which comes down to a mechanical design thing
[15:35:57] <jmkasunich> you have fets that need heatsinks, and you have a big honkin can cap
[15:36:06] <jmkasunich> and all of that stuff needs to be attached to your PC board
[15:36:12] <skunkworks> yes - I might have to buy eagle.. The 4.9Xwhatever board side.
[15:36:36] <jmkasunich> I have no idea how you intent to mount this board, or the heatsinks
[15:36:39] <skunkworks> board size is starting to be a limit.
[15:37:04] <jmkasunich> one possibility would be to put a pair of holes in the board and mount it right to the top of the can cap (screw terminal can, right?)
[15:37:17] <skunkworks> That is one of the things I usually don't worry about.
[15:37:31] <skunkworks> I wasn't sure if you would cringe at :)
[15:37:39] <skunkworks> that
[15:37:50] <jmkasunich> at what?
[15:37:59] <skunkworks> right scew terminal caps.
[15:38:12] <skunkworks> at usung the cap as part of the mounting :)
[15:38:19] <jmkasunich> all the stuff I do at work uses can caps
[15:38:36] <skunkworks> Cool. I know I have some really nice 1000uf 250 cans
[15:39:05] <skunkworks> or maybe 2200 I would have to look. Came out of some parradjust controls/
[15:39:53] <jmkasunich> what sustained/average output current do you expect to be running?
[15:40:15] <skunkworks> planning for under 20 with these componants.
[15:40:30] <jmkasunich> the desired current and voltage ratings of the drive control just about everything in the design
[15:40:30] <anonimasu> :)
[15:41:07] <skunkworks> well with these mosfets - It is the limiting factor.
[15:41:17] <jmkasunich> when you are running at 50:50 pwm duty cycle, the RMS ripple current in the capacitor is equal to half the motor current
[15:41:51] <jmkasunich> and thats only the PWM frequency ripple current, I'm not counting the 120Hz ripple current from the input diodes
[15:42:00] <jmkasunich> hence the need for a beefy cap
[15:44:01] <skunkworks> going to have to run over to the parts warehouse (polebuilding) :)
[15:46:40] <skunkworks> Thanks again. (also for putting up with me.)
[16:13:17] <anonimasu> :)
[16:56:09] <CIA-8> 03jmkasunich 07HEAD * 10emc2/src/emc/task/taskintf.cc: only call iniAxis() for axes that are configured
[16:57:00] <cradek> yay a jmk commit in task!!
[16:57:13] <jmkasunich> amazing isn't it
[16:57:20] <cradek> welcome to the brave new world
[17:04:58] <anonimasu> what's new about it?
[17:05:41] <jmkasunich> I normally don't work on 'task' or the other userspace and NML related parts of EMC, I stick to the RT code in the motion controller, (and to HAL stuff of course)
[17:05:51] <anonimasu> ah
[17:06:11] <anonimasu> jmkasunich: can you give me some pointer to where to start if I'd like to write something to talk to my plc
[17:06:23] <anonimasu> err start looking
[17:06:37] <jmkasunich> what kind of PLC, and what kind of conversation do you want to have with it?
[17:06:49] <anonimasu> a meaningful one..
[17:06:59] <jmkasunich> not helpfull
[17:07:02] <anonimasu> just kidding
[17:07:20] <anonimasu> Im going to use it to turn on coolant.
[17:07:27] <jmkasunich> do you want to just exchange some I/O bits, do you want to edit its ladder logic, or?
[17:07:27] <anonimasu> err coolant and spindle..
[17:07:47] <jmkasunich> you have an actual physical PLC installed already?
[17:07:51] <anonimasu> and use the buttons on it for start/Stop/feedhold..
[17:08:00] <anonimasu> I've got a spare off work.
[17:08:10] <anonimasu> yeah
[17:08:35] <jmkasunich> what are you using for EMC I/O? parport and step/dir, or something more complex?
[17:08:47] <anonimasu> I dont have any io to emc yet..
[17:08:50] <anonimasu> that's what Im looking at
[17:08:54] <anonimasu> yeah..
[17:08:56] <anonimasu> a usc..
[17:08:58] <jmkasunich> not even for the motors?
[17:09:04] <jmkasunich> oh, ok
[17:09:23] <jmkasunich> USC has very limited spare I/O I think
[17:09:26] <anonimasu> yeah..
[17:09:31] <anonimasu> I were thinking of using rs232..
[17:09:34] <anonimasu> rs232 > nml
[17:09:49] <anonimasu> that way I can write stuff to do online programming if I'd like..
[17:10:20] <jmkasunich> personally there's no way I'd use an external PLC for anything, I'd use classicladder in the EMC PC and figure out some way to get the number of physical I/O bits I need
[17:10:40] <jmkasunich> extra parport, 8255 card, something like that
[17:10:43] <anonimasu> jmkasunich: classicladder sucks.. really
[17:10:53] <anonimasu> :)
[17:11:04] <jmkasunich> well, if you are used to real PLCs, it probably does
[17:11:06] <jmkasunich> I'm not
[17:11:09] <anonimasu> :)
[17:11:19] <anonimasu> they are programmed in something c-ish
[17:11:19] <jmkasunich> I know less than zero about talking to real PLCs of any kind
[17:11:22] <anonimasu> or well even c..
[17:11:37] <anonimasu> and they do multitasking.. :)
[17:12:22] <jmkasunich> when does it stop being a PLC and start being "a processor with I/O"
[17:12:32] <anonimasu> I have no idea..
[17:12:50] <jmkasunich> to me a defining feature of a PLC is that you program it with ladder or some other "PLC" language, _not_ with C
[17:13:22] <anonimasu> IEC 61131-3
[17:13:25] <anonimasu> is the language..
[17:14:07] <jmkasunich> well anyway, there's not much I can offer as suggestions
[17:16:23] <anonimasu> hm.. that's too bad.
[17:17:38] <anonimasu> so you really have no idea about anything like that at all?
[17:18:08] <jmkasunich> no
[17:18:21] <jmkasunich> I don't work with PLCs
[17:18:44] <jmkasunich> I'm certainly not about to buy one for my hobby stuff, CL may suck but its free
[17:19:21] <jmkasunich> and commercial PLCs tend to be full of undocumented proprietary protocols rather than open
[17:19:21] <anonimasu> yeah.. I wouldnt either..
[17:19:39] <anonimasu> jmkasunich: canopen?
[17:20:22] <anonimasu> not that it matters though, the only propietary protocol theese use are the protocol they use for bus communication if you need high speed communication between modules..
[17:20:31] <jmkasunich> remember, my main interest is low level realtime control, not communication protocols
[17:20:56] <jmkasunich> CAN, modbus, I can't get interested in that kind of stuff
[17:21:04] <anonimasu> do you remember what I did ask?
[17:21:26] <jmkasunich> "jmkasunich: can you give me some pointer to where to start if I'd like to write something to talk to my plc"
[17:21:35] <anonimasu> yeah.. nml/hal/?
[17:21:51] <jmkasunich> hal is for stuff that can be represented as simple signals
[17:22:11] <anonimasu> so, for the things I'd like to do I'd have to use nml..
[17:22:24] <jmkasunich> if you have some PLC I/O module that has N analog inputs, you could concievable make a HAL driver for it
[17:22:35] <jmkasunich> but if you want to manipulate ladders, etc, thats not HAL
[17:22:49] <anonimasu> yeah.. I know
[17:22:52] <jmkasunich> I don't know much at all about NML, and I'm happier that way, so I can't help there
[17:25:51] <anonimasu> jmkasunich: ideally I'd like to have my stuff linked to HAL like the toolchange stuff..
[17:26:16] <anonimasu> and use nml for conveying other stuff to and from the plc..
[17:26:48] <anonimasu> though I dont know anything about hal or nml ;)
[17:40:13] <Lerneaen_Hydra> jmkasunich: that bit about commercial PLCs and proprietary comm channels is very, very true (and the one's I've played with tended to suck too)
[17:53:50] <A-L-P-H-A> jepler, it was supposed to be 'comment' not 'command'
[17:54:26] <fenn> comment is ( )
[17:55:50] <A-L-P-H-A> with in a emc2 gcode file?
[17:56:07] <fenn> yes
[17:56:11] <A-L-P-H-A> fenn, thanks.
[18:00:06] <phil_> hi guys
[18:00:10] <anonimasu> hello
[18:00:31] <phil_> i have a little trouble with EM2 & Gcode ..
[18:01:24] <phil_> i obtain a windows error : "join x following error" .. what does it mean ? please ?
[18:01:24] <A-L-P-H-A> hello
[18:01:48] <phil_> hi Alpha & anonimasu
[18:01:53] <A-L-P-H-A> physical location of motor is not moving as fast as emc wnats too...
[18:02:24] <A-L-P-H-A> I had that... and that was my understanding that my ini file was not setup properly.
[18:02:33] <phil_> not sure to right understand ! .. i dont have any encoder ..
[18:03:02] <phil_> wich parameter ?
[18:03:34] <A-L-P-H-A> umm... I don't know how I fixed it... I just used the collective brain in here to solve my setup issue.
[18:04:08] <anonimasu> I think it's "ferror"
[18:04:24] <anonimasu> and you need to check out what your max speed for your axes are
[18:04:33] <phil_> the next message now is " need to be enabled, in coord mode for linear move" ...?
[18:05:09] <phil_> u mean a speed in file is greater than the max speed in the ini file ..,
[18:05:18] <A-L-P-H-A> I just used steppers_inch.ini file, and modified it to my needs... I used www.pastebin.ca to share my ini file with the people in here to help diagnose and correctly setup. Give a detailed explaination of your system setup, and what you want your machine to do (speeds/accelerations/resolutions)
[18:05:48] <fenn> " need to be enabled, in coord mode for linear move" means that you need to set "machine on"
[18:05:57] <fenn> press f2
[18:06:27] <phil_> ok for F2 ;)
[18:06:37] <phil_> u speak french ?
[18:06:47] <fenn> no
[18:07:05] <fenn> i wanna see alpha speak french though :)
[18:07:22] <skunkworks_2> jmkasunich: still around?
[18:07:35] <skunkworks_2> http://www.electronicsam.com/images/KandT/spindle/DSC_2113.JPG
[18:07:37] <skunkworks_2> :)
[18:08:22] <anonimasu> skunkworks: cute
[18:08:26] <phil_> lol ok.. i try in english ..
[18:08:43] <fenn> * fenn secretly replaces skunkworks' capacitors with beer cans. let's see if he notices.
[18:09:12] <phil_> lool
[18:09:39] <skunkworks_2> :)
[18:10:38] <phil_> ALPHA : http://pastebin.ca/194234
[18:12:46] <phil_> this is my ini file
[18:12:50] <A-L-P-H-A> skunkworks, hey... I have the same make caps... except mine are 4600uF.
[18:13:23] <skunkworks_2> what voltage?
[18:13:53] <A-L-P-H-A> 100VDC I think.
[18:13:57] <A-L-P-H-A> all I needed...
[18:14:14] <jmkasunich> phil_: what were you doing when you got the first error "joint x following error"
[18:14:15] <skunkworks_2> cool.. I have some 50000uf but they are 10v
[18:14:53] <phil_> run a gcode file, after 1 or 2 circles ..
[18:14:57] <A-L-P-H-A> must be like a C battery
[18:15:17] <jmkasunich> phil_: I see a problem
[18:15:29] <jmkasunich> MAX_ACCELERATION = 400.0
[18:15:31] <phil_> cool !
[18:15:36] <jmkasunich> STEPGEN_MAXACCEL = 350.0
[18:15:41] <jmkasunich> should be the other way around
[18:15:57] <jmkasunich> stepgen_maxaccel needs to be _larger_ than max_acceleration
[18:16:18] <phil_> humm ok .. i change
[18:16:29] <phil_> i inverse ?
[18:16:29] <jmkasunich> MAX_VELOCITY = 30.48
[18:16:37] <jmkasunich> STEPGEN_MAXVEL = 100.0
[18:16:55] <jmkasunich> the velocity ones are OK, since STEPGEN is greater than MAX
[18:17:04] <jmkasunich> but it doesn't need to be that much greater
[18:17:24] <jmkasunich> for velocity I would use MAX_VELOCITY = 30.48 and STEPGEN_MAXVEL = 35
[18:17:46] <jmkasunich> for accel I would use MAX_ACCELERATION = 350 and STEPGEN_MAXACCEL = 400
[18:17:59] <jmkasunich> check all three axes
[18:18:07] <phil_> ok i do the change ..
[18:20:35] <A-L-P-H-A> I should relook at jepler's wizard... and add some help/comments and 'checks' to his ini generator.
[18:21:07] <phil_> is it normal in my ini file line 225 : STEPGEN_MAXVEL = 76.48559
[18:21:10] <phil_> ?
[18:21:41] <jmkasunich> everything in the ini file was either copied from the sample configuration or put there by you
[18:21:54] <phil_> not by me !...
[18:22:06] <phil_> or without know it !..
[18:22:20] <jmkasunich> then I don't know how it got there
[18:22:32] <jmkasunich> what sample did you copy?
[18:22:46] <phil_> ok i will try the mill now ..
[18:25:15] <phil_> another question please ..can i change the initial "open file" location ,.. to go directly in \home\phil ..?
[18:25:22] <jmkasunich> ?
[18:25:36] <jmkasunich> do you mean the dialog that pops up when you first start emc?
[18:26:18] <jmkasunich> the one labeled "EMC2 Configuration Selector"?
[18:27:48] <phil_> no when i want open a gcode file
[18:28:20] <phil_> yeeeaaaaahh .. cool, my work seem perfect !... :=)
[18:29:07] <skunkworks> phil_: the path to the ngc files are in the ini also.
[18:29:35] <jmkasunich> change this line: #
[18:29:35] <jmkasunich> PROGRAM_PREFIX = /usr/share/emc/ncfiles
[18:30:09] <jmkasunich> replace "/usr/share/emc/ncfiles" with "/home/phil"
[18:32:07] <skunkworks> * skunkworks wasn't going to even try to guess at the path :)
[18:32:16] <phil_> oh ok.. i searched on path .. ty
[18:33:38] <phil_> ty guys, i think all is ok now
[18:35:57] <phil_> have a good week end ;) ... ty all, bye
[19:05:04] <skunkworks> getting there http://www.electronicsam.com/images/KandT/servostart/servo2.JPG
[19:06:15] <jmkasunich> pad4 and pad5 are the cap terminals?
[19:06:53] <skunkworks> yes
[19:07:51] <jmkasunich> are you trying to do the layout with only one layer?
[19:08:21] <skunkworks> that is the plan - although - if I have to 2 sides isn't a big deal.
[19:09:07] <jmkasunich> C1 and C2 are the bootstrap caps (for the high-side gates)?
[19:09:08] <skunkworks> are you thinking I would get a lot more trace square inches if I did 2 sides
[19:09:11] <skunkworks> yes
[19:09:25] <jmkasunich> you would get lower inductance and an easier layout
[19:09:42] <skunkworks> right.
[19:09:56] <jmkasunich> I'm not sure how critical the inductance is - I try to get the very lowest possible because I work with much higher currents
[19:10:16] <jmkasunich> dunno if its absolutely needed at lower currents, or just a good thing to have
[19:10:16] <skunkworks> I understand - and your insight is very helpful
[19:11:02] <jmkasunich> if I was doing it, and I could afford to you two layers, I'd move the power traces from the left and right edge inward, and run them on either side of pad4 and pad5
[19:11:16] <jmkasunich> afford to _use_ two layers
[19:11:59] <skunkworks> not having the ability to do plated thru holes makes it a pain.. but doable.
[19:12:00] <jmkasunich> that would lower the loop area - from pad5 up to diode/fet, down to the low side diode/fet and back to pad4
[19:12:32] <skunkworks> I definatly see that.
[19:12:49] <jmkasunich> how many signals need to cross from the left side control circuitry to the right side?
[19:13:38] <jmkasunich> with two layers, they can cross on the main layer thru the gap between pad4 and 5
[19:13:55] <jmkasunich> with one layer, you'd have vertical power traces on each side, they'd have to cross with jumpers
[19:14:11] <jmkasunich> actually...
[19:14:38] <jmkasunich> what if you ran heavy jumpers from top to bottom on each side of the cap terminals
[19:15:00] <jmkasunich> from the right terminal of U$1 to the right terminal of U$3 for example
[19:16:22] <skunkworks> In adition to the traces? (are you saying to as it is?)
[19:16:44] <jmkasunich> no, once you have the jumpers the copper along the left and right edges could go away
[19:18:09] <skunkworks> Interesting. would there be a decent enough connection say with 12gauge wire connected to the board? I could use stranded and fan it out for the connection.
[19:18:28] <skunkworks> or heavier.
[19:18:34] <jmkasunich> 12AWG is good for 20A
[19:18:39] <skunkworks> yes
[19:18:56] <jmkasunich> how big is the eagle file for the board?
[19:19:26] <skunkworks> physically or file size wise?
[19:19:33] <jmkasunich> file size
[19:19:42] <jmkasunich> * jmkasunich is installing eagle
[19:21:03] <skunkworks> a little over 100k
[19:21:20] <jmkasunich> can you mail it to me?
[19:21:53] <skunkworks> could put it up on my site - would that be ok?
[19:21:58] <jmkasunich> sure
[19:22:06] <jmkasunich> its not binary is it?
[19:22:16] <skunkworks> You really don't have to do this. I am really embarased by the schematic.
[19:22:25] <jmkasunich> I know I don't have to
[19:22:26] <skunkworks> I don't know.
[19:23:30] <jmkasunich> yeah, they're binary
[19:24:53] <skunkworks> I think this is all you need
[19:24:53] <skunkworks> http://www.electronicsam.com/images/KandT/servostart/schem/
[19:25:16] <skunkworks> 2 files - the schematic and the board. seems if they are in the same directory - it sees both
[19:25:36] <skunkworks> I am doing this on xp
[19:26:09] <jmkasunich> shame on you
[19:26:11] <skunkworks> sorry
[19:26:13] <skunkworks> :)
[19:30:59] <jmkasunich> eagle takes some getting used to
[19:31:30] <skunkworks> I am getting used to eagle finally - but it is very hard to get the hang of... for me anyways.
[19:31:42] <skunkworks> Lots of help from cradek and jepler. :)
[19:32:00] <skunkworks> it isn't very intuitive to me.
[19:32:13] <fenn> * fenn hates eagle
[19:32:37] <skunkworks> * skunkworks knows fenn hates eagle.
[19:32:44] <fenn> its not as bad as it could be i guess
[19:33:34] <fenn> its probably better than geda
[19:33:45] <fenn> but certainly not good enough to warrant being the "standard" program
[19:34:03] <skunkworks> fenn: what was the program you where using at the workshop - that you where showing me.
[19:34:12] <fenn> kicad
[19:34:32] <fenn> kicad is really nice for schematics and beginner-friendly
[19:34:41] <fenn> the pcb layout tools need some work
[19:35:22] <fenn> one thing i really hate about eagle is that the footprint and the schematic symbol cant be separated
[19:35:53] <skunkworks> the only thing I have run into (and it might be something I am doing wrong) I can't do a pour in one spot for one net - then overlap another pour that is a differnt net. It doesn't separate the two like it does for pads and traces.
[19:36:44] <skunkworks> if that made sense
[19:37:06] <jmkasunich> does eagle have grids?
[19:37:11] <fenn> yes
[19:37:18] <skunkworks> type grid
[19:37:55] <skunkworks> it is kinda nice as I am used to autocad and like being able to enter commands
[19:38:09] <jmkasunich> what sickness is this? 1mm grid?
[19:38:28] <jmkasunich> everybody knows pc boards are laid out in inches
[19:38:29] <skunkworks> sorry - like I said - I am not up to speed with it.
[19:38:47] <skunkworks> I was just making it work. (it is how it was when I installed it)
[19:39:27] <skunkworks> one of those - "I need to figure out how to do that" but have not gotten around to it.
[19:39:44] <skunkworks> most of the parts in the library are in mm
[19:39:59] <fenn> whats with those smd chips in mm? whose brilliant idea was that?
[19:40:38] <skunkworks> I think it because eagle is not a native US program.
[19:40:54] <fenn> no, its because everything in japan is in mm
[19:41:32] <skunkworks> are smd actually in mm? I am used to .1 on center ic
[19:41:37] <jmkasunich> I was just being a smartass
[19:41:42] <jmkasunich> many parts are going metric
[19:41:55] <fenn> there are through hole metric DIP's right?
[19:41:55] <jmkasunich> but dips, soics, and many others are still 0.100 or 0.050 inches
[19:42:02] <jmkasunich> I don't think so
[19:42:21] <fenn> what are the little ones like you see on a computer RAM module?
[19:42:36] <jmkasunich> they're surface mount
[19:42:52] <jmkasunich> there are more packages than you can shake a stick at these days
[19:43:00] <jmkasunich> dip - 0.100 inch
[19:43:15] <jmkasunich> soic - like a dip, but half size (0.050) and surface mount
[19:43:27] <jmkasunich> then there are bunches of other ones that are stranger
[19:43:29] <skunkworks> I just used to go to radio crap and bought the dry transfer films
[19:43:40] <fenn> wonder if i should take a picture of this chip package
[19:44:09] <fenn> oh it is surface mount
[19:44:17] <fenn> it just has little legs like a dip too
[19:46:15] <jmkasunich> skunkworks: what is the terminal spacing on the can cap?
[19:46:49] <skunkworks> 27mm
[19:47:21] <jmkasunich> whats that in real units ;-)
[19:47:32] <skunkworks> 1.063
[19:47:40] <jmkasunich> 1.063
[19:47:59] <skunkworks> * skunkworks is fondling my digital caliper.
[19:48:36] <anonimasu> lol
[19:48:38] <robin_sz> evening ...
[19:48:49] <anonimasu> hey robin
[19:48:53] <skunkworks> hi
[19:50:22] <fenn> http://fenn.dyndns.org/pub/camera/DCP_0678.JPG <- little feet
[19:51:03] <Lerneaen_Hydra> fenn: you're lucky it's not bga ;)
[19:51:08] <jmkasunich> those look like J legs
[19:51:54] <jmkasunich> most commonly seen on PLCCs, usually they're square with pins on all four sides
[19:52:07] <jmkasunich> but memory packages are a species of their own
[19:52:33] <jmkasunich> so how do you actually route traces with this thing? draw->wire?
[19:52:43] <fenn> route
[19:52:59] <fenn> wait i might be thinking kicad
[19:53:26] <fenn> in the middle on the left
[19:53:57] <jmkasunich> thanks
[19:53:58] <skunkworks> yes route
[19:59:19] <skunkworks> jmkasunich: the cap distance was 1.25. I thought we where still talking about the bootstrap ones.
[19:59:33] <skunkworks> sorry
[20:00:42] <jmkasunich> no prob
[20:00:54] <jmkasunich> 1.25 is good!
[20:04:40] <jmkasunich> skunkworks: how big are the screw terminals on the cap (diameter)
[20:04:49] <jmkasunich> and how big are the screw heads?
[20:04:55] <jmkasunich> (thinking about clearance)
[20:05:28] <jmkasunich> something tells me that pad5 doesn't represent the real size of the screw
[20:06:02] <skunkworks> no - I wanted something small - just a pilot hole.
[20:06:16] <skunkworks> head size is about .375
[20:06:54] <skunkworks> biab
[20:13:05] <jmkasunich> damn - don't parts have origins in this program?
[20:18:33] <jmkasunich> why can't I place a pad on grid?
[20:18:51] <jmkasunich> pad1 stubbornly insists on moving between grid points
[20:21:16] <A-L-P-H-A> student loans... :( urk.
[20:24:10] <skunkworks> jmkasunich: same issue here. have not figured it out. I keep looking for an o-snap setting or something
[20:24:48] <Jymmm> Just curious... does anyone remember/know Pascal?
[20:25:08] <skunkworks> jmkasunich: thanks - I am going to my nieces bithday party. 9 years old. Be back in a while.
[20:28:39] <A-L-P-H-A> Jymmm, yeah.
[20:28:54] <A-L-P-H-A> delphi... but it's been a few years since I coded in it hardcore.
[20:29:08] <anonimasu> Yes
[20:29:08] <A-L-P-H-A> hardcore = in depth.
[20:29:26] <anonimasu> though it's a long time since I coded it
[20:29:33] <Jymmm> A-L-P-H-A: Do you think you remember enough to give me a hand (if needed) to convert an old .pas to .php ?
[20:29:48] <A-L-P-H-A> I could probably.
[20:30:05] <A-L-P-H-A> Jymmm, sure...
[20:30:11] <Jymmm> A-L-P-H-A ok, cool. I'll bug ya if I get into trouble.
[20:30:16] <A-L-P-H-A> 'ight
[20:30:35] <A-L-P-H-A> can't believe a d-sub 25 pin is $620CDN.
[20:30:38] <A-L-P-H-A> wtf.
[20:30:42] <anonimasu> wtf!¤#"
[20:30:45] <A-L-P-H-A> d-sub 25 punch
[20:30:47] <anonimasu> you are kidding right?
[20:30:48] <A-L-P-H-A> not pin.
[20:30:53] <anonimasu> punch?
[20:30:54] <A-L-P-H-A> No... I'm dead serious.
[20:30:57] <A-L-P-H-A> yeah...
[20:31:02] <Jymmm> A-L-P-H-A maybe $200 USD
[20:31:04] <A-L-P-H-A> I got a picture.
[20:31:05] <anonimasu> for punch machines?
[20:31:09] <anonimasu> post it :)
[20:31:14] <Jymmm> A-L-P-H-A search for greenlees
[20:31:15] <A-L-P-H-A> Jymmm, I've seen them only for $300-450USD.
[20:31:16] <Jymmm> A-L-P-H-A search for greenlee
[20:31:23] <A-L-P-H-A> Jymmm, that's what I've been looking at.
[20:31:27] <A-L-P-H-A> greenlee... @mouser
[20:31:50] <A-L-P-H-A> I could get a wire EDM guy to machine the whole SET for $600.
[20:32:05] <anonimasu> hehe
[20:32:14] <anonimasu> I'd file you a set for 600$
[20:32:19] <A-L-P-H-A> http://ec1.images-amazon.com/images/P/B000ARCCEK.01-A3NMXBC4WIK5PI._SS500_SCLZZZZZZZ_V1123556524_.jpg
[20:32:36] <anonimasu> insanity
[20:32:42] <A-L-P-H-A> I like the euro way of the "$" placement...
[20:33:41] <A-L-P-H-A> this place is local... http://www.e-sonic.com/acc/products.aspx?command=detail&partID=234&partIDExt=176 and they don't have the 25pin listed...
[20:33:48] <A-L-P-H-A> but they do have it... or will order it fo rme.
[20:33:52] <A-L-P-H-A> freak'n INSANE.
[20:34:05] <anonimasu> A-L-P-H-A do you need a punch ?
[20:34:11] <A-L-P-H-A> the bearing must be made of gold races or something
[20:34:15] <anonimasu> lol
[20:34:32] <A-L-P-H-A> anonimasu, well... I don't NEED one... but would like one. I have the dims... sec.
[20:34:59] <A-L-P-H-A> www.l-com.com/multimedia/tips/tip_greenlee_punch.pdf
[20:34:59] <anonimasu> hm, cant you have somone machine one for you?
[20:35:22] <A-L-P-H-A> use a 1/16" end mill to mill out the thing.
[20:36:04] <anonimasu> yeah
[20:36:05] <A-L-P-H-A> I just don't know how to make the male end... with the angles in... I can do the female plate super easy
[20:36:31] <A-L-P-H-A> 1/8" steel plate or C channel, mill it out with a 1/16" endmill...
[20:37:26] <A-L-P-H-A> the female end, the profile shape is what I have a little issue with... getting a 10° angle.
[20:37:54] <anonimasu> hm, triangle?
[20:38:24] <anonimasu> ah..
[20:39:41] <Jymmm> http://www.mygreenlee.com/products/det.cfm?id=4143&upc=34420
[20:41:33] <A-L-P-H-A> Jymmm, yeah. exact same thing.
[20:41:37] <anonimasu> you should be able to mill it...
[20:41:52] <A-L-P-H-A> dove tail them you mean?
[20:41:56] <anonimasu> yeah..
[20:42:05] <A-L-P-H-A> yeah... I could do that.
[20:42:12] <A-L-P-H-A> close enough, is close enough. :D
[20:42:26] <anonimasu> or you can always use a file ;)
[20:43:22] <anonimasu> *yawns*
[20:46:24] <alex_joni> anonimasu: http://www.nocircles.com/index.php?boyhave=flash
[20:47:43] <anonimasu> nice
[20:47:44] <anonimasu> :D
[20:47:54] <A-L-P-H-A> I'm stuck on level 3 or 4
[20:49:06] <A-L-P-H-A> where there is a center block spinning and this other block just orbits around with the direction of the mouse...
[20:49:11] <A-L-P-H-A> the exit target is bottom right
[20:49:21] <anonimasu> yep
[20:49:23] <anonimasu> same with me now
[20:50:07] <A-L-P-H-A> don't get that one
[20:50:15] <anonimasu> ah..
[20:50:18] <anonimasu> I threw the block ;)
[20:50:27] <A-L-P-H-A> how'd you throw it
[20:50:27] <A-L-P-H-A> ?
[20:51:02] <anonimasu> ther's this edge
[20:51:04] <anonimasu> where you can click
[20:51:06] <A-L-P-H-A> I've been spinning it... clicking, evthing
[20:51:07] <anonimasu> \
[20:51:09] <anonimasu> there..
[20:51:12] <anonimasu> now you havent :D
[20:51:23] <anonimasu> yay solved it ;)
[20:51:49] <A-L-P-H-A> I don't get where to click.
[20:51:54] <A-L-P-H-A> move + click
[20:51:56] <A-L-P-H-A> whatever.
[20:51:58] <anonimasu> follow the direction of the arrrow..
[20:52:02] <anonimasu> it's a link thingie..
[20:52:07] <anonimasu> the mouse cursor changes when you hover over it..
[20:52:13] <anonimasu> theres a half circle where you can click..
[20:52:16] <anonimasu> )
[20:52:45] <anonimasu> found it?
[20:52:50] <A-L-P-H-A> no
[20:52:54] <A-L-P-H-A> I see three objects only.
[20:53:00] <A-L-P-H-A> the spinning thing, cube, arrow
[20:53:12] <anonimasu> :)
[20:53:55] <anonimasu> i'll get you a screenshot later
[20:54:34] <A-L-P-H-A> found it
[20:54:41] <anonimasu> "fuck sphericity"
[20:54:41] <A-L-P-H-A> I could see the circle on my LCD screen at all
[20:54:42] <anonimasu> ^_^
[20:54:46] <anonimasu> circle?
[20:54:52] <anonimasu> me neither..
[20:55:54] <A-L-P-H-A> weird game
[20:56:46] <anonimasu> ahha
[20:56:50] <anonimasu> err haha
[20:57:04] <anonimasu> I've gotten really far now
[20:57:06] <A-L-P-H-A> finally got it to the end.
[20:57:33] <A-L-P-H-A> err.
[20:57:36] <A-L-P-H-A> more.?
[20:57:38] <A-L-P-H-A> now I got a donut
[20:57:42] <anonimasu> hehe
[20:57:42] <anonimasu> yeah
[20:57:47] <anonimasu> did you see fuck sphericity?
[20:57:52] <A-L-P-H-A> yup
[20:58:18] <anonimasu> im at some place with a bouncing cube
[20:59:09] <A-L-P-H-A> done that
[21:00:00] <A-L-P-H-A> catching donuts
[21:00:04] <A-L-P-H-A> done
[21:00:10] <anonimasu> it says defeated
[21:00:10] <anonimasu> :D
[21:00:18] <anonimasu> thank you for playing
[21:00:18] <anonimasu> ;)
[21:00:51] <A-L-P-H-A> now I'm at the one with the > and the block bouncing up and down
[21:01:01] <anonimasu> hehe
[21:01:09] <anonimasu> that's where i were
[21:01:10] <anonimasu> :9
[21:02:30] <A-L-P-H-A> gotta drag stuff around on that one
[21:03:13] <A-L-P-H-A> screw it I got stuff to do
[21:03:58] <robin_sz> you are in a dark narrow passage
[21:04:09] <robin_sz> behind you is a mud and stone clogged cave,
[21:04:20] <robin_sz> an iron grate covers the entrance
[21:04:20] <A-L-P-H-A> pick up penis pump
[21:04:33] <robin_sz> I cant see a penis pump here
[21:04:34] <A-L-P-H-A> whack robin_sz with it
[21:04:39] <anonimasu> lit torch
[21:04:40] <A-L-P-H-A> go west
[21:04:44] <robin_sz> heh
[21:04:47] <anonimasu> A-L-P-H-A: the grue!
[21:04:53] <Jymmm> Jymmm is now known as ImInNoMoodForBS
[21:05:04] <A-L-P-H-A> I never actually played that game
[21:05:05] <robin_sz> you know, I played that on an ICL mainframe way back in ... 78, 79
[21:05:14] <A-L-P-H-A> 78... before I was born!
[21:05:18] <A-L-P-H-A> hehehehe
[21:05:24] <anonimasu> nobody ever goes around without a light grues fear it!
[21:05:40] <robin_sz> its where the code "xyzzy" coems from
[21:06:16] <A-L-P-H-A> easter egg emc2 in "G00 zyzzy"
[21:06:50] <anonimasu> never seen that
[21:07:01] <A-L-P-H-A> someone in FL had a surplus store that was selling db25 punchs for $20... I wish I was there to have bought that.
[21:07:02] <A-L-P-H-A> damn.
[21:07:05] <robin_sz> http://computing-dictionary.thefreedictionary.com/Xyzzy
[21:07:59] <anonimasu> sweet
[21:08:22] <robin_sz> Xyzzy has actually been implemented as an undocumented no-op command on several OSes; in Data General's AOS/VS, for example, it would typically respond "Nothing happens", just as ADVENT did if the magic was invoked at the wrong spot or before a player had performed the action that enabled the word. In more recent 32 bit versions, by the way, AOS/VS responds "Twice as much nothing happens".
[21:08:41] <A-L-P-H-A> wonder if I made the punch, could I sell it for $200USD?
[21:08:48] <ImInNoMoodForBS> ImInNoMoodForBS is now known as Jymmm
[21:08:54] <anonimasu> why not..
[21:09:01] <A-L-P-H-A> should I use acme screw to drive the punch?
[21:09:04] <anonimasu> though you probably would need them in steel
[21:09:16] <A-L-P-H-A> I was going to do mild steel.
[21:09:20] <A-L-P-H-A> wonder how well that'd hold up?
[21:09:21] <anonimasu> would that last?
[21:09:46] <robin_sz> nah
[21:09:48] <A-L-P-H-A> I could get tool steel... but I don't know 2 cents for hardening.
[21:09:51] <robin_sz> use silver steel
[21:09:56] <A-L-P-H-A> what's silver steel?
[21:10:09] <robin_sz> its sold in all engineering stores
[21:10:18] <robin_sz> soft state, easy to machine
[21:10:39] <A-L-P-H-A> does it harden?
[21:10:40] <robin_sz> heat it bright red and oil quench ... hard as anything
[21:11:02] <A-L-P-H-A> if I dunk it in oil, doesn' the oil catch on fire?
[21:11:07] <robin_sz> nope
[21:11:08] <anonimasu> no
[21:11:18] <anonimasu> if you have 5cc of oil yes.
[21:11:18] <A-L-P-H-A> why kind of oil? engine oil okay?
[21:11:21] <anonimasu> yes
[21:11:22] <A-L-P-H-A> why=what
[21:11:23] <robin_sz> fine
[21:11:37] <anonimasu> oil dosent burn unless you get it hot enough..
[21:11:48] <A-L-P-H-A> red hot metal doesn't catch it on fire?
[21:11:52] <anonimasu> just have plenty of oil before you quench anything in it..
[21:11:54] <robin_sz> and theres no oxygen under the surface
[21:12:11] <robin_sz> nope .. toss it in .. its fine .. done it hundreds of times
[21:12:24] <anonimasu> if you are worried do it outside..
[21:12:32] <A-L-P-H-A> hmm... worth a try... I'll see if I can any tool steel, oil hardening from my steel supplier.
[21:12:49] <anonimasu> hardening stuff is really neat
[21:12:54] <robin_sz> yeah
[21:12:55] <A-L-P-H-A> machine it first, oil harden, grind/polish to fit.
[21:13:00] <robin_sz> yep
[21:13:03] <anonimasu> though I dont have anything that hardens well
[21:13:10] <anonimasu> just some c-shaft..
[21:13:12] <robin_sz> silve steel is very common
[21:13:13] <A-L-P-H-A> I have some drill rod... "O1" steel... I believe.
[21:13:15] <anonimasu> it hardens a little bit..
[21:13:20] <Jymmm> A-L-P-H-A: Just FYI... http://en.wikipedia.org/wiki/Fire_triangle
[21:13:22] <A-L-P-H-A> but it's only 1/2" dia.
[21:13:39] <anonimasu> why dont you try hardening something=?
[21:13:54] <A-L-P-H-A> nothing to make that requires hardening.
[21:13:59] <A-L-P-H-A> besides this punch
[21:14:10] <robin_sz> well, silver steel is the way to go
[21:14:19] <anonimasu> well, if you say so..
[21:14:23] <A-L-P-H-A> robin_sz, know buy any other names?
[21:14:39] <anonimasu> I were mostly thinking about that you seem skeptical..
[21:14:40] <anonimasu> ;)
[21:15:23] <A-L-P-H-A> nono... I'd do it in the backyard, remove some leaves, and propane torch with a bucket of oil engine oil...
[21:15:43] <A-L-P-H-A> and sand near by to dose it, if it came to that.
[21:15:56] <robin_sz> # Carbon min 1.10%, max 1.20%.
[21:15:56] <robin_sz> # Silicon min 0.10%, max 0.25%.
[21:15:56] <robin_sz> # Chromium min 0.40%, max 0.50%.
[21:15:56] <robin_sz> # Sulphur max 0.035%.
[21:15:56] <robin_sz> # Manganese min 0.30%, max 0.40%.
[21:15:57] <robin_sz> # Phosphorus max 0.035%.
[21:16:34] <anonimasu> hehe
[21:16:36] <robin_sz> it needs to be a good bright red, just put some firebrick around it
[21:17:17] <A-L-P-H-A> what could I use to hold it?
[21:17:26] <Jymmm> cotton gloves
[21:17:32] <A-L-P-H-A> Jymmm, hush.
[21:17:33] <Jymmm> channel locks
[21:17:36] <anonimasu> a plier..
[21:17:37] <robin_sz> fingers ...
[21:17:48] <robin_sz> dangle it on a bit of rod
[21:17:53] <anonimasu> or something like it
[21:18:04] <A-L-P-H-A> rod sounds best...
[21:18:12] <robin_sz> stand it up by the firbrick, heat it red, grab with pliers toss into the oil
[21:18:26] <Jymmm> s/toss/place/
[21:18:32] <A-L-P-H-A> just need the both the male and female hardened.
[21:18:44] <anonimasu> if you use a plastic container dont toss..
[21:18:49] <A-L-P-H-A> I'll try it out on tuesday when supplier opens again... thanksgiving in Canada on Monday
[21:19:00] <robin_sz> well, tossing is recommended I think ...
[21:19:10] <robin_sz> or at least quench fast
[21:19:15] <anonimasu> yeah
[21:19:17] <anonimasu> that's what I do..
[21:19:21] <A-L-P-H-A> I'll probably use two soup cans, double walled.
[21:19:32] <Jymmm> coffee can
[21:19:34] <A-L-P-H-A> I'd just drop it in...
[21:19:34] <robin_sz> or one end goes hard , the other just goes soft
[21:19:38] <robin_sz> yeah
[21:19:39] <anonimasu> single walled would be just fine
[21:19:43] <robin_sz> indeed
[21:19:44] <A-L-P-H-A> yeah, coffee can > soup can > oil > part
[21:19:51] <robin_sz> its really not as dramtic as you migh tthink
[21:19:52] <Jymmm> A-L-P-H-A the oil will never get that hot
[21:19:56] <anonimasu> as soon as it goes into the oil it's cool..
[21:20:04] <robin_sz> indeed
[21:20:15] <anonimasu> thats why you should have plenty of oil..
[21:20:20] <robin_sz> if the oil goes up 2 degrees, id be suprised
[21:20:21] <anonimasu> with that I mean more then 5cc ;)
[21:20:32] <robin_sz> yeah, about a quart would be fine
[21:20:55] <Jymmm> heh... robin_sz said quart =)
[21:20:56] <anonimasu> it's not so cool as the result you get
[21:20:57] <A-L-P-H-A> Quart ~= 250ml? pint ~= 500ml?
[21:21:02] <robin_sz> I think the mericans caul it "guage plate"
[21:21:22] <anonimasu> actually there's nothing cool about it.. :D no fire or sparks or anything
[21:21:22] <robin_sz> dunno
[21:21:33] <robin_sz> sadly :)
[21:21:43] <anonimasu> I wish there were a bit of sparks atleast ;)
[21:21:48] <anonimasu> I need a edm.
[21:21:51] <anonimasu> *grins*
[21:21:54] <robin_sz> me too
[21:22:02] <A-L-P-H-A> I would so like a wire EDM...
[21:22:12] <A-L-P-H-A> Get some tool steel, and pump these things out.
[21:22:18] <anonimasu> hehe
[21:22:19] <anonimasu> yeah
[21:22:23] <anonimasu> that's be really nice
[21:22:28] <A-L-P-H-A> Sell it for $200/pop... or $500/set.
[21:23:23] <A-L-P-H-A> wonder how they got the back C channel attached to the front plate...
[21:23:46] <A-L-P-H-A> weld or brazed.
[21:24:13] <A-L-P-H-A> I wonder could you weld something that's hardened?
[21:24:17] <robin_sz> nope
[21:24:31] <robin_sz> well, beofre you harden it maybe
[21:24:41] <alex_joni> A-L-P-H-A: or treat it before welding it :)
[21:24:46] <alex_joni> heat it up..
[21:24:59] <A-L-P-H-A> if I heated it, wouldn't that change the hardening?
[21:25:04] <robin_sz> yes
[21:25:07] <Jymmm> A-L-P-H-A: bottom of page there are two around $200USD http://www.mcmaster.com/ctlg/DisplCtlgPage.asp?ReqTyp=CATALOG&CtlgPgNbr=2163&term=D-Shape+Knockout+Punches&sesnextrep=628659004530964&ScreenWidth=1920&McMMainWidth=804
[21:25:57] <robin_sz> gee $200
[21:26:16] <Jymmm> but I have nfc whay one is $400 and the others are ~$200
[21:26:17] <robin_sz> A-L-P-H-A, you know what I would do?
[21:26:24] <A-L-P-H-A> robin_sz?
[21:26:39] <robin_sz> this is to go in the back panel of a case right?
[21:26:48] <A-L-P-H-A> Jymmm, that $200 is the backplate ones... no arms for the mounting screws.
[21:27:16] <A-L-P-H-A> robin_sz... it's already made on the case... just looks like shit... I have another case to make.
[21:27:33] <robin_sz> I'd laser cut the case out in 1.5mm sheet steel, with all the holes and stuff, then fold it up and powder coat it .. :)
[21:27:51] <A-L-P-H-A> I have greenlee punchs for circles, that my dad bought when he was young... awesome tools... just lots of surface stains from being oxidized.
[21:28:07] <Jymmm> A-L-P-H-A ah
[21:28:08] <A-L-P-H-A> robin_sz... if I had a laser! sure!
[21:28:23] <robin_sz> draw the DXF and send it to a local jobshop
[21:29:00] <robin_sz> with the DXF ... $20 maybe, as its almost zero work for them
[21:29:17] <A-L-P-H-A> you're shitting me.
[21:29:28] <robin_sz> well, $30
[21:29:41] <A-L-P-H-A> I just bring them the sheet metal, and wham they cut it all out for me?
[21:29:52] <A-L-P-H-A> + dxf
[21:30:00] <robin_sz> no, just send them the dxf, they will give you back the flat piece of 1.5mm steel
[21:30:21] <robin_sz> they will put it in on some other sheet of 1.5mm they are cutting
[21:30:23] <fenn> and etch all the markings and stuff while you're at it
[21:30:26] <robin_sz> yep
[21:30:39] <robin_sz> thats what they do
[21:30:50] <A-L-P-H-A> I wonder if they'd just punch out the whole case for me.
[21:30:58] <A-L-P-H-A> with rivit holes ready made for me.
[21:30:59] <robin_sz> if you give them a paper drawing, its a expensive business, but from dxg it can be cheap
[21:31:02] <A-L-P-H-A> that'd be awesome.
[21:31:14] <robin_sz> yeah, all the holes. every last one
[21:31:25] <robin_sz> thats what we do, every day for people
[21:31:31] <A-L-P-H-A> I'll shop for a sheet metal laser place...
[21:31:44] <robin_sz> "laer jobshop" is the phrase
[21:31:59] <Jymmm> whats that online machien shop I can never remember the name of it
[21:32:12] <fenn> i think silver steel is the same as o1 steel
[21:32:19] <robin_sz> if yo make it clear they'll get a 1:1 sized dxf and you are a hobbyist not a company and you'll wait for them to nest it in with other stuff ...
[21:32:24] <Jymmm> http://www.emachineshop.com/
[21:32:29] <robin_sz> expensive ..
[21:32:32] <robin_sz> very.
[21:32:34] <A-L-P-H-A> yeah... emachineshop is super expensive
[21:33:15] <Jymmm> yeah? has anyone downloaded the sw and tried it?
[21:33:45] <Jymmm> http://www.emachineshop.com/stamping_machines/turret-punching.htm
[21:34:36] <Jymmm> just a thought is all.
[21:37:52] <Lerneaen_Hydra> 'night all
[21:52:30] <A-L-P-H-A> why are tools so god damn expensive?
[21:52:51] <alex_joni> A-L-P-H-A: cause you can't make them yourself
[21:55:32] <A-L-P-H-A> okay... I want to learn this gcode "O" code for sub routines.
[21:55:47] <A-L-P-H-A> I've currently got a php script to produce my gcode.
[21:55:50] <A-L-P-H-A> it's just a simple loop.
[21:56:28] <A-L-P-H-A> sec to put them up
[21:56:30] <A-L-P-H-A>
[21:56:33] <A-L-P-H-A> the source doesn't work yet.
[21:56:34] <A-L-P-H-A> sec
[21:58:06] <A-L-P-H-A> k, working now
[21:58:46] <A-L-P-H-A> the dims aren't exactly critical... I'm just trying to figure out how to convert this to the O code technique.
[21:58:47] <alex_joni> A-L-P-H-A: is it supposed to say: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
[21:58:50] <alex_joni> <html>
[21:58:53] <alex_joni> <head>
[21:58:55] <alex_joni> <title> New Document </title>
[21:58:58] <alex_joni> <meta name="Generator" content="EditPlus">
[21:59:00] <alex_joni> <meta name="Author" content="">
[21:59:03] <alex_joni> <meta name="Keywords" content="">
[21:59:05] <alex_joni> <meta name="Description" content="">
[21:59:08] <alex_joni> </head>??
[21:59:35] <A-L-P-H-A> yeah... it's an HTML document for display.
[21:59:58] <A-L-P-H-A> it's not the gcode source... scroll down a bit on the gcode.php page, and it'll show you the gcode below...
[22:00:10] <A-L-P-H-A> the source file, just shows you the gcode whole file.
[22:00:14] <alex_joni> A-L-P-H-A: not getting me..
[22:00:25] <alex_joni> ahh.. ok
[22:00:40] <A-L-P-H-A> see where the gcode begins?
[22:01:42] <A-L-P-H-A> it should actually be ($tooldia/2) not just $tooldia.
[22:01:47] <A-L-P-H-A> I wrote that last night, very late.
[22:11:34] <A-L-P-H-A> optimized it better. gcode2.php and gcode2source.php
[22:11:47] <A-L-P-H-A>
[22:13:42] <A-L-P-H-A> brb... linux time.
[22:24:52] <A-L-P-H-A> logger_aj: bookmark
[22:24:52] <A-L-P-H-A> I'm feeling lazy .. but here's the log anyways:
[22:34:15] <A-L-P-H-A> Anyone know of an app that will test the speed of a harddrive in linux? Looking to test the througput/bandwidth of a raid array
[22:36:46] <alex_joni> hdparm
[22:36:52] <alex_joni> hdparm -tT /dev/hda
[22:38:13] <A-L-P-H-A> yeah... I just found it.
[22:38:22] <A-L-P-H-A> wth... my raid 0 array is POOH!
[22:38:38] <A-L-P-H-A> alpha@workstation:~$ sudo /sbin/hdparm -t /dev/hda1
[22:38:38] <A-L-P-H-A> Password:
[22:38:38] <A-L-P-H-A> /dev/hda1:
[22:38:38] <A-L-P-H-A> Timing buffered disk reads: 176 MB in 3.00 seconds = 58.60 MB/sec
[22:38:38] <A-L-P-H-A> alpha@workstation:~$ sudo /sbin/hdparm -t /dev/sda1
[22:38:39] <A-L-P-H-A> /dev/sda1:
[22:38:41] <A-L-P-H-A> Timing buffered disk reads: 180 MB in 3.02 seconds = 59.67 MB/sec
[22:38:54] <A-L-P-H-A> /sda1 is my raid0. :(
[22:39:18] <A-L-P-H-A> seems like there's ZERO point of making it into a raid0.
[22:39:54] <A-L-P-H-A> oops wrong drive to test
[22:40:05] <anonimasu> lol
[22:40:09] <anonimasu> night everyone
[22:40:14] <A-L-P-H-A> good night
[22:40:23] <A-L-P-H-A> still pooh... but better... 96.58MB/sec
[23:02:26] <alex_joni> night all
[23:41:14] <A-L-P-H-A> anyone know what I need exactly to get vmware to run on my system?
[23:41:40] <A-L-P-H-A> linux (ubuntu), vmware player, an install CD for windows or whatever else OS.
[23:57:01] <A-L-P-H-A> logger_aj: search
[23:57:01] <A-L-P-H-A> Try this address for searching the logs:
[23:59:52] <skunkworks> ate way too much,
[23:59:59] <A-L-P-H-A> same here