#emc | Logs for 2007-09-13

Back
[00:41:59] <JymmmEMC> fenn: The center line, but I gotta attach the shield "somewhere"
[00:42:07] <JymmmEMC> shield/ground wire of the feedline
[00:45:31] <JymmmEMC> Maybe I'll just get some single side PCB and connect the shiled to that. Might act as a good ground plane too.
[00:47:11] <tomp2> gcode doesnt have polar motion like heidenhain , nor centers, its not lacking in emc, its missing in gcode (ec CC X1. Y1. LP R10 A45 describes a 'pole' and a ray of motion from that pole )
[00:47:22] <tomp2> eg
[00:47:58] <tomp2> polar is real handy
[00:48:13] <SWPadnos> so CC X1 Y1 describes a cylindrical coordinate centerpoint at 1,1
[00:48:15] <skinnypuppy1334> polar is handy...
[00:48:36] <tomp2> SWPadnos: yes, a pole
[00:48:38] <SWPadnos> and LP R10 A45 means move 10 units from center, at a 45 degree angle
[00:49:00] <SWPadnos> but the CC command doesn't cause motion, it's like a G5x offset (sort of)?
[00:49:32] <SWPadnos> or more like a G17/G18/G19
[00:49:34] <tomp2> SWPadnos: yes, apole is just a reference
[00:49:36] <SWPadnos> ok
[00:50:36] <tomp2> can be 3 coords XYZ
[00:50:50] <tomp2> the pole/center
[00:50:54] <skinnypuppy1334> CC command for emc yes no?
[00:50:56] <SWPadnos> is that truly polar, or cylindrical?
[00:51:10] <SWPadnos> CC probably not until the language is redefined
[00:51:38] <tomp2> i dont se any cyclinder in it, i use short 3D vctors all around a pole myself, more spherical than cylindrical\
[00:51:45] <skinnypuppy1334> Oooh
[00:51:48] <SWPadnos> C = the C axis, and RS274NGC explicitly states that whitespace is irrelevant and is removed before interpretation
[00:52:25] <SWPadnos> tomp2, ok, so can you do LP R 10 A45 B45, or would it be LP R10 A45 Z5 ?
[00:53:03] <SWPadnos> ie, are Z coordinates still used (cylindrical coordinate system), or theta/phi (3D polar)
[00:53:03] <tomp2> either: really, i think thats an example, 2 angles and a len or 2 lens and an angle
[00:53:15] <SWPadnos> ok, interesting
[00:53:42] <tomp2> you can get the user manuals at heidenhain as biggish pdfs
[00:53:45] <SWPadnos> I don't know how that that would be possible, come to think of it
[00:54:01] <tomp2> yes, they must agree
[00:54:21] <SWPadnos> well, if you leave one ordinate off, how does the control know which one you meant to have, Z or phi?
[00:54:37] <SWPadnos> normally, unspecified ordinates are left unchanged
[00:54:42] <SWPadnos> btu that's not simple with phi
[00:54:45] <SWPadnos> but
[00:55:48] <tomp2> gets book
[00:55:57] <SWPadnos> heh
[00:56:27] <SWPadnos> I don't have time for much discussion, but it sounds interesting for some kind of interpreter
[00:58:57] <tomp2> ok, format is LP (line polar) PR (polar radius) PA (polar angle IN PLANE ) book doesnt show a change in Z (or ortho axis )
[00:59:13] <skunkworks> logger_emc: bookmark
[00:59:13] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-09-13.txt
[01:00:04] <tomp2> polar arc is CP (circle polar) PA (degrees of arc ) DR ( direction, actually Drucke )
[01:00:33] <skinnypuppy1334> I have a 3" od roundstock, upright in the mill, need to faceslot at several different degrees, all going back to origin at the center of the round face.
[01:02:37] <skinnypuppy1334> Something circle center would be nice but I guess I'll be triggin it out.
[01:03:25] <skinnypuppy1334> six slots with multiple air ports in the bottoms and alignment dowels
[01:10:22] <tomp2> Q1= YurRadPlusAbit; Q2=60; Q3=6; Q4=SlotDepth; Q5=CutFeed; Q6=ThisAng; Q6=Q2;;;LZ3FMAX (clear over part); LX0Y0; CC X0Y0; LZ.05 (close) LBL1 LZ-Q4FQ5 LP PRQ1 PAQ6; LP PR0 PAQ6; Q6=Q6+Q2 (test if done else loop to LBL1);LZ3FMAX(when done)
[01:10:33] <tomp2> skinnypuppy: can you grok that?
[01:10:52] <SWPadnos> only one with APTitude could grok that, I think :)
[01:12:24] <tomp2> i think faster & picture it faster than i can xl8 into languages.. then i thought, what if he hasnt tried parametric on heid before?
[01:12:54] <tomp2> its like O code in emc
[01:13:03] <tomp2> and macroB
[01:13:23] <SWPadnos> do you know if they have any patents/copyrights on that stuff?
[01:14:43] <tomp2> no idea, the polar isnt copyrightable, it belongs to mathematics, but the LP CP PA PR might be a 'look and feel'
[01:15:30] <SWPadnos> yeah. I'd love to start thinking about "RS274-NG" (unless that exists), but I definitely don't want to get into any legal battles
[01:17:29] <tomp2> so if a language has no words, (eg chinese , no phonetics ) then we could store a symbol ( coded like unicode/ascii/ebdic ) and we'd have less legal battles and more universality ... programming in smilies :)
[01:17:42] <SWPadnos> no, that's APL ;)
[01:17:43] <tomp2> symbols and numbers, no words
[01:21:24] <tomp2> (had to look it up ) yeah, like that, but i was thinking more universal (commonly accepted) like how everyone understands the tape recorder symbols even for axis and youtube, and the little walking man on cross walks
[01:21:45] <SWPadnos> heh. I think some of those are ISO symbols :()
[01:21:47] <SWPadnos> :)
[01:22:27] <tomp2> sorry for rambling... i was just trying to catch up backscrolling thru todays comments... thanks, byebye
[01:24:32] <skinnypuppy1334> Just got back from dinner, grok ???
[01:24:50] <SWPadnos> Robert A. Heinlein, "Stranger in a Strange Land"
[01:25:07] <SWPadnos> it more or less means "to understand the nature of something fully"
[01:25:39] <fenn> * fenn just grokked a banana smoothie
[01:26:08] <skinnypuppy1334> Alright, no that was over my head
[01:26:18] <fenn> the real meaning of grok is 'drink'
[01:26:49] <skinnypuppy1334> not grok over my head, but what tomp had posted above
[01:27:06] <SWPadnos> I think he was suggesting that you use variables and functions to accomplish your task
[01:27:32] <SWPadnos> that should be pretty easy - the EMC interpreter has trig functions
[01:28:07] <skinnypuppy1334> I'm just not that familiar with creating routines and using variables
[01:28:20] <skinnypuppy1334> or at all really
[01:28:26] <SWPadnos> G1X[#1 * [COS [#2]] Y[#1 * [SIN [#2]]
[01:28:27] <Ziegler> krash... this it what you were looking for: http://images.myonlinesite.com/tobak/
[01:28:36] <Ziegler> bah
[01:28:43] <Ziegler> sorry... second day in a row (wrong window)
[01:29:05] <skinnypuppy1334> swp, that makes some sense. I'll play with that
[01:29:33] <skinnypuppy1334> just needed to see some syntax
[01:29:37] <SWPadnos> all you have to know is the formula for where you want it to go, and then muddle through with how to pass parameters and stuff :)
[01:30:09] <SWPadnos> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Oword
[01:30:33] <skinnypuppy1334> Alright seems easy enough to incremental plunge, path, and repeat like that
[01:30:39] <SWPadnos> yep
[01:30:54] <SWPadnos> and if you need slots at equal angles, you can step the angle in a loop as well
[01:32:00] <skinnypuppy1334> I'll study and play with the examples you sent, thank you. Just trying to be productive until all my hardware arives
[01:32:51] <SWPadnos> oh. that sounds like a good idea
[05:18:30] <skinnypuppy1334> I found a Gcam bug that locks my system ... ctl alt backspace doesn't even work ... hehehe
[05:19:30] <SWPadnos> that's possibly only X being locked up. you may be able to ssh into it and kill X
[05:20:36] <skinnypuppy1334> May be, it's when opening a file, click to expand the folder and then click back on the window behind it and freeze
[05:20:59] <toast> ctrl+alt+2?
[05:21:02] <skinnypuppy1334> ctl alt f2 everything locks
[05:21:05] <toast> snap
[05:21:20] <toast> ssh that thing dude.
[05:21:22] <toast> like the man said!
[05:21:29] <skinnypuppy1334> that's why I've been on off on off
[05:22:11] <skinnypuppy1334> I'm running local, is gcam using ssh?
[05:22:20] <toast> huh
[05:22:23] <SWPadnos> no, you can log in remotely
[05:22:28] <SWPadnos> assuming you have sshd installed
[05:22:48] <skinnypuppy1334> duh sorry wasn't looking at if from that angle my fault
[05:22:49] <toast> i wish ubuntu came with sshd on by default and just root logins disabled
[05:23:00] <toast> REQUIEM FOR A DEFAULT
[05:23:09] <SWPadnos> if it only came with sshd installed bt default :/
[05:23:12] <SWPadnos> by
[05:23:48] <toast> if only.
[05:33:22] <maddash> freakin' pem certificate
[05:33:42] <maddash> I can't get wget/firefox/anything to extract it from a site
[09:25:18] <robboplus_> robboplus_ is now known as robboplus
[10:46:24] <kollabierer> hi
[10:47:00] <kollabierer> which cam/cad program do you preffere to interact with emc ?
[11:08:19] <lars`w0rk> hi all
[11:10:50] <alex_joni> hi
[11:11:02] <alex_joni> kollabierer: there's a list with cad/cam packages on the wiki
[11:11:35] <alex_joni> kollabierer: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Cam
[11:11:44] <kollabierer> I#ve seen the list, that's why I asked which you prefere
[11:12:08] <alex_joni> depends on the job
[11:12:27] <alex_joni> for PCB milling, quite a few people use eagle + some ULP to generate g-code
[11:12:39] <alex_joni> others use synergy for up to 5-axes machining
[11:15:27] <kollabierer> is there a cam program that generates g-code, EMC understands out of the box ?
[11:16:11] <alex_joni> usually the CAM programs have so-called post-processors
[11:16:32] <alex_joni> these are simple programs/scripts which parse the output from the CAM program and convert it to g-code
[11:17:05] <alex_joni> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Cam_Post <- one post-processor specifically for emc2
[11:17:36] <alex_joni> most of the time you can look in the cam program for fanuc or standard rs274 outputs, it's most likely it will work with minor corrections
[11:23:43] <kollabierer> please, don't say "standard RS-274" !
[11:23:55] <kollabierer> because EMC does NOT understand RS-274 !
[11:24:05] <kollabierer> it just understands a simple dialekt
[11:24:47] <kollabierer> I had to learn that lesson the hard way :(
[11:30:31] <alex_joni> actually it does understand the rs274 standard
[11:30:42] <alex_joni> the trouble is that rs274 standard is from 1974 or so
[11:30:56] <alex_joni> and everything after that isn't standard any more
[11:31:21] <alex_joni> most of the manufacturers use their own extensions where the original standard doesn't cover things
[11:35:20] <kollabierer> a few days ago, I was here and asked about the RS-274
[11:35:33] <kollabierer> because I have a program, that outputs RS-274
[11:35:39] <kollabierer> but EMC doesn't understand it !
[11:36:08] <kollabierer> and then I got told, that EMC doesn't understand RS-274, just RS-274bgc, a self made dialect
[11:36:22] <alex_joni> is it rs-274? or rs-274d? or rs-274x? or some other version?
[11:36:30] <alex_joni> did you put it online for someone to look at it?
[11:36:38] <alex_joni> what is the error that emc gives you?
[11:36:56] <alex_joni> what is the name of the program?
[11:37:26] <anonimasu> alex_joni: the rs274 dosent output moves..
[11:37:29] <anonimasu> just coordinates
[11:37:42] <alex_joni> anonimasu: what do you mean?
[11:37:53] <anonimasu> alex_joni: the program he's trying to run is not g-code.. remotely
[11:38:05] <anonimasu> alex_joni: it's a set of coordinates.. with aperture size
[11:38:09] <anonimasu> no gxx commands
[11:38:18] <kollabierer> like RS-274 defines....
[11:38:52] <kollabierer> nad I tried RS-274 as well as RS-274x
[11:39:07] <alex_joni> rs-274x is gerber
[11:39:20] <alex_joni> here's the standard: http://www.artwork.com/gerber/274x/rs274xrevd_e.pdf
[11:39:26] <alex_joni> (looks quite similar to g-code to me)
[11:39:51] <alex_joni> except for strange things like poly filling & such
[11:40:08] <alex_joni> anonimasu: then I guess it's not rs274
[11:40:23] <anonimasu> alex_joni: it's rs274, but not rs274ngc..
[11:40:42] <anonimasu> though who's rs274 ;)
[11:41:09] <Guest333> hello guys , i just finished my stepper motor driver!! emc i'm coming :-)
[11:41:16] <Guest333> http://www.youtube.com/watch?v=7m9jIlv-vhc
[11:41:46] <alex_joni> "RS274 is a programming language for numerically controlled (NC) machine tools, which has been used for many years. The most recent standard version of RS274 is RS274-D, which was completed in 1979. It is described in the document "EIA Standard EIA-274-D" by the Electronic Industries Association [EIA]. Most NC machine tools can be run using programs written in RS274. Implementations of the language differ from machine to machine, however, and
[11:42:32] <alex_joni> Guest333: just be carefull not to start any fires with it :P
[11:43:02] <Guest333> y?
[11:43:04] <anonimasu> oooh dangerous
[11:43:05] <anonimasu> :D
[11:43:08] <Guest333> because of the transformer?
[11:43:12] <alex_joni> Guest333: I was joking
[11:43:15] <Guest333> ok :-)
[11:43:24] <alex_joni> lots of fires around athens lately ;)
[11:43:29] <Guest333> it's very fused, it has no problem
[11:43:36] <Guest333> yes:-( and to many dead people
[11:43:45] <Guest333> 67
[11:44:39] <alex_joni> ouch :/
[11:45:26] <Guest333> anybody has any experience, where to find info for a stepper motor encoder? (missed steps)
[11:45:59] <alex_joni> what kind of info?
[11:46:14] <Guest333> what is an encoder, how to build one and stuff like that
[11:49:26] <alex_joni> Guest333: http://en.wikipedia.org/wiki/Rotary_encoder
[11:54:44] <Guest333> bb and thanks
[11:58:23] <Marcin> hi all
[11:58:45] <Marcin> i have a little question about emc
[11:59:25] <Marcin> have anyone use emc with TUR 50 DC ?
[12:10:32] <Guest729> bunny
[12:10:42] <Guest729> oops
[12:10:52] <Guest729> Guest729 is now known as skunkworks_
[12:17:57] <skunkworks_> logger_emc: bookmark
[12:17:57] <skunkworks_> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-09-13.txt
[12:23:46] <skunkworks_> Marcin: what is a TUR 50 DC ?
[12:36:32] <Marcin> how to say in english
[12:37:01] <Marcin> millingcenter
[12:37:14] <skunkworks_> ah
[12:37:38] <alex_joni> http://marketeo.com/pl/offerDetail/149548_Tokarka_TUR-50_SN_DC_x_1000mm_z_falownikiem.html ?
[12:37:57] <alex_joni> looks more like a lathe to me
[12:39:15] <Marcin> yes sorry i donk know english name :)
[12:39:29] <Marcin> this link is correct :)
[12:40:31] <Marcin> I would like to control it by ems
[12:40:42] <Marcin> emc
[12:42:45] <jepler> emc is good software to use on lathes, but it's always a lot of work to convert any machine to work with emc.
[12:43:41] <Marcin> ehe i read doc on the web side and it look dificult
[12:44:03] <Marcin> but price is very good :)
[12:44:21] <alex_joni> it's not as much difficult to make emc run, but changing wires and motors and drives (if needed), and figuring out what is already on the machine, etc
[12:44:36] <alex_joni> I'd say you'll have just as much trouble with any other control
[12:45:17] <Marcin> mhm but where is the beginig of this job where to start ?
[12:47:08] <archivist> reading, learning
[12:47:27] <Marcin> :)
[12:47:48] <alex_joni> and planning
[12:47:51] <Marcin> getiing a pice of crap ;p
[12:48:23] <Marcin> ok but one thing is iteresting for me
[12:48:24] <archivist> taking new toys to bits, see how they work
[12:49:07] <Marcin> TUR has stepping motors with stepper driver
[12:49:30] <skunkworks_> do you know if the drives take 'step and direction'?
[12:49:35] <alex_joni> it really depends on your goal
[12:49:58] <alex_joni> if you only want to control moving the axes, then it should be a couple days to make it work (including tuning)
[12:50:12] <alex_joni> if you want toolchangers and limits & all, it might get easily complex
[12:51:03] <Marcin> i think only about moving the axes
[12:51:22] <Marcin> Tur dosn't support tools container
[12:52:13] <Marcin> o i think that it gets step and direction
[12:52:29] <alex_joni> then it should be fairly simple
[12:52:34] <Marcin> :)
[12:52:36] <alex_joni> download the LiveCD from linuxcnc.org
[12:52:42] <Marcin> ok
[12:52:46] <alex_joni> start it on a computer with parallel port
[12:52:54] <alex_joni> connect the output from the parport to the drives
[12:52:58] <alex_joni> and start moving it
[12:53:47] <Marcin> hmmm but haw I shoul connect LPT to my driver it doesn't have this port
[12:55:39] <alex_joni> you need 2 wires for each drive
[12:55:44] <alex_joni> one for step, one for direction
[12:57:00] <Marcin> ok so I should cut of old control from driver plug this two wires and that is all
[12:57:12] <alex_joni> basicly
[12:57:19] <alex_joni> the problem is usually in the details :D
[12:57:59] <Marcin> you men for instane in HAL ?
[13:00:10] <Marcin> after this hardware job I gess that I should configure lpt for my machine params
[13:00:35] <Marcin> in HAL am I correct ?
[13:00:44] <alex_joni> you need to adjust steps/mm or steps/inch, velocities, accelerations, etc
[13:02:03] <Marcin> ok so now I have a picture how to do this :)
[13:02:21] <Marcin> thanks a lot
[13:02:28] <alex_joni> np
[13:03:28] <Marcin> CU I hope i Will manage :D
[13:04:20] <alex_joni> if you don't there are always people in here who can help
[13:05:36] <jepler> For those who have worked with multiple kinds of stepper drivers: do you prefer setting motor current with a resistor (like gecko drives) or with a trimmer (like xylotex)?
[13:06:16] <archivist> er trimmer is a resistor
[13:06:24] <jepler> well I know that
[13:06:35] <jepler> perhaps I wasn't sufficiently precise in my language
[13:06:37] <alex_joni> I used resistor
[13:06:43] <alex_joni> fixed value resistor
[13:06:44] <jepler> with a gecko you put a resistor in two of the screw terminals
[13:07:01] <jepler> with a xylotex you turn a trimmer resistor on the board until you get the value you want
[13:07:02] <skunkworks_> the 2 main ones I use - dip switches and resister. I think I would prefer a pot..
[13:07:09] <jepler> with a gecko you figure what resistor you need from a formula
[13:07:21] <jepler> with a xylotex you use a meter to check that you got the voltage you want
[13:07:33] <alex_joni> xylotex sounds a bit nicer
[13:08:24] <archivist> trimmer, allows removal of local tolerance
[13:11:14] <alex_joni> and you have it handy
[13:11:20] <alex_joni> jepler: any reason for asking?
[13:12:02] <jepler> since pminmo doesn't give eagle files anymore, I'm working on an l297/8 board -- I think that's one of the best choices for the hobbyist. I have experience with the xylotex and found it a pain to adjust the trimmer -- it's a 1-turn type and so near the test terminal that I had trouble fitting the probe and screwdriver in the space
[13:12:28] <jepler> I wondered if people who had used gecko style liked it or not, because I was leaning towards that in my board design
[13:13:12] <skunkworks_> * skunkworks_ hand jepler some aligator clips..
[13:13:24] <jepler> bah
[13:13:40] <jepler> in this case the test terminal is just an exposed via near the trimmer
[13:14:11] <skunkworks_> I have not had any that riquired a trimmer to adjust.
[13:14:13] <jepler> (I also don't like that the trimmer can be set to "fry driver chip" but that's easily avoidable and I'm not sure why the xylotex guy doesn't fix that)
[13:14:37] <skunkworks_> jepler: why are you creating your own stepper driver?
[13:14:44] <archivist> put a fixed r in seris with the trimmer
[13:15:11] <jepler> archivist: exactly, to set the maximum reference voltage
[13:16:06] <jepler> skunkworks_: I want something available like pminmo's single-sided l297/l298 design, but available without bullshit. not for myself, but for others.
[13:16:29] <jepler> skunkworks_: I think l297/298 is one of the best driver setups for hobbyist (all thru-hole, etc)
[13:16:57] <archivist> "they just work" TM
[13:17:04] <jepler> (pminmo doesn't permit modification, redistribution, and only furnishes .pdf format files now)
[13:17:33] <jepler> anyway .. back to work
[13:17:35] <jepler> thanks for your input guys
[13:17:51] <archivist> basic circuits are in the SGS databooks
[13:18:27] <skunkworks_> ah - I understan. the whole pminos 'I don't give out board layouts anymore'
[13:18:37] <archivist> could be an excuse to fire up PCAD
[13:19:28] <archivist> I think the reference SGS board is single sided
[13:26:39] <fenn> you can use a multiturn trimpot, or two trimpots
[13:27:21] <SWPadnos> I think trimmers have worse temperature coefficients in general
[13:27:21] <fenn> fine and coarse
[13:27:28] <SWPadnos> so as it heats up, the setting will change
[13:27:41] <alex_joni> SWPadnos: depends on the trimpot
[13:27:48] <alex_joni> but a crappy single-turn definately is like that
[13:27:50] <SWPadnos> as I said, in general :)
[13:28:02] <alex_joni> I'd use a multiturn (at least 50 turns or so)
[13:28:15] <archivist> cermet 10 turn
[13:28:19] <SWPadnos> and I don't think a lot of hobbysits are going to spend $30 on a precision Bourns 15-turn pot ;)
[13:28:28] <fenn> 50 sounds like a lot
[13:28:34] <alex_joni> http://www.solarbotics.com/products/index.php?scdfa-250100084-viewDetail-productzq3126zq4categoryzq39=true
[13:28:38] <fenn> swp that's why i suggested two
[13:28:48] <alex_joni> fenn: 20 turns ;)
[13:29:56] <alex_joni> I have some like these, which I quite like: http://www.smcelectronics.com/PT30.JPG
[13:30:20] <SWPadnos> yep - those are one of the form factors the Bourns ones are
[13:30:29] <alex_joni> http://www.smcelectronics.com/MTPOT01.JPG
[13:30:41] <SWPadnos> see! :)
[13:30:45] <alex_joni> 10 turns, 7.5$
[13:30:45] <fenn> http://futurlec.com/PotRect.shtml
[13:31:05] <fenn> $0.80-1.30
[13:31:20] <SWPadnos> yep. look at the temp. coefficients ...
[13:32:10] <fenn> data sheet what's that?
[13:32:18] <SWPadnos> yeah - looking for it now
[13:34:43] <SWPadnos> heh - powerrating at 70 C: 0.50W, power rating at 125C: 0W
[13:35:48] <SWPadnos> the temp coefficient is +/- 100 PPM / deg C
[13:36:10] <SWPadnos> that's for the 10% pots, not the 20 or 25% tolerance versions
[13:36:48] <SWPadnos> so I guess that's not too bad. a 30 C temp. change would only be 0.3% max drift
[13:37:58] <SWPadnos> the cheaper ones (+/- 25%) are 250 PPM/C
[13:38:23] <SWPadnos> and only $0.80 instead of $0.83!
[13:42:08] <skunkworks_> * skunkworks_ stole a lot off of junk circuit boards (10 turn pots)
[13:46:08] <SWPadnos> thief!
[13:49:16] <skunkworks_> they come in handy :)
[13:49:27] <SWPadnos> I'm sure they do :)
[13:51:48] <SWPadnos> wow - Murata has surface mount pots that are $0.09 in quantity, $0.20 in singles O_O
[13:52:03] <alex_joni> bbl
[13:52:09] <SWPadnos> seeya
[15:22:27] <JymmmEMC> yo
[15:37:03] <skunkworks_> jlmjvm: http://www.cnczone.com/forums/showthread.php?p=342355#post342355
[17:13:27] <Adam__> hello
[17:15:38] <alex_joni> hi
[17:15:48] <Adam__> Whats going on?
[17:26:58] <alex_joni> not much
[17:26:58] <alex_joni> :)
[17:27:20] <skunkworks_> Adam__: did you get your scales hooked up to the mesa board?
[17:28:09] <Adam__> I am waiting on the breakouts still. I did get my pinouts though for them.
[17:28:17] <SWPadnos> oh man: http://www.dailymail.co.uk/pages/live/articles/news/news.html?in_article_id=481411&in_page_id=1770
[17:29:04] <Adam__> Right now I am looking for a Quill kit for my bridgeport type clone. I want Z axis cnc, currently I just have a scale on it.
[17:32:53] <SWPadnos> Adam__, if you have lots of money and want something that's reputed to be excellent, try these guys: http://www.elrodmachine.com/
[17:33:47] <tomp2> air die grinders are rated to fairly high speeds, are suited to lateral pressure, and have bearings and collets, might they be used with a motor ( pulley and belt ) to make a 'hi-speed' spindle? ( oh, and are cheap :)
[17:49:39] <skinnypuppy34> Etchaskecth link was cool
[17:49:57] <SWPadnos> yeah - wow
[17:50:04] <SWPadnos> and no CNC ;)
[17:57:11] <tomp2> Makita Die Grinder — 28,000 RPM, 6.6 Amp, Model# GD0800C or Milwaukee Die Grinder — 14,500 RPM, 11 Amp, Model# 5196 both electric. gotta compare to routers...
[18:02:41] <lerman> Adam__: SWPadnos is correct. I did an Elrod quill conversion and am very happy with it.
[18:02:59] <SWPadnos> they're in the $2k range, right?
[18:03:58] <lerman> I forget what the whole thing cost me. I got their X, Y, and Z kits. Supplied my own servos. Jon's PWM drives and controllers.
[18:04:27] <Adam__> yeah I am trying to get a price on the quill
[18:04:46] <SWPadnos> ok. I spoke to Dwayne a couple of years ago, and I think the X+Y set was $1600 and the quill was $1900
[18:04:51] <SWPadnos> or it was the other way around :)
[18:05:03] <Adam__> I am thinking a gecko drive and not sure about the servo motor, any suggestions?
[18:05:26] <lerman> SWPadnos: sound about correct.
[18:05:36] <SWPadnos> the ones I have are Baldors, from an Anilam coversion
[18:05:53] <SWPadnos> geckos are a little undersized for them, specifically the 80V limit
[18:06:14] <SWPadnos> these are MTE-4070BLBCE (though you can use others with similar part numbers)
[18:06:37] <SWPadnos> 27 in-lb @ 9.7A, 2500RPM@100V
[18:06:48] <SWPadnos> peak of 39A and 150V
[18:07:11] <SWPadnos> (I could be a little off there, going from memory)
[18:07:35] <Adam__> I have a 110VDC powersupply in my machine right now I would like to use that.
[18:07:47] <SWPadnos> geckos are rated at 80V
[18:08:13] <Adam__> I could get the same drive as my x&y currently and jsut plug it in, but that is 900 alone
[18:08:35] <SWPadnos> Jon Elsons drives look pretty good for BP sized mills. 160V/20A, I think
[18:08:45] <SWPadnos> and I believe they're $125
[18:09:05] <SWPadnos> I'm considering using those instead of the geckos I have
[18:30:32] <skunkworks_> my servos would only spin 840rpm
[18:30:44] <skunkworks_> at 80v
[18:31:41] <Adam__> anybody know how to control a VFD from a Mesa 7i37?
[18:31:54] <SWPadnos> use a 7i33 instead :)
[18:32:01] <Adam__> heh thats what i thought
[18:32:26] <SWPadnos> the 7i37 can probably do it using software PWM, but I think the turn-on/turn-off times for the outputs are pretty slow
[18:32:49] <SWPadnos> (so you'd probably get some asymmetry unless you use PDM and a push-pull type of filter)
[18:33:03] <Adam__> Also, I need 10vdc 10ma power for the actual vfd can I get that off one of the boards?
[18:33:37] <SWPadnos> the 7i37 just closes an electronic contact, so you'd probably use the reference + terminal on the VFD
[18:34:13] <SWPadnos> one possibility is to use 2 outputs as push/pull - one contacts to +, the other to -, and both connect to the "output"
[18:34:30] <SWPadnos> output goes to a slow filter (like .1 second time constant, maybe longer)
[18:34:34] <SWPadnos> and on to the VFD
[18:35:02] <SWPadnos> the VFD should have 3 terminals for this, like you'd use for a pot (+, wiper, -)
[18:35:37] <Adam__> can you adjust the output voltage on the 7i37 (it outputs 48vdc)? I only need 10vdc
[18:35:45] <Adam__> sofware wise I mean
[18:35:55] <SWPadnos> it doesn't output anything
[18:36:07] <SWPadnos> it allows you to switch things up to 48V
[18:36:43] <SWPadnos> (this is like a light switch - it's rated for 600V, but it doesn't source anything, it just switches whatever you connect to it)
[18:37:01] <SWPadnos> it being a light switch, not the 7i37, which is rated at 48V :)
[18:37:02] <Adam__> so why does it say it has 8 48vdc 1A output drivers
[18:37:21] <SWPadnos> the drivers can be connected to a 48V supply, and can switch 1A loads at that voltage
[18:37:27] <Adam__> ok understand
[18:37:39] <Adam__> we arent elec. engs over here
[18:37:40] <SWPadnos> heh
[18:37:44] <Adam__> mechs
[18:37:50] <SWPadnos> oh, that kind
[18:38:21] <Adam__> ok what about the forward reverse and brake for the vfd
[18:38:38] <SWPadnos> those should be able to directly connect to 7i37 outputs
[18:38:46] <Adam__> ok thats what i thought
[18:38:52] <Adam__> so helpfull
[18:38:54] <SWPadnos> heh
[18:42:11] <Adam__> Its complicated the first go around thats forsure
[18:42:31] <SWPadnos> "with great power comes great responsibility" :)
[18:42:52] <Adam__> Yeah well once you do it once then its cake after
[19:11:30] <alex_joni> bye guys
[19:11:33] <alex_joni> see you on saturday
[19:11:45] <skunkworks_> have a good trip
[19:11:47] <alex_joni> thx
[19:14:55] <CIA-8> 03cradek 07TRUNK * 10emc2/src/emc/rs274ngc/interp_read.cc: fix block delete problem reported by stuart s.
[20:15:01] <lerneaen_hydra> http://www.youtube.com/watch?v=HKEuzxC4eGc
[20:23:15] <skinnypuppy1334> Interesting vid,
[20:24:25] <skinnypuppy1334> http://www.youtube.com/watch?v=zT60SkXN1UY
[21:33:48] <dmessier> hi all
[21:53:53] <jlmjvm> swpadnos:have you ever installed a 2nd parport before
[21:54:02] <SWPadnos> yes, but not for EMC :)
[21:54:20] <jlmjvm> im trying in widows right now
[21:54:39] <SWPadnos> I have a parport card that I tried out with EMC for a few minutes a couple of years ago, but I don't remember at all what I did
[21:54:57] <jlmjvm> are you supposed to be able to see the new port in bios when its installed correctly?
[21:55:03] <SWPadnos> no
[21:55:10] <jlmjvm> k
[21:55:21] <jlmjvm> thats a load off my mind
[21:55:39] <SWPadnos> if you're in Windows, then you can "look for new hardware" if it doesn't already show up in the device manager
[21:55:52] <jlmjvm> windows sees it,loaded dos drivers also
[21:56:19] <jlmjvm> followed web instructions,still no luck
[21:56:25] <SWPadnos> if you plan on using it with EMC, then you need to use "lspci" to see what address it was assigned in Linux
[21:56:33] <SWPadnos> web instructions on the EMC2 wiki?
[21:56:39] <jlmjvm> k
[21:57:01] <jlmjvm> was trying windows before i got that far
[21:57:10] <SWPadnos> no point to that, for the most part
[21:57:20] <jlmjvm> figured it would be plug n play,wrong
[21:58:12] <SWPadnos> if it's detected automatically, then that tells you it's working as far as the BIOS is concerned. if it doesn't work in Windows, then troubleshooting that is of no help in Linux
[21:58:35] <SWPadnos> one thing - it usually helps to insure that the BIOS is set to "non-plug-n-play OS" (or similar)
[21:59:20] <jlmjvm> gonna try a linux install,may get lucky
[22:00:30] <jlmjvm> the main question i had was could it be seen in the bios,thanks
[22:00:53] <SWPadnos> good luck
[22:03:38] <jlmjvm> btw,im gonna have something for the wiki in a few days,my nasa buddie is gonna look it over when im done
[22:06:35] <SWPadnos> come on, this isn't rocket science :)
[22:11:03] <jlmjvm> he wants it to be a good post
[22:12:15] <jlmjvm> he is becoming interested in the project
[22:16:03] <skunkworks> jlmjvm: http://www.cnczone.com/forums/showthread.php?t=43517
[22:17:56] <jlmjvm> thats cool
[22:18:38] <skunkworks> its a start :)
[22:18:51] <jlmjvm> mach3 cant do it
[22:19:16] <jlmjvm> with a 150 board you get read 40 times a sec
[22:20:27] <skunkworks> yeck
[22:20:42] <jlmjvm> but if you dont buy an encoder board you dont get the plugin for the software to work,should have that capability from a safety standpoint
[22:21:49] <jlmjvm> what suprised me is the encoder numbers dominate the emc numbers
[22:22:40] <skunkworks> what do you mean?
[22:22:45] <jlmjvm> instead of home at z0. everytime,now it shows -.0001,or .0002
[22:23:00] <jlmjvm> swows actual position
[22:23:05] <jlmjvm> shows
[22:23:38] <jlmjvm> know where your at for real at a glance
[22:23:54] <SWPadnos> you can switch the display between commanded and actual position
[22:24:08] <jlmjvm> k
[22:24:39] <skunkworks> you can probably twist the shaft while the steppers are on and see the numbers move
[22:24:49] <skunkworks> a little bit
[22:25:11] <skunkworks> (without making it skip)
[22:25:35] <jlmjvm> sure can
[22:26:00] <jlmjvm> make it read .0001 either way with a little pressure
[22:26:58] <SWPadnos> once you get the machine making parts, it would be interesting to use halscope or halstreamer to plot the following error
[22:27:06] <jlmjvm> i dont see why this wont work with glass scales also,or anything really
[22:27:07] <SWPadnos> err - halsampler maybe
[22:27:32] <SWPadnos> you can take the position output of the stepgen, subtract it from the encoder output, and see how they differ
[22:27:35] <jlmjvm> we can do that,would like to see
[22:27:38] <SWPadnos> I'd be interested in seeing that data
[22:27:48] <SWPadnos> both unloaded and under load
[22:27:54] <SWPadnos> (ie, cutting steel :) )
[22:28:17] <jlmjvm> we can cut it
[22:28:30] <SWPadnos> I'll bet there's some oscillation - you can probably see whether you're climb or conventional milling, and the number of flutes on the endmill
[22:28:51] <jlmjvm> that would be interesting
[22:29:10] <SWPadnos> well, if it's X or Y. if this is Z, then it would show something else
[22:29:19] <SWPadnos> (not exactly sure what at the moment)
[22:30:33] <SWPadnos> well, time for dinner. bbl
[22:30:55] <jlmjvm> same here,later
[22:34:20] <jmkasunich_> jmkasunich_ is now known as jmkasunich
[22:34:46] <jmkasunich> jepler: fwiw, I much prefer the gecko approach to setting current
[22:34:48] <jmkasunich> no meter needed
[22:34:49] <jmkasunich> no chance of it changing by accident
[22:34:51] <jmkasunich> and if its a build-your-own board, you don't even need terminals for the resistor - just provide pads, and choose the resistor value at build time
[22:35:07] <skunkworks> Hi john
[22:35:12] <jmkasunich> hi sam
[22:35:52] <skunkworks> trench filled in?
[22:35:57] <jmkasunich> yep
[22:36:01] <jmkasunich> filled monday
[22:36:06] <skunkworks> nice
[22:36:18] <jmkasunich> shockingly larger than expected bill presented monday
[22:36:20] <skunkworks> my 2 piles of sand are flat as of today
[22:36:25] <jmkasunich> shouting tuesday
[22:36:28] <skunkworks> ew
[22:36:38] <jmkasunich> somewhat smaller bill paid wednesday
[22:36:42] <skunkworks> heh
[22:36:53] <jmkasunich> still lots more than the original quote
[22:36:56] <skunkworks> they charged you for the 3 man days that you didn't get?
[22:37:11] <jmkasunich> but I do have a new copper water line and new storm sewer that weren't originally planned
[22:37:35] <skunkworks> oh - forgot about that.
[22:37:55] <jmkasunich> original quote $6800, bill $13300, final amount $11700
[22:38:22] <skunkworks> yikes
[22:38:37] <skunkworks> but - something you won't have to worry about probably in your life time
[22:38:41] <jmkasunich> right
[22:39:08] <skunkworks> any plugs will be easily cleaned :)
[22:39:28] <jmkasunich> yep, and should be very unlikely in the first place
[22:39:32] <skunkworks> yes
[22:40:07] <skunkworks> did you see jlmjvm setup steppers with encoders - and is seems to work. That is pretty neat.
[22:45:04] <jepler> jmkasunich: I have a feeling lots of hobbyists love to do things like turn the trimpot a bit this way or that to see if they get 1ipm better performance before their stepper motors burn up </1/2 wink>
[22:45:34] <jmkasunich> how about putting a hole pattern on the board that allows either a trim pot or a 1/4 watt resistor
[22:45:40] <jmkasunich> holes are cheap
[22:45:56] <jepler> yeah
[22:47:54] <jepler> people like you will put in the resistor, people like me will endlessly fiddle the pot
[22:48:35] <jepler> * jepler tries to remember if he last set his xylotex to 100% current or 75% current
[22:48:43] <jmkasunich> heh
[22:49:09] <jmkasunich> time for dinner, bbl
[22:49:15] <jepler> see you
[22:56:28] <skunkworks> woo whoo http://www.electronicsam.com/images/house/garagesand.JPG
[22:57:33] <skunkworks> the back yard and house is in the background.. (cedar shake siding)
[22:57:33] <skunkworks> http://www.electronicsam.com/images/house/newhehhouse.JPG
[22:57:42] <skunkworks> finally a front view
[22:57:49] <skunkworks> needs a little tlc
[22:58:25] <skunkworks> for some reason the train was parked when I walked to get the mail http://www.electronicsam.com/images/house/trane.JPG
[23:00:18] <LawrenceG> JymmmEMC: boink!
[23:06:20] <LawrenceG> JymmmEMC: any chance you can open up your oem750 so we can compare with oem750x?
[23:38:42] <LawrenceG> JymmmEMC: http://imagebin.org/10418 and http://imagebin.org/10417
[23:43:44] <skunkworks> LawrenceG: this is my oem650x-> oem650 :)
[23:43:46] <skunkworks> http://www.electronicsam.com/images/OEM650.JPG
[23:44:44] <skunkworks> the official oem650's have a jumper 'block' that shorts out the first 5 teminals.
[23:45:47] <skunkworks> I used seagate scsi drive jumpers
[23:46:19] <LawrenceG> hmmm.... wonder if the 750s are the same... they look very similar although mine has 2 dip switches instead of your red jumpers
[23:46:36] <skunkworks> ah
[23:49:45] <skunkworks> I would 'think' that they would use the same indexer :)
[23:49:53] <LawrenceG> was wondering if sw1 in http://imagebin.org/10418 is used in oem750
[23:50:42] <LawrenceG> good thought on the indexer card.... on mine it looks like the first 5 pins are special
[23:50:58] <skunkworks> I don't have a 650 here.
[23:51:08] <skunkworks> I mean the x card
[23:51:26] <LawrenceG> I was surprised to see how nice and compact the drive is
[23:51:56] <skunkworks> small huh - not as small as the gecko's - everytime I see one I still am supprised as how small it is
[23:53:07] <LawrenceG> yea... it has an 80c52 processor and descrete fet ouput stage t0220.... nice, but hard to clone for ~$50usd ebya prices
[23:53:10] <skunkworks> you can follow back the step and direction pins from the 25 pin connector. those 4 should be hooked strait thru I would think.
[23:54:08] <skunkworks> I may be talking out of my ass - I bet the plug is after the optos.