#emc | Logs for 2009-03-27

[00:02:36] <SWPadnos> hey seb_kuzminsky, are you following the m5i20_pidtest conversation on the user list?
[00:51:06] <seb_kuzminsky> SWPadnos: i dun seen it
[00:51:14] <SWPadnos> ok :)
[00:51:21] <seb_kuzminsky> but i'm not convinced it's related to hm2
[00:51:24] <SWPadnos> I've just realized for sure that he's not using hm2
[00:51:32] <seb_kuzminsky> yeah :-/
[00:55:48] <john_f> Hey, I think I would like to go to Fest. I had fun last year, but I would like to learn more. Would this be a good opportunity for someone to mentor me?
[00:56:41] <SWPadnos> hmmm. that's a tough question to answer
[00:57:01] <SWPadnos> what you attended last year was a combination of the CNC workshop and the EMC fest
[00:57:27] <SWPadnos> I think the idea of this gathering is more of a "pure" EMC fest,
[00:57:47] <SWPadnos> which means lots of programming, discussion about code organization, new features, etc. etc.
[00:58:21] <SWPadnos> we'll also have access to some machines in Stuart's shop, for testing things like joint limits and the like (on non-trivial kinematics machines)
[00:58:43] <SWPadnos> there will likely be some classes on using/configuring EMC2
[00:58:44] <john_f> Yes that was what I thought so maybe I would just be in the way.
[00:58:54] <SWPadnos> well, you're welcome to come
[00:59:08] <SWPadnos> I just don't know that there would be a lot of mentoring going on ;)
[00:59:14] <SWPadnos> (and it depends on what you mean also)
[00:59:42] <SWPadnos> I think, but could be wrong, that this is geared more toward development and less toward education/promotion
[00:59:43] <john_f> well if I had a project to bring and work on I would have lots of experts to ask questions
[00:59:48] <SWPadnos> sure
[01:00:02] <john_f> Is this the right forum?
[01:00:24] <SWPadnos> yes
[01:00:46] <SWPadnos> if it's related to using or setting up EMC2, this is the place
[01:01:08] <SWPadnos> #emc-devel is the place if you want to discuss some aspect of the code, like if you want to add a component or something
[01:02:54] <john_f> Well I would like to get to that point but I still haven't got my machine running. I would eventually like to learn more details of the inner workings etc.
[01:03:16] <SWPadnos> luckily, this is a self-paced class :)
[01:04:11] <john_f> good my pace is slow. I think I would like also to talk about future network connected drives etc.
[01:04:40] <john_f> Like ether-cat connectivity.
[01:05:14] <SWPadnos> well, there may be some of that at the Fest
[01:05:32] <john_f> My problem is I think there is a steep learning curve for me to ever be able to contribute
[01:05:49] <SWPadnos> well, that's true in some places, but not in others
[01:06:18] <SWPadnos> for example, you may not be able to contribute code for a while, but you can certainly contribute with your knowledge of smart drives and how they communicate
[01:16:50] <cradek> making a hal driver for new hardware is one of the times you can ignore 99% of the code in the project
[01:17:42] <SWPadnos> with the caveat that new classes of hardware could require massive changes to the core of EMC
[01:17:58] <SWPadnos> such as ethernet/USB/Profibus/SERCOS smart motor drives :)
[01:18:16] <cradek> yeah, possibly I guess
[01:18:48] <SWPadnos> well, unless no feedback is needed, in which case they're just "send it and forget it"
[03:56:13] <ds3> have I ever mention I hate machining acrylic?
[04:07:59] <seb_kuzminsky> why is that ds3?
[04:08:16] <seb_kuzminsky> dont like those long stringy chips?
[04:10:22] <tomp> or the melted wads when tool/feed/speed is wrong ;)
[04:12:11] <tomp> any 'chip breaker' grind for tools cutting plastic?
[04:15:48] <ds3> all the welded stuff
[04:15:55] <ds3> unlike ABS which chips nicel
[04:16:09] <tomp> this page has no real info on this years fest, where to look? http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?EMC_Fest_2009
[04:16:45] <tomp> ds3, ah some plastic cutting technology is needed... grind feed speed lube.. out of my experience
[04:17:05] <ds3> tomp: tried different speeds
[04:17:20] <ds3> and it seems to have dulled the carbide cutter pretty quickly too
[04:17:40] <tomp> really! dulled carbide or loaded it up?
[04:17:46] <ds3> dulled
[04:18:06] <tomp> gack! fibre reinforced or straight acrylic?
[04:18:12] <ds3> I cut abs before the acrylic and burring wasn't bad. did some acrylic now the back to ABS and it burrs like mad
[04:18:37] <ds3> I donno... the backing sheet was clear and it was in the plastic dealer's rems
[04:18:41] <seb_kuzminsky> tomp: the fest talk has all been on the mailing list
[04:19:17] <tomp> haha somehow i was booted from the mailing list... some automagic too many bounces
[04:19:29] <seb_kuzminsky> oops
[04:19:38] <tomp> np, i can web read
[04:19:38] <ds3> oh well... at least that was a cheap cutter (reground carbide)
[04:20:29] <tomp> ds3 id ask in some forum closer to cutting ( emc is control end )
[04:20:45] <ds3> tomp: *nod*
[04:20:55] <tomp> and let us know ;)
[04:20:58] <seb_kuzminsky> http://article.gmane.org/gmane.linux.distributions.emc.user/13048
[04:21:06] <tomp> thx seb
[04:26:50] <tomp> and thx to Stuart
[04:27:09] <seb_kuzminsky> yeah thanks stuart :-)
[04:30:53] <tomp> tips on milling acrylic including cutters http://www.deskproto.com/gallery/perfume.htm
[04:31:42] <tomp> ds3: ^^^
[04:32:57] <ds3> Oh
[05:32:43] <ren_encee> ren_encee is now known as renesis
[07:39:45] <Guest246> is there anyway to enable the modbus communication feature of classicladder? seems to be disabled in the emc2 version
[07:57:37] <pjm_> good morning
[07:58:34] <micges> good morning
[10:52:18] <pjm_> wow, just tried 2.3 beta!!! amazing
[10:52:29] <pjm_> the rigid tapping is as smooth as you like
[10:52:54] <pjm_> none of the Z axis velocity ringing at the start of the op
[11:55:39] <alex_joni> pjm_: that's really good to hear
[12:03:36] <pjm_> yes it is excellent, works perfectly, M3 X 0.5 tapping even at 1000rpm
[12:04:10] <pjm_> i did a short video of it tapping 5 holes in a lump of nylon, just fiddling with video before uploading it
[12:13:48] <skunkworks_> pjm_: are you rigid tapping thru the printer port? (reading the encoder back?)
[12:13:59] <pjm_> yeah
[12:14:07] <pjm_> 360ppr spindle encoder
[12:14:12] <pjm_> quadrature thing
[12:14:16] <skunkworks_> Cool! :)
[12:14:27] <pjm_> yeah i am dead impressed!
[12:14:27] <skunkworks_> do you have any pictures of your setup?
[12:14:39] <pjm_> just making a video of the tapping
[12:14:56] <pjm_> but there are some foto's of the machine on wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Pjm
[12:15:10] <pjm_> i'll make some updates and put new photos on in a minute
[12:15:17] <JymmmEMC> pjm_: Don't forget the half-naked saline-solution model too
[12:15:18] <pjm_> once this utube upload has completed
[12:15:59] <pjm_> ughh ok ;-)
[12:16:18] <pjm_> damn, brb lunch, eating is a pain
[12:16:32] <JymmmEMC> Take the blue pill!
[12:20:08] <skunkworks_> pjm_: very nice
[12:26:26] <archivist> makes you want to stab the comment at the bottom thought
[12:38:09] <pjm_> * pjm_ back
[12:38:34] <pjm_> ah i tried the fast pecking drill cycle too, very nice
[12:41:42] <skunkworks_> is the video there? Huh? Huh? huh? Is it?
[12:41:57] <JymmmEMC> what skunkworks said
[12:42:00] <pjm_> its there!!!
[12:42:01] <pjm_> one sec
[12:42:38] <pjm_> http://www.youtube.com/user/uhfdashsatcomdotcom
[12:42:46] <anonimasu> what was the comment?
[12:44:05] <pjm_> probably the one about the rigid tapping / spindle encoder
[12:44:24] <skunkworks_> Very nice work!
[12:44:35] <anonimasu> yeah
[12:47:12] <pjm_> I'm gonna try tapping some holes in a bit of perspex next so I can see that the thread is clean, then once happy, move on to some bits of ally
[12:47:50] <skunkworks_> pjm_: how much of the 5 printer ports have you used? :o
[12:48:28] <pjm_> 3 so far
[12:48:35] <skunkworks_> neat
[12:49:21] <pjm_> i found a joystick at a junk fest, it was from an electric wheelchair, so I'm gonna add that to move the tabie X / Y plus a rapid button
[12:49:29] <pjm_> which will use another port probably
[12:50:38] <pjm_> but also i have found a bigger metal cabinet for the electronics, so that will cut down some of the noise too, the VFD is a nightmare, even with good quality mains filters I can see a lot of 'fuzz' on voltage rails
[12:58:03] <archivist> comment was from a mach3 user!
[12:58:27] <pjm_> wash yer mouth out!!
[12:59:12] <archivist> delete that comment from your youtube chan :)
[13:03:17] <skunkworks_> pjm_: are you on cnczone?
[13:04:15] <pjm_> i have a login for it somewhere
[13:05:00] <skunkworks_> Thats ok - would you have a problem with me posting the video and a link to the wiki?
[13:06:48] <Optic> hello!
[13:07:42] <skunkworks_> Optic! burn any tattoos off lately?
[13:08:35] <archivist> pjm_, I would not use straight flute taps , get spiral they can get the swarf out a lot better
[13:08:41] <pjm_> skunkworks help yourself to that video etc, its all public domain stuff
[13:08:51] <Optic> nope, but we made some gears last night
[13:09:00] <Optic> in acrylic
[13:09:11] <pjm_> archivist ah yes, i was being lazy, it was the first one in the tool box ;-) - i will do it properly next time !
[13:12:03] <anonimasu> what was the comment?
[13:12:33] <anonimasu> (im curious)
[13:12:55] <pjm_> it was on the Youtube page, from a **Mach3** type...
[13:13:40] <anonimasu> yeah but what did he say?
[13:14:57] <pjm_> "I was looking at EMC2 as well and have since decided that the money spent using Mach3 will be worth it. "
[13:15:48] <anonimasu> hm.. that's like saying I were looking at servos, and I since decided that steppers are worth it
[13:16:45] <pjm_> but then given the thread in DIY-CNC yahoo group I dunno what to think, any concept of running something that could be safety critical on windows leaves me in a cold sweat
[13:17:06] <jepler> mmm rigid tapping videos
[13:17:07] <jepler> tasty
[13:17:15] <anonimasu> hehe
[13:17:25] <anonimasu> indeed
[13:17:26] <jepler> pjm_: so 2.3beta is tapping better than 2.2 for you?
[13:17:47] <jepler> I know cradek and jmkasunich did some work in that area but I don't know the details
[13:17:52] <pjm_> jepler it is unbelievably smooth!!!
[13:18:04] <jepler> your spindle stops and starts pretty fast
[13:18:29] <pjm_> with 2.2.8 the Z velocity used to 'ring' at the start of the tapping cycle, like it was PLL'ing to the spindle speed
[13:18:58] <pjm_> but with 2.3 the motion is perfecly smooth
[13:21:53] <pjm_> jepler ah yes the VFD is set for 0.5second stop time, so it is fairly quick on the reverse, yet the new EMC2.3b tracks that perfectly, even at 1000rpm with the m3 X 0.5
[13:23:06] <skunkworks_> pjm_: have you tapped anything harder yet?
[13:23:17] <skunkworks_> wow - that sounded bad
[13:23:25] <archivist> :)
[13:23:36] <Optic> i was just going to say something like "I tapped your mom last night!" but then decided it would be in poor taste
[13:23:42] <pjm_> hahh
[13:24:03] <jepler> Optic: in poor taste? what, do you have something against skunkworks_'s mom?
[13:24:23] <Optic> not yet!
[13:24:43] <Optic> * Optic goes back to programming
[13:25:12] <skunkworks_> ;)
[13:25:52] <anonimasu> Optic: it usually takes a while before thoose things show up..
[13:26:02] <JymmmEMC> jepler: Did you mean other than her having skunkworks?
[13:26:07] <anonimasu> if it starts to tingle beware..
[13:26:15] <alex_joni> skunkworks_: around?
[13:26:45] <jepler> alex_joni: I assume we shamed him into leaving
[13:26:46] <jepler> .. finally
[13:26:57] <JymmmEMC> lol
[13:26:59] <alex_joni> somehow I doubt that
[13:27:07] <JymmmEMC> jepler: and I thought *I* was bad =)
[13:27:08] <alex_joni> it's gonna take much more than that
[13:28:03] <JymmmEMC> jepler: AXIs is in python or TCL or ?
[13:28:54] <alex_joni> python mostly
[13:29:01] <alex_joni> Tk for the graphical interface
[13:29:22] <JymmmEMC> ok, thats what I thought...
[13:29:39] <alex_joni> some extensions in C
[13:29:44] <JymmmEMC> I wanted to know how Tk is/was for gui dev
[13:29:46] <alex_joni> and C++
[13:29:59] <jepler> JymmmEMC: If you're just learning GUI programming, I wouldn't recommend Tk
[13:30:02] <alex_joni> I think all the toolkit shave cave-eats
[13:30:10] <jepler> learn something more modern like gtk or qt instead
[13:30:15] <JymmmEMC> jepler: No, for dev a kiosk iface
[13:30:17] <alex_joni> s/toolkit shave/toolkits have/
[13:30:31] <Optic> Swing!
[13:30:34] <alex_joni> Qt is OSS lately, so that's probably a good advice
[13:30:38] <alex_joni> Optic: boo
[13:30:45] <alex_joni> is that Java stuff?
[13:30:50] <alex_joni> or some other Swing?
[13:30:50] <Optic> yep
[13:31:00] <tomp> tcl/tk has a long history of making front ends for text apps, independence of the app and gui was foremost feature
[13:31:03] <Optic> i'm a java guy
[13:31:26] <tomp> is that a song title?
[13:31:46] <JymmmEMC> tomp: That would work. Especially if the gui is text and not binary based so could easily be edited as neede
[13:31:47] <JymmmEMC> d
[13:32:53] <jepler> back in the late 80s and maybe early 90s, Tcl+Tk was a great combination because you could write huge swaths of your GUI in a language where you didn't have to worry about explicitly allocating and freeing memory, or making sure you didn't overrun your arrays of characters (e.g., all the problems of C)
[13:33:20] <tomp> tcl me elmo
[13:33:49] <jepler> but now C++, Python, Java, Ruby, and all those languages free you from those troubles in varying degree, and let you access graphical toolkits that integrate better with the desktop environment, look nicer, and have the widgets people expect such as comboboxes and trees built in
[13:34:41] <tomp> is there some thinking of removing the tcltk underpinnings in axis?
[13:35:00] <JymmmEMC> To be honest, I really don't need all that overhead. Might try REBOL
[13:35:12] <alex_joni> this is cool: http://www.youtube.com/watch?v=46sMbPZadPo
[13:35:18] <jepler> (or maybe I'm mistaken that Tk goes back into the late 80s .. ))
[13:35:20] <alex_joni> too bad it's not the actual video
[13:35:26] <tomp> javascript & svg, nice kiosk stuff, no compiler needed
[13:35:27] <jepler> (it's very old, anyway)
[13:36:54] <Optic> http://www.capybara.org/~andrew/temp/laser/gears.jpg
[13:36:56] <alex_joni> http://www.tcl.tk/about/history.html
[13:37:00] <jepler> (yeah, Tk 1.0 was released in 1991)
[13:37:04] <alex_joni> jepler: 80s
[13:37:22] <alex_joni> "I began work on Tk in late 1988, but it was a part-time project, so it took about two years before there was enough functionality for Tk to be useful."
[13:37:56] <JymmmEMC> Hmmm, maybe flash as there are toolkits to write flash using python
[13:38:19] <tomp> flash is pretty, just not open
[13:38:33] <JymmmEMC> tomp: Actually is is as of a few years ago
[13:38:38] <tomp> ?
[13:38:48] <tomp> tomp googles
[13:39:44] <tomp> btw: the stone carving is nice, must be pretty rigid nutating head, but where's the captive ball? ;)
[13:40:09] <JymmmEMC> tomp: http://www.adobe.com/devnet/swf/
[13:41:02] <JymmmEMC> python tools for creating swf files... http://www.swftools.org/
[13:42:48] <tomp> JymmmEMC: thx, reading now
[13:47:17] <tomp> the file format has been published, and there's opensrc tools to manipulate files but i dont see the Adobe/Macromedia app 'Flash' havinf its src code published. np. cool stuff
[13:47:47] <jepler> tomp: there is an open source effort to create a compatible plugin, but it's got a lot of work ahead. google "gnash"
[13:47:55] <tomp> i've watched the vector grfx market for years, svg was as close to open flash as i saw
[13:48:11] <tomp> yes, gnash ( a gnother gnash ;)
[13:49:10] <pjmcnc> cool, just tapped 5X M3X0.5 in a lump of ally - perfect
[13:49:25] <tomp> way cool
[13:49:43] <archivist> stainless steel next?
[13:50:04] <archivist> titanium perhaps
[13:50:22] <jepler> <canvas> is pretty powerful, too. for example, there's the processing.js lib built on it: http://ejohn.org/apps/processing.js/examples/topics/chain.html
[13:50:41] <tomp> i had to use m3s to mount a panel today, thats small enuf for my old eyes
[13:51:06] <jepler> though performance doesn't seem to be that great in my ff3
[13:51:09] <tomp> <canvas> :)) good response
[13:52:53] <jepler> and maybe in a few years we'll get canvas 3d in everyday browsers. http://blog.vlad1.com/2009/03/24/3d-on-the-web-its-go-time/
[13:56:33] <BJT-Work> or http://aclu.org/pizza/images/screen.swf
[13:57:37] <pjm_> archivist - i will look for something else to tap, the ally is a really good thread, no slop or anything with a thread gauge
[13:58:09] <skunkworks_> A few of you keep trying to convince the rest of us that it does. It doesn't.
[13:58:10] <skunkworks_> Mach can bang 100k per channel out the parallel port. Matter of fact it is
[13:58:20] <skunkworks_> capable of 200k per channel out the parallel port.
[13:58:58] <skunkworks_> uh huh
[13:59:45] <tomp> what 'it' do you refer to?
[14:00:53] <jepler> skunkworks_: better to just skip pasting stuff to which there can be no polite response
[14:03:55] <pjm_> archivist i was looking for a bit of nymonic 75 to try ;-)
[14:04:59] <BJT-Work> pjm_ why not some CPM15V
[14:05:02] <archivist> evil stuff
[14:05:45] <JymmmEMC> tomp: There is a new thing from Adobe that is like flash but text based. I kinda stick towards SWF just becaue it's been around for so long, and is actually a very nice single-file container for rich content (text, data, graphics, video, audio, etc)
[14:05:51] <tomp> where can i paste binaries? ( a zip of some demo svg that runs in firefox/moz/opera/w$ )
[14:06:15] <JymmmEMC> filebin.ca iirc
[14:06:37] <tomp> JymmmEMC: whats name of text based flash app?
[14:06:37] <JymmmEMC> tomp: are you talking the animated svg stuff?
[14:06:53] <JymmmEMC> tomp: I dont know on adobe's website.
[14:06:54] <tomp> yep, i did some user i/f with it, wanted to paste it
[14:07:18] <archivist> I had a go with the VRML but that seems to be stuck at windows only
[14:07:38] <tomp> data lookup by clicking on charts, writes code for an app to execute
[14:09:33] <JymmmEMC> tomp: It's Adobe Flex, which I *think* is what is M$ Silverlight. but not sure
[14:09:43] <archivist> js seems the lowest common denominator though http://www.uselesspickles.com/triangles/demo.html
[14:10:25] <tomp> the svg run in browser bundle, unzip in one dir, open with about any newer browser http://filebin.ca/juvfcd/SVG4edmPlanner04.zip
[14:10:55] <JymmmEMC> http://flex.org/showcase/
[14:11:24] <tomp> a guy kevlindev has great svg widgets tho they get broke at every browser update, shame, really slick stuff
[14:13:39] <JymmmEMC> tomp: http://www.adobe.com/devnet/flex/?tab:samples=1
[14:16:39] <tomp> cool, i get lost in those kind of demos, madly clicking on a picture of a button ;)
[14:18:27] <pjm_> skunkworks_ thanks for posting that stuff, i can see hits coming from cnczone
[14:18:55] <JymmmEMC> tomp: Flex uses XML formatting, which I'm personally not happy with. but to each their own.
[14:19:01] <Optic> cnczone is awesome
[14:23:05] <tomp> hmm i was moving towards xml for some files that describe lists of params needed for a process. used to use structs, heh, tomorrow another language will be born
[14:24:07] <JymmmEMC> tomp: Well, if XML was used as it was intended, that would be one thing.
[14:24:40] <JymmmEMC> tomp: There are very few cases that I've seen where XML was used for that.
[14:27:15] <pjm_> btw, does anyone here do closed loop spindle speed control, so monitoring the spindle speed with encoder, comparing that to commanded speed and adjusting as appropriate?
[14:27:45] <tomp> the svg chart range to file screenshot http://imagebin.ca/view/wa1l6DT.html
[14:28:22] <JymmmEMC> tomp: Not sure what platform you're on, and is NOT open source, but check out REBOL/View
[14:28:49] <tomp> REBOL? idiot bastard son of COBOL? ;)
[14:29:42] <JymmmEMC> Well, not sure about the C$%*^L part, but the binary is only 500KiloBytes, and the script CAN be compiled as self-exe
[14:30:39] <JymmmEMC> and does vector graphics very well
[14:30:53] <tomp> i went to svg for clean graphics, i'm stuck with a system that uses BMPs ( puke!)
[14:31:03] <JymmmEMC> M$?
[14:31:17] <JymmmEMC> tomp: http://www.rebol.com/view-platforms.html
[14:31:22] <tomp> yeh
[14:31:38] <JymmmEMC> tomp: http://www.rebol.com/downloads/v276/rebview.exe
[14:31:57] <JymmmEMC> tomp: I think you'll be impresses, espacially with it's size
[14:32:39] <JymmmEMC> tomp: If you dev on M$ the EXACT same code will run on any other platform
[14:32:47] <tomp> JymmmEMC: i'll grab the .deb
[14:32:54] <tomp> thx
[14:33:27] <JymmmEMC> look at the examples, they provide the source code for the scripts too
[14:34:36] <tomp> i like scripting languages ( i think gcode is a scripting language ;) you get stuff done w/o compiling and anyone can read it
[14:35:11] <tomp> well any english speaking person, sorry to be an anglo-bigot
[14:35:41] <tomp> here in taiwan, people code in english and live chinese
[14:42:57] <tomp> and i like viewing the machine inside a browser, i like the control in hdwr, and the browser is just a terminal view of the control. shoot the computer and the control keeps going. ( you still got estop )
[14:49:32] <tomp> wow, i already was using xml for the data ( i forgot! ) <circle id= "CUST10025POS" cx="-2.00" cy="-2.08" r="0.25" fill="rgb(223,123,0)"
[14:49:32] <tomp> peak="100" pol="pos" wkpc="skd11" tool="Cu" form="bore" ont="25" vdi="18" ovb="87" vw="228" ve="7" clicked="false"/>
[14:49:55] <tomp> that was back in 2004
[15:14:08] <JymmmEMC> SVG is XML structured, yes.
[15:24:23] <tomp> i meant struct as in C ( the thing that attracted me to C wayback when )
[15:24:48] <tomp> argh, falling asleep, bye & thx
[15:37:33] <hugomatic> hello guys, is the author of image-to-gcode around?
[15:49:53] <JymmmEMC> ug
[16:29:52] <leit1> hello everybody, i need if someone can tell me please, what is the difference between the MOTION DIGITAL INPUT and the MOTION ANALOG INPUT, the difference that i've found is that the digital needs a wait time and the analog input doesn't need it is that correct?
[16:29:58] <leit1> thanks in advance
[16:30:28] <leit1> so i guess the analog input is always waiting for the signal..
[16:32:15] <BJT-Work> digital is off or on and analog is a variable number AFAIK
[16:33:00] <BJT-Work> yep analog in is a float
[16:33:11] <BJT-Work> and digital in is a bit
[16:33:43] <BJT-Work> both are used by M66
[16:33:43] <leit1> thank you, and is there any chance to use it as an interrupt mode_
[16:33:46] <leit1> ?
[16:34:26] <BJT-Work> like an e-stop input?
[16:34:56] <leit1> yes but, i would need not to stop the machine, i would need to retire an axis to his home position for example
[16:36:12] <leit1> when the signal comes true, but for that i would need the program executing and also waiting for the input signal to be true
[16:37:05] <BJT-Work> yep
[16:37:22] <leit1> do you think it's possible?
[16:38:18] <BJT-Work> so your wanting to do some motion then stop and wait for an input to come on then move to a predefined position?
[16:38:51] <leit1> no what i need is while the machine is machining also is waiting for the input signal
[16:39:25] <BJT-Work> perhaps you should look at using ClassicLadder for that
[16:39:26] <leit1> and when the signal comes true, the machine has to do something that i programed
[16:39:54] <BJT-Work> and stop the machining?
[16:40:05] <leit1> yes, and continue with other line of program
[16:40:42] <leit1> this is for measurement, because i need to measure while i'm machining the part
[16:41:55] <BJT-Work> so the user would press a "measure" button at any time and you want the tool to move to a predefined position while you measure then continue the program?
[16:43:15] <leit1> something like that, but the signal gets true automatically with a measurement system so when the part diameter reaches the wanted value the measurement unit sends the signal
[16:43:18] <leit1> to the emc
[16:43:33] <leit1> but this has to be with the axis moving, not in a wait mode
[16:44:07] <archivist> measurement probe interferes with the cutting tool, separate in gcode or something
[16:44:12] <BJT-Work> I don't know how to do that, but I don't know everything about EMC that I don't know yet
[16:45:56] <BJT-Work> are you using something like a laser measure for diameter or something like that?
[16:45:58] <leit1> archivist i didn't understand what you say please tell me again because my english is not so good
[16:46:12] <leit1> i'm using a measurement system with an encoder
[16:46:20] <leit1> is for diameter measures
[16:46:39] <leit1> like when you are machining a shaft
[16:46:46] <archivist> grinding machine?
[16:46:47] <leit1> is for cylindrical parts
[16:46:49] <leit1> yes
[16:47:10] <archivist> now I understand more
[16:47:24] <BJT-Work> so you have to compensate for the grinding wheel getting dressed?
[16:47:45] <archivist> contact measuring wheel rests on work
[16:48:23] <leit1> i see, so i need to wait or get into the emc code and try to make my own program
[16:48:34] <leit1> i mean i need to modify what i need to do that
[16:48:57] <leit1> BJT-work , for compensation and for get the exact diameter
[16:49:19] <archivist> I dont know enough about hal to say if it can interrupt the cut
[16:50:38] <leit1> ok no problem, and the other thing is if i can make an excentric shape using the interpolation ?
[16:50:38] <BJT-Work> leit1: I understand better now
[16:50:52] <archivist> plunge grind till input goes high
[16:51:40] <leit1> but this could be done with the m66 function?
[16:52:15] <leit1> because all the tests that i made resulted in the stopping of the program until the input comes true
[16:52:22] <BJT-Work> no, AFAIK you have to be on that line waiting for M66 to change states
[16:52:42] <leit1> sorry i don't know what is AFAIK
[16:52:55] <archivist> as far as I know
[16:52:57] <BJT-Work> as far as I knonw
[16:53:01] <leit1> but i understand what are you saying
[16:53:10] <leit1> thanks :)
[16:54:14] <leit1> ok no problem, the other thing, that i've only tried with the motors but separate from the machine is, if i can make complex forms with the syncronized motion like oval shapes and excentrics like the ones in the crankshafts?
[16:54:17] <BJT-Work> you might do something with Classicladder using halui.program.pause and mdi
[16:54:50] <BJT-Work> I would think you can but I don't know for sure
[16:55:14] <leit1> i could make an excentric shape (only testing) and it worked
[16:55:49] <leit1> i think it's possible since is the same pattern that i need for making a thread
[16:56:16] <leit1> i mean is feed per revolution
[16:56:53] <archivist> leit1, what is your language, sometimes people can help in your native language
[16:57:51] <leit1> spanish
[16:58:24] <leit1> sorry if my english is not quite correct sometimes, but i'm not used to write this fast
[16:59:43] <pjm_> archivist http://www.youtube.com/user/uhfdashsatcomdotcom - rigid tapping a lump of ally!
[17:01:07] <archivist> nice
[17:01:26] <pjm_> its getting there!
[17:01:38] <pjm_> I just need faster IO for the spindle encoder
[17:03:57] <leit1> well i have to go know, thanks again to you people like always you're very helpful and useful :) keep going with this i would help when i learn more about and i show you when i finish the machines that i'm working on
[17:04:13] <archivist> usually someone here
[17:04:53] <leit1> have a nice day we'll talk later :)
[17:06:25] <BJT-Work> ok keep us posted
[17:06:56] <leit1> i will goodbyyy
[17:06:59] <leit1> e
[17:07:56] <BJT-Work> pjm_ nice
[17:08:05] <pjm_> thanks
[17:08:21] <pjm_> i must get some spiral fluted taps tho
[17:08:39] <pjm_> the only ones i have in my tool box which are M3 are blimmin left hand ones! typical
[17:08:46] <BJT-Work> I still need to test out my thread mills
[17:09:40] <archivist> pjm_, that looked like a forward facing taper cut tap, just right for sheet metal
[17:09:56] <pjm_> yeah it was the other M3 one laying around ;-)
[17:10:00] <pjm_> i will get there
[17:10:16] <pjm_> anyway it was tapping a 8mm thick sheet of ally ;-)
[17:10:34] <cncjerry> hello, all, is there a developers doc that would point me to the layout of EMC2, axis specifically, as to where the files are distributed?
[17:10:35] <archivist> the swarf came out the bottom nicely curled
[17:11:33] <pjm_> i've asked me dad to find me some spiral fluted taps of various metric sizes though, so should have some more by monday
[17:12:37] <BJT-Work> cncjerry: I think so, one moment
[17:13:02] <BJT-Work> http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/directory.map?rev=1.28;content-type=text%2Fplain
[17:13:12] <BJT-Work> is this what your looking for?
[17:13:28] <cncjerry> bjt-work, thanks, wont know till i look at it :)
[17:14:21] <cncjerry> k, was looking for the axis.tcl file, found it in the directory specified in your link so that maps correctly.
[17:15:17] <cncjerry> I am just loving my new axis jog keyboard commands. my productivity went up about 500% with no keystroke errors.
[17:15:30] <cncjerry> I want to add to it and change the axis interface
[17:15:34] <BJT-Work> cool
[17:15:45] <BJT-Work> you running Trunk?
[17:15:52] <cncjerry> trunk?
[17:16:06] <BJT-Work> then no
[17:17:00] <cncjerry> bjt - don't know if you were here yesterday, but I made some simple mods to axis to allow for shifted and control key jogs like in mach 3 and I hear turbocnc.
[17:17:35] <BJT-Work> no, I missed that
[17:17:58] <BJT-Work> isn't open source wonderful...
[17:18:20] <cncjerry> I had used mach3 for about 8yrs and found it frustrating to have to hit i and C for jog modes. yes, open source is the way of the future - howard hughes
[17:38:32] <hugomatic> http://eclipse-night-club.com/ang/show.php?id=80_1&top=2
[17:39:42] <hugomatic> http://www.eclipse-night-club.com/calendrier/calendar/samplecalendar_ang.php?calview=event&event_id=105&prevview=month&year=2009&month=03&day=28
[19:13:02] <alex_joni> hi SWPLinux
[19:13:08] <SWPLinux> hi
[19:13:18] <alex_joni> travelling?
[19:13:22] <SWPLinux> yep
[19:13:28] <alex_joni> fun?
[19:13:31] <SWPLinux> all 4 time zones today :)
[19:13:38] <alex_joni> oh my :)
[19:13:40] <SWPLinux> (for the continental US anyway)
[19:13:42] <SWPLinux> yeah
[19:13:45] <cncjerry> hello, jepler, still here? are the latest axis scripts in axis historical on cvs?
[19:13:55] <SWPLinux> started in EST, went to Chicago, now I'm in Denver waiting for the flight to San Jose
[19:13:56] <alex_joni> latest?
[19:14:10] <alex_joni> cncjerry: axis historical is the stuff before AXIS got merged with emc2
[19:14:11] <SWPLinux> axis is part of the main emc2 source
[19:14:14] <alex_joni> about 2-3 years ago
[19:14:18] <alex_joni> so very old stuff
[19:15:34] <cncjerry> can't find axis.py in emc2 or the axis.tcl, looking again.
[19:15:49] <alex_joni> emc2/src/emc/usr_intf/axis
[19:15:51] <SWPLinux> src/emc/user_intf/axis
[19:15:57] <SWPLinux> right, usr_intf
[19:17:18] <cncjerry> I guess just making an axis directory was too difficult?
[19:17:34] <cncjerry> :)
[19:18:27] <alex_joni> cncjerry: making a directory where?
[19:19:41] <cncjerry> I don't know, somewhere glaringly obvious for dolts like me.
[19:19:58] <SWPLinux> heh
[19:20:28] <cncjerry> I've been just modifying the code on my machine and didn't have axis to it (access) and wanted a fresh version while working.
[19:21:12] <alex_joni> cncjerry: the issue was that there are about 5-6 other itnerfaces to use with emc2 besides AXIS, so sticking AXIS in the same place was kinda obvious
[19:22:20] <cncjerry> ok, now I am looking for the axis.tcl file which i would normally have put in the axis directory if I was setting this up.
[19:22:50] <alex_joni> axis.tcl is in a more obscure place :)
[19:23:03] <alex_joni> it's in emc2/share/axis/
[19:23:33] <cncjerry> is this open_source or hidden_source?
[19:23:53] <SWPLinux> greppable source
[19:26:32] <BJT-Work> grep is my friend...
[19:28:31] <alex_joni> and find
[19:28:49] <alex_joni> find . -name "*axis*"
[19:28:50] <goat-lab> greetings gents - does anyone have an opinion on the lumenlabs micro x-y table, and the options it comes with? I am curious to know if it's worth purchasing the upgraded spindles
[19:31:16] <goat-lab> http://www.lumenlab.com/store/robloks/microbotics/micro.html/
[19:31:57] <alex_joni> goat-lab: my only oppinion is that it "looks" ok
[19:33:30] <BJT-Work> goat-lab: IMHO the base plate "looks" floppy and thin... lots of UHMW (plastic)
[19:34:18] <goat-lab> yeah the use of plastic puts me off a little
[19:34:19] <alex_joni> goat-lab: it also depends a lot on what you plan to do with it
[19:34:32] <goat-lab> nothing too fine
[19:34:38] <alex_joni> if you're into plastic signs & such, then it's probably ok
[19:34:49] <alex_joni> I don't think it'll mill metals
[19:34:54] <MrSunshine> anyone know when chopping a stepper, do i have say 24V in all the time and 1A .. even tho the motor is rated 5V 1A ?
[19:35:20] <alex_joni> MrSunshine: actually the thing you need to worry about is the 1A
[19:35:28] <BJT-Work> maybe mercury...
[19:35:46] <alex_joni> if your stepper driver has current limiting, then it will only output 1A
[19:35:59] <alex_joni> which at the motors resistance will correspond to 5V
[19:36:38] <MrSunshine> alex_joni, ok :)
[19:36:48] <BJT-Work> goat-lab: they don't say much about it do they... can't tell from the pictures but it might have all thread leadscrews...
[19:37:03] <alex_joni> BJT-Work: and maybe some soft aluminum
[19:37:51] <BJT-Work> alex_joni: did you scroll all the way down? for $500 you get not much
[19:39:42] <alex_joni> BJT-Work: I saw it a couple days ago
[19:39:47] <alex_joni> I think jepler linked it
[19:40:52] <alex_joni> it might be ok for PCB routing or such
[19:42:14] <goat-lab> I guess I'd be more interested in using it with a an extruder of some sort for prototyping work, or carving
[19:43:26] <goat-lab> though I'd rather something with a bajigawatt laser that could cut metal :)
[19:45:45] <BJT-Work> * BJT-Work goes to work on his tractor
[19:45:52] <BJT-Work> talk to you guys later
[20:03:51] <motioncontrol> i'm study the xemc interface for curiosity.when i push the menĂ¹ file and select quit, emc open another windows "quit ".I don't understant this code location , not is in xemc.cc.wherw is the code for open the quit windows at menu file ?
[20:12:23] <FaxModem> hi all, I found something werid with stepconf. I am not sure if it is normal.
[20:13:36] <FaxModem> I assigned 2 and 3 to be x-axis step and direction, AND 8 and 9 to also be x-axis and direction (because I have 2 stepper motors driving the x axis), but only one of them move. Is that just for stepconf or this will also happen to the actual machine.
[20:14:00] <FaxModem> I could disable pin 2 and 3, then pin 8 and 9 will work.
[20:14:10] <FaxModem> so the hardward is OK... I think.
[20:14:13] <FaxModem> any suggestions?
[20:15:49] <pjm_> cant u just hook up both drivers to the same step/direction pins?
[20:16:47] <FaxModem> yes, which is what I have done now.
[20:17:34] <FaxModem> I just worry if the opto isolator (2 together) will draw a bit more current... which may cause problem on the parallel port.
[20:17:34] <FaxModem> but then again...
[20:17:44] <pjm_> i'm sure u could do what you want to do, but via manually editing the hal file
[20:17:52] <FaxModem> it will have to drive all opto isolator anyway. I hope paralleling two will not cause problem.
[20:18:12] <FaxModem> I just wonder why stepconf dose that.
[20:18:25] <FaxModem> you are right, I think it should work by manual editing.
[20:18:45] <FaxModem> I am just trying to make sure because stepconf is not doing it.
[20:18:51] <Optic> does emc2 have some logging features? Can it record stats of each job?
[20:19:19] <FaxModem> Optic: you mean me?
[20:19:24] <Optic> nah
[20:19:27] <Optic> just general question
[20:19:31] <FaxModem> ah.
[20:20:12] <FaxModem> it will show a log when it crash. ;p
[20:20:50] <Optic> maybe I could make some HAL connections that write to a log file
[20:21:11] <Optic> halui.program.is-running state changes could log a timestamp
[20:21:13] <Optic> or something
[20:21:24] <FaxModem> Optic: what are you trying to accomplish?
[20:21:43] <Optic> recording how much use the machine gets
[20:21:52] <FaxModem> oh
[20:21:56] <Optic> in order to schedule maintenance or billing
[20:22:30] <FaxModem> that would be a good feature to have.
[20:37:34] <FaxModem> I know people have used emc2 to perform 5-axis motion. I can't find the acutal "code" or configuration file that they developed. Have I been not looking at the right place?
[20:42:12] <goat-lab> goat-lab is now known as pub-toob
[20:42:43] <jepler> FaxModem: in emc2.3beta2 you can find the sample configuration "5axis". but I don't think there are any demo 5-axis ngc files
[20:44:23] <jepler> hmm I expected that configuration to show U and V axes but it doesn't
[20:44:34] <FaxModem> really! cool!
[20:44:46] <FaxModem> I am using the older version. I haven't updated it yet.
[20:45:03] <FaxModem> I think I will re-download the new version.
[20:45:45] <FaxModem> for some reason, using script to install the old version didn't give me stepconf. I had to install from LiveCD to get it.
[20:47:09] <FaxModem> jepler: where do I get the lastest emc2.3beta2?
[20:47:25] <FaxModem> 2.2.x is what's in the lastest LiveCD
[20:47:30] <jepler> FaxModem: hold on, I'll give you a link to the upgrade instructions
[20:47:39] <FaxModem> great! thanks!
[20:48:25] <jepler> announcement: http://article.gmane.org/gmane.linux.distributions.emc.devel/2050 upgrade instructions: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?UpdatingTo2.3
[20:48:48] <jepler> it looks like the sample configuration "max5kins" under "vismach" is a better example
[20:49:17] <FaxModem> OK. is that a "head-head" 5-axis?
[20:49:51] <FaxModem> I better go downaload Ubuntu 8.04 now.
[20:50:06] <jepler> i'm not sure of the nomenclature
[20:50:37] <FaxModem> the ones I found on YouTube with emc2, many are xyz + moving table.
[20:50:47] <FaxModem> that's xyz + table+table.
[20:50:50] <jepler> the simulated machine in max5kins looks like this: http://emergent.unpy.net/files/sandbox/max5kins.png
[20:51:17] <jepler> it's modeled after the machine that cut the wood sphere, if you saw that movie
[20:51:29] <FaxModem> ohhhh.
[20:51:47] <FaxModem> cool! thanks! I think I have lots to read on now!
[20:51:58] <FaxModem> thanks, jepler
[20:52:26] <jepler> the 5-axis stuff is cool but I only have a 3-axis machine myself
[20:53:09] <FaxModem> someone what's to "upgrade" to 5-axis by adding a 2-axis head... so I am intersted in that possibility
[20:53:11] <jepler> "stustev" is the fellow who is closest to doing real work with our 5-axis code
[20:53:40] <FaxModem> where can I find him? I mean just to look at what he has.
[20:53:40] <jepler> seems like the problem with adding 2 rotary axes in whatever configuration is using up a lot of Z travel to do it ..
[20:54:23] <FaxModem> well, isn't that the problem with the CAM software? ;p
[20:54:40] <archivist> travels get me on 4 axis
[20:54:47] <jepler> these are his videos: http://www.youtube.com/user/sws1253
[20:55:42] <FaxModem> ah! that's a head-head 5-axis.
[20:55:48] <FaxModem> that's what I want to learn about
[20:59:36] <FaxModem> duty call.
[20:59:37] <FaxModem> later.
[20:59:39] <FaxModem> thanks all.
[20:59:51] <jepler> FaxModem: well if you can travel to the midwest US in May, you have a perfect opportunity -- the guy who has that machine in his shop is hosting a meeting for emc2 people. http://article.gmane.org/gmane.linux.distributions.emc.user/13048 http://article.gmane.org/gmane.linux.distributions.emc.user/13049
[21:00:48] <jepler> hm, the way gmane shows his e-mail address is totally illegible. it's: stuart AT mpm1.com
[21:10:02] <pjm_> archivist , evening, btw u have one of the microstep drivers from Arc?
[21:10:12] <archivist> yup
[21:10:22] <pjm_> driving what, your 4th axis?
[21:10:30] <archivist> all 4
[21:10:39] <pjm_> i'm just looking at possible drivers for my 4th axis stepper
[21:13:57] <archivist> http://www.archivist.info/cnc/stageone/P1050048.JPG
[21:14:11] <archivist> while building
[21:14:51] <pjm_> nice yeah
[21:17:33] <archivist> testing and one failed http://www.archivist.info/cnc/stagetwo/P1190002.JPG
[21:18:03] <jepler> failed? how?
[21:18:37] <archivist> intermittent in the driver
[21:18:47] <pjm_> that is a pretty significant PSU, the lambda one
[21:18:59] <archivist> mass test on the table http://www.archivist.info/cnc/stagethree/P1260003.JPG
[21:20:38] <pjm_> nice, btw where did u get the stepper to A axis driver coupling? made it?
[21:27:50] <archivist> I erm "got" the A axis from someone I know and he got it at one of the shows
[21:28:30] <archivist> have not found the real supplier of that adapter
[21:29:07] <archivist> others are from solid ally from local scrap yard
[21:33:15] <alex_joni> good night all
[21:44:47] <jepler> http://www.kongregate.com/games/krispykrem/kohctpyktop-engineer-of-the-people
[22:08:26] <jepler> (I can barely believe I just spent half an hour playing that ..!)
[22:08:42] <cradek> didn't seem interesting to me...
[22:08:59] <jepler> yeah but you have an engineering degree
[22:09:18] <archivist> crashed my firefox flash
[22:09:23] <cradek> hmm.
[22:09:31] <jepler> no crash here (ff3, flash9)
[22:10:27] <archivist> firefox had been up for a week or two
[22:25:50] <pjm_> a cluster of helicals might do the job, one on AORW/E/IOR
[22:25:57] <pjm_> sorry wrong channel!
[22:29:48] <archivist> ! /me missing something
[23:14:00] <dareposte> hi all
[23:30:05] <toastydeath> Anybody looking for a ovely wordy, inexperienced machinist in the newark/wilmington DE area
[23:30:09] <toastydeath> *overly
[23:30:25] <toastydeath> and as i've just demonstrated, those words are in quantity, not necessarily quality.
[23:33:42] <dareposte> whats up toasty?
[23:33:48] <dareposte> hard day at work?
[23:43:07] <toastydeath> dareposte: nah
[23:43:14] <toastydeath> the company is slowly going to pot
[23:43:22] <dareposte> oh
[23:43:23] <toastydeath> so i figure i might as well start looking
[23:43:33] <dareposte> i've been looking too
[23:43:33] <toastydeath> i've survived 5 rounds of layoffs so far
[23:43:37] <dareposte> got a couple offers
[23:43:39] <toastydeath> hot
[23:43:40] <dareposte> good work
[23:43:54] <dareposte> 5 rounds is a lot
[23:44:16] <toastydeath> 4yeh
[23:44:26] <toastydeath> we are on minimal staffing now
[23:47:53] <dareposte> that sucks on a couple levels
[23:51:47] <dareposte> i managed to not get fired today
[23:51:51] <dareposte> which is good
[23:52:24] <L84Supper> anyone know the highest precision application (sub micron, nanoscale accuracy etc) that EMC has been used for?
[23:55:01] <archivist> its easily capable of fine control but can you keep the machine in a temperature controlled environment
[23:55:28] <toastydeath> dareposte: that is good
[23:56:30] <toastydeath> l84supper:
[23:56:40] <L84Supper> archivist : was just thinking about PR for EMC, yeah, the nanometer stages we use are kept within a deg C
[23:56:52] <toastydeath> that's a complicated question, i don't know of a specific machine, but it depends on how you want to do feedback
[23:57:34] <toastydeath> using a second scale/interferometer is only just now being worked on and should be done soon (iirc?)
[23:58:10] <toastydeath> if your primary positioning/feedback control is accurate enough without a secondary scale, then it'll be as accurate as that
[23:59:06] <L84Supper> I'm looking at EMC for mods to 6-7 axis robots as well
[23:59:49] <toastydeath> i don't know a thing about multiaxis robots, plenty of other people here do though
[23:59:56] <dareposte> i work on them quite a bit