#emc | Logs for 2006-03-22

[00:00:39] <cradek> fenn: jellovision (Xgl) obviously
[00:03:11] <fenn> doesn't sound like a bad idea..
[00:03:37] <fenn> although i bet overall reduction in bloat would help a lot more
[00:04:51] <fenn> gl wasnt really designed as a general purpose graphics display though was it?
[00:05:16] <fenn> i notice blender uses all opengl windows and menus
[00:05:27] <cradek> does that really matter? today, it does a lot of things fast and in hardware
[00:06:03] <fenn> Fast, simple screen rotation/flipping/panning/magnification. thank god!
[00:06:15] <cradek> yeah
[00:06:38] <cradek> I'm worried, though, that it takes us one step closer to not being able to use realtime on our desktop machines
[00:07:04] <cradek> today Xgl relies on the closed-source nvidia driver, which is not realtime compatible
[00:07:25] <fenn> how can they still be keeping that secret? it's been like 5 years
[00:07:26] <bill2or3> you're running emc on your desktop?
[00:07:46] <cradek> sure
[00:08:01] <cradek> and I also run it on a machine hooked to the mill, that's the same as a desktop-type machine
[00:08:26] <bill2or3> ahh. so the desktop isnt actually hooked to the mill?
[00:08:34] <bill2or3> for a minute I imagined a mill in your office.
[00:08:43] <fenn> i'm still waiting for a decent VR interface
[00:08:45] <cradek> you're not getting my point - I mean ordinary user-class everyday hardware
[00:08:59] <bill2or3> yeah, I understand that part.
[00:09:02] <cradek> desktop hardware: PCs, graphics cards, etc
[00:10:45] <fenn> of course then everything would be so fast that it would be impossible to run normal programs on a non-accelerated box
[00:10:56] <cradek> exactly
[00:10:58] <fenn> because the bloat would expand exponentially
[00:11:13] <bill2or3> uhm. I think that allready happened.
[00:11:37] <cradek> well, not necessarily - as soon as apps use gl instead of XMoveRectangle, they're going to be slower since you have to use software GL to translate GL into X
[00:11:52] <fenn> only for screensavers and video games and 3d programs
[00:12:13] <bill2or3> * bill2or3 goes home, finally.
[00:12:40] <fenn> doesnt xgl translate x commands into gl commands?
[00:12:45] <cradek> well anyway, don't throw out your old hardware (or OS discs)
[00:13:01] <cradek> fenn: I suppose it does
[00:13:02] <fenn> * fenn looks over at the pile of rotting computers
[00:13:10] <cradek> haha me too
[00:13:17] <cradek> I'm on one of them now
[00:13:41] <cradek> :q
[00:13:42] <fenn> unfortunately i forgot which ones were supposed to work and which were for parts
[00:13:43] <cradek> oops
[00:13:55] <fenn> yucky vi :)
[00:14:05] <cradek> I write on parts with a sharpie "good" and the date
[00:14:15] <cradek> ... if I take them out of service, working
[00:14:42] <cradek> saves me from "I wonder which of these hard drives I should try" a year later
[00:16:03] <fenn> do you know what the purpose of a "cpu board" is?
[00:16:09] <fenn> like an isa card
[00:16:44] <cradek> some machines have/had plug-in CPU cards and simple backplanes
[00:16:54] <cradek> oddball server-class stuff usually
[00:17:24] <fenn> i got a big box of motherboards donated to me
[00:17:38] <fenn> trying to at least figure out what it all is
[00:17:56] <cradek> long ago I had a 286-on-a-card that plugged into my XT
[00:18:08] <cradek> it also had a ribbon cable that plugged into the old processor socket
[00:18:29] <fenn> that would never work today
[00:18:48] <cradek> and why bother? motherboards are $40
[00:19:06] <fenn> ah yes i forgot.. "why bother"
[00:57:05] <giacus> night
[01:42:04] <tfmacz> su - repeater
[01:52:29] <tfmacz> oops
[03:12:52] <SWP_Away> SWP_Away is now known as SWPadnos
[03:38:43] <staggerlytom> hello
[03:39:41] <staggerlytom> just setup ubuntu & emc2 with axis, all went well but got q about irc client
[03:40:40] <staggerlytom> it couldnt connect to irc.freenode.net, it wanted to connect to that buit to 'newsserver/6667' .. normal?
[03:41:32] <cradek> I think by default xchat connects to freenode
[03:41:57] <staggerlytom> it wouldnt connect, saying it couldnt connect to 'newsserver/6667'
[03:42:08] <cradek> did you run applications/internet/xchat irc?
[03:42:08] <staggerlytom> inet was fine
[03:42:25] <staggerlytom> yes, that was the menu sequence
[03:42:49] <cradek> then you scrolled down to freenode?
[03:42:58] <staggerlytom> i'm back on magma now
[03:43:23] <staggerlytom> no, I did not see irc.freenode.nety, so maybe it was freenode?
[03:43:37] <cradek> I just tried it, and it worked
[03:44:18] <staggerlytom> ok, not irc.freenode.net ... start with f, thanx cradek
[03:44:29] <cradek> welcome
[03:44:35] <cradek> ?
[03:44:42] <cradek> none of them start with irc....
[03:47:35] <tomp_> cradek: thanx, I'm just too literal dangit, works fine :-)
[03:47:42] <cradek> haha
[04:10:24] <bpmw> Evening folks! cradek you there?
[04:13:07] <cradek> yes but just about on my way to bed
[04:15:52] <bpmw> A quick question?
[04:15:57] <cradek> sure
[04:16:27] <bpmw> What was the name of the guy that does the mill/drill conversions?
[04:17:00] <cradek> umm, I'm not sure
[04:17:07] <cradek> are you still looking for help with your encoder thing?
[04:17:10] <jmkasunich> roland?
[04:17:18] <bpmw> Yes.
[04:17:29] <cradek> heh, I was just going to ask if you talked to john k
[04:17:40] <bpmw> I think Roland was it.
[04:18:25] <jmkasunich> google "CNC workshop roland"
[04:18:28] <bpmw> cradek yes, It didnt go so well. jmk didnt think he could help me.
[04:18:49] <cradek> bpmw: is that code language for he thinks it's a bad idea?
[04:19:01] <bpmw> LOL
[04:19:06] <cradek> (because he can help with just about anything)
[04:19:33] <cradek> sorry, that was just a little feeling I had
[04:19:54] <jmkasunich> http://www.cnc-workshop.com/
[04:19:59] <bpmw> I think I'll have to bite the bullet on this one and try the ballscrew thing...
[04:20:19] <jmkasunich> bpmw: if you are anywhere near Illinois, you should go to the workshop
[04:20:30] <cradek> no kidding
[04:20:45] <bpmw> Sorry I'm way up in Canada.
[04:20:57] <cradek> some of canada is "near" illinois
[04:21:16] <jmkasunich> I'm driving about 9 hours for it
[04:21:22] <bpmw> not the part i'm in.
[04:21:29] <cradek> I'm lucky and only driving 6-7
[04:21:54] <bpmw> I'd love to go, but no time or money :)
[04:22:18] <cradek> me too, but I'm going anyway :-)
[04:22:37] <bpmw> It would be nice to actually meet you guys.
[04:22:54] <cradek> it'll be a first for me too - I've never met any of these folks
[04:23:03] <bpmw> Another time perhaps!
[04:24:07] <cradek> I'm actually an 8 year old black lab
[04:24:29] <SWPadnos> sit
[04:24:37] <cradek> already sitting
[04:24:41] <bpmw> I Live about 9.5 hours straight north of Minniapolis.
[04:24:49] <SWPadnos> good dog
[04:25:02] <cradek> everyone has seen this right? http://www.unc.edu/depts/jomc/academics/dri/idog.jpg
[04:25:16] <SWPadnos> heh
[04:25:44] <cradek> wow, july 1993
[04:26:13] <cradek> damn
[04:27:26] <cradek> bpmw: minneapolis to fest would be another 6 hours
[04:27:39] <cradek> ... of boredom (iowa)
[04:27:50] <bpmw> Thanks, jmk the link has a contact for for Roland. I'll drop him a note about taking out lash from quill.
[04:28:40] <bpmw> or I might just do the ball screw thing!
[04:28:56] <cradek> "when in doubt, do it right"
[04:29:28] <bpmw> If i do the ballscrew thing what size stepper would you reccomend?
[04:29:56] <cradek> depends on the screw pitch and the mass it's moving
[04:30:28] <bpmw> ok, so I should go 1200 or so OZ/in
[04:30:30] <cradek> I bet you can find app notes of some kind from the ballscrew manufacturer
[04:30:42] <SWPadnos> those mills are about 1/2 the size of a bridgeport, right?
[04:30:42] <cradek> I have really no idea, sorry
[04:31:22] <SWPadnos> and a lot of what's missing is in the base and back side, so the table and head are similar in size to a BP (from what I've seen)
[04:31:29] <bpmw> swp, not sure on that one, it weight 850 lbs.
[04:31:42] <SWPadnos> that is about 1/2 a BP - they're about 1900-2200 pounds
[04:31:54] <bpmw> Ok.
[04:31:54] <SWPadnos> what size is the table?
[04:32:18] <SWPadnos> it looked like maybe 8x36 or so
[04:32:47] <bpmw> travel 12x20 table size 10x 28
[04:33:03] <cradek> http://money.cnn.com/2006/03/21/news/companies/microsoft.reut/index.htm?cnn=yes
[04:33:12] <SWPadnos> ok. let's assume that the quill weighs as much as on a BP
[04:33:17] <cradek> I know nobody cares about the story, but what's the deal with that picture?
[04:33:54] <SWPadnos> they're pointing at the vista you'll have ... ??
[04:34:16] <SWPadnos> my Z drive is a 28 in-lb servo, with a 5TPI ballscrew (and the clockspring at "normal" tension)
[04:35:18] <SWPadnos> since servos have reserve, the 1100 oz-in may be about right (the servo is 448 continuous, ~1800 peak)
[04:36:15] <cradek> SWPadnos: does 5tpi self-lock?
[04:36:21] <bpmw> I think I would go 10 tpi and keep things simple (both othe screws are 10 tpi.)
[04:36:25] <SWPadnos> I don't think so
[04:36:54] <cradek> bpmw: no reason to keep them the same, emc doesn't care
[04:37:01] <SWPadnos> 10TPI may, with the smaller weight of a quill, but a brake is a good idea (though the detent torque of a stepper may possibly be enough)
[04:37:22] <cradek> yeah if a stepper can move it, it can easily hold it
[04:37:39] <SWPadnos> there's more torque under power than unpowered
[04:37:58] <cradek> oh you mean when the machine is off? good point.
[04:38:02] <SWPadnos> yep
[04:38:03] <cradek> I meant when it's on, but not moving
[04:38:12] <SWPadnos> right - that should be fine
[04:38:13] <cradek> it would suck to have it fall when you turn it off
[04:38:16] <bpmw> The 1280 oz/in steppers I bought from Nylor were quite reasonable around 300.00
[04:38:32] <SWPadnos> that should be fine, I think
[04:38:33] <cradek> what do you use to drive a stepper like that?
[04:38:39] <SWPadnos> gecko
[04:38:51] <bpmw> Geckos
[04:39:09] <cradek> wow, there's a lot of drive power in that little box
[04:39:20] <bpmw> I have 1280 oz/in on x and y now.
[04:39:39] <SWPadnos> yeah, 80V at 7A/phase
[04:39:49] <SWPadnos> even more for the servos, 80V at 20A continuous
[04:40:13] <mshaver> cradek: I talked with john k tonight & he said I ought to ship you the Sherline lathe with the spindle encoder before the cnc workshop
[04:40:16] <bpmw> SWP. 80v at 1.4 A.
[04:40:36] <SWPadnos> the gecko can do 80V at 7A + 7A
[04:40:46] <cradek> mshaver: I'm making real progress on threading and it would be a big help to be able to cut
[04:40:59] <bpmw> Sorry I thought you were talking about steppers.
[04:41:09] <SWPadnos> nope, the drives ;)
[04:41:18] <mshaver> i'll see what I can do; send me your mailing address (mshaver@erols.com)
[04:41:24] <bpmw> Got it.
[04:41:36] <mshaver> i'll call fred at nist tomorrow
[04:41:58] <cradek> I appreciate it
[04:42:02] <cradek> you'll be at fest right?
[04:42:10] <mshaver> no problem
[04:42:26] <mshaver> yep, coming with steve stallings
[04:42:41] <cradek> have you used cnc lathes before?
[04:42:51] <bpmw> Ok guys thanks for the link I'll look into info from Roland. ANyone want to buy a NINE NEW LINEAR ENCODER :)
[04:42:54] <mshaver> nope, just mills
[04:43:13] <SWPadnos> what length? :)
[04:43:29] <cradek> same here... I feel like I'm shooting in the dark a bit
[04:43:37] <bpmw> Rated 2" but will do 5 easy.
[04:43:52] <mshaver> i could try to copy some operator manuals from commercial machines
[04:44:04] <cradek> that might be a big help too
[04:44:18] <cradek> ray has started a wiki page with some information - be sure you don't duplicate any effort
[04:44:31] <SWPadnos> do you think fred might have access to some programs that do threading?
[04:44:53] <mshaver> yep, i was just going to say that ray might have some info
[04:45:10] <SWPadnos> err- G-code programs, that is
[04:45:33] <cradek> instead of trying to do everything certain commercial machines do, I think we need to know what's commonly in use (what comes out of cam packages for instance).
[04:45:43] <bpmw> Thanks again guys for the info, much appreciated! good night.
[04:45:49] <mshaver> i'll have to ask fred, but i don't think they ever got that far
[04:45:52] <SWPadnos> see you
[04:45:54] <cradek> we need that basic functionality first
[04:46:16] <SWPadnos> I was thinking of stuff they may have done on the CNC lathes they have, for other reasons
[04:46:41] <mshaver> i'll ask - they have some in the shop at nist
[04:46:52] <SWPadnos> so I saw ;) (really big ones)
[04:47:02] <mshaver> in fact, i can ask at another shop that has a bunch of cnc lathes
[04:47:17] <mshaver> a real actual working shop
[04:47:22] <SWPadnos> cool. I think some samples of typical G-code would be helpful
[04:47:29] <SWPadnos> they have work? :)
[04:47:33] <cradek> it would be great if you could ask a couple guys who use them all the time what codes they use
[04:47:56] <mshaver> i can call big bob at edk machine, he's the programmer
[04:48:06] <cradek> heh "big bob"
[04:48:30] <mshaver> there was a little bob too, but he died a couple years ago
[04:48:35] <cradek> we have programmers, I think we need people who use lathes all the time
[04:49:00] <mshaver> no, i mean he's the g-code programmer for the lathes
[04:49:13] <mshaver> he sets up the jobs
[04:49:14] <cradek> oh sorry, I'm dense
[04:49:16] <cradek> ok
[04:49:22] <cradek> yeah him
[04:49:59] <mshaver> i'll look into this stuff tomorrow & get back to you here on irc
[04:50:06] <cradek> that's great, thank you
[04:50:13] <mshaver> for now, it's bed time...
[04:50:14] <cradek> I sent my address too
[04:50:18] <cradek> me too, goodnight
[04:50:23] <mshaver> thanks!
[04:50:31] <SWPadnos> argh - me three
[04:50:35] <SWPadnos> see you later, Matt
[04:53:20] <SWPadnos> oh yeah, and cradek (how rude of me)
[04:53:28] <SWPadnos> SWPadnos is now known as SWP_Away
[08:49:00] <Jymmm> Anyone know what a SMT regulator can handle current wise?
[08:51:20] <anonimasu> that should be spec:ed in the datasheets
[08:51:48] <Jymmm> I mena most of them (never touched SMT in my life)
[08:52:05] <anonimasu> that depends entirely on your part
[08:52:06] <anonimasu> :)
[08:52:24] <Jymmm> like 2n2222 sized VR would handle around 100ma
[10:51:26] <anonimasu> well, check for datasheets :)
[12:33:09] <giacus> giacus is now known as giacus_afk
[15:04:15] <alex_joni> hi bob
[15:14:50] <anonimasu> *aywn*
[15:16:02] <alex_joni> *aywn*
[15:16:25] <SWP_Away> *ywna*
[15:16:29] <anonimasu> :)
[15:24:20] <tomp_> hello all
[15:24:50] <rayh> Hi tom
[15:24:56] <tomp_> the wiki about halui refers to emc2/src/emc/usr_intf/halui, but it's not included w std install, how to get ?
[15:25:18] <cradek> tomp_: it's not at all done
[15:25:43] <tomp_> wanted to read about it, seems more done than crapahaloc
[15:25:54] <tomp_> :-)
[15:27:11] <rayh> If you want to go further into the issues read the comments at the top of emcsh.
[15:27:30] <SWP_Away> SWP_Away is now known as SWPadnos
[15:32:48] <tomp_> ray: when you get time: emcsh is a binary, and again where is the src ? ( same q new wrapper )
[15:33:04] <alex_joni> tomp_: emcsh is in emc2/src/emc/usr_intf/emcsh.cc
[15:33:23] <alex_joni> halui.cc is in the same place, but at the moment only the comments on top of the file are ok
[15:35:34] <tomp_> alex_joni: please what is the prefix on this Ubunto system? the stuff before .....emc2/blah?
[15:36:00] <alex_joni> if you talk about ubuntu and debs, you probably don't have the sources
[15:36:23] <tomp_> alex: yes, that was my 1st q, do I need them & where do I get them?
[15:36:32] <jepler> you can get the source with 'apt-get source emc2'; that will create a directory named emc2-<something> in the current directory
[15:36:33] <alex_joni> simplest way is to look at: http://cvs.sourceforge.net/viewcvs.py/emc/emc2/src/emc/usr_intf/
[15:36:49] <tomp_> thanx alex
[15:36:59] <jepler> the source you get with 'apt-get source' is going to be older than what you see in CVS
[15:37:13] <alex_joni> tomp_: you don't NEED them ;) only if you want to rebuild emc2, but it is ok to read it for reference
[15:37:31] <alex_joni> http://cvs.sourceforge.net/viewcvs.py/emc/emc2/src/emc/usr_intf/?only_with_tag=TESTING <- that's exactly what is in the debs
[15:38:05] <alex_joni> tomp_: but I must warn you that CVS is sometimes icky with the viewcvs interface
[15:38:56] <tomp_> and maybe the regular anonymous cvs command will do it? ( and what is that with TESTING )?
[15:39:31] <alex_joni> if anonymous cvs command works, then the viewcvs stuff usually works too (they are closely related)
[15:39:49] <rayh> For anyone at tom's level of interest, I recommend that he run apt-get build-dep emc2
[15:39:51] <tomp_> Alex: have a nice lunch & I'll browse cvs over http, thanx
[15:40:04] <alex_joni> you can read more about TESTING here: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?Testing
[15:40:34] <rayh> and start compiling from sf directly.
[15:40:46] <alex_joni> tomp_: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?EmcVersion2_Release_Status
[15:41:06] <alex_joni> gone home
[15:41:07] <alex_joni> later guys
[15:41:21] <tomp_> thanx Ray ( bugger in oscillator pic chip code last night, maybe bad chip as hdwr checked out ok.. pgmbl on/off/peak via EMC )
[15:42:27] <SWPadnos> pgmbl?
[15:42:52] <tomp_> programmble (gui & mcode i hope)
[15:43:37] <tomp_> just 3 bytes predestined to ports for oscillator and current selector
[15:43:41] <rayh> That's got the cat curious
[15:43:43] <SWPadnos> ah. what is it you're making?
[15:43:48] <tomp_> edm
[15:43:53] <SWPadnos> oh, cool
[15:44:04] <tomp_> new ways to make smoke
[15:44:15] <SWPadnos> mmmm - micron particle smoke
[15:44:56] <tomp_> any electronic guys, pleeze look at schema & pcbs on wiki, need jury
[15:45:16] <SWPadnos> I can help with the code, but maybe not the schematics ;)
[15:45:49] <jepler> tomp_: what schematics?
[15:46:00] <tomp_> the pic is a hdwr issue, an output wouldnt drive 2 latches holding on & off, I think the chip is bad
[15:46:19] <tomp_> theres a poswer supply (edm) and an oscillator and a gap input section
[15:46:23] <tomp_> power
[15:46:44] <SWPadnos> you're sure the port is set for output?
[15:48:09] <tomp_> pretty sure, PortB-2 , it's a working deisgn, I just pulled an old chip off a board ( doh, mayb i got the board cuz it's bad :-(
[15:48:17] <SWPadnos> heh
[15:48:39] <SWPadnos> where are the schematics?
[15:49:50] <SWPadnos> what format are they in?
[15:49:52] <tomp_> 16c54 , else I'd just burn a 16c84... schema bear bottom of page in blue , eagle format, one library for DS0026 chip
[15:50:01] <SWPadnos> hmmm
[15:50:04] <tomp_> the free vrsn of eagle
[15:50:09] <tomp_> for linux
[15:50:27] <SWPadnos> ok. I think I uninstalled it when I bought Altium ;)
[15:50:46] <jepler> oh, this? upload:osc.sch
[15:50:49] <SWPadnos> nope - I'm not that organized
[15:50:59] <tomp_> the pcb of the power supply is also eagle but the schema is Xcircuit and the website for Scott Ananian is on the page too
[15:51:19] <tomp_> he designed the ps for an MIT course
[15:51:41] <tomp_> whacking big igbit (60A 300Vdc)
[15:52:08] <SWPadnos> can eagle import or export anything?
[15:52:18] <SWPadnos> (other than its own format)
[15:52:37] <tomp_> orcad i think, I didnt move it to Ubunto yet, so cant check
[15:53:01] <cradek> it can plot in postscript
[15:53:16] <tomp_> i can post as ps if it'll help
[15:53:34] <SWPadnos> I've got Eagle 4.13r1 installed still
[15:53:57] <tomp_> dunno vrsn, just got it 2 weks ago off net
[15:54:22] <SWPadnos> it seems to load, but warns me:
[15:54:34] <SWPadnos> "No forward/back annotation will be performed!"
[15:54:37] <SWPadnos> well!
[15:54:49] <cradek> you have to load the schematic AND board for that to work
[15:55:03] <jepler> SWPadnos: that means there's no layout to go with the schematic, or there's a board but it doesn't match
[15:55:05] <SWPadnos> not that I was planning on it, you see :)
[15:55:14] <tomp_> dont worry, none are both schema and pcb
[15:56:07] <tomp_> there's no pcb for the 2 schem, there no schema for the 1 pcb(as above)
[15:56:21] <SWPadnos> portb should be an input, not output
[15:56:56] <SWPadnos> with RA0 selecting which latch to read
[15:57:13] <tomp_> uhh, checking... RA1 is what my prob is, and is output
[15:58:15] <tomp_> and it drives an 04 gate, the non inverted signal to one chipselect of one 574, the inverted to another 574
[15:58:35] <SWPadnos> that's RA0 on the schematic I'm looking at
[15:58:44] <tomp_> so it gets data to a mini bus to the pic (on then off then on then off )
[15:58:59] <tomp_> your schem show RA0 thru Ra3, and i used RA1
[15:59:27] <SWPadnos> my schematic shows RA0 connected to the '04
[15:59:46] <SWPadnos> RA1 is labelled as Trio, but I don't see that connected anywhere else
[16:00:13] <tomp_> oh oh you're looking at >my< schema, doh! and I cant look cuz new UBUNTU, case closed till i catch up (yes RA0 to gates)
[16:00:22] <SWPadnos> heh
[16:00:39] <SWPadnos> yep - the schematic from the wiki
[16:01:06] <tomp_> that scheam also is unfinished as I knew i was moving machine to machine, and just wanted to freeze it someplace, ok
[16:01:20] <SWPadnos> is this on an etched PCB, or proto board?
[16:01:32] <tomp_> w
[16:01:40] <tomp_> wire wrapped :-)
[16:01:44] <SWPadnos> ok :)
[16:02:03] <SWPadnos> it looks like IC4A is unused - is that still true?
[16:02:31] <tomp_> cant tell, no eagle here
[16:02:35] <tomp_> yet
[16:02:52] <SWPadnos> hmmm - probably not, it's placed near a transistor and LED, like gate B
[16:03:43] <tomp_> the 2 buttons hadnt been connected yet ( needed a bit o dbounz ) and the were the clock to the latch
[16:04:12] <SWPadnos> ok, but you put that under software control
[16:04:31] <tomp_> once latched, the pic just constantly read the 2 data values, ran 2 timer loops, and output 2 osc signals to DS0026 at 12V
[16:04:49] <tomp_> the data setting is asynch
[16:04:58] <SWPadnos> right
[16:05:30] <SWPadnos> you could use a single 2:1 mux chip as well (but that's loads harder at this point, with wire-wrap and all ;) )
[16:05:31] <tomp_> the idea is that the entire top portion can be whacked off and connected to 2 8 bit ports under emc later
[16:05:48] <SWPadnos> right
[16:05:52] <tomp_> 2:1 mux chip?
[16:06:04] <SWPadnos> yeah - lemme look that up
[16:06:26] <SWPadnos> has a set of inputs A B, and outputs Q, one bit selects which input goes to the outputs
[16:06:40] <SWPadnos> one control bit selects ...
[16:06:58] <tomp_> yeh yeh, i seen them lots, 74????
[16:07:06] <SWPadnos> 157 is a 4-bit
[16:08:03] <SWPadnos> 251 is octal
[16:08:12] <SWPadnos> hmmm - wait
[16:08:19] <SWPadnos> that can't be true, it's only got 16 pins ;)
[16:08:20] <tomp_> 'data-selector' 'bus-selector'? gotta get out Don Lancaster :-) ok, ls251 thanks
[16:10:00] <tomp_> oh, 74150 says Dan
[16:10:49] <tomp_> pretty old stuff, hey i could use a pic to do that :-)
[16:11:01] <SWPadnos> hmm - it would need 32 pins or so (24 for I/O, 2 power, plus a couple for control)
[16:11:03] <SWPadnos> heh
[16:11:26] <SWPadnos> actually, don't laugh. a slightly larger micro would eliminate those two chips
[16:12:30] <tomp_> thanx, yeh maybe a good time to look at atmega... rabbit...
[16:12:38] <SWPadnos> atmega, no rabbit
[16:12:52] <SWPadnos> (in my opinion)
[16:13:12] <cradek> you can get atmega16/atmega32 in dip40
[16:13:36] <SWPadnos> 28-dip, for some of them (the tiny, I think)
[16:13:41] <cradek> they have four full 8-bit ports
[16:13:59] <cradek> if you want tiny, you can get 8 pin avrs
[16:14:02] <tomp_> ok, will look there, but for now can save some time if i fix it rather than ( i used to use 68hc11's in FORTH :-)
[16:14:12] <SWPadnos> eeeewwwww
[16:14:13] <SWPadnos> :)
[16:14:45] <tomp_> forth is very good for machine control ( big telescopes and big numbers )
[16:15:03] <tomp_> again, just opinion
[16:15:11] <SWPadnos> heh. I remember when forth was supposed to be "the next big thing"
[16:15:33] <SWPadnos> there were forth chips and everything
[16:15:40] <SWPadnos> kinda like java now
[16:15:48] <tomp_> there's too many 'member whenz' in computing
[16:16:08] <SWPadnos> true enough (remember the transputer?)
[16:16:42] <skunkworks> Quick question regarding servos - we have a big mill that we could put over a ton on the table. Would the pid have to be retuned for higher weights?
[16:16:57] <cradek> yes
[16:17:07] <SWPadnos> how heavy is the knee / saddle / table?
[16:17:40] <skunkworks> the pallets weigh a ton or there about (would have to look again)
[16:18:59] <skunkworks> talking about this machine again http://www.electronicsam.com/images/KandT/DSCCurrent.JPG
[16:19:52] <SWPadnos> that's a lotta toolchanger positions :O
[16:19:58] <skunkworks> 62
[16:20:07] <tomp_> tune for the heaviest mass, is that pallet a ton?
[16:20:15] <skunkworks> so do people have differnt ini then?
[16:20:33] <skunkworks> tom - I would have to look but I am pretty sure it is close to that
[16:21:11] <skunkworks> random postion tool changer ;) (bar code rings)
[16:23:21] <Roguish> Excurse me guys, SWPadnos have you heard from peterv concerning the fpga programming for the m5i20?
[16:23:52] <SWPadnos> yes, he sent me some other info, and I haven't had a chance to massage it into the right format yet
[16:24:25] <SWPadnos> all the source is in cvs, but there's a timing file that I don't think is there (which I'll need to ask Pete at Mesa for)
[16:25:05] <Roguish> how 'bout a schematic file?
[16:25:58] <SWPadnos> nope. it's all in VHDL text, one file instantiates the "devices" in the other files
[16:26:21] <SWPadnos> there isn't a "project" file, but all the source files can be added to a project in whatever tools you would use
[16:27:12] <Roguish> i wonder if there is any way to synthsize a schematic from the vhdl.
[16:27:32] <SWPadnos> a schematic for what program?
[16:29:00] <tomp_> thanks guys, old fart off to find old part
[16:29:08] <Roguish> well you can layout the logics in the fpga via a schematic. kinda drag and drop of components and connecting the pins, etc.
[16:29:31] <SWPadnos> yes, but the "schematic library" format is different for every program
[16:30:09] <SWPadnos> I'll be trying to make a schematic in Altium, though that'll be useless for any Linux program
[16:30:31] <SWPadnos> I may be able to use the Xilinx ISE software, which would probably make files that can be used with the Linux version
[16:30:33] <Roguish> Zilinx's freeware is pretty good, and affordable.
[16:30:43] <SWPadnos> very affordable ;)
[16:32:39] <rayh> skunkworks, It would be possible to set up a bunch of load-ton.hal, load-halfton.hal
[16:32:55] <rayh> each one would simply have a bunch of setp commands for axis tuning.
[16:33:05] <Roguish> my Zilinx expert friend is off to Malaysia again for a week to support a production line (see www.OQO.com). when he returns i hope to get a bit of assistance.
[16:33:10] <rayh> Start the machine with the noload.hal
[16:33:21] <SWPadnos> cool. see if he can bring back samples ;)
[16:33:32] <rayh> and call the others as you add load to the table.
[16:33:46] <bill2or3> those oqo's are sort of neat.
[16:34:11] <rayh> Not quite dynamic tuning but better than starting, editing, running.
[16:34:14] <SWPadnos> rayh, maybe there can be a script (accessible from GUI menus) to load alternate tuning parameters
[16:34:25] <SWPadnos> which were saved with your tuning tool ;)
[16:34:45] <rayh> You bet.
[16:34:56] <SWPadnos> a save as ... button, that just saves setp commands to a file would be good
[16:35:08] <SWPadnos> rather than putting the values back in the ini
[16:35:21] <rayh> Now you're getting personal;
[16:35:26] <SWPadnos> heh
[16:35:47] <SWPadnos> I can't really help with the TCL, but I can increase your workload
[16:35:56] <SWPadnos> I'm good that way ;)
[16:36:23] <rayh> np
[16:37:58] <SWPadnos> though I might be able to add that function, if all the information is already there
[16:43:08] <anonimasu> :)
[16:45:37] <rayh> SWPadnos, Sure a script could be written and put into the emc2/tcl/scripts directory.
[16:45:55] <rayh> Make it executable with the .tcl extension and it will show up in tkemc's menu.
[16:46:02] <SWPadnos> yep
[16:46:29] <SWPadnos> actually, it might make sense to put a load... button into your program as well, then you don't need a new script ;)
[16:46:55] <SWPadnos> though it's a one-liner anyway "halcmd -f the_new_pid.hal"
[16:53:15] <skunkworks> thanks ray - this is a way off - still need drives. among other things :)
[16:53:55] <skunkworks> (was hydrolic)
[16:57:04] <SWPadnos> hey - I just noticed something at the bottom of a wiki page:
[16:57:13] <SWPadnos> "Warning: Database is stored in temporary directory /tmp/persistent/e/em/emc/wiki/wiki"
[16:57:33] <SWPadnos> considering that sourceforge has been flaky lately, is this a bad thing?
[16:59:24] <cradek> SWPadnos: that's a consequence of sf's strange setup - but a backup is made regularly by cron
[16:59:36] <SWPadnos> ok - /tmp just worries me
[17:00:33] <cradek> the only filesystem where the webserver can write is /tmp
[17:00:42] <SWPadnos> ah, ok
[17:00:45] <cradek> (they really want you to use sql instead)
[17:01:05] <SWPadnos> that would be better, assuming that their sql servers are reliable
[17:01:39] <cradek> I don't know what to trust there
[17:02:01] <SWPadnos> heh, it's a tough one, without access to the admin procedures
[17:03:34] <jepler> maybe the warning should be disabled in the wiki software, if we know about it but don't intend to change it
[17:03:48] <cradek> I'll do that
[17:07:30] <cradek> done
[17:24:42] <anonimasu> ,,,
[17:50:50] <giacus> hello
[17:51:09] <giacus> SWPadnos: I bought a laser level ..
[17:51:14] <giacus> http://giacus.altervista.org/video/img000.jpeg
[17:51:34] <giacus> http://giacus.altervista.org/video/img002.jpeg
[17:52:08] <giacus> it should be ok for the 'scanner'
[17:53:06] <giacus> I hope .. :)
[17:55:08] <alex_joni> giacus: http://www.ercim.org/publication/Ercim_News/enw44/rocchini.html
[17:56:25] <giacus> mm ok
[17:56:40] <giacus> old article ..
[17:57:13] <giacus> I'd try this http://www.sjbaker.org/projects/scanner/
[17:57:33] <giacus> with a small turntable
[17:57:44] <alex_joni> http://www.instructables.com/ex/i/03D27BB4F6F410279383000D61419BA3/
[17:59:00] <giacus> yeah, should use the same process
[18:00:31] <giacus> wonder why they use green light laser
[18:00:42] <giacus> it should be dangerous for what I know
[18:02:04] <bill2or3> probablly just because that's what they had.
[18:02:20] <giacus> btw, I do not have to scan humans objects
[18:02:41] <giacus> bill2or3: I doubt ..
[18:03:18] <giacus> I bought a laser level for E. 18 this afternoon
[18:03:35] <giacus> double laser
[18:03:53] <bill2or3> I cant really think of any reason to use green over red, maybe it's thier favorite color or something.
[18:03:59] <giacus> the only issue could be the line thickness ..
[18:04:11] <giacus> it seems to be around 1.5-2 mm
[18:04:24] <giacus> but regular enough
[18:09:14] <giacus> f you are scanning something like a human face, it may be impractical to rotate the person being scanned. In this case, you should probably consider rotating the laser and the camera in a circle around the persons head rather than spinning the person.
[18:09:31] <SWPadnos> green may be better because it holds more of the luminance signal than red
[18:09:52] <CIA-8> 03flo-h * 10emc2/tcl/bin/pickconfig.tcl: Added mor i18n support (msgcat)
[18:11:02] <giacus> SWPadnos: that's ok
[18:11:41] <giacus> this scanner type should be ok for 'rounded' and small models If I understand
[18:11:49] <SWPadnos> nice level ;)
[18:12:35] <giacus> to scan a 3D relief (20x20) should be impossible with the laser
[18:12:40] <giacus> right ?
[18:13:02] <SWPadnos> 20mm x 20mm?
[18:13:08] <giacus> inches
[18:13:23] <SWPadnos> that would probably be hard
[18:13:56] <giacus> also 20x20 cm should be hard I suppose..
[18:14:06] <giacus> always, 3D relief
[18:14:13] <giacus> horizontal
[18:15:08] <SWPadnos> you should see how "thick" the beam is at the distance you'd need for a 20x20 (cm or inch) scanning area
[18:15:47] <giacus> yeah, what seems to me strange is that turntable go turning around of 360 �
[18:15:54] <giacus> and cant get it
[18:15:55] <SWPadnos> point your camera at the area you want to scan, and see how many pixels the beam is - I think that'll be the limiting cfactor
[18:15:58] <SWPadnos> factor
[18:16:14] <giacus> uhm
[18:16:18] <giacus> the distance
[18:16:23] <giacus> too
[19:10:52] <lerman_> lerman_ is now known as lerman
[19:30:38] <anonimasu> *yawn*
[19:30:43] <anonimasu> maybe I should go outside and do some work
[19:46:27] <tfmacz> Hello everyone, can anyone tell me if I can mix a quadrature stepper drive with step and direction stepper drive in the same machine???
[19:48:12] <SWPadnos> yes, you can
[19:48:49] <jepler> tfmacz: With emc2 you can
[19:48:58] <jepler> tfmacz: You can specify the step type for each stepgen component
[19:49:03] <SWPadnos> yeah - that's what I meant ;)
[19:49:08] <tfmacz> is there some instructions somewhere???
[19:49:22] <SWPadnos> loadrt freqgen step_type=1,2,0,3,1,2 ...
[19:49:25] <jepler> tfmacz: Refer to the "hal introduction" pdf or the source code for the meaning of the different step types
[19:49:46] <SWPadnos> jepler, is that up to date for freqgen?
[19:49:52] <jepler> SWPadnos: I'm not sure if the pdf is up to date
[19:50:01] <jepler> the source code says that "type 2" is quadrature
[19:50:18] <jepler> type 0 is step+direction
[19:50:47] <jepler> so you might want 'step_type=0,0,2' if X and Y are step+dir and Z is quadrature
[19:50:57] <tfmacz> I am playing with a pic-servo controller the LG designed and would like to try it on an axis of my machine.
[19:50:58] <SWPadnos> an excerpt from the source, on how to do this:
[19:50:59] <jepler> (typically, freqgen.0 is x, freqgen.1 is y, etc)
[19:50:59] <SWPadnos> a command line like this:
[19:51:01] <SWPadnos> insmod freqgen step_type=0,0,1,2
[19:51:02] <SWPadnos> will install four step generators, two using stepping type 0, one using type 1, and one using type 2.
[19:52:41] <tfmacz> very good, I have the info now. will let you know how it works.
[20:05:42] <cradek> you guys keep saying freqgen - you mean stepgen, right?
[20:05:58] <SWPadnos> nope
[20:06:20] <cradek> hmm ok
[20:06:39] <SWPadnos> they seem to do the same thing, and I'm not sure which is "better"
[20:07:24] <cradek> for steppers, I think you want to use stepgen - it takes as input the commanded position
[20:08:22] <SWPadnos> ah, true
[20:08:28] <SWPadnos> same function, different inputs
[20:08:55] <cradek> yes
[20:10:50] <jepler> oops. I must be stuck on freqgen; stepgen is what you want.
[20:11:32] <jepler> seems like there should be just one, and if you have a velocity input then use blocks to integrate it
[20:11:36] <jepler> or vice versa
[20:14:01] <alex_joni> fenn: you around?
[20:16:22] <fenn> nice timing
[20:17:11] <alex_joni> fenn: do you have more of that bebop ?
[20:17:22] <fenn> heh only 5 cd's worth
[20:17:35] <alex_joni> * alex_joni wants some ;)
[20:17:49] <fenn> http://fenn.dyndns.org/pub/Cowboy Bebop OST's/
[20:17:56] <anonimasu> nice series
[20:17:57] <fenn> ost 1 2 and 3 are pretty good
[20:18:05] <anonimasu> yoko kanno :)
[20:18:19] <fenn> and cowgirl ed
[20:18:27] <alex_joni> http://www.instructables.com/ex/i/1C57F09C0300102985CB001143E7E506/ <- laser carved chocolate
[20:18:30] <alex_joni> rofl
[20:19:58] <fenn> you need a waterjet for that
[20:20:45] <giacus> why do not use press ?
[20:20:50] <giacus> for that
[20:20:53] <ValarQ> alex_joni: nice
[20:20:57] <alex_joni> for cutting yes, but not for etching
[20:20:58] <giacus> what the advantage ?
[20:21:09] <alex_joni> giacus: this is for people with too much spare time
[20:21:19] <giacus> uhm
[20:21:20] <giacus> :)
[20:21:24] <fenn> alex_joni: be warned cowboy bebop goes all over the map
[20:21:30] <giacus> really too much ..
[20:22:39] <alex_joni> all over the map?
[20:22:52] <giacus> famale inlay could be simple maded with aluminium
[20:23:04] <giacus> female
[20:23:10] <anonimasu> inlay for what?
[20:23:15] <giacus> for chocolate models
[20:23:35] <giacus> 2D
[20:24:26] <anonimasu> why not mold the chocholate in it from the start?
[20:24:38] <anonimasu> ah well, that's cool one off stuff..
[20:24:44] <anonimasu> cheaper then making a mold/press
[20:25:05] <giacus> I cant machine aluminium for that yet :(
[20:25:15] <fenn> next thing you know they'll be putting liquid chocolate in inkjet printers
[20:25:21] <giacus> haha
[20:25:25] <fenn> and printing flexible tft chocolate displays
[20:25:39] <alex_joni> lickable flexible tft displays
[20:25:40] <alex_joni> :D
[20:25:53] <fenn> actually a certain 3d printer uses starch and sugar water..
[20:26:48] <giacus> bleah
[20:27:57] <anonimasu> hm
[20:28:10] <anonimasu> * anonimasu likes SLS
[20:28:13] <anonimasu> that's cool stuff
[20:28:33] <giacus> anonimasu: what is ?
[20:28:39] <anonimasu> selectice laser sintering
[20:28:49] <giacus> never eard
[20:29:34] <anonimasu> rhttp://www.tms.org/pubs/journals/JOM/9812/Das-9812.html
[20:30:18] <anonimasu> laser sintering.. stuff..
[20:30:26] <anonimasu> 3d printing :)
[20:30:32] <anonimasu> in metal
[20:30:41] <giacus> a bit a dream ..
[20:30:49] <giacus> too much expensive for hobbist
[20:31:10] <anonimasu> http://www.materialise.com/prototypingsolutions/laser_ENG.html
[20:31:13] <fenn> giacus: http://www.bathsheba.com/
[20:31:45] <giacus> fenn: cool
[20:31:58] <giacus> it would be great for jewels I suppose
[20:33:03] <anonimasu> laser sintering?
[20:33:10] <anonimasu> oh, great for anything almost ;)
[20:33:47] <giacus> I'd like to see some 'simple' project in emc, like a printer to paint in a wall
[20:33:58] <giacus> os robotic arm to paint :(
[20:34:00] <giacus> or
[20:34:05] <fenn> so do it :)
[20:34:13] <giacus> :D
[20:34:14] <anonimasu> yep
[20:34:15] <fenn> * fenn throws a cheese doodle at giacus
[20:34:15] <anonimasu> :)
[20:34:21] <giacus> hehe
[20:34:24] <giacus> i'll try
[20:34:32] <fenn> you could have robots that play with food
[20:34:52] <giacus> yeah , already have it :)
[20:35:15] <fenn> or make an additive rapid chocolate prototyping machine
[20:35:31] <anonimasu> that wouldnt be impossible ;)
[20:35:34] <giacus> bah ..
[20:35:41] <fenn> am i thinking too small?
[20:35:50] <alex_joni> fenn: with molten chocolate you could use a 3D printer
[20:36:05] <fenn> uh oh i just ate 1 cup of chocolate chips
[20:36:14] <fenn> * fenn hides the chocolate from hisself
[20:36:18] <rayh> We do have a guy in Belgium that makes chocolate molds with an emc pick-and-place.
[20:36:45] <giacus> wonder how a robotic arm can be controlled by emc in a 3D space
[20:36:48] <rayh> He gave SO a kilo of chocolate when we visited.
[20:36:53] <giacus> if is possible
[20:37:11] <anonimasu> giacus: yes it's possible
[20:37:12] <rayh> Sure. Sagar ran a SCARA with EMC
[20:37:14] <jepler> giacus: emc2 supports "non-trivial kinematics"
[20:37:21] <jepler> giacus: alex checked in some support for "hexapods" recently
[20:37:28] <anonimasu> I so much wish I had a running mill
[20:37:30] <giacus> who tried something of pratical ?
[20:37:34] <giacus> about it
[20:37:39] <anonimasu> building a robot arm would be so simple hardware wise.
[20:38:02] <anonimasu> but well, with a cnc mill you can build almost anything..
[20:38:20] <giacus> jepler: thats the stuff I like
[20:53:26] <giacus> anonimasu: http://www.robot-italy.com/index.php/cPath/11_20
[20:53:40] <Jymmm> rayh: Do you know what material he's using for the dies?
[20:53:44] <giacus> could be nice, but i'm not able to machine metals as I sayd :(
[20:54:12] <giacus> but I have to build an arm for my i-droid
[20:54:19] <giacus> hi jepler
[20:54:24] <giacus> ops Jymmm
[20:54:33] <giacus> seen les ?
[20:55:06] <giacus> I have few questions for him :(
[20:57:07] <anonimasu> giacus: nice
[21:00:21] <giacus> anonimasu: yeah, I think I'll use plastic for now :(
[21:05:16] <fenn> * fenn wishes he would hurry up and build this dumb hexapod already
[21:05:31] <rayh> Jymmm, Who is using for the dies?
[21:05:56] <jepler> Jymmm: You were asking earlier about the current limit of linear regulators. IMO the limit is generally not current, but power dissipation, which is approx. (Vin-Vout) * Iout.
[21:06:34] <jepler> Jymmm: the allowable power dissipation of a SOT23 package is not much
[21:06:36] <cradek> fenn: it would be great if you'd have something by fest
[21:06:43] <Jymmm> rayh: You said someone in Belgium is making molds. I assumed thermoforming on top of some machined die.
[21:07:01] <fenn> i'm aiming to have all the mechanical crap together and at least some attempt at a motor driver
[21:07:11] <Jymmm> jepler: It took me a while because I never touched SMT, but I found it to be 100mA
[21:08:20] <fenn> you can thermoform with styrofoam even
[21:08:26] <fenn> i saw it on mythbusters :)
[21:08:48] <rayh> Right the die was aluminum.
[21:09:05] <Jymmm> fenn: but chocolate?
[21:09:12] <Jymmm> rayh ah, ok.
[21:09:22] <fenn> chocolate probably requires better detail than you can get with styrofoam
[21:09:32] <rayh> Then they vacuform some sort of clear plastic into it.
[21:09:39] <fenn> although detail on lost foam castings can be amazing
[21:09:54] <rayh> Then ultrasonic weld magnets to the plastic.
[21:10:02] <Jymmm> rayh Yeah, PET is food grade plastic
[21:10:32] <fenn> magnets?
[21:11:02] <rayh> To hold the plastic in the chocolate casting machine.
[21:12:06] <rayh> The magnets were about .25 diameter round and about .125 tall.
[21:12:18] <rayh> Inch that is.
[21:13:10] <fenn> ah dammit i wanted to go to the scrapyard today.. *sigh*
[21:13:44] <jepler> Jymmm: for this random sot-23-5 surface mount regulator: http://www.mouser.com/index.cfm?&handler=data.listcategory&D=*512fan2500s25x*&terms=512-fan2500s25x&Ntt=*512fan2500s25x*&Dk=1&Ns=SField&N=0&crc=true
[21:14:02] <jepler> Jymmm: You'll see that theta_JA is 220 degC/W
[21:14:33] <jepler> Jymmm: so if your power dissipation is 100mA * 1V = 100 mW then the temperature rise is 22C
[21:16:17] <jepler> Jymmm: but if you're trying to get, say, +5V from a rectified 9VAC supply, that's not 1V but 7V drop, giving an out-of-spec 154C rise
[21:18:58] <Jymmm> jepler: Nah, I just needed to know the max current draw surface mount packages typically handle is all
[21:19:31] <Jymmm> I found a TO-92 that can handle 250mA; that should work better for my needs.
[21:25:05] <jepler> case study: I designed my etch servo board to take the drive voltage and pass it through a 5V linear regulator (to92 package actually) to get the logic supply. With vdrive=12v, it works fine; with vdrive=24v the regulator overheats (though it didn't seem to damage anything before I unplugged it)
[21:25:30] <jepler> even though in both cases I was using around the same current for the logic
[21:28:03] <Jymmm> all that heat has to go soemwhere =)
[21:29:29] <jepler> yeah -- for instance, it could melt the regulator 'till it shorts Vin and Vout together, and then you roast the logic section of the driver, encoder, as well as your parallel port since you skimped and didn't isolate it.
[21:29:45] <jepler> (I lucked out)
[21:29:46] <Jymmm> I might just say screw it and just use drop resiters instead. No that big a deal.
[21:30:04] <Jymmm> jepler sounds like you did =)
[21:30:28] <alex_joni> fenn: anything that's like track #14?
[21:30:35] <Jymmm> and I can get shitloads of resistors for free - by the roll
[21:30:36] <alex_joni> I downloaded a few, but nothing similar
[21:37:12] <Jymmm> oh gawd... I forgot how to calculate voltage drops! lol
[21:37:24] <Jymmm> (it's been a rough week)
[21:42:11] <CIA-8> 03rayhenry * 10emc2/tcl/bin/emccalib.tcl: Will now apply changes to HAL. No revert or save yet.
[21:48:45] <Jymmm> Is 300mW alot of heat?
[21:49:20] <Jymmm> Would I feel that on my fingertips?
[21:49:49] <cradek> Jymmm: probably
[21:50:08] <Jymmm> ok, thanks.
[21:50:19] <Jymmm> sorry guys, just out of it today.
[21:51:24] <jepler> in that sot-23-5 package I mentioned earlier, it would cause the junction temperature to rise 220 * .3 degrees C. The case temperature would rise somewhat less, 39 degrees. (theta_JC = 130C/W)
[21:51:40] <jepler> you'd notice if you touched your finger to something 25+39 degrees C.
[21:51:59] <Jymmm> 102F
[21:52:24] <Jymmm> * Jymmm hugs google calculator
[21:56:50] <Jymmm> * Jymmm looks around for his 100 Watt resistors
[22:11:46] <giacus> * giacus give an iron to Jymmm
[22:12:27] <Jymmm> * Jymmm takes the iron and makes grilled cheese
[22:17:18] <giacus> Jymmm: take a look my city :P http://giacus.altervista.org/photo/vacanze/usa2002/page002.html
[22:17:43] <giacus> here I was dancing on the street http://giacus.altervista.org/photo/vacanze/usa2002/img021.jpeg.html
[22:17:46] <giacus> hahah
[22:19:38] <giacus> hollywood was wonderful
[22:20:08] <giacus> next year, if all its ok, i'll be there with k4ts ;P
[22:20:22] <giacus> without would be better , but ..
[22:20:26] <giacus> I cant :(
[22:20:36] <giacus> K4ts: helloo
[22:20:42] <giacus> you around ? ^_^
[22:21:04] <K4ts> sigh
[22:21:17] <giacus> please dont cry
[22:21:27] <giacus> forget the dog
[22:21:32] <giacus> isnt for you :(
[22:21:42] <giacus> haha
[22:22:47] <giacus> K4ts: waht the placement in the latest paltalk competition ?
[22:22:59] <giacus> got it ?
[22:23:03] <K4ts> perso
[22:23:10] <giacus> posizione ?
[22:25:00] <alex_joni> night all
[22:25:42] <giacus> night alex
[22:27:10] <K4ts> no position
[22:27:20] <giacus> good
[22:27:23] <giacus> hehe
[22:27:37] <K4ts> hi hi?
[22:27:44] <giacus> K4ts: listen to me
[22:27:45] <K4ts> ih ih?
[22:27:56] <K4ts> non volgio pui il cane
[22:27:57] <giacus> forget the stipid peples
[22:28:08] <giacus> they cant understand the good sound :P
[22:28:37] <giacus> dont want the dog anyomore ?
[22:28:39] <giacus> O_O
[22:28:42] <giacus> sure ?
[22:29:02] <giacus> I think not
[22:29:54] <giacus> anna show must go on
[22:29:56] <giacus> hehe
[22:30:28] <giacus> guys, K4ts got a a dog puppy 3 month ago ..
[22:30:40] <giacus> after 3 week the puppy died
[22:31:01] <giacus> now got another, for E. 1000 or so ..
[22:31:09] <giacus> oh
[22:31:19] <K4ts_> ops sorry
[22:31:32] <giacus> ah ! you there
[22:31:33] <giacus> :D
[22:32:10] <giacus> and also the second puppy don stay well in healt ..
[22:32:54] <giacus> K4ts: I think youre not lucky with dogs
[22:33:03] <giacus> but be patience, wait ..
[22:33:10] <giacus> lets see the next
[22:33:12] <giacus> :D
[22:33:43] <giacus> but after 3 dogs stop
[22:33:47] <giacus> hehe
[22:34:14] <K4ts_> grrrrrr
[22:34:21] <giacus> ;)
[22:34:23] <K4ts_> much patience
[22:34:46] <giacus> seems not enough :(
[22:35:24] <giacus> K4ts: first cnc I tried to build died
[22:35:27] <giacus> second too
[22:35:37] <giacus> :P
[22:35:44] <giacus> dont lost the hope
[22:36:02] <anonimasu> how did it die?
[22:36:16] <giacus> who the cnc ?
[22:36:22] <Jymmm> anonimasu in the explosion
[22:36:29] <giacus> burned :/
[22:36:40] <giacus> really
[22:36:40] <anonimasu> burned? eh?
[22:36:44] <Jymmm> anonimasu told ya
[22:36:56] <giacus> :)
[22:37:11] <giacus> burned anything
[22:37:15] <giacus> drivers and other
[22:37:24] <Jymmm> giacus needs to stop making things out of magnesium and flint
[22:37:36] <giacus> heh
[22:39:15] <K4ts_> k4ts tomorrow to shave hair!
[22:39:54] <K4ts_> kojak
[22:40:07] <giacus> K4ts: I doubt the dog was italian
[22:40:23] <K4ts_> is italian
[22:40:28] <giacus> that dogs comes from east, for very low cost
[22:40:33] <giacus> mmm
[22:40:53] <K4ts_> ufff
[22:41:18] <giacus> U just have to wait some week
[22:41:21] <giacus> came on
[22:41:23] <giacus> :)
[22:42:50] <K4ts_> Jymmm: you sleep?
[22:42:58] <Jymmm> K4ts yes
[22:43:03] <K4ts_> working?
[22:43:08] <Jymmm> tired
[22:43:51] <K4ts_> i also
[22:43:54] <K4ts_> I
[22:44:26] <anonimasu> giacus: you pay that much for a dog you'r not sure where it comes from?
[22:44:31] <anonimasu> :/
[22:44:43] <giacus> anonimasu: absurd, but true ..
[22:44:48] <anonimasu> giacus: that's insane..
[22:44:51] <Jymmm> Trying to recover from lack of sleep... Had baby cockatiel that needed feeding every hour; he died yesterday <sniff>
[22:45:09] <anonimasu> over here they are about the same but that's from professional breeders
[22:45:09] <giacus> theyre selling dogs as 'italian' 250 E. more
[22:45:12] <SWPadnos> bummer
[22:45:17] <giacus> but they comes from ungary
[22:45:29] <anonimasu> :/
[22:45:34] <giacus> often theyre not in good healt
[22:45:40] <giacus> as K4ts know ..
[22:45:46] <SWPadnos> Jymmm, were you talking about SMT regulators or transistors earlier?
[22:45:47] <anonimasu> giacus: isnt thoose italian dogs checked by vet's too?
[22:45:48] <skunkworks> jepler: - did you see they make those 3 leg voltage regulators that are switching?
[22:46:00] <giacus> anonimasu: it should be
[22:46:01] <skunkworks> same package
[22:46:04] <Jymmm> SWPadnos VR's
[22:46:07] <anonimasu> hm, ok
[22:46:11] <giacus> teorically ..
[22:46:20] <SWPadnos> Jymmm, ok, for what? (sounded like you wanted to drive LEDs with them)
[22:46:39] <Jymmm> SWPadnos I was, but just gonna use resistors instead.
[22:46:51] <anonimasu> night everyone
[22:46:56] <Jymmm> nite anonimasu
[22:46:57] <giacus> she has to leave it to veterinary this afternnon
[22:47:03] <SWPadnos> see you anon
[22:47:06] <giacus> night anon
[22:47:07] <SWPadnos> Jymmm, what's the source voltage?
[22:47:16] <Jymmm> SWPadnos 5vdc
[22:47:29] <K4ts_> ?
[22:47:42] <SWPadnos> in that case, a resistor is good. if they're 2V LEDs, then put 2 in series
[22:47:51] <giacus> K4ts_: strange,, are you firewalled ?
[22:47:52] <K4ts_> kats has left chat
[22:47:59] <K4ts_> mah'
[22:48:00] <giacus> haha
[22:48:17] <giacus> maybe your client dont reply to the pings server
[22:48:40] <Jymmm> SWPadnos Yeah, 47 ohm @ 120mW
[22:48:54] <SWPadnos> what kind of LED?
[22:49:04] <SWPadnos> (forward drop)
[22:49:17] <Jymmm> SWPadnos oh, 3.3 to 3.6V @ 20mA
[22:49:22] <Jymmm> times 2
[22:49:26] <CIA-8> 03rayhenry * 10emc2/tcl/bin/emccalib.tcl: Enabled chanded variables test at quit.emccalib.tcl
[22:49:41] <Jymmm> so 3.3v @ 40mA
[22:49:43] <SWPadnos> you need a separate resistor for each LED (or series chain of LEDs)
[22:50:03] <K4ts_> giacus: we go at Jymmm ?
[22:50:11] <giacus> what ?
[22:50:16] <Jymmm> SWPadnos Was gonna run em in parallel
[22:50:22] <SWPadnos> slight variations in the production provess can make one get much more current than the other
[22:50:28] <SWPadnos> process
[22:51:17] <K4ts_> hollywood was wonderful [{google:en|it} il hollywood era meraviglioso ]
[22:51:35] <giacus> thats what I sayd ..
[22:51:37] <Jymmm> SWPadnos: Well technically it needs 42.5 ohm, so next available would be 47 ohms
[22:52:08] <SWPadnos> Jymmm, the problem is that diodes have a nonlinear current response to varying voltage
[22:52:40] <Jymmm> SWPadnos so, calc the resistor per individual led?
[22:52:46] <SWPadnos> yes
[22:52:57] <SWPadnos> can you use a 12V source?
[22:53:03] <Jymmm> sure
[22:53:10] <SWPadnos> then I'd do that
[22:53:20] <Jymmm> then a lot of heat loss
[22:53:26] <SWPadnos> put 3 LEDs in series, for a total of ~9.9V
[22:53:50] <SWPadnos> use a single resistor, and get 3x the light output (with nearly the same loss in the resistor as for a single LED)
[22:54:38] <SWPadnos> so, you put 20 mA through the 3 LEDs. figure a 2V drop across the resistor, and you have 100 ohms and 40 mW
[22:55:02] <Jymmm> SWPadnos only using 2 LEDs though
[22:55:13] <SWPadnos> try 3 ;)
[22:55:38] <SWPadnos> are you turning them on with a transistor, or are they always on?
[22:55:46] <Jymmm> SWPadnos Insert credit card here --> [ ]
[22:56:01] <Jymmm> always on
[22:56:02] <SWPadnos> xxxx yyyy zzzz oooo exp never/ever
[22:56:10] <Jymmm> invalid, try again
[22:56:17] <SWPadnos> hmmm - try it again
[22:56:38] <Jymmm> Revolk card and destroy
[22:56:48] <SWPadnos> ok, see you in Jamaica
[22:56:52] <Jymmm> lol
[22:56:54] <SWPadnos> anyway ...
[22:57:18] <SWPadnos> did you see the eBay listing for the 110000 mcd white LEDs?
[22:57:25] <Jymmm> no
[22:57:28] <SWPadnos> one sec
[22:57:48] <Jymmm> I ordered some 40k and 80k ones
[22:59:39] <SWPadnos> damn. anyway, it was 50 LEDs (10mm, 110, 115, or 130 cd), ~$19 shipped
[23:03:06] <SWPadnos> Ah, here it is:
[23:03:11] <SWPadnos> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=7597846430
[23:04:38] <Jymmm> I'll see what the 40k and 80k one perform, then maybe order those
[23:04:52] <SWPadnos> I'm sure they'll be bright :)
[23:04:57] <Jymmm> 10mm too
[23:06:36] <Jymmm> bbiab... lunch
[23:10:07] <SWPadnos> yeah, gotta run
[23:41:48] <K4ts_> night