#emc | Logs for 2006-03-10

[00:16:24] <LawrenceG> tfmacz: Hey Ted.... It works.... cheap rotary encoder from mouse intestines http://members.shaw.ca/swstuff/mouse.html
[00:17:51] <LawrenceG> Anybody need an RPG to interface to EMC2?
[00:20:34] <LawrenceG> * LawrenceG brb
[00:31:36] <fenn> nice page LawrenceG
[00:34:45] <giacus> yup :)
[00:34:48] <giacus> very nice
[01:01:56] <giacus> hello jmk
[01:02:09] <jmkasunich> hi
[01:02:37] <giacus> i'm looking at wiki
[01:02:44] <giacus> looks nice now :)
[01:29:19] <giacus> can Emc control an arm like this? http://www.parallax.com/html_pages/robotics/roboarm.asp
[01:30:54] <LawrenceG> fenn: giacus Thanks....
[01:31:13] <giacus> LawrenceG: great page :)
[01:31:48] <LawrenceG> Its amazing what can be done with a dremel cutoff disk
[01:31:59] <giacus> nice
[01:34:20] <jmkasunich> giacus: emc can't control such an arm today, but it would be doable with some work
[01:34:50] <jmkasunich> that arm uses RC servos, EMC currently has no driver for those
[01:35:03] <Jymmm> LawrenceG: you should try a angle grinder sometime =)
[01:35:13] <jmkasunich> but it could be done (in fact, I should do that, it might be fun)
[01:35:26] <jmkasunich> the other issue is kinematics
[01:35:58] <giacus> yeah.. I read something about kinematics .
[01:38:28] <LawrenceG> Hey Jymmm ... I like angle grinders as well except when I set my overalls on fire!
[01:38:50] <Jymmm> LawrenceG Hey, no exceptions... where's the fun in that!
[01:39:21] <Jymmm> LawrenceG as long as it's not 4 foot flames of course
[01:39:49] <giacus> LawrenceG: are you a Radio Operator ?
[01:40:04] <Jymmm> cq cq cq dx
[01:40:10] <giacus> :)
[01:40:16] <giacus> Im an swl
[01:40:28] <LawrenceG> giacus: yes.... into ham radio since about 1965..... licensed in 1971
[01:40:47] <giacus> LawrenceG: dou you go in ssb band too ?
[01:40:59] <Jymmm> 2m/70cm for me
[01:41:10] <LawrenceG> swl is still fun... I have a big old military tube radio for swl
[01:41:12] <giacus> I did it for a bit of years
[01:41:30] <giacus> Ive about 30 countries postacard
[01:41:42] <giacus> canada too I think
[01:41:44] <giacus> 4-5
[01:41:45] <LawrenceG> ssb... yes... waiting to check into a net on 3.729mhz
[01:42:01] <giacus> but it was many years ago ..
[01:42:08] <Jymmm> LawrenceG general?
[01:42:29] <LawrenceG> Canadian Advanced....
[01:42:41] <Jymmm> ah cool
[01:42:54] <LawrenceG> they have 3 grades here... basic, code and advanced
[01:43:26] <LawrenceG> now you can get on hf if you have a high mark on your basic exam or code or advanced
[01:44:19] <LawrenceG> advanced lets one run amplifiers, build stuff and sponsor repeaters and club stations
[01:45:20] <CIA-8> 03jepler * 10emc2/src/ (Makefile Makefile.inc.in configure configure.in):
[01:45:20] <CIA-8> If there's an "axis" subdirectory (or axis-1.3a0, etc), run "setup.py install"
[01:45:20] <CIA-8> in it. This actually only works with AXIS 1.3, which doesn't install any files
[01:45:20] <CIA-8> outside the emc2 directory. It also requires (but does not check for)
[01:45:20] <CIA-8> --enable-run-in-place.
[01:45:22] <LawrenceG> Jymmm: do you have an HF antenna up?
[01:45:51] <giacus> recently we got a company here (TIM telecom italia mobile) with a new card type for cellphones, it allow to use the cell as walkie talkie
[01:46:00] <giacus> for free of course
[01:46:13] <giacus> but in short distance..
[01:46:45] <LawrenceG> I think Nextel in the USA has the same
[01:47:34] <LawrenceG> a good feature if you are traveling with someone on the highway....
[01:47:42] <giacus> yeah ..
[01:48:20] <giacus> I have to try it yet, but seem interesting
[01:48:25] <LawrenceG> We dont have it here yet.... we still have to make a long distance cell call to a car 20m in front of us.
[01:49:56] <ace> logger_aj: bookmark
[01:49:57] <ace> See http://solaris.cs.utt.ro/irc/irc.freenode.net:6667/emc/2006-03-10#T01-49-56
[01:51:46] <robin_sz> meep?
[01:51:53] <giacus> hello robin_sz
[01:51:57] <robin_sz> ideed
[01:52:09] <robin_sz> bah ... just got in from work
[01:52:12] <robin_sz> LOOOOOONG
[01:52:12] <Jymmm> LawrenceG: No, just a tech =)
[01:52:14] <robin_sz> day
[01:52:19] <giacus> :-)
[01:52:46] <robin_sz> i was just watching all the ham talk ;)
[01:53:03] <giacus> hehe
[01:53:36] <robin_sz> if I dont get this laser to run .. I'm going to tune the rf amp down to 10m band and be very loud :)
[01:53:47] <LawrenceG> one needs a cnc machine to make antenna parts, panels and other goodies
[01:54:45] <LawrenceG> robin_sz: cool.... you have enough power to make your own propagation
[01:54:54] <robin_sz> indeed :)
[01:55:00] <robin_sz> 2kkw out, cw
[01:55:04] <robin_sz> 25kw
[01:55:15] <giacus> robin_sz: I learned about plasma toolpath .. is it new to me know that plasma cant do the same toolpath as laser machine
[01:55:19] <LawrenceG> I'm glad I dont pay the power bill
[01:55:40] <robin_sz> giacus: ideed, you need to keep moving with plasma, or it goes out
[01:55:49] <giacus> yeah :(
[01:56:01] <giacus> what cad/cam could be ok for that ?
[01:56:03] <robin_sz> and then theres corner compensation
[01:56:16] <LawrenceG> and a laser burns a hole in the floor if you stop
[01:56:18] <Jymmm> LawrenceG I only have a TM-D700a
[01:56:34] <jepler> Does "Oscar A. Dalem" use the IRC channel? What's his nick?
[01:56:35] <robin_sz> you need to do "clover leaf" corners on thin sheet with plasma
[01:56:51] <robin_sz> its not a problem over 3mm though
[01:57:32] <robin_sz> LawrenceG: you know, some over-the-horizon radio links on 1.3ghz do make their own propagation
[01:57:39] <giacus> yeah, my cousin was explain that to me this evening ..
[01:57:52] <giacus> btw, we're starting to build the machine
[01:58:09] <giacus> I also found few docs about grounding
[01:58:17] <giacus> hope is enough :(
[01:58:58] <robin_sz> the ones ot to oil rigs apparently, noticed that the rx signal versus tx power was non linear. hit the air hard enough and it creates it own little active spot in the tropo
[01:59:48] <robin_sz> right ... bedtime for bonzo
[02:00:00] <giacus> what the power transimission of a norla cellphone ?
[02:00:08] <giacus> few mw ?
[02:00:22] <giacus> normal
[02:00:27] <LawrenceG> Yea... the USA has some megawatt transmitters up in Alaska (see HAARP) for atmospheric research (which means who knows what)
[02:01:13] <LawrenceG> cellphone = about 0.5watt.... some old bag phones were 3watts
[02:01:29] <giacus> 800 mhz ? if I remember well
[02:02:07] <LawrenceG> here most cells are 800-900mhz range, rx and tx 45mhz apart
[02:03:35] <Jymmm> broadband over mains
[02:03:50] <giacus> dunno here.. first motorola I had was rocks
[02:03:59] <LawrenceG> Jymmm: be nice... thats a touchy subject
[02:04:07] <Jymmm> giacus and probably wieght just as much too
[02:04:15] <giacus> after that I bought an ericsson t68i and having a lot of issue
[02:04:16] <Jymmm> LawrenceG Yeah, I know it is.
[02:04:47] <giacus> after bought an LG with camera.. too bad too ..
[02:04:54] <giacus> sometime lost the signal
[02:05:05] <giacus> semms ok, but isnt able to receive any call
[02:05:07] <giacus> :(
[02:05:10] <Jymmm> LawrenceG any testing in your area?
[02:05:27] <LawrenceG> not yet... brb got to check into net....
[02:05:47] <Jymmm> Can't type and talk huh? *chuckle*
[02:06:10] <LawrenceG> not very well...
[02:06:13] <giacus> oh.. my radio was President Jackson :)
[02:06:24] <giacus> qith quad cubical antenna
[02:06:36] <giacus> was nice :P
[02:06:37] <Jymmm> giacus lol, what was your handle?
[02:06:44] <giacus> hehe
[02:06:59] <giacus> Jymmm: I had to done..
[02:07:18] <giacus> to much peoples around cant see the tv
[02:07:24] <giacus> too much interference :/
[02:07:28] <Jymmm> lol
[02:07:31] <giacus> haha
[02:07:45] <giacus> they was knocking any day to my door
[02:07:49] <Jymmm> Yeah... 100 Watts on 11m will do that =)
[02:08:13] <giacus> I was using a valvles linea amplifier too ..
[02:08:19] <giacus> to much power
[02:08:41] <giacus> damn keyboard, I have to change it
[02:08:47] <giacus> keys are too high
[02:10:19] <giacus> btw I got aorund 30 countries in some year
[02:10:23] <Jymmm> giacus: put them on cnc and make them shorter
[02:10:30] <bill20r3> yo.
[02:10:37] <Jymmm> -yo
[02:10:38] <giacus> and thousand of postcards
[02:10:42] <giacus> hi bill20r3
[02:11:04] <bill20r3> I scored big at the freight-damage store, got a server-rack
[02:11:27] <bill20r3> the open kind with 2 rails, gonna cut it down for my gantry.
[02:11:33] <bill20r3> $20.
[02:16:01] <Jymmm> bill20r3 is this an independant place or do they have more than one location?
[02:17:15] <bill20r3> just one I think
[02:17:49] <bill20r3> This is thier page: http://www.nationalproductsales.com/industrial_sales/
[02:17:50] <Jymmm> Bummer... I'd love to find one out here.... Hell next to SFO OAK and SJC
[02:18:05] <bill20r3> there's probablly something similar
[02:19:21] <bill20r3> they have totally wierd sales, todays good sale was "40% off anything over six feet long"
[02:19:29] <jepler> hah
[02:24:13] <cradek> jepler: was your following error caused by something like g0x1y.2a3600?
[02:30:48] <Jymmm> bill20r3: Sounds like a storage issue thing.
[02:31:40] <Jymmm> bill20r3: Is this from one carrier or do you see labels from all different carriers?
[02:41:23] <giacus> night
[02:41:28] <Jymmm> nite
[02:45:11] <CIA-8> 03cradek * 10emc2/src/emc/task/emccanon.cc: accel for mixed linear/rotary were still wrong; pretty sure this is right.
[03:10:31] <SWP_Away> SWP_Away is now known as SWPadnos
[04:02:08] <cradek> does anyone know the url for alex's new experimental website?
[04:02:41] <fenn> juve.ro (let it load) then add joomla/ to the url
[04:05:43] <Jymmm> This is pretty cool... http://www.heatermeals.com/
[04:07:11] <cradek> fenn: thanks
[04:07:50] <SWPadnos> the cncgear site may be more recent, but I;m not sure
[04:08:00] <fenn> low in sodium.. warning contains ammonium chloride! :P
[04:08:47] <SWPadnos> heh "Large 3/4 pound Tasty entree"
[04:08:53] <fenn> stupid java
[04:10:09] <skunkworks> cradek: How long do you think before your tp will be in "testing"?
[04:12:02] <Jymmm> Well, for a hot meal and no frig needed, not too shabby.
[04:12:38] <SWPadnos> it looks expensive at $6 each, but I guess it isn't too bad
[04:12:57] <SWPadnos> if only they had the BBQ Beef entree like the *real* MREs
[04:13:14] <skunkworks> I have a linksys ap communicating with a d-link router - wireless link. (linksys says that it will only work with 2 linksys ap)
[04:13:18] <SWPadnos> and freeze dried strawberries - those are great
[04:13:18] <Jymmm> SWPadnos: And latative gum?
[04:13:22] <fenn> if only they had RTG's built in to keep it hot and ready for 1000 years
[04:14:00] <fenn> skunkworks: congratulations
[04:16:18] <skunkworks> here is the odd thing - the dlink router is set up without dhcp and has an ip The access point has ip of 245. When I reverse these and hook the ap to my wired network and the dlink wireless router to my portable it doesn't work. (I would think it didn't matter)
[04:17:02] <Jymmm> Well, the ONLY thing I have found that has an indefinite shelf life is Spam
[04:22:39] <fenn> skunkworks: why are you using DHCP on a router?
[04:26:29] <skunkworks> I am not using dhcp in the router..
[04:29:03] <skunkworks> I was hoping that I could use 2 of the dlink 39 dollar wireless routers as a wireless link - between dlink and the research that I had done I found out I could not use them as a wirelesss bridge.
[04:30:08] <skunkworks> they where only 39 dollars. the linksys ap's are around 69 - so if it works between the dlink and the ap I save a little money ;)
[04:31:50] <skunkworks> I am posting this with my portable wired to the ap - it transmitting wirelessly to the dlink router to get to my network and the internet.
[04:32:29] <skunkworks> time to make some antannas ;)
[04:34:32] <Jymmm> skunkworks: Got Pringles?
[04:38:20] <samco> samco is now known as skunkworks
[04:38:31] <skunkworks> changed back to my wireless card.
[04:39:08] <skunkworks> I have seen the pringles can cantinas. thinking more along the lines of a yagi or similar
[04:40:26] <skunkworks> Ok - way past my bedtime. Take it easy.
[04:40:35] <Jymmm> G'Night skunkworks
[04:40:44] <SWPadnos> see you
[05:38:46] <CIA-8> 03swpadnos * 10emc2/src/hal/utils/halcmd.c: Added thread execution time and max thread execution time to show thread output.
[05:39:47] <SWPadnos> SWPadnos is now known as SWP_Away
[14:06:58] <giacus_> lag :(
[14:13:02] <giacus_> jepler: cradek ?
[14:17:32] <giacus_> question about Axis
[14:18:22] <giacus_> and i18n files
[14:20:11] <giacus_> there are few terms to change
[14:21:21] <giacus_> for the IT i18n file :
[14:22:05] <giacus_> 1) under the menu View, term 'mostra' is repeated 3 times, there's no reason for that
[14:22:17] <Jymmm> Anyone know what "enamel paint" really is?
[14:22:31] <giacus_> hi Jymmm
[14:22:36] <giacus_> nope..
[14:22:38] <Jymmm> hi Jacky!
[14:22:42] <giacus_> :)
[14:23:45] <giacus_> first time I eard it ..
[14:24:01] <giacus_> you dreamed it tonight ? :P
[14:24:21] <Jymmm> I've seen it for plastic model paint, but those are usually just 1/2" ounce jars.
[14:24:29] <CIA-8> 03cradek * 10emc2/src/emc/kinematics/tp.c: eliminate warning
[14:24:29] <Jymmm> 1/2 ounce
[14:24:37] <alex_joni> Jymmm: I know ;)
[14:24:47] <Jymmm> alex_joni Tell me more don PArdo!
[14:24:51] <alex_joni> but there are many kinds ..
[14:25:01] <alex_joni> there is traditional enamel
[14:25:17] <alex_joni> the one which is like powder, and then you put the part into an oven
[14:25:23] <Jymmm> k
[14:25:30] <alex_joni> and the paint melts and it gets an uniform layer
[14:25:39] <Jymmm> like the old signs on metal for years ago
[14:25:58] <alex_joni> like pots
[14:26:00] <Jymmm> yeah, baked enamel
[14:26:02] <Jymmm> ok
[14:26:03] <alex_joni> the white ones
[14:26:14] <giacus_> alex_joni: already seen the robo-penguin? http://www.parallax.com/html_pages/robotics/penguin.asp
[14:26:15] <alex_joni> ok, and enamel paint is normal alchidic paint
[14:26:32] <Jymmm> ?
[14:26:32] <giacus_> haha look at the video, thats funny :D
[14:27:54] <Jymmm> alex_joni: google doesn't turn up anything relative for 'alchidic paint'
[14:28:55] <alex_joni> I might have spelled it badly
[14:29:25] <Jymmm> no you spelled it right, just having a hard time getting significant info on enamel paint in general.
[14:29:55] <alex_joni> ok..
[14:30:02] <alex_joni> * alex_joni goes home
[14:30:02] <Jymmm> alex_joni: Even wikipedia is limited http://en.wikipedia.org/wiki/Enamel_paint
[14:41:58] <jepler> giacus_: Do you know how to edit "po" files? If so, just take the one from axis 1.2rc3, fix it up, and e-mail it to me <jepler@unpythonic.net>
[14:43:18] <giacus_> jepler: ok.
[14:43:29] <jepler> giacus_: there may be at least one translation that is completely missing, too
[14:44:21] <giacus_> I used the latest. Btw i'll tell you the vers. too
[14:44:42] <giacus_> for what I seen translation is 90 % right
[14:44:51] <giacus_> just few terms can be changed
[14:44:57] <giacus_> i'll do it
[14:46:18] <jepler> giacus_: I appreciate it. I don't have the necessary language skills, so I have to rely on others.
[14:47:53] <giacus_> np, I seen Ernesto did a good translation too.
[14:48:22] <giacus_> an help section could be nice too ..
[14:48:26] <giacus_> lets see
[14:58:46] <jepler> help section?
[15:00:58] <giacus_> yeah, a short tutorial
[15:02:20] <alex_joni> back
[15:06:53] <giacus_> hi alex_joni
[15:07:08] <giacus_> are you able to upload a file on linuxcnc.org ?
[15:07:11] <alex_joni> hello jacky
[15:07:20] <alex_joni> giacus_: I should be able to ;)
[15:07:24] <giacus_> K
[15:07:29] <alex_joni> mail it to me
[15:07:40] <giacus_> I done the it vers. of Introduction to HAL
[15:07:46] <alex_joni> nice
[15:07:57] <giacus_> it is not updated yet to the latest Emc2 release..
[15:08:01] <giacus_> sorry ..
[15:08:11] <giacus_> but should be a good start point
[15:08:35] <giacus_> I have to find the time to update it and manually test every command in the tutorial
[15:08:37] <alex_joni> giacus_: ok
[15:08:57] <giacus_> for now, you can upload the 'beta' vers. if you like it
[15:09:07] <giacus_> a moment..
[15:09:45] <jepler> giacus_: I'd rather see the EMC Handbook updated for emc2 .. I thought it had a section on the GUIs but I don't see one now.
[15:10:15] <giacus_> jepler: yeah, you're right
[15:10:47] <alex_joni> jepler: there are 3 handbooks (developer, integrator & user)
[15:11:09] <alex_joni> and all 3 are pretty badly out of shape :(
[15:11:38] <jepler> alex_joni: I know; who wants to write documentation?
[15:11:50] <alex_joni> jepler: certainly not developers :)
[15:12:00] <jepler> and when we do, it's incomprehensible to our users.
[15:12:00] <cradek> I wrote a wiki page
[15:12:10] <alex_joni> cradek: that doesn't count
[15:12:10] <cradek> ... but it might be incomprehensible
[15:12:15] <cradek> eh??
[15:12:16] <cradek> why not?
[15:12:19] <jepler> which wiki page?
[15:12:22] <alex_joni> it's just one
[15:12:36] <cradek> 1 > 0
[15:12:39] <alex_joni> if it exceeds 200 pages, it counts
[15:12:49] <alex_joni> :P
[15:12:50] <cradek> if it exceeds 200 pages, it's probably shit
[15:12:55] <alex_joni> lol
[15:17:23] <fogl> is emc2 still in deveopping stage...the notes about that on linuxcnc.org are almost one year old?
[15:17:49] <jepler> fogl: It's very good for daily use, especially if you stick with "Testing".
[15:17:53] <giacus_> http://giacus.altervista.org/Hal_IT_08.pdf
[15:17:56] <giacus_> alex try it ..
[15:18:24] <alex_joni> giacus_: I'll change the name to HAL_Introduction_it.pdf
[15:18:41] <giacus_> yeh feel free to do it ..
[15:18:58] <giacus_> ny hosting seems reject some '_' or .
[15:19:02] <giacus_> :/
[15:19:13] <giacus_> also check for capitalized letters
[15:19:30] <giacus_> btw change it from yourself
[15:20:18] <giacus_> according to jmk and rayh I used GFDL , hope its ok
[15:20:56] <giacus_> the next vers. will be update in robotialia wiki too
[15:21:28] <giacus_> as I sayd i've not much time now to update that doc
[15:21:43] <giacus_> but it could help someone the same
[15:21:58] <alex_joni> giacus_: it looks great
[15:22:07] <giacus_> at least as introduction to Emc2
[15:22:08] <alex_joni> the pictures are a bit small (too thin lines)
[15:22:22] <giacus_> alex_joni: I seen images resolution are not geat
[15:22:30] <giacus_> yeah
[15:22:35] <giacus_> I know ..
[15:22:51] <giacus_> soon will looks better
[15:23:30] <giacus_> I like tr introduction part
[15:23:51] <giacus_> its very friendly to introcude HAL concepts
[15:24:06] <giacus_> I cutted just few part of them ..
[15:25:54] <giacus_> Of course I've the oo file too, if anyone want to join helping the translation :D
[15:26:20] <alex_joni> http://www.linuxcnc.org/Hal_Introduction_IT.pdf
[15:26:24] <giacus_> K
[15:27:16] <giacus_> lets see Axis now ;)
[16:19:43] <gene> Mornin all;
[16:41:19] <bill2or3> g'morning.
[16:42:35] <gene> I see I'm not the only one moving slowly this morning, still working on the first cup of tea etc...
[16:42:45] <SWPadnos> just started the coffee
[16:42:55] <jepler> I've had 3 cups .. I don't think it's helped.
[16:42:59] <SWPadnos> maybe I should get dressed before continuing ... ;)
[16:43:02] <gene> Pour me a cup when its done
[16:43:10] <SWPadnos> will do
[16:43:47] <gene> Nah, this thing doesn't have pix, thanks to whomever had the foresight to leave that out..
[16:44:04] <SWPadnos> pictures didn't exist when it was invented ;)
[16:44:13] <gene> Touche`
[16:44:34] <gene> besides, my webcam is a POS
[16:45:02] <SWPadnos> I could email you a 12 MPixel photo of coffee, if you like
[16:45:40] <gene> Nah, the spam filters would remove all the caffiene I think
[16:45:57] <SWPadnos> they'd remove all the salt, but caffeine???
[16:46:25] <gene> Using Spamassassin here, well trained
[16:46:54] <gene> what TZ are you in Jeff, that you've had time for 3 cups?
[16:47:50] <gene> But then maybe I shouldn't question that, its darned near noon here...
[16:48:12] <skunkworks> 10:45
[16:48:33] <skunkworks> and getting near lunch
[16:48:52] <gene> Your watch is slow it 11:48:50 here
[16:48:59] <skunkworks> I rounded ;)
[16:49:05] <cradek> 10:48:50 < gene> Your watch is slow it 11:48:50 here
[16:49:39] <gene> Ahh, this is ntp time +- a millisecond or so
[16:50:13] <gene> Couple of things I observed yesterday about emc2
[16:51:03] <gene> 1: it tends to lose track of being in the manual mode, requireing it be set to something else, and then back before the keyboard ot mouse can run an axis
[16:51:48] <cradek> gene: need exact steps for how to reproduce that if you can figure them out
[16:52:40] <gene> 2: I was able to carve a hole using the g2 g17 commands from the mdi, but it failed when put into a program, saying that R i j k were missing.
[16:53:44] <gene> and 3: running that carving routine, I found that it went berzerkers if the feedrate was cloicked down to say 15%
[16:53:49] <cradek> gene: for that one, please put together the shortest program that reproduces the bug so someone can look at it
[16:54:11] <cradek> "berzerkers"?
[16:55:13] <gene> Thats 2 above, yeah, it would goto the rear about 1/2" and carve a very small version of the loop, recovering fully when the feedrate was raised back up to 20 or above.
[16:55:40] <gene> That showed in the backplot, and in the piece of wood I was wasteing at the moment
[16:56:29] <skunkworks> I found out yesterday that I can lock up axis by programing a never ending loop in gcode ;)
[16:57:10] <alex_joni> gene: hi
[16:57:13] <skunkworks> Cradek: didn't you fix an issue with slow feedrates that jepler had found - what are you running for emc2 (head - how old)
[16:57:21] <gene> cradek: the command was something like "g2g17x-.620 y0.0z-.010r.620
[16:57:34] <cradek> skunkworks: I fixed all sorts of problems in the last day or two
[16:57:36] <gene> HEAD about 4 days old now
[16:57:40] <alex_joni> gene: we need a complete program to be able to test
[16:57:46] <alex_joni> gene: take the latest HEAD first
[16:57:53] <alex_joni> lots of things were changed lately
[16:58:08] <jepler> gene: Some blending fixes, including one that affected changing the feed override during an arc, have been fixed recently.
[16:58:15] <jepler> (maybe more than one)
[16:58:28] <gene> Ok, that I can do, back in a bit, good Jepler
[16:58:39] <jepler> Cradek's the one that fixed the bugs .. I just helped find 'em.
[16:59:13] <skunkworks> ^ that is about all I am good for ;)
[17:00:21] <jepler> gene: here's an example of a torture test program that I ran recently to test the TP: http://emergent.unpy.net/files/sandbox/torture-xyza.png
[17:01:21] <jepler> (though that particular test was to make sure machine limits were respected, not checking that the blends were "sensble"
[17:01:23] <skunkworks> jees - how did you crate that? Modified spiro?
[17:01:39] <gene> vcs has a tummy ache again
[17:01:42] <giacus_> jepler: in the .po file there are string using '_', like this: _Front view, what if I change the translated string in Fr_ontale ? are they for control+keys combinations ?
[17:01:47] <gene> cvs has a tummy ache again
[17:02:22] <jepler> giacus_: That's for menus. If you change it to Fr_ontale, then O (or alt-O for a top-level menu label) will activate the item.
[17:02:28] <alex_joni> jepler: that's almost art
[17:02:35] <giacus_> so, can work then ..
[17:02:39] <jepler> skunkworks: yeah, that spiro to do "A" moves too
[17:02:43] <giacus_> jepler: ok,
[17:02:45] <skunkworks> wow
[17:02:52] <jepler> gene: Here's an example of a bad blend that cradek has since fixed: http://emergent.unpy.net/files/sandbox/weird-blend-1.png
[17:03:06] <jepler> gene: and another, also fixed: http://emergent.unpy.net/files/sandbox/weird-blend-2.png
[17:03:39] <jepler> alex_joni: yeah, I agree .. it's pretty
[17:04:00] <cradek> this is the fixed weird-blend-1: http://timeguy.com/cradek-files/emc/better.png
[17:04:01] <gene> Which front end is that a shot of?
[17:04:14] <cradek> gene: AXIS
[17:04:15] <jepler> gene: those photos are all of AXIS
[17:04:58] <skunkworks> axis for the win!! Gene - you realy should try it. Awsome.
[17:05:07] <gene> Oh, I have it but haven't built it. Does the configure script look for it & if so, where?
[17:05:39] <gene> and cvs finally let me in, I just kept banging on the door at 2 second intervals
[17:05:51] <skunkworks> gene: been there ;)
[17:06:13] <jepler> gene: are you still on bdi or did you make the plunge and try ubuntu?
[17:06:39] <gene> I've got the cd burnt but still on bdi-4.30 out there yet.
[17:06:57] <jepler> gene: Try the instructions here: http://axis.unpy.net/installing
[17:07:14] <jepler> sudo apt-get install xlibmesa-dev libxmu-dev python2.3-dev
[17:07:20] <jepler> sudo env EMCROOT=/path/to/emc2 python setup.py install
[17:08:01] <gene> hEAD is building again, same instructs I assume?
[17:08:40] <jepler> "./configure --enable-run-in-place && make && sudo make setuid"
[17:09:27] <gene> Yoy left out the make clean :)
[17:09:57] <gene> s/Yoy/You
[17:10:06] <gene> And I just printed that lnk, thanks
[17:12:54] <gene> And the routine I tried to write that wont run due to
[17:13:00] <gene> R i j k words all missing for arc
[17:13:06] <gene> is fairly short
[17:13:17] <alex_joni> gene: put it in pastebin.com
[17:15:29] <gene> http://pastebin.com/594762
[17:16:40] <gene> The error is still there with todays HEAD, but no idea which way to point the finger, it might be like the margolin taget pistol we've all see cartoons of.
[17:17:03] <jepler> gene: On line 13 (pastebin.com numbering) you write "Z#4"
[17:17:14] <jepler> gene: But g2 from line 12 is modal, so it "sticks" and is reused on line 13.
[17:17:24] <jepler> gene: Maybe you want to write G0 Z#4 or G1 Z#4
[17:17:37] <gene> Aha, it was that cartoon!
[17:17:53] <jepler> ummm...?
[17:18:01] <gene> Lemme see if that fixes it
[17:18:30] <gene> in that cartoon, the barrel is mounted backwards
[17:20:59] <bill2or3> does emc work w/ usb-parallel adapters?
[17:21:00] <gene> Yur right as usual. So when I want to just do a move, I have to prefix it with the moves mode then.
[17:21:19] <jepler> bill2or3: no
[17:21:21] <alex_joni> not usually
[17:21:27] <bill2or3> bummer.
[17:21:30] <bill2or3> timing issues?
[17:21:31] <alex_joni> gene: only if you want a different one
[17:21:37] <gene> I doubt it Bill2or3, the latency of the usb connection will kill you or the machinery.
[17:21:48] <alex_joni> bill2or3: latency, yes (timing)
[17:21:55] <SWPadnos> bill20r3, usb = crap for realtime anything
[17:22:03] <gene> Amen
[17:22:10] <bill2or3> gotcha.
[17:22:22] <alex_joni> bill2or3: basicly USB is good for lots of throughput, bad for big packet rate (aka small latency)
[17:22:29] <gene> about the only thing its good for is mice, but mine has betteries anyway
[17:22:34] <SWPadnos> I could expound on that: usb = total, utter, amazingly useless crap for anything RT
[17:22:36] <gene> about the only thing its good for is mice, but mine has batteries anyway
[17:22:58] <SWPadnos> in case therer was any doubt ;)
[17:23:00] <SWPadnos> there
[17:23:05] <jepler> USB is fine for stuff where "low latency" == 50ms
[17:23:07] <bill2or3> heh.
[17:23:39] <SWPadnos> true, I suppose I should have said "REAL RT, not that slowpoke stuff they tell you is RT" ;)
[17:23:46] <gene> Humm program still died at end of first move
[17:24:16] <gene> I added the g1 so maybe my test logic is foobar too.
[17:24:27] <jepler> gene: same error, or different error?
[17:24:56] <gene> none reported, highlighted line is the last Z#4 of the loop
[17:25:41] <gene> and the second m101 was shown, so my end test must be wrong
[17:25:53] <SWPadnos> if you run axis, you can see what the toolpath might be ... (hint, hint)
[17:26:14] <SWPadnos> or did you switch when I was getting coffee?
[17:27:49] <jepler> gene: The O#s of the 'do' and the 'while' need to match, don't they?
[17:28:20] <gene> Humm, I assumed they were sequential, wrong?
[17:28:40] <jepler> I don't really know, to tell you the truth
[17:28:43] <gene> running backplot
[17:28:45] <jepler> my advice to everyone is "don't use O-words"
[17:28:57] <SWPadnos> they must match
[17:29:03] <SWPadnos> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Oword
[17:29:42] <SWPadnos> at least for every other O word type
[17:31:52] <gene> matching make it work, thanks
[17:33:15] <alex_joni> jepler: would it be hard to extend axis to plot for genhexkins?
[17:33:31] <gene> I can also see I'll need to slow it down to about 1/3rd of that speed when cutting 1/8" plate with that
[17:33:32] <alex_joni> e.g. XYZ + ABC
[17:33:48] <alex_joni> in the case where the tooltip tilts in 3 directions (ABC)
[17:35:10] <jepler> alex_joni: yes. I haven't really finished even the "XYZA" case
[17:35:27] <jepler> alex_joni: When composing the ABC rotations, how do you know the order to use?
[17:35:29] <alex_joni> jepler: I'm confident in you ;)
[17:35:34] <alex_joni> what do you mean?
[17:36:06] <alex_joni> no order, they apply to the same point..
[17:36:36] <alex_joni> when moving only ABC, the tooltip stays in the same point, and only tilts
[17:36:43] <gene> sudo apt-get of that list wants to remove these:libglu1-mesa libglu1-mesa-dev mesa-common-dev mesag-dev mesag3
[17:37:14] <gene> shall i let it
[17:37:42] <gene> following the axis instructs here
[17:37:49] <alex_joni> go ahead
[17:39:43] <gene> url for axis 1.3?
[17:39:50] <alex_joni> axis.unpy.net
[17:39:56] <alex_joni> ->downloads
[17:41:22] <jepler> gene: If you want the 1.3 (development) version, click the "CVS Snapshots" link.
[17:41:41] <jepler> gene: There's also 1.1.3 and 1.2rc3 which are "stable" versions.
[17:42:46] <CIA-8> 03alex_joni * 10emc2/configs/vti/ (README vti.ini vti.tbl vti.var vti_io.hal vti_motion.hal): initial version of VTI configs
[17:43:08] <gene> I just dl'd 1.3
[17:44:20] <SWPadnos> nightly snapshots are 1.3 now, right?
[17:44:52] <jepler> SWPadnos: yes
[17:45:42] <SWPadnos> ok, thanks
[17:45:51] <gene> unpacked and in that tree, next?
[17:46:34] <SWPadnos> sudo env EMCROOT=/path/to/emc2 python setup.py install
[17:47:28] <gene> gene@shop:~/axis-1.1.3$ python setup.py install
[17:47:30] <gene> Traceback (most recent call last):
[17:47:32] <gene> File "setup.py", line 36, in ?
[17:47:33] <gene> emcroot = os.path.abspath(os.getenv("EMCROOT", None) or find_emc_root())
[17:47:35] <gene> File "/usr/lib/python2.3/posixpath.py", line 395, in abspath
[17:47:36] <gene> if not isabs(path):
[17:47:38] <gene> File "/usr/lib/python2.3/posixpath.py", line 49, in isabs
[17:47:40] <gene> return s.startswith('/')
[17:47:41] <gene> AttributeError: 'NoneType' object has no attribute 'startswith'
[17:47:50] <gene> Oh, I see, that ALL one line, ok
[17:47:52] <SWPadnos> you forgot the env EMCROOT=/path/to/emc
[17:47:53] <alex_joni> gene: use the whole line SWPadnos pasted
[17:47:53] <SWPadnos> yep
[17:48:11] <alex_joni> and the sudo
[17:49:19] <gene> Looks like that was fine, now to go change the .ini to axis
[17:49:29] <alex_joni> DISPLAY = axis
[17:51:33] <alex_joni> hi & bye
[17:51:39] <CIA-8> 03alex_joni * 10emc2/src/Makefile: hope I got it right to add VTI to the common file copying
[17:51:40] <gene> did that, got this: Traceback (most recent call last):
[17:51:42] <gene> File "/home/gene/emc2/bin/axis", line 53, in ?
[17:51:44] <gene> from rs274.OpenGLTk import *
[17:51:45] <gene> File "/usr/lib/python2.3/site-packages/rs274/__init__.py", line 19, in ?
[17:51:47] <gene> from errorlist import *
[17:51:48] <gene> File "/usr/lib/python2.3/site-packages/rs274/errorlist.py", line 18, in ?
[17:51:50] <gene> from gcode import strerror, RS274NGC_MAX_ERROR, RS274NGC_MIN_ERROR
[17:51:51] <gene> ImportError: /home/gene/emc2/lib/librs274.so: undefined symbol: _Z31GET_EXTERNAL_SELECTED_TOOL_SLOTv
[17:51:53] <gene> Shutting down and cleaning up EMC...
[17:52:03] <alex_joni> gene: you're not running the latest EMC
[17:52:09] <alex_joni> or not the latest AXIS
[17:52:14] <gene> I just built HEAD
[17:52:25] <gene> and axis-1.3, fresh dl
[17:52:57] <alex_joni> gene: use this one: http://axis.unpy.net/files/downloads/nightly/axis-latest.tar.bz2
[17:53:23] <alex_joni> gene: where did you get 1.3?
[17:53:27] <gene> at least I think I did, I used cvs up -dP and it was set to HEAD I think
[17:53:54] <gene> http://axis.unpy.net/downloads/01141952313
[17:54:32] <alex_joni> that's 1.1.3
[17:54:34] <alex_joni> not 1.3
[17:54:47] <SWPadnos> wget -O axis-latest.tar.bz2 http://axis.unpy.net/files/downloads/nightly/axis-latest.tar.bz2
[17:55:15] <jepler> gene: That GET_EXTERNAL_SELECTED_TOOL_SLOT is a bug I thought I fixed just before making the 1.1.3 release.
[17:55:21] <jepler> gene: Maybe I messed it up
[17:55:59] <gene> I just ganged on cvs again afdter 50 whacks its doing a "cvs up -dP -rHEAD"
[17:56:44] <giacus_> jepler: I done editing the p.o file, now to test it should I put it somewhere or put it in the source dir and recompile Axis to take effect ?
[17:56:47] <giacus_> .po
[17:57:04] <jepler> giacus_: Put it in the i18n/ subdirectory, then "make" in that directory, then reinstall axis.
[17:57:11] <giacus_> ok
[17:57:24] <jepler> gene: It looks like the 1.1.3 tarball from my website is broken for emc2. I'll correct it later. for now, try getting the CVS snapshot or something
[17:57:29] <jepler> I'll be back in an hour or two
[18:00:54] <gene> I just rebuilt everything from the wget you gave me
[18:03:30] <gene> Ok axis is running, one error on startup:Xlib: extension "XFree86-DRI" missing on display ":0.0".
[18:03:45] <gene> Buit I'm ssh'd into it
[18:04:08] <SWPadnos> oh no, not the SSH thing again ;)
[18:06:45] <gene> Yeah, but its running my now fixed code just fine. I LIKE this
[18:07:40] <gene> I'm not sure DRI is available to display 10:0 ndi why
[18:08:11] <SWPadnos> I wonder if DRI works remotely at all
[18:09:06] <gene> NDI Steve. Wheres the starter for halscope?
[18:09:56] <gene> Ah wait, thats not in the gui anyway, duh...
[18:10:01] <SWPadnos> nope
[18:10:15] <SWPadnos> halscope &, assuming that the scope_rt component is loaded
[18:12:38] <gene> Twould be nice, hint hint :)
[18:15:08] <gene> Yeah, I LIKE axis I think. I assume those buttons on the left move it manually in the axis dotted?
[18:15:41] <SWPadnos> they select the axis for home and stuff like that
[18:16:16] <gene> So far, 'home' just zero's the axis selected, thats still true?
[18:16:41] <SWPadnos> I think so, but you should ask someone who has an actual machine hooked to their computer ;)
[18:17:32] <gene> Well, I wouldn't want it to try to goto home when theres no home switches installed yet.
[18:17:50] <alex_joni> gene: depending on your ini, yes
[18:18:35] <gene> I'd assumed so, absent eveidence to the contrary.
[18:19:59] <gene> One thing I noted, the blending of the initial move was really obvious. For that, its not a problem, just an observation.
[18:21:17] <SWPadnos> starting from 0, and going to F15 would have a blend
[18:21:44] <SWPadnos> the other endpoints should be pretty minimal, since the feedrate is the same, and the direction of motion is pretty close
[18:21:54] <SWPadnos> other than the Z moves, of course
[18:21:56] <gene> I also see some segments that look like straight lines. I have the display update at 50millisends, is that too fast for axis?
[18:23:20] <gene> It appears to only be updateing the gfx window about 3x a second.
[18:24:53] <gene> Is there a site where I can post a shot of it?
[18:25:12] <SWPadnos> frappr?
[18:25:21] <bill2or3> uhm
[18:25:24] <alex_joni> gene: drop me an email if everything else fails
[18:25:38] <bill2or3> use scoot.net/gallery/bbs tell me when you're done with it so I can delete it.
[18:25:40] <SWPadnos> I get much faster updates than that on my Windows machine via remote X
[18:25:43] <bill2or3> it'll let you do an http upload
[18:26:29] <gene> Well, I think its just the gfx load on a 100mb connection mainly. I'll go check it in person before the day is done.
[18:26:42] <SWPadnos> that's the same as I have - 100Mb ethernet
[18:27:00] <gene> I also have a seeder going for ubuntu
[18:27:17] <gene> which may be effecting the switches throuput
[18:27:18] <SWPadnos> I'm using a USC, so I don't have a lot of step generation overhead
[18:27:53] <SWPadnos> can you put the latest version of your program on pastebin? I'd like to see it, and the axis update rate
[18:28:20] <gene> I'll go check in person before I gettoo worried. tkemc seemed to handle the 50ms nicely
[18:28:56] <SWPadnos> hmmm - maybe this doesn't update all that fast
[18:29:06] <gene> two routines really I'll put the past one on pastebin
[18:29:10] <SWPadnos> it really depends on the window size
[18:29:45] <gene> two routines really I'll put the last one on pastebin, its about 1/4 of a 1600x1200 screen in there 32 bit color of course
[18:30:00] <SWPadnos> at 1280x1024x24bit. it's pretty slow ;)
[18:31:35] <gene> http://pastebin.com/594882
[18:32:23] <gene> IIRC its running on a 1024x768 32 bit screen out there.
[18:32:57] <SWPadnos> ok.
[18:33:01] <SWPadnos> what about inside
[18:33:03] <SWPadnos> ?
[18:37:47] <gene> 1600x120032bit
[18:37:57] <gene> 19" screen these days
[18:38:19] <gene> How do I post a .png to pastebin?
[18:38:58] <SWPadnos> you don't
[18:39:17] <bill2or3> you can use scoot.net/gallery/bbs/
[18:39:24] <bill2or3> it's simple.
[18:39:43] <bill2or3> check the "don't resize" box
[18:42:52] <gene> I did, its there now at: http://scoot.net/gallery/bbs/index.html
[18:45:20] <Jymmm> gene: If you have a yahoo account, you can just login to http://flickr.com/
[18:45:29] <gene> Comment of the web page instructs for building axis. When printed, the fonts are about 4x whats needed, hence the confusion if the build/install line before.
[18:45:48] <gene> No yahoo account, political reasons, see my email sig
[18:46:26] <cncuser> hello all :)
[18:46:33] <gene> Howdy!
[18:46:53] <Jymmm> gene: Than how are yo posting to the yahoo groups?
[18:47:21] <gene> I'm only a member of the coco list
[18:47:30] <Jymmm> gene ah
[18:47:36] <gene> and have long since forgotten the password.
[18:47:49] <gene> email seems to go thru ok though
[18:48:47] <gene> that line was actually printed on 4 seperate lines!
[18:50:05] <gene> I'm gonna have to go find something for breakfast & go out there & play a bit, back later if I find anything too troublesome
[18:50:55] <gene> IIRC theres an irc client installed out tehre too, if I can recall what it is, this is firefox-Chatzilla here
[19:06:33] <giacus> what's the quickly way to change Axis language on fly ?
[19:06:36] <CIA-8> 03alex_joni * 10emc2/tcl/tkemc.tcl: fixed minor joggin bug related to -/= keys
[19:06:39] <giacus> locales ?
[19:07:27] <alex_joni> LANG=it
[19:07:32] <alex_joni> LANGUAGE=it
[19:07:39] <alex_joni> on ubuntu you might need both
[19:07:54] <giacus> ty
[19:09:04] <Jymmm> LANG=it ???? Sounds greek to me.
[19:10:24] <giacus> wonder why side view, front view, show.. etc are under Edit menu
[19:10:31] <giacus> in Axis
[19:10:42] <giacus> and not under View
[19:11:16] <SWPadnos> because "That's The Way It Is"
[19:11:22] <SWPadnos> (I agree, it's weird)
[19:11:26] <giacus> maybe plot could have his menu
[19:11:41] <SWPadnos> you're editing the plot parameters ;)
[19:11:58] <giacus> uhm .. ok
[19:13:27] <giacus> then lets see if I can find a different term then 'view' or show ..
[19:13:44] <giacus> set could be appropriate
[19:14:00] <SWPadnos> view is good, but not a translation of what's there
[19:14:24] <giacus> I should use repeated 'Visualizza' in differernt menu
[19:14:35] <giacus> that's why ..
[19:14:40] <giacus> sound bad ..
[19:16:39] <giacus> set 'imposta' under Edit should be better
[19:19:41] <jepler> I'm back
[19:19:43] <jepler> I see gene got axis to work?
[19:20:04] <jepler> giacus: Someday (maybe in axis 1.3) the menus will be reorganized. I agree the layout now is bad.
[19:20:35] <giacus> jepler: I'm thinking to use Plot instead of Edit
[19:20:49] <giacus> could be ok for all items
[19:21:02] <giacus> except the latest.. Hal configuration
[19:22:10] <giacus> and , If I can, I'd remove the repeated show terms under view also for eng
[19:22:46] <giacus> 5 time repeated
[19:22:59] <jepler> giacus: I'm not sure what to tell you. For axis 1.2 the menu layout is not going to change, and it would be more confusing if a user reads english-language instructions about the "Edit" menu, but finds only "Plot (in italian)"
[19:23:15] <giacus> uhmm
[19:23:17] <giacus> yeah
[19:23:35] <jepler> so I'd choose "Edit (in italian)" even though it doesn't describe most of the items very well at all
[19:24:09] <giacus> btw it should be replaced by modifica to get a sense
[19:24:38] <cradek> jepler and I had planned a new menu layout, but never did the reorganization. I'm not sure if we still have our notes or not.
[19:26:31] <alex_joni> yay, hexapod jogging in world view
[19:26:45] <alex_joni> with emc2 of course ;)
[19:26:59] <cradek> wow
[19:27:13] <alex_joni> that was close to trivial ;)
[19:27:14] <cradek> sounds great
[19:27:41] <giacus> can be 2 alt-R keys accelerators at the same time, one under Edit and one under View ?
[19:27:45] <alex_joni> cradek: what do you think about me commiting a config dir for that?
[19:27:46] <giacus> no, I suppose..
[19:27:54] <cradek> alex_joni: why not?
[19:28:00] <alex_joni> ok
[19:28:12] <alex_joni> the problem I have now is how to call it :D
[19:28:22] <cradek> hexapod?
[19:28:43] <alex_joni> yes and no
[19:28:50] <alex_joni> the kins are for generic hexapods
[19:29:01] <alex_joni> and based on the define inside they are of different types
[19:29:08] <SWPadnos> I'd like to see the servo-thread execution time with genhexkins
[19:29:09] <alex_joni> the config I have works for minitetra
[19:29:24] <alex_joni> maybe hexapod as a dir, and minitetra.ini ?
[19:29:42] <alex_joni> that makes the most sense
[19:29:44] <cradek> I don't know enough to comment
[19:29:49] <SWPadnos> separate dirs seems to be the way so far
[19:30:07] <SWPadnos> consider that every other sample config is for a cartesian machine
[19:30:27] <alex_joni> cradek: you need to provide special HOMEpositions (for TRAJ-world, and for the joints-AXIS_*)
[19:30:34] <alex_joni> and those need to match through the kins
[19:32:53] <alex_joni> ok, I'm putting a dir called hexapod-sim
[19:33:01] <alex_joni> with a minitetra.ini inisde
[19:58:23] <CIA-8> 03alex_joni * 10emc2/configs/hexapod-sim/ (README core_sim_6.hal hexapod.tbl hexapod.var minitetra.ini): initial hexapod config files
[19:59:17] <CIA-8> 03alex_joni * 10emc2/configs/hexapod-sim/.cvsignore: cvs add * doesn't expand .cvsignore :(
[20:01:06] <CIA-8> 03alex_joni * 10emc2/configs/vti/.cvsignore: cvs add * doesn't expand .cvsignore :(
[20:02:45] <CIA-8> 03alex_joni * 10emc2/configs/stepper-xyza/.cvsignore: initial checkin
[20:05:54] <CIA-8> 03alex_joni * 10emc2/src/ (Makefile.inc.in Makefile): first attempt at allowing easy chosing of kinematics
[20:13:04] <CIA-8> 03alex_joni * 10emc2/src/emc/kinematics/ (genhexkins.c genhexkins.h genhexkins.o): genhexkins initial commit
[20:18:03] <CIA-8> 03alex_joni * 10emc2/src/emc/kinematics/genhexkins.o: this shouldn't have been commited
[20:19:11] <SWPadnos> now here's a real machine:
[20:19:16] <SWPadnos> http://upload.wikimedia.org/wikipedia/en/8/8c/Zmachine.jpg
[20:20:17] <jepler> I've corrected the axis 1.1.3 tarballs.
[20:20:31] <jepler> It looks like nobody but gene ever downloaded it, so I won't change the version number or anything.
[20:20:37] <SWPadnos> heh
[20:21:35] <alex_joni> SWPadnos: old news ;)
[20:21:38] <alex_joni> but it's hot
[20:21:50] <SWPadnos> 2 billion K - yeah, it's hot!
[20:22:26] <SWPadnos> and they don;t know how it happened ;)
[20:22:39] <alex_joni> lol, yeah I loved that part
[20:22:48] <alex_joni> makes you wonder.. are we safe from all teh scientists?
[20:23:11] <SWPadnos> between science and the military, I'd say we're not safe ;)
[20:23:29] <alex_joni> lol
[20:23:32] <SWPadnos> but I'd rather that everyone be killed in the pursuit of knowledge than in the pursuit of power or wealth
[20:24:24] <SWPadnos> (if everyone's gonna get killed, that is)
[20:24:55] <alex_joni> cradek: still there?
[20:25:14] <cradek> yes
[20:26:11] <alex_joni> you didn't update docs/NEWS
[20:26:19] <jepler> alex_joni: I don't understand what you said earlier about the order of ABC not mattering. Matrix multiplication is not commutative in general, so if you have transformations R and S, RS may != SR.
[20:27:03] <CIA-8> 03alex_joni * 10emc2/docs/NEWS: some of the recent stuff
[20:27:29] <alex_joni> jepler: I can understand that.. but I don't see what the issue is
[20:28:57] <alex_joni> having certain ABC values, you have a very precise position
[20:29:09] <alex_joni> actually not position, tilting
[20:31:15] <SWPadnos> depending on how those axes are chained together
[20:31:22] <SWPadnos> on the physical machine
[20:31:25] <alex_joni> SWPadnos: talking hexapod here
[20:31:28] <SWPadnos> ok
[20:31:33] <alex_joni> there are no actual ABC axes on the machine
[20:31:41] <SWPadnos> right, got it
[20:31:42] <alex_joni> that's the tool-tip tilting
[20:32:17] <alex_joni> I agree on trivkins + rotaries it's not very easy to define
[20:32:48] <alex_joni> but for a full carthesian machine, or even for a trivkins but with the tilting axes on the head it is
[20:32:52] <dmessier> hi all
[20:32:53] <alex_joni> I imagine ;)
[20:32:54] <SWPadnos> there are (at least) two ways to think about it
[20:32:57] <alex_joni> bon soir
[20:33:12] <SWPadnos> one, that the G-code should specify the position and TOV, and the machine kinematics sort out the rest
[20:33:14] <jepler> so you're saying that you just make the ABC values that come out of emc plus the code in the GUI give you the answer you want?
[20:33:24] <dmessier> TOV??
[20:33:29] <SWPadnos> two, that the G-code should be created with knowledge of the machine
[20:33:29] <alex_joni> you already do that for XYZ
[20:33:34] <alex_joni> dmessier: EXACTLY
[20:33:45] <alex_joni> Tool Orientation Vector
[20:33:47] <SWPadnos> right, so #1 is preferable for portability
[20:33:55] <dmessier> tool oreintation vector??
[20:33:56] <alex_joni> just like TCP (Tool Center Point)
[20:33:56] <SWPadnos> but #2 is preferable for the programmers ;)
[20:34:08] <dmessier> APT dream
[20:34:10] <alex_joni> TCP is for XYZ, TOV is for ABC
[20:34:11] <SWPadnos> which way the tool is held against the work, including tilt
[20:34:18] <SWPadnos> (that's basically TOV)
[20:34:39] <alex_joni> jepler: what code in the gui ?
[20:34:40] <dmessier> add xyzijk.. and it all to both
[20:34:43] <SWPadnos> it should be TEP for machining - tool end point
[20:34:54] <SWPadnos> unless that's what meant - the center of the machining tip
[20:35:08] <jepler> alex_joni: I thought you talked about adding code to AXIS to show the ABC
[20:35:12] <alex_joni> yeah, something like that.. we have the center of the welding arc.. so yes
[20:35:16] <SWPadnos> ok
[20:35:18] <alex_joni> jepler: only tilt the cone
[20:35:19] <dmessier> yes... but you cal lock out axes thru the ini file
[20:35:33] <alex_joni> jepler: using the ABC as the angles to tilt
[20:35:55] <dmessier> can
[20:35:59] <alex_joni> dmessier: that doesn't really work :(
[20:36:08] <SWPadnos> C is kind of useless for the GUI, I think
[20:36:20] <SWPadnos> since it's conceptually rotation that the spindle would do
[20:36:21] <dmessier> je comprend
[20:36:22] <alex_joni> if you specify X Y A in the ini, display will change, but kinematics won't
[20:36:25] <alex_joni> SWPadnos: no
[20:36:31] <alex_joni> you assume A=B=0
[20:36:46] <SWPadnos> yes I do, since the rotation is relative to the tool
[20:36:54] <alex_joni> if you first tilt A then C will rotate it
[20:37:07] <SWPadnos> that's where the non-commutativity of matrices comes into play
[20:37:37] <dmessier> you dont want to specify xya at any time... xyz ijk ijk is a tool axis vector... kimematics HAS to sort it out
[20:37:56] <dmessier> to determine an A
[20:38:08] <SWPadnos> sure, and that A is machine-dependent
[20:38:34] <alex_joni> hmm.. think we're not talking about the same thing here
[20:38:39] <alex_joni> XYZ is clear
[20:38:43] <dmessier> depedant upon you mapping your kinematics properly yes
[20:38:59] <alex_joni> I meant Alpha, Beta & Gamma
[20:39:07] <alex_joni> which are rotations around X Y and Z
[20:39:21] <alex_joni> at least those are the ones I know as standard
[20:39:25] <dmessier> it is a init vector pointing straight UP the tool axis
[20:39:30] <dmessier> unit
[20:39:48] <alex_joni> ok, you can put it that way too
[20:39:53] <alex_joni> and call it ? ijk ?
[20:40:17] <SWPadnos> I think it depends on whether you're rotating the coordinate system, or rotating the tool orientation within that coordinate system
[20:40:30] <alex_joni> coordinate system stays fixed
[20:40:34] <dmessier> with xyz ijk... the post/ tp sorts out what axis needs to move to put me there
[20:40:35] <SWPadnos> that's the basic difference between rotary tables and pivoting heads
[20:40:54] <alex_joni> ok, I'm talking only pivoting head here
[20:41:04] <dmessier> doesnt matter .. adding a matrix... is cool
[20:41:07] <alex_joni> no rotary tables involved
[20:41:22] <alex_joni> dmessier: allowing all possibilities gets you into trouble
[20:41:32] <alex_joni> because you need to specify how the axes are stacked
[20:41:49] <dmessier> we run horizontal code on a vertical profiler with an angled head...
[20:41:58] <SWPadnos> for a pivoting head, you can either pivot and rotate (A or B, plus C)
[20:42:21] <SWPadnos> or you can pan and tilt, like a BP
[20:42:25] <dmessier> rotate... translate...
[20:42:45] <alex_joni> SWPadnos: that's basicly kinematics
[20:42:51] <dmessier> about a guage point Plus tool length...
[20:43:02] <alex_joni> I think any machine can be written to be carthesian kins converted to joints
[20:43:21] <dmessier> thats what in sayin.. IF you set it up correctly it should work
[20:43:37] <alex_joni> so basicly if you want to tilt the BP head it will try to keep the position constant by moving XZ too
[20:43:41] <alex_joni> or something like that
[20:44:06] <dmessier> if thats what you whated...
[20:44:21] <alex_joni> dmessier: why not?
[20:44:38] <alex_joni> it's about the only way of specifying 6-axes g-code without going mad I can imagine ;)
[20:45:06] <dmessier> it could roll an indicator around a ball on the table and NOT have it move but all axes would be in motion all the time..
[20:45:28] <alex_joni> yes
[20:45:43] <alex_joni> ok, lets step a bit back from actual machines
[20:45:52] <dmessier> i can do that in 200 lines or less
[20:46:05] <alex_joni> lets think about carthesian and theory
[20:46:15] <dmessier> talk to me
[20:46:18] <alex_joni> if you want to specify a tool in the carthesian space
[20:46:30] <alex_joni> you need 3 axes for position and 3 for orientation
[20:46:32] <alex_joni> right?
[20:46:48] <alex_joni> so you have XYZ distances from (0,0,0)
[20:47:00] <alex_joni> and ABC angles from Ox,Oy,Oz
[20:47:19] <alex_joni> dmessier: following so far?
[20:47:47] <dmessier> si
[20:47:57] <alex_joni> the A,B,C (actually Alpha, Beta & Gamma) are the orientation of the tool
[20:48:06] <alex_joni> and that is all you need in world space
[20:48:09] <dmessier> negative
[20:48:10] <alex_joni> aka carthesian
[20:48:14] <alex_joni> how so?
[20:50:11] <dmessier> ABC are the coresponding angular correlations to the TOOL AXIS VECTOR... if i feed the same xyzabc into a boring mill.. from your vert were screwed
[20:50:38] <alex_joni> hang on.. talking about world now
[20:51:14] <alex_joni> I start to think that a boring mill is actually a trivkins machine with some external rotary run in joint space
[20:51:15] <dmessier> ok
[20:51:28] <alex_joni> does that make any sense?
[20:51:38] <alex_joni> imagine a small mill.. 3 axes
[20:51:44] <dmessier> can be yes
[20:51:45] <alex_joni> XYZ-> trivkins, simple
[20:51:53] <alex_joni> if you add another rotary
[20:52:03] <alex_joni> it will be a XYZA or whatever you call it
[20:52:07] <dmessier> what about the facing head...
[20:52:14] <alex_joni> but if you drive it like that you run it in joint mode
[20:52:18] <alex_joni> without proper kins
[20:52:20] <dmessier> or bottle boring tool
[20:52:43] <alex_joni> if you would add proper kins, then you would have XYZABC and it won't be possible to get some tilting or whatever
[20:52:57] <alex_joni> because of the joint limitations
[20:53:02] <giacus> jepler: check your mailbox
[20:53:14] <dmessier> i can do that mechanically
[20:53:30] <dmessier> add a big fat joint
[20:54:31] <dmessier> and orbit around to an axis and drill a 5 axis hole on a 3 axis machine
[20:55:00] <alex_joni> ok, I can see that.. but you would have to specify it in kins for emc to be able to make full use of it
[20:55:03] <dmessier> been there .. done that..
[20:55:15] <alex_joni> or rather you have the knowledge, and let emc think it's something else
[20:55:25] <dmessier> yes of course
[20:55:34] <alex_joni> might work too.. but can mean trouble
[20:55:55] <dmessier> theres video from the tool show... ; )
[20:56:05] <alex_joni> ok.. so the problem is the GUI needs to know about what kind of things are inside emc to be able to plot ABC
[20:56:13] <dmessier> Montreal tool show 2001
[20:56:19] <K4ts> hello
[20:56:26] <giacus> hi K4ts
[20:56:30] <giacus> im ready ..
[20:56:34] <giacus> just a sec.
[20:56:37] <alex_joni> but the most simple case (call it trivkins6 ;) might be to only pan/tilt/rotate the tool
[20:56:45] <dmessier> yes.. the gui is blind to the machine right now..
[20:56:50] <alex_joni> and keep the tool end fixed
[20:56:52] <K4ts> yes devo dirti cos importante
[20:57:02] <giacus> skype ..
[20:57:08] <K4ts> yes
[20:57:23] <dmessier> si
[20:57:32] <alex_joni> that will work for hexapods, robots, etc, all of those machines where you have axes stacked onto another
[20:57:32] <giacus> hehe
[20:57:45] <K4ts> dmessier you are itailen?
[20:57:58] <alex_joni> but not on machines where you have a XYZ in one place, and some rotaries on the table
[20:57:58] <dmessier> yes... if you build them correctly
[20:58:01] <K4ts> italiano?
[20:58:30] <alex_joni> actually the difference is if the rotaries move the head or the workpiece
[20:58:45] <alex_joni> I'm talking about #1 which is pretty easy
[20:58:58] <alex_joni> #2 is hard/impossible without knowing specifics about setup
[20:59:26] <staggerlytom> alex is #1 moving the tool?
[20:59:38] <staggerlytom> alex: is...
[20:59:39] <alex_joni> yes, but we're not talking about moving
[20:59:51] <alex_joni> about tilting/panning/rotating
[20:59:55] <alex_joni> but yes, the tool
[20:59:57] <dmessier> do work in kinematicks.. you specify specifiks about the set-up... you dont ask... ; )
[21:00:00] <alex_joni> and #2 the workpiece
[21:00:11] <staggerlytom> ok
[21:00:26] <dmessier> rotable
[21:00:37] <alex_joni> dmessier: I'm not really eager about #2, except if it's for axes >6
[21:00:41] <dmessier> or tilaxs
[21:00:48] <alex_joni> say an 6 axes robot + 3-4 external axes
[21:01:23] <alex_joni> but preview in that case is really hard to do
[21:01:51] <alex_joni> jepler: are you following?
[21:02:04] <dmessier> its all a matter of mapping allthe rgisters and adding the necessary matrices to stack up the various components
[21:02:44] <alex_joni> and a matter of providing such hooks to the user
[21:03:10] <alex_joni> err.. such simple hooks
[21:03:32] <dmessier> could there be a database to propogate for the setup of your kins..
[21:04:02] <alex_joni> dmessier: yeah, I think this will be in emc-2.8.x
[21:04:11] <alex_joni> right now 2.0.0 is not fully done ;)
[21:04:24] <dmessier> step it up
[21:04:28] <alex_joni> hehehe
[21:04:38] <alex_joni> btw, I ran genhexkins with emc2 tonight
[21:04:40] <alex_joni> works ok
[21:04:53] <staggerlytom> on the 3 wire machine?
[21:05:05] <dmessier> i am thinkin of going after some older iron and strippin it...
[21:05:50] <alex_joni> staggerlytom: that was ages ago (that was tripodkins)
[21:06:06] <alex_joni> now I was running a 6-axes sim setup
[21:06:09] <alex_joni> it's in CVS
[21:06:14] <staggerlytom> k
[21:06:48] <dmessier> any thoughts for probe input..
[21:07:10] <alex_joni> lots, but not tonight ;)
[21:07:22] <dmessier> undertood
[21:07:38] <alex_joni> it's not that I don't want to talk about it ;)
[21:07:49] <alex_joni> I thought about adding the hooks for it
[21:07:55] <dmessier> dont you ... forget about me..... i think i see a song there
[21:07:58] <alex_joni> but didn't felt motivated enough :)
[21:08:16] <alex_joni> do you need probing?
[21:08:47] <dmessier> everyone needs probing... for lights out factory to work properly
[21:08:55] <staggerlytom> I need probing ( dont tell the aliens though ) for edm
[21:08:57] <alex_joni> dmessier: ok, noted
[21:09:09] <alex_joni> that increased my motivation :)
[21:09:41] <staggerlytom> and for cycle interrupt ( mid program inspection ) i need fast position capture on interrupt
[21:09:54] <staggerlytom> we need
[21:10:00] <staggerlytom> ;-)
[21:10:06] <dmessier> i have a plan to save the planet but i want to do it with EMC technology... so pick it up Alex...
[21:10:19] <alex_joni> dmessier: check your /msg's
[21:10:29] <CIA-8> 03alex_joni * 10emc2/src/Makefile: getting emc.nml for hexapod-sim too
[21:12:16] <alex_joni> ok, mini is basicly unuseable for nontrivkins, unless I'm missing something
[21:13:11] <dmessier> nice Bong... How DO you clean it... ; )===~~~~~
[21:14:13] <alex_joni> hehehe
[21:14:14] <SWPadnos> alex_joni, that's something that really poits to using option #1 in G-code (ie, XYZ RPW, not joint positions)
[21:14:19] <SWPadnos> points
[21:14:34] <alex_joni> what's RPW?
[21:14:39] <alex_joni> vs. ABC ?
[21:14:42] <SWPadnos> alpha beta gamma ;)
[21:14:50] <alex_joni> ohh.. those are the ones I like
[21:14:52] <SWPadnos> roll pitch and yaw
[21:14:53] <dmessier> ijk can rectify.. joints too guys
[21:15:09] <SWPadnos> (W since Y is taken, I guess)
[21:15:12] <alex_joni> jepler: I have some problems with AXIS
[21:15:23] <alex_joni> how about Q ?
[21:15:33] <alex_joni> actually those are from planes I think
[21:15:37] <alex_joni> Roll Pitch Yaw
[21:15:41] <alex_joni> or whatever
[21:16:02] <SWPadnos> whichever
[21:16:39] <SWPadnos> anyway, the G-code has to specify the position of the tooltip in XYA, and the orientation of the tool in (whatever you call the rotary axes)
[21:16:46] <alex_joni> ok, RPW
[21:16:53] <alex_joni> perfect
[21:17:02] <SWPadnos> or else everyone goes insane trying to get all the GUIs and all the machines working together
[21:17:07] <alex_joni> so for XYZRPW it's pretty easy to make it show nicely in AXIS
[21:17:16] <SWPadnos> I have no idea ;)
[21:17:18] <alex_joni> and for XYZABC you can't
[21:17:35] <alex_joni> RPW just specify the tilting of the cone
[21:17:43] <SWPadnos> I think what jepler has been working on is basically RPW, even though it's called ABC
[21:17:56] <SWPadnos> right - that's what he's been doing
[21:18:08] <alex_joni> OK, that's what I was saying all along
[21:18:15] <alex_joni> probably expressed it poorly
[21:18:18] <SWPadnos> the problem is, that there do have to be two different implementations for rotary axes
[21:18:24] <alex_joni> yeah sure
[21:18:30] <alex_joni> RPW vs. ABC
[21:18:48] <alex_joni> ideally? you would have RPW for the above case
[21:18:48] <SWPadnos> because the most common use is to have a rotary table, which effectively changes the part coordinates relative to the world coordinates
[21:18:53] <alex_joni> and ABC for everything else
[21:19:01] <alex_joni> like some other linear axes
[21:19:07] <alex_joni> or who knows what
[21:19:18] <SWPadnos> cosnider that rotating a table in A will move the tool along the part, and that's what is intended
[21:19:32] <dmessier> the machine coord are what your trackin..
[21:19:36] <alex_joni> yeah, and of course you'll want to synchronize the speeds
[21:20:01] <SWPadnos> and you want the toolpath to be in part coordinates, not world coordinates (probably)
[21:20:14] <dmessier> all you need is inverse time programmed g code
[21:20:32] <alex_joni> dmessier: ok, complicate it even more
[21:20:37] <alex_joni> my head already hurts ;)
[21:21:06] <dmessier> program part... cut machine...
[21:21:13] <SWPadnos> also remember that machine gemetry effectively changes when you change tool length
[21:21:19] <alex_joni> cut machine?
[21:21:27] <alex_joni> cut the part, not the machine ;)
[21:21:31] <dmessier> coords
[21:21:34] <alex_joni> :P
[21:22:05] <dmessier> you monitor gauge line.. plus tool length in the tranforms
[21:22:20] <dmessier> oh it sucks to be us
[21:22:37] <dmessier> my head need a beer
[21:23:24] <alex_joni> beer & bong don't mix well
[21:23:27] <alex_joni> :-P
[21:23:52] <SWPadnos> bong and IRC have a hard time as well ;)
[21:24:02] <staggerlytom> dmessier: how many 6x6 matrices do you pass thru for each line with your control ( looking for how fast how complex ) ?
[21:27:16] <dmessier> im working theoretically stagg... but ive SEEn and run stuff stacked 4-5 deep.. with the 45 degree nutatung table
[21:27:57] <staggerlytom> nutating, drool :)
[21:28:22] <staggerlytom> just left Uni Michingan, they had nice 5 axis mills in lab
[21:28:41] <staggerlytom> freebies donated tothe Dr. Wu labs
[21:28:48] <dmessier> not all its cracked up to be.. fixtures end up being 45 deg angle plates..
[21:29:13] <staggerlytom> the extra axis is meant to simplify ( or elim ) fixtures
[21:29:25] <staggerlytom> and the handle crankers behind the fixtures
[21:29:29] <staggerlytom> like me
[21:29:59] <cncuser> hello
[21:30:05] <staggerlytom> ello
[21:30:18] <cncuser> alex: can you please reset my password ? i cant login and dont remember
[21:30:39] <cncuser> alex_joni: i am feeling a little transaty :)
[21:30:43] <cncuser> translaty
[21:31:06] <SWPadnos> you should get that checked by a doctor ;)
[21:31:16] <alex_joni> cncuser: on which site=?
[21:31:19] <cncuser> joomla
[21:31:34] <alex_joni> cncuser: on which joomla? ;)
[21:31:37] <cncuser> harhar
[21:31:42] <cncuser> linucnc
[21:31:44] <alex_joni> I moved the site to cncgear.com/joomla/
[21:31:48] <cncuser> hgmm
[21:31:50] <cncuser> yes
[21:31:52] <alex_joni> but no translations
[21:31:54] <cncuser> http://cncgear.com/joomla/administrator/index.php
[21:31:58] <alex_joni> ok
[21:32:21] <dmessier> were you lab machines DMG's
[21:34:28] <staggerlytom> Taurus nutating head, other 2 machine prototypes built in the lab
[21:35:14] <K4ts> night
[21:35:39] <dmessier> i like to RUF too much for a nutating head to be any good to me... i like makin 6's and 9's...
[21:36:27] <staggerlytom> need a 6 axis droop&rein
[21:36:42] <staggerlytom> heavy duty enuf to ruf with
[21:37:53] <dmessier> but you cant finish with it... we have Mandelli's here
[21:38:30] <dmessier> we cant ruf wit... the head moves
[21:39:27] <staggerlytom> yah, they're massive.
[21:39:35] <dmessier> i just want my toshiba's back
[21:39:56] <dmessier> bigges....heaviest in class
[21:40:02] <CIA-8> 03alex_joni * 10emc2/src/emc/task/emccanon.cc: think this fixes G64Pxx for non-inch machines too
[21:40:45] <staggerlytom> big toshiba's , never seen > 1m cube, but limited experience with them
[21:40:53] <dmessier> btd120-R15... is a money macking machine...
[21:41:25] <dmessier> 25 M... bf150-250
[21:41:30] <staggerlytom> what type work?
[21:42:00] <dmessier> power gen... steel and alu mills
[21:42:15] <dmessier> foundries
[21:42:34] <dmessier> they never say toshiba on them...
[21:42:44] <staggerlytom> ok, BHEL in india, turbine were 4 to 4.5 meters long, ok
[21:43:16] <dmessier> whats the problem???
[21:43:45] <staggerlytom> huh?
[21:44:01] <SWPadnos> SWPadnos is now known as SWP_Away
[21:44:37] <dmessier> can it be kellered... all the way up... there are 6 M of y axis and a rotary too
[21:44:54] <staggerlytom> i worked with stationary power systems in india Bharat Heavy Electric something.. sounds like what you said
[21:45:24] <dmessier> should be doable..
[21:46:35] <dmessier> i gotta fly... hoeydooos
[21:46:42] <dmessier> honeydoos
[21:46:43] <alex_joni> byee
[21:47:21] <dmessier> kinematics needs a boost...
[22:15:45] <NickServ> This nickname is owned by someone else
[22:15:45] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[22:24:26] <CIA-8> 03alex_joni * 10emc2/src/emc/task/emccanon.cc: forgot to convert at testphase too
[22:47:22] <dmessier> comme s'tai
[22:47:47] <K4ts> wue dmessier ?
[22:48:48] <K4ts> night
[23:07:37] <dmessier> alex.. on kinematicks... what matrix capabilities do we have available???
[23:09:34] <dmessier> we hav many machines without kinematics,, we have some VERY sophisticated center of rotation macros
[23:28:51] <bill2or3> * bill2or3 wishes he understood.
[23:30:08] <skunkworks> * skunkworks skunkworks understands but can't explain it
[23:30:14] <skunkworks> oops
[23:30:32] <bill2or3> heh
[23:31:43] <dmessier> i hate when that happens
[23:33:39] <bill2or3> I may start construction on my mill-project this weekend, if I get the time.
[23:33:53] <bill2or3> wish me luck.
[23:34:20] <skunkworks> Good luck. Don't cut any fingers off
[23:35:02] <dmessier> do the wiring firs... dont burn you digits
[23:35:27] <bill2or3> I'm doing the mechanicals first, I'm not worried about the wiring.
[23:39:45] <dmessier> enable and chk limits and E-stops before the drives are powered up
[23:40:19] <SWP_Away> don't look twice at a cross-eyed newt
[23:40:23] <SWP_Away> SWP_Away is now known as SWPadnos
[23:41:06] <mdynac_> anyone out there?
[23:41:22] <SWPadnos> nope
[23:41:53] <mdynac_> i am looking for the new emc2 web page...anyone know the address?
[23:42:04] <dmessier> but they ARE the sweet ones in cajun cunty
[23:42:18] <SWPadnos> hmmm - not sure what the new emc2 web page is
[23:42:26] <SWPadnos> what information are you looking for?
[23:42:27] <mdynac_> k
[23:42:44] <mdynac_> just wanna see what's on it....
[23:42:53] <SWPadnos> the wiki is a good place to start, and it looks much nicer now than it did a week ago
[23:42:58] <dmessier> slow.... down.... dont move so fast..
[23:43:02] <mdynac_> working on the edm function for emc2.....
[23:43:14] <mdynac_> been there done that....
[23:43:21] <SWPadnos> there's a test site up at http://www.cncgear.com/joomla/
[23:43:26] <dmessier> i like that idea,, il listen...
[23:43:33] <SWPadnos> but I don't think any of the EDM discussions are there
[23:43:43] <mdynac_> alex joni is supposed to be making some hal pins available for edm functions
[23:44:12] <SWPadnos> I recall discussion of that idea. I don't think he's done it yet
[23:44:17] <dmessier> your in the right place
[23:45:10] <dmessier> his motivation might have been redirected for the time being
[23:45:43] <dmessier> im sure it wont fall off the waysid..
[23:46:09] <mdynac_> been out of town for a week, just checking to see what's done, wanna get something working for the cnc-workshop
[23:46:16] <dmessier> wire... & plunge??
[23:46:38] <mdynac_> just wire
[23:46:59] <dmessier> so you need power control??
[23:47:29] <dmessier> wire speed??
[23:47:31] <mdynac_> dynamic feedrate control in relation to spark gap voltage.....
[23:47:49] <SWPadnos> do you think you'll need reversal as well, or just feedrate control?
[23:48:12] <mdynac_> gonna need reversal also, but one thing at a time....
[23:48:23] <SWPadnos> ok.
[23:48:46] <dmessier> if alex told you the i/o's were available... im with him... i see a mild deficiency...in my head..
[23:49:08] <SWPadnos> I agree that feedrate isn't a complete colution
[23:49:09] <dmessier> calcs of course..
[23:49:16] <mdynac_> and that would be?
[23:49:34] <SWPadnos> solution
[23:49:42] <dmessier> the deficiency isnt IN MY HEAD...
[23:49:47] <SWPadnos> oh, sorry ;)
[23:50:14] <mdynac_> we just started this project, ray henry, alex, tom powderly, myself and some others are involved...
[23:50:21] <SWPadnos> mostly everything is there, except for the feedrate override HAL pin
[23:50:52] <mdynac_> yep
[23:50:54] <SWPadnos> I don't think there's a consensus about what teh GUI should show when an EDM HAL module is controlling FO
[23:50:59] <dmessier> HAL is wired to a power reg??
[23:51:15] <mdynac_> nope
[23:51:19] <SWPadnos> it can be - you can connect an analog input to a PID or scaling block
[23:51:30] <SWPadnos> you obviously need hte appropriate hardware
[23:51:35] <dmessier> actual... feed... like the rest
[23:51:39] <mdynac_> i am going to use the MotencLite ADC input......
[23:51:40] <SWPadnos> gah - I need to learn to spell "the"
[23:52:11] <SWPadnos> what you can't do is have a nonlinear equation on that A/D input
[23:52:24] <SWPadnos> scaler is Y=MX+b
[23:52:47] <mdynac_> just a 0 to 5vdc input to it.....
[23:52:52] <SWPadnos> or it might be (X-b) * M
[23:53:16] <SWPadnos> right, but the range of mathematical operations you can perform on the value is limited at the moment
[23:53:22] <dmessier> i could burn with that..
[23:53:59] <SWPadnos> yeah - if you can linearize the function in the external gap V / current measuring hardware
[23:53:59] <mdynac_> the adc is the maxim125 chip i think.....
[23:54:00] <dmessier> order some wire ;)
[23:54:39] <mdynac_> bronco cut!!!
[23:54:56] <SWPadnos> err - is this the rodeo channel? ;)
[23:55:00] <dmessier> your trying to build an adaptive controle guys... be happy when it cut RIGHT
[23:55:28] <mdynac_> bronco cut = edm wire
[23:55:31] <SWPadnos> EMC is already an adaptive control, in that it's the only consumer-affordable controller that actually uses feedback
[23:55:35] <SWPadnos> heh - OK.
[23:55:45] <dmessier> ive seen ROLLES a wire GONE BAD
[23:55:57] <SWPadnos> rusty?
[23:56:08] <A-L-P-H-A> hey peeps.
[23:56:15] <A-L-P-H-A> dmessier sup?
[23:56:17] <robin_sz> mmm edm!
[23:56:17] <SWPadnos> hi A-L-P-H-A
[23:56:19] <dmessier> brass doesnt rust
[23:56:20] <A-L-P-H-A> SWPadnos. hey.
[23:56:23] <A-L-P-H-A> brass tarnishes.
[23:56:29] <SWPadnos> got that tach done? ;)
[23:56:41] <A-L-P-H-A> tarnish = oxidation. oxidation = rust on iron.
[23:56:45] <A-L-P-H-A> SWPadnos. HAhahaha. no. :(
[23:56:53] <A-L-P-H-A> LOL.
[23:56:54] <SWPadnos> I don't do EDM, so I don't know what the wire is composed of
[23:57:07] <A-L-P-H-A> EDM = brass wire.
[23:57:08] <robin_sz> brass mainly
[23:57:09] <dmessier> hey you been hittin the books ALPHA
[23:57:12] <mdynac_> brass
[23:57:19] <SWPadnos> but I can think about the programming issues involved in controlling an EDM machine
[23:57:25] <SWPadnos> it's Brass, or so I hear ;)
[23:57:37] <robin_sz> edm is weird motion control
[23:57:48] <SWPadnos> from all corners of the world, I hear that EDM wire is brass, mostly
[23:57:52] <A-L-P-H-A> EDM could be graphic electrods as well. But that's not WIRE EDM.
[23:57:58] <SWPadnos> graphite?
[23:58:04] <dmessier> it needs a bump and wait.... bump and wait
[23:58:06] <A-L-P-H-A> yeah, graphite.
[23:58:13] <mdynac_> the dec pdp8\a that i am replacing with emc2 runs the edm just fine.....
[23:58:14] <SWPadnos> now that, I've heard of
[23:58:18] <robin_sz> the cut speed is govenrend by the arc voltage usually
[23:58:28] <SWPadnos> heh - that's not multitasking, is it?
[23:58:31] <dmessier> plunge is a round and round
[23:58:31] <mdynac_> graphite is for sinking
[23:58:33] <giacus> :P
[23:58:33] <A-L-P-H-A> dmessier, not if it's a bath of free flowing de-ionized water.
[23:59:04] <robin_sz> the hard bit of edm is backing off when needed
[23:59:11] <SWPadnos> sadly, a PDP/11 had faster I/O than a multi-GHz PC
[23:59:15] <A-L-P-H-A> I got EDM books, as PDF, if anyone wants them.
[23:59:25] <mdynac_> which ones?
[23:59:44] <robin_sz> edm for beginners?
[23:59:51] <dmessier> the resistance controle sys knows that
[23:59:57] <robin_sz> sally and bill do EDM?