#emc | Logs for 2006-03-26

[00:00:23] <robin_sz> we did the lower 1m in our shower in silicone .. worked well
[00:00:38] <robin_sz> 10 years on, no leaks
[00:01:51] <skunkworks> cool - everything is going to have 1/2 inch of cement board behind it
[00:02:10] <robin_sz> yeah we have that
[00:02:27] <robin_sz> "plaster board"
[00:02:40] <robin_sz> gypsum
[00:03:12] <skunkworks> nope differnt - it is actually concrete reinfoced with fiberglass.
[00:03:18] <skunkworks> board
[00:03:22] <robin_sz> right
[00:03:38] <robin_sz> lots of short random fibres in cement?
[00:03:50] <skunkworks> no - a grid
[00:03:53] <robin_sz> wow
[00:04:00] <robin_sz> we dont have that AFAIK
[00:04:15] <robin_sz> you are in USA I guess
[00:05:02] <skunkworks> http://www.sheetrock.com/navigate.do?resource=/USG_Marketing_Content/usg.com/web_files/products/prod_details/DUROCK_Brand_Cement_Board.htm
[00:05:09] <skunkworks> right
[00:05:10] <jmkasunich> ranking in terms of water resistance: bad: regular gypsum board (aka drywall), medium: water resistant gypsum board (aka green drywall), good: cement board
[00:05:20] <robin_sz> right
[00:05:33] <robin_sz> gypsum is almost universal here
[00:05:50] <jmkasunich> same here, the green is only used in bathrooms, etc
[00:05:56] <jmkasunich> cement board only for things like showers
[00:05:58] <robin_sz> by-product of powerstation de-sulpharisation of flue gases
[00:06:30] <robin_sz> people are always trying to find new uses for gypsum!
[00:06:37] <robin_sz> so much of it ..
[00:07:23] <skunkworks> jmk - where are you located?
[00:07:35] <robin_sz> in the basement!
[00:08:03] <robin_sz> * robin_sz isnt being very helpful
[00:08:13] <skunkworks> ;)
[00:08:18] <robin_sz> bet im right though ;)
[00:08:53] <jules_verne> yo, anyone alive?
[00:08:58] <alex_joni> hello, a few
[00:09:21] <jules_verne> i'm looking to maybe use EMC, but i've got a controller which I can't find as supported
[00:09:25] <Jymmm> cement board is also used for foundation to tile countertops
[00:09:36] <alex_joni> jules_verne: what kind?
[00:09:46] <jules_verne> I see you can write your own driver, so I might go that route if anyone can give me info
[00:10:00] <jules_verne> Stepperboard, BC4D15NC
[00:10:07] <alex_joni> any link for that?
[00:10:08] <jules_verne> stepperboard.com
[00:10:28] <jules_verne> i have the NC firmware, product is BC4D15 for clarification
[00:10:32] <jmkasunich> robin is right, I'm in the basement
[00:10:41] <jules_verne> it's hooked up to a small 3-axis milling unit
[00:10:43] <jmkasunich> the basement is just east of cleveland ohio usa
[00:10:52] <alex_joni> jules_verne: let me look
[00:10:55] <jules_verne> ok
[00:11:33] <skunkworks> thanks
[00:12:12] <alex_joni> jules_verne: how does it connect to the PC?
[00:12:18] <jules_verne> serial, not parallel
[00:12:22] <alex_joni> I see RS232toTTL adapters
[00:12:27] <alex_joni> and USBtoTTL too
[00:12:29] <jules_verne> yea, i've got the RS-232 one
[00:12:45] <alex_joni> ok, so basicly a serial interface, and you need to feed it some commands
[00:12:51] <alex_joni> any place where those are documented´┐Ż
[00:12:53] <alex_joni> ?
[00:12:54] <jules_verne> sure
[00:13:03] <jules_verne> let me find the manual for the firmware + unit
[00:13:37] <alex_joni> I have BC4D15NCRouterFastLoad.pdf opened
[00:13:50] <jules_verne> http://www.stepperboard.com/PDFDocs/BC4D15NCRouter.pdf
[00:13:53] <jules_verne> that's the complete one
[00:14:06] <jules_verne> i guess the fast load is just as good lol
[00:14:21] <alex_joni> jules_verne: so all the motion & co are done inside that board
[00:14:28] <jules_verne> yes
[00:14:32] <jules_verne> it keeps track of everything
[00:14:33] <alex_joni> that means from emc you only need the GUI & interpreter
[00:14:41] <jules_verne> alright
[00:14:49] <jules_verne> i need to interpret standard HPGL files
[00:15:07] <alex_joni> oh.. the current interpreter is rs274ngc
[00:15:14] <alex_joni> but HPGL can be converted to g-code
[00:15:19] <robin_sz> hmmm
[00:15:21] <alex_joni> rs274 = g-code
[00:15:31] <robin_sz> dont see how that is going to work for nc control
[00:15:35] <jules_verne> yea, i can get it into GCode
[00:15:45] <jules_verne> no problem
[00:15:51] <robin_sz> looks like its serial control of each axis ...
[00:16:02] <jules_verne> it's either independent or linked
[00:16:06] <robin_sz> ie send a serial command for axis 1 goto position X
[00:16:13] <alex_joni> jules_verne: I'm almost off to bed, but before I go some pointers:
[00:16:29] <alex_joni> inside emc there is something called canon (for canonical commands)
[00:16:32] <robin_sz> looks like 1 axis will be moving befor ethe other gets the serial command
[00:16:43] <robin_sz> oh, linked, hmmm
[00:17:05] <robin_sz> if it has some linked mode, then it should be easy I guess
[00:17:16] <alex_joni> jules_verne: these canon commands, are actually functions which get called by the interpreter
[00:17:23] <jules_verne> ok
[00:17:35] <alex_joni> they are defined in emccanon.cc (in emc2/src/emc/task/emccanon.cc)
[00:17:50] <alex_joni> if I remember it correctly (the path)
[00:17:50] <robin_sz> * robin_sz nods
[00:17:54] <jules_verne> ok
[00:18:28] <alex_joni> ok, now you need your own version of emccanon.cc which provides all the command calls to the interpreter, but instead of talking to normal emc guts
[00:18:50] <alex_joni> it generates the appropriate commands and sends them through serial to the b4d15..
[00:18:58] <robin_sz> * robin_sz nods
[00:19:01] <robin_sz> simple huh?
[00:19:08] <jules_verne> LOL
[00:19:09] <jules_verne> yea
[00:19:11] <jules_verne> right...
[00:19:13] <jules_verne> haha
[00:19:29] <alex_joni> sounds simple, but it should keep you busy a few full days (24h) :D
[00:19:38] <jules_verne> alright lol
[00:19:38] <robin_sz> well, it is a nicely defined API I guess
[00:19:46] <jules_verne> well, it's setup for parallel though
[00:19:52] <jules_verne> not sure what effect that's gonna have
[00:19:54] <alex_joni> from what I've seen yes
[00:20:02] <alex_joni> jules_verne: it shouldn't matter
[00:20:16] <robin_sz> you could look at the ppemc driver ...
[00:20:20] <jules_verne> ppemc?
[00:20:26] <robin_sz> or ehatevr its called
[00:20:34] <jules_verne> put it this way, i just want the easiest solution to get my machine up and running lol
[00:20:53] <jules_verne> i can code and such, but i'd like to avoid it if possible
[00:21:14] <alex_joni> robin_sz: this works a lot higher
[00:21:22] <robin_sz> ray at pico systems has some motion controller that is talked to dowwn the parlalle port, sned a command, watch it move etc
[00:21:28] <alex_joni> it does it's own path planning and all
[00:21:34] <robin_sz> true true
[00:21:44] <robin_sz> ignore me
[00:21:51] <robin_sz> alex is Da Man!
[00:22:01] <alex_joni> robin_sz: the ppmc does export similar stuff like any other driver (speed settings & such)
[00:22:16] <alex_joni> jules_verne: you can't make this work with emc without lots of coding
[00:22:23] <alex_joni> not complicated coding
[00:22:28] <alex_joni> but coding still
[00:22:35] <alex_joni> I'll give you one example:
[00:23:22] <Jymmm> When you're engraving aluminium to a max depth of .03" using an engraving bit; what does it mena when you're left with burrs on the engraving itself (besdies doing it wrong)?
[00:23:27] <alex_joni> page 29 of the above mentioned pdf says 'G' is for moving
[00:23:51] <alex_joni> example: you need to send *200W
[00:23:51] <alex_joni> *1000X
[00:23:51] <alex_joni> *-25687Y
[00:23:51] <alex_joni> *43156Z
[00:23:54] <alex_joni> *g
[00:24:05] <robin_sz> Jymmm: probably either blunt cutter or going too slow
[00:24:27] <Jymmm> robin_sz: BRAND new cutter; slow as in feed or speed?
[00:24:36] <alex_joni> ok, now for emc inside the canon interface there is a function called STRAIGHT_FEED() which gets called by the interpreter when a linear motion has been encountered in the g-code
[00:24:49] <alex_joni> no the STRAIGHT_FEED() will need to read something like:
[00:24:53] <alex_joni> send_x();
[00:24:58] <robin_sz> ok, so cutter is OK, probably too low feed
[00:24:58] <alex_joni> wait_*();
[00:25:06] <alex_joni> send_y(y_value);
[00:25:09] <alex_joni> wait_*();
[00:25:16] <alex_joni> send_z(z_value);
[00:25:19] <alex_joni> wait_*();
[00:25:23] <robin_sz> Jymmm: proper engraving head, or dremel?
[00:25:24] <alex_joni> send_w(w_value);
[00:25:27] <alex_joni> wait_*();
[00:25:35] <alex_joni> send('g');
[00:26:01] <Jymmm> robin_sz: 90IPM; 18000 RPM. 2.25 HP router
[00:26:13] <alex_joni> jules_verne: following me so far?
[00:26:52] <robin_sz> Jymmm: how many flutes on the cutter?
[00:27:22] <robin_sz> 1? like a real engraving cutter?
[00:27:28] <Jymmm> robin_sz: It's an actual engraving bit...
[00:27:34] <robin_sz> right .. so 1
[00:27:38] <Jymmm> half shank
[00:28:25] <Jymmm> search for 37-03 https://www.onsrud.com/xdoc/plastics
[00:28:41] <robin_sz> try around 36 ipm
[00:28:51] <Jymmm> Made for aluminum and brass as awell as plastics
[00:28:55] <robin_sz> thats .002" per chip
[00:29:13] <Jymmm> I can alter the rpm from 8K to 25K
[00:29:39] <Jymmm> so not high enough clipload?
[00:29:43] <Jymmm> chip
[00:29:50] <alex_joni> wth is a 'degroid' ?
[00:29:56] <robin_sz> .002" is a good starting point
[00:30:11] <alex_joni> 25B -- set "Begin arc angle" to 25 "degroids"
[00:30:17] <jules_verne> kinda
[00:30:23] <Jymmm> For some reason I thought engraving machines were 30K RPM.
[00:30:36] <robin_sz> noramlly, at least that
[00:30:38] <Jymmm> the manual kind.
[00:30:46] <alex_joni> jules_verne: ok, I need to catch some sleep now (unfortunately it's 2:30 am)
[00:30:52] <jules_verne> alrighty, we'll talk later
[00:30:54] <jules_verne> thanks :D
[00:30:56] <alex_joni> drop me an email if you still have questions
[00:31:01] <alex_joni> alex.joni AT robcon.ro
[00:32:19] <jules_verne> cool
[00:34:38] <robin_sz> Jym
[00:34:43] <robin_sz> Jymmm: https://www.onsrud.com/xdoc/FeedSpeeds
[00:34:57] <robin_sz> and when you have read that ....
[00:35:26] <robin_sz> https://www.onsrud.com/xdoc/faqs
[00:35:31] <robin_sz> and then ... ;)
[00:35:54] <robin_sz> https://www.onsrud.com/xdoc/MachFacts
[00:36:11] <robin_sz> quite a lot of manufacturers info on feeds and speeds there
[00:38:07] <robin_sz> .002 to .004 per tooth seems about right
[00:41:41] <robin_sz> ok bedtime
[00:41:44] <robin_sz> cya
[00:46:44] <alex_joni> night all
[00:48:34] <Jymmm> nite alex_joni
[06:07:10] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[06:11:24] <Jymmm> SWPadnos what happenin?
[06:11:34] <SWPadnos> losing my connection, it seems
[06:11:41] <A-L-P-H-A> <yawn>
[06:11:58] <Jymmm> bummer
[06:12:18] <Jymmm> * Jymmm tosses fish into A-L-P-H-A's yawning mouth... SCORE TWO POINTS!
[06:34:24] <A-L-P-H-A> <bored>
[06:34:42] <Jymmm> make something
[06:34:48] <A-L-P-H-A> what would it be?
[06:35:05] <Jymmm> what ya got for materials?
[06:35:39] <Jymmm> oh, i Know.... make a electronic tool setter.
[06:35:40] <A-L-P-H-A> various alu 6061, 20xx series. some titanium round rod 1", .75", 1.5" dias.
[06:36:10] <A-L-P-H-A> some mild steel.
[06:36:15] <A-L-P-H-A> cold/hot rolled stuff.
[06:36:21] <A-L-P-H-A> oh, some brass as well.'
[06:36:24] <A-L-P-H-A> nylon
[06:36:54] <Jymmm> Make it so that it'll fit on top of the workpiece and when the tool touches it, it'll be EXACTLY 1.000" off the material
[06:38:16] <A-L-P-H-A> Jymmm, easiest way... complete a circuit. Precision ground shims, that are 1".
[06:38:21] <Jymmm> also have leads come out of it so you can use it as a limitswitch ro zero out the tool automagically
[06:38:39] <Jymmm> s/ro/to/
[06:39:00] <A-L-P-H-A> you could protect it by using 1/2" precision ground plates, and adjusted springs underneith, so you won't kill your tool.
[06:40:58] <Jymmm> it's that cushion that makes it difficult
[06:41:17] <A-L-P-H-A> guide pins.
[06:41:25] <Jymmm> for?
[06:41:49] <A-L-P-H-A> |/\/\/\| ---- |nut| ---- |plate|
[06:41:56] <A-L-P-H-A> ---- = guide pin
[06:42:20] <A-L-P-H-A> why not just buy a light up LED touch sensor.
[06:42:29] <Jymmm> $400
[06:42:32] <A-L-P-H-A> No.
[06:42:36] <A-L-P-H-A> there are cheaper ones.
[06:42:37] <A-L-P-H-A> way cheaper
[06:42:44] <Jymmm> precise?
[06:44:21] <A-L-P-H-A> looking for it.
[06:45:04] <Jymmm> and do you really want to connect a 120VAC tool to a led/switch thats connect to a Paraport and forget to pull the wire after zeroing out?
[06:45:37] <A-L-P-H-A> http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=1666228&PMT4NO=6038087
[06:46:51] <A-L-P-H-A> mcmaster = 2039A12 $24.41USD
[06:50:26] <Jymmm> looking...
[06:50:48] <A-L-P-H-A> mcmaster page 2130
[06:52:20] <Jymmm> that's an edge finder; how would I use that to zero out?
[06:52:30] <A-L-P-H-A> from known points.
[06:53:00] <Jymmm> I'm only talking 1) tool setting and 2) set material top to 0.
[06:54:37] <A-L-P-H-A> oh
[06:54:45] <A-L-P-H-A> still could use an edge finder for those.
[06:54:51] <A-L-P-H-A> well, #1 you could.
[06:54:57] <A-L-P-H-A> not really #2.
[06:55:04] <Jymmm> I saw this http://flexbar.com/Merchant2/merchant.mvc?Screen=PROD&Product_Code=18507 but not exactly what I had in mond.
[06:55:08] <Jymmm> mind
[06:56:13] <Jymmm> http://www.flexbar.com/NewPDFs/174-%20electronic%20tool%20setter.pdf
[06:57:02] <A-L-P-H-A> looks interesting
[06:57:09] <Jymmm> and $400
[06:57:45] <A-L-P-H-A> oh! I like the flexbar, CNC tool setting gage.
[06:57:48] <A-L-P-H-A> the bottom
[06:58:33] <Jymmm> not sure how one would calibrate that though.
[06:59:11] <A-L-P-H-A> level machine. level ways. level gantry.
[06:59:56] <Jymmm> http://www.penntoolco.com/catalog/products/products.cfm?categoryID=4512
[07:00:13] <A-L-P-H-A> something like that is awesome.
[07:00:18] <A-L-P-H-A> that's cheap too!
[07:00:23] <A-L-P-H-A> $70!
[07:00:57] <A-L-P-H-A> wire the LED out. :D
[07:01:14] <A-L-P-H-A> or use an light detector to the LED output
[07:08:45] <Jymmm> http://www1.mscdirect.com/CGI/NNSRIT?PMAKA=74464041&PMCTLG=00
[07:09:50] <SWPadnos_> SWPadnos_ is now known as SWP_Away
[10:36:43] <CIA-10> 03alex_joni * 10emc2/configs/max/.cvsignore: ignore emc.nml
[10:47:43] <CIA-10> 03alex_joni * 10emc2/src/emc/usr_intf/halui.cc: changed pin naming & other formatting
[13:16:14] <lerman_> lerman_ is now known as lerman
[13:28:48] <giacus_> ohh
[13:28:59] <giacus_> angelo d'arrigo died :(
[13:29:07] <giacus_> http://www.repubblica.it/2006/c/sezioni/cronaca/darrrigo/darrrigo/darrrigo.html
[13:30:32] <ValarQ> ?
[14:00:46] <giacus_> http://www.angelodarrigo.com/El%20condor%20fra.htm
[14:01:03] <giacus_> http://www.angelodarrigo.com/home%20ing.htm
[14:01:49] <giacus_> the plane crashed in catania this morning
[14:08:02] <giacus_> he was flyng this morning around catania with his plane
[14:08:13] <giacus_> crashed without apparent reason ..
[14:08:15] <giacus_> :(
[14:09:10] <ValarQ> hmm, the language on those pages seem cryptic
[14:09:47] <giacus_> thre's the english page, sorry
[14:09:54] <giacus_> http://www.angelodarrigo.com/home%20ing.htm
[14:10:02] <giacus_> he was a 'condor' man
[14:10:23] <ValarQ> ok
[14:10:45] <giacus_> les_w: know ..
[14:12:30] <giacus_> he was the king of the sky
[14:14:01] <giacus_> not on english news yet ..
[15:41:28] <ottos> Hello Gents..
[15:58:01] <giacus_> * giacus_ woders if ottos become a spammer ..
[15:58:08] <giacus_> 17:41 < ottos> Hello Gents..
[15:58:13] <giacus_> 17:45 <*> Quits ->ottos
[15:58:15] <giacus_> hehe
[15:58:20] <giacus_> kidding ;P
[15:59:18] <les_w> hi giacus
[15:59:23] <giacus_> hi les_w !
[15:59:30] <giacus_> howdy ?
[15:59:37] <les_w> haha
[15:59:41] <giacus_> :)
[15:59:56] <les_w> was just out detailing cars more but it is cold out
[16:00:04] <les_w> now working on the bmw
[16:00:09] <giacus_> nice
[16:00:22] <giacus_> i'm transfering the website ..
[16:00:32] <giacus_> got a nice hosting
[16:00:33] <les_w> finished the lexus (just wax) yesterday
[16:00:37] <les_w> then the truck
[16:00:47] <giacus_> have you seen about angelo d'errigo ?
[16:00:55] <giacus_> the 'fly man'
[16:01:07] <giacus_> crashed this morning in catania :(
[16:01:29] <les_w> each year, I sand out the car wash scratches with 0.5 micron abrasive sheets and polish
[16:01:33] <les_w> yes I saw
[16:01:42] <SWP_Away> SWP_Away is now known as SWPadnos
[16:02:00] <les_w> I built and flew a hang glider long ago
[16:02:01] <giacus_> les_w: oh .. you're able to do that
[16:02:06] <les_w> very dangerous
[16:02:08] <giacus_> and its nice
[16:02:55] <giacus_> the car for me its just an optional ..
[16:03:18] <les_w> I flew some in chicago last weekend but had some engine roughness and turned around
[16:03:28] <giacus_> yeah, I eard
[16:03:32] <giacus_> nice
[16:03:35] <les_w> oh, I like to polish cars for some reason
[16:03:44] <les_w> most don't enjoy it but I do
[16:03:52] <les_w> I like the perfect shine
[16:03:53] <giacus_> I also know there is very cold in these days
[16:03:56] <giacus_> here not ..
[16:04:05] <giacus_> we have been in catania yesterday
[16:04:11] <giacus_> was a summer day !
[16:04:26] <les_w> It should be getting warm here but it is not
[16:04:33] <giacus_> I shot some photo in the stallman conference with anna
[16:04:49] <giacus_> http://giacus.altervista.org/photo/hobby/stallman/index.html
[16:04:49] <les_w> have a link?
[16:04:56] <les_w> looking
[16:04:59] <giacus_> look at taormina ..
[16:05:08] <giacus_> taormina is a paradise !
[16:05:13] <giacus_> a place for vip
[16:05:18] <giacus_> very very nice
[16:05:34] <giacus_> there is the place where d'arrigo died this morning :(
[16:06:02] <giacus_> I cant uderstand how that happened
[16:06:24] <les_w> ah I see you and anna with stallman
[16:06:28] <giacus_> I think it happen when peoples is too much sure abou equipments :(
[16:06:32] <giacus_> yeah ..
[16:06:39] <les_w> I spun out in a hang glider
[16:06:44] <les_w> over water
[16:06:57] <les_w> it hurt...a lot
[16:07:20] <giacus_> weather was beautiful ..
[16:07:32] <giacus_> it only was some mechanical issue ..
[16:07:36] <giacus_> on that plane
[16:08:23] <giacus_> there's no reason other ..
[16:08:42] <les_w> I think I had a bad intake valve on my flight last weekend
[16:08:55] <les_w> I messed a little with magnetos and mixture
[16:09:00] <les_w> did not help
[16:09:06] <giacus_> :(
[16:09:11] <les_w> so turned back to the airport
[16:09:36] <les_w> oh well the air was very bumpy anyway
[16:09:50] <les_w> windy and very cold
[16:11:08] <giacus_> bad day, Im afraid, we lost a champion :(
[16:11:51] <giacus_> thats is incredible to think
[16:12:00] <giacus_> Hy flyed on everest
[16:12:05] <giacus_> everywhere ..
[16:12:17] <giacus_> and was on the guiness
[16:12:21] <rayh> logger_aj, bookmark
[16:12:21] <rayh> See http://solaris.cs.utt.ro/irc/irc.freenode.net:6667/emc/2006-03-26#T16-12-21
[16:12:36] <giacus_> he*
[16:12:51] <giacus_> hello rayh
[16:14:08] <rayh> Hi giacus_
[16:14:16] <giacus_> oh .. _
[16:15:22] <les_w> morning ray
[16:16:09] <rayh> Hi Les.
[16:16:21] <rayh> Did you make a decision about trying emc2 yet?
[16:17:40] <les_w> I'll surely do it. I have to wait a bit though, because I booked myself up so much with work that I'm short of time.
[16:18:42] <giacus> rayh: how much time yet to get an official stable release of Emc2 ?
[16:18:42] <les_w> In three weeks, I have to rough out a gas range ignition system, write a white paper on UV sterilization, and work on the fingerburner.
[16:18:48] <les_w> All in 3 weeks.
[16:18:53] <rayh> I know that feeling. I'm anxious to hear your comments about tuning and running.
[16:19:00] <SWPadnos> les_w, maybe you can send someone your ini file, and they (we) can take a stab at making a configuration for you
[16:19:05] <giacus> I know, Emc2 is already stable for my needs
[16:19:19] <SWPadnos> but on that note, it's time for breakfast. bbiab ;)
[16:19:29] <giacus> but, there's an official stable version ?
[16:19:35] <rayh> giacus, Yes it seems to be working very well. Just a couple of little things yet.
[16:19:43] <giacus> nice
[16:20:07] <les_w> I have no problem configuring....It's just that I want to change out boxes and go motenc
[16:20:19] <rayh> Bridgeportio is the last thing that we really have not made a decision about.
[16:20:42] <les_w> even a better tp won't help me much if I can't get a lot more than 2k servo update
[16:20:44] <rayh> And, les_w I want to hear your reaction to only the software change.
[16:21:23] <les_w> I guess I can do that
[16:21:32] <les_w> just need to clear the backlog!
[16:22:13] <les_w> This weekend I decided to only rest. I really beat myself up with that business trip.
[16:22:14] <rayh> If we could find someone to do the config SWPadnos suggested it would speed the process for you.
[16:22:56] <les_w> another hard drive might be easiest
[16:23:08] <les_w> it's a bdi redhat install right now
[16:23:10] <rayh> I'd think so.
[16:23:28] <rayh> The new grub bootloader should handle that easily.
[16:23:53] <les_w> is a p6 200 fast enough?
[16:24:23] <cradek> how much ram in it?
[16:24:45] <les_w> not much...64meg
[16:24:59] <cradek> you won't be able to use the default ubuntu install with only 64 meg
[16:25:13] <cradek> it's a little tight in my 128 laptop
[16:25:42] <cradek> you might be able to use a custom install without gnome
[16:25:42] <les_w> I do have a spare identical box...I could install on that and switch hard drives....
[16:26:16] <les_w> but I need to get a new box and card anyway or I won't be able to fully use the new spindle
[16:26:34] <cradek> you'll sure be happier with a faster box with more ram
[16:27:06] <cradek> it seems lately you can get 400-500 mhz PII machines for around $0
[16:27:18] <jmkasunich> yep
[16:27:22] <les_w> well, I have to get the servo update up to the 10k range
[16:27:23] <cradek> when vista comes out, you'll be able to get 1000-1500 mhz PIII machines for around $0
[16:27:27] <jmkasunich> altho probably not in tiger
[16:27:29] <alex_joni> hi guys
[16:27:34] <alex_joni> * alex_joni just got home
[16:27:45] <jepler> "in tiger"?
[16:27:48] <les_w> money is not a problem....I just want the fastest box that will work
[16:28:04] <alex_joni> les_w: my athlon xp 1600+ works wonders
[16:28:14] <les_w> I suspect very modern boxes might be slower though
[16:28:22] <rayh> So we throw out my ideas of the experimental method. Change one variable at a time.
[16:28:44] <jmkasunich> tiger ga is where les lives
[16:28:51] <jmkasunich> its a bend in the road
[16:28:54] <les_w> it would be best to do that ray, but my box just is too slow to try it I guess
[16:29:08] <jepler> oh oh
[16:29:11] <jepler> you just have to mail order it
[16:29:16] <les_w> right
[16:29:21] <jepler> (well, you don't get $0 machines that way)
[16:29:24] <jmkasunich> that kinda eliminates the $0 thing
[16:29:36] <les_w> like everything else!
[16:30:30] <les_w> anyway I just got a trunk full of $0 test equipment from itw
[16:30:37] <les_w> so I don't mind
[16:31:19] <jepler> Is there anything that one of us with a "modern machine" can do to help you gauge the performance, without the particular PCI card you'll be using?
[16:31:39] <les_w> They let me just have my old b&k vibration analysis gear I used for car stuff up there
[16:31:51] <les_w> will be good for machine analysis
[16:32:38] <les_w> jepler I just need to know of a motherboard that is known to work with motenc and can do servo update in the 10k range
[16:33:30] <les_w> sounds like a 1G athlon might be the ticket
[16:33:53] <les_w> Whatever WB9 was using
[16:34:17] <les_w> I remember it was 1G and ran 8k no problem with motenc lite
[16:35:09] <les_w> alex are you running motenc?
[16:35:52] <alex_joni> les_w: no, I was running parport, and got to ~80kHz pulses, not sure how nice though (didn't scope it yet)
[16:36:02] <les_w> ok
[16:36:38] <les_w> Also, I am very interested in using HAL for factory automation
[16:36:52] <les_w> Since I may have to set one up
[16:37:52] <alex_joni> les_w: start reading HAL_Introduction.pdf to get the hang of it ;)
[16:38:27] <giacus> whats the target ?
[16:39:23] <les_w> If I have to do that I will no doubt be trying to hire you guys.
[16:39:40] <giacus> maybe someone could also build a custom iso for that
[16:39:54] <fenn> alex_joni: was that a PCI parallel port card?
[16:40:30] <giacus> puppy seems vaery nice for that
[16:40:39] <giacus> just adding some new config files
[16:41:04] <fenn> * fenn is cheap :)
[16:41:06] <giacus> coolcnc distro( if I remember wel..)
[16:41:48] <fenn> gimme room and board and i'll work for free :P
[16:42:02] <giacus> hehe :D
[16:43:25] <les_w> oh yeah fenn?
[16:43:32] <les_w> will ray do that too?
[16:43:38] <SWPadnos> as long as it's a nice room and good food ;)
[16:43:55] <SWPadnos> and no toothless women
[16:44:06] <les_w> ray would prob want to camp in the woods...
[16:44:19] <SWPadnos> anh - he does that at home ;)
[16:44:21] <SWPadnos> nah
[16:44:26] <les_w> I can find some women with teeth......
[16:44:33] <les_w> but they come out at night.
[16:44:37] <SWPadnos> hmmm
[16:44:47] <les_w> haha
[16:44:52] <giacus> lol
[16:45:00] <giacus> les_w: !! haha
[16:45:22] <alex_joni> fenn: yes it was a PCI parport card, but strangely it's not any faster than an ISA one
[16:45:27] <giacus> womans are not good around emc
[16:45:32] <giacus> and computers too ..
[16:45:36] <fenn> alex_joni: i thought you said 80khz pulses?
[16:45:49] <alex_joni> fenn: right
[16:45:56] <giacus> anna sayd she didint understand a word of what stallman sayd in the conference !
[16:45:59] <giacus> bauahaha
[16:46:05] <les_w> haha
[16:46:10] <fenn> well.. what language was he speaking?
[16:46:12] <giacus> :-)
[16:46:16] <alex_joni> fenn: 5.9 usec BASE_PERIOD
[16:46:16] <giacus> stay away ..
[16:46:24] <giacus> he speak english ..
[16:46:44] <giacus> with no translation
[16:46:47] <giacus> :D
[16:47:06] <giacus> anna was sleeping after a bit :(
[16:47:11] <les_w> Anna can sing in english well.
[16:47:14] <giacus> haha
[16:48:11] <les_w> Aw i need to be out "destressing" onthe tractor but it's just too windy and cold
[16:48:38] <SWPadnos> giacus: here's a woman who's great weith computers, and might be OK with emc as well:
[16:48:42] <SWPadnos> http://www.cs.uvm.edu/~eppstein/
[16:48:45] <fenn> i'd think with some dictation sotware and google translation you'd have a decent translation
[16:48:53] <SWPadnos> (one of my old CS professors)
[16:49:02] <giacus> SWPadnos: already know that, she's an exception :(
[16:49:06] <fenn> is she a midget?
[16:49:17] <SWPadnos> no, I think it's a big sword
[16:49:22] <les_w> gulp.
[16:49:45] <fenn> she looks like my fencing instructor actually
[16:50:05] <dmessier> allo all
[16:50:09] <SWPadnos> she was great when I was in college. (she used to wear spandex pants)
[16:50:55] <alex_joni> SWPadnos: how long ago?
[16:51:06] <SWPadnos> err - a while
[16:51:07] <SWPadnos> 1988 or so
[16:51:11] <alex_joni> ;-)
[16:51:36] <giacus> * giacus already know the end of stallman with his girlfriend ..
[16:51:49] <giacus> alice was her name ?
[16:51:55] <giacus> cant remember ..
[16:52:01] <fenn> heh the AI chatbot?
[16:52:08] <SWPadnos> Eliza
[16:52:12] <fenn> ALICE
[16:52:13] <giacus> oh, ok
[16:52:21] <dmessier> square headed girlfriend...
[16:52:31] <giacus> he was playng around Hurd If I remember wel ..
[16:52:32] <SWPadnos> no - the old "conversation" program was Eliza
[16:52:34] <alex_joni> alo dmessier
[16:52:37] <giacus> yeah
[16:52:40] <SWPadnos> written in LISP, I think
[16:53:15] <giacus> great
[16:53:31] <giacus> that why anna dont like it
[16:53:33] <giacus> lol
[16:54:08] <giacus> les_w: btw, to get Emc2 for factory use would be very nice
[16:54:12] <SWPadnos> interesting. I never knew that Maggie was into Kung Fu
[16:54:26] <SWPadnos> ah - she wasn't when I knew her - no wonder
[16:54:32] <giacus> :)
[16:55:30] <giacus> les_w: however, to play around it require lot of time I think ..
[16:55:40] <giacus> Introduction to HAL is a great doc
[16:55:51] <giacus> but require a bit of time :(
[16:55:59] <giacus> for waht I seen ..
[16:56:28] <EHJ-1> Hey, Ray, Steve, Alex, et al (Eric here) FWIW, I got a config working with both the VTI and M5I20 boards.
[16:56:28] <giacus> maybe we could help for that
[16:56:43] <giacus> depend on how much time you get on these days :P
[16:57:24] <SWPadnos> EHJ-1, cool.
[16:57:36] <SWPadnos> hopefully it wasn't much harder than loading both drivers ;)
[16:57:47] <alex_joni> EHJ-1: nice
[16:57:56] <EHJ-1> The only thing that sucks is I need an entire 50pin ribbon for one signal.:)
[16:58:00] <EHJ-1> Not hard at all.
[16:58:02] <SWPadnos> heh
[16:58:19] <SWPadnos> get a 200-pair telco wire ;)
[16:58:23] <EHJ-1> I only need one PWM off the P2 connector.
[16:58:37] <alex_joni> use a single wire?
[16:59:28] <SWPadnos> you could solder up a jumper cable, but getting a 50-pin ribbon and a breakout is wasier and more future-resistant
[16:59:34] <EHJ-1> Yea, not sure yet what to do about that, I will mook at it next week.
[16:59:35] <SWPadnos> easier
[17:00:35] <EHJ-1> It really isn't a problem, it is just going to take one more Phoenix terminal than I had anticipated.
[17:00:37] <EHJ-1> look
[17:00:48] <SWPadnos> yep
[17:01:07] <SWPadnos> I was happy with Peters response on the list.
[17:01:07] <alex_joni> EHJ-1: nice work, how do you find emc2?
[17:01:20] <EHJ-1> So far so good.
[17:01:21] <SWPadnos> I may see if I can meet up with him for a bit at ESC in april
[17:01:25] <alex_joni> I mean installing, configuring, etc
[17:01:39] <EHJ-1> I have to redo my I/O for the Mesa board this week.
[17:01:57] <SWPadnos> did you get appropriate isolation for it?
[17:02:11] <SWPadnos> I think the headers run straight to the 3.3v (5V tolerant) FPGA
[17:02:20] <EHJ-1> Tuning the smaller 60 V motors has been a real pain, but they were hard under the previous system too. The 90 V motors tune very nicely.
[17:03:16] <EHJ-1> Right now I only have it going to the terminal strips. I have not tried to wire anything up yet. Will do that this week.
[17:04:18] <alex_joni> EHJ-1: maybe one day you'll take the time and document your success ;)
[17:04:20] <ottos> Greets Gents..
[17:04:41] <EHJ-1> Yea, there are a bunch of things I would like to document.
[17:04:41] <alex_joni> greetings
[17:05:08] <ottos> Alex, cradek thank you for all the info , looks like all 5-axis all operational...
[17:05:18] <alex_joni> ottos: yay
[17:05:20] <EHJ-1> A Wiki entry for each driver type for a start (if it is already there I missed it).
[17:05:32] <fenn> anyone know of a place to get timing belt pulleys besides MSC mcmaster or sdp-si
[17:05:34] <alex_joni> EHJ-1: good thinking ;)
[17:06:07] <ottos> I have one detail ... PID tuning,, since DPI settings cannot be changed on fly how do I do?
[17:06:10] <EHJ-1> Especially the Mesa board which is just so different from everything else.
[17:06:46] <ottos> halscope? halmeter?
[17:06:50] <alex_joni> ottos: you can change on the fly
[17:07:00] <alex_joni> what config are you using?
[17:07:01] <tomp> fenn: gates has offices in a lot of cities, i used office in ElkGrove
[17:07:07] <alex_joni> * alex_joni is getting old and forgets fast
[17:07:26] <ottos> motenc..
[17:07:32] <alex_joni> ok..
[17:07:35] <EHJ-1> Alex: I can take a shot at one for the VTI board.
[17:07:41] <jepler> 6705 NO base-thread ( 5865, 24270 )
[17:07:49] <alex_joni> EHJ-1: great
[17:08:08] <jepler> are the numbers on the right cycle counts?
[17:08:10] <alex_joni> jepler: right, 5865 is 7 clocks of the timer ;) (jmk pointed that out)
[17:08:26] <alex_joni> I think it's min & max, or something like that
[17:08:36] <alex_joni> in clock counts
[17:08:51] <jepler> I think it's last and max
[17:09:14] <jepler> just trying to figure out if the system is actually running the base-thread quickly enough
[17:09:15] <alex_joni> ottos: can you look at a file ?
[17:09:22] <ottos> sure..
[17:09:27] <alex_joni> motenc_motion.hal
[17:09:31] <ottos> ok..
[17:09:41] <alex_joni> you'll see the pid's in there
[17:09:59] <alex_joni> they get set using the 'setp' (set parameter) command
[17:10:09] <jepler> on the scope, when I set a 50% duty cycle with freqgen, it looks very nice. When I set 49%, all the edges are aligned to the same spots as for 50%, but they don't line up from scan to scan.
[17:10:36] <jepler> so it looks more smear-like
[17:10:36] <alex_joni> so when emc2 is running, you can open another console and issue 'bin/halcmd setp pid.0.Pgain 100'
[17:11:04] <ottos> ok.. I'm starting to get the pic..
[17:11:07] <jmkasunich> jepler: you are doing step type 1 (pseudo PWM)?
[17:11:08] <jepler> OK -- this is weird. If I put my hands over the keyboard (not pressing any keys) the scope output goes to hell
[17:11:21] <jepler> jmkasunich: yes, and I'm not saying that's not the "expected result"
[17:11:37] <alex_joni> ottos: you can issue lots of setp to tune
[17:11:51] <alex_joni> but remember to write down the good values, you'll need to put them in the ini
[17:11:54] <alex_joni> to make them stick
[17:12:03] <jmkasunich> jepler: the code can only change the output state on an interrupt
[17:12:10] <jepler> ah -- I adjusted the trigger level just a bit, now the presenece of my hands near the laptop doesn't affect anything. It must have been right on the edge of triggering.
[17:12:13] <jmkasunich> when you ask for 50%, you get 1 on, 1 off, forever
[17:12:15] <ottos> ok that solved the prob.. just have to take the time...
[17:12:32] <alex_joni> ottos: 'bin/halcmd show param pid.'
[17:12:47] <jmkasunich> when you ask for 49%, you get 1 on, 1 off, most of the time, but sometimes 1 on 2 off 1 on, so the long term average is 49%
[17:12:48] <alex_joni> that will show you the current values of all parameters that match pid.*
[17:13:04] <ottos> I've notices one thing if I may, if setting b/c axis, (rotary ) you need to add a axis as dummy for he system to work proprely kinda ghost in the machine aproach..
[17:13:21] <alex_joni> ottos: please explain a bit more
[17:13:57] <alex_joni> ahh, yes that is true
[17:14:04] <alex_joni> it's because of the kinematics
[17:14:11] <ottos> ok when axis 0-2 are set as xyz , then if you want to add b,c , assuming that axis 3 will be b, and 4 c, when b/c do weird things...
[17:14:14] <fenn> it's because alex was lazy and didnt want to fix it :)
[17:14:16] <ottos> ok then that explains it..
[17:14:32] <alex_joni> the default kinematics assign joint0=x, j1=y, j2=z, j3=a, j4=b, j5=c
[17:14:56] <alex_joni> if a is missing, then it'll try to move j3 and j4
[17:15:15] <alex_joni> but j3 is a (which isn't there), and j4 is b (not c as it expects)
[17:15:20] <ottos> now that is fine for now.. I'm having just a 1 motor for x, when gantry will be impemented 2x will it still work?
[17:15:29] <alex_joni> so it'll go do weird things ;)
[17:15:52] <alex_joni> ottos: if you add the second X motor only in hal, then it's easy
[17:16:00] <alex_joni> so command output->x & x'
[17:16:27] <ottos> no I meak the kinematics will ok throw it off?
[17:16:40] <ottos> mean.. can't even type today...
[17:16:47] <alex_joni> if you want to add the second axis to emc, then you need another kinematics file which does: j0=x, J1=x, j2=y, j3=z, j4=a, j5=b, j6=c
[17:16:58] <alex_joni> ottos: I would do it like this:
[17:17:06] <alex_joni> have XYZABC in the ini
[17:17:13] <ottos> ok.
[17:17:24] <alex_joni> then in HAL have the X output connect to both X-axes
[17:17:44] <ottos> seems simple..
[17:17:52] <dmessier> but a dual axis gamtry should have a master and slave relationship assigned.. not anoter axis..
[17:18:03] <alex_joni> dmessier: the proper ones yes
[17:18:06] <jmkasunich> jepler: just read back, your question about the output of show thread...
[17:18:16] <alex_joni> you can also do it otherwise:
[17:18:23] <jmkasunich> 6705 NO base-thread ( 5865, 24270 )
[17:18:25] <jepler> jmkasunich: yes?
[17:18:34] <jmkasunich> 6705 is the thread period in nS
[17:18:36] <dmessier> but you can also rip you machine apart that way
[17:18:43] <jmkasunich> NO means it doesn't support floating point functions
[17:18:54] <alex_joni> x output (emc) -> axis X -> feedback -> axis X' ->feedback -> x feedback (emc)
[17:19:05] <jmkasunich> the numbers in parens are the most recent and maximum execution times in CLOCKS, not nS
[17:19:15] <alex_joni> that way the feedback to emc contains both axes ferror
[17:19:28] <jepler> but on a 1.5GHz system, 24000 clocks > 6705 nS isn't it?
[17:19:33] <ottos> ok..
[17:19:40] <jmkasunich> yes
[17:19:47] <jmkasunich> which means you are having some overruns
[17:19:58] <dmessier> i see a psudo axis .. with feedback nly .. the pulse comes from x on both??
[17:20:09] <jmkasunich> one of these days I want to do overrun detection and logging
[17:20:12] <alex_joni> ottos: but ideally (not currently done in emc2 now) you'll have XX in the ini, and both axes set up as master-slave
[17:20:30] <alex_joni> dmessier: sorry?
[17:21:03] <ottos> ok that will be something to investigate.. next toy I guess.
[17:21:10] <dmessier> so xx drives both drives.. and feedback is x and x'
[17:21:24] <alex_joni> dmessier: right, but that presumes different kins & all
[17:21:39] <dmessier> or kins set the same
[17:21:42] <alex_joni> also homing is a pita to setup
[17:22:02] <alex_joni> yeah, but still different from normal XYZABC
[17:22:17] <dmessier> need and AND OR setup
[17:22:32] <dmessier> sorry OR to AND
[17:22:32] <ottos> who uses emc2 as gantry at this time anyhow..?
[17:22:43] <alex_joni> ottos: no-one afaik
[17:22:52] <alex_joni> but a few people want to
[17:23:32] <ottos> well there is only that much you can pull out of a c frame design..
[17:24:03] <dmessier> question?? are there currently any rpm limitations to the rotary axes??
[17:24:22] <alex_joni> dmessier: you can setup max speed & accel
[17:24:31] <alex_joni> also travel distance
[17:24:40] <ottos> sure..
[17:24:56] <tomp> dmessier: the unit difference problems seen when doing helixes?
[17:25:06] <alex_joni> ottos: did you have any problems that the rotaries don't behave like you expect?
[17:25:09] <dmessier> id like to rig up a verticl turning center like the toshiba MFC
[17:26:03] <ottos> for now they are ok... I've locked the servopacks to max of 4500rpm so there is no probs ... but by default emc2 keeps it within 3k.. as spec in ini..
[17:26:08] <dmessier> were you using cartesian coords??
[17:26:46] <ottos> are you talking about the rotary..?
[17:27:07] <dmessier> and FYI a true helix isnt aesy to do om any machine
[17:27:47] <dmessier> yes..
[17:28:11] <ottos> yes.. deg for the rotary..
[17:28:27] <dmessier> deg/min & mm/pin...
[17:28:27] <tomp> dmessier: didnt know you meant me, i meant feedrates are spec'd in ipm or mpm and get weird in helixes with axis having apm
[17:29:11] <dmessier> you have to rectify ipm or mpm using inverse time to get it correct
[17:29:34] <tomp> dmesier: to get the helix's anglar velocity correct?
[17:29:46] <ottos> g93 is a real pain..
[17:29:51] <dmessier> yes..
[17:29:53] <fenn> i have a gantry.. i dont know if it counts though
[17:30:05] <dmessier> g93??
[17:30:15] <ottos> inverse feed rate..
[17:30:24] <dmessier> pain how??
[17:30:35] <dmessier> i havent tried it in emc
[17:30:52] <tomp> dmessier: how (without knowing radius or diameter) can you determine how fast the tool point should move angularly?
[17:31:02] <ottos> to recalculate the proper feed.. it's a pain.. not as easy as described..
[17:31:18] <dmessier> should be painless... define your kinematicks
[17:31:41] <tomp> dmessier: this is some of the inadequacy of gcode, it doesnt state all the needed info
[17:32:07] <dmessier> it NEEDS it ALL... that is a VERY complec G-code
[17:32:16] <tomp> yes
[17:32:35] <dmessier> it will need all tool info...
[17:33:15] <fenn> * fenn refrains from bitching pointlessly
[17:33:24] <cradek> the way emc does this is defined in the kramer doc
[17:33:36] <SWPadnos> it gets pretty complex to calculate, when you take into account angular + cartesian moves, where the tool speed changes as the radius changes
[17:33:38] <cradek> the F word specifies the velocity along the XYZ path
[17:33:50] <cradek> ABC move so as to coordinate with XYZ
[17:33:57] <dmessier> yep
[17:34:05] <cradek> figuring the right F word is up to the gcode writer who has knowledge of the part's radius etc.
[17:34:23] <dmessier> and it relies on setup info and usually absolute encoders
[17:35:03] <dmessier> neg. cradec .. you define it acording to chipload
[17:35:19] <dmessier> as if it was a flat part
[17:35:32] <cradek> I'm just saying how emc does it, as defined in the kramer doc
[17:35:44] <dmessier> oh isee
[17:35:45] <cradek> you have to know that in order to choose the right F word when programming emc.
[17:36:09] <tomp> cradek: is tkramer still around?
[17:36:12] <ottos> that where the prob comes in... since there you have to lim the rot speed. ( system limitations ...)
[17:36:18] <dmessier> so some of the kins are NOT implemented completely
[17:36:36] <cradek> ottos: yes, the rot vel/accel are limited
[17:37:03] <cradek> tomp: I have never met tom kramer, don't know what he's up to
[17:37:28] <jmkasunich> I _think_ he still works at NIST, but he has moved on from EMC
[17:37:35] <dmessier> and the whole cut should slow down to maintain a constant feed without violating the fastest moving axis
[17:37:45] <cradek> it does
[17:37:46] <alex_joni> I think they have a project involving some sort of EMC and STEP
[17:37:55] <alex_joni> he might be working on that
[17:37:59] <cradek> ack, I have to go
[17:38:06] <dmessier> be well
[17:38:10] <ottos> same in here... cheers
[17:38:18] <tomp> thanks cradek, bye
[17:38:30] <dmessier> muchos gracias
[17:40:24] <EHJ-1> I have a few questions on implementing a remote GUI on Windows.
[17:40:50] <jmkasunich> eww
[17:41:01] <EHJ-1> I posted a message to the developers mailing list with a link to that application.
[17:41:08] <dmessier> i think alex was involved in that a few months age...
[17:41:33] <EHJ-1> Yea, I know, but that side of the application is still a windows world.
[17:41:34] <alex_joni> EHJ-1: ask away
[17:41:48] <dmessier> ago... we did it while in channel... if i recall
[17:42:09] <alex_joni> dmessier: yeah, but I have no idea when that was
[17:42:16] <alex_joni> EHJ-1: want the big picture?
[17:42:29] <EHJ-1> I went through emc.h and rsc and and a bunch of other files, but I don't see what is actually being sent over the network.
[17:42:30] <alex_joni> you need rcslib (which you can compile on doze)
[17:42:41] <alex_joni> I can send you a binary
[17:42:55] <EHJ-1> k, as a separate application to call or built into mine?
[17:43:00] <EHJ-1> k
[17:43:07] <alex_joni> you need to link it to yours
[17:43:11] <dmessier> who was it you logged into... im having a brainfart
[17:43:25] <alex_joni> but the one I have is several years old ;)
[17:43:29] <alex_joni> well.. 2 years or so
[17:43:58] <EHJ-1> If it still works. I can recompile on doze.
[17:43:59] <alex_joni> but it shouldn't matter
[17:44:09] <alex_joni> recompiling is not trivial
[17:44:09] <dmessier> this wasnt so long ago
[17:44:15] <EHJ-1> k
[17:44:17] <alex_joni> dmessier: the talking was sooner
[17:44:38] <alex_joni> EHJ-1: you need to checkout rcslib from CVS 'co rcslib'
[17:44:43] <alex_joni> then you can try to compile
[17:44:59] <alex_joni> there is a VC++ project in there
[17:45:10] <EHJ-1> What about just generating the network packets directly. What is the format of the wrapper that goes around the nml messages?
[17:45:13] <alex_joni> but a bit borked, can't remember how I cured it though, just remember some issues
[17:45:22] <alex_joni> EHJ-1: that's hard, there are several
[17:45:28] <EHJ-1> k, yea I was looking at it.
[17:45:32] <alex_joni> NML means neutral message encoding
[17:45:38] <alex_joni> and it has 3 encodings afaik
[17:45:45] <alex_joni> which you can chose in the .nml file
[17:45:54] <alex_joni> XDR, ASCII and some other I forget
[17:46:23] <alex_joni> http://www.robcon.ro/emc/emc-server.zip
[17:46:28] <EHJ-1> ok, I saw the file that was generated by Java, but that should no longer be done that way.
[17:46:43] <EHJ-1> k, thanx
[17:46:43] <alex_joni> that's a full emc compiled for doze, along with an emcserver
[17:47:25] <alex_joni> darn, the lib is not in there
[17:48:43] <EHJ-1> Yea, that is what I thought you were going to send me. If I inderstand that correctly, I will need to run tcl on doze. I was trying to avoid that and run just a native windows exe.
[17:48:55] <alex_joni> EHJ-1: http://www.robcon.ro/emc/rcs/rcs.lib
[17:49:03] <EHJ-1> k, thanks
[17:49:20] <alex_joni> you can link that lib into your app, it should provide you all nml needed functions
[17:49:26] <dmessier> is that the 'doze end alex??
[17:49:45] <alex_joni> EHJ-1: take something like the "current" halui.cc (from emc2/src/emc/usr_intf/halui.cc)
[17:50:04] <alex_joni> and adapt that, it contains some parts of hal, which you can easily remove
[17:50:14] <alex_joni> and the rest are the hooks for talking to emc
[17:50:14] <EHJ-1> k, Yea I was looking at that too.
[17:51:02] <alex_joni> EHJ-1: ask if you get into specific problems
[17:51:34] <EHJ-1> Yea, I will. There are so many pieces, I was sure which ones I needed to look at.
[17:51:57] <alex_joni> hope I clarified a bit ;)
[17:52:03] <EHJ-1> wasn't
[17:52:20] <EHJ-1> Its a start, I will have to look at it in more detail.
[17:52:32] <alex_joni> ok, mail me if I'm not around
[17:52:41] <jepler> If you want to see a front-end implemented in "C", look at xemc from emc1
[17:52:44] <dmessier> later
[17:52:49] <jepler> it's mostly what I read when starting on axis
[17:53:38] <EHJ-1> yea, that is more to what I inderstand, because I have never used tcl or python.
[17:53:46] <rayh> rayh is now known as rayh-away
[17:54:36] <EHJ-1> BTW, for anyone that didn't seem my post to the developers mailing list, my GUI as it currently exists is available through my ftp site.
[17:54:56] <alex_joni> EHJ-1: already ran it, but it doesn't plot anything on the display ;)
[17:55:35] <EHJ-1> It should, unless you do not have OpenGL support on your computer.
[17:55:57] <alex_joni> I do have, but it stays black (the numbers show up though)
[17:56:34] <EHJ-1> You opened an nc file I presume?
[17:57:44] <EHJ-1> I would like to know if there are any problems running on other platforms. The intent was that it not have any dependancies other than OpenGL and the starndard win api.
[17:58:01] <alex_joni> I opened 3D-Chips.nc
[17:58:14] <alex_joni> had to rename it first, the open interface expected nc instead of ngc
[17:58:21] <EHJ-1> Yea, that should come right up, it is the file I used most for testing.
[17:58:51] <fenn> * fenn hates software that expects a certain extension
[17:58:57] <EHJ-1> You can select it from the pull down, nc is the default, but ngc is there too.
[17:59:17] <EHJ-1> Also *.* is supported.
[17:59:27] <alex_joni> display stays black
[17:59:55] <EHJ-1> k, I will check on a virgin machine. Sounds like something with OpenGL.
[18:00:06] <jepler> I don't have Windows to run this on, but I'm curious what it looks like. Is there a screenshot?
[18:00:21] <EHJ-1> I can make some.
[18:00:28] <alex_joni> jepler: hang on
[18:00:32] <EHJ-1> And post them to the FTP site.
[18:00:41] <alex_joni> EHJ-1: if you give me a passwd
[18:00:43] <EHJ-1> my
[18:00:59] <EHJ-1> No password. User name is linuxcnc
[18:01:07] <alex_joni> ok, thought that was read-only
[18:01:42] <EHJ-1> Yes it is. That directory is only set up for read only. Have to log in differently to write, and then do a change directory.
[18:02:01] <EHJ-1> Making screen shots now.
[18:04:49] <EHJ-1> There should be a file there now named axis1.jpg
[18:05:52] <giacus> giacus is now known as giacus_afk
[18:06:50] <alex_joni> EHJ-1: yes there is ;)
[18:07:05] <alex_joni> EHJ-1: take a snapshot of the code entry page
[18:07:36] <jepler> Looks familiar in some ways.
[18:08:00] <jepler> Please don't go on calling it AXIS; I think it would be confusing to have two completely different EMC front ends with the same name.
[18:08:42] <EHJ-1> It is there.
[18:09:05] <EHJ-1> No problem, I was calling Win-Axis to show the lineage.
[18:09:24] <alex_joni> waxis :)
[18:10:30] <jepler> In principle there's no reason axis couldn't be ported to windows
[18:10:44] <alex_joni> jepler: only one problem I encountered
[18:10:48] <jepler> Will your front-end be GPL?
[18:10:50] <alex_joni> and that's compiling python :)
[18:11:11] <alex_joni> but I didn't bother enough
[18:11:22] <EHJ-1> Yea. or maybe laxis, I was working on a mascot which I was thinking of calling Lux (for Tux and Lasers and of course the meaning of the word). Don't know if someone else has already thought of it.
[18:12:24] <EHJ-1> Not this version, but I was planning to port it to Java and make that version GPL.
[18:12:35] <alex_joni> ick, java
[18:12:45] <alex_joni> not really better than tcl
[18:12:57] <alex_joni> and there is a java gui for emc already
[18:13:12] <EHJ-1> Yea it is, because there are RAD development tools available.
[18:13:29] <jepler> If it's not GPL, then you may not use the toolbar images from AXIS.
[18:13:40] <anonimasu> heh
[18:13:56] <EHJ-1> k, np.
[18:15:06] <alex_joni> also emc.hh & emc.cc are GPL
[18:15:14] <alex_joni> and you need to link those into the gui
[18:15:42] <anonimasu> hm, how would it affect the license if you compiled all that kind of stuff into a GPL dll..
[18:15:51] <anonimasu> and call it from the gui?
[18:16:12] <alex_joni> I think you can do that !?
[18:16:23] <anonimasu> that wouldnt violate gpl would it?
[18:19:42] <jepler> anonimasu: If you ship closed.exe and gpl.dll together, and closed.exe only works when gpl.dll is present, I believe the distribution is a "work based on the Program".
[18:19:57] <jepler> I think the terms of the GPL would probably forbid that.
[18:20:59] <jepler> but IANAL
[18:24:19] <fenn> * fenn agrees with jepler.. that's why there is the lgpl
[18:27:53] <Imperator_> logger_aj:bookmark
[18:27:53] <Imperator_> See http://solaris.cs.utt.ro/irc/irc.freenode.net:6667/emc/2006-03-26#T18-27-53
[18:28:02] <alex_joni> hi martin
[18:28:25] <Imperator_> Hi Alex
[18:30:42] <Imperator_> out granite block is now in position :-)
[18:30:46] <alex_joni> great
[18:30:48] <Imperator_> our
[18:30:57] <alex_joni> so you can start coding on the gantry ;)
[18:31:08] <Imperator_> :-)
[18:31:44] <Imperator_> i wana do this when i have time, but i don't realy need it, because we have a 840D
[18:31:56] <Imperator_> ups maybe i don't have to say this here
[18:32:30] <alex_joni> 840D?
[18:32:43] <Imperator_> Siemens SINUMERIK 840D
[18:32:47] <alex_joni> ahh.. ok
[18:33:00] <Imperator_> digital controller from Siemens
[18:33:11] <alex_joni> well, if the pace keeps up, soon emc2 will be better than sinumerik ;)
[18:33:22] <Imperator_> :-)
[18:33:45] <Imperator_> maybe, EMC makes big steps
[18:34:03] <Imperator_> i have to test the new planner
[18:34:25] <Imperator_> thats a problem with the 840D
[18:35:13] <Imperator_> Siemens has a very complicated algorithem, which makes sometimes failtures, which you can see on complicated surfaces
[18:36:16] <fenn> i thought siemens used a simple algorithm
[18:36:27] <fenn> but who knows
[18:36:47] <Imperator_> yes, very hard to find out
[18:37:37] <Imperator_> Heidenhain uses a simpler algorithem, but at the end it is better than siemens
[18:37:59] <Imperator_> with Heidenhain you get better surfaces
[18:38:24] <anonimasu> hm heidenhain are pretty insame..
[18:38:28] <anonimasu> atleast the higher end controls
[18:39:05] <Imperator_> why ?
[18:39:31] <anonimasu> 13 axis:es + spindle
[18:39:40] <Imperator_> not much
[18:39:56] <Imperator_> Siemens 50 axes per channel
[18:40:41] <Imperator_> not only the real axis are counting, for a tool chancer you can easyly count two axis
[18:42:26] <anonimasu> hope emc will be better then thoose controllers..
[18:42:27] <anonimasu> :)
[18:42:36] <Imperator_> hope so
[18:46:01] <Imperator_> * Imperator_ has to do a bit more for EMC
[18:51:01] <EHJ-1> Alex, Jeff: Sorry, went off to more thoroughly read GPL. As my post to the developer's mailing list indicates, this version is intended as a prototype, mainly to figure out how to do everything I need to do as easily as possible. There is no intent on my part to release the prototype under GPL because it was developed with a RAD tool which would require anyone wishing to modify it to have the same RAD tool. I just don't want to incur the time pe
[18:51:01] <EHJ-1> to do it in a more GPL for just a prototype. However ISTM as long as I don't release it in this form, it is not violating GPL.
[18:51:42] <EHJ-1> conducive^
[18:51:59] <alex_joni> EHJ-1: probably, but I have nfc, and I don't care much :)
[18:52:37] <EHJ-1> :)
[18:52:44] <anonimasu> EHJ-1: the point with the RAD tool dosent really matter as long as the source is open..
[18:53:51] <alex_joni> * alex_joni goes away for a while
[18:54:33] <EHJ-1> Yes I realize that, but everything released in conjunction with EMC has been done with basically a freely available tool, and it is not my intent to change that.
[19:21:11] <rayh-away> rayh-away is now known as rayh
[19:30:30] <Jymmm> Gentlemen.... START YOUR ENGINES!
[19:52:17] <Jymmm> What be happen in BC?!
[19:54:21] <CIA-10> 03rayhenry * 10emc2/tcl/mini.tcl: removed the reference to emccalib.tcl
[19:54:39] <LawrenceG> hi Jymmm ... just logging in again
[19:54:46] <Jymmm> WB!
[20:09:46] <CIA-10> 03alex_joni * 10emc2/tcl/mini.tcl: temp fix to allow mini to work
[20:10:10] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[20:11:45] <giacus_afk> hahaha
[20:11:49] <giacus_afk> http://video.google.com/videoplay?docid=-5336561617389993359&q=linux
[20:12:06] <giacus_afk> lol
[20:12:18] <giacus_afk> giacus_afk is now known as giacus
[20:28:05] <CIA-10> 03alex_joni * 10emc2/tcl/mini.tcl: let it complain
[20:59:05] <CIA-10> 03rayhenry * 10emc2/tcl/mini.tcl: moved reference to activeAxis
[21:15:43] <NickServ> This nickname is owned by someone else
[21:15:43] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[21:17:20] <CIA-10> 03jmkasunich * 10emc2/src/emc/ (iotask/ioControl.cc task/emctaskmain.cc): changed how some debug info is printed - task and iocontrol both print the inifile 'machine' and 'version' strings, this change tags the messages with the name of the program that printed them
[22:03:04] <tomp> rayh: i'm looking up couplers to direct drive Yaskawa 400W motor to solid nut lead screw, is bellows good/smart ( HUCO ) ?
[22:05:38] <rayh> I've most often seen these things connected with bellows in case there is some misallignment.
[22:06:10] <rayh> What sized shafts are you dealing with?
[22:06:30] <tomp> 14mm motor, 10mm leadscrew (1mm pitch)
[22:06:48] <Jymmm> * Jymmm hears 'bellows' and thinks --> http://www.chipcarving.com/bellows.jpg
[22:07:15] <tomp> bellows, the corrugated tin can motor couplersd
[22:08:25] <tomp> well i put in a rfq, bet it'll be xpnsv. like 40$ or so. thanks
[22:10:09] <rayh> Helical beam is a bit less expensive than bellows.
[22:11:28] <tomp> yeh, the q really is, what kind is good if i make a pretty good mount & is short? ( looking to see if HUCO has helikal beem )
[22:12:15] <tomp> oh, the spiral spring like type? maybe not enuf stiffness for the edm's constant reversal maybe
[22:13:01] <tomp> hard to judge
[22:13:21] <rayh> Oh I think you'll find they are more than enough.
[22:13:53] <tomp> ok, will rfq that too and decide on my pocketbook :-) thanks
[22:14:01] <rayh> My McMaster cat only shows inch bores but about 20-25 for one that would accomodate those metric bores.
[22:14:19] <tomp> lookin him up now
[22:14:59] <fenn> i dont know what this is called or if it is widely used, but if you have a thin disc with an inner hub and an outer hub, the disc can flex axially but not radially
[22:15:56] <fenn> hm radial isnt the right word..
[22:16:14] <tomp> fenn: HUCO call's em membrane
[22:16:17] <fenn> but you get the idea
[22:17:35] <tomp> rayh: look like it'd be enuf meat to modify, but the modification for the bores will cost me shop time... best to buy it if i cant create the accuracy :-(
[22:19:10] <tomp> ooh mm has one just for 10 to 14 mm, only 374$ whattadeal
[22:19:30] <fenn> hey at least they give you a price up front
[22:19:31] <rayh> I don't see one in Mcmaster with those bores.
[22:19:59] <tomp> p784 'precision torque limited fleible couplings'
[22:20:56] <tomp> but, i'm gonna call them tomorow, if the inch near alike is 30$ then they can get me the euro-cousin
[22:20:58] <rayh> Is that the one with the slip clutch.
[22:21:18] <tomp> yeh, clutch must the the $
[22:21:22] <anonimasu> argh
[22:21:23] <CIA-10> 03jmkasunich * 10emc2/ (26 files in 16 dirs):
[22:21:23] <CIA-10> Changed all configs (ini files) to use nanoseconds instead of seconds for
[22:21:23] <CIA-10> BASE_PERIOD, SERVO_PERIOD, and TRAJ_PERIOD, changed the run script so it doesn't
[22:21:23] <CIA-10> load motmod, and change all the config's .hal files so they _do_ load motmod.
[22:21:23] <CIA-10> This enables halcmd save to generate a file that can recreate the complete
[22:21:23] <CIA-10> configuration.
[22:21:29] <rayh> Somebody ought to have em.
[22:21:29] <anonimasu> torque limit seems like a scary thing
[22:21:39] <anonimasu> :)
[22:22:06] <rayh> It would be unless the encoder was on the leadscrew side of the slip.
[22:22:15] <anonimasu> yeah
[22:22:19] <tomp> 'watch out for that tree'
[22:22:54] <Jymmm> George george george of the jungle...
[22:23:21] <tomp> ok ray, thanks, back to gcad3D and making the mount
[22:27:43] <dmessier> my 2 girl dogs wont listen this week.... whuts up with the pups..
[22:28:02] <Jymmm> dmessier sound like they've been talking to your Son.
[22:28:19] <dmessier> full moon a comin'
[22:28:54] <dmessier> i have 2 sons... different orbits
[22:29:03] <Jymmm> the 14yo
[22:29:20] <dmessier> and a 11--12
[22:29:36] <dmessier> package deal
[22:29:49] <Jymmm> military school
[22:30:13] <Jymmm> very nice hand carvings --> http://www.chipcarving.com/gallery.htm
[22:30:35] <dmessier> mom doesnt like that .... fatherin laws ALL for it
[22:30:54] <Jymmm> drug mom, ship kids.
[22:31:25] <Jymmm> oh, and provide cellphone so moms will stop worrying every 10 seconds.
[22:31:37] <anonimasu> :night guys
[22:31:45] <dmessier> no good ones in canada .. we breed peace keepers and the definitly AIT that
[22:32:01] <Jymmm> G'Night anonimasu
[22:32:27] <fenn> happy robot dreams
[22:33:53] <Jymmm> oh a canook
[22:34:04] <Jymmm> =)
[22:34:45] <Jymmm> dmessier: Not to worry, I won't hold the poutine againest you!
[22:37:38] <dmessier> mange de la merde
[22:38:24] <Jymmm> isn't that what poutine is?
[22:38:44] <Jymmm> =)
[22:38:50] <giacus> hehe
[22:38:56] <giacus> sound familiar
[22:39:29] <steves_logging> steves_logging is now known as steve_stallings
[22:41:10] <Jymmm> I guess the one time I had poutine it was shit... but I can't expect much from a drive thru window. But you would think that if you served so much of something it taste good.
[22:44:30] <Jymmm> dmessier: Is it always served with a brown gravy?
[23:06:59] <giacus> night
[23:24:32] <Jymmm> hey skunkworks
[23:25:03] <skunkworks> Hi
[23:25:42] <skunkworks> Think I am done for the day. (re-routing sewer pipe ;) )
[23:26:13] <skunkworks> just one more vent to go.
[23:27:05] <Jymmm> skunkworks sounds fun.
[23:27:37] <skunkworks> I have too many hobbies
[23:27:45] <skunkworks> is there such a thing?
[23:28:08] <Jymmm> hobbies / 100 = time for
[23:45:17] <dmessier> some use turkey gravy... lighter... but still brown
[23:46:42] <dmessier> giacus .. not poutane
[23:46:56] <Jymmm> ah. I'm only a fan of country sausge gravy... over buttermilk biscuits.
[23:47:17] <dmessier> hick...
[23:47:42] <Jymmm> dmessier Hey, you know how many places serve SUCK gravy...
[23:48:06] <dmessier> hmmm NONE
[23:48:08] <Jymmm> only teo places EVER I know that make a great country sausage gravy
[23:48:15] <Jymmm> two
[23:48:48] <dmessier> i can make gravy from any burnt meat
[23:49:10] <Jymmm> so can these places, but doesn't make it good.
[23:49:45] <dmessier> pur on the spices
[23:50:04] <Jymmm> black pepper should be enough.
[23:50:45] <dmessier> cuyanne
[23:50:59] <dmessier> garlick
[23:51:04] <Jymmm> I'm not into hot, as much as flavor.
[23:51:13] <dmessier> and brown sugar
[23:51:16] <Jymmm> garlic in suasge gravy?
[23:51:45] <Jymmm> what you saying sound more like the fixins for a bbq sauce
[23:51:48] <dmessier> trust me.... garlick in EVERYTHING
[23:51:52] <fenn> i can make gravy without any meat
[23:52:15] <dmessier> talk to them
[23:52:44] <fenn> gravy needs sage to be gravy
[23:55:39] <dmessier> SAGE???
[23:57:19] <fenn> try it you'll like it
[23:57:27] <fenn> "everyone's doin it"
[23:57:43] <fenn> it's just a little bit of sage, man
[23:57:52] <fenn> chill out
[23:58:13] <Jymmm> dmessier: At least he didn't say Rosemary
[23:58:44] <dmessier> i used thyme and blew it once
[23:59:10] <dmessier> still nervous
[23:59:20] <Jymmm> we use a lot of cummin and worchester sauce