#emc | Logs for 2005-05-18

Back
[00:06:38] <Jymmm> anonimasu carpet knife
[00:13:43] <Jymmm> .500KVA
[01:31:09] <Phydbleep> OK.. Pics of the testing setup for the lathe are up.. I still need to throw a box around the motor speed controller and decide what I want for a rpm readout..
[01:31:22] <Phydbleep> http://68.35.143.93/lathe
[02:20:56] <Phydbleep> fenn: http://68.35.143.93/lathe :)
[02:21:37] <asdfqwega> 'Darwin's Little Helper'?
[02:21:53] <Phydbleep> Hehehe.. I liked it. :)
[02:22:44] <asdfqwega> Darwin needs all the help he can get
[02:22:55] <asdfqwega> In fact, he needs more helpers
[02:23:35] <asdfqwega> DarwinCo(tm)
[02:23:52] <asdfqwega> 1,000,000 employees
[02:25:53] <Phydbleep> asdfqwega: What do you think of the lathe?
[02:26:06] <asdfqwega> Nice
[02:26:12] <asdfqwega> Mine's bigger ;)
[02:26:30] <asdfqwega> Made with EMC: http://www.bright.net/%7Ejanfritz/CNC/deerinwildrose.jpg
[02:26:46] <Phydbleep> Watch where you're waving that thing, You could put someone's eye out..
[02:27:15] <Phydbleep> asdfqwega: ice. :)
[02:27:19] <Phydbleep> Nice
[02:27:36] <asdfqwega> thanks
[02:28:17] <asdfqwega> I'm also doing freelance work, making programs for inlays in pool cues: http://www.bright.net/%7Ejanfritz/CNC/wrench_inlay.jpg
[02:32:20] <fenn> ooh ooh show and tell
[02:34:43] <fenn> http://fenn.freeshell.org/DCP_0327.JPG
[02:34:59] <fenn> my first piece with EMC
[02:35:30] <Phydbleep> Looks like you're just tuxing around. :)
[02:36:10] <fenn> it's pretty blasted huh
[02:36:44] <asdfqwega> You milled out Chip's kidney!
[02:37:22] <fenn> yes and i cut off his toes one by one
[02:37:41] <fenn> then *kablow* one through the top of the head!
[02:38:26] <asdfqwega> You did better than I did
[02:38:29] <asdfqwega> http://www.bright.net/%7Ejanfritz/CNC/3dchips_1st_try.jpg
[02:39:47] <fenn> it's that damn origin in the center of the block
[02:39:55] <fenn> how are you supposed to zero your work offsets?
[02:41:48] <Phydbleep> Procto-Numerology. :)
[02:59:29] <Jymmm> asdfqwega what did you make that with, a laser?
[03:01:32] <Jymmm> asdfqwega HEY ANSWER UP WHEN I'M BITCHING AT YOU!
[05:30:28] <fenn_> fenn_ is now known as fenn
[05:30:52] <fenn> is anyone here planning on going to the cnc workshop in june?
[05:33:25] <Jymmm> If it aint in Northern California, no
[05:36:29] <fenn> well i guess there's something good about living in indiana after all
[05:37:33] <Jymmm> Somehow I SERIOUSLY doubt it
[05:38:03] <fenn> heh
[05:39:16] <A-L-P-H-A> where's indiana?
[05:39:33] <Jymmm> A-L-P-H-A it's where turups come from
[05:39:36] <A-L-P-H-A> I just know it's in the middle of the usa somewhere.
[05:39:38] <Jymmm> turnups
[05:39:41] <Jymmm> turnips
[05:39:43] <A-L-P-H-A> and ohio is corn fields right?
[05:39:45] <Jymmm> there we go
[05:39:53] <Jymmm> and idaho is spuds
[05:40:02] <A-L-P-H-A> northern cali is just rocks, and sea ports right?
[05:40:13] <A-L-P-H-A> where's wheat?
[05:40:48] <Jymmm> N. Cal == Weed, Wine, Hippies, Silicon Valley, etc
[05:41:02] <A-L-P-H-A> oh, okay, you have silicon valley.
[05:41:09] <A-L-P-H-A> weed, that's EVERYWHERE.
[05:41:13] <A-L-P-H-A> hippies... you can keep'em
[05:50:22] <PhydZing> Woohoo! It's been moved to where it belongs. :)
[05:54:12] <Phydbleep> And it's laggin like mad..
[05:54:14] <PhydZing> Chatzilla seams really limited though. :\
[06:03:56] <fenn> chatzilla huh
[06:04:54] <Phydbleep> For the moment.. I need to string a hardline out there an dsee if that helps with the lag.
[06:05:01] <A-L-P-H-A> just use xchat.
[06:05:05] <A-L-P-H-A> or mirc if you're on win32
[06:05:26] <fenn> you have a radio link?
[06:05:37] <fenn> to the shop?
[06:05:42] <Phydbleep> 802.11b
[06:05:47] <fenn> neato
[06:06:33] <fenn> mozilla never really cut it for me, just kinda sucks in general
[06:06:40] <fenn> i like mosaic :)
[06:06:42] <Phydbleep> it would be if I could get > 50% signal quality at < 50 feet.
[06:07:25] <fenn> use two satellite tv dishes :P
[06:07:37] <Jymmm> prongles cans
[06:07:40] <Jymmm> pringles
[06:07:50] <Phydbleep> I'll just get a chunk of cat-5 and run a hardline.
[06:07:53] <fenn> turn off the edm machine
[06:08:12] <Phydbleep> fenn: Send me one to turn off and I will. :)
[06:08:30] <fenn> i'm gonna start the "hobby wire edm" craze next
[06:08:50] <fenn> yeah and a million other things
[06:09:05] <Phydbleep> fenn: I could use a tig, an air compressor and a plasma cutter too. :)
[06:09:16] <fenn> an air compressor!
[06:09:40] <fenn> visit scrapyard; assemble junk; repeat as necessary
[06:09:57] <fenn> hell a lawnmower engine makes a decent air compressor
[06:10:35] <Phydbleep> fenn: Been there, done that, watched the t-shirt burn down, fall over and sink into the swamp. :)
[06:11:38] <fenn> the t-shirt?
[06:12:43] <Phydbleep> Yeah, After the 50'th or so trip through the junkyard the t-shirt was ready to commit self-immolation.
[06:12:53] <fenn> heh
[06:46:27] <fenn> * fenn silently mourns the loss of his mill.
[06:46:52] <Jymmm> next time keep it on a leash
[06:47:17] <fenn> a 5000 lb chain
[06:52:39] <Jymmm> how about 500 lbs PER-LINK chain ----> http://www.colosseumbuilders.com/john/nj/images/nj-007.jpg
[06:53:15] <Jymmm> Jymmm is now known as Red70sShow
[06:53:15] <Red70sShow> Red70sShow is now known as Jymmm
[07:00:09] <fenn> what is that, a battleship?
[07:00:42] <Jymmm> no, it's a keychain.
[07:01:35] <fenn> "with the USS New Jersey, you always know where your keys are. just ask, hey, have you seen a five million ton grey behemoth floating around here anywhere? and someone's bound to know
[07:02:20] <Jymmm> and HOPEfully they wont hide themself in the couch either
[08:56:11] <Phydbleep> an0n. :)
[08:56:49] <anonimasu> good morning
[08:57:41] <Phydbleep> Almost, I need to string a hardline to the shop.. The 802.11b is wigging out at < 50 ft.
[08:59:07] <Phydbleep> Plus I can't get through the firewall from the house to the shop with a ssh. :\
[09:00:19] <anonimasu> :/
[09:01:45] <Phydbleep> And of course I can't get from the shop to the smb share with the mp3 collection either..
[09:04:20] <Phydbleep> A-L-P-H-A: You awake?
[09:05:42] <Phydbleep> Hmmm... I just need to break down and fab a pcboard mill.
[09:50:34] <anonimasu> hm, I've been compiling QT on my emc box for 12 hours now
[09:52:50] <anonimasu> * anonimasu sighs
[09:53:21] <anonimasu> it's a neverending compile
[10:56:28] <A-L-P-H-A> Phydbleep?
[10:57:16] <Phydbleep> A-L-P-H-A: What would you want to etch a couple of pcboards?
[10:58:45] <Phydbleep> 40 pin dip + support logic.
[11:00:05] <Phydbleep> * Phydbleep needs prottyping bits.. And sleep..
[11:00:16] <Phydbleep> prototyping
[11:01:49] <Phydbleep> A-L-P-H-A: 3"x5" single sided (jumpers are NOT evil), Think about it and let me know later.
[11:02:05] <Phydbleep> G'night all. :0
[11:02:08] <Phydbleep> :)
[11:19:53] <A-L-P-H-A> Phydbleep, got the drill code?
[11:20:08] <A-L-P-H-A> excellon code? artwork?
[11:44:17] <anonimas1> *yawns*
[11:46:05] <A-L-P-H-A> wuzzup?
[11:46:31] <anonimas1> working on some code
[11:46:44] <anonimas1> or well, testing some code on a real plc.. :)
[11:46:48] <anonimas1> with real sensors to see how it works
[11:47:28] <anonimas1> anonimas1 is now known as anonimasu
[11:48:15] <anonimasu> what's up with you?
[11:48:39] <A-L-P-H-A> working on a basic template for my designer to follow
[11:48:48] <A-L-P-H-A> but I'm probably gonna give up soon, as each design is unique.
[11:48:52] <A-L-P-H-A> yeah.
[11:48:54] <A-L-P-H-A> I give up.
[11:49:06] <anonimasu> your designer?
[11:49:10] <A-L-P-H-A> and now I'll probably work on pumping out some parts.
[11:49:17] <A-L-P-H-A> my graphic designer.
[11:49:24] <anonimasu> what parts are you making
[11:49:25] <A-L-P-H-A> I code, he designs.
[11:49:35] <A-L-P-H-A> that part I've been working on for the last like 2 weeks.
[11:54:33] <anonimasu> hm..
[11:54:34] <anonimasu> ok
[11:54:48] <anonimasu> want to see a image of what I do right now at work?
[11:55:31] <anonimasu> i had the cam nearby :)
[11:58:13] <A-L-P-H-A> :)
[11:58:14] <A-L-P-H-A> sure
[11:59:08] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6229.jpg
[11:59:11] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6230.jpg
[11:59:13] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6231.jpg
[11:59:17] <A-L-P-H-A> http://www.lloydleung.com/gallery/distributor.png
[11:59:38] <A-L-P-H-A> http://www.lloydleung.com/gallery/current/distributor.png
[12:00:07] <anonimasu> nice
[12:00:27] <A-L-P-H-A> I hate wiring.
[12:00:30] <A-L-P-H-A> it's so boring...
[12:00:35] <A-L-P-H-A> and I have to be awake to do it.
[12:00:38] <A-L-P-H-A> heh
[12:02:20] <anonimasu> electrician ell..
[12:02:27] <anonimasu> err hell, key stuck..
[12:02:40] <anonimasu> it's where all mean electricians go when they die..
[12:03:51] <anonimasu> that's the plc to control the machines we build
[12:03:55] <A-L-P-H-A> I'm thinking now, I should have just bought some 2.5"x1/2" flat stock, instead of the 4"x1/2" stock.
[12:04:12] <anonimasu> heh, you will have to machine it down loads
[12:04:17] <A-L-P-H-A> cool.
[12:04:43] <anonimasu> one box like that is ~3000$
[12:04:43] <anonimasu> ;)
[12:05:05] <anonimasu> the stuff on the side with few wires into them are prototype modules
[12:05:06] <anonimasu> :)
[12:05:39] <A-L-P-H-A> heh. nice.
[12:06:07] <anonimasu> there's 9 of them
[12:06:12] <anonimasu> I have them all on my desk :)
[12:07:38] <anonimasu> I am wirting code to determinate which way a shaft is spinning for some automatic disabling of some functions when going in reverse..
[12:08:11] <A-L-P-H-A> that doesn't seem too difficult.
[12:08:38] <A-L-P-H-A> bbiab... I'm going into the shop (next room). I'll load my laptop
[12:09:19] <A-L-P-H-A> nope.
[12:09:24] <A-L-P-H-A> no room... no clean area for the laptop
[12:10:12] <anonimasu> yep, it's not difficult, just requires a bit of thought :)
[12:18:53] <anonimasu> hm, works nicely :)
[12:20:12] <anonimasu> that distributor looks nice
[12:49:20] <anonimasu> http://www.emuge.com/news_events/featured/carbide_end_mills/cutting_action.html
[12:52:03] <anonimasu> horrid stuff.
[12:54:11] <anonimasu> machining stuff at rc50 ;)
[13:07:04] <anonimasu> hm is solidedge a good cam program?
[13:08:37] <anonimasu> err cad..
[13:10:58] <rayh> paul_c: Hi. How you doing today.
[13:11:13] <rayh> Saw a pile of commits.
[13:12:04] <anonimasu> rayh: hello
[13:12:09] <anonimasu> have you tried solidedge?
[13:12:14] <rayh> Hey. How you doing?
[13:12:34] <rayh> No. I'm not much of a cad guy.
[13:12:52] <anonimasu> hm ok
[13:13:10] <rayh> My favorite is fastcad but no cam for it.
[13:13:21] <rayh> And no solids at all.
[13:13:21] <anonimasu> hm, as long as it's parametric and good with solids
[13:13:49] <rayh> I've played with the solids on Synergy and like the concept a lot.
[13:13:51] <anonimasu> perferably solidworks..
[13:13:51] <anonimasu> but I got a good deal on solidedge..
[13:14:16] <anonimasu> I dont match well with synergy
[13:14:21] <anonimasu> I havent played lots with it though
[13:14:24] <rayh> You use it for cam and part preparation.
[13:14:24] <anonimasu> it's autocad-ish
[13:14:27] <anonimasu> kind of..
[13:14:31] <anonimasu> although much better
[13:14:38] <anonimasu> I love solidworks..
[13:14:42] <rayh> better?
[13:15:40] <anonimasu> solidworks?
[13:15:40] <anonimasu> yes
[13:16:31] <rayh> I've never gotten the autocad rythm so that I could produce drawings with any speed.
[13:16:39] <anonimasu> yep
[13:16:53] <anonimasu> that's why I dont match with it..
[13:18:50] <rayh> Hi alex.
[13:18:57] <alex_joni> hey rayh
[13:19:10] <anonimasu> hello alex
[13:19:28] <alex_joni> hey anders
[13:19:45] <anonimasu> have you tried solidedge?
[13:20:03] <alex_joni> nope
[13:20:06] <alex_joni> should I?
[13:20:06] <alex_joni> ;)
[13:20:11] <anonimasu> no idea..
[13:20:11] <alex_joni> rayh: what's new
[13:20:14] <anonimasu> I got a good deal on it..
[13:21:09] <rayh> Working up a new stepper box for a customer.
[13:21:48] <alex_joni> stepper driver?
[13:22:43] <anonimasu> so I am trying to decide..
[13:23:45] <rayh> Up to 4 axis with spindle fwd,rev, aux, and estop.
[13:23:54] <rayh> Limits, homes, probe.
[13:24:21] <alex_joni> nice
[13:24:31] <alex_joni> for use with emc?
[13:25:05] <rayh> Yes.
[13:25:15] <alex_joni> going with emc2?
[13:25:30] <rayh> Hey did you see that Sherline CNC won an show award in Belgrade.
[13:26:03] <alex_joni> really?
[13:26:04] <rayh> I will use emc2 if I can establish that it is ready for commercial
[13:26:30] <rayh> I put a few pics on www.linuxcnc.org/dropbox.
[13:26:43] <rayh> They all start with bel
[13:27:12] <alex_joni> can't load them
[13:27:23] <alex_joni> firefox complains that they contain errors :(
[13:27:44] <rayh> I didn't try to read them. I'll try.
[13:29:59] <rayh> doesn't work here either. musta goofed up the ftp text command.
[13:31:58] <alex_joni> * alex_joni goes home
[13:32:00] <alex_joni> later guys
[13:34:36] <les> ray still there?
[13:37:22] <rayh> Hi les
[13:41:49] <les> hi ray
[13:41:56] <les> have some questions
[13:42:27] <les> just a couple really
[13:42:41] <rayh> okay
[13:43:07] <les> I think with the test you did you made the spiral up to 180 ipm?
[13:43:48] <anonimasu> rayh: keep in mind that there's a mode change error ;)
[13:43:57] <anonimasu> bug report comming in on that tonight
[13:44:18] <rayh> There is a good bit of feedrate fall-off as the tip approaches the locus
[13:44:36] <les> well thayt is what I was getting at
[13:45:13] <les> 180 was the requested feed...but velocity adapting would drop the actual velocity yo a low value
[13:45:19] <les> to
[13:45:28] <les> I used actual velocity
[13:45:38] <les> a guess of it anyway
[13:45:39] <rayh> ??
[13:46:02] <les> the feedrate fall off
[13:46:33] <les> once you get to no cruising phase profiles emc should start lowering the actual feedrate
[13:46:42] <les> that is normal
[13:47:38] <les> I was just wondering what the actual feedrate was on your test at the point stutter started
[13:49:05] <rayh> And you find actual feedrate by ?
[13:49:48] <les> Well timing over a distance...you could log motions and get it from that but logging will perturb things
[13:50:06] <les> as will full debug
[13:50:52] <rayh> Oh. I'm running no debug.
[13:51:07] <les> so just a look at how long it takes to travel a couple inches would give an estimate
[13:51:15] <les> that's all I did
[13:52:03] <les> How much it adapts is dependent on the .ini accel value
[13:52:26] <les> lower accel, greater velocity slowdown
[13:52:40] <les> again this is normal...it should do that
[13:52:54] <rayh> Along a helix? I did modify a spiral.c to try and give a delta p for each block
[13:53:28] <les> When an unrealizable move is commanded emc slows to the point that the move is realizable
[13:54:12] <les> yeah having a delta p as a rem is handy
[13:54:14] <rayh> Yes. I've seen the accel feedrate override.
[13:54:43] <rayh> I remove any rems and the block numbers before I run.
[13:54:59] <les> I see
[13:55:29] <les> do you know your accel .ini values for the test you did?
[13:55:42] <rayh> Tom K indicatd that the parse time is lower.
[13:55:57] <rayh> Yes on the big iron it was 20
[13:56:03] <les> ok good
[13:56:07] <rayh> On my tests here I set 100.
[13:56:13] <les> different from mine a little
[13:56:19] <rayh> and 1000 but they did not make much difference.
[13:57:22] <rayh> What did you mean about "an unrealizable move"
[13:57:35] <les> I guess on freqmod it would not make much difference
[13:58:36] <les> unrealizable: Physically not possible to execute the motion with the velocity and acceleration assigned
[13:59:41] <les> there are two choices....refuse the motion or adapt and do as good as possible
[13:59:42] <rayh> What might such a pair of g-code blocks look like?
[13:59:49] <les> emc does the latter of course
[14:00:55] <les> hmmm well a G01 F300 when the max velocity in the ini is below that is an example
[14:02:03] <les> another is if the accel is set so low that there is not time to ramp up to the requested speed and back down
[14:02:29] <les> That is the triangular velocity profile mode
[14:03:21] <rayh> Okay I see what you are talking about.
[14:03:38] <rayh> Blending becomes a real issue with the triangular stuff.
[14:03:51] <anonimasu> hm, I was curious, what limits your accel besides hardware?
[14:04:00] <les> It may....
[14:04:02] <rayh> depending upon where in the traj cycle the start, middle, and end are.
[14:04:31] <les> anon... limits are in the .ini
[14:04:53] <les> MAX_ACCELERATION
[14:05:44] <les> there is a default value as well but that is not used
[14:05:45] <anonimasu> les: yeah, but can emc generate enough pulses to accelerate to the requested speed?
[14:05:46] <rayh> With steppers you want to be certain that you don't command a step or series of steps faster than the motor and drive can react.
[14:05:56] <anonimasu> yeah, but with step servos?
[14:06:00] <rayh> sounds like a saw
[14:06:07] <les> heh
[14:06:23] <rayh> With servo's you need to watch acceptable following error.
[14:07:10] <les> I have to run the .ini at about 15
[14:07:15] <rayh> That's where the ferror and min_ferror ramp comes in handy.
[14:07:30] <les> more than that and the stutter would quickly destroy the ballscrews
[14:07:35] <anonimasu> dosent matter with stepgen or does it?
[14:08:45] <rayh> Yes in emc(1) freqmod uses ferror, accel, and such.
[14:09:11] <anonimasu> well, I use emc2..
[14:09:40] <anonimasu> :)
[14:09:51] <anonimasu> does ferror have a function?
[14:10:17] <anonimasu> brb
[14:10:19] <rayh> Ferror with steppers?
[14:11:21] <les> It sure works great on srvo.
[14:11:41] <les> I have mine so tight that just a dull tool will ferror estop
[14:11:47] <rayh> les: Your issue then is not that feedrate slows down as the cutter spirals in.
[14:12:03] <les> no not at all...that is normal
[14:12:11] <rayh> It is that as it spirals something runs out of range and motion stops then restarts.
[14:12:28] <les> yes
[14:12:32] <les> the stutter
[14:13:01] <rayh> I retooled a water cooled browning as a kid and it really stuttered.
[14:13:34] <les> My machine will sound like a Browning MG if I turn up the accel!
[14:13:58] <rayh> Okay!
[14:17:58] <rayh> mornin Steve.
[14:18:12] <steve_stallings> good morning Ray
[14:20:49] <rayh> I put a bunch of bel(#).jpg images in the dropbox yesterday. I can't read 'em back.
[14:21:02] <rayh> I can read the originals here.
[14:21:39] <steve_stallings> that's not nice, let me check
[14:22:03] <rayh> No hurry. I can try to send them again.
[14:25:50] <paul_c> morning folks.
[14:26:36] <rayh> Hi Paul
[14:32:02] <paul_c> they've found a bug in rtai that has been causing circular dependencies.
[14:36:31] <rayh> Really.
[14:37:00] <rayh> Could you explain that just a bit for this greenie.
[14:37:50] <paul_c> EXPORT_SYMBOL(rt_printk) is declared in adeos.c, and again in hal.c
[14:38:28] <paul_c> depmod gets in to a paddy when it tries to resolve them.
[14:45:59] <rayh> I presume that this means we need to update the 4.20 kernel.
[14:50:12] <steve_stallings> Ray, not sure of the exact meaning, but your most recent JPGs in the dropbox used a different JPEG encoding format than you used for the earlier Hardinge pictures.
[14:51:19] <rayh> Okay. I can see them here with firefox but not from the site.
[14:51:59] <rayh> I used the ftp command to get them there. I've always used gftp before but don't have it handy.
[14:52:21] <rayh> I wonder if I messed up the files during transfer.
[14:52:36] <rayh> Hey Paul. can you help with the ftp command.
[14:54:12] <steve_stallings> The file does not look corrupt, but the JPEG format flags are something I have never seen before and none of the applications that I have can parse the JPEG header.
[14:54:50] <steve_stallings> I have moved the old files aside so I can compare after you try again.
[14:55:40] <paul_c> "put" should do it just fine.
[15:00:35] <rayh> put bel1.jpg in there
[15:02:29] <rayh> same thing here. Let me pass one of these files through gimp
[15:03:30] <paul_c> before transfering an image, type in
[15:03:38] <paul_c> binary <return>
[15:04:36] <rayh> ok will try that.
[15:06:24] <steve_stallings> That did it for bel1.jpg it now has a standard JPG header.
[15:06:55] <rayh> Okay. I'll send the others. Thanks guys.
[15:08:05] <A-L-P-H-A> what happens when you run a endmill slower than you should, SFM wise.
[15:09:12] <steve_stallings> Chip load per tooth goes up. Can chip a tooth, or overload the spindle torque.
[15:13:29] <anonimasu> yep
[15:16:28] <steve_stallings> Ray - Pictures all decode correctly now. Gee you really do have a small pipe. My traffic graph just barely shows a tiny baseline while you are uploading. 8-)
[15:17:03] <rayh> Hey quit making those kind of references to my parts<g>
[15:17:16] <paul_c> * paul_c is being inflicted with a transparent poxy server cache
[15:17:51] <steve_stallings> in English please (merican English if possible)
[15:18:30] <steve_stallings> Ray - Sherline went to a show... where?
[15:19:22] <paul_c> steve_stallings: http requests are cached by a server somewhere...
[15:19:55] <paul_c> further requests for the same url pulls data from the cache
[15:20:13] <paul_c> all done transparently to the user.
[15:20:37] <paul_c> hence "poxy transparent cache".
[15:20:51] <steve_stallings> and no way to force a refresh, that sucks
[15:21:17] <paul_c> there in is the "poxy" bit.
[15:21:55] <rayh> One of sherline's distributors in Europe set up a CNC with the sherline software and
[15:22:22] <rayh> put it in a booth in a Belgrade tech trade show.
[15:23:54] <steve_stallings> Interesting.... I got an order to PMDX from a guy in Croatia. The delivery instructions were basically put it on the mule named Sue, she knows the way.
[15:24:42] <steve_stallings> Even more remote than UP.
[15:25:04] <rayh> Okay. Got one once from Pakistan, "xx behind the water tower."
[15:25:30] <anonimasu> hm, I hate what I am watching some movies of somthing hogging titanium..
[15:25:43] <anonimasu> faster then I can mill aluminium.
[15:25:48] <anonimasu> way faster..
[15:26:30] <anonimasu> hate/love them..
[15:27:55] <rayh> I can put the text of the post to sherline here if interested.
[15:28:24] <steve_stallings> Something to explain the picture files would be nice.
[15:28:45] <steve_stallings> In the dropbox.
[15:29:11] <rayh> k
[15:29:22] <rayh> will try that.
[15:30:49] <anonimasu> hm I might have found myself a new lathe
[15:31:04] <anonimasu> candidate for emc..
[15:31:46] <anonimasu> http://www.blocket.se/view/4791789.htm?caller=nbl_s&l=0&c=1&city=0
[15:31:56] <anonimasu> there was a spare one for sale for much less..
[15:32:05] <anonimasu> 600~$ for the spartpart one..
[15:32:15] <anonimasu> and that one ~3000$
[16:22:06] <stevestallings2> stevestallings2 is now known as steve_stallings
[16:30:44] <anonimasu> ferror is only during rapids isnt it?
[16:30:51] <anonimasu> and min_ferror is during normal cuts
[16:33:40] <steve_stallings> ferror is a constant term and min_ferror is proportional to feed rate
[16:34:17] <steve_stallings> http://emc.sourceforge.net/Handbook/node24.html#997
[16:34:33] <anonimasu> ah ok
[16:34:37] <anonimasu> yeah found that at google..
[16:34:42] <anonimasu> I am trying to get my rapids up to 2m/min
[16:34:51] <anonimasu> err 2.5
[16:35:10] <anonimasu> :)
[16:36:05] <les> what is a .wps file?
[16:36:13] <les> something postscript?
[16:36:33] <anonimasu> I think it's a wordperfect file..
[16:36:46] <les> oh yeah
[16:37:05] <les> have not used wordperfect in 20 years
[16:37:08] <anonimasu> :)
[16:39:40] <steve_stallings> my searching says it is a Microsoft Works file format 8-(
[16:39:48] <anonimasu> ah was it..
[16:40:00] <anonimasu> you should be able to open it in office
[16:40:03] <anonimasu> I think
[16:40:36] <steve_stallings> http://filext.com/detaillist.php?extdetail=WPS
[16:41:29] <anonimasu> how much repetability do you get with just a home switch?
[16:41:44] <anonimasu> without using the index pulse
[16:42:34] <steve_stallings> a good switch, properly configured can manage 0.001" most of the time
[16:42:49] <anonimasu> ok sounds ok..
[16:42:51] <anonimasu> thanks
[16:43:28] <anonimasu> I need to machine a box for the optocouplers, if I want to use the index pulse..
[16:43:30] <anonimasu> :)
[16:43:38] <steve_stallings> to my knowledge, all commercial metalworking machines utilize both the switch and an index signal
[16:43:59] <anonimasu> yep
[16:44:12] <anonimasu> the index makes it very accurate
[16:45:07] <steve_stallings> sometimes on stepper motor machines, the "home" state of the driver phases is decoded instead of using a pickup for the index
[16:45:44] <anonimasu> hm, that sounds like a cool technology :)
[16:47:07] <anonimasu> I found a nice tool to design QT user interfaces..
[16:47:12] <anonimasu> QT Architect
[16:48:16] <anonimasu> so I am going to throw togther some way to generate simple operations..
[16:48:57] <anonimasu> with just a few keys of input + a jog wheel
[16:49:57] <steve_stallings> QT was mentioned by Fred Proctor at the programmers fest as a good choice for future user interface development.
[16:50:06] <anonimasu> yep
[16:50:10] <anonimasu> it seems nice..
[16:50:37] <anonimasu> I havent learned it yet, but since there's a nice gui editor, I'll manage without it a while
[16:51:10] <anonimasu> it's free as long as you can distribute the sourcecode for your stuff..
[16:51:17] <anonimasu> commercial use costs more per developer..
[16:52:38] <anonimasu> would be easy to bind it via nml to emc2 so I can grab position off it..
[16:53:46] <anonimasu> for recording position..
[16:53:59] <anonimasu> something like a learn mode
[16:54:30] <anonimasu> jog > +(add point) > set feed
[17:01:16] <paul_c> There are also some Qt<->NML status "hooks"
[17:01:36] <anonimasu> hm, very niec
[17:01:40] <paul_c> have a Qt DRO widget here somewhere
[17:01:46] <anonimasu> paul_c: what do you think of that idea?
[17:02:33] <paul_c> of a Qt GUI ?
[17:02:48] <anonimasu> yeah, perhaps not gui online programming tool..
[17:03:08] <anonimasu> I dont know how much stuff needed to make it a complete gui.. :)
[17:03:28] <paul_c> be back after tea - Ask me again then.
[17:03:32] <anonimasu> ok
[17:06:27] <anonimasu> Jymmm: evening
[17:06:33] <Jymmm> Mornin
[17:10:22] <anonimasu> did you know it takes 12 hours to compile QT
[17:10:24] <anonimasu> :)
[17:10:35] <Jymmm> $Q$T$
[17:10:59] <anonimasu> Jymmm: it's free as long as you write opensource software..
[17:11:11] <Jymmm> oh they changed huh?
[17:11:27] <anonimasu> wait a sec..
[17:12:29] <anonimasu> http://www.trolltech.com/download/opensource.html
[17:14:06] <Jymmm> Eh, I wouldn't use it just due to the licensing.
[17:14:43] <anonimasu> Jymmm: what's wrong with it?
[17:15:02] <Jymmm> I prefer BSD style licensing
[17:15:18] <anonimasu> well, I guess you can write your own alternative then :)
[17:15:37] <anonimasu> do you mean the commercial stuff or just that it's gpl?
[17:15:48] <Jymmm> To ***FORCE*** me to write GPL, I don't think so.
[17:16:07] <anonimasu> hah, if what you are writing is going to be gpl why does it matter?
[17:16:33] <Jymmm> again, I like BSD style, not gpl.
[17:16:44] <Jymmm> I would not release something GPL, ever.
[17:17:03] <anonimasu> there's always a commercial version you can use..
[17:17:38] <Jymmm> it's bait-n-switch, "Homey don't play that".
[17:20:43] <anonimasu> well, whatever that works the best..
[17:21:00] <anonimasu> it's not like licenses mean crap in the end anyway..
[17:21:17] <anonimasu> I've had competitors that came and videotaped when we ran prototypes of our product..
[17:22:12] <anonimasu> if they have enough money they can take it no matter what license you release it under..
[17:23:20] <Jymmm> anonimasu : Behold the power of cryptography =)
[17:24:19] <anonimasu> Jymmm: haha
[17:24:31] <Jymmm> anonimasu you think I'm joking, but I'm not.
[17:24:33] <anonimasu> it's not like they could pry anything out of the PLC even if they wanted to..
[17:24:47] <Jymmm> anonimasu : Whos says it would be in the PLC ?
[17:25:01] <anonimasu> the stuff that's important..
[17:25:16] <anonimasu> the iron is just scrap if you dont the algorithms for controlling it
[17:25:19] <anonimasu> :)
[17:25:58] <Jymmm> anonimasu See, your thinking the application is encrypted, when I said cryptography, not encryption.
[17:26:05] <anonimasu> nope...
[17:26:44] <anonimasu> I know what you mean, but if they copy your code straight off, if you relese it under a BSD style license..
[17:26:50] <anonimasu> if they have more money then you they will win..
[17:27:00] <Jymmm> anonimasu : Don't lie to me, I'll take a angle grinder to your ballscrews!
[17:27:09] <anonimasu> they can just fight until your funds go out..
[17:27:53] <anonimasu> :)
[17:32:27] <Jymmm> EH, that's why BSD
[17:33:00] <anonimasu> and well, if you develop somthing you want everyone to modify/steal/eat.. why care.
[17:36:03] <Jymmm> exactly
[17:36:29] <anonimasu> maybe I should order a book on developing with QT
[17:36:41] <Jymmm> books are for pussies!
[17:36:50] <Jymmm> lol
[17:37:06] <anonimasu> heh, you say so until you wonder that the 390th argument of function foo() does ;)
[17:37:54] <anonimasu> 90$
[17:38:20] <Jymmm> router bits are not cheap =(
[17:38:26] <anonimasu> that's for the book..
[17:38:52] <anonimasu> not that it matters.. I want it.
[17:39:30] <Jymmm> lol
[17:39:41] <Jymmm> you look at that thing yet?
[17:40:37] <anonimasu> at what
[17:40:39] <anonimasu> ?
[17:40:44] <Jymmm> the file
[17:40:54] <anonimasu> http://www.bokus.com/cgi-bin/more_book_info.cgi?ISBN=0596000642&pt=search_result
[17:40:59] <anonimasu> what file?
[17:41:17] <Jymmm> the one I emailed you a few days ago
[17:41:21] <anonimasu> no
[17:41:26] <anonimasu> I'll do it now before I forget it
[17:41:43] <Jymmm> lol
[17:42:06] <anonimasu> I dont think I can open it..
[17:42:09] <anonimasu> I need the other parts also
[17:42:24] <Jymmm> they weren't included?
[17:42:44] <anonimasu> no
[17:42:50] <anonimasu> just the assembly file..
[17:42:51] <Jymmm> * Jymmm sighs *
[17:43:01] <anonimasu> zip the whole folder where you keep it :)
[17:43:05] <Jymmm> ok, I'll zip em all up and email them to you
[17:43:18] <anonimasu> I have 3 lines in the centre of the screen..
[17:43:21] <anonimasu> nice drawing!
[17:43:27] <rayh> Just added an html link to the belgrade pictures...
[17:43:30] <Jymmm> are they 120 degrees apart?
[17:43:37] <anonimasu> yeah
[17:43:38] <anonimasu> I think so
[17:43:40] <rayh> www.linuxcnc.org/Dropbox/bel.html
[17:43:52] <Jymmm> well, can't you figure out the rest from that?! eeeesh
[17:44:09] <anonimasu> ^_^
[17:47:01] <anonimasu> btw, there's a deg tool
[17:47:21] <anonimasu> remembered where it was
[17:47:35] <anonimasu> you use the dimension tool..
[17:48:05] <Jymmm> I found an easier way... use SMartDimensions, and just slect two lines
[17:48:21] <anonimasu> yep
[17:48:39] <steve_stallings> Ray - much fancier than I had expected, you must have too much time on your hands
[17:48:39] <anonimasu> that's what I mean...
[17:48:47] <anonimasu> rayh: very very nice
[17:49:44] <anonimasu> Jymmm: is the mail comming?
[17:50:12] <Jymmm> working on it
[17:50:20] <rayh> Looks like I missed a pic.
[17:50:38] <Jymmm> rayh Yeah, the pic of yo promoting your 900 number =)
[17:51:00] <anonimasu> hm, girls in the booth :/
[17:51:18] <Jymmm> geriatric girls maybe =)
[17:51:23] <rayh> I figured that would attract some attention.
[17:51:30] <anonimasu> girls <3 cnc
[17:51:32] <anonimasu> ^_^
[17:51:42] <anonimasu> size isnt all, ray
[17:53:50] <anonimasu> * anonimasu wishes he could to go some show like that
[17:54:43] <anonimasu> one like IMTS or somthing like it..
[18:00:00] <Jymmm> anonimasu Did oyu get it?
[18:00:07] <anonimasu> Jymmm: not yet
[18:00:09] <Jymmm> anonimasu Did you get it?
[18:00:29] <anonimasu> I found a nice car when looking at the showsite
[18:01:13] <anonimasu> mercedes/mclaren SLR
[18:01:14] <anonimasu> ;)
[18:01:26] <Jymmm> I never thought that HF could actually fsck up something as simple as a router bit, but they did.
[18:01:43] <Jymmm> it actually looks like they GLUED it togehter
[18:01:47] <anonimasu> :/
[18:02:00] <Jymmm> Not he shank, but the actual cuttign bit itself
[18:02:04] <Jymmm> s/he/the/
[18:03:00] <anonimasu> http://www.omni4car.it/sfondi/mercedes_1024_004.jpg
[18:03:04] <Jymmm> I'd take pictures, but just not worth the time or effort.
[18:03:13] <anonimasu> that's strange
[18:03:48] <anonimasu> I dont fancy mercedes or bmw, but damn.. that car is violent :)
[18:04:16] <Jymmm> looks like a bmw roadster knock off
[18:04:34] <Jymmm> LOL restractable spolier
[18:04:42] <anonimasu> http://www.omni4car.it/sfondi/mercedes_1024_002.jpg
[18:04:45] <anonimasu> better pic ;)
[18:04:48] <Jymmm> retractable
[18:05:05] <Jymmm> front end looks like a Jaguar
[18:05:14] <anonimasu> yep, but I dont think it's in the same pricerange as a bmw roadster..
[18:05:21] <anonimasu> or as a Jaguar...
[18:05:36] <anonimasu> maybe top of the line ones..
[18:05:40] <Jymmm> and the vents on the side look like corvette
[18:07:13] <anonimasu> unexpected file format
[18:07:22] <Jymmm> what?!
[18:07:31] <Jymmm> the zip file or withing hte zip file?
[18:07:35] <Jymmm> within
[18:07:40] <anonimasu> the parts..
[18:07:43] <anonimasu> flat washer
[18:07:48] <anonimasu> radial ball bearing
[18:07:53] <anonimasu> flat head 100 screw
[18:08:41] <Jymmm> you have the toolbox installed?
[18:08:51] <anonimasu> nope..
[18:08:57] <Jymmm> thats probably why.
[18:09:09] <anonimasu> :/
[18:09:13] <robin_sz> meep?
[18:09:30] <anonimasu> aligning that whouldnt be too hard :)
[18:09:32] <Jymmm> later I'll see if I can convert them to parts and email it
[18:09:40] <anonimasu> hm ok
[18:09:45] <Jymmm> anonimasu it shouldn't be, but it's been a PITA.
[18:09:47] <anonimasu> err should..
[18:09:48] <Jymmm> howdy robin_sz
[18:09:53] <anonimasu> hey robin
[18:09:56] <robin_sz> gday
[18:09:59] <anonimasu> Jymmm: looks good
[18:10:06] <robin_sz> having SW fun are we?
[18:10:22] <Jymmm> robin_sz Oh yeah, not
[18:10:23] <anonimasu> robin_sz: Jymmm is ;)
[18:10:28] <robin_sz> heh :)
[18:10:38] <robin_sz> im pretty much up to speed on that now
[18:10:43] <Jymmm> robin_sz cant even email the freaking thing now.
[18:10:47] <robin_sz> only took ... 6 months :)
[18:11:10] <anonimasu> I think it took 4 weeks to learn.. but i've had experience with 3d modelling when I was in school
[18:11:20] <robin_sz> yeah
[18:11:28] <anonimasu> loads of playing with 3dsmax on school
[18:11:39] <anonimasu> so most was free..
[18:12:10] <robin_sz> figuring how desing in it is the fun part
[18:12:16] <Jymmm> ya know... I HATE line cards
[18:12:27] <anonimasu> cranking out a complete part feels so good :)
[18:12:51] <robin_sz> anonimasu: the key to it is realising that you only ever create any dimension once
[18:12:58] <anonimasu> yep
[18:13:21] <anonimasu> robin_sz: and make use of construction lines to drive stuff to where you want them to be..
[18:13:43] <anonimasu> :)
[18:13:47] <robin_sz> anonimasu: ie, if the holes in the block are 90 apart, dimension them on the block, the holes on the sheet that bolts to it are then just concentric with the other holes, no dimension on the sheet
[18:13:53] <robin_sz> yeah
[18:14:14] <anonimasu> yeah but that can be hell if you remove a part out of a assembly.. or modify it..
[18:14:24] <robin_sz> and make good use of library features
[18:14:26] <anonimasu> atleast until you've learned how to solve it..
[18:14:29] <robin_sz> yep
[18:14:38] <anonimasu> I've had my hours of cursing at it
[18:14:51] <robin_sz> the key to it is capturing design intent
[18:15:08] <anonimasu> have you used solidedge?
[18:15:28] <robin_sz> did you mean 10mm from the edge to the centro of the 10,, hole, or 5mm to the circumference ...
[18:15:46] <robin_sz> theres a very subtle difference
[18:16:00] <anonimasu> is it good?
[18:16:13] <robin_sz> solodedge? never tried it
[18:16:19] <anonimasu> hm ok
[18:16:20] <robin_sz> SW is all I use
[18:16:27] <robin_sz> that and acad LT
[18:16:33] <anonimasu> ew!
[18:16:35] <robin_sz> for very very basic stuff
[18:16:40] <anonimasu> * anonimasu pokes robin
[18:17:05] <robin_sz> LT is fine of you want to do "sqaure with two holes in it"
[18:17:09] <anonimasu> yeah
[18:17:20] <robin_sz> now ...
[18:17:29] <robin_sz> you should try the Trumpf drawing program
[18:17:46] <anonimasu> did you get the laser job?
[18:17:55] <anonimasu> err the production
[18:17:56] <robin_sz> that ships with their multi-thjousand pound "TOPS" program for lasers, punches, water jets etc
[18:18:03] <robin_sz> anonimasu: looks like it
[18:18:06] <anonimasu> nice!
[18:18:29] <robin_sz> we were 0.3mm of ovality on an 820mm circlular part
[18:18:39] <anonimasu> hm, bad..
[18:18:40] <anonimasu> *grins*
[18:18:51] <robin_sz> probably just heat build up in the metal and stresses
[18:18:54] <anonimasu> yep
[18:19:24] <anonimasu> robin_sz: got a name of that program?
[18:19:59] <anonimasu> jetcam perhaps?
[18:20:07] <anonimasu> hm no
[18:20:55] <anonimasu> robin_sz: how were thoose holes?
[18:26:52] <robin_sz> TOPS 100
[18:26:59] <robin_sz> by the way, it sucks :)
[18:27:47] <anonimasu> 4ah ok
[18:28:00] <robin_sz> BIG time!
[18:37:27] <robin_sz> I have absolutley NFC why quite intelligne t machine manufacturers seem to want to add their own half-baked drawing packages to their control software
[18:37:42] <robin_sz> they all seem to have on thing in common.
[18:37:49] <robin_sz> they are shite.
[18:39:08] <robin_sz> drawing and design is best left to the people like acad, weber, SW etc etc
[18:40:24] <rayh> rumor had it that weber (hint hint) was working on EMC interfaces
[18:40:41] <rayh> geometric as well as drawing.
[18:41:02] <dave-e> guess we'll find out in a month or so
[18:41:24] <rayh> be nice to add that to the mazak.
[18:41:31] <dave-e> indeed
[18:42:05] <robin_sz> shrug
[18:42:44] <dave-e> les ... you around?
[18:43:12] <robin_sz> I did try using emc in its native enviroment .. an old bridgeport with a BOSS control I wanted to rip out
[18:43:36] <robin_sz> I gave up in the end, configuration was still next to impossible
[18:44:09] <rayh> A stepper boss or servo
[18:44:13] <robin_sz> stepper
[18:44:47] <rayh> The HSM article that roland and I wrote had that machine converted with Gecko drives.
[18:45:08] <robin_sz> this was using the original drives
[18:45:09] <rayh> That was a couple years ago now.
[18:45:19] <robin_sz> so its quadrature
[18:45:30] <rayh> Oh. There is a quad option in freqmod.
[18:45:46] <robin_sz> i just got totally fed up with all that trying to guess PID values
[18:45:58] <Jymmm> PID?
[18:46:57] <robin_sz> after simple velocity/accel settign on other programs and the seemingly random guessing needed in EMC , I just lost the will to live and delted it im afraid.
[18:47:48] <robin_sz> not sure what we'll use now ... ordered a Baldor CNC control for it, so thats probably what it will end up as
[18:48:18] <rayh> P=250 is all I did.
[18:48:43] <robin_sz> and where did you discover that figure?
[18:49:41] <dave-e> ray scrounges thru his brain...
[18:49:54] <robin_sz> long long ago, we talked about a "library" of common setups
[18:51:26] <dave-e> ray...have you looked at the dev-list post?
[18:52:03] <robin_sz> and then I got confused with that input scale and output scale thing again ...
[18:52:07] <Jymmm> damn, router bits 'locally' are expensive
[18:52:37] <dave-e> yep...I find servos easier
[18:52:51] <anonimasu> input scale is the input to emc from encoders.. or servocards.. output is how many pulses you have for one step..
[18:52:55] <anonimasu> :)
[18:53:01] <robin_sz> and output scale?
[18:53:22] <robin_sz> right
[18:53:25] <anonimasu> s/output/output_scale
[18:53:43] <robin_sz> well, I have no encoders on the steppers
[18:53:57] <anonimasu> I guess you dont have to concern yourself about input_scale then
[18:54:08] <anonimasu> just keep the +/- the same as output scale..
[18:54:11] <cradek> with freqmod, the encoder input is emulated
[18:54:18] <robin_sz> well, it seems confusing .. I think you have to have it set the same in certain circumstances
[18:54:24] <cradek> input and output scale are to be set the same
[18:54:31] <anonimasu> hm, if that's the case then you have to keep them the same..
[18:54:33] <robin_sz> thats the madness I remember
[18:54:39] <robin_sz> so ..
[18:54:40] <anonimasu> let me check it how I have it
[18:54:46] <robin_sz> why not just read the same variable?
[18:54:59] <cradek> robin_sz: you have to make peace with it being a servo system hacked to pretend steppers are servos.
[18:55:22] <robin_sz> so ..
[18:55:23] <robin_sz> INPUT_SCALE = 1000 0
[18:55:23] <robin_sz> OUTPUT_SCALE = 1.000 0.000
[18:55:38] <anonimasu> INPUT_SCALE = 1000 0
[18:55:39] <anonimasu> OUTPUT_SCALE = 1.000 0.000
[18:55:43] <cradek> hmm
[18:55:47] <cradek> is that emc2?
[18:55:50] <anonimasu> yeah
[18:55:50] <robin_sz> thats as it ships in EMC2
[18:55:52] <cradek> oh
[18:55:57] <cradek> just ignore me then
[18:56:01] <cradek> obviously someone changed it
[18:56:08] <Jymmm> \ignore cradek =)
[18:56:16] <anonimasu> you set up your steps per unit in the .hal file..
[18:56:51] <robin_sz> I played with emc1 on the mill ...
[18:56:57] <cradek> well in emc2 there's no pid tuning necessary (or possible) with steppers
[18:57:01] <anonimasu> emc2 works better..
[18:57:18] <robin_sz> well, after 3 days, it just got tagged as "too hard"
[18:57:44] <robin_sz> which is a pity
[18:57:47] <cradek> I agree
[18:57:57] <robin_sz> as driving an old bridgeport should have been a piece of cake
[18:58:03] <anonimasu> robin_sz: well, it seems like you are quick to give up :)
[18:58:04] <cradek> sometimes free software is harder to use, but I use it anyway.
[18:58:16] <anonimasu> after the mess I had with turbocnc emc was a breeze
[18:58:21] <anonimasu> emc dosent die in the middle of parts
[18:58:25] <robin_sz> well, its running on Mach2 till the Baldor arrives
[18:58:25] <cradek> I don't mean that as a copout, I just mean that it's ok for me if I have to hack a little.
[18:58:50] <Jymmm> it's a give and take senario
[18:58:56] <Jymmm> ^c
[18:59:01] <robin_sz> yeah, I wouldnt actually want to make anything with Mach2
[18:59:09] <anonimasu> robin_sz: well, there are people that could help you set it up..
[18:59:10] <robin_sz> but setup is soooo simple with that
[18:59:22] <anonimasu> emc2 was very easy to set up..
[18:59:24] <anonimasu> :)
[18:59:38] <robin_sz> anonimasu: I did ask on here several times .. no one really seemed to know what worked with quad drives ..
[19:00:09] <robin_sz> and no one seemed to know if emc2 supported quadrature drive
[19:00:27] <robin_sz> so its now got geckos anyway :)
[19:00:37] <robin_sz> p[erhaps that will make it easier
[19:00:53] <anonimasu> robin_sz: very very easy..
[19:00:53] <anonimasu> :)
[19:01:06] <robin_sz> well, maybe
[19:01:08] <anonimasu> just calculate how many steps per unit you have
[19:01:13] <anonimasu> and put it in the hal file..
[19:01:22] <robin_sz> oh, it was emc1
[19:01:25] <anonimasu> and reverse the dir pins..
[19:01:34] <anonimasu> emc1 is pretty much the same..
[19:01:46] <anonimasu> just that you change output scale..
[19:01:53] <robin_sz> well, it seemed to just lock up all the time ..
[19:01:59] <anonimasu> if I remember it right..
[19:02:01] <robin_sz> soemting to do with cycle times I think
[19:02:06] <anonimasu> you had too high times if that was the case..
[19:02:31] <robin_sz> I randomly changed those to differetn values
[19:02:37] <robin_sz> unti it didn;t lock up
[19:02:49] <robin_sz> and then i just seemed to get following errors
[19:03:12] <anonimasu> then you were pushing it too fast..
[19:03:16] <anonimasu> and had a too tight ferror..
[19:03:42] <robin_sz> then I searched high and low for a sample "this works on an old bridgeport" .ini
[19:04:14] <robin_sz> failed to find one and canned it.
[19:04:21] <anonimasu> oh well if you've decided to give up I cant stop you really, but driving steppers works great..
[19:04:25] <anonimasu> stepper/step servos
[19:04:39] <robin_sz> configuration was not easy ..
[19:04:50] <robin_sz> it should have been, I drove my plasmas off this for ages
[19:05:07] <robin_sz> but I'd forgotten all the tricks :(
[19:05:08] <anonimasu> do as you wish..
[19:05:32] <robin_sz> sigh ...
[19:05:44] <anonimasu> it's tricky but so is configuring anything..
[19:05:54] <robin_sz> well, no.
[19:06:08] <robin_sz> some tings are easy to configure
[19:06:11] <robin_sz> like mach2
[19:06:16] <robin_sz> (but work badly)
[19:06:19] <anonimasu> * anonimasu sighs
[19:06:59] <robin_sz> I used to understand all this .. my brain must hav had a core dump :(
[19:07:39] <dave-e> speaking of performance...is les or paul paying attention
[19:08:11] <robin_sz> rayh: do you have a bridgeport-friendly .ini handy?
[19:08:23] <robin_sz> for an old intereact with geckos?
[19:08:34] <robin_sz> maybe I'll try one last time
[19:09:24] <rayh> * rayh phone brb
[19:10:26] <Jymmm> I know this may sound silly, but anyone know a good way to set a blade angle (table saw) to EXACTLY 30 degrees?
[19:11:03] <anonimasu> calculat it..
[19:11:18] <anonimasu> calculate..
[19:11:18] <anonimasu> :)
[19:11:34] <Jymmm> I typically use a sqaure as it has a 45 degree mark, but I have nothign for 30 degrees.
[19:12:02] <dave-e> sine bar?
[19:12:24] <robin_sz> or make a template
[19:12:33] <Jymmm> dave-e don't have one, but most of the ones I've seen don't have an "absolute" for 30 degrees
[19:12:48] <robin_sz> you have a set of compasses?
[19:12:58] <Jymmm> robin_sz a cheap one $0.99
[19:13:15] <dave-e> but the beauty is that you can make a block that is exactly right.
[19:13:41] <dave-e> I have a set of incremental angle blocks
[19:13:55] <rayh> robin_sz: with the gecko drives, I'd start with 4.20 and the sherline ini.
[19:13:56] <dave-e> just had to make a 11.5 degree angle
[19:13:59] <Jymmm> dave-e : I'm just not there yet on all the tooling.
[19:14:24] <dave-e> yes...it does take time....and then you have to keep it from getting stolen
[19:14:32] <rayh> I think it's named mill_inch_freq.ini but you could as easily use the mill_mm_freq.ini if you prefer a full metric definition of the hardware.
[19:14:45] <Jymmm> dave-e stolen isn't a problem... G17 helps with that =)
[19:14:49] <dave-e> I lost 7K$ worth of stuff one night
[19:14:52] <anonimasu> measure 4cm straight up
[19:15:00] <robin_sz> rayh: mill_mm_freq.ini from current cvs right?
[19:15:11] <Jymmm> anonimasu hmmmm?
[19:15:29] <anonimasu> then draw a line 2.22cm out from it..
[19:15:34] <rayh> Works for me.
[19:16:32] <anonimasu> then connect the bottom /
[19:16:32] <anonimasu> the start of the first line..
[19:16:32] <Jymmm> anonimasu SQRT 3
[19:16:33] <anonimasu> that's 30 degrees.
[19:16:33] <anonimasu> ;)
[19:16:33] <robin_sz> ach SF is fucked .. again.
[19:16:40] <rayh> na mill_inch or mill_mm aren't a part of sf. Except bdi-4 branch.
[19:17:17] <rayh> But you can update the emc from cvs and that will work with the bdi versions of the ini files.
[19:17:54] <rayh> Otherwise send me the number of pulses per unit and the unit type and I'll work up a custom for you.
[19:18:10] <rayh> How's that for the easiest ini configuration.
[19:18:15] <robin_sz> heh :)
[19:18:37] <robin_sz> well ...
[19:18:38] <anonimasu> hm, qt is compiled..
[19:19:15] <robin_sz> the old BP seems to be about 0.2" per motor turn
[19:19:44] <robin_sz> buand we have 10 ustep geckos ...
[19:19:45] <rayh> Yes. They were 5tpi screws.
[19:19:59] <rayh> These gecko 201
[19:20:04] <robin_sz> yeah
[19:20:27] <robin_sz> I dont expect them to spin those big old 42 frame motors that quick
[19:20:37] <rayh> So inch is 10k
[19:20:52] <robin_sz> gues so ...
[19:21:39] <rayh> Gecko makes a slight mod for the 42.
[19:21:44] <robin_sz> yeah?
[19:21:50] <robin_sz> just that jumper?
[19:21:54] <rayh> I didn't use the mod and got about 65 ipm
[19:22:03] <robin_sz> ok
[19:22:13] <robin_sz> 60ipm will be fine
[19:22:18] <anonimasu> 1600mm/min
[19:22:22] <anonimasu> hm, about my speed..
[19:22:26] <robin_sz> PC is a 450mhz athlon
[19:22:32] <rayh> Think the specific fix is different for different versions of the 201
[19:22:35] <robin_sz> erm K6 AMD
[19:22:47] <anonimasu> you should get a way faster box.
[19:22:51] <anonimasu> just a friendly advice :)
[19:22:58] <robin_sz> really?
[19:23:00] <anonimasu> yeah
[19:23:08] <anonimasu> I am running a 500 but it's a bit slow..
[19:23:31] <anonimasu> but I run a pretty high period
[19:23:48] <robin_sz> I have a dos thin running on an old 66 mhz and it does over 50K
[19:23:59] <rayh> Well a k6-2 450 or an athalon that same vintage should get you most of the way there.
[19:24:12] <robin_sz> thats it a .. k62-450
[19:25:38] <robin_sz> so 10K pulses should be OK on that?
[19:25:46] <rayh> with 10 k pulses per inch
[19:25:51] <rayh> and 60 ipm
[19:26:04] <rayh> that is 10k pulses per sec
[19:26:10] <robin_sz> yep
[19:26:30] <rayh> I can remember getting 6k easily on a k6-2 450 here.
[19:26:48] <rayh> That should get you to at least 40 ipm.
[19:26:59] <rayh> Perhaps a bit faster than that.
[19:27:05] <robin_sz> that will do
[19:27:22] <robin_sz> still below the knee point of the torque curve I guess
[19:27:29] <rayh> We can do an online tweek and try whenever you are ready.
[19:27:39] <robin_sz> ahh .. theres the rub
[19:27:54] <robin_sz> the bp is in my mates unit, no 'net
[19:28:07] <rayh> You running the gecko from the bpt supply -- 56 volts dc?
[19:28:30] <robin_sz> 60 volts .. 10A toroid
[19:28:56] <rayh> The harder you can hit the steppers the better.
[19:29:05] <rayh> I ran them at 7 amps per phase.
[19:29:06] <Jymmm> bbl
[19:29:14] <robin_sz> heyy are 8 wire devices
[19:29:31] <robin_sz> (1+1) and (1+1)
[19:29:49] <robin_sz> so right now we just wired 2 of the coils
[19:30:09] <robin_sz> series or parallel are options I guess
[19:30:20] <robin_sz> no clue which is the better option
[19:31:15] <robin_sz> 1 coil from each pair seems to be enough with 7A ...
[19:31:24] <rayh> I don't remember how I wired those steppers.
[19:31:36] <robin_sz> do you have the article still?
[19:31:44] <robin_sz> can we post it in the wiki?
[19:31:52] <rayh> The single coil each would be a good start.
[19:32:19] <rayh> Leave the current set resistor off the gecko.
[19:32:24] <robin_sz> thats what we done
[19:33:59] <rayh> When you ready to test run it?
[19:34:29] <robin_sz> well, it will be at least 10 days now .. I fly to Geneva on friday
[19:34:44] <robin_sz> I could mail my mate the .ini,
[19:34:58] <robin_sz> but installing it will be the limit of his knowledge
[19:35:12] <robin_sz> amd its a 100 mile drive ...
[19:35:27] <robin_sz> having no 'net in the unit really really sucks
[19:36:27] <robin_sz> if you can find an ini likley to work, sometime in the next ten days, I'll try again when I get back
[19:38:54] <robin_sz> so .. I should be able to find these .ini fiels only in the BDI-4.X branch on SF right?
[19:39:28] <rayh> Sounds good. I'll be away for a few days also but back well before you.
[19:39:39] <rayh> I'll get the ini worked up and shipped to you.
[19:39:58] <robin_sz> thnaks, that may save it :)
[19:40:07] <robin_sz> and ..
[19:40:18] <robin_sz> lets then post it somewhere . .onthe wiki even
[19:40:23] <robin_sz> so others can benefit
[19:40:58] <robin_sz> hmm, cant seem to see those /inis you mentioned rayh
[19:41:02] <robin_sz> on SF
[19:42:31] <rayh> They are in the bdi branch of emc2
[19:43:10] <robin_sz> nah
[19:43:25] <robin_sz> bdi-4
[19:43:27] <robin_sz> ??
[19:43:34] <robin_sz> http://cvs.sourceforge.net/viewcvs.py/emc/emc2/configs/?only_with_tag=bdi-4#
[19:43:50] <robin_sz> err without the # on the end
[19:51:56] <rayh> Well Darn. I don't see 'em either. Can I email.
[19:52:20] <rayh> I could put them on the wiki easy enough.
[19:55:42] <dave-e> ray...did you get a CD in the mail? I've not checked mine yet.
[19:57:42] <rayh> Not yet. Not quite mail time here.
[19:57:47] <dave-e> ok
[19:57:54] <dave-e> catch you later.
[20:03:34] <rayh> darn wiki doesn't honor end of lines
[20:07:50] <alex_joni> greetings everyone
[20:09:05] <anonimasu> hello alex
[20:09:15] <alex_joni> hey anders
[20:09:19] <alex_joni> what's up?
[20:13:01] <anonimasu> playing with QT designer
[20:13:24] <anonimasu> :)
[20:13:32] <Phydbleep> anonimasu: Did you design a cutie for yourself?
[20:13:40] <Phydbleep> :)
[20:14:34] <alex_joni> http://www.softlab.ece.ntua.gr/~sivann/pub/swf/may02-smilepop-soapbox4.swf
[20:16:47] <anonimasu> Phydbleep: you can track email..
[20:17:13] <anonimasu> with <img> tags ;)
[20:19:00] <alex_joni> an0n: modern email client usually block those
[20:21:18] <anonimasu> yep
[20:24:13] <anonimasu> alex_joni: starting on writing a online programming utility
[20:24:24] <alex_joni> rayh: still around?
[20:26:06] <rayh> Hi Alex
[20:26:29] <alex_joni> I wish I'd knew about the belgrade thing
[20:26:31] <alex_joni> ;)
[20:26:37] <alex_joni> it's only about 200 km from here
[20:27:51] <anonimasu> hm, qt seems lovely..
[20:28:09] <alex_joni> qt seems cute ;)
[20:28:23] <anonimasu> I liked the table widget
[20:29:12] <anonimasu> very nice for adding operations..
[20:29:27] <alex_joni> rayh: how much is a sherline mill?
[20:29:30] <rayh> Sherline sent me a copy of the post.
[20:30:15] <rayh> I've not kept up CNC ready is about $1000
[20:30:33] <alex_joni> not much..
[20:30:47] <rayh> No it really is not much.
[20:31:10] <rayh> You would need to check the site for real details.
[20:31:38] <rayh> You could also get a dealership and receive a discount from list.
[20:31:53] <alex_joni> well.. maybe some day ;)
[20:33:41] <CIA-4> 03rumley * 10emc/src/rs274ngc_new/ (canon.hh canon_pre.cc rs274ngc_pre.cc):
[20:33:41] <CIA-4> 1. Ken Lerman's changes to the subroutine code (April 15-, 2005), related to the parsing of code being skipped.
[20:33:41] <CIA-4> 2. A fix to a finer point of G92 with P<x> modifier.
[20:33:41] <CIA-4> This includes grabbing the amount of active axes from emcStatus, which GET_EXTERNAL_NUM_AXES() call involves additions to emccanon.cc in emctask, as well as canon.hh and canon_pre.cc.
[20:33:41] <CIA-4> 3. A fix/cleanup for M62-65 p-value handling in convert_m()
[20:37:00] <CIA-4> 03rumley * 10emc/src/emctask/emccanon.cc:
[20:37:00] <CIA-4> 2. A fix to a finer point of G92 with P<x> modifier.
[20:37:00] <CIA-4> This includes grabbing the amount of active axes from emcStatus, which GET_EXTERNAL_NUM_AXES() call involves additions to emccanon.cc in emctask, as well as canon.hh and canon_pre.cc.
[20:38:30] <alex_joni> somebody's been busy :D
[20:38:47] <paul_c> but no commits to the documentation.
[20:38:47] <anonimasu> alex_joni: have you seen the image's of the plc box I was building
[20:38:59] <alex_joni> an0n: no
[20:39:05] <alex_joni> paul_c: those might follow ;)
[20:39:25] <alex_joni> paul_c: but that's... wishfull thinking
[20:39:51] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6229.jpg
[20:39:54] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6230.jpg
[20:39:55] <anonimasu> http://www.bojn.net/~an0n/plc/AUT_6231.jpg
[20:42:55] <anonimasu> ^_^
[20:43:13] <alex_joni> looking now..
[20:44:28] <alex_joni> an0n: looks great
[20:44:51] <anonimasu> thanks :)
[20:46:18] <anonimasu> electrician-hell it's where all mean electricicans go when they die..
[20:47:06] <anonimasu> and some while they are alive.
[20:47:11] <anonimasu> bbl going to rest a bit
[20:47:17] <alex_joni> whoa.. some heavy storm outside ;)
[20:47:22] <alex_joni> night an0n
[20:53:59] <anonimasu> iab
[20:54:09] <anonimasu> didnt end up sleeping after all
[20:54:56] <alex_joni> :/
[20:56:28] <anonimasu> going to start coding some instead..
[21:02:01] <anonimasu> *yawn*
[21:03:50] <anonimasu> alex_joni: I love QT designer.
[21:04:09] <alex_joni> an0n: got some links?
[21:04:38] <anonimasu> www.trolltech.com
[21:04:45] <anonimasu> I can post you a sample of what I am doing
[21:05:21] <alex_joni> pls do
[21:05:27] <anonimasu> I am just playing around yet..
[21:05:34] <anonimasu> looking at how you do stuff :)
[21:09:25] <anonimasu> hm
[21:09:26] <anonimasu> I cant take one..
[21:09:31] <anonimasu> seems like I am low on memory
[21:09:51] <anonimasu> http://www.trolltech.com/products/qt/designer.html
[21:11:30] <anonimasu> alex_joni: look at it and tell me what you think
[21:11:41] <alex_joni> looking now
[21:12:06] <anonimasu> ok
[21:12:19] <alex_joni> a bit strange ;)
[21:12:21] <alex_joni> but ok
[21:12:30] <anonimasu> it's similiar to msvc++
[21:12:40] <alex_joni> yeah.. kinda
[21:12:49] <anonimasu> if you've coded ui's in it this is much better..
[21:13:13] <alex_joni> only click & drool on msvc++
[21:13:14] <anonimasu> so I guess I'll have the stuff ready as soon as I decide on how it will look and throw in some NML to grab the position
[21:13:33] <anonimasu> I hate handcoding interfaces since they seldom get as you want them
[21:13:38] <alex_joni> that should be fairly easy ;)
[21:13:43] <anonimasu> it's so much work for little payoff..
[21:13:55] <anonimasu> the fun part will be plotting toolpaths..
[21:13:59] <anonimasu> ;)
[21:13:59] <alex_joni> right
[21:14:10] <anonimasu> but well, I just care for the jog and place points for now
[21:14:58] <anonimasu> since that's what I need myself now :9
[21:16:06] <anonimasu> alex_joni: you know what I mean kind of like teach in on robots..
[21:16:15] <alex_joni> yeah
[21:16:30] <alex_joni> teach in is great
[21:16:42] <alex_joni> but in apps you don't need precision ;)
[21:17:00] <anonimasu> you can still go by the numeric output of emc.. and add your points..
[21:17:08] <alex_joni> right
[21:17:14] <anonimasu> ;)
[21:17:32] <anonimasu> I hope that this will get the mill useful for my father too..
[21:17:33] <alex_joni> hmm.. emc might be great to move a robot ;)
[21:17:36] <alex_joni> 6 axes
[21:17:38] <anonimasu> LOL
[21:17:41] <alex_joni> kinematics ;)
[21:17:43] <anonimasu> dont kill it
[21:17:49] <anonimasu> I'd use a PLC..
[21:17:53] <alex_joni> now if we add some teach in ;)
[21:17:59] <alex_joni> PLC is too dumb for that
[21:18:01] <anonimasu> 6 of thoose CIO011 modules you see first..
[21:18:11] <anonimasu> one for each axis..
[21:18:25] <alex_joni> yeah .. but can you do synchronous movement on all 6 axes?
[21:18:29] <anonimasu> yes
[21:18:30] <alex_joni> like go X+ ?
[21:19:03] <anonimasu> but you would have to code the algorithms yourself..
[21:19:10] <anonimasu> they use theese plc's lots for injection molding robots..
[21:19:34] <anonimasu> encoder inputs and stuff..
[21:19:49] <anonimasu> but the coding wouldnt be too easy..
[21:20:45] <anonimasu> or you could just hook them up to a servo module..
[21:21:03] <anonimasu> that handles the interpolation without putting load on the rest of the system
[21:21:03] <anonimasu> ;)
[21:23:23] <CIA-4> 03rayhenry * 10emc/src/emctask/tkbackplot.tcl: fix to mm plotting
[21:24:23] <alex_joni> a lot of busy people today
[21:24:28] <anonimasu> yep
[21:24:34] <alex_joni> rayh: don't forget to commit to emc2 too ;)
[21:24:59] <rayh> talking with paul about this now.'
[21:25:09] <rayh> thanks for the reminder.
[21:25:31] <anonimasu> alex_joni: do you own any robots or do you just service/program them
[21:26:54] <alex_joni> personally .. no ;)
[21:27:00] <anonimasu> ok :)
[21:27:29] <alex_joni> but we got 2 at work
[21:28:12] <anonimasu> neat
[21:28:16] <anonimasu> * anonimasu loves robots
[21:28:39] <alex_joni> yeah.. I love the new types too ;)
[21:28:43] <alex_joni> great dynamics
[21:29:30] <CIA-4> 03rayhenry 07bdi-4 * 10emc2/src/emc/task/tkbackplot.tcl: fix to mm plotting
[21:29:48] <anonimasu> yay!
[21:30:42] <alex_joni> heh
[21:30:53] <alex_joni> rayh: short question
[21:30:59] <alex_joni> what backplot is used in mini?
[21:34:06] <anonimasu> Hm, now to the tricky part of grabbing stuff from nml..
[21:35:32] <rayh> Uh. It's different.
[21:35:37] <alex_joni> * alex_joni passes anonimasu a forceps
[21:35:55] <alex_joni> rayh: ok, was wondering if it's not the same ;)
[21:36:02] <paul_c> If you just want status signals
[21:36:19] <alex_joni> use emcStatus ;)
[21:36:24] <anonimasu> yeah
[21:36:43] <paul_c> * paul_c already has a Qt NML Status lib
[21:36:46] <anonimasu> emcStatus->axis.x.position somthing isnt it..
[21:36:56] <alex_joni> you'd also want emcCommand for movement commands
[21:37:07] <alex_joni> and emcError (for estoping & co)
[21:37:25] <anonimasu> I dont know if I should generate code or throw code at emc..
[21:37:46] <anonimasu> paul_c: care to share?
[21:38:05] <anonimasu> alex_joni: what do you think?
[21:38:14] <alex_joni> throw code at emc
[21:38:20] <anonimasu> hm, ok
[21:38:38] <alex_joni> nml is pretty easy
[21:38:59] <paul_c> NML is very easy.
[21:39:10] <alex_joni> paul_c: I agree ;)
[21:39:19] <alex_joni> but I kinda stumbled on that message_nr
[21:39:25] <anonimasu> * anonimasu nods
[21:39:26] <alex_joni> never figured that out completely
[21:39:40] <paul_c> peek(buffer) - read data from a struct.
[21:39:58] <anonimasu> I just had a quick look at handwheel.cc that's about my experience with nml
[21:41:41] <alex_joni> most NML I did was for ioControl.c in emc2
[21:43:27] <anonimasu> :)
[21:47:25] <alex_joni> make that ioControl.cc ;)
[21:48:46] <anonimasu> *wishes for a c++ library for accessing nml super quickly..
[21:48:48] <anonimasu> *
[21:49:05] <anonimasu> but that's being too lazy
[21:49:05] <anonimasu> ;)
[21:49:08] <alex_joni> * alex_joni hands anonimasu libnml
[21:49:11] <alex_joni> :D
[21:49:39] <anonimasu> heh, well wrapper library.. for grabbing stuff without a thought on how it's done..
[21:49:50] <anonimasu> get(foo)
[21:49:50] <anonimasu> ;)
[21:51:00] <alex_joni> get(out(here))
[21:51:04] <alex_joni> *grin
[21:51:13] <anonimasu> heh
[21:53:56] <paul_c> writing a(nother) wrapper round NML misses the the advantages of Qt
[21:54:00] <rayh> catch you all later
[21:56:12] <anonimasu> laters rayh
[21:56:21] <alex_joni> * alex_joni leaves too
[21:56:23] <alex_joni> laters
[21:56:30] <paul_c> part timer
[21:56:46] <alex_joni> heh ;)
[21:57:02] <alex_joni> not all of us can have your commitment :P
[21:57:37] <alex_joni> it would be too good if we could
[22:01:59] <rayh> An Offer They Cannot Refuse: In an even more stunning reversal, Microsoft has invited the open source community for a sit-down to drink grappa and mangiare some cannoli. You know that scene in mob films where the godfather invites all his rivals for a meeting, excuses himself to go to the loo, and guys with submachine guns show up? My advice to open sourcers: If Ballmer leaves the room, get ready to dive under the table. Infoworld mag
[22:02:50] <rayh> http://www.infoworld.com/article/05/05/13/20OPcringely_1.html?source=NLC-OS2005-05-18
[22:03:19] <Phydbleep> rayh: Maybe all the open-source geeks should take a bathroom break first. :)
[22:03:58] <rayh> Good plan.
[22:04:14] <anonimasu> yeah..
[22:04:29] <anonimasu> hm, I dont quite get why the open source community would go there..
[22:05:05] <anonimasu> ;)
[22:19:13] <Phydbleep> anonimasu: Free cyanide punch and arsenic frosted doughnuts?
[22:21:45] <Phydbleep> anonimasu: And they had all those 220V/100A walkman headphones made with the msn logo just for this event. :)
[22:21:58] <anonimasu> ^_^
[22:22:02] <anonimasu> yep
[22:28:32] <CIA-4> 03rumley * 10emc/ (20 files in 4 dirs): (log message trimmed)
[22:28:33] <CIA-4> 1. Fix for invalid axis position readout in the case of AUTO mode, G92 in effect, and 'abort'. Involved prohibiting read-ahead past G92-G92.3, G54-G59.3 and M02/M30/M60 commands.
[22:28:33] <CIA-4> 2. Added G83.1 peck drilling cycle. A slight variation on the EMC implementation of G83. Drill retracts an incremental amount, not full retract from hole. Amount may be set in ini file.
[22:28:33] <CIA-4> 3. Added ini file initializations for the RS274NGC interpreter, making it easier
[22:28:33] <CIA-4> for non-programmers to modify the interpreter behavior to suit one's likings.
[22:28:35] <CIA-4> Added the following ini file parameters (moved some from rs274ngc.hh, also)
[22:28:36] <CIA-4> PARAMETER_FILE
[22:30:02] <anonimasu> can you run emc2 with the emc1 interpreter or are they entirely different?
[22:30:56] <paul_c> not without some changes.
[22:31:15] <anonimasu> ok
[22:33:17] <anonimasu> I am stress cleaning
[22:33:27] <anonimasu> birthday + visitors tomorrow
[22:33:27] <anonimasu> :)
[22:38:32] <paul_c> OH FFS!!
[22:40:16] <anonimasu> ?
[22:40:27] <anonimasu> grandparents ;)
[22:44:36] <paul_c> Rumley - pointless code in emcmot.c, multiple files with exactly the same changes..
[22:44:52] <paul_c> undocumented changes to G codes.
[22:45:01] <anonimasu> heh ok
[22:45:26] <anonimasu> I thought you were FFS:ing me on my cleaning ;)
[22:45:32] <anonimasu> cant have coke bottles all around the place..
[22:45:38] <paul_c> It is a relief that that interpreter is not a plugin and go for the emc2 tree.
[22:46:01] <anonimasu> yep
[22:46:31] <paul_c> Oh, and it breaks the translation files & probably codegen too.
[22:47:26] <anonimasu> horrid
[22:48:13] <paul_c> It's the undocumented G code changes I have most problems with.
[22:49:33] <CIA-4> 03rumley * 10emc2/src/emc/task/emctask.cc:
[22:49:34] <CIA-4> Fix for Bug 1171692, Odd behavoir in MDI mode.
[22:49:34] <CIA-4> Caused by rs274ngc_synch call when axis not 'inpos'
[22:50:34] <robin_sz> * robin_sz sees the breaking of codegen as a good thing
[22:54:38] <anonimasu> howcome?
[22:55:07] <robin_sz> well, codegenerators are usually a sign of bad design
[22:55:31] <anonimasu> what does the codegenerator do?
[22:55:37] <anonimasu> add g/m codes?
[22:56:06] <robin_sz> mostly it generates the large lumps of bloated code I complain about most often :)
[22:56:54] <robin_sz> its a javarey thing
[22:57:06] <robin_sz> ive never actually managed to get it to run though
[22:57:42] <anonimasu> hm ok
[22:58:06] <anonimasu> well, if you cant edit the code they output you shouldnt be coding..
[22:58:18] <anonimasu> :)
[22:58:40] <les> hey robin
[22:58:45] <robin_sz> hi les
[22:59:08] <les> i'm gonna break silence here
[22:59:13] <anonimasu> evening les
[22:59:15] <les> WTH is this?
[22:59:17] <anonimasu> or night/whatever it is
[22:59:25] <les> http://www.somtn.com/
[22:59:41] <paul_c> codegen just produces the constructors, overloads the update() function, & generates the emcFormat()
[22:59:49] <anonimasu> hm ok
[23:00:15] <paul_c> simpe & repetative code.
[23:00:22] <robin_sz> les: look slike it belongs to that ebay guy/girl
[23:00:55] <les> looks from china to me
[23:01:30] <paul_c> anonimasu: Someone was working on a Python tool to replace the java one.
[23:01:36] <robin_sz> les: you know where it comes from :) I suspect it is that guy we saw on eBay's site ..
[23:02:39] <les> hmmm
[23:02:41] <les> ok
[23:04:13] <robin_sz> that would be my guess anyway
[23:04:32] <les> their forum is in chinese.
[23:04:55] <les> Oh well still no machines in the US
[23:05:06] <robin_sz> 'k
[23:06:12] <les> well...back to the music room
[23:06:23] <anonimasu> enjoy :)
[23:06:30] <les> ty later
[23:06:43] <robin_sz> les: the pictures link is interesting
[23:17:15] <robin_sz> anonimasu: the point about "well, if you cant edit the code they output you shouldnt be coding.." is incorrect. the point is if you hand edit the output of a codegen section of the system, you stand a good chance of losingthose additions the next time someone runs the codegen
[23:18:18] <les> back for a sec
[23:18:18] <anonimasu> robin_sz: I would say that the codegen is bad then..
[23:18:26] <les> those pictures are from china
[23:18:34] <robin_sz> les: yeah
[23:18:36] <robin_sz> seems so
[23:18:41] <anonimasu> it should just spit out fairly complete code for the developer to modify to their needs.. not spit it all out..
[23:18:50] <robin_sz> ok, so I dont know if its the ebay guy anymore
[23:19:27] <anonimasu> robin_sz: since you said they was outputing bloated code..
[23:19:32] <anonimasu> was/were..
[23:20:08] <robin_sz> you know all those badly thought out classes? and the 700 line switch statements?
[23:20:17] <anonimasu> yeah
[23:20:23] <robin_sz> most of that is codegen output
[23:20:23] <paul_c> IF you make the changes in the input files, a codegen will NOT revert any modifications.
[23:20:29] <les> It's some chinese business types with a poor command of english trying to market in the US with laughable results
[23:20:49] <robin_sz> paul_c: define inptu files
[23:21:00] <anonimasu> good night everyone
[23:21:07] <paul_c> in the context of emc - emc.hh
[23:21:52] <robin_sz> right
[23:21:56] <anonimasu> going to go and sleep and figure a bit about the QT stuff..
[23:22:06] <les> night anon
[23:26:22] <robin_sz> paul_c: the problems arise when people who cant use the codegen or dont appreciate stuff is codegened manually edit codegened sections
[23:27:04] <robin_sz> or modify header input files in a "correct by C standards" way, that are not acceptable to the codegen system
[23:27:58] <anonimasu> well, too many cooks makes for a bad soup ;)
[23:28:12] <anonimasu> not nescessarily true, but gets quite multicultural
[23:28:12] <robin_sz> errm.
[23:28:30] <anonimasu> lots of codestyles ;)
[23:28:33] <robin_sz> I strongly disagree
[23:28:48] <anonimasu> robin_sz: it's the way it is with cooking.
[23:28:51] <robin_sz> well,
[23:28:57] <anonimasu> not with code
[23:29:00] <robin_sz> if the code is clean with a clear style
[23:29:03] <anonimasu> I was just kidding :)
[23:29:06] <anonimasu> yep
[23:29:08] <robin_sz> its easy to follow the recipie
[23:29:41] <robin_sz> when the code is dirty, with several styles, and a strong sprinkling of really REALLY bad use ipf C++ .. well ...
[23:29:49] <robin_sz> people try and put it right
[23:30:45] <robin_sz> even a consistent layout style would be a start
[23:31:40] <robin_sz> one of the things we discussed for emc2 was to just run the whole damn thing through indent
[23:31:56] <robin_sz> sadly, it didnt happen, and then work got done
[23:32:01] <paul_c> You have/had commit access - You could do something about it.
[23:32:15] <robin_sz> I read the warnings ...
[23:32:58] <robin_sz> about people making non-necessary changes to format
[23:33:06] <robin_sz> and what sanctions might be imposed
[23:37:01] <anonimasu> wb
[23:37:33] <paul_c> there comes a point when it is best to walk away from an argument.
[23:39:29] <anonimasu> I am trying to find stuff about how the teach in/online programming works on real mills..
[23:40:25] <paul_c> for example, conversational programming ?
[23:41:03] <anonimasu> more like panel input..
[23:42:39] <paul_c> I suspect that would be controller specific
[23:43:37] <anonimasu> yeah, but what functions do they have, what do you need..
[23:43:38] <paul_c> I know Heidenhain have a semi-auto mode for their lathe package....
[23:43:44] <anonimasu> how much settings do you need..
[23:44:05] <anonimasu> I looked at a haas lathe online they had the coolest feature.. the ability to have one axis controlled..
[23:44:17] <anonimasu> for turning radiuses and stuff..
[23:44:18] <anonimasu> very cool
[23:44:49] <anonimasu> where you turn a handwheel and the other axis is interpolated with the first axis as master..
[23:46:23] <paul_c> that gives you a jog wheel and a menu option as the minimum inputs required.
[23:47:03] <anonimasu> hm, that seems neat for quick stuff
[23:50:55] <paul_c> probably the best thing to do is to go to one of the trade shows and have a play.
[23:51:21] <anonimasu> yeah
[23:51:23] <anonimasu> but I am in sweden
[23:51:24] <anonimasu> :9
[23:51:35] <dmess> hi all...
[23:51:54] <anonimasu> so it's a bit hard
[23:52:06] <paul_c> Isn't Sandvik Swedish ?
[23:52:10] <anonimasu> yeah
[23:52:23] <dmess> german catalog???
[23:52:24] <paul_c> Go knock on their door...
[23:52:33] <anonimasu> it's in the other end of the country
[23:52:35] <anonimasu> :)
[23:52:44] <paul_c> Have a car ?
[23:52:46] <anonimasu> and the tradeshows too
[23:53:32] <paul_c> Evening dmess
[23:53:48] <dmess> hello paul.. ;)
[23:54:43] <anonimasu> I am digging control documents
[23:55:00] <dmess> out of where??
[23:55:28] <anonimasu> google....
[23:55:29] <anonimasu> :)
[23:55:35] <anonimasu> looking at online programming stuff
[23:55:41] <anonimasu> teach in mode and stuff like that
[23:55:44] <dmess> i see.. any option parameters??
[23:56:01] <anonimasu> what?
[23:56:32] <anonimasu> oh, I am just looking
[23:56:57] <dmess> all of them for say all fanuc controls... would be a nice list
[23:57:35] <anonimasu> ah, it's not what I am looking for
[23:57:37] <dmess> i have some but still a few missing..
[23:57:50] <anonimasu> I am looking at how teach in modes and stuff like that works..
[23:57:57] <dmess> teach in for what control??
[23:58:18] <anonimasu> emc.
[23:58:26] <anonimasu> in the end..
[23:58:27] <dmess> i can explain the intricacies of them..
[23:58:47] <anonimasu> but I am trying to find info on haas, and on heidenhain
[23:58:58] <anonimasu> please do
[23:59:13] <dmess> heidenhein teach in sucks..