#emc | Logs for 2009-03-14

[00:26:44] <jmkasunich> dareposte: what are you basing your "I need 3000 lbs" on?
[00:27:17] <jmkasunich> you have quite a toolpost if you are willing to put a ton and a half on the tool
[00:49:25] <Optic> is there a HAL signal for "currently running the job?"
[00:50:16] <eric_unterhause1> axis enable?
[00:50:41] <Optic> so if you hit the "stop" button (not the estop) it turns off the axis enable?
[00:51:07] <eric_unterhause1> sorry, I thought this was a game of free association
[00:55:23] <Optic> hmm
[01:03:07] <eric_unterhause1> why do you want it, would something like a spindle on signal work?
[01:22:44] <eric_unterhause1> I always feel bad when I'm the only one answering questions since I'm an idiot
[01:32:50] <cncjosh> yo yo ! anybody in the mood to help a n00b? :)
[01:33:12] <cncjosh> Just getting things running for the first time tonight. Running a G540 with EMC2 on 8.04, on an X2.
[01:33:40] <cncjosh> my leadscrew pitch is set correctly in EMC2, but I'm not moving the right distance. Does my microstepping have anything to do with this , if that value is set wrong?
[01:34:05] <cncjosh> and no - i haven't measured "exactly" how far I'm off yet -but its a magnitude.
[01:34:12] <archivist> could be
[01:34:28] <eric_unterhause1> measure and it might tell you easier than we can
[01:34:58] <jmkasunich> if you are off by exactly 5:1, 2:1, etc, that will point you in the right direction
[01:35:00] <eric_unterhause1> archivist: shouldn't you be sleeping?
[01:35:04] <cncjosh> I guess the question is - will that value have any bearing on distance travelled?
[01:35:34] <archivist> eric_unterhause1, sleep is for wimps
[01:35:41] <eric_unterhause1> overrated
[01:36:13] <cncjosh> and is that value something i need to look at my steppers for, or my stepper controller for?
[01:37:03] <eric_unterhause1> gecko
[01:38:28] <eric_unterhause1> assuming it's not a steps/rev problem
[01:38:51] <cncjosh> cool
[01:38:56] <cncjosh> yeah step/rev is correct.
[01:39:06] <cncjosh> making a change now, will know momentarily :)
[01:39:34] <cncjosh> sweet that did it. thx.
[01:39:39] <cncjosh> now to try another run :)
[01:47:18] <cncjosh> that got me in the ballpark. thx guys
[01:47:33] <cncjosh> now to start measuring and seeing what my backlash looks like. outta here. l8r
[06:14:53] <jedediah_> heyo.. I'm trying to write some gcode that will turn my laser on and off without interrupting motion
[06:15:53] <jedediah_> we've tried just about every signal.. z-pos, spindle, tool, coolant.. and they all cause a discontinuity in x/y motion
[06:17:26] <jedediah_> this is emc2
[06:37:36] <toastydeath> what about turning it on anytime the machine is in g1?
[06:37:39] <toastydeath> and off in g0.
[06:38:04] <toastydeath> although i don't get why Z pos would do it.
[06:38:11] <toastydeath> would cause discontinuity, that is
[06:38:18] <jedediah_> can we send that to a parallel port pin?
[06:38:27] <toastydeath> oh, i haven't a clue
[06:38:32] <toastydeath> you'll have to ask someone who understands emc
[06:38:56] <toastydeath> there's got to be SOME way to turn that into an output
[06:39:20] <eric_unterhause1> there was somebody on here in the last couple of weeks that was trying to do this
[06:39:29] <jedediah_> we've tried various ways of mapping on/off to the z axis and we always either get a discontinuity or inaccurate timing
[06:56:09] <dareposte> jmkasunich: I am basing it on the published specs of the haas tl-1
[06:56:35] <dareposte> although now that i think about it, that is probably their "peak" servo output
[06:56:40] <dareposte> and not the continually rated output
[06:57:07] <dareposte> i don't particularly like the haas, but it has the spec's conveniently published
[06:57:19] <dareposte> and it seems to be more than adequate for the job
[06:58:34] <hypa7ia> eric_unterhause1: jedediah_ is working with Optic, that might have been who you are thinking of :)
[07:02:04] <jedediah_> yo
[07:02:38] <hypa7ia> hai
[07:02:41] <hypa7ia> * hypa7ia is lurking
[07:03:06] <jedediah_> elmood is probably going to help me build another PIC for the laser that handles all the realtime stuff
[07:12:12] <hypa7ia> nice
[07:20:50] <hypa7ia> sleep now! nite folks :)
[08:31:02] <alex_joni> jedediah_: the only thing I know is using M64/65, but you need emc2-2.3.0-beta for that
[17:04:02] <Optic> mooo
[17:05:01] <eric_unterhause1> lah
[19:24:09] <dareposte> sorry I had to ditch out yesterday, I never did figure out if the "maximum thrust" listed in manufacturer's machine specs was the servo peak output, or if it was maximum sustained output
[19:25:19] <dareposte> for example: http://haascnc.com/specs.asp?ID=SL-20&mdl=SL-20&webid=LATHE_SL#CNCLatheTreeModel
[19:25:35] <dareposte> for the Haas SL-20 it lists X maximum thrust as 2400 lbs
[19:25:44] <dareposte> and Z as 2400 lbs
[19:25:51] <dareposte> no Z as 3700 lbs
[19:26:25] <dareposte> so since I don't have the benefit of a big R&D facility or test bed, I just want to duplciate what they have done in terms of servo and screw sizing
[19:26:49] <dareposte> under the assumption that it is a typical and acceptable setup for most cnc turning
[19:27:55] <dareposte> so if it's peak output, then the servos could be sized at 700 in-oz peak
[19:28:24] <dareposte> which seems pretty reasonable
[19:28:59] <dareposte> but if its continuous output, the servos would have to be sized for 700 in-oz continuous, which is much larger servos
[19:33:47] <fenn> it's peak
[19:34:48] <dareposte> fenn: how do you know?
[19:35:01] <fenn> common sense
[19:35:11] <fenn> if i were a specman
[19:35:22] <dareposte> good point
[19:35:24] <fenn> and someone said 'write a brochure with our technical specs on it'
[19:35:29] <fenn> then i'd use peak values wherever possible
[19:35:52] <dareposte> do you know what frame size servos are common on that type of equipment?
[19:35:59] <fenn> no
[19:36:02] <dareposte> the only servos i could find with enough torque were NEMA56's
[19:36:06] <dareposte> which seems waaaay too big
[19:36:22] <fenn> i know the servos on the mazak were about 6 inches diameter
[19:36:48] <dareposte> maybe they are NEMA56 sized then
[19:36:56] <fenn> the haas looks quite a bit smaller though
[19:37:14] <fenn> can't you just go look at one?
[19:37:24] <dareposte> i could if i had one
[19:37:29] <dareposte> i think there's a dealership in cinci
[19:37:45] <dareposte> i don't particularly care about the haas, just using it as an example
[19:38:10] <dareposte> their spec is one of the easier ones to find
[19:38:51] <dareposte> one of the bridgeports at work has a CNC kit on it
[19:38:58] <dareposte> and the servos look bigger than 700 in-oz peak though
[19:46:02] <dareposte> http://www.flashcutcnc.com/beta/php/bridgeportRetrofitKits_specsServo.php
[19:46:09] <dareposte> that says 520 in-oz continuous
[19:49:01] <dareposte> for a bridgie type conversion
[20:05:58] <Optic> hey all
[20:06:36] <dareposte> hi
[20:06:54] <Optic> We got our engraver working!
[20:07:09] <dareposte> good work
[20:07:16] <dareposte> made anything yet??
[20:07:27] <Optic> It needs some tweaking
[20:07:45] <Optic> We burned some text into cardboard. :)
[20:08:30] <Optic> We need to find a way to turn the laser on and off without emc stopping the motion
[20:09:06] <Optic> Were using z axis now
[20:13:40] <alex_joni> Optic: 10:25 < alex_joni> jedediah_: the only thing I know is using M64/65, but you
[20:13:41] <alex_joni> need emc2-2.3.0-beta for that
[20:14:12] <alex_joni> but the switching can only happen when you have a different move
[20:14:25] <alex_joni> so if you're going from X0 to X10, and want to switch laser on at 2.54
[20:14:35] <alex_joni> then you need a move G1 X2.54
[20:14:40] <alex_joni> M64 Pxx
[20:14:42] <alex_joni> G1 X10
[20:14:50] <alex_joni> or something like that
[20:19:42] <BigJohnT> Optic: can't you use the spindle on and off to control your laser?
[20:21:31] <BigJohnT> oh you would have to stop for that...
[20:22:16] <invite|MkS> Optic: Or coolant/mist on and off ?
[20:22:43] <BigJohnT> that would still stop motion...
[20:23:09] <jedediah_> m64/65 doesn't stop?
[20:57:12] <JymmmEMC> * JymmmEMC pokes alex_joni
[21:07:43] <BigJohnT> I think you missed JymmmEMC
[21:07:55] <jmkasunich> need a longer stick
[21:08:27] <JymmmEMC> heh
[21:08:54] <BigJohnT> LOL
[21:09:10] <Optic> Bigjohn: those commands affect movement
[21:09:18] <Optic> We did some tests
[21:09:34] <Optic> We also tried spindle speed
[21:09:45] <Optic> And tool change. :)
[21:09:46] <BigJohnT> I kind of thought that after I typed it
[21:10:24] <BigJohnT> what you really need is two seperate controls one to move your carriage and one to fire the laser I guess
[21:10:50] <Optic> Emc is so close to being the right thing
[21:11:19] <BigJohnT> your printing with a laser more or less right?
[21:11:33] <Optic> Yeah
[21:11:54] <Optic> So speed affects kerf
[21:11:57] <Optic> Basically
[21:12:04] <BigJohnT> we need a printer driver for EMC :)
[21:12:04] <alex_joni> err.. sorry :)
[21:12:16] <alex_joni> no, in 2.3.x M64/65 will be seamless integrated
[21:12:23] <Optic> Woooo
[21:12:25] <alex_joni> that means they don't screw up motion or blending
[21:12:41] <Optic> That is *perfect*
[21:12:58] <Optic> Are they in any betas?
[21:13:09] <alex_joni> yup
[21:13:14] <alex_joni> 2.3.0~beta1
[21:14:09] <alex_joni> Optic: pm (private message)
[21:15:10] <BigJohnT> alex_joni: is m64/65 changed in real time now?
[21:15:23] <alex_joni> BigJohnT: it was before too
[21:15:27] <alex_joni> I mean in 2.2.x
[21:15:37] <alex_joni> but it was poorly implemented, and it broke blending
[21:15:45] <alex_joni> which caused moving hickups
[21:16:28] <BigJohnT> ok then the manual is wrong
[21:16:37] <alex_joni> well.. to be frank in 2.2.x you can only use M62/63 (which are the ones that break blending), M64/65 didn't do anything (they were implemented at one point, then reverted when the TP got rewritten)
[21:16:54] <alex_joni> but in 2.3.x you have 62/63 which get issued immediately when motion receives them
[21:17:09] <alex_joni> and 64/65 which get queued and will get issued when the next move starts
[21:20:23] <BigJohnT> so is this wrong alex_joni http://www.linuxcnc.org/docview/devel/html//gcode_main.html#sec:M62-to-M65:
[21:20:58] <alex_joni> err.. that is right
[21:21:01] <alex_joni> * alex_joni was backwards
[21:21:27] <BigJohnT> does m64/65 still break blending?
[21:21:41] <alex_joni> I think so
[21:21:48] <alex_joni> but one would have to test it to make sure
[21:22:02] <alex_joni> but I don't think it can work otherwise
[21:22:22] <alex_joni> * alex_joni goes look at the code
[21:23:59] <fenn> what is the use case for m64/65?
[21:24:25] <fenn> i dont see how it could be useful to turn something on at some random time whenever the interpreter feels like it
[21:24:33] <alex_joni> fenn: not wanting to use a motion with the change
[21:24:48] <alex_joni> maybe MDI for debugging
[21:26:47] <BigJohnT> * BigJohnT goes out to plasma more deer critters out
[22:26:03] <alex_joni> good night all
[22:41:42] <invite|MkS> good night
[22:57:42] <Optic> hihi
[23:52:29] <Optic> moo
[23:55:15] <BigJohnT> ...