#linuxcnc | Logs for 2012-04-11

Back
[07:47:10] -!- logger[psha] [logger[psha][email protected]] has joined #linuxcnc
[07:52:03] <Loetmichel> hmm
[07:52:20] * Loetmichel repeats himself...
[07:52:42] <Loetmichel> [07:49] <Loetmichel> hmmm... i think i have a solution for the gearing... can someone confirm that this will Produce 1:5 if the ring gear is fixed and the sun is input? -> http://www.cyrom.org/palbum/main.php?g2_itemId=13066 (shit, thats module 0,5, that will be a PITA to mill)
[07:52:42] <Loetmichel> [07:49] <Loetmichel> mornin' btw
[07:52:42] <Loetmichel> [07:52] <Loetmichel> correction: its module 0.65
[07:54:10] <Loetmichel> ring is 40 teeth, planets are 15 teeth, sund is 10 teeth
[07:54:13] <Loetmichel> sun
[08:02:03] <archivist> the maths http://www.roymech.co.uk/Useful_Tables/Drive/Epi_cyclic_gears.html
[08:05:24] <archivist> the barrel used top diag of arrangement III which are all normal gears no internal needed
[08:06:29] <archivist> see links at bottom
[08:14:08] -!- WalterN has quit [Read error: Connection reset by peer]
[08:14:08] -!- X704 has quit [Read error: Connection reset by peer]
[08:14:38] -!- WalterN [[email protected]] has joined #linuxcnc
[08:14:58] -!- logger[mah] has quit [Ping timeout: 272 seconds]
[08:19:45] Tom_garage is now known as Tom_itx
[08:28:29] Tom_itx is now known as Tom_L
[08:28:38] Tom_L is now known as Tom_itx
[08:31:23] -!- vladimirek [[email protected]] has joined #linuxcnc
[08:34:05] -!- rob_h [[email protected]] has joined #linuxcnc
[08:34:15] <awallin> kb8wmc, Jymm: there's andy and then there's andy. try to keep them separate :) I didn't go sailing round the world...
[08:38:49] <Loetmichel> archivist: i have read that elsewhere, thanks. i just wanted to have a confirmation that i havend got it wrong again
[08:39:02] <Loetmichel> "4 eyxes see more than two"
[08:39:07] <Loetmichel> eyes
[08:39:31] <Loetmichel> ... i still want that confirmation
[08:40:37] <archivist> did you miss the link to an online calculator
[08:40:57] <archivist> I did try your numbers
[08:42:09] -!- Valen [[email protected]] has joined #linuxcnc
[08:42:19] <archivist> bottom of page http://www.wmberg.com/tools/
[08:43:18] <Loetmichel> didnt see that.
[08:43:19] <Loetmichel> moment
[08:44:28] <Loetmichel> 5, 3 times passed... seems to work, thanks ;-)
[08:44:48] <Loetmichel> "this system can be assembled"
[08:44:50] <Loetmichel> ;-)
[08:45:08] <archivist> a timing belt drive would have less backlash
[08:45:49] <Loetmichel> archivist: and wouldn fit nicely in the mount that is already there
[08:45:54] <Loetmichel> +t
[08:46:35] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=11651
[08:46:46] <Loetmichel> that is the part i got for x
[08:47:15] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=11639 here is another view
[08:47:41] <Loetmichel> its a combinde linear ways with ball bearing and a ballscrew in one
[08:59:29] -!- rob_h has quit [Quit: Leaving]
[09:01:06] <Loetmichel> archivist: and as i make the gears myself by milling i can make them for minimal backlash by using slightly compressible material (PC or POM) and slightly "oversizue" them
[09:03:10] e-ndy is now known as e-ndy|afk
[09:06:30] e-ndy|afk is now known as e-ndy
[09:07:02] -!- thinpete has quit [Ping timeout: 248 seconds]
[09:22:08] -!- tehDarkAura has quit [Read error: Connection reset by peer]
[09:23:22] -!- phantoxe [[email protected]] has joined #linuxcnc
[09:24:02] <freespace> when using the G81 canned cycle, the F rate specifies the rate at which the z-axis moves?
[09:38:53] <awallin> yes, probably. x and y will move as if G0 was programmed
[09:41:58] <freespace> ok, thanks!
[09:42:49] -!- Cylly [[email protected]] has joined #linuxcnc
[09:42:56] -!- phantoxe has quit [Ping timeout: 252 seconds]
[09:43:29] -!- Loetmichel has quit [Ping timeout: 245 seconds]
[09:54:55] -!- psha[work] [psha[work][email protected]] has joined #linuxcnc
[09:57:07] -!- pingufan [[email protected]] has joined #linuxcnc
[10:00:46] -!- phantoxe [[email protected]] has joined #linuxcnc
[10:02:47] Cylly is now known as Loetmichel
[10:03:44] e-ndy is now known as e-ndy|afk
[10:03:54] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[10:06:59] -!- logger[mah] has quit [Remote host closed the connection]
[10:14:41] -!- H264 [[email protected]] has joined #linuxcnc
[10:17:47] <MattyMatt> Loetmichel I think that epicyclic calculator assumes the planets are evenly spaced. they don't have to be as someone pointed out to me v recently. 40:15:10 may work with 3 planets if they are not exactly 120° apart
[10:18:10] -!- sumpfralle [[email protected]] has joined #linuxcnc
[10:18:17] <MattyMatt> if there's room for the planets to clear each other of course
[10:19:48] <MattyMatt> the thrust will be off centre if they're not balanced but that shouldn't matter if the bearings are solid
[10:20:03] <MattyMatt> solid/sound/good
[10:20:09] <MattyMatt> not literally solid :)
[10:21:50] <MattyMatt> I've had one cup of coffee. that may be enough to draw it in blender, got a reprappable planetary box loaded already
[10:24:20] <MattyMatt> it's fairly obvious that multiples of 36 deg will fit for sure
[10:26:15] <MattyMatt> so 108° spacing would work
[10:26:24] <MattyMatt> if they fif
[10:26:27] <MattyMatt> fit
[10:28:36] -!- pingufan has quit [Quit: Konversation terminated!]
[10:30:49] <MattyMatt> do they all have to be on the same plane? the planets could pass each other. you could have 2 planes with 2 planets each, so you get 4 at 90 deg each
[10:31:33] <Jymmm> awallin: around the room, around the world, same diff.
[10:32:40] <awallin> hm, ok..
[10:32:57] <Jymmm> awallin: all you a's look the same to me
[10:33:14] <Jymmm> ;)
[10:36:27] <awallin> Jymmm: yeah, looking from the US we all live in the same place too, UK, Finland, Romania, all the same really, in europe
[10:38:10] <Jymmm> awallin: EurAsia
[10:38:42] <bedah> "on a scale 1 to 10, how european is germany?" "nein!"
[10:38:53] -!- pingufan [[email protected]] has joined #linuxcnc
[10:40:31] <MattyMatt> I'm a federalist. EU called be a citizen while UK still called me a subject
[10:41:02] <Jymmm> MattyMatt: And I just call you a suspect!
[10:41:31] <MattyMatt> latest UK passport says citizen of UK, but they're just playing Me Too
[10:42:26] <Jymmm> MattyMatt: I bet if we turned you over we'll find a "Made In China" sticker
[10:43:43] <MattyMatt> china were still driving dung carts when I was made
[10:44:25] -!- maximilian_h [[email protected]] has parted #linuxcnc
[10:44:30] <MattyMatt> that sets me up for a putdown :)
[10:44:35] <MattyMatt> don't go there
[10:45:11] <Jymmm> I won't, I thought the dung carts already took you there and back again.
[10:48:28] <Jymmm> This is a good looking bag... http://adder-gear.com/products.php?id=136#
[10:50:38] -!- Valen has quit [Quit: Leaving.]
[10:51:13] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[10:52:14] <archivist> hmm serfs are getting restless again
[10:53:44] <jthornton> lol
[10:54:20] <MattyMatt> it's checks and balances. they rule, we riot
[10:55:13] <MattyMatt> it's funny how the BBC never once correlated the royal weddings with the inner city riots
[10:57:14] <Jymmm> MattyMatt: Royal Wedding Riots, heh... Hell, win sports game riot!, Lose sports game riot,
[10:58:09] <Jymmm> After Rodney King, everything is an excuse to riot!
[10:58:57] <MattyMatt> the 81 riots were a black thing, but last years were all white kids looting
[11:00:39] <MattyMatt> they were sparked by a black kid being shot tho
[11:01:32] <MattyMatt> kid had a gun and was on his way somewhere with it, but he was popular
[11:01:45] <MattyMatt> and photogenic
[11:08:09] -!- erasmo has quit [Read error: Connection reset by peer]
[11:09:00] <MattyMatt> that happens every year tho, the only real similarity between 1981 and 2011 is the country had just been subjected to the spectacle of the heir to the throne getting married in military uniform
[11:10:27] <Jymmm> dont most leaders serve their country at one point of their lives?
[11:10:28] <MattyMatt> it still beats presidential elections tho :)
[11:10:54] <Jymmm> No it doesn't, those are fixed before the polls even open =)
[11:12:48] <MattyMatt> australia has it best, they are spared a president, and their head of state lives on the other side of the world
[11:13:39] <Jymmm> canada?
[11:14:31] <MattyMatt> and them
[11:27:43] -!- servos4ever [servos4ever!~chatzilla@74-45-201-222.dr01.hnvr.mi.frontiernet.net] has joined #linuxcnc
[11:29:17] -!- phantoxe has quit []
[11:30:51] <freespace> yeah, but we have tony abbott
[11:31:05] <freespace> it would be so embrassing if he becomes PM
[11:34:50] -!- phantoxe [[email protected]] has joined #linuxcnc
[11:35:54] <Jymmm> freespace: Well, PM would be better than Queen.
[11:36:45] <MattyMatt> we have a PM in UK too, but I think it helps that they know there's always someone above them
[11:37:22] <MattyMatt> just one more cog in the machine
[11:37:48] <MattyMatt> checks and balances :)
[11:38:27] <MattyMatt> a few more generations of internet users and we'll have direct democracy for sure
[11:38:37] <MattyMatt> no need for leaders
[11:39:14] <Jymmm> as long as theres dialup that is.
[11:39:17] <MattyMatt> AI should make the civil service mostly redundant
[11:39:40] <freespace> well if tony abbott becomes the queen... i would check to see if i am high
[11:39:51] <freespace> if not, there is something very wrong with the world
[11:40:42] <Jymmm> freespace: wrong if you're not high, or wrong if he isn't queen?
[11:40:58] <MattyMatt> openly queen or closet queen?
[11:40:58] <freespace> wrong if i am not high AND he is the queen
[11:41:05] -!- maximilian_h [[email protected]] has joined #linuxcnc
[11:41:36] <Jymmm> Eeeesh, now yoour sounding all rational
[11:42:05] -!- adb [[email protected]] has joined #linuxcnc
[11:42:09] <MattyMatt> bashing metal makes you rational, or finger/eye less
[11:42:48] <MattyMatt> none of my machines could eat more than a finger or an eye, except my table saw
[11:43:15] <Jymmm> MattyMatt: Your just not trying hard enough then
[11:43:25] <MattyMatt> yep
[11:43:41] <MattyMatt> I need a lathe that can wrap a man around the chuck
[11:44:57] <MattyMatt> or a woman. I try not to be sexist
[11:45:14] <freespace> *shudders*
[11:45:36] <MattyMatt> I said CAN, not DOES :)
[11:45:48] <freespace> stuck to the door to the student's machine shop at uni is a graphical picture of what happens long hair is entangled in a lathe
[11:45:49] <Jymmm> who needs divorce attorneys when you have a woman eating lathe
[11:46:10] <freespace> ... all you really see is 2 set of legs stucking out, and a lot of blood
[11:46:30] <freespace> it has made me somewhat terrified of lathes
[11:46:31] <Jymmm> ob/gyn
[11:46:42] <freespace> *sticking
[11:51:05] <jthornton> how about the old timers that stopped their chuck with their hand or file a chamfer on a real short part while the chuck is spinning (I do that)
[11:51:34] <freespace> i trust they know what they are doing:)
[11:51:43] <freespace> or rather, you don't get to be old timers if you don't :D
[11:53:09] <jthornton> any clues what this error means when I run the synaptic package manager? "Failed to fetch http://linuxcnc.org/dists/lucid/Release Unable to find expected entry linuxcnc2.5-dev/binary-i386/Packages in Meta-index file (malformed Release file?)"
[11:54:53] <jthornton> never mind I found it
[11:55:47] -!- Tom_itx has quit [Ping timeout: 252 seconds]
[11:55:52] -!- MarkusBec has quit [Excess Flood]
[11:56:00] -!- zlog has quit [Ping timeout: 265 seconds]
[11:56:48] -!- pingufan has quit [Quit: Konversation terminated!]
[11:56:56] -!- MarkusBec [[email protected]] has joined #linuxcnc
[11:56:56] -!- MarkusBec has quit [Excess Flood]
[11:57:21] <TekniQue> 11:43:40 < MattyMatt> I need a lathe that can wrap a man around the chuck
[11:57:26] -!- MarkusBec [[email protected]] has joined #linuxcnc
[11:57:27] -!- MarkusBec has quit [Excess Flood]
[11:57:29] <TekniQue> I have one, albeit not CNC controlled
[11:57:32] <TekniQue> 18kW motor
[11:58:12] -!- zlog [[email protected]] has joined #linuxcnc
[11:58:26] -!- MarkusBec [[email protected]] has joined #linuxcnc
[11:58:26] -!- MarkusBec has quit [Excess Flood]
[11:58:56] -!- MarkusBec [[email protected]] has joined #linuxcnc
[12:07:55] <jthornton> how many HP is 18kW?
[12:08:24] <awallin> divide by 0.76 or something...
[12:09:02] <Jymmm> Tecan: 18000/.76
[12:09:12] <Jymmm> !18000/.76
[12:09:14] <jthornton> that's pretty big then
[12:12:47] <r00t4rd3d> this thing sounds like a air raid going off
[12:12:47] <r00t4rd3d> http://www.youtube.com/watch?v=6L-W38sd65E&feature=channel
[12:13:23] <archivist> 746 watts per hp
[12:15:09] <jthornton> I like the hose clamp beating the crap out of the wire
[12:16:11] <r00t4rd3d> lol i noticed that too
[12:16:24] <MattyMatt> acrylic gantry :p
[12:17:11] <archivist> his flexible friend
[12:17:13] <Loetmichel> BTDT
[12:17:24] <Loetmichel> some prette 15 years ago... #;-)
[12:17:48] <MattyMatt> flexible to a point, then catastrophic
[12:17:52] <r00t4rd3d> 1/2 plexi would probably work good
[12:18:11] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=2859
[12:19:01] <MattyMatt> mine sounds like a dogfight between a M109 and a Spitfire when it's doing big circles
[12:19:22] <MattyMatt> 3 octaves between X & Y leadscrews
[12:19:24] <r00t4rd3d> mine is silent
[12:19:35] <r00t4rd3d> cause I have not built it yet :D
[12:20:13] <Loetmichel> my actual sounds just like any other stepper driven CNC
[12:20:25] <Loetmichel> but quietly so
[12:21:42] <MattyMatt> plywood frame on mine gives it a resonance like it's distant.
[12:21:58] <Loetmichel> ... and even works in the car trunk: http://www.cyrom.org/MC/fraese_in_auto.mp4
[12:25:07] <MattyMatt> this one sounds disgusting to me http://v.youku.com/v_show/id_XMjM2NTMzOTk2.html
[12:26:01] <MattyMatt> it convinces me to try and get servos if I ever get a real iron mill
[12:26:55] <MattyMatt> my plywood one may be floppy, but at least it damps all the high overtones
[12:29:48] <MattyMatt> I've got 5 phase steppers ready to go one when I've made/bought drivers. they sound even worse than bipolar
[12:36:52] <asdfasd> Loetmichel: this is mobile version cnc, now you need one pocket size :)
[12:37:32] -!- rob_h [[email protected]] has joined #linuxcnc
[12:40:26] <MattyMatt> built into a helmet with big magnets to affect the motor part of the brain. grip spindle tightly in right hand, press start and when you wake up you've milled something good
[12:40:32] <Loetmichel> asdfasd: taht IS pockt size
[12:40:42] <Loetmichel> 200mm*110mm*110mm movement
[12:41:31] <asdfasd> some 1.44 floppy drives have tiny step motors like 10mm diameter
[12:41:50] <asdfasd> with leadscrew like 20mm travel
[12:41:55] <MattyMatt> torque on those is negligable
[12:42:00] <Loetmichel> and already a leadscrew, i know
[12:42:48] <MattyMatt> now your 70kB floppies often had nema17s with those integrated leadscrews
[12:43:55] <MattyMatt> I've got an IBM PC portable here with 2 broken floppies. not ripped it apart yet as I'm not sure if it's a museum piece
[12:45:01] <asdfasd> with that small machine you can engrave on rice
[12:45:06] <asdfasd> http://www.craftandartisans.com/sites/default/files/imagecache/inside-view/Rice%20Calligraphy.jpg
[12:45:07] <asdfasd> :)
[12:45:52] <MattyMatt> you can do that on a large machine if you have the resolution
[12:46:02] -!- skunkworks [[email protected]] has joined #linuxcnc
[12:46:48] <asdfasd> yeah, but will be such a fun looking that small machine fully working
[12:48:08] <archivist> if your travels are too small it becomes difficult to fit tooling
[12:48:20] <MattyMatt> a machine that fits in a 5.25" bay would be neat
[12:48:52] <MattyMatt> a lightscribe drive is almost a general purpose laser machine already
[12:49:20] <archivist> my X is about 60mm travel but I manage small http://www.collection.archivist.info/archive/DJCPD/PD/2012/IMG_1213.JPG
[12:50:09] <MattyMatt> seen that pic before :) very nice
[12:50:33] <MattyMatt> are you just repairing a clock/watch or are you making whole ones?
[12:51:09] <asdfasd> too big and too small it is always interesting to see
[12:51:11] <asdfasd> http://www.youtube.com/watch?v=5t5qb0GWnsg&feature=relmfu
[12:51:34] <archivist> was just testing if I could so I could quote one off for that little gear top left
[12:53:22] <MattyMatt> a CNC toolgrinder that does sub 1/8" tooling would be useful to me eventually
[12:53:43] <MattyMatt> I'm trying to get away from tiny tooling tho, except for PCBs
[12:54:19] <MattyMatt> new spindle I'm making has ER16 collets
[12:54:36] <archivist> I dremel hss under a stereo zoom microscope to make tooling
[12:55:36] <MattyMatt> yep that's my current option. a bit hit & miss handheld under a single big magnifying lens tho
[12:55:46] <saschi> hrhr, the flute ;D
[12:56:24] <MattyMatt> I've successfully resharpened a 0.3mm drill bit tho, enough to finish the brass hole I was drilling
[12:57:36] <asdfasd> wow
[12:58:16] <saschi> sick
[12:59:09] <archivist> watch makers type drills should be easily made on a simple tool grinder
[13:01:43] <MattyMatt> I'll turn one of my dremels into a small toolgrinder when my ER16 spindle is working right
[13:03:19] <MattyMatt> I've got my old mono microscope to use but I think 2 webcams may be better
[13:03:49] <archivist> cant find online pics of the drill bit type Im thinking of
[13:04:06] <MattyMatt> is it fluteless?
[13:04:13] <archivist> yup
[13:04:41] <asdfasd> probably pyramid sharpened?
[13:06:02] <archivist> sort of flat straight flutes really
[13:06:58] <asdfasd> http://img.inkfrog.com/click_enlarge1.php?image=New0011.JPG&username=144mph&aid=360240846
[13:07:14] -!- fenn has quit [Read error: Operation timed out]
[13:08:04] <MattyMatt> 4mm. even those are too big for my dremel
[13:08:25] <asdfasd> I mean the sharpening
[13:08:34] <archivist> http://www.esslinger.com/drills-mascot-carbon-steel-flat-pivot-12-piece-assortment.aspx
[13:08:34] <MattyMatt> 1/8" collets are very limiting
[13:08:57] <MattyMatt> yarr, just moaning generally about my current 1/8" limit :)
[13:09:51] -!- kanzure has quit [Ping timeout: 252 seconds]
[13:09:58] <asdfasd> 1/8 is a nice size, there is a lots of tools
[13:10:06] -!- ybit has quit [Ping timeout: 252 seconds]
[13:10:59] <asdfasd> currently Im using er16 and I have several collets, but only 1/8 is used, the rest still unopened
[13:11:22] -!- sumpfralle has quit [Read error: Operation timed out]
[13:11:52] <MattyMatt> I got a set of 8mm shank wood router bits. I'm looking forward to "hogging" with those
[13:11:59] <Loetmichel> archivist: when i see your workpieces i belive the stereo
[13:12:17] <Loetmichel> i can use bare eye but using TC bits
[13:12:32] <Loetmichel> hmmm
[13:12:52] -!- pingufan [[email protected]] has joined #linuxcnc
[13:13:10] <Loetmichel> bit i only use tools down to 0,3mm ;-)
[13:13:22] <Loetmichel> maybe a engraving biV-bit
[13:13:33] <Loetmichel> with 0,05 or so (tip)
[13:14:28] <MattyMatt> I don't even have a dremel collet that'll hold a 0.3mm drill bit. I needed to wrap it in tape
[13:15:12] <pingufan> Hi, I am (still) trying to get g38.2 working as expected. It stops at that z-value which I define. The problem is: It uses the coordinates-origin of my workpiece, and this can heavily vary. How can I code that it always moves down to the machine's lowest z-value?
[13:15:55] <pingufan> Background: I want to put the code for probing the tool length into a external file.
[13:15:58] <MattyMatt> then I found that the bit held sideways in a vicegrip was more effective than trying to get it centred in my tailstock chuck
[13:16:44] -!- kanzure [[email protected]] has joined #linuxcnc
[13:17:04] -!- fenn [[email protected]] has joined #linuxcnc
[13:17:15] -!- ybit [[email protected]] has joined #linuxcnc
[13:18:43] <MattyMatt> I've made an edm drilling rig now, for those nozzle holes. I'm gonna try a razor blade in it to see if I can do fine slitting, then I can make zero size dremel collets
[13:19:09] <asdfasd> pingufan: for example G28 go to machine 0, G 28.1 do almost same but it is always using current coordinate system
[13:20:00] <pingufan> G28 is identical to G38.2 except that it uses machine coordinates?
[13:20:18] <pingufan> I mean G28.2
[13:20:48] <asdfasd> pingufan: depends what you are triyng to do you may try with G53
[13:21:44] <asdfasd> g28 moves several axes one after another, G38 moves all same time
[13:21:49] <pingufan> I want to touch a sensor to measure the tool length. With first touch, I measure the reference tool, then I calculate the difference to this one on further touches.
[13:22:51] <pingufan> The point is: When I have the z0 point higher than the sensor, I have a problem. Motion of g38.2 stops at zero or where I define,.
[13:23:28] <pingufan> And when I define a point at i.e. z-20 this exceeds possibly the machine limits when I use a more flat workpiece.
[13:23:47] <asdfasd> then why not just G28 ?
[13:24:03] <pingufan> Didn't know that.
[13:24:09] -!- mhaberler [mhaberler!~mhaberler@2001:858:745:a0e4:a82d:dfa5:2141:218f] has joined #linuxcnc
[13:24:15] <asdfasd> it is going to machine 0
[13:24:16] -!- ries [[email protected]] has joined #linuxcnc
[13:24:41] <pingufan> and stops when the probe sensor is hit?
[13:24:43] <asdfasd> and it is safer than g38
[13:24:52] <MattyMatt> http://i.imgur.com/qubWX.jpg?1 my EDM drill and the pin vice I bought as a collet chuck donor (as I can't cut threads on my lathe yet)
[13:25:01] <asdfasd> no
[13:25:30] <asdfasd> it stops at machine coordinates, where the machine 0 is
[13:25:30] <pingufan> So how shall I use it to sense the tool tip?
[13:25:46] <asdfasd> just a second
[13:26:00] <MattyMatt> all delrin as I wasn't sure if alu would transmit magnetic field into the workpiece (and my lathe)
[13:26:43] <asdfasd> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ClassicLadderExamples#Single_button_probe_touchoff
[13:27:05] -!- ries_ [[email protected]] has joined #linuxcnc
[13:27:38] -!- ries has quit [Read error: Connection reset by peer]
[13:27:39] ries_ is now known as ries
[13:28:27] <MattyMatt> when I can cut threads I'll try and get the edm drill into the size of a jacobs chuck, encased in steel
[13:28:49] <pingufan> Do you suggest me to fiddle around in configuration files? :) (I am a novice)
[13:28:54] <TekniQue> what do you use for EDM electrodes?
[13:28:57] <TekniQue> tungsten?
[13:29:13] <MattyMatt> some do. piano wire suffices
[13:29:18] -!- logger[mah] has quit [Remote host closed the connection]
[13:29:24] <TekniQue> ok
[13:29:33] <TekniQue> I've never seen an EDM machine in person
[13:29:34] <asdfasd> how you do that now? MDI?
[13:29:50] <TekniQue> but I've seen them on youtube a few years ago
[13:30:01] <TekniQue> looked like many of them submersed everything in water
[13:30:31] <pingufan> I currently use an ASCII editor to write my .ngc files (to learn G-Coding) and then I load the file with AXIS.
[13:30:52] <MattyMatt> http://www.youtube.com/watch?v=uUN4_-xp1Wc my direct inspiration
[13:31:28] -!- joe9 [[email protected]] has joined #linuxcnc
[13:32:52] <pingufan> asdfasd: I actually presume(d) that there are ready made G-Codes, covering this function.
[13:33:24] <pingufan> But I was surprised that they work in user coordinate system.
[13:34:27] <Loetmichel> MattyMatt: sinche the TCbits have a 1/8" shaft that is not a problem... but being TC its a bit prone to breaking
[13:35:13] <Loetmichel> MattyMatt: http://www.cyrom.org/palbum/main.php?g2_itemId=8961
[13:35:19] <MattyMatt> yep I've got a growing bile of TC 1/8" shanks :)
[13:35:21] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=8958
[13:35:29] <MattyMatt> I call them "blanks"
[13:35:35] <MattyMatt> or "posh nails"
[13:35:45] <Loetmichel> (bought ones) http://www.cyrom.org/palbum/main.php?g2_itemId=11933
[13:36:20] <pingufan> Hi, Loetmichel! Meanwhile I bought 2x 50 pcs of tools. Thanks for your hint.
[13:36:32] <Loetmichel> hrhr, i have about 1kgh of broken tc shafts ;-)
[13:37:05] <pingufan> What the hell are you doing to collect 1kg of them?
[13:37:20] <pingufan> :D
[13:37:35] <Loetmichel> pingufan: de nada, pleased to help
[13:38:05] <Loetmichel> over 46 years, 3 employed as a CNC mill expert
[13:38:15] <Loetmichel> for a model cplane company
[13:38:16] <pingufan> You can't pherhaps help me with coding a generic tool length probing.
[13:38:36] <Loetmichel> no
[13:38:39] <asdfasd> pingufan: I read your again
[13:38:48] <Loetmichel> s/46/6
[13:39:03] <asdfasd> when Z0 is higher?
[13:39:32] <asdfasd> if you change with thinner workpiece?
[13:40:16] <pingufan> I try to do the following: I want to write an external subroutine for measuring the first tool length, and another subroutine for the other tools (I manually change).
[13:40:31] <pingufan> So I must enter there a fixed z height.
[13:40:40] <MattyMatt> that edm might be the way to turn these shanks back into tooling. edm doesn't care about the workpiece hardness
[13:40:42] <pingufan> ... for g38.2
[13:41:02] <asdfasd> ok then probably you have to reset previous offsets
[13:41:16] <asdfasd> like g10 L2 p1 z0
[13:41:36] <pingufan> Sure. But what z-value do I use?
[13:41:45] <asdfasd> before using G38.2 again
[13:42:00] <pingufan> No. You mis-understand.
[13:42:13] <TekniQue> what is a TC shank?
[13:42:26] <TekniQue> tungsten carbide?
[13:42:34] <MattyMatt> yep
[13:43:14] <TekniQue> many of the TC router and drill bits I've seen had a shaft made of something else
[13:43:27] <pingufan> My mill has 60mm (0..60mm) for Z-AXIS. Some workpieces are flat (PCBs i.e.), but the tool sensor is always in same height of ~19mm.
[13:43:34] <TekniQue> with the end welded on
[13:43:47] <MattyMatt> could be
[13:44:22] <pingufan> As g38.2 uses user coordinates, the value for Z will vary, depending on the user coordinate system.
[13:44:31] <pingufan> do you understand now?
[13:44:39] <MattyMatt> there's not a lot of material in them tho, so I'm hoping some of them are solid TC
[13:44:44] <asdfasd> yes
[13:45:26] <Loetmichel> Tecan: as the TC bits i use are called "full hard metal" in german they are not welded
[13:45:31] <asdfasd> where is your machine 0 for Z axis, up or down?
[13:45:34] <Loetmichel> tehy are purely from TC
[13:45:58] <Loetmichel> TekniQue was meant
[13:46:10] <TekniQue> aha
[13:46:35] <pingufan> If G38.2 would allow to move down to the mill's lowest Z-value, it would always work. But in my case I cannot enter i.e. z-60. This would actually always move down far enough, but in most cases I will get an abort because this would exceed the machine limits.
[13:46:41] <TekniQue> tungsten is awesome stuff
[13:46:50] <MattyMatt> WC₄ is harder to type
[13:46:50] <pingufan> This is what I am trying to get solved.
[13:46:55] <TekniQue> I wonder why you don't see knives with a tungsten blade
[13:47:04] <TekniQue> that'd stay sharp pretty much forever
[13:47:18] <ssi> maybe cause it's slightly radioactive? :D
[13:47:42] <pingufan> asdfasd: You see the problem?
[13:47:52] <TekniQue> ssi: maybe
[13:48:08] <ssi> at least, that's a compelling reason not to use it for food knives
[13:48:14] <MattyMatt> carbon steel can be sharpened and hardened just as well, you only need tungsten or even HSS for high temp cutting
[13:48:31] <ssi> also tungsten is quite brittle
[13:48:35] <pingufan> asdfasd: If I could use G53 g38.2 z0 f40 it would be fine. But this is not allowed.
[13:48:39] <ssi> I'm not sure how well it'dm maintain a keen edge that was doing work
[13:49:35] <asdfasd> pingufan: I still think that you have to reset previous offsets for Z, then your current coordinates for Z become same as your machine coordinates
[13:50:10] <asdfasd> then you know where are your Z limits
[13:50:32] <pingufan> asdfasd: and how do I then proceed working on the workpiece, when I loose my coordinates?
[13:50:46] -!- mhaberler has quit [Ping timeout: 260 seconds]
[13:50:52] <asdfasd> you loose only Z, then you do 38.2 to find the new Z
[13:51:03] -!- FinboySlick [[email protected]] has joined #linuxcnc
[13:51:48] <pingufan> asdfasd: And then I resto0re my coordinate system and calculate the tool length offset?
[13:51:56] <pingufan> Sorry, I am confused.
[13:52:16] <asdfasd> you can try carefully
[13:52:53] <pingufan> Sounds really complicated.
[13:53:48] -!- Tom_itx [Tom_itx!~Tl@unaffiliated/toml/x-013812] has joined #linuxcnc
[13:53:49] <asdfasd> no it is easy, G10 L2 P1 Z0 your Z axis offset is reset, you loose position but you are in machine coordinates
[13:54:05] <asdfasd> then G38.2 to find the new Z
[13:54:30] <asdfasd> X and Y remain unchanged
[13:54:53] <asdfasd> when you use G10 and G38.2 you enter only Z
[13:56:02] <pingufan> And how do I get back the coordinate system (especially Z) I touched off at the beginning of the job?
[13:57:37] <asdfasd> where you touched off ? on the sensor?
[13:59:28] <pingufan> My workflow is: 1. Home the mill 2. Touch off x/y/z axis with first tool mounted. 3. Start my job.
[13:59:57] <pingufan> Now the job shall first move to the tool sensor and touch it. So I know the current tool length.
[14:00:35] <pingufan> Then - after every tool change - I probe again the tool and calculate an offset (stored into #5063).
[14:00:43] -!- alex_joni has quit [Ping timeout: 264 seconds]
[14:02:34] <asdfasd> Im not sure how to store current position
[14:03:30] <asdfasd> once the sensor is touched I think new values are entered for curent position
[14:03:48] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:04:07] <JT-Shop> you get the current position from the variable it is stored in
[14:04:27] <pingufan> I am wondering why this G38.2 does not allow Machine coordinates. That would be everything I need. But I have another idea: Can I calculate the difference between "Machine Z0" and "User z0" ?
[14:04:43] <pingufan> I am wondering why this G38.2 does not allow Machine coordinates. That would be everything I need. But I have another idea: Can I calculate the difference between "Machine Z0" and "User z0" (G54 coordinate system)
[14:05:17] <pingufan> Then the subroutine could calculate the Z-value to be used by itself.
[14:06:04] <pingufan> That information must be kept in some register?
[14:06:22] <asdfasd> so you want to measure the tool ofset before changing?
[14:06:32] <asdfasd> and then make corrections after changing?
[14:07:30] <pingufan> I measure it once as reference, and then again (after every tool change) to calculate the difference.
[14:07:57] <pingufan> The difference is stored in G43.1 z[...] .
[14:08:06] <asdfasd> now I see...
[14:08:16] <pingufan> :)
[14:08:34] -!- logger[mah] has quit [Remote host closed the connection]
[14:09:38] <asdfasd> before toolchange you have to store curent offset
[14:09:53] <asdfasd> then you do touch off and store another offset
[14:10:22] <asdfasd> this difference between them is the level between the workpiece and the sensor
[14:10:24] <pingufan> Precisely. I start with g49 (tool length comp. = off). So - finally - I have two external subroutines, and I simply call them from my
[14:11:03] <pingufan> The difference is the difference in tool lengths. I pastebin you the codes - moment.
[14:11:14] <asdfasd> after toolchange you touch the sensor and add the difference
[14:12:54] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:13:08] <pingufan> http://pastebin.com/pfUTTWiJ
[14:13:58] <pingufan> The first sub is called immediately at job start. it measures the currrent tool and this is the reference length.
[14:14:13] -!- logger[mah] has quit [Remote host closed the connection]
[14:14:20] <pingufan> The second one calculates the offset to the first tool.
[14:18:00] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:19:04] <asdfasd> on the first routine you have to store one more - just before you touch the sensor
[14:19:36] <asdfasd> that will give you the workpiece Z
[14:20:01] <pingufan> brb - Customer came in...
[14:28:55] <JT-Shop> you guys know where the variable descriptions are?
[14:31:20] <archivist> I would ask the man that looks after the docs :)
[14:32:08] <JT-Shop> yea me too
[14:33:26] <archivist> I just had fun with one of the straight out of school "Im not a salesman would you like to fill in a survey" type sales people, I won
[14:34:28] <JT-Shop> LOL
[14:35:37] <archivist> callers are on a loser if they knock on my door
[14:35:39] -!- psha[work] has quit [Quit: Lost terminal]
[14:37:08] <JT-Shop> I live kinda remote to they are extremely rare around here or they never make back this far dunno
[14:38:27] <archivist> we get sales droids, itinerants after scrap and god peddlers
[14:38:46] <archivist> housing estate
[14:40:40] <JT-Shop> I remember that when I lived in a more populated area... sometimes fun sometimes annoying
[14:41:54] <pingufan> asdfasd: Ok, I am back.
[14:42:00] -!- logger[mah] has quit [Remote host closed the connection]
[14:42:08] -!- mk0 has quit [Quit: LinuxCNC 2.5.0!]
[14:43:49] <asdfasd> I think you have to store 3 variables
[14:44:00] <pingufan> Which ones?
[14:44:13] <asdfasd> 2 in the first routine
[14:44:25] <asdfasd> one after workpiece touch off
[14:44:31] <asdfasd> and one after sensor touch off
[14:44:55] <asdfasd> the difference should be added later in the second routine
[14:45:26] <pingufan> Can I read in the height of my "G54-Z0" above "machine Z0"?
[14:45:47] <pingufan> Or calculate this value?
[14:46:06] <asdfasd> Im not sure you have to try
[14:46:43] <pingufan> If yes, all I have to do is to move down z[- #value] instead of z[0]. :)
[14:47:03] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:47:21] <pingufan> Machine coordinates are always positive, right?
[14:47:54] <asdfasd> depends of your machine configuration
[14:49:07] <asdfasd> mine machine Z0 is high, any lower is negative
[14:49:21] -!- logger[mah] has quit [Remote host closed the connection]
[14:49:24] <pingufan> Well, my hom switches are on left, rear and uppermost limit. So I actually do auto-homing and then correcting Y and Z.
[14:49:58] <pingufan> This results in x/y/z=0 at front-left-bottom point.
[14:50:49] <pingufan> So when I can read in one variable the position of the G54 Z0, I only have to invert it to get the lowest possible Z value.
[14:51:34] <pingufan> And as this can be done by the subroutine right before doing the G38.2, this should work.
[14:51:41] <JT-Shop> archivist: I guess they don't read back
[14:52:08] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:52:32] <archivist> JT-Shop, current Z confusion ? :)
[14:53:38] <pingufan> JT-Shop: You are welcome to bring light into the darkness. :)
[14:53:48] -!- logger[mah] has quit [Remote host closed the connection]
[14:54:02] <jdhnc> any suggestions for why an encoder would just count up when you turn it either direction?
[14:55:13] <archivist> jdhnc, missing either A or B signal
[14:56:05] <jdhnc> ahh, that makes sense.
[14:56:41] -!- logger[mah] [logger[mah][email protected]] has joined #linuxcnc
[14:57:03] <pingufan> How can I see the current value of a variable in AXIS ?
[15:01:00] <archivist> use the debug or print commands in comments
[15:01:24] <JT-Shop> pingufan: http://linuxcnc.org/docview/html/gcode/overview.html#_parameters_variables_a_id_sec_parameters_a
[15:01:39] <pingufan> Where appears the output of "(print, hello)" ?
[15:02:04] <archivist> tell us when you read the manual or tried it
[15:02:06] <pingufan> Or "(print, Hello #1000)" ?
[15:02:11] <JT-Shop> http://linuxcnc.org/docview/html/gcode/overview.html#_print_messages_a_id_sec_print_messages_a
[15:02:37] * archivist just learnt them by reading the docs
[15:03:50] -!- zlog has quit [Ping timeout: 248 seconds]
[15:04:44] -!- Tom_itx has quit [Ping timeout: 272 seconds]
[15:04:50] <pingufan> Well, this docs are possibly sufficient for an experienced user, but not really fopr a novice. Too many details are missing or not easy to be found.
[15:06:34] -!- zlog [[email protected]] has joined #linuxcnc
[15:07:05] <JT-Shop> try reading the new docs
[15:08:45] <pingufan> Those for version 2.5 ?
[15:08:49] -!- freespace has quit [Quit: leaving]
[15:10:26] -!- freespace [[email protected]] has joined #linuxcnc
[15:10:36] <pingufan> JT-Shop: Should this always move down to machine's z0 when I am in G54 coordinate system? #5223 is the Z-value of Z --> G38.2 z[-#5223] f40
[15:12:06] -!- ktchk [[email protected]] has joined #linuxcnc
[15:12:12] <pingufan> I wonder why (MSG, hghgjgh #1000) cannot print the value of #1000.
[15:13:17] -!- ktchk [[email protected]] has parted #linuxcnc
[15:15:15] <JT-Shop> http://linuxcnc.org/docview/html/gcode/overview.html#_messages_a_id_sec_messages_a
[15:15:20] <JT-Shop> yes they are 2.5 docs
[15:16:01] <JT-Shop> to have a message with a variable http://linuxcnc.org/docview/html/gcode/overview.html#_debug_messages_a_id_sec_debug_messages_a
[15:22:46] <pingufan> Thank you. (DEBUG, ...) is what I need.
[15:24:02] <pingufan> What is the correct syntax for passing the inverse value of a variable as parameter? I.e. z [- #5223], or z-[#5223] ...
[15:24:53] -!- thinpete [[email protected]] has joined #linuxcnc
[15:25:50] <pingufan> I ask, because this gives the error "Z-value exceeds positive limit of Z-Axis": g38.2 z[-#5223] f10
[15:26:28] <pingufan> But this also does not work: g38.2 z[-1 * #5223]
[15:27:49] -!- vladimirek has quit [Remote host closed the connection]
[15:32:28] -!- cray [[email protected]] has joined #linuxcnc
[15:33:00] <cray> Hello, anyone in here used an atx pc supply as a dc power supply outside of a pc or familiar with electronics?
[15:34:06] <cray> I have hooked up my power OK and power on cables as I should and the supply puts out a steady 12v across yellow cables as it should.
[15:35:03] <cray> However, as soon as a I attach anything that will draw a decent current, small 12v motor or such, the supply just trips. I can`t understand why but I`m not 100% knowledgable on the whole subject! Any thoughts?
[15:35:39] <cray> forgot to say, I have a resistor across 5v to preload the supply.
[15:38:21] <JT-Shop> cray: I've seen some info on that when searching with google a while back
[15:39:10] <JT-Shop> pingufan: I think either one works try with a normal G1 move to verify the variable part
[15:40:25] <pingufan> My machine was confused because I aborted a former job. I continue testing.
[15:42:23] -!- syyl [[email protected]] has joined #linuxcnc
[15:43:48] <cray> JT-Shop: what, a similar problem to what I`m having??
[15:44:36] <JT-Shop> just general info on how to use a PC power supply like your trying to do
[15:46:25] -!- mazafaka [mazafaka!~ig@unaffiliated/mazafaka] has joined #linuxcnc
[15:46:33] -!- mhaberler [[email protected]] has joined #linuxcnc
[15:47:20] <mazafaka> stupid day...
[15:49:20] <cray> issue is, I`ve done it 4 times before and never had an issue!
[15:51:05] <cray> http://forums.makezine.com/comments.php?DiscussionID=28 exact same issue, mine is an AT supply, not ATX
[15:52:01] <JT-Shop> I hate when that happens, could the power supply be bad?
[15:53:41] <mazafaka> power cord could be bad
[15:53:49] <JT-Shop> Jymmm: http://imagebin.org/207508
[15:54:04] <syyl> uh
[15:54:07] <syyl> is that yours?
[15:54:15] <JT-Shop> me? yes
[15:54:21] <syyl> nice!
[15:54:30] <JT-Shop> thanks
[15:55:48] <mazafaka> JT-Shop: violin on wheels? Think it's not a 'reminiscent' of old culture?
[15:56:05] <JT-Shop> http://imagebin.org/207596
[15:56:37] <JT-Shop> you don't stand on the wrong end of this puppy
[15:57:31] <syyl> *like*
[15:57:57] <mazafaka> if it's a violin, I each time stand on the wrong side until I am anyhow related to this what in fact is a huge gun! >:0
[15:59:35] <mazafaka> And besides, what if The South will have realised the threat in a proper time?
[16:05:40] <skunkworks> cray: are you just hooking it across the +12v supply? or are you also using the -12v also?
[16:11:26] <JT-Shop> Hi skunkworks
[16:15:27] <skunkworks> Hi JT-Shop
[16:15:39] <skunkworks> Nice cannon bits
[16:18:42] -!- mhaberler_ [[email protected]] has joined #linuxcnc
[16:18:42] -!- mhaberler has quit [Read error: Connection reset by peer]
[16:18:43] mhaberler_ is now known as mhaberler
[16:21:51] <pingufan> Can I read from a variable the currently set feed rate? I experience that a G38.2 command changes the feed-rate but does not restore the original one. I would like to save the current feedrate, then do the g38.2, and finally restore the saved value.
[16:23:04] <pingufan> Also I would like to check if the spindle is currently on or off and to turn it off in my code before doing the G38.2 (only to be sure not to mill away my sensor.
[16:23:05] -!- mhaberler has quit [Read error: Connection reset by peer]
[16:24:38] -!- mhaberler [[email protected]] has joined #linuxcnc
[16:28:31] -!- mhaberler has quit [Client Quit]
[16:30:43] <archivist> pingufan, you wrote the g code so you know what speed you want
[16:31:48] <pingufan> Sure. But I write now little external modules (subroutines) and I prefer to do things bullet-proof.
[16:32:23] <archivist> add values like that to the calling parameters
[16:32:50] <archivist> Im only stating the obvious :)
[16:33:10] <pingufan> :D
[16:33:36] <pingufan> So there is actually no possibility to read in such things at all?
[16:33:59] <archivist> no idea
[16:34:44] <archivist> but I know what is obvious and bullet proof
[16:35:52] -!- syyl_ [[email protected]] has joined #linuxcnc
[16:35:56] -!- zlog has quit [Read error: Connection reset by peer]
[16:38:09] <pingufan> Btw., the touchoff for sensing tool length works meanwhile.
[16:38:31] -!- syyl has quit [Ping timeout: 264 seconds]
[16:42:40] -!- zlog [[email protected]] has joined #linuxcnc
[16:45:07] -!- Nick001-Shop [[email protected]] has joined #linuxcnc
[16:48:09] -!- syyl_ws [[email protected]] has joined #linuxcnc
[16:49:56] -!- ve7it [[email protected]] has joined #linuxcnc
[16:51:40] -!- mhaberler [[email protected]] has joined #linuxcnc
[16:55:20] <Thadius> so anyone out there using vectric products...ie cut2d, cut3d, if so, any huge hurdles/problems?
[16:56:50] -!- cray has quit [Ping timeout: 272 seconds]
[16:57:29] <archivist> Im not a user but been to a talk and not heard anything bad yet
[16:59:50] <joe9> i am trying to make the taig work with minimum limit+home switches. The stepconf seemed to have done its' stuff. http://linuxcnc.org/docs/2.4/html/config_ini_homing.html is there a better tutorial than this to understand how it works?
[17:00:04] -!- Tom_itx [Tom_itx!~Tl@unaffiliated/toml/x-013812] has joined #linuxcnc
[17:01:00] <skunkworks> http://www.youtube.com/watch?v=aflrnD9OEvA
[17:02:19] -!- tom3p [[email protected]] has joined #linuxcnc
[17:02:27] * archivist restarts firef*cks again to see teh icing
[17:02:31] -!- ries [[email protected]] has parted #linuxcnc
[17:02:43] <JT-Shop> skunkworks: thanks
[17:03:23] <JT-Shop> pingufan: it is a hal pin
[17:03:28] <JT-Shop> err halui
[17:04:33] <JT-Shop> joe9: homing? or homing settings?
[17:05:11] <tom3p> whats the name of the engraving tool supplier? bits&pcs? kibbles&bits? bit-o-honey???
[17:05:46] <pingufan> How do I read in the status of a HAL pin from a G-code program and process it in my code ?
[17:06:38] <JT-Shop> tom3p: onsrud
[17:06:42] <joe9> JT-Shop: i have switches at the end of the axes and plan on using them as "minimum limit + home" switches.
[17:06:43] <tom3p> thx
[17:06:55] <joe9> I selected that option in the stepconf wiz.
[17:07:06] <tom3p> ( i thought it had 'bits' in the name ;)
[17:07:45] <joe9> and, when I start linuxcnc, as my axes are not homed, I am trying to home it for the first time.
[17:07:55] <JT-Shop> pingufan: a good question that I do not know the answer to
[17:08:13] <JT-Shop> joe9: and?
[17:08:49] <joe9> and, when I do "Home All", it seems to go the wrong direction in Z axis.
[17:08:53] -!- sumpfralle [[email protected]] has joined #linuxcnc
[17:09:15] <pingufan> Then you have to enter a negative speed.
[17:09:20] <joe9> JT-Shop: i think I should keep experimenting and try to figure it out.
[17:09:28] <joe9> pingufan: oh, ok. that could be it.
[17:09:30] <joe9> thanks.
[17:09:32] <JT-Shop> reverse the sign of the velocity
[17:09:39] <joe9> I selected "Same"
[17:09:43] <pingufan> I have to do exactly the same.
[17:09:46] <JT-Shop> joe9: there you go
[17:10:11] <joe9> my stepconf: http://codepad.org/vM7j9158
[17:10:21] <JT-Shop> same is home up to switch, down off of switch, back up at slower speed to switch
[17:10:52] <archivist> docs system needs fixing so links dont die http://www.linuxcnc.org/docview/html/gcode_main.html#sec:M66-Input-Control
[17:11:14] <JT-Shop> archivist: old links?
[17:11:20] <joe9> jt-shop, pingufan, thanks. I will figure it out from here.
[17:11:32] -!- ktchk [[email protected]] has joined #linuxcnc
[17:11:38] <archivist> JT-Shop, that was a link from http://www.linuxcnc.org/index.php/english/component/kunena/?func=view&catid=20&id=14735
[17:11:55] <archivist> so not an old link as far as I can see
[17:12:11] <JT-Shop> joe9: if your using combined home a limit switches a tip from me is to make the home final position not right on the switch
[17:12:19] <archivist> a mere 5 months
[17:12:36] <joe9> pingufan: can you please paste your stepconf file?
[17:12:46] <JT-Shop> archivist: I'm afraid all the links to 2.4 docs will be broken now
[17:13:03] <archivist> JT-Shop, that is just not right
[17:13:19] <joe9> JT-Shop: this is my X settings: http://codepad.org/TRbD0HiR
[17:13:28] <JT-Shop> I have no control over that
[17:13:44] <archivist> note that the link had no version in the link
[17:13:56] <JT-Shop> joe9: I can't read stepconf generated code sorry
[17:14:21] <JT-Shop> archivist: right the default docs are now 2.5 not 2.4
[17:14:43] <archivist> should make no difference
[17:14:50] <joe9> http://codepad.org/vV3zNTkp JT-Shop this is the X-axis settings in the .ini file
[17:15:32] <joe9> all axes from .ini files: http://codepad.org/6kFXDRaz
[17:15:55] <JT-Shop> 2.5 docs are totally different in structure, 2.4 used lyx and 2.5 uses asciidoc as well as a lot of restructuring to try and make them easier to navigate
[17:16:01] <joe9> pingufan: would you mind pasting your .ini file too, if you do not mind?
[17:16:26] <JT-Shop> joe9: I can read that... what was the question sorry I forgot
[17:16:34] <archivist> pingufan, http://www.linuxcnc.org/docview/html/gcode/m-code.html#_m66_wait_on_input_a_id_sec_m66_input_control_a
[17:16:38] -!- Thadius has quit []
[17:17:38] <joe9> JT-Shop: the "home velocity being not negative" and the "home position not to be on the switch". wondering if those settings make sense to you?
[17:17:46] <joe9> do they appear as correct settings?
[17:17:50] -!- Thadius [[email protected]] has joined #linuxcnc
[17:18:00] <joe9> when I have the minimum limit + Home switch at the bottom of the axis.
[17:18:11] <archivist> JT-Shop, who can read the db and do a search replace to point at 2.4 docs
[17:19:00] <JT-Shop> maybe alex can, I can't get to the db
[17:19:05] Thadius is now known as TJB
[17:19:36] TJB is now known as Guest66227
[17:19:52] Guest66227 is now known as Thadius
[17:20:01] <JT-Shop> joe9: I would have a little bit of home offset if sharing home and limit switches
[17:20:08] -!- ktchk [[email protected]] has parted #linuxcnc
[17:20:08] <joe9> ok, thanks.
[17:20:37] <JT-Shop> search and latch are not usually the same value, normally you latch at a much lower speed for accuracy
[17:20:41] <joe9> JT-Shop: is there a better explanation of homing than here: http://linuxcnc.org/docs/2.4/html/config_ini_homing.html
[17:21:09] <JT-Shop> for stepconf?
[17:21:09] Thadius is now known as ThadiusB
[17:21:19] <joe9> i can do the configuration by hand.
[17:21:30] <joe9> do not really care for stepconf at this point.
[17:21:38] <JT-Shop> http://linuxcnc.org/docview/html/config/ini_config.html
[17:21:47] <joe9> the settings I have were generated by stepconf
[17:21:50] <JT-Shop> and http://linuxcnc.org/docview/html/config/ini_homing.html
[17:22:13] <JT-Shop> http://linuxcnc.org/docview/html/config/ini_config.html#_axis_lt_num_gt_section_a_id_sub_axis_section_a
[17:22:34] <joe9> JT-Shop: thanks.
[17:22:39] <JT-Shop> joe9: once you get your initial setup stepconf kinda gets in your way
[17:22:49] -!- psha [[email protected]] has joined #linuxcnc
[17:22:54] <joe9> oh, ok. thanks for that tip.
[17:23:06] <joe9> i will do the changes manually.
[17:23:16] -!- maximilian_h [[email protected]] has parted #linuxcnc
[17:24:38] <JT-Shop> check out the homing sequence chart in the homing chapter
[17:24:49] <JT-Shop> the first two pertain to stepper systems
[17:28:39] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[17:29:36] -!- Thetawaves [[email protected]] has joined #linuxcnc
[17:36:10] <Jymmm> JT-Shop: DIY Cannon Kit, available at a walmart near you
[17:36:13] <JT-Shop> archivist: I don't think a global search and replace would work as many of my posts reference the 2.5 docs but I might be wrong
[17:36:17] <JT-Shop> lol
[17:37:00] * JT-Shop heads to the Mennonite store for a loaf of Amish bread
[17:37:31] <Jymmm> JT-Shop: Grab some Amish Laptops too
[17:42:16] <pingufan> Do I have to switch toollength compensation off before I do the next sensing? (At the second and up tool change?) I wonder why my mill does not drill a hole, only a slight pocket.
[17:45:39] <archivist> JT-Shop, needs scripting to see if link exists
[17:48:44] <Dave911> Amish laptops.... Ha ha.. They have them, but they won't tell you.... My Amish next door neighbor also doesn't have air conditioning but the generator runs constantly, even at night, when it is really hot... go figure..
[17:48:46] <Dave911> My other Amish neighbor used to keep a TV in the garage... somehow that was ok as the evil device was not in the house.
[17:50:27] <Dave911> I make some nice Amish bread even though I am not Amish or Mennonite ;-) Breaking all of the rules - one at a time. :-)
[17:50:41] <Jymmm> Heh. I just figured an Amish Laptop has to be something like an abacus so hence the plural.
[17:52:04] <Tom_itx> Dave911 maybe they believe you should keep your friends close and your enemies closer
[17:53:11] <Dave911> Ha ha... I live in a very large Old Order Amish area and my next door neighbor (Amish) has a smartphone. He also has a wired phone in his office along with a fax machine.. Go figure... But he does look Amish and he has a beard and horses, buggies etc.
[17:54:35] <Jymmm> Dave911: Maybe they've seen communications as a good thing.
[17:54:45] <Dave911> I'm sure that my Amish neighbors are convinced that I am going straight to hell....
[17:55:33] <Dave911> It is really a requirement if you want to run a business, or be employed... My neighbor has a shoe store and a drywall business.
[17:55:55] <Jymmm> Dave911: you should ask him about the smartphone/fax and the "rules" these days.
[17:56:23] <Jymmm> does he pull the drywall with a horse buggy?
[17:56:46] <Dave911> Oh I could tell you some stories... none of it really makes any sense. If you try and apply any logic to any of it, it is just a waste of time.
[17:57:00] <Jymmm> hang with cordless drill?
[17:57:54] <Dave911> drywall, horse and buggy.. Heck no.. delivered via truck and forklift .... cordless drills -- of course... generators, power drills, power mixers etc.
[17:58:19] <Dave911> Power sanders.. etc
[17:58:34] <Jymmm> I remember hearing once that technology is ok if it's tool related.
[17:59:01] <Tom_itx> define tool
[17:59:14] <Jymmm> Tom_itx: <-------- Tool
[17:59:22] <Dave911> Truth be told... their ancestors would roll over in their graves if they saw how they lived these days
[17:59:54] <Dave911> Some of the kids have computerized handheld games.. that is ok as it is not wired to the power grid
[18:00:09] <Jymmm> Dave911: lol
[18:00:23] <Jymmm> Progressive Amish!
[18:00:23] <archivist> whats evil about the grid
[18:00:36] <Jymmm> archivist: "Da Man"
[18:01:02] -!- JT-Shop has quit [Read error: Connection reset by peer]
[18:01:02] -!- jthornton has quit [Read error: Connection reset by peer]
[18:01:04] <Dave911> Depending on outside stuff like the grid is considered enslavement (sort of..)
[18:01:27] -!- jthornton [[email protected]] has joined #linuxcnc
[18:01:29] -!- JT-Shop [[email protected]] has joined #linuxcnc
[18:01:45] <Dave911> Don't try and read any logic into any of this unless you want to waste a lot of time and get a headache ;-)
[18:02:19] <Jymmm> Heh, In progress of being approved by the FCC, all the wireless carriers are building a database of stolen phones/tablets, so they can't be activated if on the list.
[18:02:34] <jdhnc> I put a scope on this flakey encoder, A & B are in perfect sync
[18:03:06] <archivist> jdhnc, should be 90degrees out of phase
[18:03:11] <Dave911> Some of them have given up trying to be Amish and have joined more moderate churches. When they do, cars show up, power lines are hung, etc..
[18:03:23] <skunkworks> jdhnc: yah - what do you mean by perfect sync..?
[18:03:50] <jdhnc> archivist: the other (good) one is 90 degrees out. This one lines up perfectly
[18:03:55] <Dave911> Oh well.. gotta get back to work...
[18:04:19] <jdhnc> skunk: I mean they rise fall at exactly the same time.
[18:04:29] <archivist> jdhnc, the a shorted to b?
[18:05:29] <jdhnc> guess I could buzz it.
[18:05:36] -!- phantoxe has quit []
[18:05:52] <Jymmm> What causes 18/8 Stainless Steel to rust? http://i43.tinypic.com/2qxb4g8.jpg
[18:06:31] <DJ9DJ> hmm, some kind of acid?
[18:06:44] <Jymmm> Just tap water
[18:06:55] <Jymmm> it's a SS water bottle
[18:06:58] <Tom_itx> i wouldn't drink it
[18:07:20] <Jymmm> The mfg is replacing them.
[18:07:24] <Jymmm> mfr
[18:07:40] <skunkworks> some stainless is better than others...
[18:08:04] <Jymmm> skunkworks: I have two of them that are rusting
[18:08:40] <Jymmm> suppose to be "food grade" SS.
[18:10:53] <Jymmm> 1200ml (41oz) SS Water Bottle, nice capacity, cheap at Walgreens, scary as hell now
[18:11:15] <Jymmm> subzero brand
[18:11:43] -!- sumpfralle has quit [Ping timeout: 245 seconds]
[18:11:49] <Tom_itx> made in china?
[18:11:58] <Jymmm> yeah
[18:12:16] <Jymmm> everything is anymore
[18:13:53] <Jymmm> Kleen Kanteen, Nalgene
[18:15:08] <Jymmm> Though kleen kanteen makes a big ass 64oz ss bottle
[18:15:56] <Jymmm> Any of you pyros?
[18:17:44] -!- kb8wmc [[email protected]] has joined #linuxcnc
[18:18:47] -!- psha has quit [Quit: fontconfig]
[18:18:58] -!- psha [[email protected]] has joined #linuxcnc
[18:19:40] <Jymmm> I'm trying to find a (non explosive) material that will ignite from a tiny ember
[18:20:03] <archivist> matches
[18:20:29] <Jymmm> anything else?
[18:20:56] <archivist> spare matches
[18:21:54] <Jymmm> heh
[18:22:12] <archivist> this is a cnc channel!
[18:23:25] <Jymmm> True, but you guys have a far more reasonable usage of such things, not some anarchist cookbook sorta methodology.
[18:24:28] <skunkworks> black powder?
[18:24:29] <cradek> have you tried a match?
[18:25:04] <Jymmm> cradek: Yeah, I'm trying to find somethign that doesn't require a HazMat cert to transport.
[18:26:10] <Jymmm> like super fluffy cotton fibers or some such thing.
[18:31:55] -!- mhaberler has quit [Quit: mhaberler]
[18:31:57] <skunkworks> cap paper for a cap gun?
[18:32:07] <Jymmm> flash paper?
[18:32:16] <tom3p> 'charcloth' ? http://www.grannysstore.com/Wilderness_Survival/fire_pistons.htm
[18:32:45] <Jymmm> tom3p: fwiw, charCORD is better
[18:32:54] <tom3p> but i really want the hal widget to connect this http://www.emotiv.com/
[18:33:28] <Jymmm> tom3p: not with the base os being ubuntu you don't!
[18:34:10] <tom3p> The Vaseline is used to lubricate the plunger O-ring. It can also applied to cotton balls, or other organic material and makes a great fire-starter
[18:34:25] <tom3p> bbl
[18:34:37] <Jymmm> tom3p: Yes, but still won't ignite from an ember.
[18:34:53] <Jymmm> no matter how hard I've tried
[18:42:09] -!- syyl_ws has quit [Quit: Verlassend]
[18:58:12] -!- pingufan has quit [Quit: Konversation terminated!]
[19:01:04] <Dave911> Jymmm: Propane..
[19:01:07] <asdfasd> is there any chance using hal to write in to parameters like #1000
[19:02:19] <cradek> asdfasd: see http://linuxcnc.org/docview/html/gcode/m-code.html#sec:M66-Input-Control
[19:03:50] <jdhnc> encoder doesn't seem to be shorted. A- and B- are the same, but they are on the good one also.
[19:03:58] <Jymmm> Dave911: gun cotton
[19:04:24] <adb> MatJymmm> material prepared from this fungus and used lighter with flint and flint. >>Polyporus fomentarius, polyporus ignarius , Fomes igniarius
[19:04:43] <Dave911> gun cotton - is that cotton soaked with something??
[19:05:24] <Jymmm> Dave911: No, it's an impact ignited explosive
[19:06:08] <Jymmm> created by accident by farmers and their fertilizer I think
[19:07:12] <Loetmichel> soo, y geardon fits. Now i have top make a Stepper mount and then work on the x axis... -> http://www.cyrom.org/palbum/main.php?g2_itemId=13068
[19:07:19] <Loetmichel> -p
[19:07:29] <Loetmichel> geardown
[19:10:41] <asdfasd> cradec: I found passing parameters to hal, not oposite
[19:11:02] <asdfasd> I need to write in to #1000 for example
[19:11:16] <Jymmm> adb: I can get losts of thing to ignite from a spark ( ~2000F), but not from an ember ( ~200F).
[19:11:21] <Jymmm> lots
[19:12:42] <Jymmm> nitrocellulose would work (autoignition temp is ~80F), but in the explosives realm and handling.
[19:13:45] <Jymmm> I think I just need to hit up the local arson investigator for ideas. They have the coolest pyro stuff!
[19:15:31] <cradek> asdfasd: that link is the correct one - please read the M66 docs again
[19:15:49] <cradek> asdfasd: it's the way to get a value from hal into gcode
[19:17:15] <Loetmichel> i've seen a projector room in a cinema where a 2hrs film was accidentially copied on Zelluloid...
[19:17:19] <Loetmichel> NOT pretty
[19:18:07] -!- iwoj has quit [Quit: Textual IRC Client: http://www.textualapp.com/]
[19:19:38] <JT-Shop> cradek: is mode 0 the only one that stores the input value to 5399?
[19:21:21] <cradek> I don't know without looking, but it would make sense if so; in the other cases you know the value already: it's the one you were waiting on.
[19:21:40] <JT-Shop> yea, that makes sense
[19:22:06] <asdfasd> cradek: Im sorry but this is only when program is running, cannot work before that
[19:22:20] -!- phantoxe [[email protected]] has joined #linuxcnc
[19:22:49] <cradek> ok you didn't say you wanted it to happen before a program runs. there is no way to do that. you must use mdi or a running program to read the input.
[19:24:03] <cradek> maybe if you'd explain what you are trying to do?
[19:24:13] <asdfasd> for example the defined function touch_off() in axis
[19:24:21] <asdfasd> you press button touch off
[19:24:37] <asdfasd> then a value is passed like MDI command
[19:24:46] <asdfasd> I think it was G92...
[19:25:17] <asdfasd> can I do same in pyvcp panel
[19:25:40] -!- micges [[email protected]] has joined #linuxcnc
[19:25:50] <cradek> touch off uses G10
[19:25:57] <asdfasd> pressing a button to sent a spin value like MDI
[19:26:02] -!- micges [[email protected]] has parted #linuxcnc
[19:26:03] <Loetmichel> haha, noone killing me for using polycarbonate on gears ?
[19:26:12] -!- iwoj [[email protected]] has joined #linuxcnc
[19:26:12] <cradek> halui does have an mdi interface
[19:26:22] <Loetmichel> i expected some rants ;)
[19:26:51] <asdfasd> I read about halui bit it only send pre-written G code
[19:27:04] <asdfasd> I dont know how ot pass the spin value
[19:27:13] <cradek> what spin value?
[19:27:27] <cradek> say what you're trying to do, the whole thing
[19:27:52] <asdfasd> spinbox
[19:28:30] <asdfasd> ok, I want to make a spin box, and button next to it
[19:29:05] <asdfasd> pressing the button to do for example G10 L2 P1 [spin box value]
[19:29:35] <asdfasd> G10 L2 P1 X[spin box value]
[19:30:05] <cradek> I don't know how you'd do that, sorry
[19:30:15] <archivist> asdfasd, andypugh did something like that for http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Hobbing
[19:30:36] <cradek> do you know that touch off has a keyboard shortcut (End)?
[19:30:54] <asdfasd> Im sorry that Im so novice with so strange questions
[19:31:27] <asdfasd> yes I know but it is working with active axis , the last moved
[19:32:04] <cradek> ah yes, nice feature :-)
[19:32:22] <asdfasd> nice... every time set wrong axis :D
[19:33:03] <asdfasd> I edited axis and axis.tlc, and I added touch off Z button
[19:33:11] <asdfasd> but I will loose that feature after update
[19:33:15] <asdfasd> or reinstal
[19:33:46] <asdfasd> and I thought I can make it in custom panel which will be used with every next version
[19:34:16] <skunkworks> can mdi commands in halui run a ngc file?
[19:34:41] <cradek> skunkworks: mayybe
[19:34:41] <skunkworks> probably no.
[19:34:48] <cradek> dunno
[19:34:57] <cradek> with O...call maybe
[19:35:27] <cradek> but spinboxes are just plain terrible ways to enter numbers
[19:35:38] <cradek> I can't imagine wanting to use one for touch off
[19:35:52] <cradek> sorry, neither here nor there
[19:36:08] <skunkworks> but if it could - I could see you doing what you want.. (2 lines - one that does the m66 and a line that does the g10...)
[19:36:12] <skunkworks> otherwise....
[19:36:20] <cradek> modifying AXIS seems like what you want
[19:36:30] <cradek> perhaps you will end up with a customized build
[19:36:45] <skunkworks> You could setup the 2 mdi lines in halui - but than when the button is pressed - some hal or ladder would run one - then the next..
[19:36:54] <skunkworks> but yes - seems like a bit of work.
[19:37:33] <asdfasd> cradek: actually Im not sure what spinbox do exactly, I just need to enter a value with keyboard, just like you can do with the touch off button in axis
[19:38:00] <cradek> spinbox has tiny increase and decrease buttons so you can very tediously make it show the number you want
[19:38:13] <cradek> horrible
[19:39:05] <asdfasd> then not suitable...
[19:39:56] <asdfasd> can I call same function touch_off() with button in pyvcp panel?
[19:40:18] <asdfasd> there is a lot functions there, like set_active_axis()
[19:40:21] -!- stevegt_ has quit [Ping timeout: 252 seconds]
[19:40:26] <asdfasd> or something similar
[19:40:40] <cradek> I don't think so but I admit I'm not sure
[19:40:43] <Spida> how hard/expensive is building/buying an automatic toolchanger for a diy cnc mill/router?
[19:40:50] <asdfasd> some of them available in halui
[19:41:06] <Loetmichel> Spida: the changer is not the problem
[19:41:07] <cradek> could you maybe put tcl code in .axisrc that adds your button?
[19:41:36] <Loetmichel> the mill spindle with pneumatic tool release is the (costly) problem
[19:42:08] <asdfasd> I did it in the worst way possible, and I thought how to make it easy to apply it after reinstall
[19:42:14] -!- psha has quit [Quit: Lost terminal]
[19:43:53] <asdfasd> I didnt look into axisrc, I have to take a look
[19:47:01] <asdfasd> another usefull option which I like very much is a shortcut key, like shift
[19:47:15] <asdfasd> shift + arrow jog fast
[19:47:40] <asdfasd> ctrl + arrow jog single step, or pre defined steps
[19:47:45] <cradek> it already has that
[19:47:52] <cradek> check the help screen
[19:47:59] <asdfasd> only shift
[19:48:07] <asdfasd> ctrl is not doing that
[19:48:28] <cradek> right
[19:49:56] <cradek> doing incremental with control would be harder I think, since the increment isn't even currently being shown when in continuous mode
[19:50:33] <asdfasd> what did in mach3 before is to set slow jog speed, then press shift and move fast to the workpiece, then only arrows to jog slowly, and then hold control for single steps for precise, it is very quick
[19:51:04] <Thetawaves> are screw based tool changers any easier?
[19:52:11] <cradek> asdfasd: I always use a wheel so I don't need this, but if you can come up with a clean way to do it and submit a git patch I'd be happy to review
[19:53:18] * skunkworks seconds the jog wheel...
[19:53:34] <cradek> AXIS does seem to know the active increment, since it keeps it when switching with c/i/c/i
[19:53:45] <skunkworks> screw based tool changers?
[19:53:49] <cradek> so if you could make it show that active increment while control is held down...
[19:54:17] <asdfasd> wow :D just to say again Im novice, but I will be happy if I can make it one day... then I will contact you about that
[19:54:20] <cradek> Thetawaves: tool holders and a spindle/drawbar are the hard part. presenting the right one to the spindle can be done all sorts of DIY ways.
[19:57:48] <Thetawaves> sorry i meant... instead of pneumatic draw bar, a screw based one driven by a gearhead motor
[19:59:07] <Loetmichel> Thetawaves: woll not work
[19:59:09] <Loetmichel> will
[19:59:34] <skunkworks> http://youtu.be/KplU8hkI0AQ
[20:00:54] <cradek> some folks use air impact wrenches on bridgeport type screw drawbars - seems slightly crazy to me
[20:00:59] <Thetawaves> i believe you but why
[20:01:03] <cradek> that's generally for manual changes though
[20:01:09] <Thetawaves> because the threads may jam?
[20:03:02] -!- iwoj has quit [Remote host closed the connection]
[20:03:26] <Thetawaves> what is the basis for your aversion to this
[20:03:41] <skunkworks> http://www.youtube.com/watch?v=a5mZa9JklH0
[20:03:46] <Spida> Loetmichel: does it have to be pneumatic? I don't want to run a compressor, especially not only for toolchanging
[20:04:11] -!- maximilian_h [[email protected]] has joined #linuxcnc
[20:04:18] -!- maximilian_h has quit [Client Quit]
[20:06:28] <asdfasd> cradek: what you think about that idea: continuous removed from the drop down menu, only increments, pressing arrows move continuous, ctrl+arrowns move with the selected increment
[20:06:51] <Loetmichel> Spida: it would be the simpelest
[20:07:02] <Loetmichel> and you dont need much volume
[20:07:18] <cradek> asdfasd: I wouldn't want to force that kind of change on everyone for this feature - the tradeoff is not good enough.
[20:07:22] <Loetmichel> so a simple Refridgerator-"egg" is sufficient
[20:07:32] <Loetmichel> with a small buffer tank behind it
[20:07:45] <Loetmichel> something of an airbrush compressor
[20:07:47] <Thetawaves> skunkworks, that looks like a massive machine
[20:07:57] <cradek> to get this merged in, you'd have to do it in a way that does not break everyone's habits
[20:08:20] <cradek> I think "holding control changes the menu to show the increment" is the only way to accomplish that
[20:08:38] <cradek> (but I don't know if that's easy or hard to actually do)
[20:08:40] <Loetmichel> cradek: ha have milled with PCNC-DOS for a long time
[20:08:43] <skunkworks> Thetawaves: yes - nice large HMC
[20:08:55] <asdfasd> which increment to show?
[20:09:14] <Loetmichel> i STILL catch mysel pressing the space bar( was used to cycle throug hte increments in PCNC)
[20:09:25] <cradek> the one that's currently/last selected. change increments and then hit c/i/c/i to see what I mean
[20:09:29] <cradek> brb
[20:10:50] <skunkworks> Thetawaves: from the 1960's
[20:11:31] -!- iwoj [[email protected]] has joined #linuxcnc
[20:11:38] <asdfasd> I feel stupid.... what is c/i/c/i ?
[20:11:56] <ssi> continuous/incremental I'd guess
[20:12:40] <asdfasd> I thought is that but how to hit it...
[20:12:57] <ssi> type c for continuous and i for incremental?
[20:14:40] <asdfasd> hm...
[20:15:29] <asdfasd> it looks like I missed that
[20:22:57] <Spida> Loetmichel: do you have any idea how much such a spindle might cost?
[20:23:13] <Loetmichel> Spida: depends
[20:23:37] <Loetmichel> you can get a kress with a changer in front from usovo
[20:23:46] <Loetmichel> that is "cheam"
[20:23:49] <Loetmichel> "cheap"
[20:24:17] <Loetmichel> but the better ones are even from china abut 3k$ with VFD
[20:25:15] <Spida> ouch
[20:25:32] <archivist> there have been some home made spindles
[20:25:36] <Spida> how crappy are the chinese ones?
[20:25:38] <tom3p> see atc spindles at http://www.kelinginc.com/KL-2200ATC.pdf
[20:28:20] -!- richardraseley [richardraseley!cead2596@gateway/web/freenode/ip.206.173.37.150] has joined #linuxcnc
[20:28:35] -!- richardraseley [richardraseley!cead2596@gateway/web/freenode/ip.206.173.37.150] has parted #linuxcnc
[20:30:25] <joe9> i am trying to understand what an "index pulse" is in the homing sequence?
[20:32:33] <joe9> is it just for servos?
[20:33:50] -!- iwoj has quit [Quit: Textual IRC Client: http://www.textualapp.com/]
[20:34:16] <jdhnc> it comes from an encoder
[20:34:20] <cradek> yes encoders have a special index output that's exactly once per revolution. you can use it to home to exactly the same encoder count every time
[20:35:00] <cradek> stepper or open loop servo machines can only home to the repeatability of a switch on the slideway
[20:35:04] <joe9> ok, thanks. I have a taig with stepper motors. so, no indexer for me.
[20:35:12] <cradek> right no index pulse with steppers
[20:35:25] <jdhnc> unless you have an encoder on your stepper
[20:36:06] <jdhnc> we seem to have quite a few stepper/encoder systems here.
[20:36:10] <joe9> I have a home+minimum limit switch. and I am trying to figure out what is a good value for HOME_OFFSET and HOME values?
[20:36:15] <joe9> any suggestions, please?
[20:36:23] <joe9> http://linuxcnc.org/docview/html/config/ini_homing.html
[20:36:27] <joe9> is what I am following.
[20:36:44] <joe9> the material uses a home offset of 3.0000 and home of 1.0000
[20:36:56] <joe9> not sure if that is a good idea for me too.
[20:37:12] <cradek> you'll be using one of the top two figures in the picture
[20:37:16] <djdelorie> In your case, you have two limits, which tend to be 0 and some other number, in inches or mm. On mine, for example, I use 0 and 30. The HOME switch is at one end, so it's "30"
[20:37:35] <djdelorie> but I have the home "location" at 29 so the axis backs off a bit after it finds the switch
[20:37:45] <joe9> cradek: yes, I am using the first picture there.
[20:37:54] <joe9> sorry, the second picture
[20:37:56] <cradek> yes if it's also a limit you have to be off the switch when done homing
[20:38:04] <joe9> both searchlevel and latchlevel positive.
[20:38:21] <cradek> you can have your unoffset origin wherever you want it, it doesn't matter
[20:38:22] <djdelorie> I could have chosen the other switch for homing, by putting in a negative homing speed, setting HOME to 0 and the home "location" to, say, 1
[20:38:29] <anonimasu> tom3p: are you there?
[20:38:50] <djdelorie> for Z I ended up using -4..0, homing to 0, so that the machine origin is "up" to avoid crashes during tool changes
[20:39:15] <joe9> djdelorie: ok, thanks. I was going for home on x = y = z = 0, z is -ve down the axis.
[20:39:21] <cradek> yes 0 is usually top of travel in Z
[20:39:24] <joe9> and the others are positive on movement.
[20:39:54] <djdelorie> if you have home+limits like you and I do, you can't have the home location (where the axis moves to *after* homing) also be zero. It needs to be somewhere off the limit switches
[20:40:05] <joe9> djdelorie: the "30". is it in inches?
[20:40:10] <djdelorie> in my case, yes.
[20:40:24] <djdelorie> you should measure the actual travel of your axes between switches, and use those values
[20:40:26] <cradek> be aware your final location also has to be somewhere INSIDE your soft limits
[20:40:28] <joe9> why are you leaving out 1 inch at the home?
[20:40:37] <joe9> isn't that a waste of your table space?
[20:40:53] <djdelorie> why not? The axes have to be *somewhere* between the switches, they just can't be *at* the switches after homing
[20:41:09] <djdelorie> otherwise, the machine goes OH NO YOU'RE AT THE LIMIT! WE'RE GOING TO CRASH! ESTOP!
[20:41:21] -!- micges has quit [Quit: Leaving]
[20:42:05] <joe9> djdelorie: why not something like 0.5 inches for home?
[20:42:13] <joe9> then all your gcode dimensions can be +ve.
[20:42:47] <joe9> that does not leave it enough time to stop?
[20:42:47] -!- stevegt_1 [[email protected]] has joined #linuxcnc
[20:43:21] <joe9> djdelorie: my machine specs for table travel are 12 inches, 5.5 and 6 inches.
[20:44:07] <joe9> and I am thinking about a home of 1cm (as i use mm for linear_units)
[20:44:36] <joe9> would it be too close to the end? not giving it enough time to stop?
[20:44:42] <joe9> should I have it more than 1cm?
[20:46:51] <joe9> djdelorie: what is your home_offset and home values? or, could you please paste your .ini file?
[20:47:41] <djdelorie> if I home to 0, the carriage is in my way. Mine homes to the back of the machine
[20:48:02] <djdelorie> you only have to be far enough away from the switch that you know the switch is not active. Could be a few mm
[20:48:10] <cradek> yep a lathe really ought to home X outward (+)
[20:48:49] <djdelorie> cradek; 0,0 is next to my chair, towards what I think of as the "front" of the machine. Homing moves the carriage away from me
[20:49:23] <djdelorie> joe9: if linuxcnc knows where the switches are, and knows where the axes are (which is the whole *point* of homing), it can avoid hitting the limits anyway.
[20:49:28] <joe9> home_offset = where the home/limit switch trips correct?
[20:50:00] <joe9> and, this can be calculated from your switch properties, correct/
[20:50:04] <joe9> s,/,?,
[20:50:12] <djdelorie> HOME_OFFSET is the location of the switch. HOME is where the axis goes after homing
[20:50:21] <joe9> ok, thanks.
[20:50:45] <tom3p> anonimasu, pm
[20:50:47] <cradek> the homing documentation is pretty good...
[20:51:13] <djdelorie> one switch is at MIN_LIMIT. The other is at MAX_LIMIT. In your configuration, HOME_OFFSET is the same as one of those two values, but if you had a separate home sensor, it would be different.
[20:51:41] <djdelorie> HOME must be between MIN_LIMIT and MAX_LIMIT. The values for MIN and MAX are actual physical dimensions (get your ruler and measure them :)
[20:52:10] <joe9> djdelorie: got it. thanks.
[20:52:21] <joe9> and do you have volatile_home = 1?
[20:52:42] <djdelorie> that's not even mentioned in my ini
[20:53:14] <joe9> neither in mine, but, it seems to be useful from the docs uri: http://linuxcnc.org/docview/html/config/ini_homing.html
[20:53:24] <joe9> but, you are on servos, might not matter.
[20:53:42] <djdelorie> you'll probably need HOME_IGNORE_LIMITS whenever one switch is serving both purposes, else as soon as you hit "home" the limit fault happens too :-)
[20:54:20] <djdelorie> yeah, mine isn't volatile. It tracks position even if you unplug the motor power supply.
[20:54:28] <cradek> djdelorie does not have [what linuxcnc thinks of as] servos
[20:54:39] <djdelorie> in fact, that's what my "machine idle" plugin does, kills motor power.
[20:54:49] <djdelorie> I have servo *motors* but not a servo *configuration*
[20:54:56] <djdelorie> linuxcnc thinks they're steppers
[20:55:14] <cradek> yes that's what I intended to say
[20:55:36] <djdelorie> but, I *do* have an encoder, so my *controller* can at least keep track of the motor position, and fix it when the machine starts moving again.
[20:57:11] <djdelorie> if tuning it weren't so insanely hard, they'd be wonderful.
[20:57:11] <cradek> so when you power them back on, they jump to where they were when they were shut off?
[20:57:11] <skunkworks> but djdelorie is going to add analog input to his drives in the future... :)
[20:57:11] <djdelorie> yes, although "jump" is overkill. They don't really go anywhere when you kill power
[20:57:11] <joe9> in the ini file, is the TRAJ section a common section for all AXES_x sections?
[20:57:11] <joe9> or, is there no such relation.
[20:57:11] <djdelorie> skunkworks: only if I have some other compelling reason to respin the control boards
[20:57:11] <skunkworks> :)
[20:57:11] <cradek> some machines sure move when you turn them off
[20:57:46] <cradek> with a normal linuxcnc servo setup, the commanded tracks the feedback when amps are disabled, so there is no motion when reenabling
[20:58:03] <djdelorie> cradek: I'm sure some do, if mine did, I'd use that signal to change the tuning parameters. My problem is, if I give it enough P to drive normally, there's too much "hunting" when it stops, which makes noise and uses power.
[20:58:36] <DJ9DJ> gn8
[20:58:47] <djdelorie> so the plugin, for me, is just there to silence the machine when it's idle.
[20:59:15] <cradek> sorry, I don't get what you're talking about with this plugin
[20:59:26] <djdelorie> http://www.delorie.com/cnc/machineactive.comp
[20:59:31] -!- DJ9DJ has quit [Quit: bye]
[21:00:01] <djdelorie> one of the signals to my controllers is "machine idle" that combines estop, program running, and jog velocity to form one "doing something" signal
[21:00:38] <cradek> your motors go slack when no motion is being commanded?
[21:00:42] <djdelorie> but I had to write that one myself, HAL didn't already have a suitable signal.
[21:00:57] <djdelorie> yes, although as I said, they don't actually move when unpowered
[21:01:09] <skunkworks> heh - my y axis sure would...
[21:01:24] <cradek> yeah that's not gonna work on most machines :-)
[21:01:28] <djdelorie> my Z axis is 3/8-16 all-thread, no problem there :-)
[21:01:55] <djdelorie> well, on "most machines" that signal could just change the tuning values from "precise position control" to "low power hold" then
[21:02:45] -!- FinboySlick has quit [Quit: Leaving.]
[21:04:39] <djdelorie> call it a "energy saving GREEN power management" signal :-)
[21:05:09] <cradek> haha
[21:05:47] <djdelorie> even with hal-controlled servos, it could, for example, cut P in half or something, so you can tune closer to critical but still idle better.
[21:06:09] -!- stevegt_1 has quit [Ping timeout: 252 seconds]
[21:06:20] <skunkworks> A 'decently' tuned system doesn't make much noise or use power. My Big ass servos are stone cold. even my crappy tune doesn't.
[21:06:37] <skunkworks> even on the y axis.
[21:06:50] <skunkworks> (vertical axis)\
[21:07:15] <djdelorie> feel free to come over and tune my servos then :-)
[21:07:36] <skunkworks> heh - I keep trying to get cradek to tune mine... ;)
[21:07:47] <alex4nder> get a room
[21:07:52] <cradek> everyone but me thinks I'm an expert
[21:08:07] <djdelorie> I don't need an expert. I just need someone better at it than I am.
[21:08:13] * archivist votes for expert too
[21:08:19] -!- izua [[email protected]] has joined #linuxcnc
[21:08:19] -!- izua has quit [Changing host]
[21:08:19] -!- izua [izua!~izua@unaffiliated/izua] has joined #linuxcnc
[21:08:31] <cradek> skunkworks: one of these days I'll come visit.
[21:08:50] <skunkworks> That would be cool - We really should do a fest...
[21:09:09] <skunkworks> maybe when the cincinati lathe is about to run
[21:10:30] <cradek> bbl, time to go home
[21:11:11] <skunkworks> http://youtu.be/FgOqEz5Tk-Y
[21:11:23] <skunkworks> very happy with even the crappy tune :)
[21:15:35] <archivist> looked like a tenth correction at the end of the move
[21:15:59] <archivist> should be a foot at that weight :)
[21:16:25] <skunkworks> hard to tell - maybe a little less than a tenth.. :)
[21:18:47] <skunkworks> that was the heaviest axis - X
[21:19:30] <skunkworks> don't know exactly - but would guess around 2 ton
[21:19:47] <skunkworks> The pallet is 1 ton iirc
[21:19:49] -!- MarkusBec has quit [Excess Flood]
[21:19:56] -!- MarkusBec [[email protected]] has joined #linuxcnc
[21:19:57] -!- MarkusBec has quit [Excess Flood]
[21:20:00] <archivist> I never made vids of my tests, and I know Im getting lead screw wear so need a rebuild
[21:20:01] <skunkworks> it is more than 2 ton
[21:20:56] -!- MarkusBec [[email protected]] has joined #linuxcnc
[21:20:57] -!- MarkusBec has quit [Excess Flood]
[21:21:56] <archivist> my B axis is causing a lot of extra friction due to off centre weight, I some times add weight just to balance
[21:21:57] -!- MarkusBec [[email protected]] has joined #linuxcnc
[21:31:15] -!- skunkworks has quit []
[21:31:31] <joe9> djdelorie: when i start up with the homing configuration, it says "joint 0 on limit switch error", and all the (x, y and z) values are 0.00
[21:32:12] <joe9> is there anything I can do to tell the machine to find the actual 0.00 instead of assuming that the start (or, at power on) position is 0.00?
[21:38:17] <djdelorie> are any of your limit switches active?
[21:38:37] <joe9> yes, Y and Z, as they are disconnected.
[21:38:44] <joe9> I am trying out the X - axis now.
[21:38:57] <djdelorie> until you've actually homed the machine, the XYZ numbers on the display are irrelevent
[21:39:00] <joe9> http://codepad.org/tZFCdEEn is my config in the .ini file.
[21:39:28] <joe9> yes, that is what I am trying to do.
[21:39:36] -!- syyl_ has quit [Quit: Leaving]
[21:39:36] <joe9> homing the machine.
[21:39:59] <djdelorie> manually move the machine to the center-ish, so there's no chance of any switch being "hit". If you get the limit error as soon as you start, you probably need to invert the inputs
[21:40:14] <joe9> djdelorie: ok, thanks.
[21:41:06] <djdelorie> remember the part where I said "your HOME_OFFSET will be one of those two values" ?
[21:41:13] <djdelorie> yours isn't
[21:41:42] <joe9> i know.
[21:41:54] <joe9> i am trying to get it a little bit off of the limit.
[21:41:59] <djdelorie> since you're using one switch as both min/max limit *and* home, your HOME_OFFSET must be *exactly* the same as the min/max for that switch
[21:42:02] <joe9> experimenting at this point.
[21:42:06] <djdelorie> those are *physical* descriptions of your machine
[21:42:16] <joe9> oh, ok. thanks. will change it.
[21:42:26] <djdelorie> HOME is the one that's somewhere in the middle
[21:42:59] <joe9> what scared me was in the figure, there was travel after the home offset.
[21:43:09] <joe9> and to me that means travelling beyond the axis end.
[21:43:10] <djdelorie> a tiny bit, yes
[21:43:35] <joe9> i probably need to change the min limit from 0 to -0.01 or something like that.
[21:43:37] <djdelorie> my limit switches are a few mm away from the physical end for that reason, and you set the HOME_SEARCH_VEL low enough that the axis can stop before it hits anything
[21:44:08] <joe9> oh, ok. i understand now. thanks.
[21:46:36] <djdelorie> also a good reason to mount the switches so that the axis goes past them, rather than smashing into them. I use roller switches for that reason
[21:46:49] <djdelorie> an optical interrupter would also be appropriate
[21:47:36] <joe9> mine is mounted to the side. it is be protected by the axis end. I understood why use were using the roller switches.
[21:47:51] <joe9> djdelorie: thanks for bringing it up.
[21:51:27] <mozmck> you can get optical interrupters off old printers if you want some cheap!
[21:51:43] -!- bedah has quit [Quit: gn8]
[21:54:10] -!- izua has quit [Quit: brb]
[21:54:37] -!- izua [[email protected]] has joined #linuxcnc
[21:54:37] -!- izua has quit [Changing host]
[21:54:37] -!- izua [izua!~izua@unaffiliated/izua] has joined #linuxcnc
[21:54:51] -!- Fox_Muldr has quit [Ping timeout: 265 seconds]
[21:56:10] -!- Fox_Muldr [[email protected]] has joined #linuxcnc
[21:58:28] -!- FinboySlick [[email protected]] has joined #linuxcnc
[22:01:53] -!- FinboySlick has quit [Client Quit]
[22:02:30] -!- FinboySlick [[email protected]] has joined #linuxcnc
[22:07:54] <joe9> djdelorie: quick question, do you prefer searchvel = latchvel = +ve (second pic) or, the first pic where searchvel is +ve and latchvel is -ve.
[22:08:04] <joe9> the second seems more robust.
[22:08:11] <djdelorie> latchvel should be slower so it detects the edge more reliably
[22:09:11] <joe9> so, do you have "searchvel = latchvel = +ve"
[22:09:26] <joe9> and, would you mind sharing those values?
[22:09:32] <djdelorie> ah, I use the "same" config
[22:09:39] <joe9> oh, in stepconf.
[22:09:56] <djdelorie> http://www.delorie.com/cnc/dj_cnc_1.ini but I don't think I tweaked those values much
[22:10:29] <joe9> so, both positive. thanks.
[22:10:51] <joe9> btw, how did you come up with those velocity values.
[22:11:04] <joe9> for me, stepconf put them both at 0.05
[22:11:21] -!- Tom_L [Tom_L!~Tl@unaffiliated/toml/x-013812] has joined #linuxcnc
[22:12:37] <joe9> djdelorie: is the velocity units "LINEAR_UNITS/min"?
[22:12:47] <djdelorie> I have no idea
[22:13:19] <djdelorie> I think it's inches/sec as 0.5in/s sounds about right
[22:15:08] <joe9> from the manual, velocity in machine units per second
[22:21:29] -!- tom3p [[email protected]] has parted #linuxcnc
[22:22:05] <Tom_itx> who was asking about epoxy here the other day?
[22:22:10] <Tom_L> http://www.masterbond.com/tds/ep21fl?utm_source=epn&utm_medium=email&utm_content=ep21fl&utm_campaign=hearst
[22:23:22] -!- Tom_L has quit []
[22:34:07] -!- Farthen has quit [Excess Flood]
[22:35:24] -!- Guest83733 [Guest83733!~Farthen@2a01:4f8:101:2a4:0:bc28:b2e1:9] has joined #linuxcnc
[22:37:06] <joe9> djdelorie: quick question, is it a good idea to have 2 switches in series for each limit/home switch? this is to protect in a scenario when one switch fails.
[22:37:22] <joe9> the spec says that the lifetime of the switch is 100k
[22:37:32] <joe9> any thoughts, please?
[22:37:33] <djdelorie> are you going to home your machine that many times?
[22:37:40] <joe9> hope not.
[22:38:03] <joe9> given that alex4nder pointed out that a milling env is harsh on the switches.
[22:38:07] <joe9> just checking.
[22:38:07] <djdelorie> once homed, the switches should never be hit unless something goes wrong, and the E-Stop button is the next switch that should be hit anyway
[22:38:15] <joe9> oh, ok.
[22:41:26] -!- skunkworks__ [skunkworks__!~chatzilla@str-bb-cable-south-3-102.dsl.airstreamcomm.net] has joined #linuxcnc
[22:48:54] -!- PCW has quit [Quit: ChatZilla 0.9.88.1 [Firefox 3.6.13/20101203075014]]
[22:51:04] -!- asdfasd has quit []
[22:54:09] -!- GeorgeH [[email protected]] has joined #linuxcnc
[23:05:18] -!- mhaberler [[email protected]] has joined #linuxcnc
[23:07:17] <JT-Shop> evening guys
[23:07:27] <alex4nder> yoh
[23:09:16] <Tom_itx> hey
[23:13:03] <JT-Shop> dang Zotac MB on the CHNC started to shut down on it's own lately
[23:17:31] -!- factor has quit [Ping timeout: 264 seconds]
[23:18:19] <djdelorie> JT-Shop: so you need an E-Stop switch *and* an MB-Stop switch ?
[23:19:40] -!- factor [factor!~factor@r74-193-21-187.cnrocmta01.conrtx.tl.dh.suddenlink.net] has joined #linuxcnc
[23:19:58] -!- A0Sheds has quit [Quit: puff of smoke]
[23:23:25] <JT-Shop> MB-Stop?
[23:23:58] <djdelorie> AAAHHH! My Motherboard is on fire! KILL IT!!!!!!!
[23:24:01] <djdelorie> mb-stop
[23:24:14] <JT-Shop> I can do that with my cannon I think
[23:24:21] -!- GeorgeH has quit [Ping timeout: 246 seconds]
[23:24:29] -!- A0Sheds [[email protected]] has joined #linuxcnc
[23:24:29] -!- A0Sheds has quit [Changing host]
[23:24:29] -!- A0Sheds [A0Sheds!~ly@unaffiliated/l84supper] has joined #linuxcnc
[23:24:30] <JT-Shop> it's seems to just take a nap on it's on
[23:24:33] <JT-Shop> own
[23:24:57] <djdelorie> as long as the machine stops when that happens! Have you looked to see if the power saving stuff turned itself back on?
[23:26:38] <JT-Shop> yea, that is all off it seems to do it after long periods of idle time
[23:28:08] -!- sumpfralle [[email protected]] has joined #linuxcnc
[23:28:54] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[23:33:07] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[23:33:26] -!- sumpfralle [[email protected]] has joined #linuxcnc
[23:36:50] * JT-Shop hopes he saved the program for the cannon bolt heads...
[23:38:24] <Tom_itx> JT-Shop, is the zotiac dusty around the heatsink?
[23:38:39] <JT-Shop> Tom_itx: it might be
[23:38:45] -!- sumpfralle has quit [Read error: Connection reset by peer]
[23:38:47] <Tom_itx> my video cuts out when it gets too dusty
[23:38:49] <JT-Shop> I'd have to pop the top and see
[23:38:50] -!- sumpfralle1 [[email protected]] has joined #linuxcnc
[23:39:17] <djdelorie> my machines all need annual internal dustings too
[23:39:55] -!- pfred1 [pfred1!~pfred1@unaffiliated/pfred1] has joined #linuxcnc
[23:40:32] <Tecan> thank lord im done work
[23:40:40] <Tecan> im starting to get crazy
[23:41:02] <Tecan> going postal is the new in thing isnt it ?
[23:41:55] <Tecan> anyone know how to make custom M codes work ?
[23:43:22] <skunkworks__> tecan: the docs directions worked for me.
[23:43:43] <Tecan> did you update to 2.5 ?
[23:43:56] <JT-Shop> it's not too dusty in there but it don't smell right
[23:43:58] <Tecan> i tried sticking them in nc_files and making them executable
[23:44:32] <Tom_itx> JT-Shop, look for hot spots
[23:45:18] <Tecan> http://www.youtube.com/watch?v=zAlNrtcPCLw << insane in the membrane
[23:45:39] <Tom_itx> JT-Shop, or is it just oil mist getting sucked thru the fan?
[23:47:04] <JT-Shop> no, it is kinda sealed up sorta
[23:47:29] <JT-Shop> and defiantly not part of the inside air
[23:49:08] <skunkworks__> tecan: what is the content of the file?
[23:49:26] <Tecan> python graster script
[23:49:36] <Tecan> for hal threading the laser pwm
[23:49:52] <skunkworks__> neat!
[23:50:39] <JT-Shop> although an amazing amount of dust came out when I blew it out...
[23:51:16] <Tecan> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Rastering_With_A_Laser
[23:51:16] <pfred1> fans need their winter blankets they sit in a draft a lot of the time
[23:52:06] <Tecan> https://github.com/jedediah/hacklab-engraver/blob/4a654eac48d62e718b71c65eabd923b916ff7acb/M101
[23:54:28] <Tecan> skunkworks can you test your nc_file dir ?
[23:58:28] <Tecan> maybe its a linuxcnc 2.5 path bug
[23:58:32] -!- Nick001-Shop has quit [Remote host closed the connection]
[23:59:51] <skunkworks__> Tecan: I get this when I try to run the M101 script.. (I am sure i have nothing setup right....)
[23:59:53] <skunkworks__> http://pastebin.com/KNZpWsz3