#linuxcnc | Logs for 2013-03-06

Back
[01:40:13] <Loetmichel> mornin'
[01:40:36] <Jymmm> yo
[02:02:59] <Jymmm> r00t4rd3d: looks like it to me
[02:10:57] <DJ9DJ> moin
[04:50:45] <fomox> Does it exist any open source router plans ?
[05:34:58] <TekniQue> 09:50:47 <@johntramp> http://i.imgur.com/nVavlZw.jpg
[08:18:11] <FinboySlick> Anyone taken a peek at these? http://www.supermicro.com/products/motherboard/Xeon/QM77/X9SPV-M4.cfm
[08:18:28] <FinboySlick> There's a few red flags but SuperMicro usually knows how to make boards right.
[08:44:40] <Jymmm> FinboySlick: http://www.supermicro.com/support/resources/OS/QM77.cfm
[08:49:24] <FinboySlick> Jymmm: I meant more for latency tests. I know they'll work with linux.
[08:54:16] <Jymmm> They'll work for MOST linux
[08:56:44] <r00t4rd3d> http://www.ebay.com/itm/NEW-IN-BOX-Canon-Bluetooth-White-Calculator-3-in-1-Wireless-Mouse-Keypad-Mark-1-/111009920127?pt=Mice&hash=item19d8b4f47f
[08:57:37] <r00t4rd3d> odd
[09:03:06] <pcw_home> more than odd, verging on creepy...
[09:04:25] <r00t4rd3d> http://www.ebay.com/itm/Canon-Wireless-Mouse-Calculator-All-in-One-X-Mark-I-Mouse-Slim-Calculator-/170936971840?pt=Mice&hash=item27cca33240
[09:04:30] <r00t4rd3d> they got a new version
[09:04:33] <r00t4rd3d> i think i might get one
[09:05:06] <r00t4rd3d> i need a bt mouse
[09:05:12] <R2E4_> What do you do with it? Calculate your budget while machining?
[09:05:25] <r00t4rd3d> i guess they popular with accountants and such
[09:05:46] <Jymmm> road warriors
[09:05:54] <Jymmm> salesmen
[09:05:58] <jdh> dorks
[09:05:58] <r00t4rd3d> i just wanna look at 80085 24/7
[09:06:36] <r00t4rd3d> i use my computers calculator alot though when designing parts
[09:06:48] <r00t4rd3d> so it actually would be useful
[09:06:51] <jdh> I use bc
[09:07:25] <Jymmm> x-4/2-4
[09:08:38] <Jymmm> Ir's been so long, I can't even remember how to solve that and I have the answer too!
[09:08:52] <JT-Shop> seems like Woody is a very efficient converstion factory on paws... he can convert the smallest amount of cat food to the largest piles of poop I've seen from a cat his size
[09:09:52] <r00t4rd3d> do you weight and measure them?
[09:10:43] <r00t4rd3d> and compare those numbers to intake
[09:10:45] <JT-Shop> calibrated eyeball
[09:11:05] <Jymmm> O, how do you solve: x-4/2-4 ? http://upload.wikimedia.org/math/f/d/a/fda0f946e0dd632a80d832e8d90cc170.png
[09:13:33] <Jymmm> Is that: x-2/-2 ?
[09:13:50] <Jymmm> err x-4/-2
[09:15:09] <R2E4_> 1
[09:15:58] <Jymmm> R2E4_: ?
[09:16:46] <R2E4_> x=4 same as 2=4 4/4 =1
[09:16:52] <R2E4_> I suck at math
[09:16:58] <pcw_home> it s line with y intercept of 2 and x intercept of 4
[09:17:25] <Jymmm> pcw_home: ?
[09:21:25] <Jymmm> pcw_home: The solution is to the right, but is it solved?
[09:21:33] <Jymmm> but how is it solved?
[09:22:07] <Jymmm> 1/6x^2
[09:22:12] <R2E4_> x-6
[09:22:22] <R2E4_> x-4over2-4
[09:22:57] <Jymmm> R2E4_: and then?
[09:23:23] <pcw_home> I may have the order of operations wrong so X-6 is correct
[09:23:51] <Jymmm> pcw_home: Isn't the answer: 1/6x^2
[09:24:15] <Jymmm> ?
[09:27:40] <pcw_home> assuming standard order of operations the division is done first so
[09:27:42] <pcw_home> x -(4/2) -4 = x-6
[09:29:07] <pcw_home> which is a line with intercepts 0,-6 and 6,0
[09:49:52] <Jymmm> x-4/2-4 * x-5/2-5 and I get:
[09:49:58] <Jymmm> x-4/-2 * x-5/-3 and I'm stuck again
[09:51:31] <Jymmm> The solution is: 1/6x^2 - 3/2x + 10/3 but I'm not sure how they got there
[09:52:01] <Jymmm> err...
[09:52:25] <Jymmm> The solution is: (1/6)x^2 - (3/2)x + 10/3
[09:55:30] <R2E4_> huh?
[09:55:50] <Jymmm> R2E4_: http://upload.wikimedia.org/math/f/d/a/fda0f946e0dd632a80d832e8d90cc170.png
[09:56:22] <R2E4_> x-4/2-4x-5/2-5
[09:56:30] <pcw_home> shoudl be (x-6) * (x-(15/2))
[09:57:12] <pcw_home> unless you have dropped some parens
[09:58:27] <pcw_home> is this a test in order of operations? looks like it
[10:00:29] <pcw_home> http://en.wikipedia.org/wiki/Order_of_operations
[10:02:05] <R2E4_> thats complicated
[10:03:55] <Jymmm> pcw_home: No, but that (in part) turns into this http://upload.wikimedia.org/math/4/9/4/49402f5fbe1fa7ec8a7c4ec73fc342de.png
[10:04:40] <Jymmm> So, I'm just trying to go with the basics for now and deal with the rest later.
[10:08:18] <r00t4rd3d> i think you should stick with fuel pumps
[10:08:30] <Jymmm> rotf
[10:08:44] <r00t4rd3d> :)
[10:08:51] <Jymmm> Noooooooo, fuel pumps too hard!
[10:09:19] <Jymmm> I just haven't done this shit in so long I don't have a clue anymore
[10:09:33] <pcw_home> you need to apply parens or its to easy to get lost in the order of execution
[10:09:35] <pcw_home> (x-4)/(2-4)*(x-5)/(2-5) = (x-4/-2)*(x-5/-3)
[10:10:36] <Jymmm> pcw_home: Ok, but how did they get: (1/6)x^2 - (3/2)x + (10/3) ???
[10:10:59] <pcw_home> by multiplying fractions
[10:11:10] <Jymmm> show your work!
[10:12:18] <r00t4rd3d> ive programed a NFC tag to open this link
[10:12:20] <r00t4rd3d> http://www.youtube.com/watch?feature=player_detailpage&v=PyOFHmz-mVc#t=32s
[10:12:37] <r00t4rd3d> now I am going to stick it on the back of my buddies dock
[10:14:09] <pcw_home> (x-4)*(x-5)/6
[10:15:14] <mazafaka> r00t4rd3d: 'ive programed a NFC tag to open this link' <- is it some kind of greeting text line as in the Futurama? Then I write 'In Dolby Surround where available', and hello
[10:16:25] <pcw_home> = (x^2 -9x +20)/6
[10:17:32] <pcw_home> = x^2/6 + (3/2)x + 10/3
[10:18:18] <mazafaka> Jymmm: kids should make their home work themselve, you need to do your adult things. To sit and watch TV.
[10:19:03] <Jymmm> mazafaka: LOL, no homework actually
[10:19:47] <mazafaka> Jymmm: then what? The 'Sesame Street' TV programme?
[10:20:13] <Jymmm> mazafaka: It's a secret =)
[10:21:07] <Jymmm> pcw_home: How did you derive (x-4)*(x-5)/6 from (x-4/-2)*(x-5/-3)
[10:22:02] <pcw_home> multiplying fractions
[10:22:32] <mazafaka> Jymmm: Son... Knowing you, I would think these are puzzles for the governments to switch their attention from your CNC robot which will steal money from Fort Knox, while you will spamming e-mails of NSA clerks with a tag 'Must Resolve Today'
[10:22:43] <pcw_home> A/B * C/D = AB/CD
[10:23:29] <Jymmm> mazafaka: VERY VERY Close actually
[10:24:33] <mazafaka> Jymmm: way too complicated Captcha, please don't develop it :)
[10:24:57] <Jymmm> mazafaka: Not a captcha =)
[10:25:13] <pcw_home> so first step is re-arrange to (x-4)*(X-5)/(-2*-3)
[10:26:39] <tjtr33> I was surprised to see no mail at all from user or dev. are lists down?
[10:26:42] <Jymmm> mazafaka: Better? http://upload.wikimedia.org/math/4/9/4/49402f5fbe1fa7ec8a7c4ec73fc342de.png
[10:27:45] <Jymmm> mazafaka: Solve for z so we know you are hu-mon:
[10:32:44] <skunkworks> FinboySlick, we are using simlar supermicro atom boards for linux servers
[10:33:37] <skunkworks> oh - i7..
[10:37:54] <FinboySlick> skunkworks: I might have read wrong but they may also have xeon versions of those.
[10:38:12] <FinboySlick> I wish there was an extra pci slot though.
[10:44:22] <pcw_home> http://www.newegg.com/Product/Product.aspx?Item=N82E16813128585 the celeron 847 is really an I 3 running
[10:44:24] <pcw_home> r e a l l y s l o w
[10:45:12] <pcw_home> 17 W
[10:45:13] <FinboySlick> intel is growing increasingly bold at crippling its own hardware.
[10:46:18] <Jymmm> Intel has been doing that for decades
[10:46:29] <pcw_home> what interesting is that its about twice as fast as the fastest Atoms at 1/2 the clock speed
[10:46:34] <skunkworks> and it still needs a fan?
[10:46:35] <skunkworks> ;)
[10:47:13] <pcw_home> 17W is a bit much without a giant heastsink
[10:48:00] <pcw_home> I think some of the NUCs use the 847
[10:48:45] <FinboySlick> Jymmm: I wish AMD would start kicking ass again.
[10:49:02] <pcw_home> The ASUS AMD Hudson MB I have is passively cooled
[10:49:15] <Jymmm> FinboySlick: I know nothing about AMD, thus why I never even look at their stuff.
[10:49:28] <pcw_home> and runs circles around the Atom
[10:49:44] <FinboySlick> In the days of the athlon XP, they were great.
[10:50:07] <pcw_home> (10 usec latency and 8 KHz servo thread)
[10:50:28] * Jymmm needs a math book!
[11:13:21] <L84Supper> AMD suffered from very poor management after the arrival of the 64bit extensions
[11:13:48] <L84Supper> they considered buying nvidia before ATI
[11:13:58] <L84Supper> now nvidia could buy them :)
[11:17:11] <skunkworks> https://www.youtube.com/watch?v=IVpOyKCNZYw
[11:17:14] <skunkworks> Not work safe
[11:20:11] <t12> they may hate nvidia
[11:20:18] <t12> but sure beats the ati drivers for linux
[11:20:24] <t12> at least as of a few years ago
[11:30:14] <tjtr33> skunkworks, on same page did you see 'Hitler uses git' ? https://www.youtube.com/watch?v=CDeG4S-mJts
[11:30:57] <skunkworks> yes - that has come up a few times :)
[11:31:23] <tjtr33> rofl
[11:50:43] <Jymmm> t12: AMD now owns ATI
[11:52:50] <__Gabe__> Its this the irc for feminine products?
[11:53:11] <__Gabe__> :) just kidding
[11:55:21] <__Gabe__> So i wrote a module that Queues RTSIGMIN, on pin state changes, and i'm driving my gui with those signals instead of using a while loop. This seems to be working pretty good. What are the cons of this?
[11:56:43] <jdh> if you are sitting in a leaf and put your arm out, you can hit everyone in the vehicle
[11:56:48] <jdh> <urk>
[11:57:11] <jdh> no clue on the cons, but it sounds like a good idea.
[11:57:21] <__Gabe__> Maybe i should poll the mailing list
[11:57:40] <__Gabe__> it works incredibly well even if your using time.sleep() in python
[11:58:10] <__Gabe__> you just map your python program to its pid pin, so it sends the signal to the write process
[11:58:21] <__Gabe__> right&
[11:58:26] <__Gabe__> right*
[11:59:51] <__Gabe__> should mention that im using rtpreempt though on ubuntu 12.04
[12:04:30] <tjtr33> __Gabe__, ask on linuxcnc-devel, the min max * sigquueue stuff is beyond mere mortals
[12:05:23] <__Gabe__> im mailing it out on the list write now
[12:07:18] <IchGuckLive> hi all B)
[12:10:37] <IchGuckLive> question is there a formular for getting the movement of a pipe tp fit a pipe on the A axis HARD to Discribe
[12:11:20] <IchGuckLive> plasma cuts a pipe at the top A turns no Y only X along pipe Axis
[12:11:39] <IchGuckLive> so the cut takes the same pipe in vertical to weld
[12:12:03] <IchGuckLive> on mill it woudt be a perfect curve
[12:15:14] <IchGuckLive> so Z and Y is fix and A,X are moving
[12:21:46] <IchGuckLive> hi kiplitton
[12:26:31] <pcw_home> IchGuckLive: pretty sure I've seen this done by a LinuxCNC'er but maybe that was with a welding robot
[12:27:51] <IchGuckLive> :D
[12:29:43] <IchGuckLive> it has to be only the roloff of the corner and as it is a circle to flat it is in a formula
[12:30:02] <IchGuckLive> i get some students tomorrow on this
[12:30:33] <IchGuckLive> welcom Keknom
[12:33:58] <Keknom> hello
[12:34:08] <IchGuckLive> B)
[12:34:23] <andypugh> IchGuckLive: Viewed down the axis of the mating pipe it is a perfect circle.
[12:34:36] <IchGuckLive> what is your profession milling lathe plasma laser 3D printing ....
[12:34:54] <pcw_home> Yes the hole will be elongated so the projection of the hole is a circle
[12:35:29] <IchGuckLive> and the movement is not a strait
[12:35:42] <andypugh> It is a function of the diameter of the two pipes
[12:35:45] <IchGuckLive> (line)
[12:36:08] <IchGuckLive> diameter are most Equal
[12:36:27] <andypugh> Might as well solve for the general case
[12:37:04] <andypugh> The equation in xy of the top pipe is x^2 + y^2 = r1^2
[12:37:04] <IchGuckLive> someone has inventor at hand to get a metall sheet part to strait flat
[12:37:19] <andypugh> We can leave X alone.
[12:37:53] <andypugh> The Y displacement can be transformed into an angle, it's pretty easy I think.
[12:38:03] <IchGuckLive> yes
[12:38:57] <andypugh> Look along the axis of the second pipe. You need the angle at which the distance to the centre is the Y above
[12:39:40] <andypugh> y = r2 sin theta
[12:39:49] <IchGuckLive> unwinding
[12:40:12] <andypugh> so your shape is the locus of points x^2 + (r2 sin theta)^2 = r1^2
[12:41:03] <andypugh> But actually it is going to be more convenient to follow a top-pipe angle from 0 to 360, calculating x and theta for each point.
[12:41:34] <andypugh> x = r1 sin A, y = r2 cos A.
[12:41:45] <IchGuckLive> on etch deg is enoph for low R
[12:42:49] <andypugh> r2 cosA = r1 sinB, or B = arcsin(r2 cosA)/r1
[12:44:44] <andypugh> so, for A = 0 to 360 X = r1 sinA, B = arcsin(r2 cosA)/r1
[12:45:39] <andypugh> 1821 to 1831. You owe me £8.33 consultancy :-)
[12:46:53] <andypugh> I wonder if it is true?
[12:48:39] <andypugh> For r1 > r2 it should make one half of the "mouth" shape too.
[12:49:10] <andypugh> Darn! I want to build the machine now.
[12:52:39] <pcw_home> also fun if the pipe axis do not intersect
[12:53:06] <IchGuckLive> only a x ofset then
[12:53:14] <archivist> viesters did that for pipe welding didnt he?
[12:54:07] <pcw_home> thats who it was I vaguely remember his robot from a few years ago
[12:55:48] <andypugh> My equation is only valid for pipes at right-angles.
[12:56:15] <pcw_home> also this will only work for thin pipes (line you really should cut on with the smaller pipe axis
[12:56:43] <pcw_home> normal to the surface is not right
[12:57:04] <andypugh> In practice it is easier to let a hole-saw solve the equations :-)
[12:57:05] <archivist> working out intersections in cones can be fun http://www.collection.archivist.info/archive/DJCPD/PD/2010/2010_07_27_tinware/IMG_0797.JPG
[12:57:53] <archivist> sheet metal books have the developments the old fashioned graphical way
[12:58:44] <andypugh> The tin bath ppears to be to a different scale?
[12:59:12] <archivist> scale!!!
[12:59:32] <pcw_home> looks like my sheep water trough
[12:59:55] <andypugh> I guess it might be a loaf tin
[13:00:35] <archivist> the other picture in that dir has the jug top which was a bugger to bend
[13:01:01] <archivist> andypugh, was meant to represent a tin bath
[13:01:45] <archivist> but so long ago I cannot remember if it was to any particular scale
[13:02:21] <Jymmm> Has anyone played with airsoft?
[13:02:40] <andypugh> Leaving the plane rattling aroiund with the hammers like that feels wrong :-)
[13:05:05] <archivist> andypugh, the plane and chisel do get used
[13:05:49] <archivist> clock case veneer repair
[13:06:37] <andypugh> I was wondering, but was scarfed to ask.
[13:49:12] <archivist> hmmm Maiden screwing machines
[13:49:15] <archivist> http://www.gracesguide.co.uk/Maiden_and_Co
[14:03:45] <JesusAlos> hi
[14:06:54] <c60> Afternoon, anyone around that can help me get a joypad working as a pendant?
[14:07:47] <Jymmm> Old... http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Using_A_Joypad_To_Move_Your_CNC_Machine
[14:08:00] <JT-Shop> c60 just ask the question
[14:08:15] <Jymmm> Newer... http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Simple_Remote_Pendant
[14:08:59] <c60> well I've setup everything according to that wiki, and I can see the outputs in hal meter and have changed the values to the correct names, but it's not jogging.
[14:09:16] <c60> so I'm thinking I must be missing something, I'm using the exact same joysti
[14:09:24] <JT-Shop> are you pressing the enable button?
[14:09:33] <c60> you mean one of the number buttons?
[14:09:41] <c60> or is there another enable?
[14:09:45] <JT-Shop> yes, if you did the 3 speed one
[14:09:54] <JT-Shop> press a speed first then move the joystick
[14:10:03] <c60> yup, still not working though
[14:10:47] <JT-Shop> open up Machine > Show Hal Configuration and watch the pins as you move the joystick
[14:12:07] <c60> yup, that all works, I have input.0.abs-x-position reporting correct and all the other pins/buttons
[14:13:06] <c60> which I'm using input.0.abs-y-position and input.0.abs-rz-position
[14:14:05] <JT-Shop> and you have the manual control tab selected?
[14:14:11] <c60> yup
[14:14:21] <c60> I can jog with the keyboard no problem
[14:14:31] <JT-Shop> forgot to addf something?
[14:15:19] <andypugh> Put a halmeter on joy-speed-final and see what comes out.
[14:15:47] <andypugh> And are you remembering to press one of the back buttons?
[14:15:48] <JT-Shop> put halui.jog-speed and halui.0.analog or 1 or 2 and see that they show a change
[14:16:17] <JT-Shop> put in the watch window
[14:16:27] <JT-Shop> hi andypugh
[14:16:41] <andypugh> Hi
[14:16:46] <c60> so when I watch halui.jog-speed I see nothing
[14:16:51] <c60> or 0 rather
[14:17:21] <andypugh> I am guessing that either you are not pressing the back buttons, or you didn't addf the mux4
[14:17:47] <andypugh> Or you haven't turned the joypad on (mine only workw with the LED on)
[14:18:15] <c60> I have no joy-speed-final
[14:18:26] <c60> Just halui.jog-speed
[14:18:53] <JT-Shop> can you pastebin your postgui.hal file
[14:18:55] <c60> I have addf mux40.0 servo-thread and the loadrt mux40 count=1 in my postgui.hal
[14:19:02] <andypugh> You seem to lack the net joy-speed-final halui.jog-speed <= mux4.0.out line then?
[14:19:19] <L84Supper> which open CAM application has the most features and best support?
[14:19:52] <andypugh> L84Supper: Are there enough for there to be a contest?
[14:19:57] <c60> I have net joy-speed-final halui.jog-speed <= mux40.0 out in my postgui.hal
[14:20:04] <L84Supper> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Cam
[14:20:30] <andypugh> So you should have a joy-speed-final signal then (it's under signals, not pins)
[14:20:34] <Jymmm> L84Supper: ArtCam, if you have the money to PAY for the features and the support =)
[14:20:59] <L84Supper> Jymmm: is the source available?
[14:21:09] <Jymmm> L84Supper: Sure, just buy the company.
[14:21:37] <c60> my postgui.hal http://pastebin.com/BQBvTfFK
[14:22:02] <andypugh> L84Supper: I think the choice is likely to be between PyCAM and HeeksCNC
[14:22:16] <L84Supper> I'm looking for open source so we can modify it for scanning and imaging to g-code for Linuxcnc
[14:22:23] <JT-Shop> c60 you have some typos
[14:22:30] <Jymmm> L84Supper: you didn't say that.
[14:22:35] <andypugh> mux40?
[14:22:49] <andypugh> or20?
[14:22:54] <JT-Shop> you have loadrt or20 count=2 and it should be addf or2.0 servo-thread
[14:22:59] <JT-Shop> yep and aye
[14:23:17] <andypugh> I am surprised that LinuxCNC even starts, in fact
[14:23:22] <Jymmm> whats or2.0 ?
[14:23:30] <L84Supper> "open CAM" open source, not just open for business :)
[14:23:38] <JT-Shop> a component
[14:23:43] <Jymmm> k
[14:24:03] <Jymmm> L84Supper: Ah, I missed that part
[14:24:30] <Jymmm> L84Supper: you were tlaking too much and it scrolled off screen
[14:24:30] <L84Supper> Jymmm: no problem you're a busy guy
[14:25:11] <Jymmm> busy, ha, I wish I was busy, that would be a "slow day"
[14:25:25] <c60> JT-Shop: I do have the addf or2.0 servo-thread there too, do I not need the loadrt
[14:26:01] <JT-Shop> please read http://www.linuxcnc.org/docs/html/hal/basic_hal.html
[14:26:36] <JT-Shop> also ignore my typo about addf and loadrt
[14:26:55] <JT-Shop> you have or20 and it should be or2.0
[14:27:24] <JT-Shop> the point zero is the instance number like .0 next one is .1 etc
[14:27:36] <c60> ah yes I see I added an extra zero to a bunch of things, I'll fix and see what happens.
[14:29:42] <c60> does the machine need to be homed for this work, or should it work before I home too
[14:30:29] <JT-Shop> home has no effect on jogging except not allowing you to jog outside of the soft limits after you home
[14:33:28] <c60> Ok, I fixed the typos' hopefully, but still not working, here's the new http://pastebin.com/pXBMZiwy
[14:34:07] <c60> will closing and restarting axis reload all files, or is there a command I have to issue
[14:34:21] <andypugh> Closing and restarting axis will do it
[14:36:41] <andypugh> I don't quite understand that file, it seems over-complicated. I don't use the "or2" functions. Top button slow, bottom button faster, both buttons really fast.
[14:36:43] <c60> I still don't see a joy-speed-final under signals though.
[14:37:41] <andypugh> Exit axis, start linuxcnc from the command line (command is "linuxcnc") . You will probably see the error reported in the console
[14:37:50] <JT-Shop> pastebin your postgui.hal file again
[14:37:59] <andypugh> he just did
[14:38:19] <JT-Shop> oh missed it
[14:38:58] <c60> No errors on the console, :(
[14:39:03] <andypugh> Do you see the exact button names?
[14:39:04] <c60> usually that's a good thing.
[14:39:33] <c60> andypugh: I don't understand the file either, just going off the wiki,
[14:39:54] <L84Supper> Jymmm: I'm so busy today I got up 2 hours before I went to bed.
[14:40:41] <JT-Shop> when you put halui.jog-speed in the watch window does it change when you press the speed buttons?
[14:41:46] <c60> nope,
[14:42:08] <andypugh> can you see remote-speed-fast and friends change?
[14:42:10] <JT-Shop> add the or2 in's to the watch window
[14:43:01] <JT-Shop> do they change when you press the buttons?
[14:43:10] <c60> Where would that be, the or2 in's?
[14:43:18] <andypugh> pins
[14:43:31] <JT-Shop> or2.0.in0 etc
[14:43:43] <c60> I don't see it anywhere
[14:43:55] <andypugh> do you see the mux4s?
[14:44:11] <JT-Shop> pastebin your ini file
[14:44:11] <c60> nope,
[14:44:20] <c60> which ini file?
[14:44:26] <andypugh> do you see any of the signals or pins mentioned in that file?
[14:44:28] <JT-Shop> you only have one
[14:44:35] <JT-Shop> .ini
[14:45:08] <c60> http://pastebin.com/ytKfHvnG
[14:45:12] <JT-Shop> I suspect you didn't add the postgui to the hal file
[14:45:35] <andypugh> Blame Stepconf!
[14:45:35] <c60> oh here's the problem, I'm referencing the wrong post-gui
[14:45:36] <JT-Shop> what is the name of your postgui file?
[14:45:40] <c60> yup
[14:45:44] <JT-Shop> great
[14:46:01] <andypugh> Do you want to reprise your rant, JT?
[14:46:07] <JT-Shop> LOL
[14:46:14] <JT-Shop> once is enough
[14:46:37] <andypugh> edit the INI file to reference postgui.hal not custom_postgui.hal
[14:46:54] * JT-Shop goes back to welding
[14:46:56] <andypugh> This also explains why it was starting with the typos, the file wasn't even being read
[14:47:21] <c60> well I added the stuff in postgui.hal to the custom_postgui.hal since I had some stuff for the touchoff routine in there.
[14:47:29] <andypugh> OK
[14:47:36] <andypugh> How does it crash now?
[14:47:55] <c60> SUCCUESSS!! Thank you all
[14:49:18] <c60> Has anyone used this with a wireless joypad? I'm probably just going to get a long usb cable but,
[14:51:10] <andypugh> I think a few people are runnign that way, yes.
[15:04:05] <jdh> nothing says "I really trust chinese electronics" like jogging a spinning spindle with a cheap wireless joypad
[15:07:46] <c60> Hehehe, well most of my jogging is done without the spindle spinning ;)
[15:10:35] <r00t4rd3d> fyi its much easier to use a app like joy2key then to program hal to use one.
[15:11:09] <r00t4rd3d> its what i use
[15:11:22] <r00t4rd3d> no hal involved.
[15:19:20] <r00t4rd3d> http://qjoypad.sourceforge.net/
[15:19:36] <r00t4rd3d> thats what i have
[15:26:53] <jdh> I have one of those contour(?) jog shuttle thingies
[15:26:57] <c60> I trust hal more than I trust a keyboard emulator. plus once it's done it's done,
[15:27:07] <c60> and it's done :)
[15:43:59] <__Gabe__> <--- i agree with c60
[15:46:12] <tjtr33> pcw_home, i built this arduino based sub uSec oneshot, and found out the C prgramming stackframe killed its subuSec ability ( ate 1.8uS).
[15:46:12] <tjtr33> should i go for another language, or platform, or skip programmables and head into FPGA land? http://pastebin.com/RexbyRZt
[15:47:28] <tjtr33> works great, 63nS precision but bottom end is 1.8uS :(
[15:49:10] <JesusAlos> there are a way to create a internal variable for general use in Hal?
[15:53:35] <andypugh> Bargain! http://www.ebay.co.uk/itm/130857499320?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649
[15:54:08] <micges> ended
[15:54:14] <andypugh> tjtr33: Toy should be able to use an AVR timer rather than C
[15:54:21] <Connor1> Yea. Cause andypugh bought it.. :)
[15:56:01] <andypugh> r00t4rd3d: But HAL is realtime and deterministic, and keyboard emulation isn't. Key-emulation is just wrong. (And I say that as someone who is working on a key-emulator for HAL)
[15:56:37] <tjtr33> andypugh, nice encoder, is it thru hole? and i used C to setup Timer0 in CTC mode. as close to ASM as i could get in IDE. will try in-line
[15:57:19] <tjtr33> Toy? acronym newto me
[15:57:41] <andypugh> Yes, it's a through-hole one. Not big enough for my application (lathe spindle) but it means it can sit on its own bearings with belt tension taken externally.
[15:57:51] <andypugh> "Toy" was a typo.
[15:58:15] <tjtr33> nice buy
[15:58:22] <andypugh> tjtr33: Perhaps make the timer retriggerable?
[15:58:57] <tjtr33> need oneshots triggered by process
[15:59:25] <tjtr33> edm pulse ( yes of course its for edm, thats what i do :)
[15:59:28] <andypugh> I think you can trigger them on GPIO entirely inside the AVR?
[16:00:07] <andypugh> pcw_home: Hmm, what does it take to interface a line-driver encoder to a 7i43 input?
[16:00:22] <DaViruz> i ended up getting a http://www.ebay.com/itm/261169074895?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649 for my lathe spindle
[16:00:43] <tjtr33> andypugh, yes, http://pastebin.com/RexbyRZt is the code i wrote, the pulse is handled in hdwr once setup by code
[16:00:49] <andypugh> How do you interface it?
[16:02:27] <andypugh> tjtr33: Sorry can't help, I exhausted my knowledge when I said it was possibel hardware-trigger the timers.
[16:02:33] <__Gabe__> JesusAlos Internal variable for what?
[16:02:41] <tjtr33> np, thx andypugh
[16:10:06] <Tom_itx> andypugh, how'd the lid fix go over?
[16:10:16] <andypugh> Folk seemed happy
[16:10:21] <JesusAlos> I need turn bitA to 1 with pinB and turn 0 with pinC
[16:10:43] <JesusAlos> __Gabe__:
[16:11:24] <Tom_itx> i need to fix my pendant spindle logic
[16:11:37] <Tom_itx> been too lazy to get into it though
[16:12:00] <andypugh> JesusAlos: Can you describe in more detail?
[16:12:07] <Tom_itx> you shouldn't be able to 'run' unless the spindle is back on
[16:12:38] <Tom_itx> and you shouldn't be able to turn it off unless it's paused
[16:12:48] <andypugh> JesusAlos: I think you need "lut5" http://www.linuxcnc.org/docs/html/man/man9/lut5.9.html
[16:13:29] <andypugh> The documentation is not very clear :-)
[16:13:30] <__Gabe__> don't you just love hal!
[16:13:38] <__Gabe__> always had what you need
[16:13:42] <__Gabe__> has*
[16:14:37] <andypugh> write out the truth-table as shown, then write the rquired output in the end column, then read it as a binary number reading upwards.
[16:15:36] <andypugh> Tom_itx: lut5 is probably what you want too. Assign the input bits to interlocks and buttons, then write a 1 if you want spindle on, and a 0 if you want spindle-off in the last column, and then convert from binary readin upwards.
[16:15:36] <__Gabe__> why use google... when you can just ask Andy the living hal component encyclopedia
[16:16:34] <Tom_itx> andypugh, when i was working on it i wrote the two at separate times so i need to rework them and integrate em both
[16:17:28] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/cnc/configs/sherline/
[16:17:33] <Tom_itx> my_jog is the one
[16:17:49] <Tom_itx> i think that's fairly current anyway
[16:18:56] <Tom_itx> no, it doesn't have the program start..
[16:18:58] <Tom_itx> i'll update
[16:21:05] <andypugh> JT-Shop: Didn't you change the LUT5 docs to actually make some sort of sense?
[16:21:16] <Tom_itx> what does LUT5 do?
[16:21:40] <andypugh> It lets you define an output state for every possible input state of 5 inputs.
[16:21:40] <Tom_itx> i'll read it later.. suppertime
[16:22:47] <Tom_itx> the pendant works, i just need to remember what is still broke on it
[16:23:02] <JesusAlos> andypugh: http://imagebin.org/249226
[16:23:12] <JesusAlos> is like this
[16:23:24] <JesusAlos> but with Hal pins
[16:23:36] <JesusAlos> and internal variable
[16:24:29] <andypugh> HAL doesn't hold state. So I am not sure an internal variable concept exists.
[16:25:11] <andypugh> Or, alternatively, you can consider any signal as an internal variable.
[16:25:52] <JesusAlos> ok. I can use a GPIO output of 7i77
[16:26:09] <andypugh> I meant "hal signal"
[16:26:29] <DJ9DJ> gn8
[16:26:32] <JesusAlos> ok
[16:27:27] <andypugh> JesusAlos: Can you describe in words what behaviour you want?
[16:27:39] <andypugh> (I don't actually speak ladder :-)
[16:29:07] <Tom_itx> do you know what that's listed under in the help?
[16:29:38] <andypugh> Tom_itx: What?
[16:29:39] <Tom_itx> hal componenet list?
[16:29:41] <JesusAlos> I'm not sure that i can describe with my english beach. But attempt
[16:29:48] <Tom_itx> LUT5
[16:30:16] <andypugh> It's under HAL components -> lut5
[16:30:22] <Tom_itx> found it
[16:32:18] <Tom_itx> i'm not sure mine is that complicated that i need it but it might simplify it
[16:37:51] <Tom_itx> andypugh, do you use G97 and G95 for rigid tapping?
[16:38:31] <andypugh> What? Actually _use_ a machine? With my reputation?
[16:39:21] <Tom_itx> oh and did you see those magnified machine parts someone posted yesterday?
[16:39:29] <Tom_itx> was supposed to pass that on to you
[16:39:31] <andypugh> I think G33.1 is the usual code.
[16:39:45] <Tom_itx> does it cancel G96?
[16:40:03] <Tom_itx> or do you have to set that up ahead of it
[16:40:14] <andypugh> On a lathe?
[16:40:16] <Tom_itx> yes
[16:40:46] <andypugh> I think you need to turn it off, the radius is likely to be completly wrong when the tap is on the centreline.
[17:07:04] <yoshimitsuspeed> Does anyone know how to get Ubuntu to print 1:1? I am using an HP deskjet F2120 with HPLIP. I have tried printing straight from freecad, exported to PDF and printing straight from HPLIP, converting to SVG and printing from GIMP. The fit to page box is uncheckd in HPLIP and I can't find any other settings to force it to print 1:1. What's the easiest way to do this?
[17:11:01] <andypugh> yoshimitsuspeed: You might find an answer here, but to be honest you stand more chance on a generic Linux channel
[17:11:32] <andypugh> It might be a case of getting the dpi setting correct.
[17:14:45] <yoshimitsuspeed> andypugh thanks. This is the only IRC I knew how to find lol. I have about lost hope trying to find any help on the linux forums these days too hehe.
[17:16:04] <andypugh> Theres the #linux channel
[17:16:20] <andypugh> or #linux-help but that looks very quiet
[17:18:21] <JT-Shop> andypugh: I added some info here http://www.linuxcnc.org/docs/html/hal/rtcomps.html#_lut5_a_id_sec_lut5_a
[17:19:14] <andypugh> Ah, yes, I knoew it was somewhere
[17:27:31] <Tecan> http://storage-new.newjamendo.com/?trackid=529286&format=ogg2&u=0
[17:40:28] <JesusAlos> Why halui.program.is-running is ever FALSE ?
[17:40:48] <Tom_itx> if it is not running
[17:40:51] <JesusAlos> Program is executing and continue FALSE
[17:41:07] <andypugh> It refers to the G-code program
[17:41:31] <JT-Shop> are you watching the pin in the watch tab?
[17:41:57] <JesusAlos> only change true when change gcode line?
[17:42:11] <JesusAlos> I watching in hal configuration
[17:42:35] <JT-Shop> did you add the pin to the watch tab or just looking at the pins?
[17:43:03] <JesusAlos> I'm looking pins
[17:43:14] <JesusAlos> not add to wathc tab
[17:43:26] <andypugh> Only the "watch" tab updates live
[17:43:26] <JT-Shop> the pins don't update you need to add the pin to the watch tab to see it update
[17:43:45] <JT-Shop> what Andy said^^^
[17:44:28] <JesusAlos> ok
[17:44:38] <JesusAlos> you a're trust
[17:45:01] <andypugh> Halmeters can be useful, they take up less space and you can open several.
[17:45:03] <JesusAlos> right
[17:45:26] <andypugh> I wish the watch tap saved state. I also wish you could remove items from it.
[17:55:13] <JT-Shop> instead of just clearing the watch tab?
[17:55:36] <JesusAlos> In quick speed, hal don't turn on in watch view
[17:56:32] <JesusAlos> withc tab?
[17:56:33] <JT-Shop> do you mean when a MDI command is ran?
[17:57:31] <JesusAlos> when executing program
[17:58:02] <JesusAlos> there are a pin hal that change very quick from 1 to 0
[17:58:13] <L84Supper> andypugh: http://www.hirox-usa.com/library/machine.html check out the micro gear videos
[17:58:32] <JT-Shop> the gui update may miss a fast change
[17:58:54] <JesusAlos> yes
[17:59:16] <JesusAlos> don't are a buffer?
[17:59:23] <JT-Shop> I think the halmeter is faster but can you see that fast
[17:59:57] <JesusAlos> ah. use scope
[18:00:09] <JT-Shop> yea, use halscope
[18:01:13] <andypugh> L84Supper: Pah! I used to make X-ray machines with better magnification and depth of field than that!
[18:02:14] <L84Supper> andypugh: but these go to 11
[18:03:04] <andypugh> They do seem to get very nice images, to be fair.
[18:03:23] <andypugh> And less bother than SEM
[18:03:42] <L84Supper> what price range were the xray machines in?
[18:04:53] <L84Supper> I come across lots of used xray equipment, I don't recommend any of it for DIY
[18:08:00] <andypugh> I think we liked to get $100k for one :-)
[18:08:58] <Tom_itx> JT-Shop update us on your splitter lever
[18:09:40] <andypugh> L84Supper: http://www.nordson.com/EN-US/DIVISIONS/DAGE/PRODUCTS/EXAMPLEPRODUCTFAMILY/Pages/XD7600NTDiamondFp.aspx
[18:10:09] <Valen> got a customer who wants some M2 brass grub screws for their SEM
[18:10:21] <JT-Shop> http://imagebin.org/249232
[18:10:38] <Valen> anybody heard of M2 brass grub screws?
[18:10:44] <andypugh> That's a cool linkage :-)
[18:10:54] <andypugh> Valen: Maybe
[18:11:01] <JT-Shop> http://imagebin.org/249233
[18:11:05] <Valen> allen key too ;->
[18:11:13] <JT-Shop> aye, very cool
[18:12:05] <andypugh> Valen: Why brass?
[18:12:15] <Valen> don't know precisley just yet
[18:12:27] <Valen> i presume its either vacuum or magnetic reasons
[18:13:49] <andypugh> http://brass-screws.brass-fasteners-inserts.com/brass_socket_set_screws_flat_point_din_913.htm
[18:14:48] <andypugh> If they can use stainless: http://www.modelfixings.co.uk/cup_point_grub_screws.htm
[18:15:04] <Valen> I was going to ask them about stainless
[18:15:28] <andypugh> You would probably have to buy some and see if they were austenitic or not.
[18:15:58] <Valen> I can probably specify 316 for them which will at least help with magnetic
[18:16:25] <andypugh> You could consider making some. Drill and broach with a lathe tailstock.
[18:17:04] <Valen> i believe its under the "nice to have" heading so its worth looking for a supplier, not making them lol
[18:17:15] <andypugh> In theory my lathe can machine them. But I would need a very expensive tiny boring bar.
[18:17:34] <Valen> we were joking about how to do the broach, we figured put the broach in the mill then sneeze really hard on the head
[18:17:52] <Valen> to shape the socket for the allen key
[18:18:07] <andypugh> http://www.ebay.co.uk/itm/50-ph-Horn-R105-1802-0-05-TN35-02-Min-Bore-Boring-Bar-/400067050738?pt=LH_DefaultDomain_0&hash=item5d25dabcf2
[18:18:24] <syyl> grind a hex-punch and force it in with the quil of the lathe or mill..
[18:18:39] <andypugh> (boring bar for 0.02" holes)
[18:18:44] <syyl> the horn tools are epic :D
[18:18:51] <syyl> got a hole bag of them
[18:18:56] <jdh> can you edm a hex socket?
[18:19:02] <andypugh> The hex needs to be a fraction larger than an allen key though.
[18:19:13] <andypugh> jdh:Absolutely
[18:19:26] <JT-Shop> rotary broach
[18:19:41] <Tom_itx> JT-Shop, hand pump or hydraulic?
[18:19:46] <andypugh> Can you get one in 0.9mm ?
[18:19:53] <JT-Shop> yes
[18:20:04] <Valen> I figured just mill the thing, grind a cutter onto a pin and go at it ;->
[18:20:08] <JT-Shop> make one
[18:20:15] <Tom_itx> JT-Shop, may as well put a lever for it on the controls too
[18:20:43] <jdh> that looks a lot like the Upenders we have at work. Minus a lever
[18:22:23] <JT-Shop> it's a clever design that a friend drew up for me
[18:23:50] <andypugh> Getting to the pump lever might be an issue.
[18:23:50] <JesusAlos> GN
[18:23:55] <JesusAlos> Thank
[18:25:02] <Tom_itx> just run a hydraulic line to it
[18:25:08] <JT-Shop> it's close to the 2/3's level
[18:25:32] <Valen> heres a question for you guys, dad has been toying with an idea for a lube pump
[18:26:04] <Valen> but instead of metering orfices and the like it'll actually just be 30 odd peristaltic pumps in one little stepper driven housing
[18:26:34] <Valen> control the volume by changing the size and number of hoses you route to a specific port
[18:26:49] <Valen> good idea?
[18:27:02] <t12> sounds expensive
[18:27:15] <Valen> yes and no
[18:27:18] <andypugh> It might take a surprising level of torque.
[18:27:39] <Valen> we were going to use a 23 size stepper with ~5:1 ratio on it
[18:27:43] <t12> does everything actually need to be lubed at the same time
[18:27:57] <t12> you could do one peristaltic and a big rotary valve
[18:28:15] <Valen> i spose we were thinking our new mill, which is all ballscrews and linear rails so there shouldn't be much pressure
[18:33:02] <andypugh> There are some fairly small solenoid pumps: http://i559.photobucket.com/albums/ss35/katherinebecker/Hawke%20Oiler/h_kit.jpg
[18:33:50] <Valen> yeah, but this one will do 30 odd oil lines
[18:34:06] <JT-Shop> oil or grease?
[18:34:28] <Valen> and it'll totally look the sex with all hoses, module 2 ring gears, brass whatsits etc ;->
[18:34:29] <JT-Shop> when you said lube I assumed grease
[18:34:30] <Valen> oil
[18:34:58] <Valen> I don't think we have ever greased anything on our mills, even the HM45
[18:35:06] <andypugh> I think you should try it and see how it works out.
[18:35:12] <Valen> currently we have a gravity feed oil system
[18:35:35] <Valen> AKA 10 syringes sitting on the wall above the head
[18:36:03] <Valen> he fills them up in the morning and they drip oil through precisely tightened tek screw clamps into the mill
[18:36:28] <JT-Shop> sounds precise to me
[18:36:33] <Valen> (the tek screws clamp on the hose that runs to the mill)
[18:36:50] <andypugh> Like http://farm7.staticflickr.com/6091/6315762801_d0d3db220c_z.jpg ? SOunds pretty
[18:37:15] <jdh> those are pretty
[18:37:26] <Valen> hah it does look a bit like that
[18:37:37] <Valen> except made out of plastic and looking really ugly
[18:37:47] <Valen> what is that? 1850's motorcar?
[18:38:05] <L84Supper> world smallest church pipe organ?
[18:38:31] <JT-Shop> fire engine?
[18:38:53] <Valen> andypugh: wasn't the fire engine person here was he?
[18:39:34] <andypugh> Yes, I run an old fire engine, but that bank of oilers is something older. The fire engine has an oil pump!
[18:40:04] <andypugh> Valen: If you can find an 1850s motor car it would be unique and priceless :-)
[18:40:25] * Valen is a child of the interwebs if its older than about 15 minutes its aincent
[18:40:34] <andypugh> Talking oiler, this mill engine has fish-tanks full of oil sat on the main bearings: http://manchesterhistory.net/rochdale/ellen/steam2.jpg
[18:40:51] <Valen> lol
[18:41:13] <Valen> the big brass box or the little thing on top of the arm?
[18:41:27] <andypugh> The big brass box is full of oil.
[18:41:42] <andypugh> But the other oiler is also fun.
[18:41:44] <Valen> that is a lot of oil
[18:54:59] <archivist> hmm Ellen road engine, been to see that :)
[18:56:27] <andypugh> It's very nice.
[18:56:46] <andypugh> They have a nice workshop too, all full of lineshafting.
[19:03:23] <gabewillen> andy do you have a link to the new RTAI? i googled it with no prevail
[19:04:25] <L84Supper> somebody who is part of the Berlin Hackerspace was interested in this: http://www.indiegogo.com/projects/opensl-3d-printer-a-stereolithographic-printer-for-everyone?c=home
[19:04:46] <L84Supper> I forgot what their nick is
[19:09:40] <andypugh> gabewillen: http://thread.gmane.org/gmane.linux.distributions.emc.devel/9299
[19:09:49] <gabewillen> Thanks!
[20:15:35] <tjb1> Connor: I could help you but not with skeinforge
[20:16:04] <Connor> Yea. Using ReplicatorG with Skeinforge.
[20:16:11] <Connor> Really don't like it..
[20:16:32] <tjb1> I use Kisslicer now but it may not be the best
[20:16:38] <tjb1> Cura or Slic3r
[20:17:45] <Connor> It has Slic3r in it.. but says Experimental..
[20:18:06] <Connor> I've always used Skeinforge.. but.. it really sucks.
[20:18:37] <tjb1> I use repetier host
[20:19:00] <tjb1> I don't know if that works with makerbot though
[20:19:02] <Connor> what sort of printer?
[20:19:09] <tjb1> I am using a kossel now
[20:48:26] <MrTrick> I'm trying to become better acquainted with G-code. In the wikipedia example ( http://en.wikipedia.org/wiki/G-code#Example_program ) Lines 8-9 and 10-11 split a G01 and G00 command over multiple lines. Does that work, then?
[20:48:54] <MrTrick> So "G01 X10 Y20" is one command, and "G01 X10\nY20" is two?
[20:49:10] <jdh> it's just bad rendering of the page
[20:50:03] <jdh> nevermind, I see what you mean.
[20:50:57] <jdh> http://linuxcnc.org/docs/html/gcode/overview.html#sec:Modal-Groups
[20:51:07] <MrTrick> So it works? or it doesn't work? or it works but it's bad practice and it should really be "G01 X10\nG01 Y20"?
[20:51:36] <jdh> It is bad practice in my mind.
[20:51:55] <jdh> or, perhaps my mind is too simplistic. I would never do that.
[20:52:18] <tjb1> G00/G01 stay active until changed
[20:52:32] <jdh> see the URL above for Motion modal commands
[20:56:41] <skunkworks> Connor: how is petes machine?
[21:01:15] <Connor> skunkworks: Still in the process of wiring everything up.
[21:01:46] <Connor> He's been working on the Pendant and wiring up the AC for the drivers and computer..
[21:01:57] <jdh> what did you end up doing for spindle encoder?
[21:02:42] <Connor> Nothing yet. Was just doing research.. But I think we're going to use the prox sensor to mask the index on the encoder (installed on the spindle motor)
[21:03:50] <jdh> how big is the machine?
[21:05:17] <Connor> Well.. it's pretty big..
[21:05:50] <Connor> here isone..
[21:05:50] <Connor> http://www.exapro.com/used-cincinnati-arrow-500-cnc-vertical-machining-centre-w-acramatic-2100-pc-based-cnc-control-p20727031/
[21:06:23] <MrTrick> Thanks jdh. So, with say G00 active, it will use the newline to delimit, and say differentiate between "X10 Y20" and "X10\nY20" ?
[21:07:35] <jdh> the first is a single coordinated move, the second is two single moves.
[21:07:59] <MrTrick> yup, cool. :-) (sorry, the overview.html didn't seem to say that, at least not that I found)
[21:08:23] <jdh> but, I always explicitly put G0/G1/G2/G3/etc on all motion lines
[21:08:30] * MrTrick plans to drive a grbl controller with dynamically generated g-code
[21:08:44] <MrTrick> Yes, I'd say that's probably good practice.
[21:09:05] <jdh> and I never use Nxxx
[21:10:15] <MrTrick> I'm not sure grbl even supports them. (It's geared mainly towards CAM-generated code)
[23:29:32] <L84Supper> http://www.youtube.com/watch?feature=player_embedded&v=eIwj_ThMf4Q finally a use for Cicada wings