#emc | Logs for 2008-05-14

[00:02:58] <dmess> if its high now... it should do the trick Guest
[00:04:48] <dmess> i just love ati..
[00:33:42] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[01:13:43] <SWPLinux> ns identify monster
[01:13:46] <SWPLinux> dbsf
[01:25:05] <lewing_> lewing_ is now known as lewin1
[01:26:01] <lewin1> lewin1 is now known as lewing
[03:23:07] <Jessica508> Has anyone in here ever used Foredom flexshafts as your spindle? Any comments? How is the runout?
[03:54:48] <Jessica508> Everyone sleeping tonight?
[04:11:24] <K`zan> Trying not too, but suceeding poorly :).
[04:12:10] <K`zan> lower the RSSI the better the signal?
[04:23:57] <SWPadnos> the lower the RSSI, the lower the signal strength
[04:24:16] <SWPadnos> RSSI usually means "Received Signal Strength Indicator"
[06:46:27] <sendo> good morning
[07:16:06] <sendo> i have a question i´m doing the HAL tutorial is that based on an older version of emc2?
[11:59:58] <jepler> sendo: yes, it hasn't been heavily revised since 2.0.0 and probably could use it.
[12:02:22] <sendo> ok have found some differences but got it all working :)
[12:07:12] <Guest875> Guest875 is now known as skunkworks_
[12:11:39] <jepler> sendo: like everything else, the emc documentation is made by volunteers. we welcome contributions. http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?BeyondWiki
[12:11:45] <jepler> * jepler drives in to the office
[12:11:54] <jepler> .. to do the job that pays
[12:12:12] <sendo> i could change it as i go along
[12:12:27] <rayh> Hey. That's the spirit sendo
[12:14:05] <sendo> hehe i´m all for open source i will make my essay also public (about PCB milling) its in dutch though
[12:14:46] <rayh> It is easy to get confused by the stuff on starting a HAL.
[12:15:39] <sendo> yeah i notice
[12:16:54] <rayh> What kind of an install are you working with?
[12:18:18] <sendo> converting an old LPKF so its useable with emc2 XY axis are driven by stepper Z axis is a solenoid
[12:18:42] <sendo> its a pcb prototyping mill
[12:20:14] <sendo> X-Y axis are working properly but didnt need HAL for that the solenoid and spindle will be something different
[12:23:41] <rayh> You could use the spindle forward signal to drive the solenoid down.
[12:25:13] <rayh> motion.spindle-froward and connect that to the pin you connect to the solenoid.
[12:26:09] <rayh> But then you will need m3 and m5 whenever you have to hop over a trace or move to a new part of the board.
[12:26:39] <sendo> yes but i want to achieve the following ex for milling the paths my gcode says Z-0.02 and for drilling z-0.1 in the first case i want to drive my solenoid with a 50% duty cycle and in the 2nd 100%
[12:27:32] <sendo> so i would like to use PWM and some sort of comparison to the generated gcode
[12:27:48] <rayh> Is that two positions of the depth or what?
[12:28:48] <sendo> yes but there is no feedback whatsoever i just want to achieve 2 different press strengths of the solenoid
[12:29:24] <sendo> the problem is that i lack any info of the old setup and way of working i guess its got to be something like that
[12:29:53] <rayh> Don't mind my curiosity. What is the difference to the product between the two.
[12:29:55] <skunkworks_> I really don't think it will do what you want... Solenoids are enharently on - off devices.
[12:30:17] <ALS> maybe use CL and m3 m4 m5
[12:30:45] <rayh> I'm imagining some sort of spring loaded plunger that fights back.
[12:30:51] <sendo> i got a datasheet of the used solenoid its capable receiving PWM cycles 25to 100%
[12:31:20] <sendo> yes there is a damper (or however it is in english)
[12:31:38] <sendo> i will first try the on/off way
[12:31:41] <rayh> Okay. Never seen such a thing.
[12:32:13] <ALS> sendo: did you post a picture?
[12:32:32] <rayh> Like ALS, I can imagine several ways to do this.
[12:32:59] <rayh> But I don't see the advantage to the product.
[12:33:06] <sendo> but i thinkk when drilling there is no problem but when engraving the solenoid will go down and the engraving needle will be on the same sport for too long before X an Y go into action
[12:33:36] <sendo> wait got 2 pictures somewhere
[12:33:45] <rayh> How do you set the depth to cut?
[12:33:56] <rayh> Some sort of mechanical stop?
[12:34:40] <jepler> looks like the pictures were: http://img519.imageshack.us/my.php?image=img0669pb9.jpg http://img132.imageshack.us/my.php?image=img0675qe6.jpg
[12:35:47] <ALS> couldn't the tool lenth take care of the diff depths?
[12:35:55] <sendo> thats my poblem i think you cant there is a 2 micrometer/notch wheel where you can finetune depth but i dont think its ment for setting depth
[12:37:32] <sendo> yeah well another problem my school doesnt want to order the tools before the machine works... there is a difference between drill and needle length yes but its just a guesstimate as to wether that is going to be enuf
[12:37:36] <rayh> Thumbwheel just above the tool.
[12:37:58] <sendo> yes the thumbweel is 2micrometer per notch
[12:37:59] <rayh> Interesting device.
[12:38:20] <sendo> jepler got the coreect links btw thanks
[12:38:29] <rayh> And the duty cycle to the solenoid selects the distance to travel down.
[12:38:41] <rayh> So the same tool can drill and mill.
[12:39:12] <rayh> Thanks for the links jepler
[12:39:28] <sendo> no the dutycycle is a way to set the pressure the solenoid will give on the surface about too be milled
[12:40:19] <sendo> iff you would drive the solenoid at 50% it will also drill a hole only slower then at 100%
[12:41:18] <rayh> My ignorance of circuit boards is showing.
[12:43:12] <sendo> but i will try it first with on/off i will use it as a pen plotter then first to test drive
[12:44:33] <sendo> first i thought this was going to be a piece of cake. the X and Y were damn Z-axis :)
[12:45:32] <rayh> Since your program generator uses z for that, using z to create duty cycle will be easier.
[12:48:30] <sendo> i´m using EAGLE with PCB-gcode script wich works beautifully iff i have time left i´m going to tweak the script
[12:49:20] <rayh> We can read Z position from axis.0.motor-pos-cmd or fb.
[12:49:38] <sendo> thanks writing that down
[12:49:51] <jepler> axis.2 maybe
[12:49:54] <rayh> oops
[12:49:57] <rayh> Yep
[12:50:36] <rayh> And then set some steps to compare.
[12:51:22] <jepler> there are no shortage of ways you can convert the "Z" value into a PWM value
[12:52:10] <rayh> Yea I'm trying to think of a simple one. How about directly driving the pwm from the position pin.
[12:52:36] <sendo> i´m reading hal manual cover to cover i´m sure i´ll find something there
[12:52:59] <rayh> You'd just have to set the PWM range to match the max Z
[12:53:16] <jepler> yes if a simple linear relationship between Z and pwm is acceptable, that's sure the easiest way
[12:54:08] <SWPadnos> if you want "off" then a couple of specific drive forces, then a mux4 with two comparators is likely to be the easiest way
[12:54:31] <rayh> Yep that would do it also.
[12:54:39] <jepler> other ways that come to mind are: wcomp+mux4, wcomp+tristate_float, and a custom component which basically says "if(in>=0) out=0; else if(in>=-0.02) out=0.5; else out=1.0;
[12:54:46] <jepler> er I see that SWPadnos already covered my "first way"
[12:55:02] <rayh> This is not going to be as difficult as it seemed at first.
[12:55:24] <SWPadnos> oh, tristate_float - didn't know about that one :)
[12:55:36] <rayh> sendo, can you manipulate the z position in the eagle software?
[12:55:56] <jepler> SWPadnos: I think I put that one in; I can't remember
[12:56:47] <SWPadnos> yep, it's there
[12:56:53] <SWPadnos> cool
[12:57:06] <rayh> It's there but I don't quite see how you'd use it?
[12:58:07] <sendo> it is possible rayh but not thru the UI i got to do it in the code of the script
[12:58:08] <SWPadnos> connect a bunch of the outputs to the PWM, with the inputs driven by a one-of-N bit
[12:58:25] <SWPadnos> those only update their output pin if the input is true
[12:58:35] <jepler> yes, what SWPadnos said
[12:58:35] <cradek_> cradek_ is now known as cradek
[12:59:04] <rayh> I was thinking an external power supply with several output voltages triggered by pins.
[13:00:14] <rayh> IMO we will need to match our solenoid driver to whatever eagle puts out for this.
[13:02:06] <rayh> sendo, How many different Z positions do you see?
[13:02:19] <rayh> in the gcode?
[13:02:32] <sendo> 3
[13:02:32] <rayh> * rayh needs more coffee!
[13:03:08] <sendo> 0, -0.02 for making tracks, -0.1 for drilling
[13:03:35] <sendo> so thats 2 for the solenoid as 0 is just solenoid off
[13:04:01] <SWPadnos> sounds like a couple of wcomps would do it
[13:04:14] <rayh> Yes it does.
[13:04:43] <rayh> Or the custom jepler suggested.
[13:04:53] <SWPadnos> though there would probably be a "thunk" at the transition between the two on forces
[13:07:38] <rayh> I wonder if the program does that. Mill then drill or drill then mill without a 0 between.
[13:08:16] <SWPadnos> it probably does go to 0 in between - you're unlikely to drill in the same spot as you just milled
[13:08:29] <rayh> Maybe the drill is really pushing the mill tool through the foil.
[13:08:44] <SWPadnos> but there will still be transitions on the way "down", as the SZ motor pos command goes from 0 down to -0.1
[13:08:57] <sendo> the milling and drilling is done in 2 seperate nc files there is also a tool change needed so it wil not go from one state to the other without going to 0 first
[13:08:58] <SWPadnos> it will transition through the 50% setting
[13:09:21] <rayh> Okay. That clears that up.
[13:10:02] <rayh> This doesn't sound all that difficult.
[13:10:07] <SWPadnos> and if the wcomps aren't perfectly aligned (not sure how the hysteresis works there), then you can have neither or both wcomps active near the transition from 50% to 100%
[13:11:28] <sendo> is it possible to get info directly from the loaded nc file where it says G1 Z-0.1 that the solenoid would go directly to 100%?
[13:12:14] <rayh> Yea but I've done that and don't like it much.
[13:12:47] <rayh> I color the backplot doing that and it sometimes doesn't work.
[13:13:06] <SWPadnos> you could set the Z axis accel insanely high, so the TP is likely to output 0 then -0.1 the next cycle
[13:14:03] <rayh> And use a very small input scale.
[13:14:15] <rayh> Like 5.
[13:14:31] <rayh> so that .2 is a 1
[13:14:48] <sendo> ok thats a nice idea
[13:14:50] <SWPadnos> also, with the wcomp idea, you should use two wcomps - one for each of the two forces you want, then use an or gate (and inverter I think) so that you always get the zero output when neither of the others is activated
[13:15:11] <rayh> There you go.
[13:15:38] <sendo> ok some more ideas to try with this info i should be able to make it work
[13:15:41] <sendo> thanks
[13:15:47] <SWPadnos> sure. have fun :)
[13:15:57] <rayh> Good luck.
[13:17:15] <rayh> Years ago we had some problems with very small input scale values. Anyone tested these lately?
[13:17:35] <SWPadnos> small as in not many counts per unit?
[13:17:43] <rayh> Right.
[13:18:02] <SWPadnos> ok. I think now that means very small encoder ticks :)
[13:18:19] <SWPadnos> oh no, not in the ini - nevermind
[13:18:22] <rayh> Few counts per unit
[13:18:53] <SWPadnos> scale is one of those things that has two perfectly good definitions, which are opposites of each other
[13:19:15] <SWPadnos> in any case - wasn't that usually due to differentiation of a very coarse step-wise input?
[13:19:25] <rayh> There were some small numbers in input scale that would produce an oscillator.
[13:19:33] <SWPadnos> I'd imagine the problems still exist for PID, but not necessarily for stepper machines
[13:19:50] <SWPadnos> but I don't know for sure
[13:20:24] <rayh> PID might have been the issue with freqgen.
[13:20:37] <rayh> One loop on one loop off.
[13:20:46] <SWPadnos> yep, could be
[13:27:46] <rayh> I don't see any problem with very small numbers in input scale with steppers.
[13:28:06] <rayh> But it was a quick test.
[13:33:55] <micges> hello
[13:34:15] <BigJohnT> hi
[13:44:54] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/Plasma%20Cutter/Plasma012.jpg
[13:45:15] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/Plasma%20Cutter/Plasma013.jpg
[13:45:31] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/Plasma%20Cutter/Plasma014.jpg
[13:45:55] <skunkworks_> no issues with noise?
[13:46:12] <BigJohnT> I just turn the radio up louder
[13:46:37] <BigJohnT> or do you mean from the plasma torch?
[13:46:55] <skunkworks_> :) torch
[13:47:33] <BigJohnT> no problem, Jethro Tull keeps the stray electrons in line :)
[13:47:40] <skunkworks_> sweet
[13:47:54] <skunkworks_> It is nice he does that for you..
[13:48:22] <BigJohnT> btw, I solved the lack of understanding of the touch off from yesterday
[13:50:15] <jepler> really? how do you solve "lack of understanding"? I have problems with that a lot.
[13:51:01] <BigJohnT> I kept beating my head against the Z axis till it sunk in
[13:51:02] <archivist> more beer?
[13:51:13] <BigJohnT> glass of homemade wine
[13:58:47] <skunkworks_> bunch of goof balls http://www.electronicsam.com/images/KandT/cncworkshop07/DSC_3749.JPG
[14:02:15] <ALS> Jethro Tull and the Band?
[14:06:13] <skunkworks_> rayh: http://www.electronicsam.com/images/KandT/cncworkshop/rayh.JPG
[14:07:43] <jepler> yes officer that's the man who stole my ice cream cone. im sure of it
[14:09:26] <sendo> hmm lyx takes a long time to install windows installs faster....
[14:09:30] <SWPadnos> ah - the good old days, when those shirts still looked new
[14:09:52] <SWPadnos> and that's Aram behind me, I think
[14:10:03] <SWPadnos> err - next to me
[14:10:42] <rayh> That old guy can't be me skunkworks
[14:13:36] <SWPadnos> it wasnt me - there isn't enough hair :)
[14:14:32] <rayh> I was wondering who belonged to that hand behind you. Suit jacket?
[14:15:12] <SWPadnos> no, just a sweater
[14:15:12] <archivist> * archivist note he's not the only hairy
[14:19:07] <skunkworks_> aram - I forgot he was at the 2006 fest also.. (first one I went to)
[14:19:43] <SWPadnos> oh hmm. I was thinking of last year, so that may not be him (but it does look like hum)
[14:19:46] <SWPadnos> him
[14:20:32] <skunkworks_> I vaugly remember him being there.. I think my father got cornered by him :)
[14:20:47] <SWPadnos> heh
[14:25:30] <skunkworks_> dave didn't notice his part was not scaled right... (z scaling on the mazak must have been off then also - caused problems when cradek tried rigid tapping the first time)
[14:25:41] <skunkworks_> http://www.electronicsam.com/images/KandT/cncworkshop/mazakpart.JPG
[14:27:00] <jepler> yes he was there in 06.
[14:42:24] <skunkworks_> archivist: I think it is a linux thing (hairy) http://www.electronicsam.com/images/KandT/cncworkshop/cncworkshopemc.jpg
[14:43:33] <SWPadnos> http://linux.ues.edu.sv/servidor/maracosas/bruno2d/richard-stallman.jpg
[14:44:37] <archivist> thankfully no pics of me online yet!
[14:46:51] <archivist> exept for ...http://ccgi.archivist.plus.com/php/showresult.php?searchv4page=1&searchstr=caroline&srcdata=title&srcprog=http%3A%2F%2Fccgi.archivist.plus.com%2Fphp%2Fsearchv4.php&dir=sgood&file=rim00077&subject=613
[14:48:21] <archivist> silly hat hiding the evidence
[14:49:12] <skunkworks_> that doesn't count :)
[14:50:48] <archivist> Im in an american amateur tv mag somewhere
[14:51:06] <archivist> about 10-12 years ago
[15:26:46] <BigJohnT> What a crew...
[15:27:22] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/hunting/HPIM0281.jpg
[15:28:50] <seb_kuzminsky> nice! with a front-loader no less
[15:29:00] <seb_kuzminsky> is that you?
[15:29:10] <BigJohnT> with electronic sparker
[15:29:12] <BigJohnT> yes
[15:29:45] <seb_kuzminsky> "we're eatin' tonight kids!"
[15:30:00] <BigJohnT> gonna have new coat this winter ma
[15:30:10] <seb_kuzminsky> heh
[15:42:03] <Vq^> BigJohnT: i like the backseat of your car :)
[15:42:45] <BigJohnT> lawn chairs and ice chest full of beer redneck tree stand on wheels!
[15:43:05] <rayh> Good looking huntin' shack
[15:43:54] <rayh> Guy up here uses a camo scissor lift.
[15:44:24] <skunkworks_> heh
[15:44:30] <BigJohnT> cool
[15:45:07] <BigJohnT> you don't even need camo where I hunt but you better not shoot one of the cows or you bought it...
[15:45:26] <rayh> Mich made elevated blinds legal a couple years ago.
[15:45:50] <rayh> camo is a macho thing.
[15:46:10] <BigJohnT> ha, my wife has more camo than I do!
[15:46:36] <BigJohnT> we sold her bass boat to by her more guns
[15:51:42] <alSMT_> besides you throw them bass back
[15:52:26] <BigJohnT> you got that right bluegill and red eared sunfish are much better
[15:52:42] <BigJohnT> we only keep the bass when we can't catch them
[15:53:31] <alSMT_> I like catching them better then northerns
[16:36:43] <fenn> BigJohnT: now that you've got a functioning plasma table you can make all sort of useless stuff like this: http://fennetic.net/pub/irc/25-hedron.png http://fennetic.net/pub/irc/Mesh.010.scaled.png http://fennetic.net/pub/camera/DCP_0804.JPG
[16:37:41] <BigJohnT> fenn: then what would I do with it once it was made?
[16:38:18] <fenn> well, the point is that you can flatten any arbitrary polygon mesh and cut it out, fold it
[16:38:36] <BigJohnT> I was looking at that, coo
[16:38:59] <BigJohnT> I could cut perfs along the bend lines
[16:39:03] <fenn> mesh_unfold.py comes with blender
[16:39:10] <fenn> outputs svg
[16:39:15] <BigJohnT> cool
[16:39:38] <BigJohnT> btw, did you know SheetCam was developing a Linux version?
[16:39:56] <alex_joni> cool.. really?
[16:40:30] <BigJohnT> yes, I've been e mailing the programmer about different things
[16:40:44] <BigJohnT> the reason for the TNG version is to port it to linux
[16:41:56] <alex_joni> http://www.sheetcam.com/featureDetails.php?id=18&func=show
[16:42:18] <BigJohnT> there you go
[16:42:35] <alex_joni> date submitted: 2006 ;)
[16:45:27] <BigJohnT> I talked to him yesterday but he didn't give me any idea of when the linux version would be ready
[16:46:00] <BigJohnT> so everyone here needs to vote on it!
[16:46:23] <fenn> eh, from my perspective it's an impediment to real progress
[16:46:45] <fenn> can you not run sheetcam in wine already?
[16:46:55] <BigJohnT> dunno
[16:48:55] <seb_kuzminsky> fenn: real progress == FOSS CAM?
[16:49:01] <fenn> yeah
[16:49:07] <seb_kuzminsky> amen brother
[16:49:12] <BigJohnT> what id foss cam
[16:49:13] <alex_joni> it's not FOSS
[16:49:15] <BigJohnT> is
[16:49:20] <fenn> ideally something that can be used algorithmically
[16:49:24] <SWPadnos> Free Open Source Software
[16:49:25] <seb_kuzminsky> Free/Open Source
[16:49:27] <SWPadnos> ==FOSS
[16:50:02] <fenn> let me tell you all a story
[16:50:07] <alex_joni> sheetcam is proprietary
[16:50:24] <fenn> one day a group of concerned citizens in my town decided to start an independent media center
[16:50:47] <fenn> we all got together, signed up on mailing lists, discussed the battle plan, and started _doing_ stuff. things were going swell
[16:50:49] <seb_kuzminsky> alex_joni: yes, that's what fenn & i dont like about it
[16:51:40] <BigJohnT> until someone/s creates a FOSS CAM then SheetCam is the most cost effective I have found
[16:51:43] <fenn> then, one day, a restaurant owner decided to fund his friends' pet newspaper, "the free press", which was actually just another newspaper
[16:52:40] <fenn> since it appeared to meet many of the requirements (except the freedom part) most of the people who were working on the real free newspaper lost interest, and the whole project failed (personally i wasn't interested in newspapers anyway)
[16:53:01] <seb_kuzminsky> BigJohnT: available and cost effective is good :-)
[16:53:15] <alex_joni> BigJohnT: I kinda agree..
[16:54:05] <fenn> but we still dont have an ad-hoc mesh network in bloomington!
[16:54:32] <seb_kuzminsky> fenn: i see your point
[16:55:28] <fenn> oh, and the other newspaper died a quiet death a couple weeks later
[16:57:57] <seb_kuzminsky> after i finish the hostmot2 driver and my mill conversion i want to concentrate on an open-source cam project
[16:58:39] <BigJohnT> for 2.5 or 3D?
[16:58:52] <seb_kuzminsky> start small and build up i think
[16:58:58] <seb_kuzminsky> i dont know the first thing about cam yet :-)
[16:59:25] <fenn> seb_kuzminsky: its ok, neither do CAM software programmers
[16:59:45] <BigJohnT> my pet peeve is when programmers try and make you do what they want and not what you want.
[17:00:29] <BigJohnT> I know enough about both to be real dangerous ...
[17:00:48] <fenn> like, uh, not giving you the source code so you can't make changes?
[17:01:13] <seb_kuzminsky> "That's the thing about people who think they hate computers. What they really hate is lousy programmers."
[17:02:25] <BigJohnT> you really can't fault a guy for not giving away his rice bowl if he makes a living on it
[17:02:35] <fenn> seb_kuzminsky: if you want to get up to speed on CAM check out anderswallin.net
[17:02:59] <BigJohnT> and take a look at what I've done with g code generators on wiki
[17:03:00] <SWPadnos> add to that the fact that it's nearly impossible to make substantive changes to a large piece of software, even when you do have the source
[17:03:13] <SWPadnos> (ever looked through the Mozilla source?)
[17:03:30] <fenn> SWPadnos: i'd argue that's a programming language problem
[17:03:36] <SWPadnos> I don't think so
[17:03:38] <fenn> or design philosophy (same diff)
[17:03:42] <SWPadnos> it happens with all languages
[17:04:21] <SWPadnos> I suspect it's more from the fact that feature creep can't result in complete reqrites very often if you have to pay for programmer time
[17:04:27] <SWPadnos> rewrites
[17:05:10] <fenn> well, there's something to be said for standardized/fixed interfaces
[17:05:27] <SWPadnos> for competing companies, that's often seen as a death wish
[17:05:32] <SWPadnos> if you can't differentiate, you lose
[17:05:34] <BigJohnT> yea, they never seem to fit
[17:05:51] <fenn> and now we have a thousand of every identical product, whoo
[17:05:58] <SWPadnos> I'm only pointing out that it's not a simple problem
[17:05:59] <fenn> all mutually incompatible
[17:06:11] <BigJohnT> but is nice to know what to expect when you press a button
[17:06:17] <SWPadnos> it's not true that having access to the source fixes all software problems
[17:06:39] <SWPadnos> it's also not true that *not* having access to the source fixes all problems
[17:06:47] <SWPadnos> so there's no easy answer
[17:06:51] <fenn> it's a symptom of flaws in human competition
[17:07:06] <SWPadnos> that could be
[17:07:49] <fenn> it's possible (and quite easy actually) to see the commonalities, define some standards.
[17:08:09] <fenn> our culture doesnt know how to do that yet
[17:39:19] <Jessica946> Hey folks, would .005" rollout be acceptable for PCB work?
[17:40:12] <archivist> thats a bit loose, depends on work to be done
[17:40:24] <Jessica946> hmmm...
[17:40:25] <Jessica946> ok
[17:41:27] <Jessica946> I'm thinking of using a foredom rototool hand piece as the basis for my spindle. That is their published rollout.
[17:41:27] <archivist> eg .1" pitch no tracks between pads, near enough
[17:41:28] <cradek> runout?
[17:41:39] <Jessica946> yeah, chris
[17:41:49] <skunkworks_> fenn: http://www.electronicsam.com/images/KandT/cncworkshop/fennpour.JPG
[17:41:52] <Jessica946> runout, roll out...
[17:41:55] <skunkworks_> Runout?
[17:41:57] <cradek> that would ruin the tools I use
[17:41:59] <Jessica946> heheh
[17:42:08] <Jessica946> .005?
[17:42:25] <SWPadnos> that's a lot
[17:42:29] <Jessica946> what do you use as a spindle, chris?
[17:42:32] <cradek> yes traces and separations are about .008 on the boards I make
[17:43:02] <cradek> it's just some ball bearing thing that I'm probably running way overspeed
[17:43:12] <skunkworks_> Jessica946: what are you using for a spindle?
[17:43:41] <Jessica946> skunk, i've got a crap spindle. I'm shopping for something better. that's why I'm asking
[17:43:44] <cradek> do you have a good size air compressor? I'd be tempted to try some of the air powered die grinders that run at 30-50 krpm
[17:44:05] <fenn> .005" is huge for a dremel
[17:44:07] <cradek> if nothing else, they are cheap and very light
[17:44:14] <skunkworks_> Or a dremel at 20k would probably work well
[17:44:26] <cradek> skunkworks_: jepler has had very bad luck with his (new) dremel
[17:44:32] <skunkworks_> Ew
[17:44:39] <skunkworks_> runoff?
[17:44:41] <skunkworks_> ;)
[17:44:55] <fenn> you could add a gyroscopic stabilizer :)
[17:45:08] <skunkworks_> We use a sears version of the 'rotozip' which seems to work well
[17:45:14] <fenn> a ring that press-fits over the carbide shaft
[17:45:27] <Jessica946> I'm thinking of having a seperate spindle from my drive source instead of an all-in-one thing like a dremel
[17:46:28] <fenn> looks like a lot of people use a regular motor (sewing machine motor for example) with an o-ring belt and pulleys to speed up a separate ball bearing spindle
[17:47:00] <fenn> that way the motor itself doesn't have to go 20krpm and so it cuts down on the noise a lot
[17:47:09] <Jessica946> ok, so re: the ball-bearing spindle, did you all make your own or did you buy someone elses part?
[17:47:11] <cradek> rick (o-matic) had a very nice small spindle at fest last year. I wish I knew where to get one.
[17:47:22] <cradek> it was totally silent at 30? krpm
[17:47:30] <fenn> rick made everything from scratch no?
[17:47:40] <cradek> I think the spindle was commercial
[17:47:59] <Jessica946> http://www.blackstoneind.com/foundations/store/shopdetail.asp?product=H%2E44T&ncluster=1&ngroup=2&name=H%2E44T++Handpiece%2C+General+Use&desc=Collet+type+handpiece+that+accepts+accessories+with+up+to+1%2F4%22+diameter+shanks%2E+Prelubricated+ball+bearings+require+no+maintenance%2E+Comes+with+3%2F32%22%2C+1%2F8%22+and+1%2F4%22+collets%2C+pin+and+wrench+for+changing+collets+and+accessories%2E%0D%0A%0D%0A+++++++%0D%0A%0D%0A+++++++%0D%0A+++++++%0D%0A+++++++&ima
[17:47:59] <skunkworks_> http://www.rickomatic.com/
[17:48:13] <Jessica946> This is what I was thinking of using as a spindle.
[17:48:37] <skunkworks_> that site sucks
[17:48:41] <fenn> skunkworks_: um, is it just me or is it just a picture of him?
[17:48:42] <skunkworks_> (rickomatic)
[17:48:43] <cradek> haha what a website
[17:48:45] <Jessica946> ???
[17:49:59] <fenn> quick disconnect handpieces?
[17:50:07] <archivist> worst website award 2008 rickomatic
[17:50:27] <OoBIGeye> archivist: i'm voting for it
[17:50:55] <skunkworks_> Now you know how to identify him at the workshop :)
[17:50:56] <Jessica946> fenn, yes, quick disconnect handpieces. I was thinking of modding one for a spindle. I'm here wondering if you all have a better idea.
[17:52:43] <fenn> no, not really
[17:53:02] <Jessica946> chris, did you build your own spindle?
[17:53:30] <cradek> nope
[17:54:08] <Jessica946> who's are you using?
[17:54:35] <archivist> * archivist scrapped a lathe for the spindle
[17:55:43] <lerman> On my list of things to use for a spindle motor is an "outrunner".
[17:55:58] <Jessica946> outrunner?
[17:56:18] <lerman> They are three phase motors with an incredible weight to HP ratio. Used in model planes.
[17:57:22] <lerman> http://en.wikipedia.org/wiki/Outrunner
[17:58:27] <fenn> outrunner motors have more leverage on the magnets so generally higher torque for the same motor size
[17:59:02] <fenn> i dont really see why that matters for a PCB engraver though
[18:00:31] <archivist> pcb engraver needs speed
[18:02:46] <lerman> http://www.maxxprod.com/pdf/HC6332-230.pdf -- 24.3 oz wt. 2200 Watts(max) 10,000 RPM
[18:03:41] <skunkworks_> yikes 30-80A Continuous
[18:04:21] <skunkworks_> Max RPM: 10,000 RPM
[18:04:51] <SWPadnos> also needs lots of cooling, I'd bet
[18:05:01] <SWPadnos> it's meant to be run outside, behind a high speed fan :)
[18:05:20] <fenn> liquid helium!
[18:05:39] <SWPadnos> or liquid metal ;)
[18:06:06] <lerman> For the DIYers, there are places on the web that sell part and have instructions on building your own outrunner.
[18:06:34] <fenn> SWPadnos: somehow i think that'd interfere with the operation
[18:06:56] <SWPadnos> yeah, that was more a result of not having enough liquid helium :)
[18:07:14] <fenn> is there an analogous effect for magnetic superconductivity? otherwise you'd still have iron losses
[18:07:39] <fenn> (unless you use a coreless motor)
[18:56:03] <rayh_> rayh_ is now known as rayh
[19:18:43] <BigJohnT> Gamma-X what's up buddy
[19:25:10] <Gamma-X> yo!
[19:25:24] <Gamma-X> whats up BigJohnT
[19:25:34] <Gamma-X> yo i just got home from a trip to irland
[19:25:42] <BigJohnT> cool
[19:25:56] <BigJohnT> you ever get your machine running
[19:26:39] <Gamma-X> nope
[19:26:44] <Gamma-X> cant get the whole pid thing
[19:27:38] <BigJohnT> darn, I got my plasma going
[19:27:40] <BigJohnT> http://www.youtube.com/watch?v=qlXCdfPsQUU
[19:27:50] <BigJohnT> http://www.youtube.com/watch?v=EC1_TTlsOrM
[19:29:25] <Gamma-X> damn dude
[19:29:28] <Gamma-X> thats awsome!
[19:29:32] <Gamma-X> how precise is it?
[19:29:32] <BigJohnT> thanks
[19:29:58] <OoBIGeye> nice work BigJohnT
[19:30:02] <BigJohnT> +- 0.005 more or less
[19:30:06] <BigJohnT> thanks
[19:30:09] <Gamma-X> wow
[19:30:12] <Gamma-X> that is awsome
[19:30:15] <Gamma-X> u make anything yet?
[19:30:21] <cradek> I want one
[19:30:22] <BigJohnT> the videos I was cutting the slats for the support
[19:30:37] <BigJohnT> so the machine was making the machine
[19:30:46] <OoBIGeye> sweet
[19:31:03] <OoBIGeye> how many amps is it in the torch?
[19:31:05] <BigJohnT> I have to cut some header flanges for corvetes from 1/2 plate
[19:31:09] <BigJohnT> 80 amps
[19:31:46] <OoBIGeye> sweet, then you can make som serious stuff...
[19:35:56] <BigJohnT> yep
[19:40:03] <skunkworks_> cradek: I think you have the technology..
[19:40:40] <BigJohnT> It's kinda simple really when you get down to the nuts and bolt
[19:43:43] <BigJohnT> cradek: you can borrow mine till you get yours built
[19:46:32] <BigJohnT> Gamma-X what did you do in ireland?
[19:46:40] <Gamma-X> got drunk
[19:46:47] <Gamma-X> went to the guinnes factory
[19:47:02] <Gamma-X> saw the book of kelles. made out with a chick from london
[19:47:52] <Gamma-X> went to the cliffs of moare
[19:48:00] <Gamma-X> rented a car and drove on the other side of the road
[19:48:22] <BigJohnT> cool
[19:50:46] <archivist> no you lot drive on the other side
[19:58:13] <OoBIGeye> archivist: only when drunk :)
[19:59:12] <BigJohnT> we drive on the right side you drive on the wrong side
[19:59:26] <BigJohnT> when I was little I had a right foot and a wrong foot
[20:00:30] <archivist> iirc you have trains on the left?
[20:03:57] <rayh> and I always started marching with the wrong foot.
[20:58:49] <BigJohnT> time to start drinking
[22:17:40] <dmess> hi all.any ne now what libatk is all about??
[23:40:18] <dmess> hi all