#emc | Logs for 2008-05-13

[00:01:20] <dmess> libatk1.0-0 violations..... im scared.. i llove livux...but :-! what to do now.... its good to have friends...:)
[00:06:25] <tomp2> "vio‭lations"? what did it say
[00:22:16] <dmess> Error: Dependancy is not saisfiable:libatk1.0-0
[00:23:05] <dmess> im a linux screw-be
[00:28:00] <dmess> with a pretty honkin' linux box too.... i can MAKe g-code for 5 axis stuff and EMC2 can run it if i get a little help from my friends.... post processor's are LawrnenceG's fortee
[00:28:02] <tomp2> i'm using ubuntu 6.06 lts right now & have libatk1.0-0 installed ( tho the version is listed by Synaptic as 'l1.11.4-0ubuntu1' ) so it >is< satisfiable (somehow)
[00:28:40] <dmess> where did you find that ver??
[00:29:08] <tomp2> try using Synaptic to install that file, and see what it reports ( it wont screw up anything to try, as Synaptic will stop shy of destroying your system....afik )
[00:29:41] <tomp2> i found it by opening Synaptic ( Syetm"admin|Synaptic then searching for 'libatk'
[00:30:06] <tomp2> i found it by opening Synaptic ( Syetm|admin|Synaptic then searching for 'libatk'
[00:30:46] <tomp2> i likely have universe & multiverse enabled as sources
[00:31:17] <dmess> same stuff
[00:31:27] <dmess> me too
[00:31:33] <tomp2> as repositories and yes i also have 'restricted' allowed
[00:31:59] <tomp2> then try 'update', then search
[00:40:26] <dmess> bbl... wife's home .... its fightin' time.... ;0
[00:42:19] <tomp2> again, best o luck ;)
[00:46:05] <dmess> it ain't luck....it's concentrated power and skill....if all i went in was with luck i'd have been outta dodge 5 yrs ago
[00:48:33] <dmess> still a downward spiral WITH altitude........ i can still save ME...and THE KIDS.....
[00:50:54] <Anastasia> what is it one runs to update library paths!?!?
[00:51:43] <Anastasia> sheesh, windoz solution initiated :).
[00:53:12] <tomp2> dang, do 'man ldconfig'
[00:53:41] <toastyde1th> DO NO SUCH THING
[00:53:46] <toastyde1th> !!!!
[00:53:53] <toastyde1th> i kid.
[00:54:18] <tomp2> i man (man ldconfig)
[00:54:36] <toastyde1th> hahaha zing
[00:54:53] <tomp2> (bored, soldering)
[00:55:11] <tomp2> (board soldering... see?)
[00:55:29] <toastyde1th> i wish i didn't.
[00:55:38] <toastyde1th> :D
[01:04:13] <fenn> K`zan: ldconfig
[01:19:18] <eric_U> I wonder if Cadence will import part libraries from eagle
[01:21:11] <eric_U> I was doing a design rules check on my part, and it gave me 3 error messages
[01:21:51] <eric_U> one for "documents", one for "and", and one for "settings" in my path
[01:22:26] <dmess> David Ball - Riding With Private Malone(1).mp3.. find that
[01:23:47] <dmess> cadance.... tell me if you get it to work...
[01:23:55] <eric_U> I'm almost there
[01:24:03] <jmkasunich> heh, I had "strata" of sawdust under the chute on my table saw - I was alternating between cutting tempered hardboard (dark brown) and pine
[01:24:19] <jmkasunich> cleaning up the pile - it had stripes, like sandstone at a road cut
[01:24:26] <eric_U> but the board layout software rejected all the parts I made myself
[01:24:53] <dmess> she will spark up TOO... be carefull
[01:25:08] <eric_U> probably not
[01:25:43] <eric_U> there have been reports of explosions, but they are incredibly difficult unless you are forcing small particulates into the air
[01:26:05] <jmkasunich> I've been forcing them into my shopvac
[01:26:09] <jmkasunich> no boom yet
[01:26:43] <eric_U> that would be a thump in your case
[01:27:31] <tomp2> the guy that did the 'chips' penguin, his mail list has been talking about using a mechanism for a satellite dish for an A B head. but they never mention the part, any ideas? sounds like a cheap way to get a 'knucklehead'
[01:27:42] <dmess> i can lite up coffee whithener... dont fack with FINE "particulate"... as i call it.. and dont BLOW UP
[01:27:48] <tomp2> rainea?
[01:28:01] <eric_U> yeah, but those things have gotten fairly rare
[01:28:25] <eric_U> it's a dc motor with a big gear
[01:28:57] <eric_U> he has a nice web site
[01:29:34] <tomp2> just a motor & worm? oh, i thought it was some azimuth, other thingy, thingy
[01:29:39] <dmess> perfect..
[01:29:48] <eric_U> the biggest explosion I ever was close to was a 8oz planter's peanut can full of acetylene
[01:30:10] <jmkasunich> acet-oxy mix I bet
[01:30:27] <jmkasunich> (if it was actually an explosion and not just poof
[01:30:28] <eric_U> not really, it was a pilot light
[01:30:43] <eric_U> it shook the building, probably just a very energetic poof
[01:31:12] <jmkasunich> pure acet (or any other fuel gas not mixed with air or oxy) just burns fast - foomp, not boom
[01:31:36] <dmess> hell... ive been 100 m from a full 18' x 24' drift.... PLUG YOUR EARS AND STAYBACK......
[01:31:55] <jmkasunich> drift?
[01:32:04] <eric_U> we had these torches that would leak acetylene so there was a pilot light and you didn't have to strike each time
[01:32:04] <dmess> a mine
[01:32:11] <jmkasunich> ah
[01:32:23] <eric_U> there probably was a little oxy in there
[01:32:31] <eric_U> but it was a yellow flame
[01:32:35] <dmess> not bench top stuff
[01:33:09] <jmkasunich> dmess: you get enough of it, and even a foomp turns into a boom
[01:33:43] <dmess> it was digiing 8 feet into the direction of the blast..
[01:35:15] <dmess> over-break.... 3-5 ft was sweet enuf for us.... straight DOWM
[01:35:22] <eric_U> that doesn't compare to a 8 ounce planter's peanut can
[01:36:10] <dmess> thru HARD ROCK... PLS
[01:36:53] <eric_U> anybody can do a large explosion with a large explosive
[01:37:10] <dmess> thats on top of an 8' drill and break
[01:38:10] <eric_U> I wonder if Trek keeps good enough records that they would be able to figure out I used to work there
[01:38:17] <dmess> we worke small and personal..... its mine in the morning
[01:38:53] <dmess> trek??
[01:39:12] <dmess> you mined?
[01:39:15] <eric_U> bicycle
[01:40:27] <dmess> im sure your name is good enuf to be where it should be...
[01:40:57] <dmess> i recall.... find me a 2nd
[01:41:10] <dmess> for cheap
[01:41:25] <dmess> proto or whatever...
[01:42:21] <dmess> id ride an old fat one these days......
[01:43:28] <eric_U> I was just reading some history of the company, talk about revisionism, whew
[01:43:39] <tomp2> not nice to say that about her
[01:44:55] <eric_U> for some reason they call this building a barn: http://blog.guykawasaki.com/2007/08/my-visit-to-tre.html
[01:45:21] <eric_U> it is a steel pole building that was attached to a concrete building that you can see in the background
[01:45:58] <dmess> seen it 2 x in the last 2 yrs... it sucks for us who have to make the changes
[01:46:11] <eric_U> I was in the concrete block half of the building when the peanut can blew, both buildings shook
[01:46:52] <dmess> was it a stress member??
[01:47:15] <dmess> the peanut can
[01:47:36] <eric_U> I think someone had it to put water in
[01:47:46] <eric_U> but it was upside down on the table
[01:50:19] <tomp2> just add water http://www.cannon-mania.com/bigbang.htm
[01:56:22] <eric_U> I have a cannon, never used it
[02:04:47] <toastyde1th> fffff
[02:12:35] <tomp2> g'nite
[02:32:58] <eric_U> it's funny having worked for a company that some people care enough about to keep history
[02:34:11] <steve_stallings> steve_stallings is now known as steves_logging
[02:34:20] <K`zan> fenn: Thanks, I use it so seldom anymore I forget :-).
[03:53:53] <toastyde1th> toastyde1th is now known as toastydeath
[08:29:40] <micges> hello]
[08:47:24] <pjm> good morning
[09:39:33] <alex_joni> 'lo
[12:30:17] <Guest316> Guest316 is now known as skunkworks_
[13:23:40] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/Plasma%20Cutter/Plasma012.jpg
[13:24:14] <BigJohnT> I have a micro switch on the floating head of my plasma cutter that is connected to probe in
[13:24:42] <BigJohnT> my plan was to use this for both touch off on plasma cutting and for centerpunching
[13:25:04] <BigJohnT> I've ran into a problem with centerpunching with a spring punch
[13:25:25] <BigJohnT> I get an error if I try to go in the same direction as the touch off
[13:26:10] <BigJohnT> any thoughts on this?
[13:26:16] <alex_joni> what error?
[13:27:24] <BigJohnT> I don't recall the exact verbiage but something like this probe all ready tripped
[13:28:01] <BigJohnT> for example when I touch off with the torch and go back up all works as expected
[13:28:25] <alex_joni> ah.. you can't move while the probe is tripped
[13:28:35] <BigJohnT> not down only up
[13:28:40] <alex_joni> you need to unhook it from HAL if you want that
[13:28:44] <alex_joni> BigJohnT: correct
[13:28:52] <SWPLinux> or use one of the other probe moves
[13:29:05] <alex_joni> SWPLinux: won't those halt immediately?
[13:29:15] <alex_joni> as the probe is hit?
[13:29:16] <SWPLinux> yes, but no error :)
[13:29:29] <BigJohnT> * BigJohnT looking at them
[13:30:44] <BigJohnT> ok I see a G38.3 gives you no error thanks SWPLinux
[13:31:09] <SWPLinux> sure. it also won't move though, so ...
[13:31:15] <SWPLinux> (if the probe is tripped)
[13:31:31] <SWPLinux> you can also probe looking for the input to deactivate
[13:31:35] <BigJohnT> with the centerpunch I need to find the surface then move 3/4" down more
[13:33:26] <BigJohnT> using G38.5?
[13:33:36] <SWPLinux> so g38.<xx> move up until probe not tripped
[13:33:41] <SWPLinux> then move down until tripped
[13:33:45] <SWPLinux> then G91Z-0.75
[13:34:15] <SWPLinux> if it's not tripped, the first one will perform no motion
[13:34:42] <BigJohnT> so the second one exits me from the probe routine?
[13:34:58] <SWPLinux> the first one insures the punch is clear
[13:35:06] <SWPLinux> the second brings the punch to the surface
[13:35:45] <SWPLinux> though it seems you'd be making all sorts of bad marks if the punch were dragging around on the surface :)
[13:35:58] <BigJohnT> I move it up after the punch
[13:36:05] <BigJohnT> to Z0
[13:37:07] <SWPLinux> is the switch getting stuck, or is it just that you may try to punch with the torch near the work?
[13:37:15] <BigJohnT> so all I need to do is move to my spot then G38.3 till I touch then G91z-0.75
[13:37:29] <BigJohnT> no the switch is good
[13:37:38] <BigJohnT> and the torch is off
[13:37:47] <BigJohnT> removed
[13:38:26] <SWPLinux> ok. I guess I don't understand why you'd get the error them
[13:38:28] <SWPLinux> then
[13:39:03] <BigJohnT> after probing the switch is still closed and I try to g91z-.75 and I get and error
[13:39:22] <SWPLinux> ah, right
[13:39:29] <SWPLinux> hmmm
[13:39:33] <BigJohnT> do I need to g91z.25 then g91z-1
[13:39:44] <BigJohnT> to reset the probe?
[13:40:01] <BigJohnT> then punch?
[13:40:19] <BigJohnT> I didn't have much time last night to try different options...
[13:40:22] <SWPLinux> I'm not sure how to get around that actually. when the probe input is tripped, you don't want to move down, or you'll break it
[13:40:45] <BigJohnT> hmmm, not in my case...
[13:40:49] <SWPLinux> actually, I have no time now to help out - gotta go get breakfast before getting on a plane
[13:40:58] <SWPLinux> see you later
[13:41:01] <BigJohnT> thanks and have a safe trip
[13:41:08] <SWPLinux> thankd
[13:41:09] <SWPLinux> s
[13:41:16] <Guest804> hello
[13:41:28] <BigJohnT> hi
[13:42:26] <Guest804> I have tried to use emc2 for a couple of years, but always the problem is hardware and unbuntu 6.06 live cd aren't compatable
[13:42:41] <Guest804> anything going on to update to a later version?
[13:43:33] <BigJohnT> my guess is if your hardware is so old that ubuntu won't work it's not going to get better...
[13:43:57] <Guest804> no the problem is the hardware is newer than 6.06
[13:44:11] <BigJohnT> however, on one machine I just changed the hard drive and all was ok and on another I had to change the video card and then all was ok
[13:44:43] <BigJohnT> I see its the other way around
[13:45:13] <SWPLinux> there's an 8.04 liveCD, but it isn't "official" yet: http://www.linuxcnc.org/hardy/ubuntu-8.04-desktop-emc2-aj07-i386.iso
[13:45:37] <BigJohnT> my guess is if you can load the newer version of ubuntu on your machine then when emc comes out on 8... you will be ok
[13:45:49] <SWPLinux> so test it out and report back :)
[13:46:40] <micges> hi all
[13:46:47] <Guest804> I'm using the java client to connect to irc and the url you posted isn't fully shown
[13:47:14] <SWPLinux> it should be clickable, but here's the directory: http://www.linuxcnc.org/hardy/
[13:47:22] <SWPLinux> http://www.linuxcnc.org/hardy/
[13:47:30] <SWPLinux> http://www.linuxcnc.org/hardy/ubuntu-8.04-desktop-emc2-aj07-i386.iso
[13:48:08] <SWPLinux> ok, now it's really time for breakfast
[13:48:41] <Guest804> got it will give it a try
[13:48:44] <Guest804> thanks
[13:49:05] <BigJohnT> can I unhook the probe input with g code?
[13:50:26] <BigJohnT> or do I need to use M66 for touch off and punching...
[13:52:36] <cradek> what do you mean unhook?
[13:53:23] <skunkworks_> digital output (m62-m65) with an and gate in hal?
[13:54:22] <cradek> oh I should have read back
[13:54:24] <cradek> ...
[13:56:16] <cradek> which error are you getting BigJohnT?
[13:57:19] <ALS> I loaded 8.04 on a Dell 566 cell. triple booting W2K & 6.06 320mb ram intel 810 e no hitchs exept I deleted a partition and needed to renumber my boot.ini partition thanks jepler and all
[14:00:08] <cradek> I don't see any problem with moving down after the probe trips.
[14:02:19] <BigJohnT> cradek sorry stepped away
[14:02:51] <BigJohnT> when I try to move down after touch off I get and error and can not move down
[14:03:41] <cradek> what error? move down using jog? mdi?
[14:04:04] <BigJohnT> using a g1 move
[14:04:07] <BigJohnT> <BigJohnT>I don't recall the exact verbiage but something like this probe all ready tripped
[14:04:53] <cradek> if the probe trips during a non-probing mdi move, the move will abort
[14:05:10] <cradek> but it sounds like it should have already been tripped in your case.
[14:05:25] <BigJohnT> yes it is tripped when I try to move
[14:05:58] <BigJohnT> should I try and move up then back down after the probe?
[14:06:14] <BigJohnT> I'm not at the machine now so I will have to test this evening
[14:06:18] <cradek> no, that will trigger the error for sure
[14:06:32] <cradek> if it stays in contact after the probe move, I think you should be ok
[14:06:44] <cradek> maybe you have a bit of noise on the switch. halscope would show for sure.
[14:07:16] <BigJohnT> I can check this evening...
[14:07:16] <cradek> bbl
[14:07:20] <BigJohnT> ok thanks
[14:25:08] <Jessica579> Morning all. I'm tryin to get my head around helical motion in 3-axes. if I'm at x1 y0 z0, and I want to make a one turn spiral ending at x1 y0 z1 with a radius of 2 units, the code would be "g2 x1 y0 z1 i-1 j0 k.5"? err... umm... right?
[14:26:46] <alex_joni> a helical motion is just like a regular circle
[14:26:53] <alex_joni> only that you add a Z-word
[14:27:09] <alex_joni> you don't need the K in the code you wrote
[14:27:33] <alex_joni> G2 x1 y0 z1 i-1 j0 <- should be enough
[14:27:53] <Jessica579> ok. what is the "k" used for if not that?
[14:28:03] <BigJohnT> Jessica579 take a look on the emc wiki site for the socket head cap screw g code generator. It outputs helical motion
[14:28:18] <Jessica579> thx, john
[14:28:33] <alex_joni> Jessica579: the K is used when you have an circle in the XZ or YZ plane
[14:28:47] <alex_joni> usually circles are in XY (so you only use I & J)
[14:30:13] <Jessica579> alex.... I think I get it. In my example, my z is a linear progression from z0 to z1. so no need for k. Now, if i wanted to cut a curved dome shape, I'd use "K". right?
[14:37:16] <SWPLinux> you'd use a lot of circles
[14:37:40] <cradek> yes for a helix, two axes move in a circle, the other moves straight. the two that move in circles can be any two; you use the appropriate I,J,K letters for those two
[14:37:57] <skunkworks_> Jessica579: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Oword#Sample_1_One_side_of_a_ball_in_cage
[14:38:05] <cradek> also you have to select the right plane (g17,18,19)
[14:44:32] <Jessica579> Ok. Thanks guys!
[14:54:38] <solexious1> solexious1 is now known as solexious
[14:56:33] <Guest156> Guest156 is now known as skunkworks_
[15:20:51] <alex_joni> this is really cool: http://openfarmtech.org/weblog/?cat=13
[15:46:41] <BigJohnT> they make it sound so simple...
[16:00:22] <alex_joni> BigJohnT: at least they use emc ;)
[16:00:34] <BigJohnT> yes i saw that
[16:03:36] <skunkworks_> where did you see that?
[16:04:06] <BigJohnT> in the section where they have a cnc torch
[16:16:50] <skunkworks_> boundry layer turbine looks interesting
[16:27:30] <fenn_> fenn_ is now known as fenn
[16:30:52] <alex_joni> bbl
[17:13:30] <sendo> hello is this where i can ask questions about setting up an emc system?
[17:24:02] <tomp2> sendo: yes, just ask away ( but most people here at USA night-time ) if anyone can help you, they will
[17:28:24] <BigJohnT> I resolved the G38 problem by just moving up and off the probe switch to finish the G38 move then I could Z down as far as I liked! SWEET!
[17:41:49] <fenn> for whoever is interested in solar-powered boundry layer turbines: http://dspace.mit.edu/bitstream/1721.1/37400/1/122931954.pdf
[17:42:17] <fenn> (and lots of 3d printer fun)
[17:44:37] <sendo> ok i´m converting an old LPKF mill (PCB mill) X and Y axis were no problem but Z axis is actually a solenoid and i´m kind off stuck on how to resolve that
[17:44:56] <sendo> i´m going to use PWM to achieve that
[17:45:17] <sendo> but how do i tell emc the axis is PWM driven?
[17:47:03] <fenn> it's just up/down?
[17:47:42] <fenn> you can use the 'comp' window comparator to turn a solenoid on if Z is within a certain range
[17:48:01] <sendo> i´m not sure i will have to experiment with that to get proper trackwidth and the LPKF info i have is inadequate
[17:48:39] <BigJohnT__> fenn: me?
[17:49:13] <sendo> i´m guessing that to drill a hole in a pcb it will use 100% duty cycle but for making the tracks i´m not sure
[17:49:47] <jepler> sendo: does the Z axis have any provision for "feedback", so that the position the solenoid has been driven to can be measured in the PC?
[17:50:05] <sendo> no feedback
[17:50:38] <archivist> does it have damping for feed
[17:52:01] <jepler> in emc, the way an axis moves is a black box -- emc produces a commanded axis value on the pin called axis.#.motor-pos-cmd and monitors a feedback axis value on a pin called axis.#.motor-pos-fb.
[17:52:06] <fenn> sendo: i think the depth of a drill is set deeper with the little plastic ring/collar on the bit
[17:52:41] <archivist> because most pcb drills would shatter if slammed into the board, there has to be a feed mechanism
[17:52:53] <fenn> archivist: i was thinking some kind of spring linkage
[17:53:07] <fenn> with a damper
[17:53:14] <jepler> some people hook stepgen in there, some people hook PID in there to control servers, but you can do anything that you can express in hal or with a custom hal component.
[17:53:37] <gefink> fenn: i use a modell servo for drilling PCB
[17:53:43] <gefink> but its bad
[17:54:19] <archivist> I would change the feed from solenoid to something else
[17:55:02] <jepler> seems like pcb drilling would also need to take place at a controlled speed
[17:55:18] <sendo> fenn there is a ring to adjust depth but its at 2 micrometer per notch or something i think that is to finetune trackwidth nothing more
[17:56:34] <sendo> there is however something i have never seen before it looks like a adjustable button with no wires attached as i said i really have nearly no info on the machine
[17:57:01] <fenn> take some pics and maybe we can help
[17:57:33] <fenn> the button is probably a damper
[17:57:33] <sendo> the "button" pushes in the Z+ direction
[17:57:45] <fenn> and return spring
[17:58:07] <sendo> yes i think so there is indeed a return spring i´ll upload a picture
[17:59:59] <archivist> hmm lvdt/opto and pwm to the solenoid may produce a workable servo
[18:00:19] <jepler> archivist: but without feedback?
[18:00:38] <archivist> with lvtd or opto grating
[18:01:07] <jepler> oh I should read what you're saying :-P
[18:01:09] <archivist> but if the damper is unidirectional
[18:02:42] <archivist> can the software servo deal with different damping rates dependent on direction
[18:03:28] <fenn> yes of course, you just can't (yet) do FF in one direction
[18:04:47] <jepler> in this specific case, having the drill or router bit snap back up quickly doesn't seem like it would hurt anything
[18:04:53] <jepler> or am I not thinking things through again?
[18:06:41] <sendo> http://img519.imageshack.us/my.php?image=img0669pb9.jpg
[18:07:36] <BigJohnT> dang internet is lumpy today
[18:09:24] <sendo> the solenoid is on the fixed part of the machine and puches the spindle down
[18:09:47] <sendo> the threaded thing is the damper i think
[18:15:29] <sendo> i think jepler is correct about snapping up there will be only 2 tools used a spiral drill and a universal pcb cutter (engraving needle)
[18:16:09] <sendo> http://img132.imageshack.us/my.php?image=img0675qe6.jpg
[18:16:18] <sendo> thats from another angle
[18:44:19] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[19:24:30] <tomp2> sendo: if the 'z axis' only has 2 positions ( up and down ),
[19:24:30] <tomp2> then stop thinking of it as an 'axis',
[19:24:30] <tomp2> begin thinking of your machine as XY with a special function ( plot on/ plot off ).
[19:24:30] <tomp2> The members here might think of a real positionable axis when you say 'z axis'. ;)
[19:24:30] <tomp2> To do this you need a special ( tho simple ) setup ( it has X Y but no Z ) and maybe an M function to do Plot on / Plot off
[19:24:52] <tomp2> it's just a word, but may help you get a solution ;)
[19:25:03] <BigJohnT> like pen down and pen up
[19:25:10] <tomp2> right ( HPGL)
[19:25:38] <sendo> thanks thats an idea i got to look into
[19:26:24] <BigJohnT> sendo look at M64/65
[19:27:30] <sendo> but i was thinking that i might have nothing to do with actual positioning but rather a difference in duty cycle press down soft for making tracks and press down hard for drilling
[19:28:03] <alex_joni> sendo: that won't work without feedback
[19:28:36] <tomp2> oh, if you want to do a 'floating head' then BigJohnT and Dallur have done similar for torches
[19:29:39] <tomp2> btw, nice dust collection collar on the unit
[19:31:40] <sendo> hehe its an old machine from the school i attend and i got to do an essay on PCB milling and when i asked to do some testing they said they didnt have a machine for it
[19:31:47] <gefink> sendo: you can send the y-coordinate using pwmgen
[19:31:53] <gefink> sorry z
[19:32:22] <sendo> a few days ago they said we found something but you got to fix it first.....
[19:32:48] <tomp2> gefink: sendo said "Z axis is actually a solenoid"
[19:33:29] <gefink> tomp2: yes he want pwm to move slow
[19:34:03] <sendo> yeah annyway i would like to thank you all for your input but i got to go now
[19:34:09] <tomp2> like close a relay slow with pwm? ok, cool
[19:34:11] <gefink> tomp2: i think on a voicecoil or how this is named
[19:34:30] <tomp2> yes voice coils use pwm, and are solenoids
[19:34:58] <sendo> i got the datasheet of the solenoid and it can take a dutycycle of 15%
[19:35:13] <sendo> 100-15%
[19:36:27] <tomp2> still need feedback ( and coil quality like voice coil motor ) you might make it into an r/c airplane like servo where it seekssome value ( like a sewing machine foot with pot tell it where the surface is ) but this is pretty custom
[19:37:04] <tomp2> if you can be sure where the surface is, then you can skip all this
[19:37:19] <sendo> well thanks everybody but really got to go now
[19:37:21] <gefink> tomp2: thats right. i think thats not worth
[19:37:27] <sendo> bye
[19:37:29] <tomp2> and mechaniclly set up & down
[19:37:33] <tomp2> bye have fun
[19:53:31] <BigJohnT> alex_joni: did you see that I solved the probe issue?
[19:55:38] <alex_joni> BigJohnT: briefly
[19:55:41] <alex_joni> but.. yay
[19:57:30] <alex_joni> heh.. nice name for a package "gstreamer0.10-plugins-ugly-multiverse"
[19:58:42] <alex_joni> * alex_joni heads to bed
[19:58:44] <alex_joni> good night all
[20:00:32] <gefink> good night
[20:00:53] <K`zan> night alex_joni
[20:10:47] <gefink> by
[20:11:50] <fenn> tomp2: i like the sewing machine foot idea, better than an IR noncontact probe (less noise and immunity to dirt/swarf)
[20:12:31] <fenn> there are those spring loaded engraving spindle-carriers but it seems like trouble to me
[20:13:31] <fenn> what kind of repeatability and precision can one expect from a linear pot?
[20:13:49] <fenn> i bet you need < 0.001"
[20:14:09] <fenn> over short distances
[20:14:20] <pjm> hi fenn, a linear pot should be reasonable but no where near what you would need for precision cnc
[20:14:39] <pjm> for a start they are usually fairly linear, but not 100% linear over their track
[20:14:41] <fenn> pjm: this is just for measuring the height of engraving surface
[20:15:11] <pjm> in that case it could give you an idea certainly
[20:15:21] <pjm> what sort of accuracy u looking for ?
[20:15:22] <fenn> so as long as it's locally continuous (no backlash or switch bounce)
[20:16:00] <lewin1> lewin1 is now known as lewing_
[20:16:13] <fenn> pjm: not entirely sure.. sixty degree cone, 0.005" features, somewhere around .001" i guess
[20:16:16] <pjm> plus of course the res would be limited to the resolution of the ADC measuring the output of the pot
[20:17:36] <fenn> can one use the term backlash in describing a sensor? or is that called hysteresis?
[20:17:45] <pjm> it might be easier to find one of the cheap ebay digital vernier based scales that has a serial output and modify that
[20:18:03] <fenn> ugh
[20:18:28] <fenn> for one thing, it's discrete steps of 0.0005", which will cause hell
[20:18:37] <fenn> and they're too slow
[20:19:02] <archivist> tooooo effin slow
[20:19:08] <pjm> how is the height of the table set? anything u could bolt an encoder on to?
[20:19:11] <fenn> anyone know exactly how slow they are?
[20:19:25] <fenn> i'd guess something like 1Hz
[20:19:34] <archivist> fenn i stuffed one on a lathe
[20:19:46] <archivist> too slow to work with
[20:19:57] <fenn> never used th serial connection, maybe the display is just slow
[20:20:27] <archivist> I think the electronics is slow to save battery power
[20:20:31] <fenn> pjm: this is a hypothetical machine, discussing sendo's options
[20:20:33] <pjm> i have one on the Z of my milling machine and it must update at around 10Hz at a guess, its quite responsive
[20:21:00] <fenn> he's retrofitting a plotter-thingy
[20:21:46] <pjm> ahh ok
[20:22:23] <pjm> i did find in my 'crap' an old Z axis off a multicam CNC plotter, only problem is the travel is only about 150mm or so
[20:23:12] <fenn> btw i think your mill's column is too skinny
[20:23:25] <pjm> my project mill?
[20:23:28] <pjm> with 100mm box?
[20:23:31] <fenn> yes
[20:23:52] <pjm> yeah i cant decide about that, still umm'ing and ahh'ing about what to do re the Z
[20:23:56] <fenn> eh, i guess either it's satisfactory or it isnt
[20:24:13] <pjm> the only concern i have at the moment is high frequency vibrations
[20:24:25] <pjm> caused by the lack of mass supporting the Z
[20:24:42] <archivist> basic spring of the col will be a pain
[20:24:54] <pjm> a big cast iron lump would be better but this is what I have to hand
[20:24:58] <fenn> yeah try sand, or sand mixed with gravel and oil
[20:25:09] <pjm> yeah that is a good idea
[20:25:20] <fenn> (toxic waste disaster when you want to move though)
[20:25:36] <pjm> i'm gonna use one of the 'little machine shop' mini-mill heads
[20:25:48] <fenn> ah i guess one could mix soap with the oil before mixing it into the sand
[20:25:49] <pjm> which isnt massive in terms of motor HP
[20:26:14] <pjm> so i'm not sure that would bend the actual box section
[20:26:30] <fenn> the ones with plastic gears?
[20:26:40] <pjm> i have the belt drive mod kit..
[20:26:45] <fenn> heh ok
[20:26:48] <pjm> i read all that stuff about the gears ;-)
[20:26:56] <K`zan> pjm: HF uMill?
[20:27:07] <fenn> same thing pretty much
[20:27:14] <pjm> its the same as http://www.mini-lathe.com/Mini_mill/Reviews/Homier/Homier_mill_p3.htm
[20:27:17] <fenn> your choice of mt3 or r8
[20:27:23] <pjm> MT3
[20:27:28] <K`zan> Yeah, that plastic gear crap went away REAL quick here to a belt drive.
[20:27:29] <pjm> only because its the same as my manual mill
[20:27:47] <pjm> but that particular mini-mill head looks quite interesting in terms of the motor controller
[20:27:48] <fenn> i think i'll go with mt3 also
[20:27:54] <pjm> since its PWM apparnently
[20:28:01] <fenn> or maybe mt2
[20:32:46] <pjm> well i'm collecting this mini-mill head on thursday so hopefully i'll have it running pretty soon
[20:32:57] <pjm> for the PWM I'm gonna try and get EMC to control it
[20:36:06] <pjm> K`zan, so u have one of the same type of mini-mill head as that one in the above link?
[21:31:46] <K`zan> pjm: Nope that is mines big brother, I have the micro-mill, that is the mini-mill.
[21:31:57] <pjm> ahh ok got ya
[21:32:10] <pjm> and what u think of the micromill head in general?
[21:32:21] <pjm> i've only be able to go on the reviews i've read of the mini-mill
[21:36:26] <K`zan> Original SUCKED, much better with the belt drive kit.
[21:36:46] <K`zan> Between the variable speed and the pulley system it works out pretty good now.
[21:37:12] <pjm> ah yes i read a lot of people got tired of replacing the plastic gears
[21:37:21] <pjm> also the noise of them was commented about quite a lot
[21:37:29] <K`zan> Plastic gears, what in hell where they thinking? I would have suspected they would have been offering a belt drive kit to make up for it, err, grub a few exta $$$.
[21:37:39] <pjm> well yeah exactly
[21:37:40] <K`zan> Noise wasn't horrid, but it did NOT sound well at all.
[21:37:57] <pjm> the annoying thing is that i looked in the UK for one of those mini-mill heads
[21:38:01] <pjm> and the cost was just a joke
[21:38:10] <K`zan> Overall the uMill sucks, but it is a hell of a lot better than no mill at all :).
[21:38:17] <pjm> less than 1/2 the price in the USA even taking the $/£ into account
[21:38:47] <K`zan> Interesting, guess it is set up for what they market will bear rather than cost+profit.
[21:39:07] <pjm> it was cheaper to buy in the US and buy a plane ticket to go and get it
[21:39:25] <pjm> so i'm going tommorow to atlanta to pick it up and coming back monday
[21:39:47] <K`zan> My new wireless antenna (bi-quad) gets me about 2x the number of sites I get with the stock antenna, but all weak and none open :-/.
[21:39:53] <K`zan> Ahh good old ATL :).
[21:40:01] <K`zan> Great place to be away from.
[21:40:20] <pjm> hehh i'm only going there as a stop gap to visit a ham radio fest in dayton
[21:40:51] <K`zan> Good, been a LONG time since I went to Dayton. Miss the flea market to be sure.
[21:40:59] <K`zan> de wv9k/7
[21:41:17] <pjm> ah cool, m0eyt here ;-)
[21:41:22] <K`zan> m0?
[21:41:27] <pjm> england
[21:41:42] <pjm> amazing how many hams are into machine tools too
[21:41:53] <K`zan> Live in an apartment with high voltage power lines and an industrial park across the street, so I am not real active as you might guess :).
[21:42:00] <pjm> near where i live the local hams have either a lathe or mill in their garages
[21:42:20] <K`zan> Ah, thought that was G, guess enough folks coming aboard they had to expand.
[21:42:25] <archivist> ew /me was licensed
[21:42:34] <pjm> yeah it used to be G but apparently it ran out...
[21:42:49] <K`zan> Here I know no hams at all. NEver get out and FM I just have a real hard time with.
[21:43:02] <pjm> but anyway one of the drivers behind making a CNC is to make parts for microwave receivers
[21:43:18] <pjm> that need a fair amount of precision, things like scalar feeds and conical feeds etc
[21:43:21] <K`zan> Tried to get active on 6m, 2m and 70cm ssb/cw but antennas are a problem :-(.
[21:43:37] <K`zan> The higher in frequency the better :-).
[21:43:59] <fenn> pjm: wtf you are flying to the US to get a chinese mini mill part??
[21:44:01] <archivist> pjm I used to do 10g tv
[21:44:04] <pjm> yeah exactly.. my lowest freq band is 2.3GHz and higest 20.2GHz
[21:44:04] <K`zan> HF holds almost no interest for me, I can do that on my cordless or cell phone :).
[21:44:15] <pjm> fenn, cos the cost in the UK is about 4X that of the USA
[21:44:18] <K`zan> .me envies pjm
[21:44:23] <pjm> so its cheaper to combine it with a trip ;-)
[21:44:25] <K`zan> * K`zan envies pjm
[21:44:31] <pjm> hehh
[21:44:41] <fenn> that's just so wrong
[21:44:49] <pjm> well that is rip off britain for you
[21:44:53] <K`zan> I stuck on the DC bands, highest is 70cm...
[21:45:16] <pjm> it was cheaper to buy a plane ticket and physically go and get it, than buy it from a UK outlet
[21:45:24] <fenn> can't you just have it shipped?
[21:45:29] <K`zan> Well out of date, but: http://wrlabs.shacknet.nu/~vw/hamradio/
[21:45:48] <pjm> fenn, yeah i could have it shipped, but 40.5lbs to the UK aint cheap
[21:45:59] <pjm> it was about 2/3 the price of the plane ticket...
[21:46:22] <pjm> plus i have friends in atlanta so its an excuse to go and visit them for a catchup and some beers
[21:46:24] <K`zan> I should get off my arse and go screw with the pipedream...
[21:46:43] <archivist> * archivist waits for the duty and vat charges on pjm's return
[21:46:59] <pjm> hmm but it was 2nd hand and i have the receipt to prove it!
[21:46:59] <K`zan> or be screwed by the pipedream, as the case may be :).
[21:47:15] <fenn> K`zan: have you cut anything with it yet?
[21:47:18] <K`zan> LOL, give the gov what it gives you so well.
[21:47:36] <pjm> K`zan, my radio pages are at www.uhf-satcom.com
[21:48:06] <archivist> pjm I imported my 100meg to 26gig sig gen from the states
[21:48:06] <K`zan> fenn: Ran the EMC program it defaults to with a pen in the chuck, I got a LOT of work to do as far as alignment and strengthening goes.
[21:48:16] <K`zan> pjm: I'm afraid to look :)
[21:48:18] <pjm> archivist, handy item...
[21:48:26] <archivist> yup sure is
[21:48:44] <archivist> and some network analyser bits
[21:48:55] <pjm> useful indeed..
[21:49:08] <pjm> i only have a few bits of test gear here, mostly old stuff
[21:49:16] <pjm> well 1990's vintage
[21:49:35] <archivist> 1990's thats new
[21:50:36] <pjm> but yes hopefully when i get back with this mill head in one piece
[21:50:54] <pjm> it should only be a matter of weeks before the machine is in a state to start cutting metal
[21:50:58] <K`zan> wxtrack only wincrap :-(.
[21:51:17] <K`zan> Too many nice pix there :).
[21:51:21] <pjm> ah yes that is a problem, it does run under wine..
[21:51:23] <K`zan> envy...
[21:51:58] <K`zan> Cool, will give it a go, have my ancient xp under vmware, but I do so loath to run anything wincrap.
[21:52:26] <pjm> ah btw i have a Q on the EMC live cd perhaps someone will know the answer too
[21:52:34] <pjm> its regarding services that run at boot time
[21:52:42] <K`zan> Gotta go make a hole to get to work on the pipedream.
[21:52:50] <K`zan> laters es 72/73 :-)
[21:53:01] <pjm> and i noticed from looking at the output of dmesg that stuff like bluetooth is started, so how can i stop all that stuff i dont need?
[21:53:19] <pjm> should i edit the /etc/rcx.d/whatever or is there a tool to do that?
[21:53:23] <K`zan> I wish it would support my wireless card...
[21:53:38] <K`zan> that damn SysV init crap is insane...
[21:53:46] <pjm> yeah i'm a slackware guy too
[21:53:53] <pjm> so these modern OS's are a pain ;-)
[21:54:03] <ds2> so ditch the %@##@%#$%$@#%$ distros
[21:54:07] <ds2> write your own rc scripts
[21:54:09] <K`zan> It used to be better, but some damn geek had to wave his weenie to get it in, sigh.
[21:54:30] <ds2> use whatever init you want
[21:54:32] <K`zan> ds2: In my copious free time, right? :-)
[21:54:42] <archivist> pjm did you ever go to the batc events about 10 years ago
[21:54:43] <ds2> it is quicker then dealing with that crap
[21:54:53] <ds2> <-- did it once for SLS
[21:54:57] <pjm> archivist, no not really been into the ATV stuff
[21:55:03] <K`zan> ds2: I can't argue that....
[21:55:05] <pjm> archivist, mainly uWave etc
[21:56:00] <fenn> there is likely some silly gnome/gui tool to edit which services get started
[21:56:04] <archivist> pjm I was doing uWave testing at the annual rallies for a couple of years
[21:56:18] <pjm> archivist, this was a nice story on /. http://science.slashdot.org/article.pl?sid=05/12/19/041205
[21:56:24] <ds2> think on the F'ed stuff, sysconfig had something for it
[21:56:26] <pjm> archivist, ah that is interesting... perhaps I know you
[21:56:58] <pjm> fenn, yeah i looked at that 'boot up manager' but havnt installed it as yet, but it seemed to be a handy UI for editing boot time services
[21:58:31] <pjm> ok i need to go, gn all cul
[21:58:43] <archivist> seeya
[21:58:49] <pjm> cheers
[22:05:12] <rayh> 6.06 silly gnome/gui tool 1 -> System ->Preferences->Sessions is where I put an emc --(ini) so my machine starts on boot up.
[22:05:25] <rayh> But then I also autologin
[22:10:06] <rayh> tool 2 -> System->Administration->Services also has a few we might not need.
[22:13:16] <HighOctane> I have emc2 installed (used live cd for install). I want to install vismach. Anyone here that can walk me through that real quick?
[22:14:10] <rayh> tool 3 System ->Administration->System Monitor. Then View All Processes.
[22:14:25] <rayh> But I still don't see bluetooth
[22:15:14] <SWPadnos> what are these tool 1 / tool 2 / tool 3 things you're referring to?
[22:16:21] <rayh> 6.06 silly gnome/gui tool
[22:16:38] <rayh> fenn, thought there might be such tools.
[22:17:01] <SWPadnos> oh, the services manager also, but that's more for apache / sshd etc
[22:17:12] <SWPadnos> that would be too l2 :)
[22:17:15] <SWPadnos> tool 2
[22:17:44] <rayh> Sure. I'm not seeing anything quite like the init.d manager from some other distributions.
[22:18:42] <SWPadnos> hmmm. I saw one on some Ubuntu, but it might be a package I installed
[22:18:57] <SWPadnos> had columns for the runlevels and services/daemons as rows
[22:19:13] <rayh> Seems like I had one of those a while back also.
[22:20:13] <HighOctane> Excuse me. Could someone answer a few quick questions?
[22:20:23] <SWPadnos> yes
[22:20:32] <SWPadnos> you have two more questions. :)
[22:21:24] <HighOctane> lol... First, I want to install vismach. How do I go about that. I mean... where does vismach.py belong, etc...
[22:21:43] <SWPadnos> I can't answer that one. next!
[22:22:22] <HighOctane> Where does emc keep all of its files (executables)?
[22:22:40] <SWPadnos> (I think vismach can be more or less anywhere, but it may need to know the ini file or something)
[22:22:48] <SWPadnos> installed or run-in-place?
[22:23:25] <HighOctane> I don't know what you mean, installed or run-in-place. I have installed emc2 on my hard drive (i installed from the livecd
[22:23:33] <SWPadnos> ok, that's "installed"
[22:23:49] <SWPadnos> you can find out where all the files are by running the command `dpkg -L emc2`
[22:24:02] <SWPadnos> I believe configs are in /etc/emc2/
[22:24:08] <SWPadnos> binaries are in /usr/bin
[22:24:23] <SWPadnos> manuals in the correct man dirs ...
[22:25:34] <SWPadnos> gotta run for a bit. see you later
[22:25:41] <HighOctane> thanks.
[22:44:56] <jepler> HighOctane: if you use the sample configuration "scara" you will get a configuration which uses vismach to display a representation of a machine
[22:46:38] <jepler> HighOctane: in the file /usr/bin/scaragui (it's a text file; a program written in the python language that uses the vismach module) you can read the description of the machine. In the file scara_sim_4.hal you can see how the "joint-pos-fb" values are connected to pins of the "scaragui"; this is how the visualization knows what position to display
[22:47:37] <jepler> HighOctane: if you want to write your own machine description, you can use scaragui, hexagui, and pumagui as examples, and see a list of available primitives by executing 'pydoc vismach' at the commandline.
[22:47:37] <HighOctane> O.K. - Trying it now.
[22:48:00] <jepler> HighOctane: outside of what you see in pydoc there's essentially no other documentation.
[22:48:10] <HighOctane> Will the scara config run on a virtual machine?
[22:48:41] <jepler> yes, it should run OK (though you'll get the "unexpected realtime delay" messages)
[22:50:59] <HighOctane> It runs really slowly. Like 2-3 frames per second.
[22:54:32] <jepler> it uses opengl; if you are in a setup without hardware-accelerated opengl it is likely to not be fast
[22:54:55] <skunkworks> I bet glxgears runs really choppy also
[22:55:18] <toastydeath> all those sentences would have made great your mom jokes
[22:58:04] <skunkworks> Yo mamma?
[22:59:50] <toastydeath> "i gave your mother an unexpected realtime delay last night"
[22:59:51] <toastydeath> etc
[23:00:34] <skunkworks> you're weird :)
[23:00:34] <jepler> looks like simple matrix math is patentable. well, who here is surprised at that. http://www.google.com/patents?id=k1eAAAAAEBAJ
[23:01:33] <jepler> (or if it's not the matrixes that are patented, it's "a method of putting multiple pieces of electronics in a single plastic shell"
[23:01:42] <jepler> )
[23:01:49] <skunkworks> heh
[23:07:30] <archivist> USA patent system needs a kick in the bits
[23:08:50] <tomp2> put the patent lawyers in front, ( come de revolution)
[23:15:05] <dmess> VAPT is all matrix math
[23:16:13] <HighOctane> Jepler: That's pretty cool, vismach is. I gotta get that on a real machine, with a good graphics card. What do you recommed. 512 VRAM and 2 gigs ram, or what?
[23:16:28] <dmess> well ATI have done it with the all-in wonder series..
[23:17:13] <skunkworks> ^ have had good luck with the ati rage and 128 for opengl
[23:17:25] <skunkworks> (old stuff)
[23:17:51] <dmess> i live on the old ATI cards...
[23:18:32] <dmess> always say unsupported but sometimes shit just works
[23:19:02] <dmess> the remote works on the linux box... but we lots it ....
[23:19:43] <dmess> i never did get the tv-in working... but who cares
[23:33:57] <jepler> HighOctane: well, for running emc and controlling real hardware, you want a simple graphics card and you don't want to install the proprietary opengl libraries (e.g., nvidia)
[23:34:42] <jepler> HighOctane: for visualization but not running real hardware, any of the chipsets with accelerated opengl support will work nicely (intel, ati, and nvidia being the majors there)
[23:35:36] <jepler> almost all users report that installing the nvidia closed-source drivers mean they get constant "unexpected realtime delay" messages, and often they find that after uninstalling nvidia they can't get any opengl programs to work anymore without major surgery
[23:38:42] <skunkworks> Even the open source NV driver has been known to cause problems. Might need to use the generic vesa driver
[23:54:52] <Guest863> hey guys quick question, gecko's want an active low signal right? can this be done by just inverting the parport step pin?