#emc | Logs for 2007-02-21

[00:04:37] <a-l-p-h-a> neither am I.
[00:05:00] <a-l-p-h-a> vectors make things easy to do... drawing on the computer is much easier than on paper...
[00:05:04] <a-l-p-h-a> ctrl-z to undo. :D
[00:43:40] <Jymmmm> If anyone cares.... http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93977
[00:46:10] <SWPadnos> that won't be too useful for pins that go into connectors
[00:46:20] <SWPadnos> that looks like a faston / bullet crimper
[00:46:35] <Jymmmm> No, but ratching stack on connector crimpers are not easy to come by for that price.
[00:46:41] <Jymmmm> stake
[00:46:52] <SWPadnos> that's not a Stack On
[00:47:09] <Jymmmm> stake on, quick connect, etc
[00:47:22] <SWPadnos> Stack On, the company that would like you to confuse them with Snap-On ;)
[00:47:32] <Jymmmm> heh
[00:47:36] <SWPadnos> yes - faston == quick-connect, I think
[00:47:40] <Jymmmm> yeah
[00:48:15] <SWPadnos> it even looks like it can take dies for other crimp types
[00:48:20] <Jymmmm> yep
[00:48:40] <Jymmmm> I might even pick up two of them. if I ever get a mill, I could make my own dies
[00:49:25] <Jymmmm> leavign for HF in 10 minutes
[00:49:58] <SWPadnos> I was going to say - get me one, but that was when I thought it might be useful for pins for MIL-C or D-Sub connectors
[00:50:16] <Jymmmm> seeing if they might have the pin crimpers I really need.
[00:50:25] <Jymmmm> SWPadnos well, do you want one?
[00:50:43] <SWPadnos> no thansk, I have a Lucent interchangeable die ratcheting crimper :)
[00:50:54] <SWPadnos> (eBay to the rescue)
[00:51:09] <Jymmmm> for $10, you can't go wrong having a spare imo
[00:51:25] <Jymmmm> not like you won't use it
[00:51:34] <SWPadnos> I won't use it, I tell you!
[00:52:00] <Jymmmm> ah, ok. I know I will.... I use them all the time. and my old crimpers are getting up there in usage.
[00:52:17] <Jymmmm> only been used and abused for the last ten years heavily.
[00:52:26] <SWPadnos> how far away from you is HF?
[00:52:46] <Jymmmm> 20 miles, but work is 4 miles from HF =)
[00:53:02] <SWPadnos> oh, that's convenient
[00:53:16] <SWPadnos> I was thinking it might be fun to go there
[00:54:01] <Jymmmm> SWPadnos: Yeah we can. I just like to go before work, instead of on my days off.
[00:54:15] <SWPadnos> then again, if it's anything like the "Homier tool acutions", I'd probably get sick
[00:54:23] <SWPadnos> auctions
[00:54:36] <Jymmmm> SWPadnos store doens't carry EVERYTHING that website has.
[00:54:58] <SWPadnos> yeah - Fry's or the AMD dumpster might be betteer for a guy like me :)
[00:55:06] <Jymmmm> harborfreight.com is the main site, harborfreightusa.com is the retail stores.
[00:55:18] <Jymmmm> AMd dumpster/campus is behind a 10 ft fence
[00:55:35] <Jymmmm> easy to get into intel than amd
[00:55:43] <SWPadnos> yeah, but who wants to?
[00:55:53] <Jymmmm> heh
[00:55:55] <SWPadnos> how about one of those palettes of "tested" hard drives?
[00:56:02] <SWPadnos> thrashed, I mean
[00:56:25] <Jymmmm> heh, you and me both
[00:56:34] <SWPadnos> did you see the article on Slashdot about Google's report on hard drive reliabilit?
[00:56:40] <SWPadnos> reliability
[00:56:45] <Jymmmm> nope
[00:56:59] <SWPadnos> oh - it's interesting. something you may want to peruse at work
[00:57:22] <Jymmmm> then wait till I get to work in about 120m
[00:57:35] <Jymmmm> I gots to go.... laters!
[00:57:36] <SWPadnos> it's many pages long, but they go into detail about what kinds of failure rates they've seen, what SMART indicators actually show impending doom, ...
[00:57:38] <SWPadnos> see you
[01:03:43] <jepler> SWPadnos: the stuff the drive manufacturers probably know too, but aren't telling us
[01:04:01] <SWPadnos> heh
[01:04:10] <SWPadnos> yes, but they'd tend to know only their own drives that well
[01:04:42] <SWPadnos> the Google report doesn't actually mention manufacturer names - they weren't statistically different from one another
[01:05:15] <SWPadnos> or at least, the differences were well below that of drive age (for instance)
[01:36:46] <skunkworks> so my companies owner had a hunting museum. The place is huge. They have wildblue sat internet in one of thier kids bedroom. There is a stair way that goes down to the museum part. They wanted internet at the museum office. no cables - took 2 linksys repeaters to get from the bedroom to the bottom of the stair well. and it still didn't make it to the office.
[01:37:35] <crepincdotcom> aight im sitting here with my motor/encoder and an oscope
[01:37:48] <crepincdotcom> i have one output that gives a varying period square wave
[01:37:51] <crepincdotcom> but only one.
[01:38:14] <crepincdotcom> all the other pins are either tied to low or high, or seem to have a tiny sine on them, i'm guessing interferance from the motor coils
[01:38:31] <crepincdotcom> will that sort of encoder still work, if EMC assumes that the motor will go the correct direction it tells it?
[01:38:40] <skunkworks> no
[01:38:44] <crepincdotcom> :-(
[01:39:19] <jepler> imagine EMC wants the motor to stand still, but some external force causes it to turn one way or the other
[01:39:23] <jepler> how can EMC know which it was?
[01:39:26] <crepincdotcom> ah
[01:39:28] <crepincdotcom> point taken
[01:39:42] <skunkworks> sorry - what jepler said :)
[01:40:13] <crepincdotcom> im probing around inside now, maybe there are other places i can read it with the quadrature business
[01:40:13] <jepler> take a look at the inputs to the comparators -- are they swinging or are they stuck?
[01:40:28] <jepler> it could be that 3/4 of the comparators are fried
[01:40:34] <jepler> who knows
[01:40:52] <skunkworks> have you looked at any of the other ones?
[01:41:06] <crepincdotcom> nah, this is the first motor i pulled
[01:41:15] <crepincdotcom> the comparator is an LM339, testing outputs now
[01:41:47] <crepincdotcom> first && nicest btw
[01:43:01] <skunkworks> every encoder that I have gotten cheap with lm339's in them have been doa.
[01:43:07] <skunkworks> but that is just me.
[01:44:10] <crepincdotcom> the encoder itself or the 339, cuz i can easily replace a few 339s
[01:44:33] <jepler> there's another IC besides the 339?
[01:45:28] <skunkworks> the total encoder - the last one had surface mount 339's. Not worth it for me.
[01:45:31] <crepincdotcom> not on this one, no
[01:45:56] <crepincdotcom> aha! i have 2 squares
[01:46:04] <crepincdotcom> putting them on separate channels to compare nowe
[01:47:19] <crepincdotcom> yay! it works!@
[01:47:59] <skunkworks> great!!
[01:48:03] <jepler> second encoder, or did you "fix" something?
[01:49:19] <crepincdotcom> i went straight to the board with the scope, instead of the cable and such connected to it
[01:49:30] <crepincdotcom> ill poke around later, something funky with the cable
[01:49:51] <jepler> ah
[01:49:59] <crepincdotcom> now: the two squares both vary their size (period), but keep the same relationship to each other, as opposed to changing the offset
[01:50:12] <crepincdotcom> is that proper quadrature, or are the waves supposed to move?
[01:50:20] <crepincdotcom> s/move/change offset
[01:50:23] <jepler> they should be offset about 90 degrees
[01:50:26] <jepler> try another comparator output
[01:50:27] <crepincdotcom> they are
[01:50:53] <crepincdotcom> but they stay at that offset the entire time then, its the period not the offset that gives rotational speed?
[01:50:56] <skunkworks> that sounds right - they will always stay 90 degrees out of phase
[01:51:00] <crepincdotcom> ah ok
[01:51:05] <crepincdotcom> thats what i was asking
[01:51:06] <crepincdotcom> great
[01:51:15] <jepler> if you rotate CW it should be A leading B by 90 degrees, and if you rotate CCW it should be B leading A by 90 degrees
[01:51:37] <crepincdotcom> if i cant figure out which is which, can i invert it in EMC later? ie, SCALE -n
[01:51:44] <jmkasunich> yes
[01:51:48] <crepincdotcom> ok
[01:52:15] <jmkasunich> you can also swap the wires from A to B and that will reverse the direction
[01:52:26] <crepincdotcom> true, i guess that's just as easy
[01:52:52] <crepincdotcom> well thanks again all, I have to keep working on this paper
[01:52:58] <jmkasunich> darned schoolwork
[01:53:01] <crepincdotcom> can't use the schools things if I fail out ;-)
[01:53:25] <jmkasunich> good point
[01:53:27] <jmkasunich> get busy
[01:53:31] <crepincdotcom> heh
[01:53:34] <crepincdotcom> later guys
[03:23:39] <ejholmgren> badgers badgers badgers badgers ...
[03:42:45] <jtr> Anyone still use Metrabyte digital I/O cards? I've got some, but they seem to be obsolete.
[03:43:19] <jtr> These are the old Metrabus boards - MEM-32, MIO-32, and MII-32.
[03:54:20] <jtr> probably not worth writing a driver for, but if you're trying to fix an old machine that uses them...
[04:10:46] <Jymmm> lo
[04:41:09] <tomp> jmkasunich: maybe your 50 pin breakout is phoenix contact FLKM50, would be number on rear, expensive & nice, used by compumoter et others
[04:46:11] <mic50k> hey my nvidia 5200 gfx sucks with emc2. any suggestions?
[04:46:47] <Jymmm> tomp I hit HF today, they show 9 in stock, but couldn't find em - bastidges! LOL
[04:47:27] <tomp> arf! order one? ask for free shipping ;)
[04:47:44] <Jymmm> they'll give free shipping just for asking?
[04:48:07] <tomp> i mean tell them they were a pita and should make it up to a willing customer
[04:48:55] <Jymmm> Ah, the sympthaty bs, gotcha =)
[04:56:36] <ejholmgren> I ordered 20 or so of those $.99 casters from HF
[04:57:04] <ejholmgren> 2 weeks later a card comes in the mail saying that I can cancel the order cause they were on backorder
[04:57:17] <ejholmgren> so I did .... and the came in the mail the next day
[04:57:20] <Jymmm> yeah, that part always sucks
[04:57:22] <Jymmm> lol
[04:57:43] <ejholmgren> in the meantime ... I had alraedy bought 20 more at a local surplus store
[04:58:10] <ejholmgren> bastards
[04:58:35] <ejholmgren> so now I have to figure out 5 items at home that I want to put wheels on :)
[04:58:52] <Jymmm> ejholmgren" The Wife's side of the bed?
[05:00:00] <ejholmgren> I don't want to waste _all_ of them
[05:00:10] <ejholmgren> I kid, I kid
[05:00:23] <ejholmgren> ... but she will probably still kill me later
[05:00:42] <Jymmm> Ok, the wife's shoe collection
[05:14:16] <ejholmgren> the puppy fixes that problem :)
[05:14:48] <Jymmm> heh
[05:15:37] <ejholmgren> "4 GB May Be Vista's RAM Sweet Spot"
[05:15:39] <ejholmgren> O_o
[05:15:53] <Jymmm> dont know, dont care =)
[05:15:56] <ejholmgren> that's heinous
[05:16:12] <ejholmgren> good thing I stopped at win2k
[05:19:32] <jmkasunich> good thing I stopped at w95
[05:26:34] <Jymmm> have you seen this? http://runkel-it.de/download/steppercard.zip
[05:27:56] <Jymmm> oh it's unipolar, nm
[05:30:35] <ejholmgren> I still have yet to understand what exactly Windows ME was supposed to be ...
[05:31:24] <ejholmgren> does anyone know where I can find reference to using mdi?
[05:33:08] <Jymmm> pdf is the closest clue I could give
[05:33:55] <ejholmgren> just found it, thanks
[05:48:11] <jmkasunich> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TweakingSoftwareStepGeneration
[05:48:37] <Jymmm> bbiab
[05:52:55] <tomp> computabuser ( latency test )
[08:02:31] <K`zan> Night all.
[08:04:57] <Jymmm> boo
[08:10:20] <alex_joni> woo
[08:11:04] <Jymmm> I just got some spare rails, think I'll make a hot wire foam cutter
[08:11:39] <Jymmm> I already have 2 x 6' acme rod
[08:12:28] <Jymmm> Just ned to figure out hot to make a digital temperature controller
[08:15:15] <ds3> rip apart a soldering iron
[08:16:12] <Jymmm> and that gets me digital how?
[08:16:42] <Jymmm> digital , as in PC control.
[08:17:32] <ds3> they make digitally controlled irons
[08:18:49] <ds3> think a few issues ago, Circuit cellar had an article on using the heating element resistance as a feedback for temperature control
[08:19:11] <Jymmm> that might be interesting
[08:19:37] <Jymmm> I have these rails already http://www.rcgroups.com/articles/ezonemag/2003/feb/xy/xytable1.jpg
[08:21:50] <Jymmm> http://www.8linx.com/cnc/cnc.htm
[08:23:59] <Jymmm> ds3 http://www.hobbycnc.com/4aupcwhc.php
[08:24:09] <Jymmm> built in heat control
[08:27:18] <Jymmm> WHAT THE FUCK?! http://www.hobbycnc.com/downloads.php
[08:27:29] <Jymmm> How to modify an AT and ATX computer power supply for use as a 12VDC CNC power supply.
[08:29:27] <ds3> hmmm no idea what they'd expect
[08:29:59] <Jymmm> which one are you talking about? the PS or heat controller?
[08:31:16] <ds3> the heat controller
[08:31:23] <ds3> it is a single bit input
[08:35:44] <Jymmm> Yeah, I was thinking that emc could do the controlling, replaicng the spindle control for a heat control.
[08:44:35] <ds3> could always run it openloop
[08:49:01] <Jymmm> ?
[09:04:09] <alex_joni> huh, where did jmkasunich come from?
[09:10:27] <Jymmm> netsplit of leguin
[09:10:45] <Jymmm> [02/21 01:03] * SWPadnos (n=Me@emc/developer/SWPadnos) Quit (leguin.freenode.net irc.freenode.net)
[09:10:45] <Jymmm> [02/21 01:03] * jepler (n=jepler@emc/developer/jepler) Quit (leguin.freenode.net irc.freenode.net)
[09:10:45] <Jymmm> [02/21 01:03] * jmkasunich (n=jmkasuni@emc/board-of-directors/jmkasunich) Quit (leguin.freenode.net irc.freenode.net)
[09:10:45] <Jymmm> [02/21 01:03] * SWPadnos (n=Me@emc/developer/SWPadnos) has joined #emc
[09:10:45] <Jymmm> [02/21 01:03] * jmkasunich (n=jmkasuni@emc/board-of-directors/jmkasunich) has joined #emc
[09:10:47] <Jymmm> [02/21 01:03] * jepler (n=jepler@emc/developer/jepler) has joined #emc
[09:10:49] <Jymmm> [02/21 01:04] <alex_joni> huh, where did jmkasunich come from?
[09:11:47] <Jymmm> alrigth, which geckos should I buy (if needed)
[09:14:44] <tray> Does anyone know any tips for identifying if your PCI parallel port is working?
[09:15:54] <Jymmm> print from it?
[09:16:09] <Jymmm> http://www.epanorama.net/circuits/parallel_output.html
[09:18:19] <tray> Ahh, yes I've connected a LED & resistor across pins 2-25 & the LED is lit as soon as the PC starts up. Haven't been able to compile the code on the epan site to test operation. Thnaks for the reply
[09:18:34] <alex_joni> Jymmm: I meant this:
[09:18:36] <Jymmm> thers always dos
[09:18:38] <alex_joni> 10:58 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: SWPadnos,
[09:18:39] <alex_joni> jepler
[09:18:39] <alex_joni> 10:59 -!- Netsplit over, joins: jmkasunich, SWPadnos, jepler
[09:18:57] <Jymmm> alex_joni you have a suck client is all
[09:19:06] <Jymmm> mine shows jmkasunich got booted too
[09:19:08] <alex_joni> Jymmm: no I don't
[09:19:16] <alex_joni> it's as good as it gets :P
[09:19:17] <Jymmm> jmkasunich (n=jmkasuni@emc/board-of-directors/jmkasunich) Quit (leguin.freenode.net irc.freenode.net)
[09:19:19] <alex_joni> irssi :)
[09:19:39] <Jymmm> ok, then either it or you phuked up, so which is it?
[09:19:56] <alex_joni> you did something.. I'm sure of it
[09:20:07] <Jymmm> no, but I'mabout to
[09:20:12] <alex_joni> heh
[09:20:39] <tray> Thanks Jymmm
[09:20:51] <Jymmm> tray what os are you using?
[09:21:22] <tray> ubuntu, installed from the LiveCD
[09:21:36] <tray> the liveCD being EMC live
[09:22:09] <Jymmm> linux sucks, DOS RULES!
[09:22:54] <Jymmm> tray * In many linux systems you can get information on I/O devices and ports they use with command "cat /proc/ioports". This list should include your parallel port or parallel ports in it. The first parallel port (port in motherboard) typically has name parport0. There are some cases where you might not see your device in your list or sometimes you see the sama name twice (usually the first one is the right one).
[09:22:54] <Jymmm> * In some systems you might have additional parallel ports on I/O cards in the PCI bus. In many new Linux systems you can run "lspci -v | more" to get information of each device on your PCI bus (this should tell make and model, interrupt and port and memory block assignments).
[09:23:09] <tray> emc runs on linux, and you're on a emc forum??
[09:23:21] <Jymmm> tray and your point is?
[09:24:13] <tray> a funny place to support dos, still each system has it's strengths
[09:24:58] <Jymmm> There is a dos version of EMC ya know!
[09:25:16] <tray> no I didn't know that
[09:25:29] <Jymmm> It's a piece of shit, but still =)
[09:25:31] <alex_joni> for machine control DOS might be one of the few alternatives to an RT linux
[09:25:39] <alex_joni> Jymmm: you talking about turbocnc ?
[09:25:58] <Jymmm> alex_joni no, there is a variant of emc (PD stuff) written by someone
[09:26:06] <tray> There's a few dos cnc pieces of code around
[09:26:10] <alex_joni> Jymmm: got some pointers?
[09:26:30] <Jymmm> nope, can't find it, looked at it and went straight back to TCNC
[09:26:47] <tray> pspci lists the card, manufacturer & IO port address
[09:26:56] <alex_joni> TCNC is also based on emc iirc
[09:26:59] <alex_joni> at least the interp
[09:27:04] <tray> lspci that is
[09:27:21] <alex_joni> tray: you searching for a secodn parport?
[09:27:30] <tray> I've read the note on the emc website relating to NETMOS cards
[09:27:33] <alex_joni> ok :)
[09:27:52] <alex_joni> Jymmm: if you ever find it again, let me know
[09:27:55] <tray> What does the second parport look like?
[09:28:09] <tray> the second address used by that card?
[09:28:16] <Jymmm> alex_joni: Why? It's closed source.
[09:28:23] <alex_joni> err, yeah, that's what I meant
[09:28:32] <alex_joni> Jymmm: just curiosity
[09:29:34] <tray> I then enter this addess into the config file. Then I should be able to run emc, & use the hardware monitor to watch the led?
[09:30:24] <tray> Watch the LED when I put the machine ON, and manually jog the x axis, when using the xylo config
[09:31:39] <alex_joni> tray: yes, something like that
[09:32:17] <tray> such a simple plan but it doesn't work :(
[09:33:03] <alex_joni> what doesn't ?
[09:35:01] <tray> the led connected doesn't change state
[09:35:12] <tray> LED across pins 2-25
[09:36:42] <Jymmm> alex_joni http://www.deskam.com/deskncrt.html
[09:37:48] <tray> I should be able to use the HALISCOPE to monitor the pin, but I can't get that to work either. So I need to employ some tools to debug what is happening, which is of course what the HALSCOPE is for
[09:38:26] <Jymmm> alex_joni if you get it to work, let me know
[09:44:12] <alex_joni> Operating NC Machines is dangerous. Under this license agreement, you may NOT use DeskNCRT to directly control the turning ON of a machines motor.
[09:44:14] <alex_joni> hah
[09:45:08] <Jymmm> well, that's a good way to CYA =)
[09:45:42] <tray> to monitor the xen channel what Halscope settings should I use?
[09:46:30] <alex_joni> tray: you don't need halscope for xen
[09:46:33] <alex_joni> halmeter is enough
[09:46:47] <tray> Using axis, I would have thought that I can monitor the X en & X dir channels, even if I haven't configure the parallel port correctly?
[09:47:35] <tray> Sorry Alex, I haven't come across the Halmeter? Can you explain
[09:48:20] <alex_joni> tray: how do you start halscope?
[09:48:51] <alex_joni> Jymmm: $Id: rs274ngc.cc,v 1.35 2000/02/24 15:30:13 proctor Exp $
[09:48:53] <tray> menu item - machine, halscope in axis
[09:48:55] <alex_joni> sure is emc inside
[09:49:03] <Jymmm> alex_joni read french?
[09:49:03] <alex_joni> tray: look for halmeter in the same menu
[09:49:10] <alex_joni> Jymmm: hardly..
[09:49:12] <alex_joni> but I can struggle
[09:49:30] <Jymmm> alex_joni Next thing you'll say is that you don't read Romania either, eeeeesh!?
[09:50:23] <tray> I see a calibration,emc status, show hal config
[09:51:40] <tray> calibration is a little slow in loading
[09:53:10] <alex_joni> you don't need calibration
[09:53:17] <alex_joni> show hal config will help
[09:53:22] <alex_joni> but there must be a halmeter around
[09:54:15] <alex_joni> maybe it's under View ?
[09:54:44] <tray> No I think you mean goto HAL configuration
[09:55:00] <tray> Then I add a watch for Xen & X dir
[09:55:09] <alex_joni> .menu.machine add command \
[09:55:09] <alex_joni> -command {exec halmeter &}
[09:55:09] <alex_joni> setup_menu_accel .menu.machine end [_ "H_al Meter"]
[09:55:27] <tray> the Xen turns on when I turn on the machine
[09:55:31] <alex_joni> it should be right below Hal configuration
[09:55:49] <alex_joni> and before Hal Scope
[09:56:05] <alex_joni> but Halshow should do just fine
[09:56:18] <alex_joni> Jymmm: had somethign for me to read?
[09:57:01] <tray> can I add the menu as you listed, at the command line?
[09:57:54] <alex_joni> tray: no, it should be there
[09:58:00] <alex_joni> just look closer :)
[09:59:51] <tray> Looking closer, but someone has stolen it??
[10:00:22] <Jymmm> alex_joni like what? playboy?
[10:02:29] <alex_joni> < Jymmm> alex_joni read french?
[10:02:46] <alex_joni> tray: you're running emc2.1 ?
[10:02:50] <Jymmm> alex_joni google does a better job at translating than you do =)
[10:03:05] <alex_joni> Jymmm: probably so
[10:03:08] <Jymmm> alex_joni Just looking at some hotwite cnc machines
[10:03:38] <Jymmm> alex_joni: I got 3 pairs of 1U rails tonight, so why not =)
[10:04:30] <tray> running emc -v last line says 2.0.3
[10:16:00] <alex_joni> tray: then maybe you want to upgrade :)
[10:16:17] <alex_joni> Jymmm: I like this one I just found: http://www.goindustry.com/en/equipmentdetails.asp?sOfferID=1073923
[10:16:33] <tray> ahh, the ol'pgrade to fix ya problems eh?
[10:16:43] <alex_joni> tray: depending what your problem is :P
[10:16:50] <alex_joni> so far I didn't hear of one
[10:17:08] <alex_joni> I just suggested you probably want an upgrade because of new features
[10:17:23] <tray> You're probably right, at this stage in my emc career. The more help the software can give me the better
[10:17:27] <alex_joni> btw, "it doesn't work" is not really a problem :)
[10:17:56] <Jymmm> alex_joni http://foamlinx.com/
[10:18:06] <tray> """it doesn't work" is not really a problem"" I think I know what you mean.
[10:18:54] <tray> Alex, thanks for you help. I'll download again & see what I can come up with.
[10:19:18] <tray> Is it worth using the latest ubuntu & trying to get the RTAI woking?
[10:23:46] <Jymmm> where was this circuit cellar article?
[10:27:07] <alex_joni> tray: what latest Ubuntu?
[10:27:36] <tray> 6.10
[10:29:00] <alex_joni> I wouldn't advise on that
[10:29:36] <alex_joni> it's not real fun to get a proper RT system going
[10:29:46] <alex_joni> I'd stick with the setup you have
[10:30:06] <alex_joni> just connect the box to the internet and follow this: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?UpdatingTo2.1
[10:30:47] <alex_joni> it should upgrade to 2.0.5 automatically once connected to the internet
[10:30:56] <alex_joni> maybe even that is enough to fix parport related problems
[10:32:13] <tray> I started downloading the Jan 2007 liveCD. Kinda like the disc, the I can return to a clean slate.
[10:32:42] <tray> Should only take 6hours
[10:33:03] <alex_joni> ok
[10:33:19] <alex_joni> but there's not much new on the new LiveCD
[10:33:30] <alex_joni> an old one + update gives you the same things
[10:33:39] <alex_joni> (probably 100MB updates)
[10:34:06] <tray> Getting the 2.03 version was a little rough, so I'll try to upgrade
[10:40:52] <tray> the knowledge base note refer to breezy but the liveCd has referred to dapper
[10:42:30] <tray> No worries, just needed to adjust connection settings
[10:59:30] <alex_joni> tray: can you tell me what page still talks about breezy ?
[11:00:53] <tray> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?UpdatingTo2.1. Sorry a little confused here, I will post the repo location in a sec. I've upgraded to 2.0.5.
[11:01:25] <alex_joni> tray: that page describes things for dapper, in [] you have the name for breezy
[11:01:53] <alex_joni> ' Channels ["Software Sources" in breezy] '
[11:02:09] <alex_joni> dapper calls it Channels, breezy calls it "Software Sources"
[11:02:37] <tray> For 6.06 I used url www.linuxcnc.org/emc2, Distribution=dapper
[11:02:45] <alex_joni> that is correct
[11:02:59] <alex_joni> but you must be sure you are runnning Dapper
[11:03:07] <alex_joni> I mean _already_ running Dapper
[11:03:12] <alex_joni> otherwise this won't work
[11:03:34] <alex_joni> Changing Distribution=breezy to Distribution=dapper is not a good plan
[11:05:30] <tray> No, I left the distribution as was already set.
[11:05:39] <alex_joni> ok, then it's ok
[11:05:56] <alex_joni> you want to change the url to www.linuxcnc.org/emc2.1
[11:06:02] <alex_joni> if you want the latest release
[11:07:54] <tray> changing to 2.1 I get a error that the repo might not be available
[11:08:43] <alex_joni> err sorry
[11:08:51] <alex_joni> the url is www.linuxcnc.org/emc2
[11:09:15] <alex_joni> on Components you want emc2.1 instead of emc2
[11:09:34] <tray> yep, that's what I found, and this upgrades me to 2.0.5
[11:09:57] <tray> sorry, just got the last bit about components
[11:10:40] <alex_joni> you just need to read things more carefull
[11:13:04] <tray> yep, another reason for buying a second monitor. makes it easy to double check things. Still first time I've upgraded any packages on ubuntu
[11:13:27] <alex_joni> tray: if the PC is connected to the internet you should get daily/weekly updates
[11:13:33] <alex_joni> various packages
[11:15:16] <tray> Sounds good, as long as the changes are 'good'
[11:16:49] <tray> This attempt looks better axis is being removed, as it's part of the base code as per the release notes of 2.1
[11:19:08] <alex_joni> right
[11:27:04] <tray> (alex) will need to work my way through the updating process. Again thank you for your help
[11:27:20] <jlmjvm> good morning alex
[11:27:34] <alex_joni> 13:22 < jlmjvm> good morning alex
[11:27:37] <alex_joni> jlmjvm: hi :)
[11:27:46] <alex_joni> tray: no problem
[11:28:40] <jlmjvm> alex:do you know how to install gdepth-0.2
[11:28:53] <alex_joni> jlmjvm: nope, never tried
[11:29:05] <jlmjvm> k
[11:29:09] <alex_joni> what did you try so far?
[11:29:43] <jlmjvm> downloaded tar to desktop,then extracted to a folder on desktop
[11:30:35] <alex_joni> ok, did you read the README ?
[11:31:19] <alex_joni> If you wish to use the acceleration module (see below), run "make".
[11:31:21] <alex_joni> Otherwise, no compilation step is required.
[11:31:39] <alex_joni> I suggest you start using it without accel first
[11:31:56] <alex_joni> (chances are there's no real accel available if this is an emc2 machine)
[11:33:13] <jlmjvm> yes,but dont really understand,is it installed at this point
[11:33:57] <alex_joni> jlmjvm: you can start using it
[11:34:06] <jlmjvm> k
[11:34:13] <alex_joni> python gdepth.py -x "min max" -y "min max" -z "bottom top" -r dpi \
[11:34:13] <alex_joni> [-t tool-radius] [-o output.png] [-d] input.ngc
[11:34:20] <alex_joni> e.g. Ö
[11:34:20] <alex_joni> e.g. :
[11:34:53] <alex_joni> python gdepth.py -x "10" -y "10" -z "1" 3DChips.ngc
[11:39:35] <jlmjvm> ?
[11:39:41] <jlmjvm> cant find it
[11:40:16] <jlmjvm> where are you getting into it
[11:42:16] <jlmjvm> where are you inputting those parameters or values
[11:44:42] <jlmjvm> on gdepth.py do i run in terminal,or run when i click on it
[11:46:28] <alex_joni> jlmjvm: you open a terminal
[11:46:33] <jlmjvm> k
[11:46:38] <alex_joni> then you change to the folder where you unpacked it
[11:46:45] <alex_joni> then you typed the command I wrote above
[11:48:16] <jlmjvm> k,i opened in terminal,terminal opens for a sec,then goes away
[11:55:40] <alex_joni> I didn't say you should run in terminal
[11:55:48] <alex_joni> I said you should open a terminal
[11:56:05] <alex_joni> from the top menu, applications -> accessories -> terminal
[11:56:09] <alex_joni> or something like that
[11:56:33] <jlmjvm> k,im in terminal now
[11:56:46] <alex_joni> then you change to the folder where you unpacked it
[11:56:51] <jlmjvm> just got to dir folder
[11:56:59] <jlmjvm> typing command
[11:57:23] <alex_joni> you probably need a full path to a ngc file
[11:58:09] <jlmjvm> prolly,said no such file or dir
[11:58:47] <jlmjvm> is this how this program runs,i thought it was inside emc
[12:00:57] <alex_joni> no, not inside emc
[12:01:09] <alex_joni> completely separate
[12:02:10] <jlmjvm> /home/jlmjvm/Desktop/gdepth-0.2,this is where my folder is,how do i get to it?
[12:03:04] <jlmjvm> cd doesnt get me there
[12:06:16] <alex_joni> cd /home/jlmjvm/Desktop/gdepth-0.2
[12:08:37] <jlmjvm> dude,i typed that same syntax before,but i didnt have a space between cd and /home
[12:09:07] <jlmjvm> it makes a difference,now in in dir
[12:10:32] <jlmjvm> its trying to work now,need to find a ncg path
[12:11:03] <jlmjvm> thanks
[12:19:46] <alex_joni> jlmjvm: cool
[13:27:11] <alex_joni> morning samco
[13:28:53] <skunkworks> morning alex - how is it going?
[13:28:59] <alex_joni> pretty well
[13:30:12] <skunkworks> home inspection today with the buyer.. might be a little stressed :)
[13:31:08] <alex_joni> heh.. wish you all the best :P
[13:31:21] <skunkworks> thanks :)
[13:34:07] <skunkworks> should be 45F here today -
[13:34:15] <alex_joni> that sounds warm
[13:34:29] <skunkworks> it will be after below 0f for the last few weeks
[13:34:36] <alex_joni> heh
[13:35:08] <skunkworks> Have you done anything more with your mill?
[13:35:21] <alex_joni> hmm.. no, not yet
[13:36:47] <skunkworks> me either :(
[13:37:23] <skunkworks> I am giving it a month or 2 - then things should slow back down
[13:38:05] <CIA-6> 03jepler 07TRUNK * 10emc2/tcl/emc.tcl.in: use hal.so instead of invoking halcmd
[13:38:06] <CIA-6> 03jepler 07TRUNK * 10emc2/tcl/bin/ (emccalib.tcl halshow.tcl): use hal.so instead of invoking halcmd
[13:38:20] <alex_joni> http://dsplabs.cs.utt.ro/~juve/blog/index.cgi-files/sandbox/IMG_9720.JPG
[13:39:22] <skunkworks> cool groving fest? :)
[13:39:30] <alex_joni> not quite
[13:39:34] <alex_joni> a client of mine
[13:41:56] <skunkworks> looks like it works - was that done with an end mill or v-cutter?
[13:43:02] <alex_joni> ball mill :)
[13:43:07] <alex_joni> didn't have any other
[13:43:12] <skunkworks> :)
[13:43:28] <alex_joni> it's also a bit low on rpm, it looks a bit melted
[13:50:52] <jepler> you'll edge-light it?
[13:51:24] <alex_joni> jepler: hmm.. not sure it's deep enough for that
[13:51:27] <alex_joni> what do you think?
[13:51:32] <alex_joni> the cut I mean
[13:52:47] <jepler> alex_joni: for stippler, "the dots are drilled 0.004" (.1mm) into a 0.089" (2.3mm) piece of plexiglass"
[13:53:12] <alex_joni> this is 2mm plexiglass
[13:53:24] <alex_joni> guess it's more than .1mm deep (I think I used .25mm)
[13:53:49] <alex_joni> you used a black background.. right?
[13:55:06] <jepler> yes
[13:55:34] <jepler> and I wanted the dots to appear as small as possible too
[13:55:42] <alex_joni> right..
[13:55:49] <alex_joni> how many led's did you use?
[13:55:56] <jepler> if you have a LED or LED flashlight it's not hard to shine it in the edge and see how it looks
[13:56:11] <jepler> 12 or 16 LEDs I think
[13:56:15] <jepler> running at about 20mA each
[13:56:20] <jepler> it's actually nearly 1W
[13:56:45] <SWPadnos> sounds like 1/4 to 1/3 watt to me :)
[13:57:00] <SWPadnos> (into the LEDs anyway)
[13:57:49] <jepler> 16 * 2V * 20mA = 640 mW
[13:57:56] <jepler> maybe 2V is overestimating Vf
[13:58:22] <SWPadnos> ah yes - my de-caffeinated state made me forget to multiply by voltage
[13:58:24] <jepler> 4 * Vf_led <= 4 * 2.8V = 11.2V
[14:08:16] <alex_joni> jepler: how did you place the leds ?
[14:08:47] <alex_joni> I'm thinking that SMD leds would be nice, as they are about the same size as the plexi thickness
[14:09:00] <jepler> alex_joni: I used a special SMD LED which shines through a hole drilled in the PCB
[14:10:12] <jepler> alex_joni: so on the copper side I had the LEDs laid out in series, and on the other side of the PCB I milled (I guess I'm supposed to say "MCB" not "PCB") there was a channel a little wider than the plexiglass so that it "fit over" it
[14:10:33] <jepler> I don't really have a good photo of it unfortunately
[14:10:55] <jepler> http://emergent.unpy.net/files/01109127822/img_9015.jpg
[14:10:55] <alex_joni> http://emergent.unpy.net/files/01109127822/img_9015.jpg
[14:11:01] <alex_joni> :)
[14:11:07] <alex_joni> I see it on the bottom
[14:11:22] <jepler> look at the top left one -- the copper side is towards the camera, and you can see the channel cut in the other side
[14:12:59] <alex_joni> nice, thank you..
[14:13:07] <alex_joni> I might waste some time on it someday
[14:13:41] <skunkworks> http://emergent.unpy.net/files/stippler/img_9023.jpg
[14:14:38] <skunkworks> jepler: have you ever thought about changing the program to use a v style cutter - so z depth would be relation to how big the dot would be - more like a half tone?
[14:15:16] <skunkworks> that is cool btw
[14:16:00] <jepler> skunkworks: I think it was a "V" cutter, but all the holes were at the same depth
[14:16:09] <jepler> skunkworks: but yeah that sure would be possible
[14:16:41] <jepler> alex_joni: the LEDs are package "I": http://www.mouser.com/catalog/629/95.pdf
[14:17:37] <alex_joni> jepler: thanks.. a friend of mine said he might have some
[14:26:05] <alex_joni> did you guys ever use %i with printf ?
[14:36:08] <SWPadnos> I think I always use %d instead
[14:40:49] <jepler> the printf(3) manual page makes it sound as though %d and %i function identically
[14:40:52] <jepler> I always write %d
[14:40:53] <alex_joni> same here.. but I found it hard to read :)
[14:41:07] <alex_joni> yeah, was just surprised about %i
[14:58:22] <alex_joni> bbl
[15:13:36] <CIA-6> 03eric-johnson 07TRUNK * 10emc2/src/hal/utils/halrmt.c: Initial commit. Telnet interface for HAL
[15:14:30] <CIA-6> 03eric-johnson 07TRUNK * 10emc2/src/hal/utils/Submakefile: Update make file to include halrmt
[15:19:42] <SWPadnos> alex_joni, the DOS version of EMCis probably Mach1, or whatever Art called the DOS version of Mach
[15:21:30] <jepler> while I hear that mach incorporated some of the public domain parts of emc, I don't see why you'd call it "the DOS version of EMC"
[15:22:06] <SWPadnos> I wouldn't - I was trying to shoehorn something into Jymmm's definition
[15:22:24] <SWPadnos> though it looks like even Master 5, which was the precursor to Mach, was written for Windows
[15:23:32] <skunkworks> I remember there being a dos verions that touted using emc parts. I can't remember what that was called. hmmm
[15:25:43] <jepler> in the business of non-Free software, why would you tout "an important part of our software wasn't even written by us -- it came from the public domain"?
[15:25:45] <jepler> it makes no sense to me
[15:26:18] <SWPadnos> "standards-compliant"
[15:26:20] <SWPadnos> (we stole the code)
[15:26:25] <skunkworks> found it http://www.deskam.com/deskncrt.html
[15:27:17] <SWPadnos> "This is a screen shot showing the wealth of information displayed while processing your files.
[15:27:19] <SWPadnos> The coordinate screen is updated at 8 times per second so you will always know exactly where your tool is located"
[15:27:21] <SWPadnos> how funny
[15:27:55] <skunkworks> it 'does' have 150 block look-ahead :)
[15:28:42] <SWPadnos> that reminds me - Stece Hardy has released an interface to the G-Rex. The specs are in the Geckodrive files section
[15:28:47] <SWPadnos> does anyone want to have a look?
[15:28:51] <SWPadnos> Steve Hardy, that is
[15:29:14] <SWPadnos> there will be a Linux version of the library at release time
[15:29:44] <skunkworks> ah - I think your the only one with a g-rex ;)
[15:30:31] <SWPadnos> jmk has one, Matt S has one ...
[15:31:37] <skunkworks> I stand corrected. what do you think of the g-rex?
[15:31:52] <SWPadnos> dunno - never used it ;)
[15:32:05] <SWPadnos> luckily it was free
[15:32:43] <jepler> I suspect they made trade-offs that are inappropriate for emc, just like pete w.'s step generating mesa firmware did
[15:32:43] <skunkworks> I can't believe the z80 is still around.
[15:33:04] <jepler> that's not much crazier than the 8086 still being around
[15:33:11] <skunkworks> right
[15:33:14] <SWPadnos> there is one major tradeoff - the thing uses some sort of serial bus for communication (ethernet or USB)
[15:33:56] <SWPadnos> if I ever get off my butt and design an ARM-based Rabbit replacement, the PID can run on the G-Rex
[15:34:09] <jepler> ethernet has good electrical isolation properties doesn't it?
[15:34:12] <SWPadnos> (actually, Steve's got a small version of EMC on the Rabbit anyway)
[15:34:15] <SWPadnos> yes
[15:34:19] <SWPadnos> 1500V is the spec, I think
[15:34:44] <jepler> as long as you don't mind toasting g-rexes when something goes wrong that is nice
[15:35:33] <SWPadnos> the baord-based one (the G101 I think) is $208 in single quantity, so it's comparable to the Mesa card
[15:35:56] <SWPadnos> but it's got some isolation and high-current drive built in
[15:36:02] <SWPadnos> (also 4 channel analog in and out)
[15:36:54] <SWPadnos> but less total digital I/O, I think (and though the FPGA is reprogrammable, they don't provide information on which pins connect to what)
[15:39:25] <skunkworks> was it the initial z80 that had an instruction that would cause it to smoke? or there was that an urban myth
[15:39:53] <skunkworks> who ever this site is says they prefer emc/ubuntu :) http://www.probotix.com/cnc_software
[15:40:25] <SWPadnos> I don't know about a "kill" instruction :)
[15:44:00] <jepler> there's the legendary 'hcf' instructio
[15:44:01] <jepler> n
[15:44:18] <jepler> http://en.wikipedia.org/wiki/Halt_and_Catch_Fire
[15:44:25] <SWPadnos> heh
[15:46:48] <skunkworks> maybe that is what I was thinking of.
[15:58:47] <awallin> hi, anyone know about putting a filter inside a PID loop ?
[15:59:10] <awallin> I'm trying it right now, but the loop oscillates whenever the phase delay of the filter is -180 degrees
[15:59:32] <awallin> so I'm searching for a digital filter with 'nice' phase behaviour for this purpose
[16:03:35] <SWPadnos> it's "inside the loop" because it's on the output of the PID (which affects the control signal and therefore the feedback)?
[16:32:29] <awallin> SWPadnos: hi, I'm filtering the error signal, before it goes to the P/I/D gain blocks
[16:33:00] <awallin> I want the PID loop to compensate for low(high) frequency disturbances, and leave the high(low) frequency stuff alone
[16:33:35] <SWPadnos> hmmm
[16:33:59] <SWPadnos> what kind of filter are you using?
[16:34:26] <awallin> the PID alone works fine, it runs at 200kHz sampling f, and seems to compensate well up to about 10kHz (that's as far as I need)
[16:35:18] <awallin> wait, I'll check... (but when scoping the signals, I see there are many frequencies where the phase delay is -180 deg. that can't be right) those frequencies match approximately the oscillation fs I see in the PID loop
[16:36:16] <SWPadnos> it does make sense. FIR filters (for example) have constant time delay, which means that the phase delay is proportional to frequency
[16:36:22] <awallin> the current ones are elliptic filters, about 15-order (15 different filter coeffs). either 1kHz low-pass or 1kHz high pass
[16:38:07] <awallin> any other smarter ideas for how the PID loop could correct for only high or low frequency stuff ?
[16:38:09] <SWPadnos> the very inefficient but possibly functional way to do it is to do a small FFT (imaginary, since you need phase info), then do an inverse FFT after setting the magnitude of whichever frequency components you want filtered to 0
[16:38:26] <SWPadnos> this is in a DSP, right?
[16:38:29] <awallin> hm. this needs to run on an FPGA
[16:38:36] <SWPadnos> oh, then that won't work ;)
[16:38:43] <awallin> 40MHz clock frequency, and the AD runs at 200kHz
[16:39:37] <SWPadnos> hmm. for a sufficiently small sample set, you may be able to do an integer FFT in hardware, and even have a new set of answers every A/D sample
[16:40:40] <SWPadnos> for a small sample size, like 16 samples or so, you can figure out the set of operations to do on each sample, and count up the resources you'd need
[16:40:46] <awallin> for the scheme you describe (FFT, set some amplitudes to zero, then I-FFT), isn't there an implementation that uses the standard multiply and accumulate algorithm?
[16:40:52] <SWPadnos> damned if I know ;)
[16:40:58] <SWPadnos> I was just brainstorming
[16:41:03] <awallin> ok.
[16:46:39] <SWPadnos> awallin, you may want to check on opencores.org for some filter / DSP / other math blocks
[16:51:22] <awallin> I'm doing this in Labview, with a NI-FPGA card, so I have pretty much all the standard filter implementations available, maybe even some FFT stuff, the challenge is finding the correct filter type I could use
[16:57:21] <SWPadnos> don't they have a filter designer?
[16:58:32] <SWPadnos> I'm surprised you got it to a 200 KHz loop rate. is this with the CompactRIO or a PC-based FPGA card?
[17:04:12] <tomp> for svg on firefox 1.5, there's an extension that enables zooming/panning http://www.treebuilder.de/zoomandpan/index.htm . a long time ago, Adobe's plugin allowed zoom in/out and pan, firefox and adobe dont play nice tho, so this xpi almost works as well as it used to ;) then visit http://upload.wikimedia.org/wikipedia/commons/4/40/Sine_sqare_triangle_generator_blocks.svg
[17:26:58] <jepler> tomp: adobe killed their svg plugin about the time they bought macromedia (flash) -- coincidence?
[17:56:57] <awallin> SWPadnos: sorry I was away for a while. It's an R-series PCI card, but I think it's identical hardware as in the compact RIO modules
[17:57:23] <SWPadnos> ok. I think they are very similar
[17:57:37] <awallin> Yes, there is a filter designer, but but with the 200 kS sampling rate I seem to get about 1 ms time delay, which translates into a linear phase delay as you said
[17:57:49] <SWPadnos> there's a "discrete filter" component that does a 5th-order low pass filter (always at 1/10 the sample freq, I think)
[17:58:32] <SWPadnos> you should be able to do something like subtracting the LP filter output from the input signal (delayed by some amount) to get the HF components
[17:58:36] <awallin> 1 ms seems awfully long, 200 cycles at 200 kHz
[17:59:17] <awallin> even thought the filter has only 15-20 coefficients
[17:59:19] <SWPadnos> well, you can open that VI to give you ideas for a custom one
[17:59:51] <SWPadnos> it shouldn't be 1 ms. it shouldn't be able to be more delay than a simple shift register would be
[18:00:10] <SWPadnos> ie, if you have 20 taps , then the delay shouldn't be capable of being more than 20 periods
[18:00:28] <SWPadnos> since you have 20 samples in the FIFO
[18:00:28] <lerneaen_hydra> 'lo
[18:02:35] <awallin> SWPadnos: I agree, I need to look into this tomorrow... thanks for the chat, going home now.
[18:02:55] <SWPadnos> see you later. I should get back to accounting anyway :)
[18:06:21] <tomp> jepler: i can still download the adobe svg plugin, tho they say "Please note that Adobe has announce that it will discontinue support for Adobe SVG Viewer on January 1, 2008." and firefox is committed to stay web standard, so its built in. adobe dload at http://download.adobe.com/pub/adobe/magic/svgviewer/linux/3.x/3.01x88/en/adobesvg-3.01x88-linux-i386.tar.gz
[18:07:13] <tomp> yes, adobe wants to kill the free/open flash tools like svg
[19:00:38] <lerman__> lerman__ is now known as lerman
[19:12:27] <alex_joni> hi guys
[19:13:01] <skunkworks> Hi alex. Home inspection guy is walking around the house. :)
[19:13:11] <lerneaen_hydra> spooky
[19:13:16] <skunkworks> I'll say
[19:13:17] <lerneaen_hydra> what's he inspecting?
[19:13:35] <skunkworks> we are selling. The buyers are having the house inspected.
[19:13:41] <lerneaen_hydra> ah, moving, ok :)
[19:14:06] <lerneaen_hydra> it's always scary when that inspector comes and looks around
[19:14:13] <lerneaen_hydra> rather nerve-wracking
[19:14:20] <skunkworks> a bit ;)
[19:14:41] <skunkworks> the buyers are stopping by in an hour also..
[19:14:59] <lerneaen_hydra> heh
[19:15:08] <skunkworks> 'potential buyers'
[19:35:05] <alex_joni> I'm thinking about getting this one
[19:35:08] <alex_joni> http://www.goindustry.com/en/equipmentdetails.asp?sOfferID=1073923
[19:38:17] <skunkworks> so the end has vertial movement?
[19:38:23] <skunkworks> (z)
[19:39:53] <skunkworks> wow - that is an awful lot of electronics.
[19:40:05] <skunkworks> (cabinates
[19:40:07] <skunkworks> )
[19:45:00] <alex_joni> yeah, that's a scara
[19:45:08] <alex_joni> but it seems awfully cheap
[19:46:32] <alex_joni> brb
[20:12:20] <xemet> hi to all
[20:14:46] <jepler> hi xemet
[20:17:11] <xemet> hi
[20:17:28] <xemet> I've posted a page in the wiki about my experience with the joypad
[20:17:59] <xemet> it is the first page I post in the wiki so if there are errors please let me know
[20:18:22] <xemet> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Using_A_Joypad_To_Move_Your_CNC_Machine
[20:19:03] <xemet> also my english is not very good...so if someone finds errors please let me know
[20:21:10] <jepler> great, thank you
[20:21:58] <xemet> welcome, I hope I can do more in the future!
[20:22:03] <skunkworks> very nice. Good work.
[20:33:54] <crepincdotcom> jepler: im trying to export from eagle, and I'm getting the error "Error: Unrecognized Argument. \n\n CHAN"
[20:34:03] <crepincdotcom> any idea what would cause that?
[20:49:44] <crepincdotcom> :-(
[20:54:19] <jepler> no, it doesn't ring any bells
[20:54:26] <crepincdotcom> drat
[21:01:38] <crepincdotcom> sigh same thing on a different schematic
[21:01:44] <crepincdotcom> i must have messed something upo
[21:01:45] <crepincdotcom> -o
[21:02:09] <crepincdotcom> OH
[21:02:13] <crepincdotcom> haha
[21:02:36] <crepincdotcom> the output filenames have to be one word ;-)
[21:03:01] <jepler> * jepler would never do that
[21:03:09] <jepler> (put spaces in filenames)
[21:03:45] <crepincdotcom> * crepincdotcom is copying them fron his m$ box since there's no monitor on the nix box and VNC makes it look un-nice
[21:08:11] <Carlo> quit
[21:09:56] <crepincdotcom> ok i have a better question jepler
[21:10:10] <crepincdotcom> why does it make (0,0) so far from the actual board?
[21:10:23] <crepincdotcom> its 3" from the center of my board, and my y-axis isn't 3" long
[21:10:59] <cradek_> in eagle, your origin must be 3" from the center of the board
[21:11:05] <cradek_> the origin is marked with a + in eagle I think
[21:11:17] <cradek_> just group everything and move it so the origin is where you want
[21:11:18] <crepincdotcom> is the origin the little cross thing? im playing with it now...
[21:14:17] <crepincdotcom> nope that wasnt it
[21:22:08] <alex_joni> cradek_: bump
[21:22:38] <alex_joni> cradek_: ping
[21:22:51] <cradek_> cradek_ is now known as cradek
[21:33:20] <crepincdotcom> milling in progress, lets see if this board comes out any better
[21:34:57] <crepincdotcom> hm, bit is a little deep but i think theres still enough conductor there
[21:35:10] <cradek> how deep did you cut?
[21:36:26] <crepincdotcom> 0.017"
[21:36:41] <cradek> try .004
[21:36:57] <cradek> I think that's what I use
[21:40:41] <crepincdotcom> it shouldnt matter really how deep, as long as it knows how wide the cut is at that depth
[21:40:54] <crepincdotcom> so what i really meant was, its a little too wide still
[21:41:13] <crepincdotcom> but the REASON is that its too deep, since my stepper misses steps going up for some unknown reason
[21:41:25] <cradek> true, I guess I always wanted to get the cut as small as possible
[21:41:26] <SWPadnos> runout will manifest itself as a seemingly wider cutter
[21:41:33] <crepincdotcom> ahhhhhhhhhhhhhh
[21:41:35] <cradek> that's true too
[21:41:37] <crepincdotcom> i didnt think of that
[21:41:42] <crepincdotcom> i bet thats playing a pary
[21:41:43] <crepincdotcom> *part
[21:42:11] <alex_joni> http://www.thecooltool.com/files/dateien_527.pdf
[21:42:15] <crepincdotcom> got 2 more nice encoders from the lab today, im milling a pwm driver as we speak
[21:45:11] <SWPadnos> alex_joni, do you think they can correct a typo in that document?
[21:45:58] <cradek> alex_joni: do you know how they home the gantry?
[21:46:42] <cradek> heh EMC2 is available in "Roman"
[21:47:10] <cradek> hmm, there are a lot of typos actually
[21:47:29] <alex_joni> cradek: I think they can correct it
[21:48:25] <alex_joni> cradek: guess we need a latin translation to make the doc fit :D
[21:48:32] <alex_joni> then it would be Roman :))
[21:48:49] <jepler> maybe they mean the font (times roman)?
[21:49:17] <cradek> "and so finds it's was into the software" should be "find its way"
[21:49:23] <SWPadnos> "AXIS is an interface fort he EMC system" ...
[21:49:34] <SWPadnos> (or similar)
[21:50:04] <cradek> it looks like they have very cool machines
[21:50:13] <cradek> they just need an editor :-)
[21:50:23] <SWPadnos> they're pretty darned neat, actually
[21:50:35] <SWPadnos> I played with one of their manual lathes at Photokina last year
[21:50:53] <SWPadnos> the CNC system is sold in a metal briefcase
[21:51:02] <cradek> heh, neat
[21:51:36] <cradek> wonder if they are using EMC2.1 yet
[21:52:02] <SWPadnos> Alex would be closer to knowing that than anyone else here, I think
[21:52:15] <SWPadnos> he knew I had been to the Photokina booth before I got home from Germany :)
[21:52:53] <alex_joni> hmm, I think so, but I am not sure
[22:05:01] <col> hi
[22:05:21] <alex_joni> hi
[22:15:20] <alex_joni> good night all
[22:21:50] <col> night
[22:26:05] <skunkworks> well - I think that went well
[22:31:29] <col> ?
[22:31:51] <skunkworks> sorry - house inspection and meeting the buyers.
[22:32:03] <col> ah right
[22:32:32] <col> iv been a busy bunny today
[22:32:38] <col> ripping a hurco mill apart
[22:32:39] <col> lol
[22:36:30] <skunkworks> sounds like more fun :)
[22:40:41] <col> yeah
[22:41:15] <col> managed to find the servo drives manual and wiring diagrams on net aswell so trying to figure that out
[22:41:26] <col> hopefully be able to use the old amps
[22:41:45] <col> and the spindle drive is a new eurotherm control so that should be easy enough
[22:52:30] <sebastienbailard> Hello all. Can anyone recommend a source of extra-long ball nose 3MM or < 1/8" diameter end mills?
[23:00:03] <jlmjvm> has anyone here run ubuntu and emc 2.1 on a computer with a pci express video card
[23:06:06] <Jymmmm> Any comments? http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93981 $62.99 -OR- http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=38123 $69.99 -OR- ???
[23:07:52] <SWPadnos> Jymmmm, #93981 says the motor is 1/2 HP instead of 1/3 forthe other one
[23:08:15] <Jymmmm> Yeah, also notice one doe s60deg and the other does 90 deg
[23:08:20] <SWPadnos> yep
[23:08:35] <Jymmmm> cheaper one does 1/2hp 60 deg.
[23:08:36] <SWPadnos> 45 and 60 though :)
[23:08:43] <Jymmmm> ?
[23:09:04] <SWPadnos> 45 degrees for the other one, 60 degrees tilt for the cheap one
[23:09:30] <SWPadnos> they weigh about the same, so I'd say they're pretty comparable
[23:10:17] <Jymmmm> I've never used one, just know I need a disc sander
[23:10:35] <Jymmmm> so I'm not sure about the belt stuff and if the 90 deg is important
[23:12:48] <SWPadnos> oh - 90 degree tilt of the belt part - I was looking at the table for the disc part
[23:13:34] <Jymmmm> Yeah, I'm dling the manual for the cheapy now.
[23:14:46] <Jymmmm> the manual says it's 1/3 hp, not 3/4hp
[23:15:17] <SWPadnos> it say s3/4 peak on the website, 1/2 "rated", whatever that means
[23:15:31] <SWPadnos> either will be pretty crappy actually
[23:15:45] <Jymmmm> how crappy?
[23:16:29] <SWPadnos> well, I have a Craftsman model that's almost identical, 1/3 HP
[23:17:04] <SWPadnos> if I try to sand the end of a 1.5 x 1/8" aluminum flat, the motor sometimes stops
[23:17:17] <SWPadnos> or sanding something like the end of a 2x4 will make it slow down a lot
[23:17:23] <SWPadnos> (or stop, if I keep pressing)
[23:17:42] <SWPadnos> this is with "medium" grit paper
[23:18:16] <Jymmmm> Ah, ok. Well I bought a bench grinder and attempted to convert it to a disc sander... not too pretty.
[23:18:24] <SWPadnos> nope - wrong loading
[23:18:58] <Jymmmm> it wasn't that, it was the wheel I made, not exactly square in multiple axis =)
[23:19:10] <Jymmmm> once mounted on the shaft that is
[23:19:11] <SWPadnos> heh
[23:19:29] <SWPadnos> you can get replacement backing plates for the disc portion, one of those could work in a grinder
[23:19:40] <SWPadnos> but I wouldn't do that myself
[23:20:16] <Jymmmm> Yeah, that's why I was thinking cheapy from HF
[23:35:57] <CIA-6> 03cradek 07TRUNK * 10emc2/bin/.cvsignore: new binary
[23:51:45] <col> anyone here know anythng about sinumerik system 8 lathe control?
[23:51:57] <col> made by siemens
[23:52:43] <jmkasunich> crepincdotcom: saw your comment about missing Z steps...
[23:52:52] <jmkasunich> you might want to read this http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TweakingSoftwareStepGeneration
[23:53:11] <jmkasunich> (may or may not be relevant)