#emc | Logs for 2010-04-09

Back
[00:00:29] <andypugh> Anyone ever looked at http://pokylinux.org/ ?
[00:01:27] <morfic> wow, only 2 years in gentoo, felt like longer, going to be 4 yrs this year since i retired? wow, shouldn't have looked, cause it makes you realize how time flies and how you get old
[00:01:28] <andypugh> I get the impression it is designed to be a minimalist embedded linux, and so might be a good fit for some EMC2 installs (if it works with RTAI, of course)
[00:01:41] <pfred1> SWPadnos plus I've heard some disturbing things about what is in store for Ubuntu in the future
[00:02:08] <morfic> pfred1: like?
[00:02:24] <SWPadnos> oh. well everyone is disturbed by something about Ubuntu :)
[00:02:34] <pfred1> morfic some advertising I forget all the gory details all I remember is when I heard it it didn't sound good to me
[00:02:52] <SWPadnos> andypugh, gentoo wasn't/isn't so much meant for embedded, it's meant for optimized builds for each machine it's run on
[00:02:57] <SWPadnos> which embedded needs, of course,
[00:03:18] <andypugh> I was talking about Poky, not Gentoo
[00:03:21] <pfred1> SWPadnos gentoo is meant for people who rarely clibm the basement stairs
[00:03:26] <pfred1> climb even
[00:03:31] <SWPadnos> heh
[00:03:46] <pfred1> I mean get real
[00:04:35] <pfred1> I seen them gentoo maniacs on IRC always rebuilding their systems over and over
[00:04:42] <SWPadnos> thinking about it from a purely cost accounting perspective, it's unlikely that optimization would get back the time spent compiling (even outside of compiling updates all the time)
[00:05:13] <SWPadnos> but it is pretty cool to make a system that lets people customize and build everything on their machine easily
[00:05:34] <pfred1> SWPadnos if you really want to do that you can do it on any Linux distro
[00:05:45] <SWPadnos> missing the "easily" part
[00:05:52] <pfred1> its not like they have the market corners on gcc or anything
[00:06:14] <SWPadnos> ebuilds are an interesting solution to a perceived problem
[00:06:49] <SWPadnos> and the fact that you can literally "emerge world" (or similar) to rebuild everything on the machine is pretty impressive
[00:06:52] <pfred1> I still don't think the ultimate Linux distro has been invented yet
[00:06:57] <SWPadnos> probably not
[00:07:03] <morfic> ebuilds sound less strange than "spells" that you "cast" as in sourcemage, but that aside, sourcemage guys are a lot more fun to hang around with than too many of the gentoo dev kids
[00:07:15] <SWPadnos> and the definition of ultimate is (a) different for everyone and (b) changes over time for most people
[00:07:22] <SWPadnos> heh
[00:07:25] <morfic> SWPadnos: portage 2.2 now has sets like "emerge @kde-live"
[00:07:49] <SWPadnos> cool. I wondered about subsets as I was typing that last thing
[00:08:18] <pfred1> 'emerge at the bottom of another core dump
[00:08:39] <SWPadnos> what was a real pain though was going through all the config files whenever you
[00:08:42] <VernonM> What's the Ubuntu equivalent of ctrl-alt-delete? :) EMC just locked up on me.
[00:08:44] <SWPadnos> did an update
[00:08:45] <morfic> i read the email i sent to gentoo-devel mailing list, sounds corny 4 yrs later
[00:08:50] <pfred1> welcome to the bleeding edge i hope you packed a lot of bandages
[00:08:51] <SWPadnos> VernonM, ctrl-alt-delete
[00:08:56] <VernonM> oh, duh :)
[00:09:07] <morfic> etc-update? not like ubuntu is not trying to sneak a config file past you once in a while :)
[00:09:15] <pfred1> VernonM try alt+F2 first
[00:09:40] <SWPadnos> yeah, it's a great thing to help you migrate configs, but if you're on the bleeding edge, or you have gone a few updates without fixing it all, it's a big job
[00:09:57] <morfic> it can be fun for the night, yes
[00:10:01] <VernonM> wow, it's locked solid. Wife clicked "show hidden files and folders" while opening an NGC, and it just...hung.
[00:10:24] <pfred1> VernonM you try to switch to another console?
[00:10:27] <SWPadnos> and it didn't seem to be able to coalesce changes, so if there were 10 updates to a single file, you had to go through 10 diffs
[00:10:33] <SWPadnos> VernonM, that's not good
[00:10:40] <SWPadnos> do the keyboard lights work?
[00:10:44] <VernonM> NOTHING works. First time this thing's ever locked up.
[00:10:51] <SWPadnos> can you ping it?
[00:11:06] <VernonM> yeah the keyboard lights work
[00:11:12] <SWPadnos> hmmm
[00:11:13] <pfred1> yeah i haven't seen a kernel crash in like 10 years now
[00:11:17] <SWPadnos> ctrl-alt-F1 then
[00:11:31] <VernonM> yeow, thanks haha
[00:11:40] <SWPadnos> did that do something?
[00:11:42] <VernonM> THAT sure did it
[00:11:43] <VernonM> yes
[00:11:46] <SWPadnos> huh
[00:11:53] <pfred1> still haven't seen one either i guess
[00:11:54] <morfic> pfred1: tuesday morning i had to power down the redhat server, something with it's spinlock made it show a nice bug trace, that thing is on 2.6.18-something
[00:11:56] <VernonM> punched me out to desktop login
[00:11:59] <SWPadnos> from there, log in, and then do the following command:
[00:12:09] <SWPadnos> sudo /etc/init.d/gdm restart
[00:12:16] <pfred1> init 1 ?
[00:12:19] <SWPadnos> you mean text mode login, or desktop?
[00:12:27] <VernonM> text mode login
[00:12:32] <SWPadnos> ok, phew :)
[00:12:34] <VernonM> haha
[00:12:36] <VernonM> sorry
[00:12:41] <SWPadnos> no problem
[00:13:32] <SWPadnos> I guess they have ctrl-alt-del disabled or something. that should have worked if the keyboard and kernel were still working (and ctrl-alt-f1 shows they were)
[00:13:46] <pfred1> by default Ubuntu disables ctrl+Alt+backspace to kill X doesn't it?
[00:13:52] <morfic> pfred1: right
[00:13:57] <pfred1> booo!
[00:14:04] <SWPadnos> but not ctrl-alt-del, I'd think
[00:14:04] <morfic> can't zap it, hate it
[00:14:19] <SWPadnos> it's a setting somewhere (ctrl-alt-bksp)
[00:14:25] <pfred1> yeah why the hell would they get rid of something like that?
[00:14:31] <morfic> pfred1: someone said it can be hit too easy because it is similar to something people use in their editors?
[00:14:34] <SWPadnos> so people don't hurt themselves
[00:14:46] <SWPadnos> (silly)
[00:14:56] <SWPadnos> silly Ubuntu that is, not silly you
[00:14:57] <pfred1> I can
[00:15:01] <VernonM> I wish I could make the HOME key a two-key command on my little mill with no home switches
[00:15:06] <morfic> Option "DontZap" "False" in theory, but i think it's an actualy two part problem
[00:15:15] <VernonM> that reboot worked, thanks a lot.
[00:15:18] <pfred1> I can't seem to change resolution with ctrl+alt+KP + and - in Ubuntu either
[00:15:26] <SWPadnos> VernonM, did you re-run EMC2?
[00:15:33] <SWPadnos> (don't, if you haven't)
[00:15:41] <morfic> i think they actually unmapped it? i am not sure one distro unmapped it, one just defaults to dontzap true
[00:16:30] <VernonM> trying but it wont start
[00:16:37] <SWPadnos> figured :)
[00:16:53] <SWPadnos> open a terminal (Applications -> Accessories -> Terminal)
[00:17:09] <pfred1> make a shortcut on your desktop for gods sake!
[00:17:25] <VernonM> hhahaha I already have
[00:17:25] <andypugh> OK, log-off time
[00:17:26] <SWPadnos> if you're not familiar with http://pastebin.ca/ , please go there now :)
[00:17:29] <SWPadnos> heh
[00:17:31] <SWPadnos> ok
[00:17:33] <pfred1> I got one for an -rv rxvt with a font I like
[00:17:44] <SWPadnos> please pastebin the output of the following command:
[00:17:54] <SWPadnos> lsmod | grep rt
[00:18:16] <SWPadnos> incidentally, what version of EMC2 do you have?
[00:19:24] <VernonM> the newest version! I'll look in a sec. rebooting.
[00:19:40] <VernonM> This thing has worked flawlessly for more than a year
[00:20:09] <SWPadnos> bummer. before rebooting, it's possible to try to see what went wrong
[00:20:11] <VernonM> the only thing new today, ( I run two keyboards) is a different second keyboard.
[00:20:15] <SWPadnos> but not after :)
[00:20:26] <SWPadnos> huh. shouldn't be an issue
[00:20:57] <VernonM> yeah, I know... but it's the only thing new, like, five minutes ago new.
[00:21:40] <pfred1> ulimit -c unlimited
[00:22:32] <pfred1> core files are always fun to rummage through
[00:22:55] <SWPadnos> dmesg is sometimes interesting as well
[00:23:04] <MarkusBec> MarkusBec is now known as MarkusBec_away
[00:23:24] <SWPadnos> as well as trying to do things like stop the realtime system and things like that
[00:23:40] <pfred1> SWPadnos all dmesg ever does is remind me of all the broken stuff on my system I never bother to fix
[00:23:47] <SWPadnos> heh
[00:24:00] <SWPadnos> well, if you just had a breakage, and you wanted to fix it, dmesg would be your friend ...
[00:24:17] <pfred1> I see it and go oh yeah I was going to do something about that someday ...
[00:24:46] <pfred1> thats the other really annoying thing about ubuntu hiding all the boot up messages
[00:25:31] <pfred1> I've always setup systems to text login then I'd run startx
[00:25:50] <pfred1> just in case X ever got borked I wouldn't be "locked out"
[00:26:12] <SWPadnos> debian kind of made that odd. they mixed runlevels 2 and 3, so there's no "text mode, multiuser" by default
[00:26:21] <SWPadnos> 2 and 3 are identical
[00:26:23] <pfred1> eeewww
[00:26:31] <SWPadnos> yeah, been that way for some years now
[00:26:35] <pfred1> well all I really need is 1 3 and 6
[00:26:44] <SWPadnos> 1 is single user, isn't it?
[00:26:50] <pfred1> indeed it is
[00:26:55] <SWPadnos> single user/text
[00:27:01] <pfred1> 1 is oh junk things have gone off the rails
[00:27:10] <SWPadnos> yeah, "maintenance mode"
[00:27:21] <pfred1> I drop to it to kill everything then reinit to another run level usually
[00:27:27] <SWPadnos> 2 is where most servers should sit, or PCs that you don't always want X on
[00:27:59] <pfred1> init 1 is almost a soft boot but doesn't kill your uptime
[00:28:12] <VernonM> Well it's working fine on just one keyboard. Thanks, all.
[00:28:14] <pfred1> pretty much wastyes all userland programs
[00:28:42] <SWPadnos> sure VernonM, better luck next time (or next keyboard or whatever :) )
[00:29:08] <pfred1> my keyboard goes south once in a while I unplug it then plug it back in
[00:30:49] <pfred1> they should auction off convicted spammers on ebay
[00:31:00] <pfred1> I'd probably buy a few ...
[00:31:27] <VernonM> I have one keyboard on the lathe, and one on the mill, and just changed the mill keyboard to a second of the flat rubber roll-up keyboards. perhaps two of the same device are causing some sort of conflict.
[00:31:40] <SWPadnos> nope, shouldn't be
[00:31:52] <SWPadnos> what emc2 version was that?
[00:32:12] <pfred1> SWPadnos should i ready the rack?
[00:32:24] <SWPadnos> well, if you like. I don't need it
[00:32:32] <SWPadnos> got better things to buy than spammers ;)
[00:32:38] <pfred1> VernonM confess!
[00:33:14] <VernonM> oh yeah. version 2.3.0
[00:33:25] <SWPadnos> eek. that's a ways out of date
[00:33:32] <VernonM> what? no way
[00:33:38] <VernonM> really?
[00:33:39] <pfred1> yeah thats not that old
[00:33:39] <SWPadnos> 2.3.5 is the latest
[00:33:43] <VernonM> argh
[00:34:20] <SWPadnos> almost exactly a year old
[00:34:41] <pfred1> SWPadnos so how do you upgrade it?
[00:34:47] <VernonM> wow, I just got it, maybe two months ago
[00:34:52] <SWPadnos> apt-get update
[00:35:01] <SWPadnos> and then upgrade whatever you want
[00:35:06] <SWPadnos> huh
[00:35:09] <SWPadnos> which OS?
[00:35:18] <SWPadnos> liveCD from linuxcnc, or something else?
[00:35:28] <pfred1> thats what i got
[00:35:40] <VernonM> started off with the live cd, but updated recently
[00:35:52] <SWPadnos> did you update first?
[00:36:02] <SWPadnos> (does that computer live on the net?)
[00:36:20] <pfred1> only when i turn it on
[00:36:34] <SWPadnos> err, that was for VernonM
[00:36:40] <VernonM> mine is only connected to update.
[00:36:46] <SWPadnos> ok
[00:36:56] <SWPadnos> using synaptic?
[00:36:57] <pfred1> VernonM time to connect
[00:37:32] <VernonM> yep yep
[00:37:51] <VernonM> ugh, wife took my 25foot cable
[00:38:10] <pfred1> kinky!
[00:38:11] <SWPadnos> ok, remember to hit "reload" before "mark all upgrades for installation" (or however that's spelled)
[00:38:13] <VernonM> hahaha
[00:38:37] <VernonM> roger that
[00:38:42] <pfred1> SWPadnos isn't it just easier to apt-get update?
[00:39:02] <SWPadnos> depends on whether you're used to using synaptic or the terminal
[00:39:06] <SWPadnos> reload does the update
[00:39:54] <Valen> ubuntu has nerfed ctrl + alt + backspace now
[00:39:57] <pfred1> I like synaptic for browsing new packages but for specific stuff the console seems more efficient
[00:40:11] <SWPadnos> the crash on "show hidden files" was fixed last June, and should be in 2.3.2 or later
[00:40:26] <VernonM> yeah, that's the problem too, I just tried it
[00:40:37] <VernonM> and now my wife's all "told you so"
[00:40:40] <pfred1> VernonM ls -a
[00:40:51] <VernonM> it's locked
[00:40:54] <VernonM> again.
[00:41:00] <SWPadnos> well don't do that
[00:41:05] <SWPadnos> does the mouse move?
[00:41:16] <VernonM> yes\
[00:41:23] <SWPadnos> ok, then kill axis
[00:41:36] <pfred1> xkill !
[00:41:37] <VernonM> cant
[00:41:38] <SWPadnos> may be hard, since it's probably using all the CPU or something
[00:41:47] <SWPadnos> terminal -> sudo kill -9 axis
[00:41:55] <SWPadnos> or killall -9 axis
[00:41:58] <pfred1> killall -9
[00:42:03] <SWPadnos> (may work better)
[00:42:07] <SWPadnos> sitll with sudo, maybe
[00:42:32] <pfred1> I always wanted to install the quake process killer
[00:42:38] <SWPadnos> heh
[00:42:47] <SWPadnos> BFG9000, baby!
[00:43:01] <Valen> system monitor in system administration might be a gui way of doing it
[00:43:13] <pfred1> I was screwing around with an old PC here the other day and played some original doom on it
[00:43:30] <pfred1> man that music can get annoying
[00:43:37] <SWPadnos> I really should get that dual Opteron running again. Wolf:ET really looks great at 3840x2400
[00:43:57] <pfred1> SWPadnos I'm lethal with he barney Mod
[00:44:06] <SWPadnos> it's pretty amazing how well the sounds worked in Doom II
[00:44:26] <pfred1> I got a whole CD someplace of wolfenstein mods
[00:44:35] <VernonM> hahahaha yeah I remember those
[00:44:42] <SWPadnos> the cacodemon and whatever the pink horned thing was called really could raise the hair on your neck
[00:44:57] <SWPadnos> I had the barney mod - it got almost as annoying as Barney
[00:45:16] <pfred1> well I'd kill barney like he was going out of style
[00:45:31] <pfred1> I can play a mean game of xbill too you know?
[00:46:01] <VernonM> haha
[00:46:19] <VernonM> xbill
[00:46:36] <pfred1> it gets bad when the fires start on the network cables
[00:46:53] <pfred1> then you don' know whether to squash bills or bucket
[00:49:26] <VernonM> well, I've gotta disconnect to update. thanks again for the help!
[00:49:39] <SWPadnos> sure. enjoy
[00:49:45] <VernonM> haha
[00:49:47] <SWPadnos> and the keyboard isn't the problem :)
[00:49:54] <VernonM> yeah I SEE that :)
[00:49:57] <VernonM> hahahah
[00:50:21] <pfred1> dang Malcolm McLaren dead at 64
[01:08:21] <morfic> wow, pfred1 sounds a lot like me.... (re: dmesg reminding him of things to fix)
[01:11:28] <pfred1> morfic I always have too many brands in the fire to get around to doing everthing I should
[01:12:43] <morfic> i should change things to be automatic on boot, but as rarely as i boot, i may as well bring this or that up manually *cough*
[01:13:19] <pfred1> this ubuntu is so different from what I'm used to I donno what works anymore and what doesn't now
[01:13:32] <Valen> 10.04?
[01:13:40] <pfred1> 9.1
[01:14:04] <pfred1> my last distro before this was suse 8.2
[01:14:17] <pfred1> which was almost 7 years old
[01:14:43] <pfred1> and I thought that was new fangled!
[01:22:00] <pfred1> I have to prototype this tomorrow: http://img52.imageshack.us/img52/6181/tb6560ahqr3.png
[01:24:51] <john_f_> pfred1: what is it?
[01:25:11] <pfred1> john_f_ I'll let you know but it's supposed to be a stepper motor driver
[01:26:08] <pfred1> it could just be an early 4th of July celebration device
[01:26:26] <john_f_> I have built some of those a long time ago
[01:26:42] <john_f_> made my own microstepping drive
[01:26:44] <pfred1> oh I still build suprises every now and again
[01:27:08] <pfred1> oh yes this is supposed to microstep 16th
[01:28:42] <john_f_> pfred1: http://www.machineability.com/Stepper.html
[01:28:58] <pfred1> john_f_ this is the main IC in that board: http://www.toshiba-components.com/motorcontrol/pdfs/TB6560AHQ_AFG_E_2003_20080407.pdf
[01:30:54] <pfred1> john_f_ you should be real familiar with whats in the 6560 datasheet
[01:31:55] <john_f_> yes I used all descrete FETs What you have should make it easier
[01:32:21] <pfred1> john_f_ yeah especially being as they're only $4.73 a module
[01:33:14] <pfred1> I hear you're lucky if you can get 1.75 amps per phase out of them though
[01:33:57] <john_f_> my steppers were old and needed 5 amps
[01:34:20] <pfred1> yeah you'd be smoking the place up trying to run those on these
[01:35:36] <pfred1> my plan is to put two motors each on the X and Y axises
[01:35:58] <john_f_> two motors on the same lead screw?
[01:36:09] <pfred1> no on parallel lead screws
[01:36:44] <john_f_> I always wondered if you could put 2 motors on the same axis
[01:36:49] <pfred1> though i did toy with the idea of 4 motors on an axis I may even try it someday
[01:36:53] <Valen> you can
[01:37:03] <pfred1> I hear if you power them up and tighten them it may be possible
[01:37:08] <Valen> do you mean on the same screw or in a gantry system?
[01:37:29] <john_f_> I mean on the same screw
[01:37:29] <pfred1> dual gantry system
[01:38:22] <pfred1> john_f_ i hear what people do is have each motor hold then tighten them to the shaft
[01:38:45] <Valen> if its steppers you can easily, just link them togther, for servo's you need some more fancy ness but it all seems to work
[01:39:18] <john_f_> so if the stepps dont match exaclty then you wont get 2 times the torque
[01:39:31] <john_f_> but maybe close to that
[01:39:45] <pfred1> beats having a system that racks
[01:40:01] <pfred1> dual X is common today
[01:40:08] <pfred1> on moving gantry at least
[01:41:30] <pfred1> john_f_ like this: http://www.joescnc.com/themachines-hybrid.php
[01:42:16] <pfred1> but if its such a good idea why not do it on the Y axis too?
[01:42:36] <morfic> morfic is now known as Machinist
[01:43:10] <john_f_> probably not needed on the Y axis
[01:43:36] <pfred1> I hate how the Z looks so unbanalced with those setups
[01:43:42] <pfred1> unbalanced even
[01:44:10] <john_f_> OK you could fix that with 2 Y screws
[01:44:16] <john_f_> I see your point
[01:44:38] <pfred1> sort of like how this looks: http://www.mechmate.com/
[01:44:47] <pfred1> just looks more stable to me
[01:45:01] <john_f_> yep
[01:46:21] <john_f_> Looks like my dog wants out of the shop so I guess I will call it a day and go back to the house
[01:46:33] <pfred1> that and you cna get weak stepper motors dirt cheap compared to the bigger ones everyone wants for their CNC machines
[01:47:50] <Machinist> pfred1: that mechmate has custom cabinets written all over it
[01:48:17] <Machinist> cabinets as in speakers
[01:48:27] <calcite> can one drive power multiple steppers within the drives amp range? parallel stack motors ;-)
[01:49:18] <SWPadnos> no
[01:49:19] <calcite> I couls see an array of four 23 frame motors on a common belt
[01:49:34] <calcite> too bad ;-( I din't think so
[01:49:51] <SWPadnos> the motors will move, but it will suck
[01:50:18] <calcite> suck like in "that's no good" ? ;-)
[01:50:22] <SWPadnos> yep
[01:51:23] <SWPadnos> if you parallel them, you get uneven current (no two motors are identical, and they'll have different loading too), if you put them in series, you increase inductance, which will reduce the maximum speed for a given voltage
[01:51:39] <SWPadnos> in addition to having them not share power equally
[01:52:12] <SWPadnos> and the anti-resonance circuitry (if there is any) won't know what the heck to do
[01:52:22] <SWPadnos> all that's AFAIK
[01:52:24] <calcite> 1 8 amp motor just looks so elegant compared to a plethora of lesser devices
[01:52:33] <SWPadnos> I haven't tried it :)
[01:53:08] <pfred1> calcite yeah until you see the power supply for it
[01:53:16] <calcite> Regarding the power sharing, as a descrete step device, power sensing is immaterial
[01:54:04] <SWPadnos> not for the driver
[01:54:16] <calcite> pfredl Parker S6 drives can be picked off the -bay for less than $50
[01:54:35] <calcite> oem750s for similar
[01:54:44] <pfred1> calcite i can make 5 drivers for that
[01:54:45] <SWPadnos> don't those use a DC supply?
[01:54:57] <calcite> How much for the smoke?
[01:55:32] <calcite> the OEM's require a DC supply, the S6 is plug in to a wall outlet.
[01:55:46] <SWPadnos> ah
[01:56:32] <calcite> pfred1 8 amp units, at $10 each Puff proof?
[01:56:43] <pfred1> calcite or already poofed
[01:56:51] <calcite> or should that be 6 amp
[01:57:56] <pfred1> I'm just doing this because its an interesting electronics project
[01:58:10] <pfred1> same reason I built my own computer out of chips once
[01:58:11] <calcite> I admire your creative juices, I'm not much for transistors and such
[01:58:31] <pfred1> same reason i build audio stuffs
[01:58:48] <calcite> Something wants to do the electronics, but.......
[01:58:56] <pfred1> though I have a marshall next to one of my amps and i have to admit I like my amp a little better ;0
[01:59:05] <calcite> something about reinventing the wheel
[01:59:26] <calcite> You like it cause you know all about it.
[01:59:33] <pfred1> hardly
[01:59:55] <pfred1> I like it because i find it challenging
[02:00:13] <pfred1> making stepper motors run well is not easy
[02:00:13] <calcite> the alternative is what I call the "Black box frustration" If we don't know how a device works, It frustrates at some level
[02:00:38] <pfred1> out of all the drives I've built so far only one design performs
[02:01:08] <Valen> I was meaning when I said to parallel them to use multiple drivers and steppers one driver per stepper and just feed them differently
[02:01:22] <pfred1> but hopefully with these new modules i jus got i can put another notch on my stepper handle there
[02:01:22] <Valen> alternatly if you connected the phases in series it'd probably work ok
[02:02:01] <pfred1> Valen in minumum configuration you can make these modules run with like 6 external parts
[02:02:46] <pfred1> heck they even threw pull down resistors into these things
[02:02:47] <calcite> pfred1 How many parts to provide protection from "idiots"?
[02:03:16] <pfred1> calcite I don't even bother no matter what you do a bigger idiot always comes along
[02:03:50] <calcite> I'm looking at a pair of RTA 6 amp drives, heck, there are a couple thousand parts in each (well, perhaps a few less)
[02:04:24] <calcite> Idiot is greater than smart, there is no end in idiot
[02:05:12] <calcite> Did you get your package of devices yet?
[02:05:21] <pfred1> one of them
[02:05:25] <pfred1> still waiting on another
[02:05:33] <calcite> progress
[02:05:46] <pfred1> yeah my library element pads line up and everything
[02:05:52] <pfred1> I did it from the datasheet
[02:06:09] <calcite> mirror mirror on the board.....
[02:06:22] <pfred1> I got the whole circuit all worked out to prototype just have to build it now see what needs to be tweaked
[02:07:05] <pfred1> I was going to machine up a heatsink before they arrived but I figured that'd be pushing it
[02:08:25] <Jymmm> Now, THIS is a heatsink... http://farm2.static.flickr.com/1064/980074783_b772ce34e7.jpg
[02:08:37] <Jymmm> (two of em for sale too)
[02:08:58] <pfred1> Jymmm look in high watt stereos
[02:09:10] <pfred1> I've seen better
[02:09:19] <pfred1> I got a whole box of heatsinks
[02:09:35] <Jymmm> Was for BP, been cut down to 8" long from 48"
[02:10:28] <pfred1> well some of the best and easiest to find heatsinks are in cast off stereos
[02:10:44] <calcite> I thought you might try the tubof veggie oil and plumbing to the kitchen refridgerator like the overclockers do
[02:12:31] <calcite> I wonder what level of current your little devices could carry then?
[02:12:46] <calcite> 3.5 A?
[02:12:57] <pfred1> nah
[02:13:13] <pfred1> I hear they're only good for like 14 watts
[02:13:25] <pfred1> Jymmm http://img88.imageshack.us/img88/6150/p7100098.jpg
[02:13:35] <calcite> smoke bubbles in oil nice visual
[02:14:15] <pfred1> calcite unless you're really careful you can't expect more than 1.75 amps per phase out of these modules
[02:14:51] <calcite> I do need to find some 80V DC caps for the supply to power these RTAdrives. Any good leads?
[02:15:20] <pfred1> again high power stereos
[02:15:37] <calcite> pfred1 what does "careful" mean. Be nice, I'm no electrinics guy ;-)
[02:15:39] <pfred1> anything over 150 watts a channel should have 80VDC caps in it's power supply
[02:16:23] <calcite> I snagged some from similar...45 V on the lable..;-(
[02:16:25] <pfred1> calcite careful would mean external freewheeling diodes and some other protection
[02:16:36] <calcite> they are big though ;-)
[02:16:48] <calcite> OK
[02:17:36] <calcite> Every time I hear the term "Freewheeling diodes" I think of the Fabulous Furry Freak brothers" ;-)
[02:17:51] <calcite> Freewheeling Franklin
[02:21:27] <pfred1> I liked that heatsink so much I had to use it again: http://img713.imageshack.us/img713/2244/pict0579s.jpg
[02:22:39] <Machinist> same board as i used for my headphone amp it looks like
[02:22:52] <pfred1> yeah its a rat shack protoboard
[02:23:19] <pfred1> that power supply is so mean its like a little constant voltage welder
[02:23:33] <Machinist> i "mislooked", the RS one is close, but not the one i have
[02:23:46] <Machinist> that does sound mean
[02:24:05] <pfred1> when i connect stuff to it it arcs and makes contact stick
[02:24:36] <Machinist> it's a good one :)
[02:24:45] <pfred1> here's me load testing it and it didn't break a sweat: http://img196.imageshack.us/img196/6414/pict0576h.jpg
[02:25:14] <pfred1> theoretically it should put out 3 times what its doing there
[02:25:24] <pfred1> but I sort of didn't want to fry it
[02:25:55] <pfred1> but its putting out 100 watts there
[02:27:07] <pfred1> well 95.9
[02:27:26] <Machinist> oh wow, you almost tricked me there, big time
[02:27:28] <Machinist> :P
[02:27:31] <Machinist> Machinist is now known as morfic
[02:27:55] <pfred1> I've found out recently that my ammeter can read a little low though
[02:28:07] <pfred1> well one of the twins does i can't remember which
[02:28:45] <morfic> still havent upgraded my multimeter, i want one i can trust more, and add capacitance, but i worry about that when i actually need that again
[02:28:47] <pfred1> the needle is getting hung up on the upper range of the scale but only for amperage
[02:29:15] <pfred1> it does fine when i use the ohms
[02:29:35] <morfic> how odd
[02:29:44] <pfred1> yes it is
[02:32:01] <pfred1> here's the schematic for that power supply: http://hmin.tripod.com/als/andysm/drawings/supplies/ps2-30v.gif
[02:32:20] <pfred1> I built a few off that guy's site i liked this one the best
[02:33:05] <pfred1> its nice and clean
[02:38:21] <calcite> What is special about a BD243C?
[02:38:44] <pfred1> you got me i think I used a 2n3904
[02:38:54] <calcite> and are all four power transistors on the sink?
[02:39:03] <pfred1> or 2n3906 whichever is the npn I always get them confused
[02:39:19] <pfred1> the 243 is a small signal
[02:39:38] <pfred1> the 3055s are power transistors i used some NECs i had laying around
[02:39:43] <calcite> yes I see that now
[02:40:29] <calcite> help me understand the term 2V-30V
[02:40:34] <pfred1> any TO-3 NPNs will do just about 3055s are pretty common
[02:40:44] <pfred1> V = volts
[02:40:50] <calcite> yes
[02:41:06] <calcite> and the 2V?
[02:41:12] <pfred1> 2 volts
[02:41:26] <calcite> superimposed?
[02:41:46] <pfred1> its adjustable from 2 to 30 volts
[02:42:10] <calcite> OH! Sure (hitting myself on the forehead...sheesh!
[02:42:20] <pfred1> and its current limiting too
[02:42:43] <calcite> U adj 20K 10rev .....
[02:42:50] <pfred1> you can adjust whatever voltage and current you want it to put out
[02:43:14] <calcite> the 1K ten turn?
[02:43:26] <pfred1> yeah thats the current
[02:43:35] <pfred1> and the 20K is the voltage
[02:43:42] <calcite> Whao that is nice!
[02:43:56] <calcite> How clean does the input need to b?
[02:43:59] <pfred1> yeah it even works
[02:44:24] <pfred1> I used a transformer and bridge and a cap
[02:45:05] <pfred1> the big beer can looking thing in this pic: http://img713.imageshack.us/img713/2244/pict0579s.jpg
[02:45:29] <calcite> What does the 723 cost qty 2?
[02:45:32] <pfred1> the bridge is the little square thing to the far right with the red wire coming off it
[02:45:43] <pfred1> maybe a dollar
[02:45:46] <calcite> Yes Yes, I see
[02:45:55] <pfred1> I got a pile of them
[02:46:15] <calcite> do the pots need to be anything special? carbon?
[02:46:25] <pfred1> for 2 maybe a dollar a quarter a piece wouldn't suprise me either
[02:46:40] <pfred1> from the right supplier
[02:47:13] <pfred1> they're pretty nice for what they are they only put out like 150 ma on their own though
[02:47:30] <pfred1> most people just use the 3 terminal regulators today
[02:47:36] <calcite> a 2-24V current limiting PS is a useful device
[02:47:52] <pfred1> it has a place of honor on my workbench thats for sure
[02:48:02] <pfred1> its my go to power supply
[02:48:18] <pfred1> though I've made dozens over the years now
[02:48:21] <calcite> how does it handle dead shorts or opens?
[02:48:45] <pfred1> well it sits open on my bench a lot there's a resistor load on it all the time
[02:48:56] <pfred1> and it should current limit on a short
[02:49:11] <calcite> I've got a commercial 100V DC supply that does not play well if left un connectd
[02:49:25] <pfred1> slap a resistor across it
[02:49:36] <calcite> Yep
[02:49:48] <pfred1> the 390 ohmer on the schematic I linked to is the constant load
[02:50:09] <pfred1> 390 ohms 5 watts
[02:50:22] <calcite> mostly the bench supply does everything I need (Sorenson 30V 20A)
[02:50:45] <calcite> Ahh the five watt
[02:51:17] <calcite> Would that be a "free wheeling diode " just next to it?
[02:51:34] <calcite> 10A!
[02:51:36] <pfred1> I believe it would be i got mine out of a PC power supply
[02:51:56] <pfred1> every PC switching power supply has a heavy back to back diode pair in it
[02:52:11] <calcite> OH Good fact!
[02:52:18] <pfred1> looks like a power transistor but they're really 2 back to back diodes rated at 20 amps or so
[02:52:30] <pfred1> and they even have pics of diodes on them too
[02:52:37] <pfred1> like the symbol
[02:53:01] <calcite> Have you had a look at the scope when this circuit is pulled to max?
[02:53:10] <pfred1> yeah they're sweet I htink they're used with the ciol somehow
[02:53:32] <pfred1> its clean power
[02:53:35] <pfred1> its linear
[02:53:45] <morfic> how does it sound? (no seriously, i wonder what it would do on audio equipment)
[02:53:51] <calcite> certainly a motor would run on one.
[02:54:07] <pfred1> this is it as far as power goes
[02:54:19] <pfred1> maybe a choke thrown at it wouldn't hurt it
[02:54:29] <calcite> Linear? does that mean a straight line until it falls flat?
[02:54:32] <pfred1> but as it is it's clean linear power
[02:54:45] <pfred1> it's not a switching power supply
[02:54:55] <pfred1> its analog all the way
[02:55:17] <pfred1> it is basically an amplifier
[02:55:21] <calcite> are there any critical value components?
[02:55:37] <pfred1> the current sharing resistors
[02:55:49] <pfred1> the .1 ohms off the power transistor emmiters
[02:55:55] <calcite> the .1's?
[02:55:58] <pfred1> yes
[02:56:07] <pfred1> they're so the transistorsd share the load equally
[02:56:24] <pfred1> you wouldn't want one to do all the work
[02:57:03] <pfred1> I used pairs of .25s because its what i had a lot of
[02:57:05] <calcite> 2%?
[02:57:46] <pfred1> just make sure they're all really the same is all
[02:58:17] <pfred1> I mean when i put this thing together I built it from junk i had lying around
[02:58:37] <calcite> Have you seen "The Power Supply Handbook" by the editors of 73 magazine? a TAB publication 1979. I have it here on my desk, just got it from the shop in reference
[02:58:51] <pfred1> dark brown book?
[02:59:02] <pfred1> maroon even?
[02:59:07] <calcite> paper back for me.
[02:59:29] <pfred1> I like national semiconductors a bit better honestly
[02:59:51] <pfred1> my nat semi book is just about falling apart
[03:00:28] <pfred1> I wonder if its online?
[03:01:27] <pfred1> all it says is Voltage Regulator handbook National Semiconductor Corporation
[03:01:37] <pfred1> oh and 1982
[03:01:37] <calcite> just paged until I found a design using the 723, that device goes back
[03:01:50] <pfred1> oh yeah the 723 is from the 70s
[03:02:01] <pfred1> I pull them out of old equipment all the time
[03:02:59] <pfred1> this book has no code or publicatin number or nothing
[03:04:23] <pfred1> http://siris-libraries.si.edu/ipac20/ipac.jsp?uri=full=3100001~!104761!0
[03:05:06] <calcite> designs based on devices?
[03:05:36] <pfred1> its mainly datasheets but it has chapters on designing power supplies
[03:05:52] <pfred1> all i know is my copy is about to fall apart i use it so much
[03:06:37] <pfred1> years ago before the net I used to buy the databooks
[03:06:43] <pfred1> and while I'
[03:07:00] <pfred1> and while I'm working on something I still like to have a book open to a device to check the pinouts
[03:07:31] <pfred1> I have a fairchild TTL databook that has seen some usage too
[03:07:37] <calcite> To change perspective, I wired up the 480 Volts to the VFD powering the lathe today. Single phase 240V in, 480 V 1ph out of a 3ph transformer. Supply to a 22kVa VFD at 400V and 120 cycles. I have "overclocked" my lathe motor Viola 10HP where there used to be but 5!
[03:08:48] <calcite> I would like to spend more time at the electronics bench....I just don't
[03:09:15] <pfred1> ah I go in cycles I'll do it a lot then other things come up and i don't
[03:09:29] <calcite> I have a few other handy design bookson the shelf....waiting
[03:09:38] <pfred1> but off and on I've been into electronics since i was about 7
[03:09:59] <calcite> But these new devices keep popping up...like your motor driver
[03:10:17] <pfred1> yeah i didn't even know about it until recently myself
[03:10:40] <pfred1> for what it is within limits it seems like hot stuff to me
[03:12:04] <pfred1> at $4.73 a device i mean what else can anyone ask for?
[03:12:11] <calcite> too bad all my ideas tend to include the concept of HorsePower;-)
[03:12:25] <calcite> not inch #
[03:12:51] <mozmck> like, how many horsepower does that chip put out?
[03:13:20] <Valen> with the exception of a few fets , not many chips put out "horsepower"
[03:13:20] <calcite> can that device be palced in a power amplifier circuit?
[03:13:25] <pfred1> 0.018666667
[03:13:47] <pfred1> whopping horsepower!
[03:13:49] <calcite> convert!
[03:14:10] <calcite> could you express that in kVA?
[03:14:17] <calcite> ;-)
[03:14:25] <calcite> or NM
[03:14:45] <pfred1> if i needed to
[03:14:51] <calcite> no need!
[03:15:16] <pfred1> hey when these things blow up I hear they are a hell of a bang!
[03:15:43] <pfred1> some kid posted a pic of one he blew up thing had a big hole in it
[03:16:00] <calcite> I have a pair of rectifing diodes that must weigh a pound. DC Welder attachment for the buzz box...someday
[03:16:27] <calcite> Oh did the smoke get away?
[03:16:34] <pfred1> I have a miller CC TIG that does stick too
[03:16:43] <calcite> I want that!
[03:16:52] <pfred1> its my big power supply
[03:17:00] <pfred1> 200A
[03:17:02] <calcite> backwards compatability
[03:17:23] <pfred1> an econotig
[03:17:44] <calcite> I've used the buzz box to demagnitize some pretty good sized chunks of metal
[03:17:50] <pfred1> you can't set the frequency on it and it sucks at aluminum or I suck at it
[03:18:02] <pfred1> I have a degaussing plate
[03:18:03] <calcite> need hi freq
[03:18:17] <calcite> for alu
[03:18:17] <pfred1> I use it too
[03:18:40] <calcite> A hammer works for some applications
[03:18:40] <pfred1> well they say it does it and i have but its pretty spotty to the point where i don't even bother trying anymore
[03:19:03] <pfred1> i got aluminum stick rods that work pretty good
[03:19:16] <calcite> I have not tried
[03:19:26] <pfred1> yeah well at $150 a box ...
[03:19:52] <pfred1> and that was years ago who knows what they charge for the things now
[03:20:07] <calcite> a box might last a long time
[03:20:25] <pfred1> not really
[03:20:34] <pfred1> just 5 pounds
[03:21:09] <pfred1> I'v met people who swear by the aluminum solder too
[03:21:42] <pfred1> I had someone give me a stick once I never bothered to evne try it out I pretty much avoid aluminum
[03:21:47] <calcite> from the LM723 data sheet 40V in is MAX, what have you been giving it?
[03:22:30] <calcite> I've hundreds of hours behind a tig tourch, but I ned a good machine to do good work.
[03:22:32] <pfred1> hmmm over 27.4
[03:22:41] <calcite> A poor craftsman blames his tools
[03:22:59] <calcite> 27.4 is safe compared to 40
[03:23:03] <pfred1> I hardly ever weld anymore
[03:23:22] <pfred1> 27.4 would be RMS its like a third over that peak
[03:23:23] <calcite> Only if I need to repair something
[03:23:47] <pfred1> yeah thats where I'm at with welding if its broke I'll weld it back together but I don't fab with it like i used to
[03:24:02] <calcite> yep. is the data sheet rms or peak? I don't know I'll read again
[03:24:11] <pfred1> I'm sure its max peak
[03:24:48] <pfred1> I tied 2 windings of a transformer together wait they're only 22.7 I think
[03:25:03] <pfred1> they were the 5 volt rails for an old mini computer
[03:25:34] <pfred1> so they were probably each good for about 30 amps or so
[03:26:11] <calcite> the "data sheet" I have open is only a sales schpeel No data just comments
[03:26:12] <pfred1> like 10 gauge wires coming out of that transformer
[03:29:20] <calcite> 10 ga and 30 amps is a match
[03:29:37] <pfred1> this is a big transfirmer
[03:29:42] <pfred1> transformer
[03:29:49] <calcite> I was using #8 THHN today
[03:29:50] <pfred1> computer was the sixe of a closet it ran
[03:30:15] <calcite> 15kVA for me today
[03:30:28] <pfred1> it had like garden hose sized 220V power cord
[03:31:54] <calcite> time to go! thanks for the electronics inspiration!
[03:32:25] <pfred1> another Mr. Wizard is born!
[05:36:13] <tom3p> opinions on refurb dell dimension 2400 for emc2 stepper system? http://tinyurl.com/ybpnsgf
[05:36:40] <tom3p> i need more than the single pci slot on the micro form factor pcbs
[06:27:21] <renesis> okay case, psu, mem kit, hard drive
[06:27:29] <renesis> what else do i need for an atom board
[06:27:33] <renesis> thats it i guess
[06:28:08] <alex_joni> spares..
[06:28:55] <renesis> wtf spares
[06:29:29] <renesis> okay what like they will stop selling ddr2
[06:29:34] <renesis> okay probably
[06:30:15] <renesis> guys there is some crazy stepper mill at work
[06:30:32] <renesis> big fat steppers
[06:30:48] <renesis> for X and Y i cant see what there is for Z
[06:31:08] <renesis> the people that know what it is are gone so i prob have to figure it out
[06:33:01] <alex_joni> what do you mean by big ?
[06:33:45] <renesis> i dunno like you can prob fir 3 of my nema 23 motors inside
[06:34:59] <renesis> http://www.anaheimautomation.com/products/stepper/stepper-motor-item.php?sID=16&pt=i&tID=75&cID=19
[06:35:05] <renesis> maybe big like nema42
[06:35:28] <renesis> the mill looked maybe 20" by 10" or something
[06:36:02] <renesis> on a stand, mill wasnt very big
[06:45:50] <renesis> guys fuck the intel doesnt have parport connector
[06:45:59] <renesis> has a header tho
[06:48:34] <renesis> damn like every other d510 board has a parport dsub
[06:51:30] <renesis> heh @ dongle shopping
[06:53:09] <renesis> http://www.newegg.com/Product/Product.aspx?Item=N82E16815340004&cm_re=parallel_port-_-15-340-004-_-Product
[06:53:14] <renesis> =(
[07:23:52] <cnc_valen> so i'm looking to see if I can find an index pulse on these linear scales
[07:24:20] <cnc_valen> in hal meter I have an axis.2.index-enable and its false
[07:24:31] <cnc_valen> any think its relevant?
[07:25:50] <micges_work> what hw you have?
[07:29:28] <cnc_valen> mesa
[07:29:54] <cnc_valen> hmm index enable seems to only be for when hitting an index will re zero the machine
[07:30:12] <cnc_valen> I want to see the actual index pulse in halscope
[07:30:31] <micges_work> yes
[07:30:59] <micges_work> you must find it on encoder.nn.Z or sth
[07:33:37] <cnc_valen> I dont actually see them
[07:34:13] <cnc_valen> i've got count index-enale position rawcounts reset velocity
[07:38:22] <cnc_valen> I dont see a or b of the encoders
[08:25:36] <MrSunshine_> so when connecting a driver, if i have a peak current of 4.2A and a RMS of say 3.5A, i have a motor that can take 4A can i set that driver to 4.2A then ? :)
[08:26:38] <herron> drivers set the rms current not peak
[08:28:54] <MrSunshine_> well the settings are for peak up to 7.9A
[08:39:46] <WalterN_work> this shop is full of ugh
[08:39:57] <WalterN_work> lol
[08:40:56] <WalterN_work> ppl like to spin up the drill presses for deburing parts... ruining the deburring tools
[08:42:05] <WalterN_work> and use files backwards (like they were stroking with a hacksaw)
[08:42:43] <WalterN_work> things like that :-/
[08:43:15] <MarkusBec_away> MarkusBec_away is now known as MarkusBec
[08:48:01] <renesis> okay
[08:48:10] <renesis> so i could be an ultra guinea pig
[08:48:16] <renesis> and get a foxconn mobo
[08:52:33] <MarkusBec> MarkusBec is now known as MarkusBec_away
[08:53:13] <MrSunshine_> how high voltage do we get out of a parport ?
[08:55:52] <sealive> 5,04V as max with 52mA
[08:56:30] <sealive> a max of 200mA in total at the same time
[08:57:32] <alex_mobile> Sometimes only 3.3
[08:58:59] <MarkusBec_away> MarkusBec_away is now known as MarkusBec
[09:06:54] <MrSunshine_> so with a 270ohm resistor i would get a maximum of 0.019A then
[09:06:56] <MrSunshine_> or 19mA
[09:07:35] <MrSunshine_> but the logic can take max 16mA
[09:07:36] <MrSunshine_> hmm
[09:09:16] <toast__> toast__ is now known as toastydeath
[09:25:54] <sealive> i use 330ohm for switches
[09:28:50] <MrSunshine_> sealive, well the msd980 drivers have 270ohm resistors built in for the optical thingies
[09:31:05] <sealive> that shoudt work
[09:31:19] <sealive> your not the first one using this!
[09:37:15] <MrSunshine_> huh ? :)
[09:38:55] <MarkusBec> is it possible to measure the tool length by a webcam ?
[09:42:38] <herron> MarkusBec, not trivial, resolution of camera, and lens needs to be a telecentric for best results
[09:43:29] <MarkusBec> does anybody test it with emc?
[09:45:06] <herron> not aware of anybody else, I had a play
[09:45:51] <herron> I learnt what the problems were and stopped till I can get or make better stuff
[09:51:18] <MrSunshine_> http://paste.cplusplus.se/paste.php?id=11849
[09:51:20] <MrSunshine_> tops
[09:51:22] <MrSunshine_> wrong channel
[10:42:11] <piasdom> g'mornin all
[10:55:45] <proxxon> what does this mean: 7782: ERROR: joint 0 on limit switch error
[10:57:27] <JT-Dev> you machine has tripped the limit switch for joint 0 which is usually the X axis
[11:00:34] <proxxon> but when i press the switch it is the same error
[11:01:18] <proxxon> i have connected the switch to 5 volt and then to the lpt port is this correct?
[11:01:50] <herron> with a pullup up resistor?
[11:02:17] <herron> or pull down in your case
[11:02:30] <proxxon> no with an led and a 100ohm resistor
[11:03:21] <herron> with a meter measure the two states of the pin
[11:04:43] <herron> sounds a bit like you cannot get a 0 or 1 due to the voltages
[11:04:58] <proxxon> when i press thebutton i have 5 volt on it
[11:06:20] <proxxon> the led ison when nothing ispressedand when the switch is pressed the led is off
[11:06:54] <herron> and when not pressed what voltage at the pin
[11:08:04] <herron> * herron is expecting 5v-led_voltage, which is probably still high for the port
[11:08:40] <proxxon> 5 volt when not pressed on the pin
[11:11:11] <herron> 5v in both cases...!
[11:11:41] <herron> wiring / circuit error
[11:15:47] <proxxon> from the lpt pin to ground i have 5volts when the switch is not pressed. when the switch is pressed i have o volts from the lpt pin to ground so i dont think that it is a wiring or curciut error
[11:17:37] <JT-Dev> If you open up the Hal Configuration window and watch that pin on the watch window then cycle the switch then disconnect the switch you can tell if you have a wiring issue
[11:19:42] <JT-Dev> or a port problem
[11:21:25] <proxxon> ok i can see the both-x switch in the "watch" window but it doesn't change when i pressthe switch
[11:22:02] <JT-Dev> is it on or off in the watch window
[11:22:14] <proxxon> it is yellow
[11:22:27] <proxxon> so i think this should be on
[11:22:28] <JT-Dev> disconnect the wires from the pin
[11:23:29] <proxxon> should it change in realtime?
[11:24:02] <alex_joni> kinda realtime
[11:24:07] <alex_joni> maybe < 1 sec delay
[11:24:13] <proxxon> nope itis still on
[11:24:49] <JT-Dev> you might have a problem with that pin on the parallel port then
[11:25:07] <JT-Dev> what pin are you using?
[11:25:29] <proxxon> pin 10
[11:26:04] <JT-Dev> I ask because I have had some parallel port pin 1 problems in the past
[11:27:51] <JT-Dev> i just did a sanity check and without anything connected to my parallel port here pinxx-in is yellow
[11:28:10] <JT-Dev> pin-nn-in-not is red
[11:31:16] <JT-Dev> so I assume what your doing with the wiring is not changing the states of the pin then
[11:38:28] <proxxon> when nothing isconnected tothe lpt pin the 0,78volts is messured
[11:38:35] <proxxon> on pin10
[11:49:15] <JT-Dev> I used a breakout board on my installation so I can't help with direct wiring of the inputs
[11:50:46] <JT-Dev> http://www.cnc4pc.com/Store/osc/product_info.php?cPath=33&products_id=49
[12:08:36] <proxxon> ok i found the problem
[12:08:50] <proxxon> the problem is that it is not working with pin 10
[12:08:56] <proxxon> i use now pin 11
[12:09:13] <WalterN_work> woo, work is almost over for the week... 45min left
[12:09:51] <WalterN_work> an I still cant believe that drill is still drilling
[12:10:01] <WalterN_work> man
[12:10:54] <WalterN_work> about 400 holes in 1/4" 316 plate now
[12:13:30] <JT-Dev> what kind of drill is it?
[12:15:13] <WalterN_work> cobalt
[12:15:32] <JT-Dev> my favorite kind
[12:15:44] <WalterN_work> yeah
[12:15:48] <WalterN_work> good stuff
[12:16:03] <WalterN_work> but still... did not expect it to last that long
[12:16:31] <WalterN_work> even if it was me doing the sharpining (which it was :P)
[12:22:28] <WalterN_work> anyway, time to start wrapping up
[12:22:37] <WalterN_work> payday, yay
[13:02:34] <H_MrSun> would pin 2 and 3 be the default X axis step/dir?
[13:05:10] <atmega> they are mine
[14:17:43] <atmega> does anyone use the zero reference on an encoder for homing?
[14:18:33] <SWPadnos> probably not, there's no real support for absolute encoders in EMC2
[14:18:54] <SWPadnos> (assuming that's what you meant)
[14:19:05] <SWPadnos> or are you talking about the index mark?
[14:19:23] <atmega> the index/tdc/wtf mark
[14:19:34] <SWPadnos> ok, then yes, there are people using that
[14:19:45] <atmega> what does that do for you?
[14:19:58] <Jymmm> absolute encoders???
[14:20:01] <SWPadnos> it makes homing as accurate as it can be
[14:20:06] <atmega> no, homing to the 0 ref
[14:20:14] <SWPadnos> since the index mark is usually one encoder count long
[14:20:16] <atmega> oh, so home to switch then look for the 0?
[14:20:23] <SWPadnos> there are several options
[14:20:46] <SWPadnos> you can do it without a switch even - jog to a mark on the table, hit home, and the home position is the next index point
[14:20:59] <SWPadnos> move to switch, back off, move back to index point
[14:21:05] <SWPadnos> move to switch, move back to index
[14:21:06] <SWPadnos> ...
[14:21:26] <atmega> gotcha
[14:21:43] <SWPadnos> there's a lot of information in the "homing" section of the integrators manual
[14:21:50] <SWPadnos> complete with diagrams
[14:21:58] <atmega> I was looking for more general info, not emc specific
[14:22:42] <SWPadnos> considering we're talking about machine controls, I don't know how much "general info" there oi :)
[14:22:44] <SWPadnos> is
[14:23:06] <Jymmm> vey
[14:23:16] <SWPadnos> oy
[14:24:00] <atmega> a machine was off by one rev (looking for 0-ref) just curious how common that was as a homing procedure
[14:24:58] <SWPadnos> in the wider CNC world, I have no idea
[14:25:39] <atmega> this isn't really a cnc machine, just an automated system
[14:26:03] <atmega> drives are via an APM in a GE/Fanuc PLC
[14:26:27] <SWPadnos> APM?
[14:26:37] <atmega> axis positioning module
[14:26:44] <SWPadnos> ah
[14:26:57] <atmega> an indexer in a PLC slot
[14:27:17] <SWPadnos> yep
[14:27:32] <SWPadnos> being one revolution off is a danger of an index system
[14:27:59] <SWPadnos> if there's a switch plus index, it's possible that the switch moved a little bit, or wore out enough that the activation point is now past the intended index
[14:28:02] <SWPadnos> (sometimts)
[14:28:04] <SWPadnos> grr
[14:28:05] <SWPadnos> sometimes
[14:28:24] <atmega> they find the edge of a prox switch, then go to the 0-ref, it seems to have been right on the edge of the 0-ref at the same time
[14:28:28] <SWPadnos> or a bent actuator would do it too
[14:28:32] <SWPadnos> oh, a prox. hmmm
[14:28:51] <SWPadnos> yeah, a slight nudge "away" would probably fix it
[14:29:14] <SWPadnos> is there a metal tab that actuates the prox?
[14:29:23] <atmega> yeah
[14:29:23] <SWPadnos> activates
[14:29:27] <SWPadnos> ah
[14:29:34] <SWPadnos> it may be bent away
[14:29:34] <atmega> slotted flag, slotted prox switch mount
[14:29:39] <SWPadnos> hmm
[14:30:10] <atmega> barrel prox switch with lock washers, when you tighten them, they tend to move slightly
[14:30:59] <SWPadnos> so the tab moves past the prox, not toward it?
[14:31:06] <SWPadnos> or alongside it anyway?
[14:31:23] <atmega> yeah, over in this case
[14:32:30] <SWPadnos> was some service recently done on that machine?
[14:32:39] <SWPadnos> or cleaning or something
[14:32:42] <atmega> brand new machine, acceptance testing
[14:32:50] <SWPadnos> ah
[14:33:35] <SWPadnos> if you have a scope, you can see where the index mark is
[14:34:00] <SWPadnos> just probe both the prox and the index mark, and turn the screw by hand (if that's possible)
[14:34:14] <SWPadnos> in both directions
[14:34:24] <atmega> I have a similar machine that just homes off a prox switch, it is repeatable to < 0.001" anyway (compumotor drives/indexer)
[14:34:37] <SWPadnos> wow, that's great
[14:35:47] <Jymmm> SWPadnos: 6 wheel drive... http://www.aircompressorsdirect.com/images/stories/submitted/bluetruckreview_338.jpg
[14:36:02] <SWPadnos> dumbass-wheel drive
[14:36:22] <SWPadnos> it looks like 8, but we just can't see the inside two in the rear
[14:36:28] <Jymmm> ya
[14:36:43] <SWPadnos> "I think we're gonna need a bigger jack"
[14:37:11] <atmega> and another case of PBR
[14:37:24] <SWPadnos> coors!
[14:37:30] <SWPadnos> Miller Lite!
[14:37:39] <atmega> don't be gettin' all fancy on us now
[14:37:45] <SWPadnos> heh
[14:37:57] <Jymmm> Bud Weis Er
[14:38:13] <atmega> naw, there ain't no company comin' over
[14:38:19] <SWPadnos> I don't drink beer, but when I take a sip, it's always something dark, like a good German brew, Guinness, that kind of thing
[14:39:07] <SWPadnos> Jymmm, I prefer to drive these: http://tri.army.mil/LC/cl/CLC/m113/M113Fov.htm
[14:39:22] <SWPadnos> my license has probably expired though :)
[14:40:55] <Jymmm> ant tell if the same truck or not... http://www.pirate4x4.com/forum/attachment.php?s=4b14f67844e0b3e6b19ce8826d2ff1af&attachmentid=193156&stc=1&d=1119286569
[14:41:45] <Jymmm> Now, this is a nice looking truck... http://www.pirate4x4.com/forum/attachment.php?s=4b14f67844e0b3e6b19ce8826d2ff1af&attachmentid=192998&stc=1&d=1119235096
[14:42:03] <SWPadnos> ooooh, shiny
[14:42:16] <Jymmm> What you think.... 3, maybe 4 mpg?
[14:42:37] <atmega> or gpm?
[14:43:20] <Jymmm> i see that spare tank as a foot rest too
[14:44:11] <SWPadnos> that gives it a cruising range of OVER 100 miles!
[14:44:30] <atmega> F650 9/12mpg
[14:44:54] <Jymmm> eeeesh
[14:45:48] <atmega> I get 14/16
[14:45:51] <Jymmm> 2000 F650
[14:46:52] <Jymmm> Ah... http://www.differentialeng.com/CompanyTrucks.htm
[14:48:32] <Jymmm> whoa... 6x6 http://www.differentialeng.com/InstalledVehicles.htm
[14:50:59] <Jymmm> I guess if you mfg monster differentials, you really can't drive around in a toyota pickup =)
[14:52:03] <Jymmm> That's kinda like making ship parts and you use a dremmel as your spindle =)
[14:59:53] <grommit> I am plodding through trying to configure EMC to run my machine. I have steppers connected to a 7i47 (off a 7i43) and have a base config that I am trying to modify. I can turn my coolant relay on/off so I know something is working. But I am having trouble getting other things to work, like my steppers and spindle. I am sure it is my lack of understanding. Taking baby steps... I want to get my X-Axis stepper working. I posted
[15:01:11] <Jymmm> grommit: you got cut off at "...X-Axis stepper working. I posted"
[15:01:52] <grommit> Must have been long winded :-) I posted my config at http://pastebin.com/pbFU9uBT. Could someone look and comment?
[15:12:14] <grommit> am I still connected..?
[15:12:27] <Jymmm> Nope.
[15:12:34] <grommit> oh, good, I thought I was :-)
[15:12:58] <grommit> lots of traffic today, I see.
[15:24:32] <PCW_> PCW_ is now known as PCW
[15:43:12] <skunkworks_> logger_emc: bookmark
[15:43:12] <skunkworks_> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2010-04-09.txt
[16:06:51] <kbarry> to "Ring" soemone else, i do what?
[16:07:27] <frallzor> call
[16:10:24] <frallzor> anyone using solidworks and some cam for it?
[16:40:11] <PCW_> PCW_ is now known as PCW
[16:46:31] <kbarry> funny frallzor
[16:46:33] <kbarry> i mean
[16:46:44] <kbarry> sometimes people say things and it makes a noise for me
[16:46:55] <kbarry> is it because they used my nickname?
[16:47:07] <SWPadnos> yes
[16:47:23] <SWPadnos> chatzilla adds your nickname to its list of "stalk words" by default
[16:47:50] <SWPadnos> if any of those words are used, it will ring the bell and highlight the tab label in red
[16:47:54] <SWPadnos> (by default)
[17:30:33] <proxxon> how can i scale the example EMC AXIS2 that it fits in my limits?
[17:39:16] <alex_joni> proxxon: edit the file, and look for a parameter called scale
[17:48:04] <SWPadnos> damn: http://singularityhub.com/2010/04/05/5-axis-robot-carves-metal-like-butter-video/#more-14509
[17:57:52] <Jymmm> Daaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaamn!
[18:03:32] <Jymmm> Not as cool, 2nd video http://singularityhub.com/2009/08/06/stainless-steel-printing-from-shapeways/
[18:06:58] <SWPadnos> cool. they madea SOMA cube :)
[18:07:17] <Jymmm> SWPadnos: I don't know how they call that a 5 axis mill... The base of it has like 3 axis, then the head has another 4
[18:07:24] <Machinist> Machinist is now known as morfic-
[18:08:54] <SWPadnos> yeah, it may have been more than 5
[18:09:21] <Jymmm> I just noticed that the base has XY plus rotate and tilt, that's 4 alone.
[18:09:58] <archivist> I noted the counting error too
[18:12:01] <SWPadnos> hypermill looks very nice too
[18:12:15] <SWPadnos> but it probably costs about as much as a medium-sized sedan
[18:13:07] <archivist> a collection of large sedans
[18:13:19] <morfic-> morfic: look
[18:13:23] <archivist> I asked...
[18:13:40] <SWPadnos> oh
[18:14:47] <archivist> iirc it was 60k£ to moare for a sensible number of axes
[18:14:55] <SWPadnos> ouch
[18:15:00] <SWPadnos> so a very nice sedan ;)
[18:15:12] <SWPadnos> bbiab
[18:15:19] <archivist> 3 axis was a more normal price
[18:16:52] <archivist> machine building show in a week or two in the UK, £++++ will probably be there
[18:18:47] <Jymmm> Look, it made a Mini Mill! LOL http://www.gildemeister.com/ino/lieferprogramm_09/en/fr_dmu_pfd_210.htm
[18:19:56] <Jymmm> (ok, the caption DOES say 'processing examples')
[18:21:39] <Jymmm> I'm ot sure I'd want to have to walk up four stairs to load/unload a machine THAT big
[18:21:44] <Jymmm> s/ot/not/
[18:24:49] <frallzor> * frallzor hates solidcam
[18:25:27] <Jymmm> Honestly, it looks like it could be a time machine or transporter pod... http://www.gildemeister.com/ino/lieferprogramm_09/en/ul_hsc_10.htm
[18:26:22] <Jymmm> Chrome-cobalt teeth -- bling bling
[18:27:29] <frallzor> http://pici.se/p/dMiyUIcnV/ a first dust shoe to make a proper one later :P
[18:30:22] <Jymmm> No slot for brushes?
[18:30:29] <Jymmm> err groove
[18:30:46] <frallzor> that small differense you see on top
[18:30:57] <Jymmm> yeah
[18:31:01] <frallzor> no brushes but a stripe of some type of plastic
[18:31:05] <Jymmm> ah
[18:31:12] <frallzor> and "slots" so its like ripped
[18:31:14] <frallzor> you know
[18:31:24] <frallzor> think piano =)
[18:31:32] <Jymmm> is the little hole for mount or vac hose?
[18:31:46] <frallzor> temp vac
[18:31:50] <Jymmm> k
[18:32:34] <frallzor> hmm wonder if I can find brushes locally...
[18:33:56] <Jymmm> * Jymmm watches SWPadnos video for the 7th time...
[18:34:26] <SWPadnos> I wonder if the mention on Slashdot has affected linuxcnc.org bandwidth :)
[18:34:46] <frallzor> seen my video Jymmm? =P
[18:35:05] <bill2or3> people still read slashdot?
[18:35:15] <SWPadnos> only a few hundred thousand of them
[18:35:52] <SWPadnos> or maybe it's a few million
[18:36:44] <bill2or3> * bill2or3 re-checks it.
[18:37:17] <Jymmm> Never heard of it.
[18:39:04] <morfic-> what would be the best diagram to have to figure out what our "digital" Fanuc AC servos really are, i have a ton of 8.5 x 11, which seem to refer back and forth, but i see the pinout of the X and Z drives, spindle drive, nowhere is a +-10V, all i can see is 0V and 5V
[18:39:31] <SWPadnos> X and Z drive pinouts are probably the most telling
[18:39:46] <SWPadnos> unless they have stupid names like "AF_001.2313"
[18:40:14] <morfic-> i wanna make sure i get everything i need scanned, i am tempted to get the whole section, some seem to repeat itself over 2 or so pages, would be best to see what difference they are showing
[18:41:45] <morfic-> SWPadnos, think i can link you to the scans over the weekend, the ones i think tell most to start, and you check if they have everything needed?
[18:42:06] <SWPadnos> well, I'll tell you what I can figure out :)
[18:42:09] <morfic-> not sure if they use sensical descriptions or just stuff useful to Daewoo techs
[18:42:15] <morfic-> that would be great
[18:42:33] <SWPadnos> please put them on a website where they can be downloaded, I have a stupidly small inbox size
[18:42:46] <morfic-> boss wonders why camsoft never even asked about this info, he doesn't like that
[18:42:50] <SWPadnos> I can't read any asian languages, just so you know ;)
[18:42:59] <SWPadnos> maybe they have it already
[18:43:05] <SWPadnos> you never know what it means
[18:43:06] <morfic-> SWPadnos, no, i'd up it somewhere, need to get guy started on scanning now
[18:43:14] <SWPadnos> ok
[18:43:49] <morfic-> SWPadnos: no, never know, but seeing how Fanuc can't even put on the same keyboard twice, why would they use the same drive twice ;)
[18:43:57] <SWPadnos> heh
[18:47:04] <morfic-> designer is still on phone trying to get more exact info, if not i take the stuff i have
[18:47:09] <frallzor> http://pici.se/p/large/HEOqhFNVH/ think this will work? =)
[18:47:10] <morfic-> guy startting to scan now
[18:47:13] <frallzor> as a temp
[18:49:41] <skunkworks_> frallzor: yes
[18:50:04] <frallzor> some rubber and 2 holes for spindle and vac on mdf =)
[19:05:04] <frallzor> baaaah
[19:05:10] <frallzor> rubber to short =/
[20:14:50] <Machinist> ALL Daewoo/Doosan sent us is a settings bitmask
[20:15:14] <frallzor> is that bad?
[20:16:10] <Machinist> if you want a wire diagram/pin out, yes
[20:16:21] <Machinist> Machinist is now known as morfic-
[20:19:09] <morfic-> at least i got the scans out of the manual now, to upload when i get home
[20:38:44] <billykid2> hello
[20:39:36] <micges> hi
[20:39:37] <JT-Hardinge> hi
[20:40:41] <billykid2> I built a dedicated keyboard for EMC2
[20:42:07] <micges> cool
[20:42:15] <micges> any pictures?
[20:42:29] <frallzor> Im thinking about doing that too, to implement with touchy
[20:42:33] <billykid2> the home button brings home the micro axis active
[20:42:56] <frallzor> so I can use usb to replace the lack of room for any extra pci-cards
[20:43:24] <billykid2> axis help says control- home
[20:43:36] <billykid2> for all axis
[20:44:03] <billykid2> home work
[20:44:11] <billykid2> control home no
[20:44:57] <micges> ctrl home works only if you have homing sequence defined in ini file
[20:45:16] <billykid2> ahhhhhhh
[20:45:26] <billykid2> thanks
[20:45:36] <billykid2> where you post photos?
[20:45:52] <micges> http://www.imagebin.ca/
[20:46:04] <billykid2> ok
[20:46:31] <frallzor> wonder how small you can get the electronics from a keyboard...
[20:46:57] <billykid2> I just put pictures of my CNC with DC servo motors
[20:47:49] <billykid2> and servo controller by elm chan
[20:50:54] <billykid2> imagebin not respond ...
[20:51:24] <frallzor> whats the name of that usb-device you can configure as you like with buttons and such?
[20:52:35] <billykid2> is a ps2 circuit
[20:53:51] <frallzor> Im asking a different question :P
[20:54:34] <Dave911> Hi guys.... How do you get a joint axis following error with a stepper axis? Using a 5i20 card.... If I do a MDI G1 X100 F350 or faster it errors out ...
[20:55:54] <billykid2> Emulatore tastiera PS/2 per joystick arcade o Mach3 CNC
[20:56:22] <billykid2> http://cgi.ebay.it/Emulatore-tastiera-PS-2-per-joystick-arcade-o-Mach3-CNC_W0QQitemZ170462829513QQcmdZViewItemQQptZGiochi_da_Bar_Flipper_Jamma?hash=item27b0605bc9
[20:57:57] <micges> Dave911: to much acc I thinl
[20:58:06] <tom3p> morfic, try to get the wiring diagrams ( not user manuals ). thats what i used on the emc mazak conversion to see what was needed for fanuc motors & drives to emc
[20:58:11] <micges> Dave911: pastebin your config
[21:00:28] <Dave911> To much accel setting in the hal config? I thought it might be not enough so I bumped it up.. didn't help and it didn't change either. Seems to be related to the speed.
[21:01:38] <micges> * micges is tired
[21:01:40] <morfic-> tom3p: it's wiring, out of the manual, not an operator manual, i saw no label on it, it has ladder diagrams in there too, like a maintenance documentation, just not the fold open see it in detail large format print
[21:02:18] <tom3p> Dave911, when there IS feedback, the error is when the device doesnt keep up with the command, when there ISNT feedback, the error is the when the soft pulse generator cant keep up with the command ( like hi acc )
[21:03:03] <frallzor> baaaah I cant figure out that the so called keyboard emulator is called im thinking about :(
[21:03:15] <frallzor> its like a usb-unit with lots of inputs
[21:03:54] <morfic-> frallzor: i have no idea, but now i wanna know too, sounds interesting
[21:03:57] <tom3p> morfic, did you you find labeled wires to the drive with signal names? thats what the integrator will need, pins and names
[21:04:19] <morfic-> named pins, that i also found on the diagrams
[21:04:20] <frallzor> morfic I knew its name when I was into flightsims and planning to make a nice panel =)
[21:04:24] <tom3p> morfic, jon elson and stuart (in nebraska) have done fanuc to emc
[21:04:41] <tom3p> morfic, great
[21:04:43] <JT-Hardinge> Dave911: http://www.linuxcnc.org/docview/html//common_Stepper_Diagnostics.html#r1_2_1
[21:04:46] <morfic-> i just don't know if they are universal or fanuc specific, more 2 letter abreviations
[21:04:52] <andypugh> frallzor: Were you thinking of this?: http://generichid.sourceforge.net/
[21:05:22] <morfic-> frallzor: yeah, good use there too
[21:05:27] <frallzor> andypugh im thinking of making one of those
[21:05:28] <morfic-> tom3p: oh good info, thanks
[21:05:30] <billykid2> bye
[21:05:35] <frallzor> but with the unit I cant find
[21:05:36] <billykid2> thanks
[21:05:47] <frallzor> and just bind the buttons like any other keyboard key
[21:07:59] <morfic-> frallzor: blue button == control - home ?
[21:08:12] <andypugh> frallzor: http://www.delcomproducts.com/productdetails.asp?productnum=902770 then?
[21:08:31] <Dave911> OK.. let me try lowering accel below the standard hal setting ..... BBIAM
[21:08:36] <frallzor> the thing Im thinking about is a readymade unit
[21:08:42] <frallzor> just plug in and then connect buttons
[21:08:49] <tom3p> frallzor, good url, i imagine a cnc pendant shaped like a fender stratocaster, and the whammy bar jogs the axis :)
[21:09:13] <frallzor> I havent linked anything? :P
[21:09:41] <tom3p> oops andypugh
[21:10:37] <frallzor> bah I give up
[21:11:40] <PCW> Dave911: ferrors with stepgen can be because the hardware timing (1/(steplen+stepspace)) wont allow the velocity (not enough headroom)
[21:14:59] <Dave911> Hi PCW... I have the step timing set so I should be good for 50 khz I believe...
[21:15:31] <tom3p> JT-Hardinge, in that link about stepper ferrors, how do these agree? "usually an indication that BASE_PERIOD...is set too low" and "your realtime threads are set too fast you will get this error"
[21:15:38] <tom3p> they sound opposed to me
[21:15:46] <Dave911> I just ran another test with an MDI G1 X1000 F300 and I get up to speed and about a second later it trips out
[21:16:04] <SWPadnos> tom3p, low base period = higher base frequency
[21:16:13] <morfic-> time to go home, and put up the scan pdf
[21:16:21] <tom3p> uh does that make both statement true?
[21:16:30] <SWPadnos> it makes them consistent at least :)
[21:17:04] <frallzor> http://martzis.wippiesblog.com/martzis-usb-hid-interface-card-for-emc-and-mach/ this looks terrific
[21:17:21] <tom3p> so one isnt talking about the value entered
[21:17:46] <Dave911> I've got a 10 us step length and step space, so 20 us all together which I think is 50 khz
[21:18:27] <tom3p> need more room for jitter?
[21:18:37] <SWPadnos> tom3p, I don't know. I'm too impatient to read the whole page ;)
[21:18:46] <JT-Hardinge> tom3p: I'm not sure
[21:18:47] <SWPadnos> Dave911, yes, that sounds like 50 kHz to me
[21:18:55] <SWPadnos> this is a Mesa stepgen?
[21:19:08] <Dave911> I'm using a 5i20 board with a 7i47 card (RS422 high speed I/O)
[21:19:08] <JT-Hardinge> base period set too long not low
[21:19:56] <tom3p> JT-Hardinge, ok, thx
[21:20:10] <andypugh> Hang on, this is ringing a bell..
[21:20:25] <JT-Hardinge> I hear bells too
[21:20:39] <Dave911> I could probably shorten the step length and space length since I'd driving a servo in step mode....
[21:20:40] <frallzor> I hear voices telling me to burn things
[21:20:40] <SWPadnos> Dave911, what is the accel in [AXIS_xx] sections, and what is the max accel for the stepgen (and vel)?
[21:21:35] <Dave911> Currently I am at 10 and 10 for both max vel and max accel, but I have tried higher values ..
[21:21:49] <andypugh> I think the bell it is ringing is a comment in the hostmot stepgen source code (eek!)
[21:22:14] <SWPadnos> Dave911, I guess I'm trying to find out if the stepgen has higher limits than the TP
[21:22:32] <Dave911> What is goofy is that it trips out AFTER I hit the programmed speed, not during accel...
[21:22:39] <SWPadnos> if you don't have that, the stepgen can never quite catch up to where the TP thinks it should be
[21:22:56] <SWPadnos> I guess that can happen if the speed is slightly wrong
[21:23:01] <SWPadnos> due to roundoff or something
[21:24:52] <Dave911> How fast can the TP go? I thought I asked that a long time ago and now I have no idea what was answered ..... I thought it was a lot faster as this is just one move.. not a bunch of curves etc...
[21:24:53] <Dave911> I'd like to be able to go at 25 inches per second in a linear move (1500 ipm)....
[21:25:04] <SWPadnos> it's a setting
[21:25:21] <Dave911> Oh...... a setting .. ok.
[21:25:33] <andypugh> Dave911: Steps per rev?
[21:26:09] <SWPadnos> but if the TP is limited to 100 IPS^2, and the stepgen is limited to 100 IPS^2, then the stepgen will never be able to catch up to the TP, because by definition the stepgen doesn't move until the TP has already decided the motor needs to be somewhere else
[21:26:20] <Dave911> I have it set for 10000 right now I believe.
[21:26:45] <SWPadnos> if you don't have separate MAXACCEL and something like STEPGEN_MAXACCEL settings, then they're probably set to the same thing, which can cause an err
[21:26:47] <SWPadnos> or
[21:28:48] <Dave911> I just have Maxaccel at 10, I don't have a stepgen_maxaccel... but I can add one ;-)
[21:29:52] <andypugh> I don't know where I read this, and I am possibly misremembering, but in the unlikeley event that I do recall correctly something odd happens with Mesa cards if steplen + stepspace > thread jitter (or was it <?)
[21:31:12] <SWPadnos> hmmm. shouldn't be an issue
[21:31:35] <SWPadnos> the mesa stepgens aren't synched to thread execution, they just get new commands every servo cycle
[21:31:51] <SWPadnos> so what's the difference if that happens every 1 ms or 1.01 ms?
[21:31:59] <SWPadnos> (or 0.99 / 1.01)
[21:33:59] <Dave911> Here is the 5i20 .ini file http://pastebin.com/j27r7BEi
[21:34:39] <SWPadnos> can you post the hal file(s)?
[21:34:56] <Dave911> sure one moment
[21:35:16] <SWPadnos> you're testing on X?
[21:37:23] <Dave911> I'm on the Master as of last night.. 2.5 pre
[21:38:26] <micges> I had issue that if you don't set stepgen_maxaccel on hm2 strange things going on
[21:38:27] <SWPadnos> err, I meant the X axis, or AXIS_0 :)
[21:39:19] <Dave911> http://pastebin.com/Md0rSZz3 Here is the Hal file... sorry just working the X axis or Axis 0 for now.. :-)
[21:40:27] <tom3p> no MAX_VELOCITY in [TRAJ] ? no MAX_LINEAR_VELOCITY ? thought that was needed
[21:40:27] <micges> Dave911: for one test: set max acc of stepgen 20% higher that axis max accel
[21:41:00] <Dave911> I think this is the box stock hal file in the configs of the master... I don't think I touched it...
[21:41:32] <SWPadnos> huh. the accel/vel are set to 0, which should mean no limit
[21:41:49] <andypugh> That sounds wrong
[21:41:53] <Dave911> Huh?
[21:42:47] <SWPadnos> oh well duh
[21:43:09] <Dave911> That does look wrong, I wonder if it really is?
[21:43:11] <SWPadnos> 50 kHz / 10000 steps/inch = 5 IPS max for the stepgen
[21:43:17] <andypugh> If this is a default HAL file, what's with the (deprecated) Newsig statements?
[21:43:29] <SWPadnos> but the limit is set to 10 IPS, so the stepgen just can't go that fast
[21:43:52] <SWPadnos> set MAX_VELOCITY to 4.9 instead of 10
[21:44:31] <Dave911> OK.. lets see..
[21:44:34] <SWPadnos> and leave the hal file alone
[21:46:29] <Dave911> OK, now it is running to X1000 at 294 ipm... no trip out..
[21:46:47] <SWPadnos> exactly as expected
[21:47:01] <SWPadnos> is the scale actually 10000 steps/inch?
[21:47:43] <Dave911> So do I need to crank up my max khz to 100Khz plus ... so I can go back to max speed of 10? No I'm just trying to get 1350 rpm out of the motor which is what I need
[21:48:24] <SWPadnos> are you going to be using this for positioning or speed control?
[21:49:22] <Dave911> I am running at 294 rpm right now on the motor... Positioning but in a contorted kind of way... Peter suggested using a stepgen in velocity mode with a PID loop around that. I have a tracking encoder to correct the position..
[21:49:52] <SWPadnos> ok, I was going to suggest vel mode as well
[21:50:17] <Dave911> So I feed the X command to the PID, which drives the Stepgen in velocity mode and the encoder closes the loop .. good suggestion. I think ;-)
[21:50:18] <SWPadnos> in any case, you may want to set the scale to 1 revolution (unless you're actually moving a thing linearly)
[21:50:25] <SWPadnos> I think so
[21:50:35] <SWPadnos> that's how the Pico Systems USC has been doing it for ages
[21:50:56] <andypugh> I guess that will recover lost steps, though my experience is that you don't just drop the odd step, the motor tends to just stall. Then cranking up the speed won't help at all.
[21:51:46] <Dave911> OK, so set scale to 1 for the velocity loop Stepgen.. that makes sense...
[21:51:47] <SWPadnos> with the USC, you can either read back the position registers or an encoder input
[21:52:05] <SWPadnos> Dave911, well, it depends on how you want to program the speed/position
[21:52:34] <SWPadnos> if you're controlling RPM, then "steps per revolution" makes sense, so that the input to the stepgen is in RPS
[21:52:35] <Dave911> I want to program the positon and speed conventionally IPM and inches
[21:52:39] <SWPadnos> ok
[21:52:58] <SWPadnos> in that case you should use a scale that represents 1 inch
[21:53:24] <Dave911> Even on the velocity stepgen?
[21:53:29] <SWPadnos> yes
[21:53:36] <Dave911> OK....
[21:53:40] <SWPadnos> since your unit is an inch, not a revolution
[21:54:39] <SWPadnos> and I'd set the stepgen parameters to whatever your driver actually needs, maybe plus a little safety factor
[21:55:11] <SWPadnos> so for a gecko, that's ~2 uS and 0.5 uS (or close)
[21:55:39] <SWPadnos> hmmm, no, it's a total of 4 uS, but I forgot the split
[21:55:57] <Dave911> Ok. I'll look at the specs on this drive..
[21:56:09] <Dave911> Is this at all important then ?? STEPGEN_MAXACCEL
[21:56:24] <SWPadnos> no, I think those are OK as they are, at 0
[21:56:42] <SWPadnos> you'll use the AXIS_# settings to control accel and vel
[21:56:53] <Dave911> Oh.. thats right that is in the hal file... I was thinking I needed it in the .ini file
[21:57:14] <SWPadnos> it would be good to have it in the ini file, if you wanted to change it
[21:57:23] <SWPadnos> just makes it so there's one place to make changes
[21:57:32] <Dave911> So I was just bumping the limit a bit and that is why it was tripping out..
[21:57:49] <SWPadnos> yes, you went past the physical limit, as PCW suggested earlier
[21:58:00] <Dave911> So you can move those settings between the hal file and the ini file - no problem?
[21:58:01] <SWPadnos> the physical limit being determined by steplen and stepspace
[21:58:25] <SWPadnos> yes, if the hal file has something like blahblah=[AXIS_0]STEPGEN_MAXACCEL
[21:58:54] <SWPadnos> the [SECTION]NAME constuct looks in the ini file for section SECTION, key NAME to get the value
[21:58:57] <SWPadnos> construct
[21:59:33] <Dave911> Yep.. Peter was right.. I had a wrong number stuck in my head ...
[22:00:22] <SWPadnos> you had the right number, 50 kHz, but maybe hadn't related to how the 10000 SCALE affected things :)
[22:00:43] <andypugh> I like to put everything in the HAL file, but I can see that probably isn't sensible.
[22:00:44] <SWPadnos> hmmm. I need to go get cream if I want to have good coffee tomorrow morning
[22:00:45] <Dave911> OK.. that is the .ini processor..... I remember that..
[22:01:30] <Dave911> Better go.. THANKS for the help!
[22:01:48] <SWPadnos> sure. see you
[22:02:05] <andypugh> These servos sat on the floor next to me are whispering "Buy us a Bridgeport" to me.
[22:02:28] <Dave911> Andy.... you know that could be in your head .. ;-)
[22:02:58] <archivist> andypugh, SWPadnos has a bridgeport waiting for luv and EMC2
[22:03:03] <atmega> do the voices in peoples heads ever say good things?
[22:03:38] <andypugh> Would 500W servos work in a Bridgeport?
[22:04:06] <Dave911> There have been several $1000-$2000 running CNC Bridgeports for sale around me recently.... some even come with tooling for that price. I think those would be plenty big.
[22:04:12] <andypugh> I guess with 6000rpm to play with you can afford to gear down a fair way.
[22:04:12] <andypugh> ]
[22:05:11] <andypugh> The thing is, I have put so much time into the Mini-Mill now.
[22:05:12] <Dave911> Thanks for the help guys .... going to do more testing on the X axis .. :-)
[22:05:50] <atmega> I'll give you $68.00 for the mini-mill
[22:05:56] <Dave911> The bridgeports are a little hard to shuffle around also... better have a good floor
[22:06:18] <Dave911> gee... where did you come up with $68?
[22:06:22] <andypugh> That is another issue, I really don't have the room.
[22:06:50] <andypugh> I need to start making the clock and see how it goes, any machine will struggle when it is making parts for itself.
[22:07:03] <archivist> amend the walls till you do have room
[22:07:11] <Dave911> BBL...
[22:08:29] <andypugh> I am thinking of building a garage-like structure in the garden, to separate bikes and swarf. A simpler plan is a dividing wall in the garage.
[22:11:30] <archivist> Im off to fetch a window for my garage tomorrow
[22:15:32] <ries_> ries_ is now known as ries
[22:26:02] <izua> izua is now known as JamesDalton
[22:26:11] <JamesDalton> JamesDalton is now known as izua
[22:36:30] <skunkworks> someone want to explain how this work.... (it isn't a resolver...) looks like you excite each coil in series - then the center tap gets sum'ed together to get the actual postion shift...
[22:36:32] <skunkworks> http://electronicsam.com/images/KandT/conversion/accupins.jpg
[22:36:59] <skunkworks> center taps (plural)
[22:39:45] <andypugh> Is it even slightly like an LVDT?
[22:39:47] <andypugh> http://en.wikipedia.org/wiki/Linear_variable_differential_transformer
[22:40:18] <archivist> * archivist thinks so too
[22:43:32] <skunkworks> hmm
[22:43:47] <skunkworks> except this has 4 coils.
[22:44:17] <skunkworks> and reads off of these.. http://electronicsam.com/images/KandT/conversion/accupins.JPG\
[22:44:30] <skunkworks> http://electronicsam.com/images/KandT/conversion/accupins.JPG
[22:44:48] <andypugh> I am wondering if it is effectively a multi-core LVDT.
[22:44:50] <skunkworks> good thing my site is case sensitive.. ;)
[22:45:12] <skunkworks> I would love to get these working...
[22:45:39] <archivist> skunkworks, thats close to a system that uses balls in a tube
[22:46:23] <skunkworks> I am wondering if I sent the same excitation signal to 2 of the coils - I could get the sin/cos out of the other 2 for a resover>quadture converter
[22:46:47] <andypugh> http://www.newall.com/LEDs/leds.htm
[22:47:39] <archivist> yup thats the system
[22:47:40] <andypugh> Sounds worth a try
[22:48:04] <andypugh> Do you have $20 to burn on an AD2S1200 chip?
[22:48:23] <skunkworks> That is a thought (i have looked at those before)
[22:49:22] <andypugh> Tell you what, if you get one and it doesn't work I will buy it off you for my own experiments.
[22:49:30] <skunkworks> :)
[22:50:14] <andypugh> Though I am pondering an Arduino-based resolver decoder too.
[23:06:59] <morfic> SWPadnos: at your convenience: http://www.thirtyfivemillimeter.net/12-LC_CIRCUITS.pdf page 36 of the pdf has what i can see inside the machine, RX, SX, TX, power to servo has labels: A1X, A2X, A3X, aside from those shown, there is E (Yellow), and G (Green) (I couldnt decide if it's E Earth or G Ground, just because Ground is Erde in german, which literally is Earth :>)
[23:08:15] <andypugh> We use E for Earth in the UK too. Mains supplies are referred to as LNE for example.
[23:09:29] <andypugh> Those diagrams use E and G interchangeably for added fun :-)
[23:09:30] <morfic> line, neutral, earth
[23:09:38] <morfic> andypugh: i have both
[23:09:42] <andypugh> Live Neutral Earth.
[23:10:07] <morfic> page W18 is preceeded by W06 and W07, crap, i did not think to QC the QC guy's work!
[23:10:16] <morfic> i hope W16 is there somewhere
[23:10:32] <andypugh> Page 5 shows G and E connected together, and both with the Earth/Ground symbol.
[23:10:48] <kbarry> Anyone here running a Mechmate?
[23:11:19] <andypugh> Frallzor?
[23:11:56] <frallzor> frallzor is
[23:12:18] <frallzor> is he around?
[23:13:29] <archivist> only frallzor knows
[23:13:42] <kbarry> When i cut Plywood, i use a 2 flute 1/8 Upspiral Fishtail Solid Carbide bit.
[23:14:11] <kbarry> The plywood surface tends to "Peel" infrnt of the cutter, instead of being cut clean,
[23:14:29] <kbarry> Is there a solution to this problem? (Downspiral bits?)
[23:14:39] <andypugh> You can get hybrid up/down cutters
[23:14:46] <kbarry> You mea
[23:14:48] <kbarry> Compression bits>?
[23:14:54] <andypugh> Maybe.
[23:14:57] <kbarry> But that requires one pass cutting
[23:15:00] <kbarry> right?
[23:15:10] <andypugh> I saw a picture once, that's as far as my experience goes.
[23:15:40] <morfic> ah W16 is on page 80, phew, he just got it out of order, and W71 is on 81
[23:15:42] <kbarry> when cutting plywood, what tool diameter to pass depth do you follow?
[23:16:02] <frallzor> me?
[23:16:19] <kbarry> (as in, i typically use 1:1, so .125 [ass for a .125 bit.
[23:16:22] <frallzor> Havent done it yet
[23:16:42] <kbarry> Anyone?
[23:16:48] <frallzor> in MDF I have used both 3 and 10mm at 8mm depth
[23:17:15] <kbarry> so less than 1 to 1
[23:17:25] <andypugh> http://www.newwoodworker.com/updowncutbits.html
[23:17:37] <frallzor> less than and more than
[23:17:38] <morfic> kbarry: 3diam. is deeper at 8 deep
[23:17:48] <kbarry> ohh
[23:17:50] <kbarry> hahahhaa
[23:17:51] <kbarry> sorry
[23:18:00] <frallzor> no issues with any as of yet
[23:18:05] <frallzor> but mdf is mdf
[23:18:08] <kbarry> right
[23:18:23] <andypugh> Perhaps you need a toolchanger, start with down-cut then do the rest with upcut?
[23:18:26] <frallzor> the spindle doesnt even chuckle at that
[23:18:30] <kbarry> Well, i might try some tests on plywood at 1 to 6
[23:18:33] <morfic> can't say i cut much wood, but the router ripped through 1/8 aluminum with 1/8 2 flute endmills (we never could find router bits to make us happy)
[23:18:56] <kbarry> Hmmmm.
[23:19:10] <frallzor> I use some nice for wood
[23:19:11] <frallzor> CMT
[23:19:13] <frallzor> good shit
[23:19:20] <kbarry> CMT?
[23:19:30] <frallzor> brand of nice bits
[23:20:10] <frallzor> a bit pricey, but by far the best for wood I've tried, but there are alot of brands in the world =)
[23:20:56] <andypugh> http://www.routertooling.co.uk/images/RTL%20Super%20C%20Compression%20March%202010%20-%20min.pdf
[23:21:17] <andypugh> Though not the oddest cutter I have seen.
[23:22:55] <frallzor> odd mix of metric and imperial
[23:23:11] <kbarry> seen
[23:23:17] <kbarry> precisebits.com
[23:23:33] <kbarry> (thats who i use)
[23:23:51] <kbarry> they carry bits on the smaller end of the spectrum.
[23:24:05] <kbarry> .01 diamter cutters
[23:24:23] <kbarry> 1/16 round (ball-nose) bits
[23:25:03] <andypugh> Crikey, I didn't think that cutters go much smaller than 0.3mm
[23:25:14] <kbarry> check them out,
[23:25:38] <kbarry> they carry a 1/14 shank that tapers to 1/16" ball nose....
[23:25:46] <kbarry> I havn't worn mine out yet,
[23:26:05] <kbarry> I have about 400 hours of tooling on that once cutter.
[23:26:38] <andypugh> i was playing games with units. 0.3mm is 0.01" Sorry.
[23:27:29] <kbarry> http://precisebits.com/products/carbidebits/precisebit-soft.asp
[23:27:40] <kbarry> .125mm diameter
[23:28:15] <kbarry> .15 ballnose
[23:28:19] <kbarry> .15mm
[23:30:52] <kbarry> guess i've never really pushed it with my bits
[23:30:59] <kbarry> but sure would save allot of time
[23:31:05] <kbarry> for me to take more per pass
[23:31:20] <morfic> can't say i tinker in little stuff, been a while since i handled a .03125" end mill, don't go much bellow 1/8" now, for surfacing
[23:31:29] <morfic> 400 hours on one bit does sound wasteful though
[23:31:38] <kbarry> wasteful?
[23:31:43] <kbarry> I don't understand.
[23:32:37] <morfic> i agree with you, about saving time if pushing the bit, machining to make tools last never is efficient
[23:32:49] <kbarry> ohh,
[23:32:58] <kbarry> i hve never been nice to that taperd bit
[23:33:02] <kbarry> that tapered bit
[23:33:08] <kbarry> i run it 1" depth
[23:33:16] <andypugh> I suspect he is saying that you could have worn out the tool in 50 hours and made 10x the stuff.
[23:33:22] <kbarry> yea
[23:33:33] <kbarry> well,
[23:33:37] <kbarry> my lack of business
[23:33:43] <kbarry> means i have the luxury of dpiognm things slowly.
[23:41:20] <morfic> i just need to hope now someone can figure out anything about our drives, Boss made it clear today, he favors EMC, and he got things moving when he heard i still had zero info on the drives
[23:41:54] <andypugh> The existing drives, or putative new ones?
[23:43:48] <morfic> ideally existing, hence the pdf, but i am tempted to price new ones in comparison
[23:44:08] <morfic> they are 1996 models, why spend to make them work, when it's easier to make new ones work
[23:44:26] <morfic> unless the difference is a $100 daughter board to switch
[23:45:56] <frallzor> bosses tend to think recycling is the best way no matter what
[23:46:08] <frallzor> "use the old ones" blablsa
[23:47:02] <andypugh> I am currently intrigued by the possibility of not using drives at all.
[23:47:49] <mozmck> skunkworks: you here?
[23:47:50] <andypugh> You end up paying a lot of money for hardware that basically gets in the way.
[23:49:01] <andypugh> Whereas in theory EMC2 could drive something like this (30A / 600v) pretty much directly.
[23:49:02] <andypugh> http://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=6710104
[23:57:17] <tom3p> PCW can you look at http://www.thirtyfivemillimeter.net/12-LC_CIRCUITS.pdf page 71 of 82? what would 6 pairs of PWM signals be used for?