#emc | Logs for 2007-05-10

[00:19:40] <JymmmEMC> Class action lawsuit... "Engineer developes work place diabities for prototypo creation"
[01:11:40] <skunkworks> tomorrow. that is when I will hook up the switch. yah
[01:14:11] <jepler> skunkworks: what switch is that?
[01:16:20] <skunkworks> probe switch. for tool setting
[01:17:08] <jepler> ah
[01:17:25] <jepler> I'm milling a board on chris's machine so I'm about to see it in action .. I think the first toolchange is coming up
[01:17:46] <skunkworks> Nice.. You need a live web cam - cnc porn.
[01:18:09] <skunkworks> what is the board for?
[01:18:12] <skunkworks> if I may ask
[01:18:48] <jepler> It's the second try of an AVR microcontroller board with USB connector (I bridged a trace on the first one and couldn't clean it up)
[01:19:59] <jepler> the project is to develop a USB HID firmware that supports several digital inputs, several analog inputs, and a jog wheel
[01:20:17] <skunkworks> Very nice.
[01:22:17] <jepler> talk to me again after I've written the firmware :-P
[01:24:09] <jepler> the main thing I'm concerned about is whether performing the USB communication (which is done entirely in software, without interrupts permitted) will take too long and not leave enough time to reliably count quadrature, even at jogwheel speeds
[01:26:25] <toastydeath> fff
[01:29:29] <skunkworks> Is this another fest project?
[01:36:14] <jepler> skunkworks: depends whether I finish it first
[01:36:18] <jepler> or lose heart
[01:38:48] <skunkworks> I never have that problem
[01:38:57] <skunkworks> ;)
[01:39:49] <skunkworks> I find that a nice new razor blade takes care of bridges nicely. just don't cut yourself
[01:41:25] <skunkworks> the very first version of the h-bridge I had inverted a few pins.. So I ended up carving the traces out and running new jumpers. wasn't pretty but worked.
[01:53:02] <jepler> board done (long ago) -- the switch worked like a charm.
[01:53:35] <skunkworks> nice
[01:56:11] <toastydeath> it would be really darn cool if someone made a full-featured CNC controller that also had an amazing manual interface
[01:57:10] <toastydeath> dual turret, sub-spindle, C axis lathe for instance
[01:57:31] <toastydeath> use one turret as a follower rest, or set up balanced turning
[01:58:01] <toastydeath> milling/etc
[01:58:19] <a-l-p-h-a> anyone use win32?
[01:58:25] <a-l-p-h-a> what do you use to mount an iso image as a driver?
[01:58:38] <toastydeath> ...as a driver?
[01:58:54] <petev_> daemontools
[01:59:07] <a-l-p-h-a> thanks
[01:59:10] <a-l-p-h-a> forgot what it was called.
[01:59:55] <jepler> hi I'm jepler and I should have logged out my account!
[01:59:59] <jepler> haha
[02:00:19] <cradek> err, hi
[02:00:31] <a-l-p-h-a> so it's a pretend Jeff?
[02:02:01] <cradek> well he was using my computer but left
[02:02:49] <skunkworks> skunkworks is now known as credak
[02:02:53] <credak> Hi
[02:03:05] <credak> credak is now known as skunkworks
[02:03:06] <cradek> hmm, people do call me that sometimes
[02:15:05] <Jymmm> Someone's on crack... http://www.alltronics.com/cgi-bin/category.cgi?item=93I002
[02:16:54] <cradek> those are totally unnecessary if you have emc2
[02:17:04] <cradek> you can generate any of those step patterns directly
[02:17:18] <Jymmm> I thought it was a driver
[02:17:58] <cradek> oh you're right
[02:18:17] <cradek> forget what I said then
[02:18:32] <Jymmm> forget what?
[02:18:37] <Jymmm> ;)
[02:18:55] <cradek> no chopping then, it's not great for much
[02:19:07] <Jymmm> not at $17 it aint
[02:19:10] <cradek> might be nice for tiny or slow motors
[02:19:32] <Jymmm> if it was bipolar I might consider it
[02:19:55] <Jymmm> then I could always give it prozac.
[02:20:09] <cradek> what are you building?
[02:20:53] <Jymmm> A PS, just came across that is all. I thought about have a lil board that'll drive a stepper though, nothing fancy, just like a speed control.
[02:21:00] <Jymmm> no PC
[02:21:40] <Jymmm> I see these tiny pancake steppers all the time, just thought I'd put them to use.
[02:30:59] <Jymmm> cradek: Maybe like a winder or some such thing. just a lil somethign to tinker around with.
[02:35:22] <skunkworks> you could make a miniture hexapod
[02:38:30] <Jymmm> Yeah, or a portable engraver or some such thng.
[02:39:01] <Jymmm> Just a lil something i can play aorund with, not so much for "production" type stuff.
[02:41:23] <Jymmm> Hey, how repeatable are hall effect sensors?
[02:42:55] <Jymmm> http://www.alltronics.com/cgi-bin/item/04T029/47/Ball%2DBearing%2DMousetraps
[02:48:01] <Jymmm> Ok, I found the PERFECT stepper motor.... http://www.alltronics.com/mas_assets/full/24M021.jpg
[02:51:32] <skunkworks> I have a bunch of floppie drives with those in or similar
[02:52:58] <Jymmm> heh
[02:53:01] <skunkworks> http://www.electronicsam.com/images/desk1.jpg
[02:53:31] <skunkworks> if you look to the right - you will see a nice stack of floppy drives
[02:54:15] <skunkworks> you never really know when they will come in handy - do you
[02:56:05] <skunkworks> DO YOU
[02:56:09] <Jymmm> weird... http://www.alltronics.com/cgi-bin/item/24L008/47/Honeywell%2DHPX%2DA1%2DH%2Dproximity%2Dsensor
[02:58:04] <Jymmm> I DONT!
[03:04:22] <Sapote_reloaded> Sapote_reloaded is now known as Sapote
[03:29:08] <toastydeath> ffff
[03:32:00] <Jymmm> 65535
[03:33:19] <toastydeath> lol integers
[03:34:11] <Jymmm> 1111111111111111
[04:06:18] <Jymmm> Has anyone seen a blade fuse block for mounting in an enclosure? Inline I've seen.
[06:36:28] <a-l-p-h-a> a-l-p-h-a is now known as a-l-p-h-a2
[08:09:03] <kwajstabo> hello
[08:10:36] <kwajstabo> can ylo please tell me, how can i set/change spindle frequency. What output can i get on LPT pins?
[08:24:26] <anonimasu> you can drive a VFD with a voltage signal
[10:47:08] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[11:27:49] <martin_lundstrom> Hello everyone
[11:32:20] <Dallur> hey Martin, I started the work last night but I fell asleep in front of my computer :P
[11:32:56] <martin_lundstrom> lol
[11:33:16] <Dallur> Martin_Lundstrom: at least I got a good nights sleep
[11:33:42] <martin_lundstrom> I imagining a nice keyboard pattern in your face ;)
[11:34:10] <Dallur> Martin_Lundsrom: yeah, got a pretty windows icon imprinted into my neck :P
[11:34:56] <martin_lundstrom> I will try to be patient
[11:45:07] <alex_joni> haha
[12:07:49] <martin_lundstrom> Hello alex_joni, how are you doing?
[12:11:07] <alex_joni> busy
[12:28:34] <jepler> skunkworks: did you ever use pwm.0 on the pluto?
[12:28:39] <jepler> skunkworks: besides looking at the LED, anyway
[12:28:45] <skunkworks> Not yet.
[12:29:04] <skunkworks> Issues?
[12:29:08] <jepler> maybe
[12:29:17] <jepler> my user says I should wait for a full bug report before panicking
[12:29:28] <cradek> ha
[12:29:29] <skunkworks> cradek?
[12:29:32] <skunkworks> ;)
[12:29:57] <skunkworks> almost got the lathe back togather?
[12:41:19] <jepler> skunkworks: when you were finding the bugs in the encoder, how were you turning the encoder -- some other way besides with pluto PWM?
[12:41:45] <skunkworks> drill press
[12:42:58] <skunkworks> well - it was very bad initially - you could turn it by your fingers and see it wasn't counting correctly
[12:43:08] <jepler> indeed
[12:43:27] <skunkworks> I had just hooked it into the drill press after you had fixed it to see how fast i could spin it. Think I was still working after 6krpm
[12:43:32] <jepler> which is why it was nice to see chris produce this screenshot .. http://timeguy.com/cradek-files/emc/counting.png
[12:44:11] <skunkworks> Nice
[12:44:18] <skunkworks> is the lag bad?
[12:45:03] <jepler> 07:23:32 <cradek> siggen -> pwm.1 -> motor -> encoder
[12:45:10] <jepler> so this isn't a position loop at all
[12:46:23] <skunkworks> ah - ok
[12:47:35] <skunkworks> was chris up late?
[12:47:52] <skunkworks> logger_emc: bookmark
[12:47:52] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-05-10.txt
[12:48:14] <jepler> this wasn't on the channel
[12:48:20] <jepler> so no use looking for it
[12:48:40] <skunkworks> aw
[12:48:45] <skunkworks> crap :)
[13:08:08] <rayh> We have encoder issues?
[13:22:45] <skunkworks> There was an issue with the comunication with the pluto a while back that caused the pluto to act goofy.
[13:23:05] <skunkworks> encoder counting made no sense
[13:34:46] <rayh> ah okay. Long ago we had a similar issue with the code on Abdul's motenc card
[14:08:04] <cradek> skunkworks: found the problem - the board is wrong (connector upside-down)
[14:10:29] <skunkworks> Cool. so the polarity was wrong?
[14:11:22] <cradek> no the whole ribbon cable jumper is wrong on one end
[14:11:33] <skunkworks> and why arn't you at work?
[14:11:36] <cradek> I trusted the numbering in the eagle library and I should have checked it
[14:11:42] <cradek> I am...
[14:11:49] <skunkworks> oh.. ;)
[14:12:10] <skunkworks> this is a board between the pluto and your hardware?
[14:12:23] <cradek> yes
[14:12:46] <cradek> I think I can make a "special" (read like "special education") cable to go between them
[14:14:33] <skunkworks> nice. I still have not finished the breakout board. Thinking maybe of making the pluto plug into the breakout board.
[14:15:17] <cradek> remind me to give you my pluto eagle library (after I fix it)
[14:15:51] <skunkworks> oooh - that would be cool
[14:17:52] <skunkworks> Interesting that you first thought it was jeplers issue.
[14:18:02] <skunkworks> * skunkworks being an instigator
[14:18:04] <skunkworks> ;)
[14:18:55] <cradek> oh no I didn't - I would never blame someone else if something I'm doing at midnight doesn't work
[14:19:38] <skunkworks> can you flip the plug - or do you have to make a new board
[14:19:50] <skunkworks> or flip the ribbon
[14:20:08] <SWPadnos> make the pins stick out the bottom ...
[14:20:24] <skunkworks> you have 4 combinations of putting the ribbon into the plug...
[14:20:53] <SWPadnos> only two, actually (for a plug that's already mounted)
[14:21:25] <cradek> SWPadnos: we thought of that, but I think making a "special" cable is easier
[14:21:34] <SWPadnos> that's probably true
[14:21:38] <cradek> in fact, it's almost done
[14:23:13] <skunkworks> I guess it is only 2
[14:23:24] <SWPadnos> yep - you can only rotate it 180 degrees
[14:24:19] <skunkworks> I was thinking flipping and rotating but the pairs are the same
[14:25:17] <SWPadnos> right - even making the cable with the connector on "top" vs "bottom" doesn't represent a flip
[14:25:26] <SWPadnos> the cable just exits in a different direction :)
[14:27:06] <skunkworks> right
[14:29:38] <skunkworks> Experience is the BEST Teacher. Is that why it usually arrives in a shower of sparks, flash of light, loud bang, a cloud of smoke, AND -- a BILL to pay? You usually get it -- just after you need it.
[14:30:32] <SWPadnos> heh
[14:31:45] <skunkworks> one of the few things I remember from highschool is a teacher that said something similar. 'School has it all backwards - you learn the lesson then take the test. In real life - you take the test and then learn the lesson"
[14:32:45] <SWPadnos> that's what we need - more testing!
[14:32:49] <SWPadnos> oh wait
[14:54:56] <skunkworks> http://bash.org/?760257
[14:58:20] <skunkworks> omg http://bash.org/?758032
[14:58:22] <skunkworks> :)
[14:58:36] <SWPadnos> heh
[14:58:37] <skunkworks> that concludes the bash quotes for the day
[14:58:51] <SWPadnos> taht's a good show-stopper
[14:58:53] <SWPadnos> that
[14:58:59] <cradek> wow
[15:00:13] <SWPadnos> it's funny, when I took the ASVAB, I got scores high enough that the recruiter didn't know they were possible (he had never seen a score above 120 or so, I had 140+ in all categories)
[15:00:37] <SWPadnos> of course, I thought it would be a percentile thing, so I was surprised to see anything above 99 ;)
[15:00:38] <cradek> SWPadnos: lol @ you
[15:01:13] <SWPadnos> yeah - the joke's on me - I did join the Army national guard, after all ;)
[15:01:35] <cradek> at least your timing was better than the poor saps who did it recently
[15:01:40] <SWPadnos> yeah
[15:01:49] <SWPadnos> I narrowly escaped Gulf War I
[15:02:55] <skunkworks> I was a senior during the first gulf war.. I was a bit worried about it at the time... but then agian - I am a worrier by nature. Got it from my mom ;) Hi dad ;)
[15:03:31] <cradek> I have to say there has not been one day of my life where I didn't know better than to sign up for military service
[15:03:47] <SWPadnos> yeah - all I had to do was help with some repairs to get another unit ready to go. My unit was a maintenance unit, so we weren't slated to go for a couple of months, then the whole thing ended in 4 days :)
[15:04:29] <cradek> either I don't have what it takes, or I have something it doesn't take, not sure which
[15:04:41] <SWPadnos> yeah, the military didn't have much pull for me, but I figured I'd be better off choosing where to go for myself, arther than letting someone else do it
[15:04:52] <SWPadnos> s/arther/rather/
[15:05:11] <SWPadnos> this was the Reagan years, so we had lots of crap to look forward to
[15:17:07] <gene> Question for the one who called my attention to the connection beteen that time warning and usb?
[15:17:24] <SWPadnos> skunkworks did it!
[15:17:26] <gene> Can I plug in hub and get awy with it?
[15:17:26] <SWPadnos> I think
[15:17:30] <SWPadnos> no
[15:18:06] <gene> so the only way I get this scanned file off this box is samba then?
[15:18:35] <SWPadnos> you can plug in the USB key, copy the file to HD, then remove the USB key
[15:18:54] <SWPadnos> it's usually changes in the USB configuration that cause problems
[15:19:00] <SWPadnos> ie, plugging / unplugging devices
[15:19:20] <gene> and that rrequires I pull it off the shelk cause the only usb connector is on the mobo, on the back
[15:19:28] <gene> shelf
[15:19:37] <gene> bummer
[15:19:53] <SWPadnos> then how do you access the USB key now?
[15:19:53] <gene> I'll setup samba and mount it then.
[15:20:15] <skunkworks> you need a usb cable extention ;)
[15:20:36] <SWPadnos> I didn't mean to say that you can't have a USB hub - adding a hub shouldn't change the situation
[15:20:55] <gene> by pulling the machine down from teh shelf, its about 6 feet up, getting it turned far enough to find the porrt and plug it in, takes about 4 or 5 minutes
[15:21:01] <SWPadnos> so if you already have one, then you need to avoid plugging/unplugging things into that
[15:21:32] <SWPadnos> ok, so add a hub - it shouldn't make things worse unless you have a crappy cable (causing plug/unplug events ;) )
[15:21:37] <gene> well, just leaving the key in, unmounted, made all sorts or errors
[15:22:12] <SWPadnos> hmmm. maybe leaving the key unplugged is a better idea
[15:22:13] <gene> the hub idea is to get it where I can reach it
[15:22:18] <SWPadnos> do you have a USB mouse?
[15:22:31] <gene> yes, no problems that I can see there
[15:22:53] <skunkworks> others have said they only have problems with usb memory sticks
[15:23:11] <gene> ok, I'll try the hub then
[15:23:19] <skunkworks> petev for sure iirc
[15:28:01] <gene> Hu8b plugged in, doesn't seem to be bothering it
[15:28:52] <skunkworks> Great :)
[15:30:11] <gene> stick plugged in, got requestor but it fails as only root can mount, grrrr
[15:30:57] <gene> and plugged in stick, in the hub, does not seem to be bothering emc, at all???
[15:31:33] <gene> can I fix mount so I can mount that thing, once and for all time?
[15:32:00] <cradek> on my system plugging in a usb stick automatically mounts it
[15:32:16] <cradek> then you just right click to "eject" (unmount) it
[15:32:27] <gene> it tries to here, but fails cause I'm not logged in as root
[15:32:52] <gene> thats a udev rule, right?
[15:33:12] <cradek> I don't know how it all works, something about hotplug
[15:33:38] <cradek> maybe it doesn't work in kde or something
[15:35:16] <jepler> on ubuntu, the commandline way for a user to mount a filesystem is with the "pmount" command, not "mount" as you may have experienced on other systems.
[15:35:30] <jepler> DESCRIPTION
[15:35:30] <jepler> pmount ("policy mount") is a wrapper around the standard mount program
[15:35:30] <jepler> which permits normal users to mount removable devices without a match-
[15:35:30] <jepler> ing /etc/fstab entry.
[15:38:35] <cradek> oh, interesting
[15:45:49] <gene> well, I've been off-screen, and have it mounted, but even as sudo root I can't chaneg the friggin perms so gene can write to it.
[15:46:04] <gene> but that wasn't using mpmount. I'll try that
[15:48:03] <gene> and thats a no go. I'm about to make me a root account on this box dammit
[15:50:06] <gene> ok, theres an override in fstab, how do you folks hae your fstab setup?
[15:51:17] <jepler> I don't change my fstab; I use pmount when I use removable media
[15:51:22] <gene> what do I change so the gui can do it for me?
[15:51:37] <SWPadnos> gene, is this a new install?
[15:51:48] <gene> thats because udev creates the device file on the fly
[15:52:01] <gene> yes, from the live dvd vesion
[15:52:09] <Dallur> gene: I use automount, plug a device in and it auto mounts, automatically mounted with sync so I can just rip it out without unmounting
[15:52:45] <SWPadnos> I haven't changed anything from the default, and USB sticks automagically mount (and pop up a Nautilus window) when I plug them in
[15:52:59] <SWPadnos> this is with the standard Gnome Ubuntu though, not KUbuntu
[15:53:01] <Dallur> swpadnos: thats the gnome mount agent though, not the kernel
[15:53:09] <SWPadnos> right
[15:53:15] <gene> or kde
[15:53:52] <SWPadnos> it's possible that the KDE agents aren't set up right
[15:53:55] <Dallur> there is an automount kernel agent which can do the same thing for you, automatically unmounts if you have not written to a device and mounts it when you try to write to it
[15:53:56] <gene> according to ps, automount isn't running, but isn't that called on demand from hal?
[15:54:15] <Dallur> so after a timeout of for example 1 sec the device is automatically unmounted
[15:54:18] <gene> whats its name
[15:54:21] <Dallur> if there are no open files on it
[15:54:22] <gene> ?
[15:54:32] <Dallur> automount v.4
[15:54:34] <SWPadnos> automounter?
[15:54:48] <SWPadnos> aut<tab><tab> ;)
[15:54:51] <gene> I'll see if adept can find it.
[15:54:55] <Dallur> also works with nfs/samba shares
[15:58:32] <gene> there is not an automount in the listing, installed or otherwise
[15:59:10] <gene> There is an 'asmounter' which reads like a tool of that ilk though.
[15:59:12] <Dallur> gene: you have to recompile the kernel with automount support and then install the automount userspace application to use i t
[16:00:03] <gene> I think I'll pass, one machine with a bleeding esge kernel is a big enough headache
[16:00:14] <Dallur> autofs v.4 is the name of the kernel stuff
[16:00:20] <Dallur> but the userspace app is called automount
[16:02:00] <gene> thats installing now, and it claims 'debian' kernels are already automounter awre?
[16:02:09] <gene> aware?
[16:03:06] <Dallur> hehe okey, so now you just need to configure :)
[16:03:10] <SWPadnos> there are kernel and userspace parts of the automount system, so the kernel parts have to provide the right services/triggers for the userspace parts to work
[16:03:12] <Dallur> man automount should do the trick
[16:03:22] <SWPadnos> hence "aware" ...
[16:03:24] <gene> ok
[16:08:14] <gene> man automount gives a bunch of requited options, but can these not be put into a file someplace? The man page is sorely lacking in that bit of detail...
[16:08:31] <gene> required
[16:10:03] <gene> would it not be easier just to give gene ownership of it in fstab?
[16:10:25] <Dallur> gene: probably, I just like how cool automount is ;O
[16:11:20] <gene> I insstalled it, but it ddoesn't seem to be doing anything, does its demn need started yet?
[16:11:27] <gene> demn demon
[16:11:49] <SWPadnos> the daemon may not be started by the install script
[16:11:49] <cradek> gene: maybe you should ask on #ubuntu how to fix this, I don't think anyone here knows
[16:12:03] <SWPadnos> (other than installing "normal" gnome/ubuntu ;) )
[16:13:19] <cradek> yeah it does seem to work automatically on "normal" ubuntu
[16:13:32] <gene> I started it, but it says "no automount maps defined", and the manpage doesn't mention them by filename
[16:13:47] <Dallur> gene: look under /etc/automount/
[16:14:25] <gene> if you knew how much I hated gnome and its incessant, you can't tell it to shut the fuck up, nagging...
[16:14:28] <cradek> fwiw my system does not have automount/autofs. I think you're on the wrong track actually.
[16:14:53] <SWPadnos> gene, I guess I'm lucky I don't know :)
[16:17:11] <gene> I may have found it, there isn't an /etc/automount, but there are a quartet of auto.* files in etc, one of which is labeled auto.master, and apparentlt serves as an on-off switch for the other 3.
[16:17:31] <gene> I'll prod it and see if it squeals.
[16:17:38] <gene> :)
[16:18:00] <SWPadnos> did you just run the automounter daemon, or did you do the equivalent of /etc/init.d/automounter start ?
[16:23:06] <gene> sudo init.d/auutfs restart
[16:23:21] <gene> sudo init.d/autofs restart
[16:23:31] <SWPadnos> ok
[16:24:32] <JymmmEMC> ko
[16:25:25] <gene> and after enabling the misc menu, then configuring the misc file with this: usbstick -fstype=vfat :/dev/sda1 <typu there, I'll fix
[16:25:38] <JymmmEMC> "You know your a redneck when you go and test if something squeals"
[16:27:34] <gene> yup
[16:28:36] <skunkworks> so - who else eats the chocolate off of the outside of a 3 musketeer and eates the center last? anyone?
[16:28:38] <gene> I think there is some sort of a kde daemon getting in the way though, so I need to figure out how to kill that puppy
[16:31:19] <gene> can't kill it, and turning off all the options didn't help
[16:33:56] <gene> humm, stopping and restarting init.d/mountall.sh did mount it, as /media/sda1
[16:34:13] <gene> now see if I, gene, can write to it
[16:36:33] <gene> It worked! Time for a celebratory cup of coffee.
[16:36:47] <JymmmEMC> s/coffee/moonshine/
[16:36:56] <Dallur> gene: autofs pretty cool right ?
[16:37:42] <JymmmEMC> gene: Sounds like your ready to start automounting remote SSH shells and samba shares now.
[16:37:47] <gene> Only thing I have there would be a half gallon of black jack, about 12 years old now (whats left of it that is)
[16:38:33] <JymmmEMC> or remote mount sftp exported as a samba share =)
[16:38:34] <gene> Yeah, but it put it where fstab said, not where auto.misc said
[16:39:07] <gene> so I don't think AAUTOFDS DID IT.
[16:39:13] <Dallur> hmf
[16:39:18] <Dallur> ;)
[16:39:31] <SWPadnos> just wait 'til you reboot ;)
[16:39:36] <SWPadnos> who oknows what'll happen
[16:39:40] <SWPadnos> -o
[16:40:09] <skunkworks> wh oknows what'll happen ?
[16:40:20] <JymmmEMC> The Shadow knows!
[16:41:02] <SWPadnos> exactly!
[16:41:42] <gene> the reboot, unless emc2 screws the moose, mnight be months down the log
[16:41:54] <skunkworks> http://clipcast.wpr.org:8080/ramgen/wpr/otr/otr070414c.rm
[16:43:34] <gene> is that link completew, no load here
[16:44:49] <JymmmEMC> REAL Media file
[16:46:15] <gene> probablt don't have realplayer installed, is it available to adept
[16:46:49] <JymmmEMC> You *MIGHT* see if VLC will play RM files.
[16:48:12] <gene> yeah but somethingn is stoppoing me from running anything now. wtf?
[16:49:49] <gene> gene@shop:~/emc2/configs/stepper$ emc
[16:49:52] <gene> EMC2 - 2.1.5
[16:49:57] <gene> Application initialization failed: couldn't connect to display ":0"
[16:50:00] <gene> Error in startup script: invalid command name "image"
[16:50:04] <gene> while executing
[16:50:08] <gene> "image create photo -file $f/$i.gif"
[16:50:13] <gene> invoked from within
[16:50:16] <gene> "if [file exists $f/$i.gif] {
[16:50:21] <gene> return [image create photo -file $f/$i.gif]
[16:50:24] <gene> }"
[16:50:29] <gene> (procedure "emc::image_search" line 7)
[16:50:33] <gene> invoked from within
[16:50:37] <gene> "emc::image_search emc2-wizard"
[16:50:41] <gene> invoked from within
[16:50:45] <gene> "set logo [emc::image_search emc2-wizard]"
[16:50:48] <gene> (file "/usr/share/emc/tcl/bin/pickconfig.tcl" line 36)
[16:50:48] <gene> Application initialization failed: couldn't connect to display ":0"
[16:50:49] <gene> and that took about a full minute
[16:50:49] <gene> maybe that reboot is closer than we thought, here goes
[16:54:13] <gene> back
[16:55:00] <gene> left the stick in, and its there again after the boot
[16:57:28] <JymmmEMC> You and your stick *sigh* (rolls eyes)
[16:57:56] <JymmmEMC> you have more issues with that dman thing
[16:58:18] <JymmmEMC> just setup file share like the rest of us
[16:59:15] <JymmmEMC> petev: So, hows the sprint broadband treating you?
[16:59:15] <gene> I wonder what turned into a cpu hog? Thats what it acted like, but emc is running ok now, with the stick still plugged in, one of those erros as I started it executing the logo is all.
[16:59:52] <gene> I think I will, it would be simpler in the long run thats for sure.
[17:00:24] <JymmmEMC> If you were booting off the thing, I might understand...
[17:00:52] <JymmmEMC> I'm assuming you DO have netorking on the box
[17:01:59] <gringos_> where are the optical encoder code ?
[17:02:17] <JymmmEMC> it's a secret
[17:02:25] <gene> yup, sure do
[17:02:32] <gene> ++++
[17:02:54] <gene> repeat that link & lemme see if FF can play it
[17:03:08] <gringos_> can we hack a usb mouse for our optical encoder ?
[17:03:15] <JymmmEMC> http://clipcast.wpr.org:8080/ramgen/wpr/otr/otr070414c.rm
[17:03:39] <JymmmEMC> gringos_: what is your SPR ?
[17:03:39] <gringos_> ir mouse
[17:03:57] <gringos_> what
[17:04:14] <JymmmEMC> IR?! as in optical mouse, or wireless IR mouse
[17:04:24] <gringos_> optical
[17:04:37] <JymmmEMC> SPR = Steps per revolution of your motors
[17:04:46] <gringos_> 200
[17:04:56] <JymmmEMC> Optical mice have a CCD camera in them
[17:05:05] <JymmmEMC> are you microstepping?
[17:05:20] <gringos_> was thinking about it
[17:05:27] <JymmmEMC> 8x ?
[17:05:42] <gringos_> I need to make emc run on 64 bit
[17:06:36] <JymmmEMC> 200 SPR * 8Microstepping = 1600 encoder (higher to be safer)
[17:06:57] <JymmmEMC> Most mice are 300-500
[17:07:07] <JymmmEMC> mechanical
[17:08:18] <gene> thanks for the link, I don't see any mention of real in the vlc stuff thats installing now though
[17:08:34] <gene> opticals are now above 1000dpi
[17:08:52] <JymmmEMC> gene: they are using cameras in them
[17:09:17] <gene> ppir type gizmo's I assume
[17:09:24] <gene> pir type gizmo's I assume
[17:09:38] <JymmmEMC> gringos_: http://www.usdigital.com/products/e6/ with cable = $100/ea
[17:10:13] <JymmmEMC> 2000 ppr. 1/4" thru-hole shaft
[17:11:12] <JymmmEMC> $97.52 + tx & shipping
[17:12:21] <gringos_> even an old ps2 mouse
[17:12:27] <gringos_> they have optical encoders on em
[17:12:41] <JymmmEMC> Most mice are 300-500
[17:12:54] <JymmmEMC> you need 1600 minimum
[17:12:54] <gringos_> thats enough for low speed cnc
[17:13:03] <JymmmEMC> if you micrstep
[17:13:57] <skunkworks> http://members.shaw.ca/swstuff/mouse.html
[17:14:17] <gene> looks like they bstart at $60.71 for onsies
[17:14:25] <skunkworks> LawrenceG on here -->
[17:15:15] <skunkworks> you can buy the bits also - the encoder disk and sensor for a bit less.
[17:16:03] <gringos_> wheres the encoder code ?
[17:16:20] <skunkworks> http://www.electronicsam.com/images/KandT/servostart/Encoder.JPG
[17:16:35] <skunkworks> encoder code?
[17:16:48] <gringos_> for emc
[17:18:59] <gene> skunkworks:now thats a cute deal
[17:20:21] <gene> I'm going to go play with this file & see what I can make out of it
[17:20:26] <gene> bbl
[17:25:00] <SWPadnos> you don't need the high res ones
[17:25:12] <SWPadnos> that's 2000 CPR, not PPR
[17:25:20] <SWPadnos> those would give 8000 counts/rev
[17:28:02] <ds2> crazy question - can EMC act as a DDS if I put a DAC on the parallel port?
[17:28:36] <SWPadnos> it should be pretty easy to write a DDS module if you need it
[17:28:56] <ds2> (and the follow on question is, can it do it in 3 phase so it can as a VFD if I put amps on it)
[17:28:56] <SWPadnos> but the update rate and jitter may not be so great (compared to a microcontroller)
[17:29:11] <ds2> don't think the jitter is that big of a deal for use as a VFD
[17:29:25] <SWPadnos> heh - gonna need 3 parports (for 8-12 bit resolution) if you're thinking of an R-2R ladder DAC
[17:29:52] <ds2> why 3? I think there are enough pins that I can use have 2 extra ones for use as address lines
[17:30:08] <SWPadnos> ok, if you're making custom hardware, then sure :)
[17:30:25] <ds2> 74138 + DAC + transistors... not very custom
[17:30:56] <ds2> but this is one way of generating a synchronous spindle
[17:32:07] <ds2> oh and what exactly is 'pretty easy'? any idiot off the street easy or ?
[17:32:19] <SWPadnos> any idiot who can write a HAL module ;)
[17:32:23] <ds2> oh
[17:32:58] <SWPadnos> you could probably make a siggen variant that has a "phase shift" input, and use 3 of them (woth 0,120,240 shifts)
[17:33:10] <ds2> so this will not be a 1 weekend project for me then. :/ got a pile of tiny 3PH motors to play with (hard drive spindle motors) and it should be small enough a few 2N2222's should do it
[17:33:29] <SWPadnos> then you'd need a custom driver for the addressed parport output (similar to but much less complex than PPMC)
[17:33:32] <ds2> don't that give me 3ph square waves?
[17:33:42] <SWPadnos> siggen does several waveforms
[17:33:52] <ds2> Oh... Hmmm
[17:33:53] <SWPadnos> sine, square, triangle, sawtooth I think
[17:34:07] <ds2> and siggen understands a single DAC?
[17:34:35] <SWPadnos> siggen has pins (now) for amplitude and frequency, and has a float output (for each signal type - they're all done simultaneously)
[17:35:03] <ds2> hmm
[17:35:16] <SWPadnos> so you'd run the same amplitude and freq into all of them, and set phase shifts according to need (+/- 120 or whatever)
[17:35:35] <SWPadnos> getting the float signal out the parport is another job that I'm not sure has an existing solution
[17:35:45] <ds2> I do have a few 4S3P boards around
[17:36:12] <SWPadnos> so you'd need a driver for those, including the addressable DACs on the outside
[17:36:27] <ds2> if I use the 4S3P board, I don't need addressing
[17:36:47] <SWPadnos> I'm not sure there are still HAL components to do format conversion, from float -> int -> bits, so you may need to visit that as well
[17:36:48] <ds2> /STROBE to /CS and D[0-7] to D[0-7]
[17:36:56] <ds2> ohhhh by float you mean that
[17:37:01] <SWPadnos> yes
[17:37:38] <SWPadnos> a simple scaling HAL block would be good (and has been discussed before, and may exist in one form or another)
[17:37:55] <ds2> too much of a time drain for the next few weeks :/
[17:37:58] <SWPadnos> yep ;)
[17:38:33] <ds2> but it would turn EMC into a brushless servo driver with just a plain amp
[17:39:00] <SWPadnos> there's a simpler way, if you have a 5i20
[17:39:08] <gringos_> whats spindlesync for ?
[17:39:23] <ds2> I suspect a 5i20 is more expensive then a handful of transistors
[17:39:26] <SWPadnos> they have a BLDC motor driver card, but I'm not sure it's compatible with the EMC FPGA config
[17:39:40] <SWPadnos> hmmm. yep (depending on the transistors)
[17:39:41] <ds2> but if I am really clever, it would not use a DAC
[17:39:47] <ds2> just PWM direct
[17:40:20] <ds2> otherwise, transistors will get hot and I am back to designing icky analog circuits :/
[17:40:31] <skunkworks> and if you wait a few months - jmkasunich will probably have a 3 phase spindle control in the mesa ;) iirc
[17:40:48] <SWPadnos> the 5i20 would allow that directly with a modified siggen
[17:41:03] <ds2> but all those require extra hardware... transistors are cheaper
[17:41:12] <SWPadnos> time is more expensive ...
[17:41:17] <ds2> *nod*
[17:41:27] <SWPadnos> it all depends on priorities
[17:41:36] <ds2> the other thing is - I have some of the transistors around... think CA calls them hazmat ;)
[17:41:58] <SWPadnos> heh - you're a non RoHS criminal!
[17:42:01] <ds2> and performance is not that big of a deal for me... I am more interested in avoiding stepper stall
[17:42:09] <ds2> $@#!#(#*!# RoHS
[17:43:43] <JymmmEMC> * JymmmEMC hands ds2 20 gallons of lead based paint
[17:43:59] <ds2> is it dried out? ;)
[17:44:05] <JymmmEMC> nope
[17:44:12] <ds2> you can keep it then :P
[17:44:25] <JymmmEMC> No, no I insist...
[17:44:45] <ds2> * ds2 calls the EPA to report an illegal dumping ;)
[17:45:23] <JymmmEMC> Hey, you painted in your house, not my problem
[17:45:40] <ds2> * ds2 has never seen it before ;)
[17:45:47] <gringos_> gringos_ is now known as VteCk
[17:45:56] <ds2> it was hard enough disposing of latex paint....
[17:46:13] <ds2> damn paint store won't take it... had to do the full dry it out and....
[17:46:18] <VteCk> VinTeCk
[17:46:24] <VteCk> VteCk is now known as VinTeCk
[17:46:49] <Dallur> lol, I got 3 metric tons of lead bricks in my workshop as we speak, any takers ?
[17:47:17] <JymmmEMC> Dallur: Free Delivery?! =)
[17:47:39] <ds2> Jymmm: he might send it express after casting them into little round peices ;)
[17:47:43] <slundell> Dallur I'll take 300Kg if its free shipping :)
[17:47:45] <Dallur> JymmmEMC: I actually need those for ballast but hey ...
[17:48:17] <JymmmEMC> lol
[17:49:13] <Dallur> nahh ,, just makes me sad thinking about regulations like that they ban lead from electronics causing no end of problems with wiskers and such, when the replacement materials cause huge headaches for the metal recycling industry.. :(
[17:49:56] <VinTeCk> while taping something there should allways be a proper direction to go
[17:50:33] <VinTeCk> I hate removing gobs of tape
[17:50:52] <ds2> heat it to around 2500C
[17:50:52] <VinTeCk> electrical is notbad
[17:50:57] <ds2> no more tape ;)
[17:51:01] <VinTeCk> huhu
[17:51:19] <VinTeCk> no more wire coating either
[17:51:25] <ds2> may even be profitable
[17:51:44] <ds2> should be enough to boil off most things and leave just gold and all the other good stuff }:-)
[18:58:40] <gene> well, that bombed, I got the stick to the main machine and its not a storage deviced anymore.
[19:01:01] <SWPadnos> I don't think that's a problem that can be caused by automount (or any other mounting)
[19:35:51] <eric_u> isn't that a problem that can be caused by writing and not syncing before removing the device?
[19:37:11] <SWPadnos> not for the USB class
[19:37:27] <SWPadnos> you can certainly screw up the data (ie, the partitions), but not the class of the device
[20:27:53] <toastydeath> bo jangles
[20:28:08] <SWPadnos> that's Mr. bo jangles to you!
[20:42:44] <toastydeath> srsly
[21:50:30] <cradek> jepler: whee, the new cable fixes pwm.0
[21:51:00] <cradek> and din.00
[21:51:08] <anonimasu> hehe
[22:08:38] <jepler> cradek: oh good
[22:11:55] <cradek> jepler: whee I have closed-loop
[22:12:16] <cradek> everything appears to work - time to start building
[22:12:36] <cradek> well I didn't try the differential yet - better check it first
[22:22:24] <cradek> anyone ever heard of a 504 line encoder?
[22:27:04] <CIA-8> 03cradek 07TRUNK * 10emc2/configs/lathe-pluto/lathe-pluto.hal: looks like some names changed
[22:27:09] <jepler> is this a trick question?
[22:27:25] <cradek> that's what this motor appears to have
[22:27:41] <cradek> I marked it and turned it 100 times, and the counts read 201600
[22:28:03] <jepler> the existing motors on your lathe?
[22:28:06] <cradek> when I set scale 2016, the mark stays lined up when I "jog" to any mm
[22:28:07] <jepler> or some other motor?
[22:28:12] <cradek> no, a different motor
[22:28:16] <cradek> unknown encoder
[22:28:18] <jepler> oh good
[22:28:22] <cradek> it just seems so unlikely a number
[22:28:24] <jepler> I thought it was a subtle way of submitting a bug report
[22:28:37] <jepler> 504=2*2*2*3*3*7
[22:28:39] <cradek> nope
[22:29:08] <cradek> I'll open it and see if it's marked somehow
[22:30:00] <jepler> are you going to try to auto-tune your lathe?
[22:30:05] <jepler> or is it not quite that close to ready
[22:30:05] <cradek> yes
[22:30:11] <cradek> no it's not at all ready
[22:30:13] <SWPadnos> 504 is for metric screws, I bet
[22:30:16] <cradek> but I might auto-tune this motor
[22:30:25] <anonimasu> hm
[22:30:29] <SWPadnos> hmmm, that should be 508 though
[22:30:29] <anonimasu> does autotuning work well?
[22:30:52] <cradek> anonimasu: I think it's only tested on one machine so far
[22:30:55] <jepler> SWPadnos: yeah I had that thought too
[22:31:01] <anonimasu> cradek: let me know how it works :)
[22:31:06] <jepler> anonimasu: it's pretty new but I think it's based on well-tested theory (ziegler-nichols?)
[22:31:16] <anonimasu> yep
[22:31:20] <anonimasu> I like "auto" stuff
[22:31:51] <cradek> grumble grumble torx
[22:32:42] <cradek> well whaddya know - it says 504 right on the wheel
[22:33:25] <SWPadnos> maybe it was supposed to be a 404
[22:33:33] <SWPadnos> error - encoder pitch not found
[22:34:21] <cradek> * cradek shrugs
[22:34:31] <cradek> I think that means pluto counts right (I was a little concerned)
[22:39:07] <skunkworksemc> updating 43 of 57 27.9kB/s
[22:39:27] <cradek> skunkworksemc: now that it's wired right, my breakout/motor driver board works
[22:39:35] <skunkworksemc> The probe is hooked up - seems to work. Just not mounted
[22:39:44] <skunkworksemc> cradek: Very nice
[22:39:55] <cradek> cool, are you going to probe tool lengths?
[22:40:33] <skunkworksemc> I hope to (not like you did - just setting the z coordinate offset for each tool)
[22:40:52] <skunkworksemc> knowing the distance between the switch and top of circuit board
[22:41:05] <skunkworksemc> I don't have a z home switch yet
[22:41:37] <skunkworksemc> I am glad jmkasunich_ isn't around.. I did a linkpp ;)
[22:41:39] <cradek> your scheme will really mess up the axis preview
[22:41:58] <cradek> I mostly use "net" now - it's nice
[22:42:11] <skunkworksemc> It didn't seem to - it seemed to move the part.. but time will tell. (I was using g10....)
[22:42:26] <cradek> we'll see :-)
[22:44:03] <skunkworksemc> cradek: are you going to try to tune it with the new script/whatever that pete was working on?
[22:44:05] <cradek> SWPadnos: this motor has a 21 tooth pulley on it, and 504/21 is an integer
[22:44:10] <cradek> skunkworksemc: yes definitely
[22:45:22] <SWPadnos> hmm. 24 counts/tooth. ok :)
[22:46:35] <cradek> I bet it gives some nice dpi if I knew the diameter (motor is from a printer or plotter)
[22:47:20] <SWPadnos> does it look like a rack/pinion or belt drive thing?
[22:47:29] <cradek> it was surely a belt
[22:48:18] <cradek> pitch diameter might be 0.5 inch
[22:48:47] <anonimasu> sioraiocht: File.Exists("file.txt")
[22:48:52] <SWPadnos> try to estimate the number of teeth per inch
[22:51:56] <cradek> might be 16
[22:52:38] <skinnypuppy1334> Anyone here running EMC on feisty 7.04?
[22:53:15] <cradek> using dapper here
[22:53:30] <cradek> should work fine if you get a realtime kernel built
[22:58:53] <skinnypuppy1334> What is the current version of EMC, the 2.1.5 release, or the emc2 live cd? I don't mind compiling RT if it would be newer version than live cd.
[22:59:19] <skinnypuppy1334> I'm taking that info from http://www.linuxcnc.org/content/view/2/4/lang,en/
[22:59:31] <cradek> if you install from the live cd, when you connect to the net, emc2 will update to 2.1.5 which is the latest release
[22:59:41] <cradek> it's very easy
[23:02:35] <skinnypuppy1334> I'm only wondering b/c 7.04 has a lot of things working for me that never would in 6.06 and 6.10. Beryl and VMserver are pretty cool items, and my .FLV files saved from youtube have audio this version...
[23:03:18] <cradek> you cannot use the nvidia driver and hope to run machinery with emc, because it is incompatible with realtime. That means beryl is right out.
[23:03:40] <cradek> do you want to control machinery, or just experiment with emc? there's a simulator mode that does not require realtime and can therefore run on any OS
[23:03:54] <skinnypuppy1334> SIm mode now.
[23:04:30] <cradek> then you could just build in simulator mode from source after installing some dependencies. See the wiki for instructions.
[23:05:19] <skinnypuppy1334> I've got a bridgeport I am interested in fitting with steppers geckos..but it will get a dedicated low tech box
[23:06:15] <skinnypuppy1334> As of now I was thrilled to be able to create a Vm and use mastercam and winderz without leaving Ubuntu...
[23:06:19] <cradek> yes that's the right approach
[23:06:32] <cradek> get an old PIII for the mill
[23:07:23] <skinnypuppy1334> Got atleast two p3's in the attic and a 1ghz amd
[23:08:08] <cradek> somehow I ended up with a stack of matrox millenium II cards - in a P3 they are great for emc
[23:08:25] <cradek> low tech is right
[23:08:38] <cradek> I've gotta run for a bit, let us know how the simulator build goes
[23:08:59] <skinnypuppy1334> nice to meet ya, thanks!
[23:09:05] <cradek> welcome
[23:20:19] <skinnypuppy1334> Any Bridgeport stepper retrofit users here?
[23:20:36] <SWPadnos> steppers. hahahahah
[23:20:40] <SWPadnos> no, not at the moment :)
[23:21:46] <cradek> do you mean the series-1 type original steppers with the fins?
[23:22:32] <skinnypuppy1334> I've got a 10x54 enco mill , not ser1 bp
[23:24:27] <skinnypuppy1334> Home shop with 13" enco lathe, 17" Clausing and a B&S 6x18 grinder.
[23:24:49] <cradek> all manual so far?
[23:25:46] <skinnypuppy1334> Yep. I have been too busy at the tech school, 60 miles one way, Haas sl20 with live tooling and BP with heidenhein...
[23:26:19] <skinnypuppy1334> I got all my machines at the right price at the right time..
[23:26:50] <anonimasu> skinnypuppy1334: how do you like the heid controllers?
[23:27:14] <skinnypuppy1334> I've got a blacksmith/metal artist as a friend and do a lot of machining to keep his machines projects going
[23:28:12] <skinnypuppy1334> I like what can be done with a heidenhein as far as polar offsets and scaling. Lables make it easy to repeat things without a lot of code.
[23:31:07] <skinnypuppy1334> more than anything I think EMC/steppers would be good for me in combination with the DRO on the mill, just to speed up repetative drilling
[23:32:22] <SWPadnos> do you have the motors already, or are you concerned about the price / complexity of servo systems? (or something else)
[23:35:35] <skinnypuppy1334> I've heard widely varying info on how many oz-in I need for the x-y steppers mostly. I'm not scared about wiring some geckos or making a good powersupply
[23:36:17] <SWPadnos> the torque you need is very dependent on how fast you want your rapids to be, and how deep cuts you expect to make
[23:36:42] <SWPadnos> I have a BP series 1 manual machine, and I got servos from an Anilam retrofit
[23:36:43] <skinnypuppy1334> basically anywhere b/t 450 oz in and 1500
[23:37:33] <SWPadnos> they're 27 in-lb, and most servo retrofits use servos in the range of 27-30 in-lb (40-60 on the knee, if that's motorized)
[23:38:08] <SWPadnos> 27 in-lb is ~430 oz-in, but the servo can deliver ~4x that torque in short bursts, and it can deliver full torque at any speed, unlike a stepper
[23:38:55] <SWPadnos> yeah - I'd say 430 would let you move the mill and do some cutting, but closer to 1000 would be better for "real machining"
[23:39:29] <skinnypuppy1334> I don't really have the money for ballscrews so I know it limits
[23:39:38] <ds2> my my those are big motors
[23:39:56] <SWPadnos> that increases the required torque, since acme screws are less efficient
[23:40:10] <SWPadnos> ds2, which, mine or roughly equivalent steppers? ;)
[23:40:20] <skunkworksemc> question
[23:40:25] <ds2> =)
[23:40:48] <ds2> 430oz-in motors would probally turn my lead screws into pretzels
[23:40:58] <SWPadnos> what kind of machine?
[23:41:02] <ds2> Taig =)
[23:41:14] <SWPadnos> heh, they'd turn your column into a pretzel ;)
[23:41:38] <ds2> there was an interesting thread on 305oz-in steppers on the CAD_CAM_DRO list
[23:41:59] <skunkworksemc> How would I display the #5063 if I am not running trunk. It doesn't seem to be in the var file and (msg, #5063) or (msg, [#5063]) doesn't work
[23:42:01] <ds2> people are arguing that these stronger motors won't do that much better
[23:42:09] <skinnypuppy1334> I wish I had time to keep up with all those lists.
[23:42:25] <ds2> so do i ;)
[23:42:35] <SWPadnos> I read through that thread
[23:42:37] <cradek> skunkworksemc: one way is to make M101 print to stdout, then M101 #5063
[23:43:59] <SWPadnos> too bad ballendo's original reply had so much incorrect information in it ;)
[23:44:00] <skunkworksemc> cradek: uh huh..
[23:44:00] <cradek> skunkworksemc: also, if you manually make a 5063 line in the var file, it will save it in there
[23:44:00] <SWPadnos> M101 P#5063
[23:44:00] <cradek> right, P
[23:44:00] <ds2> comments on the thread?
[23:44:00] <ds2> really? thought most of what he said is about the same as what Tony has been saying for a while
[23:44:04] <SWPadnos> the thread was mostly about those specific motors, and the fact that the higher inductance would require much higher voltages to get similar speeds as the 200 oz-in motors
[23:44:25] <SWPadnos> he was totally wrong about why steppers lose torque at high speed - talking about back EMF being the culprit
[23:44:43] <SWPadnos> if that were the case, then servos would have the same problem, because they have back EMF too
[23:45:04] <ds2> Hmmm
[23:45:22] <skinnypuppy1334> Motor windings are a coil their resistance goes up as the freq does... current drops, intern power is effected
[23:45:48] <ds2> eh? 'resistance'?
[23:45:56] <SWPadnos> reactance ...
[23:46:13] <skinnypuppy1334> it is inductive reactance
[23:46:18] <SWPadnos> yes, but steppers lose torque because they have to overcome detent torque. this is negligible at low speeds, since torque is limited by current limits
[23:46:39] <ds2> but reactance has no "power" consumption
[23:46:55] <SWPadnos> at higher speeds, the detent torque curve overlaps the current-limited torque curve, and reduces torque available to the load
[23:47:42] <SWPadnos> for the most part, on a torque vs. speed graph you have a flat horizontal line that represents the current limit of the windings
[23:47:59] <skinnypuppy1334> Sorry wasn't trying to say the reactance ate some of the power...
[23:48:16] <ds2> Hmmm
[23:48:24] <SWPadnos> and a decreasing straight line that represents the detent torque (actually, it's -detent torque, since that's subtracted)
[23:48:33] <SWPadnos> where they cross, you have the knee ...
[23:48:43] <SWPadnos> and the motor output is the lower of the tow lines
[23:48:44] <ds2> isn't the detent torque fixed for a given motor?
[23:48:49] <SWPadnos> two, not tow
[23:49:09] <SWPadnos> yes, but it's a certain amount of power *every time* you move the rotor out of a detent
[23:49:18] <SWPadnos> so 200 times per revolution ...
[23:49:55] <SWPadnos> if that's 1/1000 watt, then at 1 RPM, detent torque uses up 200 mW per minute
[23:50:09] <SWPadnos> err - joules ...
[23:50:52] <ds2> and this power is going torward heating the motor then?
[23:51:09] <SWPadnos> it should heat the iron, I think
[23:51:21] <skinnypuppy1334> I'm gonna get outside for a few before it gets dark. Thank you all again
[23:51:56] <SWPadnos> have fun. see you later
[23:51:59] <ds2> so that heat (and curie temp) would be one constraint on max speed then
[23:52:20] <SWPadnos> sure, but I think you're well beyond the knee at that point
[23:52:22] <ds2> not necessarily just the current driven into the coils and the resultant I2R losses
[23:52:52] <SWPadnos> the main reason it's a limit on max speed is that you're removing load driving ability
[23:53:19] <SWPadnos> the detent torque should be on the datasheets, though I'm not sure how to use it in an equation :)
[23:53:32] <ds2> but for smaller machines where you don't have much need for the torque...
[23:53:52] <ds2> i.e. motors running in overkill mode for the few corner cases where you need the torque
[23:53:58] <SWPadnos> the torque or speed will limit you, otherwise there would be no limits ... :)
[23:54:27] <SWPadnos> I agree with one main point in that thread - that it's a very application-dependent decision
[23:54:41] <ds2> I think the problem that happens way before even that is resonance... think that's Jymmm's problem and it is also what I have seen
[23:54:54] <SWPadnos> that's solvable with the right drives
[23:55:09] <SWPadnos> but it would be a limit without anti-resonance drives
[23:55:18] <ds2> hmmm
[23:55:23] <skunkworksemc> cradek: adding it to the var file worked for me. looks like min to max is .0007
[23:55:31] <skunkworksemc> over about 10 hits
[23:55:35] <ds2> I just slowed things down to work around it but is annoying
[23:55:45] <SWPadnos> the thing people have to remember (and often seem to forget) is that the lowest limiting factor in the system will be the controlling factor in the system
[23:56:40] <ds2> but based on everything... would a normal 200oz-in motor do just as well as those "305 oz" ones in most "typical hobby apps"
[23:56:55] <SWPadnos> it depends on the driver, I think
[23:57:23] <ds2> assuming an average, typical, hobby app driver =)
[23:57:35] <SWPadnos> heh - is that a gecko or xylotex? ;)
[23:57:51] <ds2> Xylotex's think class is more common
[23:58:04] <ds2> geckos are the cadalliac drivers and $$$$
[23:58:44] <SWPadnos> the 305 should give better push at low speeds. if the machine is meant to be run at higher speeds, then the 200's may end up being better - assuming the same supply voltage (fairly low)
[23:59:56] <ds2> so with all this said, am I correct in assuming you would pick a servo setup over steppers if it was the same $$?