#emc | Logs for 2005-11-22

[00:02:30] <les_w> well, off for some dinner for me. biab
[00:13:46] <alex_joni> * alex_joni is gone ... but he'll be back
[01:22:29] <fenn> hmmm... guessing the root password on a system i havent used in 3 years...
[01:22:41] <SWPadnos> 'password'
[01:22:51] <fenn> i coulda swore it was pb4ugo2bed
[01:22:52] <Jacky^> fenn: use a floppy ?
[01:23:00] <Jacky^> chroot
[01:23:05] <SWPadnos> init 1
[01:23:57] <fenn> telinit: must be superuser
[01:24:16] <SWPadnos> bummer
[01:24:32] <SWPadnos> can you reboot, and pass the kernel S or single as a parameter?
[01:24:55] <SWPadnos> (or is ctl-alt-del limited to root?)
[01:25:07] <fenn> i'm about to wipe it anyway.. just wanted to see if anything cool was on the disk
[01:27:37] <fenn> the idea is to install puppy-emc2 over ethernet
[01:27:47] <SWPadnos> ah
[01:28:02] <SWPadnos> wouldn't a CD be easier?
[01:28:10] <fenn> i'm hoping to hijack the redhat 7.3 so i can use my pcmcia ethernet card
[01:28:13] <fenn> no cd drive
[01:28:22] <SWPadnos> well - then it may noe be easier :)
[01:28:26] <SWPadnos> not
[01:28:41] <cradek> fenn: that's what single user mode is for
[01:29:52] <Jacky^> G night guys
[01:30:00] <SWPadnos> bight
[01:30:01] <Jacky^> Jacky^ is now known as Jacky^afk
[01:30:05] <SWPadnos> night
[01:31:31] <fenn> ok so i've got a boot: prompt
[01:31:38] <fenn> how do i pass options to the kernel?
[01:31:43] <cradek> linux init=/bin/sh
[01:31:52] <SWPadnos> or linux single
[01:32:01] <SWPadnos> or possibly linux initrunlevel=1
[01:32:51] <fenn> cool it worked
[01:33:04] <Jymmm> fenn you're not done yet =)
[01:33:16] <fenn> well now i can look around at least
[01:33:22] <Jymmm> * Jymmm dont remember the rest to reset the password though =)
[01:33:46] <SWPadnos> rm /etc/passwd ;)
[01:34:00] <Jymmm> fenn and dont listen to SWPadnos
[01:34:14] <SWPadnos> yeah - he's full of shite
[01:34:22] <SWPadnos> (that's polite for shit)
[01:34:41] <Jymmm> fenn ^ from the horses smartass direct!
[01:36:43] <fenn> i can just delete the encrypted passwd field and it should let me login with no password
[01:37:17] <SWPadnos> yeah - something like that, unless shadow passwords or PAM or something else was set up (I think)
[01:38:27] <Jymmm> you need to run passwd ro reset root
[01:38:34] <Jymmm> you need to run passwd to reset root
[01:40:29] <jtr> What's the best way to pull down emc2 to see if you could aid in the development effort?
[01:41:09] <jtr> Oh - i.e. anonymous or sign up as a developer?
[01:43:11] <jtr> Been lurking (well, the system has) since the beginning of yesterday's meeting. jtr = Jim Register - long time lurker on emc-users from the old nist list.
[01:43:26] <SWPadnos> anonymous CVS is fine, it just lags behind by a couple of hours
[01:45:08] <jtr> cool, thanks. Been programming the last mumble years in the AMX Axcess language - don't know if I remember any c
[01:45:55] <jtr> off to sourceforge now.
[01:55:28] <jtr> hmmmn - it asks for a password, which I assumed to be my email addy - should have just been a <cr> maybe?
[01:58:07] <jtr> Scrolled down on the wiki page, saw the link for the cvs book - I'll start there.
[02:00:51] <skunkworks> just an update - for some reason I thought my ini file period was set to .000016. It wasn't - it was set to .000022. Now I can get 25ipm from z. If I go 1:1 on the drive I should be able to double that.
[02:00:57] <skunkworks> that would be fine for me.
[02:01:04] <jepler> until you're a developer or want to submit patches, CVS is pretty easy to use. "cvs co" to check out files (the first time) then "cvs update" to get changes since your last update or checkout (from time to time)
[02:10:49] <jtr> Ok - the command line at http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?CVS is a little more involved.
[02:12:11] <jepler> I probably over-simplified
[02:13:43] <jtr> Seems like I read something somewhere about adding some lines to a "sources" file on my system
[02:14:14] <jepler> That might be /etc/apt/sources.list
[02:14:39] <jepler> apt is the software debian uses to fetch software updates and install new packages
[02:14:45] <jepler> bdi is based on debian
[02:14:53] <jtr> Thanks, But I think there's a lot of reading to be done first re:cvs.
[02:15:51] <jtr> I'm responding about a line or two behind, but yeah I finally got around to putting the BDI on a system.
[02:16:06] <jtr> Had to buy my bride a new one first.
[02:17:38] <jtr> I managed to mess up the BDI install - thought I knew better than X's probe what kind of video card I had.
[02:19:52] <jtr> had to go find out what kind of X it was running - XSetup weren't the answer. I took notes - maybe that should go in the wiki
[02:20:25] <jtr> somewhere where it doesn't distract from the main point, which is EMC.
[02:23:24] <jtr> jepler: thanks for the help; sorry to have hogged the channel
[02:26:34] <jepler> jtr: any kind of notes are welcome in the wiki
[03:07:11] <jepler> someone know the new mirror of paul_c's deb files? mirrors.neuron.com (listed in my months-old sources.list) doesn't seem to work these days
[03:08:46] <jepler> maybe this is what I want: http://solaris.cs.utt.ro/emc/debian/
[03:10:31] <jepler> no, that doesn't seem to work. :(
[03:12:10] <jepler> oh well
[03:14:03] <SWPadnos_> jepler: I'm trying to install axis on a mostly screwed up machine - do you know what package(s) I need to install when setup.py gives an error about _tkinter in togl_setup.py?
[03:14:59] <jepler> SWPadnos_: debian-based? python-tk redhat-based? tkinter
[03:15:30] <SWPadnos_> screwed-up BDI - partially (mostly) failed upgrade from 2.18-ish
[03:15:55] <jepler> 2.18? that's so old, I don't even know
[03:16:03] <SWPadnos_> pyhton2.3-dev is installed, and I can't install python2.3-tk because of a dependency
[03:16:19] <SWPadnos_> well - it's almost 4.30 now :)
[03:16:35] <jepler> if the error is short can you paste it here or in /msg?
[03:17:02] <SWPadnos_> I'll try - it's on a different machine, so it'll take a sec
[03:19:21] <SWPadnos_> # env EMC2_DIR=/Project/emc2-latest python setup.py install
[03:19:23] <SWPadnos_> Traceback (most recent call last):
[03:19:24] <SWPadnos_> File "setup.py", line 27, in ?
[03:19:26] <SWPadnos_> from togl_setup import get_togl_flags
[03:19:27] <SWPadnos_> File "setup/togl_setup.py", line 2, in ?
[03:19:29] <SWPadnos_> import _tkinter
[03:19:30] <SWPadnos_> ImportError: No module named _tkinter
[03:27:13] <SWPadnos_> well - I'm downgrading python2.3 from 2.3.5-8 to 2.3.5-3sarge1 so I can install python2.3-tk
[03:29:59] <fenn> jepler the solaris server has been down a couple days but it should work usually
[03:40:17] <rayh> must seen me coming!
[03:42:52] <SWPadnos_> I don't suppose anyone knows what OpenGL / Mesa packages are needed to build AXIS?
[03:46:16] <Jymmm> jepler and cradek might
[03:46:40] <fenn> SWP this all sounds pretty familiar
[03:47:11] <fenn> try installing the nvidia hardware acceleration :)
[03:47:30] <fenn> redhat did a terrible job packaging the opengl libraries
[03:47:54] <SWPadnos_> I'm not sure that nv acceleration would be good for a C&T video chip
[03:48:03] <SWPadnos_> and this is debian, BDI
[03:51:17] <rayh> I've got three here that built it axis today
[03:51:52] <SWPadnos_> any clue which of the 17000 mesa packages needs to be installed (or removed)?
[03:52:11] <rayh> let me switch boxes.
[03:52:26] <SWPadnos_> ok
[03:54:29] <rayh_emc2> xlibmesa-dev xlibmesa-gl xlibmesa-gl-dev xlibmesa-glu
[03:54:49] <rayh_emc2> xlibmesa-glu-dev
[03:54:55] <SWPadnos_> ok - how about opengl, and xlib?
[03:55:38] <rayh_emc2> python2.3-opengl python-opengl
[03:57:06] <SWPadnos_> actually - xlib or X11 would be useful (I have all the ones listed so far)
[03:57:11] <rayh_emc2> xlibs xlibs-data xlibs-dev xlibs-static-dev
[03:57:48] <rayh_emc2> I have no clue how many of those are essential or what each does.
[03:58:13] <rayh_emc2> Once I got the instructions for install it went without a hitch.
[03:58:39] <SWPadnos_> ok - thanks - some of the xlib packages are missing. I'll ask again if that doesn't fix it
[03:58:44] <rayh_emc2> I saw a goofy reaction from one of the axis menu lists but it only happened once.
[04:05:31] <rayh_emc2> looks like part of linuxcnc.org is down.
[04:19:12] <jmkasunich> what part is down?
[04:40:44] <CIA-5> 03jmkasunich * 10emc_HAL/farm_scripts/run_farm: changed farm scripts to save a history of good and bad compiles
[04:41:51] <rayh_emc2> I was missing images from org/images but they are there and being displayed now.
[05:10:50] <CIA-5> 03jmkasunich * 10emc_HAL/farm_scripts/run_farm: changed farm scripts to save a history of good and bad compiles (maybe it will work this time)
[05:26:57] <jtr> Still here, working my way through compiling EMC2 on a BDI-4.30 system
[05:27:45] <jtr> apt-get install cvs... updated cvs, which is asking for the path to the repository
[05:28:30] <jtr> is that going to be exactly this? :pserver:anonymous@cvs.sourceforge.net:/cvsroot/emc
[05:29:45] <CIA-5> 03paul_c * 10emc2-auto/wiki/ (32 files in 22 dirs): "Auto update wiki from a cron job. Tue Nov 22 05:30:02 GMT 2005 "
[05:58:21] <jtr> just a little more to figure out and document.
[07:06:50] <SWPadnos> SWPadnos is now known as SWP_Away
[22:22:25] <fenn> cool motion control q&a site: http://www.control.com/1026213484/
[22:59:03] <Jymmm> bookmarked
[23:08:33] <SWPadnos> Hi Dave
[23:09:06] <dave-e> hi swp
[23:09:48] <dave-e> I got on here to bug les but he doesn't appear to be around
[23:10:03] <SWPadnos> nope -doesn't look that way
[23:10:25] <SWPadnos> you just posted to the user and dev lists, right?
[23:10:27] <dave-e> I'm hoping someone will check out my last post
[23:10:38] <SWPadnos> checking now ;)
[23:10:38] <SWPadnos> how did you measure velocities in your testing?
[23:11:12] <dave-e> using the emc timing....and checking against the wall clock to make certain emc wasn't lying
[23:11:41] <SWPadnos> hmmm - I'd use a couple of marks on the table, and a stopwatch
[23:11:59] <dave-e> I've done that too
[23:12:02] <SWPadnos> also, you have to account for the accel and decel phases of the move
[23:12:17] <dave-e> ah yes but my accel is at 20 ....
[23:12:33] <SWPadnos> 20 IPS^2?
[23:12:44] <dave-e> faster than most machines, if you discount les
[23:13:04] <dave-e> a lot of BP's are at 4 or 6
[23:13:35] <dmess> but is the machine actually achieving that.. 20
[23:13:44] <dave-e> I know that is still fractional G's but reallistic for the age of the machine...
[23:13:49] <SWPadnos> ok - so it should only take 0.15 seconds to get to full speed
[23:13:57] <dave-e> I think so...
[23:14:16] <SWPadnos> and you (of course) tested each axis individually?
[23:14:22] <SWPadnos> ie, no compound moves
[23:14:33] <dmess> how much axis travel do you have??
[23:14:41] <dave-e> if it was struggling then the difference should show up at lower speeds
[23:14:57] <dave-e> I used 20 on x , 10 on Y and 5 on z
[23:15:23] <SWPadnos> sure, I'm just confirming that you ddid moves on each axis separately
[23:15:29] <dave-e> I have about 28 on X, 15 on Y and 7.5 on z
[23:15:43] <dave-e> yes, not combined moves
[23:15:48] <SWPadnos> ok.
[23:16:02] <dmess> higher speeds would be more critial due to the longer accel decel zones
[23:16:19] <dave-e> indeed
[23:16:23] <SWPadnos> as for the max_vel settings, I think the one in TRAJ is supposed to be the fastest combined tool speed, and each axis is further limited by the AXIS_N setting
[23:16:39] <SWPadnos> so it's effectively the lower of the two
[23:16:51] <dave-e> that is my thinking
[23:17:29] <dave-e> I asked Ray if anyone actually checked the Mazak at Cardinal Eng and he doesn't think so.
[23:17:41] <dmess> but if x & y are set differently your g17 circles will only be as fast as the sloest programmed axis
[23:17:54] <SWPadnos> hopefully :)
[23:18:15] <SWPadnos> it may be that the moves along the faster axis are faster, making an uneven finish (I'm not looking at the code, so I don't know)
[23:18:15] <dmess> big hope...
[23:18:27] <dave-e> well most of my real cutting is done below 20 (actual ) ips
[23:19:13] <dave-e> I'm just hoping a couple of people will go do real timing both with emc2 and with emc1 and post
[23:19:24] <dmess> accel is 20..ips... mot max speed
[23:19:27] <dmess> not
[23:19:39] <SWPadnos> 20 IPM, I'd assume ;)
[23:19:57] <dave-e> accell is set at 20 ips/sec/sec
[23:20:11] <dmess> oh thats more realistic then...
[23:20:48] <dave-e> please note that the speeds are relatively linear with commanded .... just not scaled correctly
[23:20:59] <dmess> but g02/g03 use accel/decell on 2 axes
[23:21:37] <dave-e> my circles are just fine....
[23:21:57] <dmess> hmm ok then never mind
[23:22:01] <dave-e> the only problem I have is chip load...if I believe the commanded velo
[23:23:01] <dmess> how well controlled is your spindle speed??
[23:23:03] <SWPadnos> are your input and output scales correct?
[23:23:13] <SWPadnos> (I guess tey'd have to be)
[23:23:19] <dave-e> they are
[23:23:31] <dmess> circles are round... ; (
[23:23:34] <dave-e> yep
[23:23:37] <SWPadnos> what hardware do you have?
[23:24:18] <dave-e> Mazak v5 with Servo Dynamaics amps running STG card on 2.18
[23:25:09] <SWPadnos> ok.
[23:26:36] <dave-e> JMK...you have any comments?
[23:27:22] <jmkasunich> the emails?
[23:28:14] <dave-e> yeh....
[23:28:19] <dave-e> axis speed
[23:28:24] <jmkasunich> did you see it on emc1, or 2?
[23:28:34] <dave-e> emc1
[23:29:07] <dave-e> and since a lot of the code is the same I'm wondering if anyone checked actual velo on the Mazak
[23:29:18] <jmkasunich> trying to remember
[23:29:28] <jmkasunich> I dunno if we timed a move or not
[23:29:57] <SWPadnos> I'd like to see more precise timing
[23:30:17] <dave-e> I'm guessing not....I don't think anyone does that kind of thing until a problem surfaces.
[23:30:22] <SWPadnos> something where only the constant velocity portion is measured, using physical switches for gating
[23:30:40] <jmkasunich> stepper machine is trivial to test
[23:30:46] <jmkasunich> don't need the machine, I can do it here
[23:30:51] <jmkasunich> use halscope
[23:31:05] <SWPadnos> it's an STG in Dave's machine
[23:31:11] <dave-e> guess I should plot the accel...
[23:31:17] <jmkasunich> on emc2, it will make no diff
[23:31:23] <jmkasunich> on emc1, I don't care ;-)
[23:31:44] <dave-e> I should think this should show under sim
[23:32:08] <SWPadnos> I can say that running under VMWare gives you significantly slower moves ;)
[23:32:33] <jmkasunich> running RT code under VMware is a joke
[23:32:50] <SWPadnos> yes - it's not really RT anymore ;)
[23:33:03] <SWPadnos> in fact, it's *ReALLY* not RT
[23:33:10] <SWPadnos> s/e/E/
[23:33:12] <dave-e> I'm working on limit switches ,etc now for bringing an old tracer mill up under emc2 ... but that is still a ways away.
[23:33:45] <jmkasunich> you were just doing MDI moves to test?
[23:33:59] <dave-e> jmk.... yep...emc1 is not your problem.
[23:34:09] <dave-e> mdi ...
[23:34:34] <dave-e> don't think auto changes it.
[23:36:02] <jmkasunich> hmm
[23:36:18] <jmkasunich> is G0 cutting feed or rapid
[23:36:30] <dave-e> g0 is rapid
[23:36:34] <jmkasunich> duh, ok
[23:37:38] <jmkasunich> oops... and F1 = 1 ipm, not 1 ips ;-)
[23:37:48] <SWPadnos> sloooooooow :)
[23:37:48] <dave-e> that is a good point tho.... I didn't try g0 to see if it bypasses the problem
[23:38:20] <dave-e> yep units are important
[23:38:44] <jmkasunich> ok, according to halscope, a 1 inch G1 move at F30 took a hair over 2 seconds
[23:38:48] <SWPadnos> you weren't in cm mode, were you?
[23:38:58] <dave-e> inch
[23:39:19] <SWPadnos> yeah - it shouldn't matter anyway - you would have noticed the short moves
[23:39:26] <SWPadnos> (I hope)
[23:39:31] <dave-e> I should hope so
[23:40:25] <dave-e> jmk ... what is your accel set at?
[23:40:34] <jmkasunich> have to check
[23:40:58] <jmkasunich> using the stock emc2 emc.ini
[23:41:23] <icee> so jmka, how bad is dc bias on the power lines? I'm not going to blow up lots of electronic stuff in my house am i?
[23:41:32] <dave-e> which probably defaults to 20
[23:41:33] <jmkasunich> TRAJ: default vel 0.0167, maxvel 1.2, default accel 15, max accel 20
[23:41:39] <icee> I've modelled it with SPICE and after the inrush currents it's not that bad
[23:41:50] <jmkasunich> icee - just a sec
[23:41:53] <icee> OK
[23:42:02] <jmkasunich> axis: maxvel 1.2
[23:42:05] <jmkasunich> max acc 20
[23:42:25] <jmkasunich> I'm setting up a differentiator on position to get velocity to scopee
[23:42:39] <dave-e> nice toys
[23:44:23] <SWPadnos> I should make a doohickey that has an encoder and a wheel, and records movement every millisecond or so, so you can use an external measurement tool
[23:44:52] <SWPadnos> measuring with a possbly broken thing soesn't sound like the best approach to me
[23:45:04] <SWPadnos> doesn't
[23:45:04] <jmkasunich> ok, accel from zero to ,5 inches/sec took 50ms
[23:45:25] <SWPadnos> sounds right to me
[23:45:39] <jmkasunich> good point - if the servo thread isn't ticking at 1ms, then scope wouldn't have a clue
[23:45:48] <SWPadnos> no - off by a factor of two
[23:46:10] <SWPadnos> 2 IPS should be 100 ms, so 0.5 IPS should be 25 ms
[23:46:54] <jmkasunich> something interesting - the velocity isn't a sharp ramp
[23:47:05] <icee> 15 * ((inches / second) / second) * (50 milliseconds) = 0.75 inches per second
[23:47:07] <jmkasunich> there is some rounding at beginning and end
[23:47:19] <SWPadnos> vel has an S-shape?
[23:47:20] <SWPadnos> odd
[23:47:34] <SWPadnos> that's PID, with no FF, right?
[23:47:41] <jmkasunich> yeah, at best the middle third is linear
[23:48:03] <icee> rounding isn't so strange
[23:48:04] <jmkasunich> I'm looking at the command out of the motion controller, before the PID
[23:48:10] <SWPadnos> oh
[23:48:20] <jmkasunich> let me try a faster speed
[23:48:35] <icee> oh, i didn't think the motion controller did any jerk limiting
[23:48:55] <jmkasunich> I didn;t either, thats why I'm confused
[23:49:30] <jmkasunich> ok, rounding is the same in absolute terms, when I doubled the speed the linear section got longer
[23:49:39] <jmkasunich> the first 20mS is rounded
[23:49:59] <SWPadnos_> is that an interpolation artifact?
[23:50:01] <jmkasunich> in 20mS it gets to about .200 ips
[23:50:28] <jmkasunich> then it takes 60mS to go from .2ips to .8ms
[23:50:37] <jmkasunich> correction
[23:50:45] <jmkasunich> 40ms to go from .2 to .8
[23:50:55] <jmkasunich> then another 20ms to level off at 1.0
[23:51:13] <icee> so 40ms is exactly the rate you'd expect for .6ips
[23:51:26] <icee> 15 * ((inches / second) / second) * (40 milliseconds) = 0.6 inches per second
[23:51:36] <jmkasunich> yeah, the linear ramp is right, but why the rounding?
[23:51:46] <SWPadnos_> cubic interpolation
[23:51:52] <jmkasunich> over 20mS?
[23:52:06] <SWPadnos_> maybe - I'm just throwing out (up) suggestions
[23:52:10] <icee> and the ramp rates for the first 20 milliseconds and last 20 milliseconds is exactly half
[23:52:15] <icee> i don't think the interpolation should cause that
[23:52:19] <icee> the average rates
[23:52:26] <icee> so it sounds like there may be some magic jerk limiting in there somewhere
[23:52:30] <jmkasunich> yeah
[23:52:42] <SWPadnos_> so it's always 20 ms? - try going to 10 IPS
[23:52:44] <jmkasunich> gonna try an accel to 0.2 ips
[23:53:02] <jmkasunich> need to edit ini and restart for anything over 1.2 ips
[23:53:06] <icee> i'd expect that to take 40 ms and be perfectly rounded
[23:53:08] <SWPadnos_> also, change the max_accel to some ungodly high number
[23:53:21] <icee> can you take a second derivative too, jmka?
[23:53:27] <icee> it would help to validate your methodology
[23:53:43] <jmkasunich> yes,
[23:53:45] <icee> if you see acceleration ramp up, level, and ramp down linearly
[23:54:09] <jmkasunich> accel from 0 to 0.2 took 30ms
[23:54:21] <SWPadnos_> note that you'll have artifacting due to the differentiator, though it should be only one scope cycle per level
[23:54:37] <icee> yah, i know.. this is still at 1KHz or something right?
[23:54:47] <SWPadnos_> (you don't have smoothing on the scope, do you?)
[23:54:54] <icee> e.g. nyquist is >500Hz
[23:55:14] <SWPadnos_> it would show up as a delayed spike
[23:55:56] <icee> second derivative should show you pretty nicely for sure what's happening
[23:56:10] <dave-e> I'm going to run.... maybe go back and try g0 and auto mode just for grins
[23:56:25] <icee> if -it- is rounded significantly, then there's probably a methodology problem
[23:56:43] <icee> if it is mostly trapezoidal, then what jmka is seeing is probably what's really happening
[23:57:38] <jmkasunich> the scope is sampling at 1KHz, the differentiator also runs at 1KHz
[23:58:07] <jmkasunich> I think I have an ini for simulation, I'm gonna restart with that, after adding two differentiators for vel and acc
[23:59:18] <SWPadnos_> you'll just have one or two invalid samples at the beginning and end