#emc | Logs for 2007-09-16

[01:45:30] <jmkasunich__> jmkasunich__ is now known as jmkasunich
[01:59:02] <rff> gees, long time since i been here, never seen so many logged on before!!
[02:07:18] <jmkasunich> so of us are just lurking
[02:07:22] <jmkasunich> * jmkasunich hides again
[02:09:32] <eric_u> tomp did you draw us any more gEDA symbols?
[03:45:14] <Unit41> whats a good mod and pitch to go with for percision wood work ? on flexible rack
[03:45:22] <Unit41> 1 mod or 2 ?
[03:45:28] <Unit41> and what pitch
[04:12:40] <mook> What kind of cad/cam programs are popular for use with EMC2?
[05:46:03] <Unit41> autocad and corel draw
[05:46:26] <Unit41> corel draw uses vector fonts
[05:46:50] <Unit41> autocad does not :P
[05:48:16] <Unit41> hire some slaves to run the cnc too
[05:48:46] <Unit41> a crate full of immigrents is lower than it was in 05
[05:49:17] <Unit41> shit you buy the crate and its like half off
[05:50:03] <Unit41> j/k I dont know what im talking about lol
[05:50:20] <Unit41> sheit
[11:27:46] <tomp> eric_u: see bottom of http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?M5i20_Halvcp_Test_Panel it has 5i20 7i33 7i37 in the tarball, breakout & yaskawa amps to come
[11:44:01] <tomp> eric_u: the 5i20 is config specific to HOSTMOT54E
[14:08:43] <tomp> any US feeds for the emc2 iso that might be faster than lised on wiki? ( i got eta of 7hrs :(
[14:09:03] <SWPadnos> the linuxcnc.org site is in california
[14:09:12] <tomp> thanks
[14:09:16] <SWPadnos> sure
[14:09:18] <JymmmEMC> the romanian site is fastest
[14:09:30] <SWPadnos> that would be strange
[14:09:33] <tomp> oh,,, thats the 7+hrs
[14:09:38] <tomp> ro
[14:09:57] <JymmmEMC> which iso?
[14:10:04] <JymmmEMC> live?
[14:10:09] <SWPadnos> JymmmEMC, are you serious that you don't get very high speeds from linuxcnc.org?
[14:10:24] <SWPadnos> you should be getting a few meg/second
[14:10:31] <tomp> yes live, must be my end of the stick, you'd think an abandoned industrial park would be ok on weekends
[14:10:43] <rayh> By being ahead of us time wise, those ro sites get a jump on sending.
[14:11:14] <SWPadnos> plus the rotation of the earth helps - we catch up to the buts faster
[14:11:16] <JymmmEMC> SWPadnos: I was being serious, but I'll test right now
[14:11:16] <SWPadnos> bits
[14:11:18] <alex_joni> it's a university network, so they have plenty of bandwidth
[14:11:25] <alex_joni> SWPadnos: haha
[14:12:46] <JymmmEMC> .ro is is jumping all around.
[14:12:57] <JymmmEMC> from 40K to 120K/s
[14:13:06] <JymmmEMC> eta right now is 2hr
[14:13:12] <alex_joni> that's extremely slow
[14:13:17] <JymmmEMC> Let me try from work....
[14:13:40] <alex_joni> mm.. perfect after-lunch snack (grapes & camembert)
[14:14:53] <SWPadnos> JymmmEMC, what about this: http://linuxcnc.org/iso/emc2-ubuntu6.06-desktop-i386.iso
[14:15:51] <JymmmEMC> SWPadnos: that last link, from work 200KB/s
[14:16:08] <SWPadnos> hmm. that's not as fast as I'd like
[14:16:55] <SWPadnos> just to be sure, you aren't transferring to home through your work connection, are you ? :)
[14:16:58] <alex_joni> I get more on that from over here
[14:16:59] <JymmmEMC> SWPadnos: LOL, from homw 300KB/s
[14:17:18] <JymmmEMC> SWPadnos: I'm sheeled into work
[14:17:20] <JymmmEMC> shelled
[14:17:31] <SWPadnos> weird
[14:17:45] <alex_joni> hmm.. it's down to 320kB now
[14:17:49] <SWPadnos> it sounds to me like there's something funny with your remote login
[14:17:49] <JymmmEMC> I'm using lynx too, so that may be an issue
[14:17:57] <alex_joni> * alex_joni is using wget
[14:18:17] <tomp> haha ping to linuxcnc.org 64 bytes from basic-ugly.dinero.dreamhost.com ( icmp_seq=1 ttl=48 time=3181 ms
[14:18:38] <SWPadnos> that's a loooooong ping
[14:18:41] <alex_joni> 64 bytes from basic-ugly.dinero.dreamhost.com ( icmp_seq=1 ttl=58 time=177 ms
[14:18:42] <JymmmEMC> where wget usually located?
[14:18:48] <tomp> pung
[14:18:54] <alex_joni> JymmmEMC: didn't get that
[14:19:03] <JymmmEMC> path
[14:19:10] <alex_joni> oh, along with lynx
[14:19:16] <alex_joni> I think /usr/bin
[14:19:17] <SWPadnos> maybe /usr/bin, along with everything else :)
[14:19:24] <alex_joni> juve@proxy:~$ which wget
[14:19:24] <alex_joni> /usr/bin/wget
[14:20:52] <JymmmEMC> heh, now it's 5.5M/s
[14:21:02] <JymmmEMC> eta 2m
[14:21:08] <SWPadnos> I thought there was something screwy going on
[14:21:11] <alex_joni> sounds a bit more like it
[14:21:55] <alex_joni> I get nice ping times from teh .ro site :D
[14:21:59] <JymmmEMC> I had to shell into another box
[14:22:06] <alex_joni> 64 bytes from dsplabs.cs.upt.ro ( icmp_seq=3 ttl=60 time=0.661 ms
[14:22:07] <tomp> what happened? 64 bytes from basic-ugly.dinero.dreamhost.com ( icmp_seq=4 ttl=48 time=120 ms
[14:22:21] <JymmmEMC> alex_joni: the 5M/s was from us
[14:22:29] <alex_joni> JymmmEMC: I figured
[14:23:12] <JymmmEMC> Connecting to dsplabs.utt.ro[]:80... connected.
[14:23:12] <JymmmEMC> HTTP request sent, awaiting response... 303 See Other
[14:23:12] <JymmmEMC> 07:47:45 ERROR 303: See Other.
[14:23:28] <alex_joni> heh, wtf is 303?
[14:24:08] <JymmmEMC> http://www.checkupdown.com/status/E303.html
[14:24:46] <alex_joni> JymmmEMC: did you get the complete link>
[14:24:48] <alex_joni> ?
[14:25:02] <JymmmEMC> http://dsplabs.utt.ro/~juve/emc/get.php?file=emc2-ubuntu6.06-desktop-i386.iso
[14:25:08] <alex_joni> http://dsplabs.cs.utt.ro/~juve/emc/emc2-ubuntu6.06-desktop-i386.iso
[14:25:21] <alex_joni> remove the 'get.php?file='
[14:25:31] <JymmmEMC> wth is it there in the first place?
[14:25:43] <alex_joni> statistics
[14:26:08] <JymmmEMC> eta 2hr
[14:26:17] <JymmmEMC> eta 2:30
[14:26:36] <JymmmEMC> from my work
[14:26:38] <alex_joni> pretty ok, given the distance
[14:26:49] <JymmmEMC> not really
[14:26:58] <alex_joni> can you count hops?
[14:27:22] <SWPadnos> no, they get too well mixed with the barley
[14:27:30] <SWPadnos> or malt
[14:28:00] <JymmmEMC> Ah, telia is the culprit
[14:28:11] <JymmmEMC> lots of timeouts too
[14:30:19] <JymmmEMC> checking from NL to ro I get 52ms
[14:31:00] <jlmjvm> Good Morning Everyone
[14:31:02] <alex_joni> hi
[14:31:04] <JymmmEMC> de to ro I get 58ms
[14:31:39] <JymmmEMC> UK to RO 60ms
[14:31:48] <alex_joni> depends on the site
[14:32:02] <JymmmEMC> alex_joni: at the univ you mean?
[14:32:06] <alex_joni> it's way less if it's one inside GEANT
[14:32:08] <alex_joni> JymmmEMC: yeah
[14:32:19] <alex_joni> geant is some european research network
[14:32:29] <alex_joni> jlmjvm: any luck?
[14:32:46] <JymmmEMC> NY to RO 131ms
[14:33:14] <alex_joni> http://www.geant.net/upload/pdf/Topology_Oct_2004.pdf
[14:33:31] <JymmmEMC> alex_joni: looks like the hop between telia and roedu has issues
[14:33:51] <jlmjvm> alex joni:just got up,havent tried it yet
[14:34:06] <alex_joni> might be
[14:35:23] <alex_joni> I get about 40msec to the last europe hop of telia
[14:35:34] <alex_joni> and about 210 to across the pond
[14:36:29] <JymmmEMC> alex_joni: from the uni or from work?
[14:37:19] <JymmmEMC> SWPadnos: 9ms from work to linuxcnc.org on a traceroute ;)
[14:37:51] <SWPadnos> not bad ;)
[14:38:14] <JymmmEMC> 9 hops
[14:38:23] <SWPadnos> oh - I was thinking it would be around 3 or so ;)
[14:38:27] <JymmmEMC> 3 internal to us
[14:38:32] <SWPadnos> ah
[14:38:50] <JymmmEMC> then it exits the 10GigE =)
[14:39:03] <JymmmEMC> 3 hops in internap
[14:39:10] <alex_joni> JymmmEMC: from uni
[14:39:18] <JymmmEMC> alex_joni: ah
[14:40:13] <alex_joni> 19 hops to linuxcnc.org from home (200msec)
[14:41:08] <alex_joni> 7 hops from work (that is quite less than I expected :)
[14:41:16] <SWPadnos> funny - it's only 9 hops from here also, but 100ms ping
[14:41:28] <alex_joni> about 180-190 from here
[14:42:26] <SWPadnos> of course, I've got 30 ms just to my ISP (over DSL)
[14:43:39] <JymmmEMC> lol
[14:44:01] <jlmjvm> how are yall checking your time?
[14:44:19] <JymmmEMC> jlmjvm: cesium-137 clock
[14:45:10] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: preselecting mdi command text makes it easier to clear and enter a new command
[14:45:10] <JymmmEMC> jlmjvm: oh you mean lag?
[14:45:22] <jlmjvm> yes,lag
[14:45:27] <JymmmEMC> traceroute
[14:45:41] <jlmjvm> is that a program?
[14:46:09] <JymmmEMC> yes, open a shell and type: traceroute yahoo.com
[14:46:15] <jepler> "mtr" is the traceroute-like program on ubuntu
[14:47:11] <SWPadnos> I think there's a gui-fied set of tools under System -> Administration -> Network tools
[14:47:14] <SWPadnos> or somethinig like that
[14:48:02] <JymmmEMC> Goo Eeee?
[14:48:14] <JymmmEMC> what is this Goo eee you speak of?
[15:02:04] <jlmjvm> pretty cool
[15:11:20] <JymmmEMC> dumb question, when you do xfwd, is it like screen ?
[15:11:52] <JymmmEMC> you can pick up where you left off with someplace else?
[15:39:24] <jlmjvm_> http://pastebin.ca/699635
[15:40:25] <jlmjvm_> alex_joni:what do we need to add to my hal file to try the adaptive feed?
[15:42:34] <alex_joni> I can give you some hints, can't test it right now though
[15:42:42] <alex_joni> (so don't blame me if it's wrong :P)
[15:42:58] <jlmjvm_> i can test,im at my machine
[15:43:27] <jlmjvm_> no blame,its all in the name of science,lol
[15:43:29] <alex_joni> let me write something first
[15:43:42] <alex_joni> can you look around if there's a ferror pin?
[15:43:48] <jlmjvm_> i pasted my hal file
[15:43:56] <alex_joni> or maybe we need to calculate it from cmd-feedback
[15:48:44] <alex_joni> you only have the encoder on the Z axis?
[15:49:07] <jlmjvm_> yes,havent mounted my new x and y motors yet
[15:50:41] <alex_joni> I see
[15:59:02] <alex_joni> http://pastebin.ca/699645
[15:59:04] <alex_joni> something like that
[15:59:32] <jlmjvm_> k
[16:08:42] <skunkworks> so - 0 to .005 range for 100% to 0%
[16:09:32] <skunkworks> ?
[16:09:37] <jlmjvm_> http://pastebin.ca/699657
[16:09:43] <jlmjvm_> heres an error
[16:11:13] <jlmjvm_> HAL: ERROR: signal 'Zpos' not found
[16:12:13] <alex_joni> skunkworks: yup
[16:12:56] <alex_joni> jlmjvm_: make that Zpos-cmd
[16:13:10] <alex_joni> and change Zfb to Zpos-fb
[16:13:51] <jlmjvm_> k
[16:17:14] <jlmjvm_> the machine is working now
[16:18:34] <jlmjvm_> how can we test,give it a big accel and velocity that would normally stall the motor and see if it keeps it from stalling?
[16:20:37] <alex_joni> I'd first open a halcmd
[16:20:51] <jlmjvm_> in terminal?
[16:20:58] <alex_joni> and look at the 'adaptivef' signal
[16:21:04] <alex_joni> (you can do that from the AXIS menu)
[16:21:14] <jlmjvm_> k
[16:21:20] <alex_joni> next program some MDI line moving Z
[16:21:51] <alex_joni> and try to make it go slower by dragging at the axis (if it's not too big of an axis)
[16:22:17] <alex_joni> the feed reported by AXIS and the value in the halmeter should both change
[16:25:52] <jlmjvm_> the value for adaptivef in hal meter is changing
[16:26:35] <alex_joni> can you be more specific?
[16:27:00] <alex_joni> is it going down a bit? staying there? going back up?
[16:27:17] <alex_joni> did you program G<mumble> to enable adaptive feed-override?
[16:28:41] <jlmjvm_> no
[16:28:56] <jlmjvm_> is the m52 a modal command?
[16:30:38] <jlmjvm_> i think it is,got different numbers when its on
[16:38:39] <jlmjvm_> i just jacked it up to 250 ipm,this shoild show something
[16:38:49] <jlmjvm_> should
[16:46:47] <Unit41> has anyone ever made a cnc from timing belts off vehicles ?
[16:47:05] <Unit41> or fan belts
[16:50:45] <CIA-8> 03jepler 07TRUNK * 10emc2/bin/.cvsignore: add stepconf
[16:50:46] <CIA-8> 03jepler 07TRUNK * 10emc2/debian/ (control.in emc2.files.in): add stepconf
[16:50:45] <CIA-8> 03jepler 07TRUNK * 10emc2/docs/man/man9/.cvsignore: add stepconf
[16:50:46] <CIA-8> 03jepler 07TRUNK * 10emc2/docs/src/ (Master_User.lyx Submakefile index.tmpl): add stepconf
[16:50:47] <CIA-8> 03jepler 07TRUNK * 10emc2/docs/src/config/ (7 files): add stepconf
[16:50:48] <CIA-8> 03jepler 07TRUNK * 10emc2/share/.cvsignore: add stepconf
[16:50:50] <CIA-8> 03jepler 07TRUNK * 10emc2/src/Makefile: add stepconf
[16:50:51] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/ (Submakefile TODO stepconf.glade stepconf.gladep stepconf.py): add stepconf
[16:50:55] <CIA-8> 03jepler 07TRUNK * 10emc2/src/hal/components/steptest.comp: add stepconf
[16:57:04] <CIA-8> 03jepler 07TRUNK * 10emc2/debian/extras-Ubuntu-6.06/usr/share/applications/emc2-stepconf.desktop: application menu entry for stepconf
[16:59:13] <CIA-8> 03jepler 07TRUNK * 10emc2/src/hal/components/steptest.comp: markup fix
[17:00:11] <owhite> hey folks.
[17:00:41] <owhite> would anyone have a suggestion if I have two analog inputs that I want emc to read, what kind of board or adaptor I might buy?
[17:01:03] <owhite> both are on the 0-5volt range.
[17:01:18] <owhite> (oygen pressure and temp)
[17:01:46] <jepler> owhite: the motenc-lite has ADC inputs. http://www.vitalsystem.com/web/motion/motionLite.php
[17:02:09] <jepler> owhite: I don't know whether any other I/O cards do, and I don't have any direct experience with motenc
[17:03:51] <alex_joni> I think SWPadnos is working on a card for the m5i20 for analog inputs
[17:03:56] <alex_joni> jepler: yay stepconf
[17:03:57] <owhite> looking at the web site.
[17:05:38] <jepler> yeah, analog inputs on the mesa PCI cards are on the long list of "could do, with the right driver, firmware, and add-on card"
[17:06:16] <alex_joni> owhite: the STG has some analog inptus aswell
[17:07:59] <owhite> that motenc-lite is $550.
[17:08:04] <owhite> too rich for my blood.
[17:10:14] <alex_joni> do you need lots of speed on reading the stuff?
[17:10:36] <owhite> nope.
[17:11:30] <owhite> it certainly doesnt have to be happening at the same rate as most things like encoders etc. I just want to -display- temp and pressure.
[17:11:50] <SWPadnos> I think my analog card would be overkill for that :)
[17:12:07] <alex_joni> cheapest would be to look for some serial connected microcontrolelr with analog inputs
[17:12:23] <alex_joni> then some userspace program to read the stuff through the serial port
[17:12:49] <alex_joni> maybe there is some usb tingie that can be used with hal_input
[17:13:04] <owhite> alex_joni: and how how hard would it be to direct the output of the program to emc, or would I just make a completely separate program to display the readings?
[17:13:39] <alex_joni> owhite: not that hard, you can export a hal pin from that program
[17:13:57] <alex_joni> then use that hal pin to display it in PyVCP inside AXIS
[17:14:16] <owhite> * owhite is still a die hard TkEMC user. :-)
[17:14:32] <alex_joni> then I guess you need some program to display the values
[17:14:37] <owhite> wait wait. sorry.
[17:14:41] <alex_joni> pyvcp can work even outside of axis
[17:14:43] <owhite> I -do- use pyvcp.
[17:14:49] <alex_joni> right
[17:16:47] <owhite> is there an example of stand-alone programs that exports data to hal_input?
[17:16:57] <owhite> er, of -a- stand-alone program.
[17:17:14] <SWPadnos> it's the serial part that's hardest
[17:17:40] <SWPadnos> if you write it in python, all you need to do is import hal and make a pin
[17:17:55] <SWPadnos> it's a few lines added to the code that actually reads the value :)
[17:18:18] <alex_joni> owhite: there are lots of sources which export HAL stuff, but they are pretty big (not because of the HAL export)
[17:18:23] <jepler> 'pydoc hal' (terminal command) will show you a short example which creates a HAL pin from a python program
[17:18:35] <SWPadnos> if you have a developer checkout, look in src/hal./user_comps (I think)
[17:18:36] <jepler> in this case it just copies an input value to an output but you'd want to get the output value in some other way..
[17:19:17] <SWPadnos> hmmm. those may be more complex than you need - so I'd take a look at the docs jepler mentioned
[17:19:20] <jepler> there's a modest amount of boilerplate, followed by a loop that writes the output value frequently
[17:20:03] <owhite> right, so there are probably drivers for reading the serial or USB -- you guys expect I could make calls to them in python?
[17:20:23] <owhite> * owhite would normally approach it using C.
[17:20:34] <jepler> google "pyserial" (also should work for usb serial ports)
[17:20:34] <SWPadnos> you can do that too
[17:20:45] <jepler> you can use C but there is more boilerplate code to write
[17:20:57] <Unit41> http://cgi.ebay.com/Berman-Astrea-II-Remote-Vibrating-Panty-Fit-upto-Size10_W0QQitemZ110168975072QQihZ001QQcategoryZ63854QQrdZ1QQssPageNameZWD1VQQcmdZViewItem
[17:20:58] <SWPadnos> check out using the program "comp" to compile and install the C module
[17:21:13] <owhite> pyserial seems reasonable.
[17:21:25] <jepler> I don't think 2.1.x's "comp" works for userspace components, but 2.2's does.
[17:21:33] <alex_joni> Unit41: is that emc2 controlled?
[17:21:52] <Unit41> not stock
[17:22:01] <owhite> I observe there is a PyUSB too.
[17:22:05] <Unit41> its got its own controlls
[17:22:33] <SWPadnos> it may be easiest to hook something up to an analog joystick
[17:22:45] <SWPadnos> that should be readable with hal_input
[17:22:52] <alex_joni> yup, 2 axes too
[17:23:00] <owhite> I'm awaiting the flurry of "hey I got an analog joystick" remarks.
[17:23:16] <SWPadnos> nah - digital with force feedback here ;)
[17:23:20] <alex_joni> how about soundcard?
[17:23:41] <jepler> yes, if you can modify a linux-compatible analog joystick to read your analog value, it'll work right away with hal_input or hal_joystick
[17:25:06] <owhite> okay people. as always I really appreciate your help.
[17:25:12] <alex_joni> bye
[17:26:21] <alex_joni> http://dsplabs.cs.utt.ro/~juve/blog/index.cgi/photography/01189957474
[17:28:23] <jlmjvm_> alex_joni:it appears to be working in the z- direction
[17:29:06] <jlmjvm_> i set a high acceleration in the ini file that will make me get a ferror instantly
[17:29:53] <jlmjvm_> give it a m52p1 and g0z-1. and it works,try to go back to zero and get feror
[17:30:33] <jlmjvm_> if you dont have the m52p1 it will ferror when you try to go z-
[17:30:52] <alex_joni> ahh.. I see
[17:31:00] <alex_joni> I didn't use an abs()
[17:31:24] <alex_joni> if the move is in the other direction ,it will use an adaptive feed override of -xx
[17:31:28] <alex_joni> which will cause your error
[17:31:42] <jlmjvm_> so its actually working
[17:32:06] <alex_joni> add the following:
[17:32:11] <alex_joni> loadrt abs count=1
[17:32:20] <jlmjvm_> gotta be,or it wouldnt work in z-
[17:34:22] <jlmjvm_> still doing it
[17:34:57] <alex_joni> http://pastebin.ca/699760
[17:37:18] <CIA-8> 03alex_joni 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/stepconf.glade: typo
[17:37:59] <alex_joni> jepler: is stepconf using doublestep?
[17:38:45] <jepler> alex_joni: yes
[17:40:18] <alex_joni> any reason for the min base period to be double of the latency test result?
[17:40:31] <CIA-8> 03jepler 07TRUNK * 10emc2/configs/common/ (emc.var tool.tbl): files used by stepconf
[17:40:34] <alex_joni> err. not double
[17:40:55] <jepler> it's latency + step time + 5000
[17:40:59] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/stepconf.py: look in the right place for files
[17:42:36] <alex_joni> hmm, I get some errors when trying "Test this axis"
[17:42:48] <alex_joni> seems like it looks for installed modules instead of rip
[17:42:58] <alex_joni> halcmd: halcmd: HAL:2: ERROR: Can't find module 'steptest' in /usr/realtime-2.6.15-magma/modules/emc2
[17:43:00] <alex_joni> halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: HAL: ERROR: function 'parport.0.reset' not found
[17:43:12] <SWPadnos> did you . emc-environment?
[17:43:20] <SWPadnos> (silly question, I know)
[17:44:00] <alex_joni> hmm.. is it needed?
[17:44:10] <SWPadnos> probably, if you have RIP and installed
[17:44:29] <alex_joni> yup, makes a difference
[17:44:43] <alex_joni> I only get: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: halcmd: ha
[17:44:48] <SWPadnos> heh
[17:44:53] <SWPadnos> that seems wrong :)
[17:44:57] <jepler> that's what it does for me too
[17:44:59] <alex_joni> nope it's not :)
[17:45:02] <jepler> but with a motor connected I can spin it
[17:45:18] <alex_joni> great
[17:46:04] <jepler> I forget why, but passing the -s flag to halcmd made it not work right
[17:46:07] <alex_joni> next thing will be to put a link to stepconf somewhere in the configchooser
[17:46:25] <jlmjvm_> alex_joni:HAL: ERROR: signal 'Zpos' not found
[17:46:25] <jlmjvm_> HAL:26: link failed
[17:46:37] <alex_joni> jlmjvm_: Zpos-cmd and Zpos-fb
[17:46:53] <jlmjvm_> k
[17:49:43] <SWPadnos> strange. script mode shuoldn't have any effect
[17:50:24] <jlmjvm_> it comes on now
[17:50:55] <SWPadnos> I'd think the isatty(0) check would prevent the block of code from executing, even if scriptmode is 1 (halcmd_main.c, function get_input)
[17:56:06] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/stepconf.py: remove debugging statements
[17:56:53] <jlmjvm_> still stalling when you move z+ with m52p1 on,but will move in rapid both ways with no m52 with mdi command,but will get ferror if you jog manually
[17:58:01] <jepler> hm I wonder whether the HAL adaptive feed is applied to jogs or not
[17:59:08] <jlmjvm_> can the adaptive feed be working without the m52 command?
[17:59:25] <alex_joni> I think it stays on, once you commanded it
[17:59:31] <alex_joni> m54p0 to turn it off
[17:59:35] <alex_joni> 52
[18:00:15] <SWPadnos> you should be able to add M52P1 to the ini file "RS274NGC_STARTUP_CODE" (or similar - check the docs for the actual name and section)
[18:00:29] <jlmjvm_> k,its modal,but when you turn emc on its not active
[18:00:38] <SWPadnos> see above ;)
[18:01:28] <jlmjvm_> the machine will take a mdi rapid with no m52,but with m52 stalls moving z
[18:01:31] <SWPadnos> the danger is that someone could put M52P0 in their G-code, which that will turn it off, and cause that program and any that run after it to fail (until you catch the problem)
[18:02:10] <alex_joni> jlmjvm_: that sounds quite backwards
[18:02:12] <jlmjvm_> unless you restart emc
[18:02:31] <jlmjvm_> i know it does
[18:04:35] <jlmjvm_> can mdi z axis both dir,manual jog will give ferror after 50%,m52p1 will move z-,gives erroe moving z+
[18:04:45] <jlmjvm_> error
[18:05:40] <alex_joni> I bet my formula has an error for negative
[18:06:14] <jlmjvm_> i dont understand why its working with no m52p1
[18:07:09] <alex_joni> jlmjvm_: try this: http://pastebin.ca/699792
[18:07:42] <SWPadnos> alex_joni, is the f-error output from motion a pin or a param?
[18:07:53] <jlmjvm_> if you have accel turned up high enough to get ferror in manual jog,shouldnt it do the same thing in mdi with no m52p1?
[18:07:54] <alex_joni> dunno, didn't look it up
[18:07:56] <SWPadnos> heh
[18:07:58] <alex_joni> I think it's a param
[18:08:13] <SWPadnos> I think so too, but that would sure make this easier if it were a pin :)
[18:08:30] <alex_joni> jlmjvm_: it might use max_ferror for g0 vs. ferror for jogs
[18:08:46] <jlmjvm_> ahh
[18:09:06] <jlmjvm_> tah would make sense
[18:09:10] <jlmjvm_> that
[18:09:12] <alex_joni> I'm just guessing here
[18:09:27] <jlmjvm_> sounds good
[18:18:09] <jlmjvm_> works in mdi with g1,feed set to rapid rate,no m52p1
[18:19:22] <jlmjvm_> turn on m52p1 with same moves and it stalls moving z+
[18:23:59] <jlmjvm_> im gonna try these moves with my old hal file and see if it stalls with the high accel value
[18:28:23] <lerneaen_hydra> anyone still here?
[18:29:47] <lerneaen_hydra> I've been having some strange behavior with EMC recently; the latency test reports an ovl_max of 23µS, and I have a base period of 28.5µS, yet when starting EMC I *always* get a "unexpected realtime delay, check dmesg" error/warning window
[18:30:07] <lerneaen_hydra> dmesg reports quite a large clock count diff
[18:30:14] <alex_joni> how large?
[18:31:49] <lerneaen_hydra> I don't recall exactly but it was in the range of 2.0e8 for "ok" clock counts, and 2.3e8 for the anomoulus clock count
[18:34:05] <lerneaen_hydra> emc "seems" to run correctly if I ignore the error message
[18:34:16] <lerneaen_hydra> ie. I haven't noticed that the stepper stall
[18:35:59] <SWPadnos> that error is generated in the traj (slowest) thread. I wonder if it needs to wait some number of periods before checking tor overruns
[18:36:34] <SWPadnos> it seems that the error occurs at startup pretty often, but it's hard to know if it continues to occur because the messages get suppressed
[18:36:46] <SWPadnos> I think there's a counter though, so you can halmeter/halscope that
[18:37:13] <lerneaen_hydra> hmm, anything I can do to stop getting it at startup?
[18:37:34] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/ (stepconf.glade stepconf.py): allow stall test to be run on -, + or, +- sides of the initial point; get rid of halcmd: prompts
[18:38:55] <CIA-8> 03jepler 07TRUNK * 10emc2/src/hal/components/steptest.comp: allow stall test to be run on -, + or, +- sides of the initial point; get rid of halcmd: prompts
[18:41:11] <jlmjvm_> went back to a regular halfile and it does the same thing
[18:41:38] <jlmjvm_> does jog not use accel?
[18:46:09] <jlmjvm_> why do i get a feeling that this is prolly working and its operator error
[18:46:39] <Unit41> why am I still living in the stone age ?
[18:46:54] <Unit41> all these fucking cave people roaming around
[18:48:48] <Skullworks-PGAB> Unit 41 must have gone "clubbing"
[18:49:27] <skinnypuppy1334> So easy a caveman can do it???
[18:55:16] <fenn> there's a full 19" rack cabinet for $20 down the road, i'm hemming and hawing
[18:56:02] <alex_joni> good for you
[18:56:06] <alex_joni> full with what?
[18:56:10] <fenn> bah you guys all have milling machines, its no big deal
[18:56:23] <fenn> its empty, but it's 6' tall
[18:56:53] <alex_joni> nice
[18:57:01] <alex_joni> mine's a bit smaller
[18:57:06] <SWPadnos> I could use a short rack for my power supplies
[18:57:24] <SWPadnos> I have a "studio rack", but it's not too sturdy
[18:57:32] <SWPadnos> (and it doesn't fit under the table, which is a pain)
[18:57:39] <alex_joni> 8U ?
[18:57:54] <SWPadnos> I could use a 12-15U rack
[18:58:11] <alex_joni> mine's something like that.. but stuffed in the mean time
[18:58:28] <SWPadnos> the one I have has some front-facing spots, then an angled area that you'd use for audio knobs or patch panels
[19:11:05] <jlmjvm_> jepler:where do you turn off the program exceeds machine limits message
[19:13:57] <alex_joni> jlmjvm_: you need to increase limits
[19:14:48] <jlmjvm_> limits are correct,we did this a couple months ago
[19:15:44] <lerneaen_hydra> SWPadnos; you didn't know any way of fixing the RT delay problem?
[19:16:09] <SWPadnos> nothing short of changing the source code or possibly turning mesasges off completely
[19:16:15] <jlmjvm_> it thinks it will overtravel,but the z number it sees is when the g43 is turned on,so it never actually moves z+
[19:16:16] <SWPadnos> messages
[19:16:45] <jlmjvm_> works fine with message turned off
[19:16:46] <lerneaen_hydra> SWPadnos; oh. right
[19:16:52] <lerneaen_hydra> so is it a bug?
[19:17:05] <alex_joni> lerneaen_hydra: nope, a too restrictive safety measure maybe
[19:17:23] <SWPadnos> lerneaen_hydra, no, it's probably not a bug. just a possible issue with a startup glitch
[19:17:31] <jlmjvm_> mach3 used to do it,emailed art a few times and it works now
[19:18:09] <SWPadnos> I think there's a parameter that counts overruns - you should look at that with EMC running. it may go up to 5 or 10 at startup, but should never go up again
[19:18:16] <alex_joni> jlmjvm_: you can always get your money back with emc :P
[19:18:22] <SWPadnos> jlmjvm_, are you running a CVS version of EMC?
[19:18:29] <jlmjvm_> no
[19:18:39] <jlmjvm_> jlmjvm luvs emc
[19:18:49] <lerneaen_hydra> shouldn't the feature be modified so it doesn't report overruns in the beginning?
[19:18:50] <SWPadnos> I remember some discussion of G43 problems with preview, and I think they may be fixed in CVS (Which will be 2.2 sometime)
[19:18:58] <jlmjvm_> agreed
[19:19:01] <SWPadnos> lerneaen_hydra, that's the "modifying the source" option ;)
[19:19:50] <lerneaen_hydra> hmm, so I take it it will be in the 2.2.x version?
[19:20:00] <lerneaen_hydra> or am I the only person that's seen this effect?
[19:20:09] <SWPadnos> no, I didn't say anyone had modified the source for that
[19:20:26] <jlmjvm_> preview doesnt bother me,just wanna be able to cycle start without error msg
[19:20:27] <SWPadnos> I think Ive heard it before, but I haven't had a chance to look into it
[19:20:33] <lerneaen_hydra> hmm, ok
[19:20:44] <lerneaen_hydra> is there any way to suppress the messages completely?
[19:20:51] <lerneaen_hydra> without recompiling ;)
[19:20:52] <jlmjvm_> just dont remember where the axis folder was to turn it off
[19:20:57] <alex_joni> not without touching the code
[19:21:07] <alex_joni> jlmjvm_: don't think you can turn it off
[19:21:21] <alex_joni> jlmjvm_: but I feel it's a bug that it doesn't take G43 into account
[19:21:29] <jlmjvm_> either jepler or cradek told me how to do it 1 day
[19:21:29] <alex_joni> jlmjvm_: it should work as advertised :)
[19:21:41] <jlmjvm_> agreed
[19:22:15] <jlmjvm_> whats happening is the number isnt being modified by the offset value
[19:22:28] <alex_joni> jlmjvm_: right
[19:22:44] <skinnypuppy1334> I don't know if any of this would help, a link I still had open about resetting offsets. http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems
[19:23:11] <SWPadnos> there are complexities with offsets. the CVS version of EMC has the ability to do touch-off from within G-code (maybe possible with 2.1.7 as well)
[19:23:15] <jlmjvm_> ex:g0g43z.1
[19:23:31] <jlmjvm_> g0g43z.1h1
[19:23:42] <SWPadnos> there's no way for the preview generator (which is the same code that knows the program goes out of limits) to know what the final offset will be
[19:24:12] <jlmjvm_> i just wanna turn off the message
[19:24:25] <jlmjvm_> works great with it off
[19:25:01] <SWPadnos> I bet if you look for the message text in axis.py, you can see where to get rid of it :)
[19:25:44] <jlmjvm_> where is axis.py
[19:26:09] <alex_joni> http://dsplabs.cs.utt.ro/~juve/blog/index.cgi/photography/01189958841
[19:26:49] <SWPadnos> hmmm. actually, you may not be able to just modify axis.py - I think that gets compiled, not run as a script (could be wrongthere though)
[19:31:40] <jlmjvm_> i cant find axis.py
[19:32:03] <SWPadnos> don't bother looking. I'm not sure that method will help anyway
[19:32:13] <alex_joni> better wait for cradek or jepler
[19:32:17] <SWPadnos> yep
[19:32:19] <jlmjvm_> k
[19:32:43] <jlmjvm_> we did it awhile back,worked perfect
[19:33:01] <SWPadnos> I see where the function is (it's called run_warn), but I'm not sure (a) if there's a settable option somewhere or (b) if you need to compile once you change axis.py
[19:34:03] <jlmjvm_> didnt have to compile
[19:34:25] <jlmjvm_> just changed the file
[19:34:41] <roel1> Hi all
[19:34:44] <SWPadnos> well, then if you can find axis.py (somewhere under /usr I'd bet), then you can change the body of run_warn to just return 0
[19:42:20] <SWPadnos> jlmjvm_, axis is at /usr/bin/axis
[19:42:51] <SWPadnos> if you load that in an edotor and look for the word "exceeds", you'll see where the error is checked for
[19:42:57] <SWPadnos> or an editor
[19:43:29] <SWPadnos> if's just after the line "def task_run(*event):"
[19:44:32] <SWPadnos> you can delete the 15 lines starting with "warnings = []" and ending with "if r: return"
[19:44:36] <alex_joni> hi roel1
[19:44:51] <SWPadnos> I'd make a backup first ;)
[19:46:42] <jepler> there's not an option
[19:46:48] <jlmjvm_> only changed 1 thing that day,didnt delete anything,gonna go look
[19:48:15] <alex_joni> Tools->Options->Preview (uncheck [ ]Check program extents on loading)
[19:51:32] <jepler> there's Show Extents but that doesn't change whether the warning is issued
[19:52:48] <jlmjvm_> jepler:do you remember where you told me to turn the message off?
[19:53:02] <jlmjvm_> we did it 1 day awhile back
[19:53:22] <jepler> seems like it was to change 'if warnings:' to 'if 0:' but I don't remember the details
[19:53:43] <jepler> this time make a note :-P
[19:53:56] <SWPadnos> heh - that's the simple but inefficient method :)
[19:53:59] <jlmjvm_> you can bank on that
[19:54:35] <jlmjvm_> i reloaded emc on a new hard drive,saved my hal and ini files but forgot about this
[19:54:46] <jlmjvm_> being changed
[19:56:08] <jlmjvm_> i found the if warnings
[19:56:54] <SWPadnos> you can either delete those lines (a little harder to do now but prevents the checks from being run, which may save a little time on large programs), or change the if warnings to if 0
[19:58:52] <jlmjvm_> will try that
[20:04:13] <jlmjvm_> wont let me save
[20:04:20] <SWPadnos> use sudo
[20:04:30] <SWPadnos> you need to be root to edit that file
[20:11:59] <alex_joni> or use money
[20:14:40] <jlmjvm_> i give up
[20:14:58] <alex_joni> jlmjvm_: how so?
[20:16:11] <jlmjvm_> dont know how to change this file,dont know the terminal commands to be root and get there
[20:16:52] <jlmjvm_> i think ive done too much emc this week
[20:16:58] <alex_joni> sudo gedit /pat/to/file
[20:17:25] <alex_joni> I mean /path/to/file
[20:20:47] <alex_joni> SWPadnos: http://blog.medallia.com/2007/06/dyesight.html
[20:25:38] <jlmjvm_> thanks guys that did it
[20:26:48] <fenn> jlmjvm_: you arent going to be our lab rat?
[20:27:18] <alex_joni> fenn: no, he needs feeding for that
[20:27:49] <jlmjvm_> the fat rat
[20:27:55] <fenn> alex this is a wacky cool picture http://dsplabs.cs.utt.ro/~juve/blog/index.cgi-files/photography/01189958841/le_mans_&_castele 947.jpg
[20:27:58] <skunkworks> jlmjvm_:z is going to go down easier than up anyways..
[20:28:18] <jlmjvm_> what do you mean?
[20:28:18] <alex_joni> fenn: thanks :)
[20:28:34] <alex_joni> mr. newton pulling at things
[20:28:42] <alex_joni> gravity doing it's business
[20:29:02] <jlmjvm_> you talking about the adaptive feed
[20:29:08] <alex_joni> fenn: would have like more contrast on the gargoyle
[20:29:40] <skunkworks> you have to remember z- will act different from z+ from the get go..
[20:31:25] <jlmjvm_> r u talking about the adaptive feed stalling when moving z+?
[20:32:20] <alex_joni> jlmjvm_: yeah, I guess so
[20:32:38] <alex_joni> anyone playing tetris around here?
[20:32:42] <skunkworks> sorry - yes. I think there is still an issue :)
[20:33:39] <jlmjvm_> i dont think that was the problem,its a quill machine,and was moving in the air at 60ipm
[20:34:15] <jlmjvm_> im running 100 ipm now with ferror and no adaptive feed,works great
[20:34:27] <alex_joni> jlmjvm_: better leave it at that :)
[20:34:36] <alex_joni> http://www.mediablog.cz/story.php?id=4042 <- this is insane
[20:34:41] <SWPadnos> jlmjvm_, is that with the encoder feedback?
[20:34:46] <jlmjvm_> yes
[20:34:49] <SWPadnos> cool
[20:34:59] <SWPadnos> what's your ferror setting?
[20:35:08] <jlmjvm_> had it set at 240 ipm earlier
[20:35:16] <jlmjvm_> .050 and .01
[20:35:25] <SWPadnos> oh, that's fairly wide
[20:35:41] <jlmjvm_> it was the default settings
[20:36:01] <SWPadnos> still wide :)
[20:36:04] <tomp> alex_joni: thanks for the dyesight link, i've been following ftir and hadnt heard of that.
[20:36:27] <alex_joni> tomp: not sure how real it is ;)
[20:36:35] <alex_joni> but it sounds like a good idea
[20:36:50] <tomp> yes, getting the apple mung code now
[20:36:56] <SWPadnos> the description of the algorithm seems plausible, but I'm sure it isn't as easy as it looks to get the software right :)
[20:37:30] <jlmjvm_> would be cool if the ecoders could read at 240 ipm
[20:37:48] <SWPadnos> also, it relies on brightly backlit hands, not the refraction changes the other multi-touch screens sense
[20:39:25] <tomp> a shiny colored computer interface that only works on bright sunny days, are we all hippys yet:)
[20:39:55] <SWPadnos> and only if you're not wearing blue gloves ;)
[20:40:04] <SWPadnos> (or whatever the dye color is)
[20:42:48] <jlmjvm_> SWPadnos:what would you shoot for on the ferror setting
[20:43:09] <SWPadnos> I'd be doing metal, and I'd want ferror below 0.005 I think
[20:43:13] <SWPadnos> maybe more for rapids
[20:43:31] <SWPadnos> even 0.005 is crap though
[20:43:58] <alex_joni> hahaha "This is so gorgeous. Is there an organ I can donate to get to look at the code?"
[20:43:59] <jlmjvm_> i basically want it to catch a stall
[20:44:51] <jlmjvm_> and it appears to be doing that
[20:45:03] <SWPadnos> then it's good enough :)
[20:45:42] <jlmjvm_> agreed
[20:46:27] <jlmjvm_> nothing to do with accuracy,just stopping a runaway or stall
[20:46:40] <jlmjvm_> which could save a part
[20:47:08] <jlmjvm_> usually would be off way more than the ferror
[20:47:34] <SWPadnos> as long as it's doing what you want, it's good
[20:47:43] <SWPadnos> time for me to run some errands. bbl
[20:47:52] <jlmjvm_> later
[20:48:54] <alex_joni> tomp: http://acg.media.mit.edu/people/tom/projects/liquid/
[20:49:37] <skinnypuppy1334> http://youtube.com/watch?v=vCHPo3EA7oE
[20:51:48] <tomp> alex_joni: the paper from an another earlier experimenter is interesting, with feedback, and a pic x/y controller/reader http://acg.media.mit.edu/projects/thesis/TWthesis-lossy.pdf
[20:51:57] <tomp> alex_joni: and thanks
[20:52:34] <tomp> oh, same paper/author :)
[20:55:35] <alex_joni> skinnypuppy1334: http://youtube.com/watch?v=f2XQ97XHjVw
[20:55:39] <alex_joni> tomp: yup
[20:56:47] <jlmjvm_> alex_joni:is working at 100 ipm with ferror set at .010 and .002,if i use .001 it errors
[20:58:15] <alex_joni> jlmjvm_: nice
[21:01:09] <jepler> 100 inch/min ~= .00166 in/ms (servo period)
[21:01:32] <jlmjvm_> is that good
[21:02:20] <jlmjvm_> my base period is 10000
[21:03:06] <jlmjvm_> havent really tried to go any lower ,have a better cpu now
[21:05:35] <skunkworks> as jepler mentioned - if you want to try to tighten up the error a bit more - you would need to lower the servo period a bit more. Another experiment.. :)
[21:06:11] <skunkworks> at some point you're going to run into the accuracy of the steppers.
[21:06:24] <anonimasu> hm
[21:06:29] <anonimasu> does anyone have cutting data for pcb?
[21:06:52] <anonimasu> 150mm/min at 2000rpm cuts nicely,.
[21:07:02] <anonimasu> with a 4 flute 2mm endmill
[21:09:14] <jlmjvm_> my servo period and traj period are both 1000000
[21:09:26] <jlmjvm_> never changed them
[21:10:32] <tomp> poly crystalline boride or printed circuit board?
[21:11:28] <anonimasu> poly cristalline ofcourse :p
[21:11:32] <anonimasu> no, pcb
[21:11:38] <anonimasu> circuit boards ;)
[21:11:40] <JymmmEMC> tomp: (Poor cheating Bastard)
[21:11:54] <tomp> then dont submerge it and use high voltage :)
[21:12:07] <JymmmEMC> tomp: that's what she said
[21:12:39] <anonimasu> lol
[21:12:41] <alex_joni> http://www.youtube.com/watch?v=YocnQ0NMTUA&NR=1
[21:14:00] <anonimasu> what's up?
[21:15:11] <anonimasu> so, nobody has any idea?
[21:16:53] <tomp> theres people who do cut pc boards, sorry not me. do you mean to separate or to cut the copper coat?
[21:17:48] <anonimasu> separate
[21:17:56] <JymmmEMC> score and snap?
[21:18:04] <fenn> anonimasu: iirc they use .0005"-.001" depth of cut, with a 1 flute 90 degree conical cutter
[21:18:31] <fenn> er.. s/depth of cut/feed per rev
[21:19:01] <fenn> and a roughing 1/8" endmill for cutting the board edges
[21:19:17] <fenn> double cut preferably
[21:19:55] <fenn> double cut bur i mean
[21:20:01] <fenn> past my bedtime
[21:20:15] <anonimasu> feed per tooth?
[21:20:21] <fenn> ya
[21:20:21] <skinnypuppy1334> non-newtonian fluid pretty cool alex
[21:20:31] <anonimasu> fenn: I tool a mm at a time.. :)
[21:20:57] <anonimasu> JymmmEMC: yeah
[21:21:08] <anonimasu> JymmmEMC: well machine a shape out of it..
[21:21:40] <JymmmEMC> anonimasu: other than a rectangle?
[21:22:04] <anonimasu> JymmmEMC:yeah
[21:22:27] <anonimasu> rectangle with tabs for alignment..
[21:22:28] <JymmmEMC> cut the shape and use a couple of bridges maybe
[21:22:34] <alex_joni> JymmmEMC: you seemed to know a bit about medication
[21:22:45] <anonimasu> JymmmEMC: um. that's not remotely the problem(not that there is one)
[21:22:51] <anonimasu> what feeds speeds do you guys use?
[21:22:52] <JymmmEMC> anonimasu: ah
[21:23:01] <JymmmEMC> alex_joni: pick one
[21:23:02] <anonimasu> JymmmEMC: I can cut it like a ~^~ if I like :)
[21:23:09] <alex_joni> JymmmEMC: ever heared of Ma Huang
[21:23:55] <alex_joni> ephedrine I think
[21:24:11] <JymmmEMC> alex_joni: Yeah,
[21:24:24] <JymmmEMC> alex_joni: killed a few pple here iirc
[21:24:25] <anonimasu> http://en.wikipedia.org/wiki/Ephedrine
[21:24:45] <JymmmEMC> was sold as a weight loss product
[21:24:50] <tomp> i have cut disks out of frp, and it's hell on hispeed tool steel ( 2 fluke 1/8 dia dunno what feed ) , had some luck with carbide 1 fluke
[21:25:03] <JymmmEMC> alex_joni: what about it?
[21:25:46] <JymmmEMC> alex_joni: http://www.mayoclinic.com/health/ephedra/NS_patient-ephedra
[21:26:09] <anonimasu> tomp: it seemed like it would cut well with my tooling
[21:26:50] <skunkworks> I think corn starch and water makes a non-newton liquid. I remember playing with it whan I was a kid. So is silly puddy
[21:27:15] <tomp> i ended up using a trammeling tool with carbide tooling, 60 opcs 2" dia disks ate the bit up ( smoked thru last disk :(
[21:28:09] <tomp> i took the fuzzy off the edge with a file, the frp ate a groove in the nicholson file
[21:28:24] <anonimasu> I have 2 pices :)
[21:28:33] <tomp> np
[21:28:52] <anonimasu> and new tooling
[21:30:39] <anonimasu> though feeds/speeds is what I would like more specific ones
[21:31:41] <fenn> silly putty is elastic, corn starch is not
[21:33:33] <fenn> i found 10 x 50lb bags of corn starch in a dumpster once but i didnt know what to do with it
[21:33:54] <alex_joni> fenn: toss it in a swimming pool nearby
[21:34:03] <fenn> well, now i know
[21:34:08] <alex_joni> haha
[21:34:10] <alex_joni> good night all :P
[21:34:12] <anonimasu> that would be funny
[21:34:25] <alex_joni> anonimasu: not if someone tries a dive head first
[21:34:36] <fenn> and then drowns
[21:34:45] <alex_joni> most likely
[21:34:50] <fenn> answer: put it in your annoying neighbor's pool
[21:34:50] <anonimasu> :/
[21:35:03] <alex_joni> * alex_joni is off to bed
[21:35:09] <anonimasu> night
[21:35:31] <anonimasu> fenn: drown?
[21:35:33] <anonimasu> http://www.youtube.com/watch?v=f2XQ97XHjVw
[21:35:48] <alex_joni> tomp: http://www.whitenoiseaudio.com/touchlib/
[21:36:00] <fenn> that pool is only a foot deep
[21:36:20] <alex_joni> it's deeper, they are standing up inside
[21:36:32] <fenn> i cant believe nobody's selling those ftir screens yet
[21:36:35] <alex_joni> but still if you jump head first, I suspect you don't have time to turn around
[21:37:55] <fenn> we should ask mythbusters to try it out for us :)
[21:39:06] <fenn> does anyone know why your hands smell weird after touching keys?
[21:39:24] <anonimasu> keys
[21:39:33] <fenn> like for opening locks
[21:40:03] <fenn> or coins sometimes
[21:41:09] <anonimasu> fenn: you usually keep your keys in your pocket near your genitals.. it's not surprising if they do smell.
[21:41:27] <fenn> no, its a metallic smell
[21:41:59] <anonimasu> balls of steel..
[21:42:07] <fenn> "When touching objects made of iron, perspiration from skin causes the iron atoms to gain two electrons. The doubly negative iron atoms react with oil in skin, causing them to decompose, forming 1-octen-2-one."
[21:42:17] <anonimasu> fenn: im kidding..
[21:42:21] <fenn> but, this is a brass key..
[21:42:53] <anonimasu> night
[21:43:09] <fenn> me too
[21:45:02] <jmkasunich_> jmkasunich_ is now known as jmkasunich
[21:45:51] <tomp> alex_joni: thanks again, making me interested in doing the iBar http://www.youtube.com/watch?v=iaKehq6qsdY ( i figger i'd get free drinks during the install ;)
[21:47:16] <alex_joni> tomp: this might help http://www.whitenoiseaudio.com/blog/
[21:47:31] <tomp> fenn: its the 'touch'. during the year some toolmakers get 'the touch', they get slightly acidic and couldnt touch steel without marking it... old timers took thier vacations when they got 'the touch'
[21:51:48] <tomp> alex_joni: wow, i wonder if the open-source flash-like stuff will work with it , again again thanks
[21:51:59] <tomp> walkies bbl
[21:59:17] <jlmjvm_> alex_joni: is the feed hold in emca controlled stop?
[21:59:29] <jlmjvm_> emc a
[21:59:41] <alex_joni> jlmjvm_: what do you mean?
[22:00:04] <jlmjvm_> does it lose it place
[22:00:28] <alex_joni> nope
[22:00:31] <jlmjvm_> and does it use the accel values
[22:00:36] <alex_joni> yes
[22:01:01] <skunkworks> * skunkworks thinks jlmjvm_ is a emc2 convert..
[22:01:12] <jlmjvm_> excellent,gonna have to put a feed hold button on the trade school machine then
[22:01:34] <jlmjvm_> skunkworks is correct
[22:01:36] <alex_joni> oh, sure you could
[22:02:14] <jlmjvm_> prolly need a cycle start and a feed hold
[22:02:16] <jlmjvm_> cool
[22:02:24] <jlmjvm_> and estop of course
[22:04:05] <SWPadnos> be careful with feedhold
[22:04:16] <SWPadnos> I believe it's a bit input to the motion controller
[22:04:22] <jlmjvm_> how so?
[22:04:53] <SWPadnos> you should set up either classic ladder or some HAL components so you can use a momentary contact pushbutton (and probably an indicator lamp) to control it
[22:05:22] <SWPadnos> you don't wnt someone wondering why the machine isn't moving, and while they're poking around have someone else hit the feedhold button and have the machine take off
[22:06:02] <jlmjvm_> true
[22:06:18] <jlmjvm_> or have a pyvcp led
[22:06:19] <SWPadnos> that scenario would be easy to have if you just stick a toggle switch on the panel
[22:06:22] <skunkworks> normally feed hold is a momantary push button. where feed hold is only on when your finger is on it.
[22:06:28] <skunkworks> from what I see
[22:06:36] <SWPadnos> no, you should have an indicator and a momentary contact at any point that could control FH
[22:07:28] <SWPadnos> the temp thing is fine too
[22:07:31] <jlmjvm_> all the machines i have run if you pressed feed hold 1 time it stopped till you hit cycle start
[22:07:50] <skunkworks> hmm - I have used too old of machines then :)
[22:08:04] <SWPadnos> then you need extra logic to get cycle start to turn off feed hold if it's on, or to do a cycle start if it's off
[22:08:16] <SWPadnos> (again, classic ladder should work for that)
[22:08:47] <jlmjvm_> i will never be able to do a classic ladder setup by next week
[22:09:01] <SWPadnos> heh - then I'd leave the feedhold out of the equation
[22:09:24] <jlmjvm_> just show em where the keyboard is
[22:10:37] <jlmjvm_> my stepper setup is running good,gotta get a usc+gecko interface+spindle dac from jon going next week
[22:11:01] <jlmjvm_> would never make it to classic ladder
[22:12:47] <jlmjvm_> i thought you could have 2 momentary button linkeded to 2 input pins and do a pause and resume
[22:13:01] <jlmjvm_> linked
[22:13:39] <jlmjvm_> wouldnt that work
[22:13:46] <SWPadnos> could be, if you use halui
[22:13:49] <SWPadnos> bu tI'm not sure
[22:14:10] <jlmjvm_> those functions are already there
[22:14:34] <SWPadnos> I don't know if there's a "logic input" for cycle start
[22:14:47] <SWPadnos> that's usually done with an NML message from the UI to the motion controller
[22:14:57] <jlmjvm_> run
[22:14:59] <SWPadnos> and when you want to interface buttons to NML, you use HALUI
[22:17:11] <jlmjvm_> cool,this will keyboard setup
[22:17:34] <jlmjvm_> get to that stuff later
[22:20:30] <JymmmEMC> SWPadnos: HALUI.... Sounds like an Hawaiian dish
[22:20:56] <SWPadnos> that would be something like haiuluiluilui
[22:21:52] <SWPadnos> here's the Hawaiian state fish: Humuhumukununukuapua'a
[22:22:01] <SWPadnos> n, I never pronounced it correctly
[22:22:02] <SWPadnos> no
[22:22:21] <JymmmEMC> SWPadnos: You sure that's not the name of Gawd?!
[22:22:29] <SWPadnos> yes, quite
[22:22:32] <SWPadnos> I fed some
[22:23:46] <SWPadnos> in fact, I may have gotten some photos of them while snorkeling
[22:27:12] <JymmmEMC> uou fed Gawd?! Cool
[22:27:23] <JymmmEMC> you
[22:27:45] <tomp> hallui loo-eye oh baby we gotta go
[22:28:03] <tomp> kingsmen 1950-ish
[22:28:41] <JymmmEMC> Whoa... I didn't even consider all the bible thumpers and their fish bumper stickers
[22:30:14] <tomp> http://www.youtube.com/watch?v=_FhwTZPSe0c