#emc | Logs for 2007-10-18

[00:33:18] <jlmjvm> does anyone have an example of how to use the lube signal?
[00:34:20] <cradek> it's to run a way lube pump etc
[00:34:35] <cradek> it actually just comes on whenever amp-enable does, so it's not that useful
[00:35:11] <jlmjvm> it doesnt give an alarm?
[00:36:00] <cradek> sorry what kind of alarm?
[00:36:07] <jlmjvm> low lube
[00:38:23] <cradek> the lube signal is an output from emc. sensing low lube and doing something smart would be a separate issue
[00:38:32] <cradek> but, you can do whatever you want in HAL.
[00:38:58] <cradek> my bridgeport allows the current run to finish on low lube, but then won't let you turn the spindle on again - that seems like a good compromise
[00:39:04] <cradek> or, you could estop if you want
[00:40:38] <jlmjvm> I thought emc was already setup to do something with the lube signal
[00:41:07] <cradek> if so, I'm not sure, sorry
[00:41:42] <cradek> there may be a lube level input with a hardcoded behavior on iocontrol.0
[00:41:48] <jlmjvm> thanks for the info,no need to try and figure out something that isnt there
[00:42:04] <jlmjvm> really
[00:42:39] <jlmjvm> didnt emc1 use a lube sensor
[00:42:47] <cradek> but I'd do it in HAL so I can make it do what I want
[00:43:05] <cradek> if emc1 had a certian behavior that may still be part of iocontrol
[00:43:55] <jlmjvm> i remember seeing a lube sensor polarity at the bottom of the .ini file
[01:26:41] <SkinnyPuppy34> jlmjvm: you have g203's right?
[01:28:39] <jlmjvm> skinnypuppy34:yes
[01:29:16] <SkinnyPuppy34> What did you do for supplying ground to the common input?
[01:31:28] <jlmjvm> pulled it from pin 24 on my breakout board
[01:31:43] <jlmjvm> straight from the parport
[01:32:16] <SkinnyPuppy34> Is that a standard ground or did you have to drive it low in hal?
[01:32:51] <jlmjvm> didnt have to do anything in hal
[01:33:57] <jlmjvm> i have a breakout board that lets me use all 25 pins
[01:34:21] <jlmjvm> and i think 18 thru 25 are ground
[01:34:43] <jlmjvm> what kinda breakout board you got?
[01:35:07] <SkinnyPuppy34> I've got one that does 25, its from winford
[01:35:17] <jlmjvm> thats what i have
[01:35:33] <SkinnyPuppy34> I'm checking voltages on 18~25..
[01:35:44] <jlmjvm> all my drives are tied to pin 24 for common ground
[01:37:04] <SkinnyPuppy34> Pin 24 isn't hooked to anything now but is .007 v above the computer case, I need to round up that paralell port doc.
[01:37:31] <SkinnyPuppy34> Keep getting distracted and got a guy stopping in later with a driveshaft to put on the lathe....
[01:37:58] <jlmjvm> are your drives not working?
[01:39:02] <SkinnyPuppy34> I'm just now wiring, got my motors and xfmr last week
[01:39:08] <jlmjvm> cool
[01:39:26] <jlmjvm> what kinda xfmr did you get
[01:39:36] <SkinnyPuppy34> Definately glad just double checking ...
[01:40:31] <SkinnyPuppy34> KL- 7220, from http://www.kelinginc.net/SwitchingPowerSupply.html
[01:42:18] <jlmjvm> thats a hoss
[01:43:13] <SkinnyPuppy34> Yeah bought a little extra
[01:43:46] <jlmjvm> i did a servo mill that moves 250 ipm rapid with a 1000 watt supply
[01:44:30] <CIA-8> 03jepler 07TRUNK * 10emc2/docs/man/man1/ (comp.1 .cvsignore): manpage for the comp command
[01:46:01] <jlmjvm> those kinda look like the ps from antek
[01:47:55] <SkinnyPuppy34> It seems to be a good unit, the torroid is about 7" by 3" tall just eyeballin it
[01:48:27] <jlmjvm> should be a good 1
[01:52:20] <SkinnyPuppy34> How is your pico stepper card doing for you?
[02:04:09] <SkinnyPuppy34> Schweet !! geckos green led and steppers running.
[02:34:35] <jlmjvm> the usc is doing good
[02:35:21] <SkinnyPuppy34> Here's a few pics
[02:35:22] <SkinnyPuppy34> http://imagebin.org/11111
[02:35:32] <SkinnyPuppy34> http://imagebin.org/11112
[02:35:39] <SkinnyPuppy34> http://imagebin.org/11113
[02:35:45] <SkinnyPuppy34> http://imagebin.org/11114
[02:36:43] <SkinnyPuppy34> Good to hear the usc is working I got dual shafts in hopes of finding some encoders.
[02:37:21] <jlmjvm> nice looking unit
[02:37:45] <fenn> i wonder why we're still in the dark ages with screw terminals on everything
[02:38:01] <fenn> even light switches
[02:38:12] <SWPadnos> because screw terminals are more reliable for power transfer than microwave waveguides?
[02:38:19] <SWPadnos> or other high-tech doodads
[02:38:26] <Jymmm> or telekenisis
[02:38:40] <fenn> bah
[02:39:15] <SWPadnos> I can't believe we're still in the stone age with browbras and everything
[02:39:32] <SWPadnos> you'd think someone could have figured out a better way of prying things than with a lever, in the last 50000 years
[02:39:48] <fenn> * fenn straps on his browbra
[02:40:00] <fenn> keeps my glasses from getting all greasy
[02:40:01] <SWPadnos> ouch! :)
[02:40:06] <Jymmm> * Jymmm walks into 7-eleven.... Let me have a slurpy, lottyo ti9cket, and that 256bit 4PHz cpu
[02:40:13] <SWPadnos> and I even saw that and corrected it once
[02:40:14] <Jymmm> That'll be $34.95
[02:40:22] <SkinnyPuppy34> Anyone else read recently about the nut stealing wire? all they found was his hand and bolt cutters...cnn reuters somewhere...
[02:40:36] <SWPadnos> not that one, but it's a big problem in a lot of places
[02:40:47] <Jymmm> and copper pipe in new construction
[02:40:55] <SWPadnos> they have to put up fences so people don't steal the feed wires for electric trains
[02:41:13] <fenn> they should put up fences anyway so people dont go poking at 440+V
[02:41:26] <SWPadnos> they have to put up bigger fences ;)
[02:41:46] <fenn> why dont they just use aluminum wire
[02:42:03] <SkinnyPuppy34> SOme jackass climbing a pole...
[03:44:31] <CIA-8> 03petev 07TRUNK * 10emc2/configs/boss/boss.hal: -Fixed problem with pid.x.maxoutput setting.
[07:32:00] <Rabbitbunny> aluminum wire isn't as good. copper wire is better, hence the use of copper.
[07:33:31] <Rabbitbunny> 02:32:31 up 5 days, 22:56, 3 users, load average: 0.49, 0.50, 0.45
[08:51:33] <Jymmm> alex_joni:
[08:56:08] <alex_joni> Jymmm: ?
[08:59:08] <Jymmm> alex_joni: question...
[08:59:21] <alex_joni> waiting for it..
[08:59:32] <Jymmm> alex_joni: when you have a subnet say /24 how do you add that to ubuntu?
[08:59:43] <Jymmm> not using aliases I wouldn't think
[08:59:59] <alex_joni> I don't follow..
[09:00:06] <alex_joni> can you be a bit more specific?
[09:00:22] <Jymmm> alex_joni: if you have 200 ip's, how do you add them to ubuntu
[09:01:27] <alex_joni> 200 ip's on one machine?
[09:01:32] <Jymmm> yes
[09:01:40] <alex_joni> why would you have 200 ip's?
[09:01:50] <Jymmm> does it matter?
[09:01:52] <alex_joni> but I would think aliases..
[09:01:54] <alex_joni> curiosity..
[09:02:15] <Jymmm> you mean eth0:1, eth 0:2......eth0:200 ?!?!?!?!
[09:04:00] <alex_joni> yes
[09:04:19] <Jymmm> hmmm, that doens't seem right.
[09:05:19] <alex_joni> you have one ip/interface
[09:05:28] <alex_joni> at least as far as I know..
[09:10:55] <alex_joni> you can have a small script to generate the interfaces file
[09:14:06] <alex_joni> but I'd still like to know why you need 200 ip's :D
[09:14:53] <alex_joni> (you could maybe do it at iptables level, but it depends on the routing scenario)
[09:15:11] <alex_joni> say you have some client PC's in 192.168.0.x, and the 200 ip's in 192.168.1.x
[09:15:37] <alex_joni> then the router from 0.x to 1.x could also have the 200 ip's
[09:16:00] <alex_joni> it would just do some DNAT to his IP
[09:21:02] <Jymmm> no nat at all
[09:28:40] <alex_joni> Jymmm: so any of the other 55 machines are in the same subnet, and need to contact this machien directly?
[09:30:55] <alex_joni> you could also have a DHCP server, and dhcp client on the machien with the need for the 200 ips
[09:31:11] <alex_joni> and *maybe* the DHCP can be configured to return a lot of addresses
[09:33:26] <alex_joni> but you still need to define them as aliases
[09:33:30] <alex_joni> http://ubuntuforums.org/archive/index.php/t-483375.html
[09:44:30] <alex_joni> then I guess you need to write them down
[09:44:51] <alex_joni> is this for shared hosting?
[09:45:02] <Jymmm> YEP
[09:45:28] <Jymmm> less caplock
[09:46:14] <alex_joni> I can't think of another way
[09:46:21] <alex_joni> maybe #networking might be helpfull?
[09:46:29] <alex_joni> (this is not really ubuntu specific..)
[12:00:37] <JanVanGilsen> hi
[12:02:02] <alex_joni> hi
[12:03:15] <JanVanGilsen> i was hoping that somebody could help me with my following errors :)
[12:07:10] <JanVanGilsen> I actually don't understand why I get them, because i didn't change much in the .ini file
[12:09:25] <JanVanGilsen> I only removed the Z-axis, I did'nt change the base_period, max_velocity and max_accelleration parameters
[12:09:36] <jlmjvm> alex_joni:good morning
[12:26:19] <jlmjvm> JanVanGilsen:is this the first time you have used emc?
[12:32:11] <fenn> JanVanGilsen: are you getting any 'unexpected realtime delay' messages when you type 'dmesg'?
[12:32:16] <JanVanGilsen> no, i have desktop mill that works with emc
[12:34:05] <jlmjvm> k,was wondering if your velocity was set too high
[12:36:14] <JanVanGilsen> i did try to set velocity verry low (1 mm/s) but that didn't help...
[12:39:28] <jlmjvm> what kind of motors and drives do you have?
[12:40:42] <JanVanGilsen> on the thing i'm building now, im using some old stepper motors from a floppy drive
[12:41:30] <JanVanGilsen> http://www.motors.wrobots.com/MSJE200A53.php
[12:44:14] <JanVanGilsen> And im driving them with the parallel port, like jepler does with his etch-a-sketch
[12:54:56] <JanVanGilsen> fenn: no, i don't get realtime delay messages in dmesg
[12:56:54] <skunkworks> could you pastbin.ca your ini file?
[12:59:46] <jlmjvm> do you have a motor driver
[13:00:44] <JanVanGilsen> i use darlington transistors :-)
[13:04:25] <alex_joni> JanVanGilsen: you said you are getting following errors?
[13:04:41] <alex_joni> that is probably *because* you removed the Z axis form the ini
[13:05:00] <alex_joni> (I mean if you removed the [AXIS_2] section)
[13:06:34] <JanVanGilsen> yes, i did remove the [AXIS_2] section, but i don't understand why this would result in following errors
[13:11:12] <JanVanGilsen> i'll try to copy/paste it back, but i remember having ferrors even with a [AXIS_2]
[13:11:18] <alex_joni> JanVanGilsen: because in the hal file you have things defined for the 3rd axis
[13:12:39] <JanVanGilsen> i removed those
[13:18:54] <JanVanGilsen> http://pastebin.ca/741068
[13:19:57] <JanVanGilsen> http://pastebin.ca/741070
[13:20:58] <cradek> what emc version is this?
[13:23:14] <alex_joni> sounds like CVS from about 2 weeks ago
[13:23:31] <JanVanGilsen> yes =)
[13:23:33] <cradek> FE on joint 2?
[13:24:05] <JanVanGilsen> no joint 0 and joint 1
[13:24:23] <cradek> when you do what in particular?
[13:24:57] <JanVanGilsen> G0 or G1 moves... (didn't try G2, G3)
[13:25:03] <cradek> oops brb
[13:25:09] <JanVanGilsen> I don't have them when is jogg
[13:25:17] <alex_joni> did you try different feedrates for G1>
[13:25:18] <alex_joni> ?
[13:25:27] <alex_joni> G1 x10 F10
[13:25:30] <alex_joni> G1 x10 F20
[13:25:32] <alex_joni> ..?
[13:25:46] <JanVanGilsen> yes, from 0.00000000000000000001 to 200
[13:25:54] <alex_joni> and you get ferrors on all?
[13:26:12] <JanVanGilsen> yes
[13:26:52] <alex_joni> you can probably decrease BASE_PERIOD
[13:27:08] <alex_joni> right now it's at 50000 (which is quite conservative)..
[13:27:13] <alex_joni> what kind of PC do you have?
[13:28:04] <JanVanGilsen> atm i'm usig a 850mhz duron so i don't think i should lower the base_period
[13:32:07] <alex_joni> I think you can at least go to 25000
[13:32:20] <alex_joni> 50000 should work on 2-300 MHz
[13:34:19] <JanVanGilsen> At 25000 I get "unexpected realtime delay on task 1"
[13:34:38] <alex_joni> ok, then increase it slightly until that goes away
[13:40:25] <alex_joni> did you really want step type 11?
[13:40:36] <alex_joni> from your pinout I think you really want step type 5
[13:41:24] <JanVanGilsen> isn't step type 6 better then 5?, it has more torque..
[13:42:08] <alex_joni> well, one of those anyways
[13:42:37] <alex_joni> maybe even 9
[13:42:41] <JanVanGilsen> ok, i dit change the base_period to 32000 (31000 gives realtime delay errors)
[13:42:55] <alex_joni> ok, now check for g1 ferrors
[13:45:37] <JanVanGilsen> even if I command "G0 F0.0000000000001 X1" I get "joint 0 following error"
[13:46:55] <alex_joni> and jogging works?
[13:47:02] <JanVanGilsen> yes
[13:47:06] <alex_joni> G0 doesn't use F-codes
[13:47:09] <alex_joni> only G1 does
[13:47:28] <alex_joni> G0 goes at max speed
[13:47:47] <JanVanGilsen> sorry, typo, i did comand G1 F0.0000000000001 X1
[13:48:18] <alex_joni> when you try to move Y, do you get joint 1 ferror?
[13:48:41] <cradek> try putting back the [AXIS_2] entries but leaving all the HAL unhooked
[13:49:07] <cradek> let's troubleshoot
[13:49:15] <alex_joni> JanVanGilsen: I assume you run it form a terminal?
[13:49:27] <alex_joni> maybe there are some warning messages there?
[13:49:53] <JanVanGilsen> no i'm noot running it from a terminal ..
[13:49:57] <JanVanGilsen> *not
[13:50:59] <JanVanGilsen> I'm quite new to linux, i stared using it about a month ago when i did a retrofit of a desktop mill ...
[13:51:17] <alex_joni> how do you run it then?
[13:52:53] <JanVanGilsen> i made a shortcut on the desktop to /scripts/emc/
[13:53:47] <alex_joni> ok, then let's try to run it in a terminal
[13:53:59] <alex_joni> start a terminal (should be on the menu, under apps)
[13:54:54] <JanVanGilsen> ok i started emc in the terminal :)
[13:56:34] <JanVanGilsen> stange, it starts the 2.1.7 version
[13:57:39] <alex_joni> right
[13:57:48] <alex_joni> you need to go to the path where you have the CVS version
[13:57:59] <alex_joni> and run 'scripts/emc' instead of 'emc'
[13:58:07] <JanVanGilsen> ok it worked :)
[13:58:21] <alex_joni> any strange messages while starting up?
[13:59:20] <JanVanGilsen> no, only the "Bad character 'Z' used" popup from axis
[13:59:54] <alex_joni> hrmm.. that's odd (we'll keep that in mind for now..)
[14:00:00] <alex_joni> next I would try to change step_type to 2
[14:00:14] <alex_joni> or 1
[14:00:22] <JanVanGilsen> thats because there is a Z in the spash G-code from axis
[14:00:22] <jepler> that error may appear because the splash gcode has "Z" words in it
[14:00:25] <alex_joni> (those are certain to work 100% ok)
[14:01:08] <alex_joni> JanVanGilsen: try changing steptype, and comment out the links to the parport for now
[14:01:29] <alex_joni> "the ' connect signals to step pulse generator outputs' part"
[14:07:29] <JanVanGilsen> I commented out the phase-# lines and the parport.0.pin-##-out lines
[14:11:15] <alex_joni> right
[14:17:03] <JanVanGilsen> still got those ferrors ...
[14:17:30] <JanVanGilsen> even when i use step/dir signals ...
[14:17:40] <cradek> I still think you should put [AXIS_2] back but not change hal
[14:18:03] <cradek> or, undo the last thing you did before you had the problem
[14:19:14] <JanVanGilsen> i think it maybe wiser to make a fresch copy of the stepper config and start over ... :)
[14:19:35] <cradek> that could be. then, change one thing at a time and test.
[14:19:46] <alex_joni> until you can reproduce the error
[14:19:55] <cradek> first sign of a problem, stop changing things :-)
[14:21:20] <JanVanGilsen> =)
[14:52:09] <cradek> I only say that because people tend to get it wrong
[14:52:29] <cradek> when there's a problem, they go on to change a dozen other things they hadn't even touched before
[15:13:51] <JanVanGilsen> jipiiiieee!! it works
[15:16:12] <cradek> yay!
[15:16:55] <JanVanGilsen> cradek, you were right, AXES = # can't be less then 3, and [AXIS_2] TYPE=... is also obligated
[15:17:51] <JanVanGilsen> maybe this should be added to the documentation :)
[15:17:58] <cradek> that's strange...
[15:18:17] <cradek> when you change only AXES=2, does it fail with those following errors?
[15:19:29] <JanVanGilsen> even a lathe has AXES=3
[15:21:45] <cradek> ouch, if I take Z out of sim/axis and run a program with only XY, it runs but the X/Y often say 'nan'
[15:21:50] <JanVanGilsen> when AXES=2 you'll get following errors with controled motion on both axes (not with jog-moves)
[15:21:55] <cradek> as the cone flashes around in various places on the screen...
[15:22:06] <cradek> yeah there's definitely a problem if you remove Z.
[15:22:57] <cradek> let me see if I can find it
[15:24:08] <JanVanGilsen> could it be that the trivkins must have a joint 2 in order to function?
[15:26:18] <cradek> it is somewhere in the motion controller... the issued moves look fine
[15:29:56] <cradek> Breakpoint 1, tpAddLine (tp=0xb7be9880, end={tran = {x = 0.99999999999999989, y = 0.99999999999999989, z = 0}, a = 0, b = 0, c = 0, u = 0, v = 0, w = 0}, type=11, vel=0.016666666666666663, ini_maxvel=1.541031987208572, acc=0, enables=11 '\v') at emc/kinematics/tp.c:327
[15:30:04] <cradek> I see a smoking gun...
[15:41:42] <skunkworks> I see a bunch of numbers...
[15:43:08] <CIA-8> 03cradek 07TRUNK * 10emc2/src/emc/task/emccanon.cc: fix configs that have some or all of x,y,z missing
[15:43:40] <cradek> fixed
[15:49:20] <skunkworks> nice work. :)
[15:58:12] <JanVanGilsen> :)
[16:06:24] <SWPadnos> cradek, did you intend to leave debug_velacc on in that last commit
[16:06:27] <SWPadnos> ?
[16:06:36] <cradek> arg
[16:07:30] <CIA-8> 03cradek 07TRUNK * 10emc2/src/emc/task/emccanon.cc: oops
[16:09:22] <JanVanGilsen> Is it possible to change the AXIS_OPEN_FILE in the ini file? because i keep getting the "Bad character 'Z' used" error every time I start Axis
[16:09:57] <SWPadnos> you should be able to set an environment variable or change the ini var
[16:10:20] <SWPadnos> but I think AXIS has an internal default as well, so you need to have sometihng specified or it'll load that
[16:10:42] <SWPadnos> you can also replace the file it loads to be something else
[16:10:44] <cradek> looks like you can also specify a file on the command line
[16:11:00] <cradek> but I agree that should come from the ini, not the environment
[16:11:18] <SWPadnos> it comes from any of them, but you have to specify something to prevent the internal default from loading
[16:11:37] <cradek> SWPadnos: no it doesn't, look at the code
[16:12:07] <SWPadnos> hmmm. I had, but then again I'm remembering this BC (before coffee)
[16:12:24] <cradek> :-)
[16:12:43] <cradek> sorry I didn't mean to snap at you
[16:14:06] <SWPadnos> oh - I didn't even notice
[16:14:22] <cradek> LOOK AT THE CODE FOR ONCE, DAMN YOU
[16:14:26] <cradek> oops there it was again
[16:14:35] <SWPadnos> it does work as I thought, except that it doesn't have an ini file option
[16:14:51] <cradek> yeah, and that would probably be nice
[16:15:07] <JanVanGilsen> i agree :)
[16:15:09] <SWPadnos> open arg[0] else evnvar("AXIS_OPEN_FILE") else "internal default file"
[16:15:21] <SWPadnos> envvar
[16:15:28] <SWPadnos> eh - whatever. amybe I'll mae coffee now
[16:15:32] <SWPadnos> maybe
[16:15:34] <SWPadnos> make
[16:15:35] <SWPadnos> see?
[16:16:03] <cradek> SWPadnos: please feel free to fix it... I don't know python
[16:18:38] <SWPadnos> if I knew the specifics of what you get when you use inifile.find and the item doesn't exist, I'd be able to do it
[16:18:42] <SWPadnos> but I don't
[16:29:24] <JanVanGilsen> is it possible to set it in ~/.axisrc ?
[16:29:40] <SWPadnos> it doesn't appear to look there
[16:31:11] <jepler> hm somewhere I have a not-checked-in change to make this an inifile item
[16:31:23] <jepler> but I dunno where
[16:31:31] <SWPadnos> I thought there had been a discussion about that in the last few weeks
[16:31:40] <jepler> yes there was -- that guy set AXIS_OPEN_FILE
[16:38:11] <CIA-8> 03jepler 07TRUNK * 10emc2/docs/src/config/ini_config.lyx: [DISPLAY]OPEN_FILE allows a different 'splash gcode' to be opened
[16:38:11] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: [DISPLAY]OPEN_FILE allows a different 'splash gcode' to be opened
[16:40:58] <jepler> ^^ lightly tested
[16:42:02] <SWPadnos> there should be an inifile function that's equivalent to "environ_has_key"
[16:42:17] <SWPadnos> but I don't know that it would actually be any more efficient, considering what an ini file is
[16:43:33] <jepler> since it happens once at startup, I'm unlikely to be worried about the efficiency
[16:43:54] <SWPadnos> yeah - all the ini stuff is at startup, so it's pretty irrelevant to overall efficiency
[17:34:32] <Roguish> jepler: nice addition of the 'splash gcode file', where would one need to put that file?
[17:34:54] <SWPadnos> anywhere. just put the full path to it in the ini file
[17:35:20] <Roguish> will try that.
[17:35:38] <SWPadnos> but only with the latest CVS, of course ;)
[17:36:13] <Roguish> just did an update and it shows in the axis code.
[17:36:24] <SWPadnos> oh good :)
[17:42:25] <Roguish> works. thanks.
[17:42:37] <Roguish> like it.
[19:14:26] <CIA-8> 03cradek 07TRUNK * 10emc2/src/emc/task/emccanon.cc: normals are unitless, don't scale them
[20:05:49] <lerneaen_hydra> random sanity check; what's the limiting factor for a two-stroke's unloaded, full throttle rpm? a slow burn of fuel? insufficient fuel flushing? something else?
[20:06:41] <cradek> failure of the flap valves to keep working?
[20:07:03] <lerneaen_hydra> oh, this would be for smaller motors that don't have any valves
[20:07:17] <lerneaen_hydra> moped and/or RC engines
[20:07:40] <cradek> I guess I don't know
[20:07:48] <cradek> (it's surprising that they work at all)
[20:08:19] <lerneaen_hydra> hmm, IMO they're so robust and simple there's not much that can break
[20:08:21] <alex_joni> yeah, even without batteries
[20:08:45] <lerneaen_hydra> only a crankshaft, piston rod and piston
[20:08:51] <lerneaen_hydra> no valves or anything
[20:10:27] <SkinnyPuppy34> lerneaen I'm no expert but we used to play with the port heights. Been a while since my motorcycle days
[20:13:31] <lerneaen_hydra> oh, sweet :)
[20:14:20] <SkinnyPuppy34> What kind of engine is on your cart? I ran across your site last night
[20:16:27] <lerneaen_hydra> it's a radne raket 85cc
[20:16:35] <lerneaen_hydra> 8.5kw at 8500rpm
[20:16:53] <lerneaen_hydra> maxes out at about 12000rpm apparently
[20:18:00] <lerneaen_hydra> I'm planning to rebuild it for a parade-esque thing my college has in the spring, with a large prop that gives thrust
[20:19:30] <SkinnyPuppy34> That would be interesting to see what the piston sfm is for your stroke/rod combo. Perhaps a longer rod with a cylinder base spacer would be a place to start. But that is ignoring the port/exhaust tuning..
[20:20:02] <skunkworks> you can't measure gas engine output in KW - what are you thinking?
[20:20:20] <lerneaen_hydra> sure you can, at a given rpm
[20:20:24] <skunkworks> ;)
[20:20:45] <SkinnyPuppy34> Were the pics where you were in the lathe at your college? Clean shop !
[20:21:00] <lerneaen_hydra> the lathe is at home, the mill at school
[20:21:12] <lerneaen_hydra> well, an electronics club of sorts
[20:21:30] <lerneaen_hydra> http://www.radne.se/Store/Product/ProductDetails.aspx?PartNo=3001-1 <-- the motor
[20:22:53] <SkinnyPuppy34> Neat engine, can't say I've seen one of those.
[20:23:25] <lerneaen_hydra> apparently they're very common for gokarts here in sweden, as well as propeller-driven hang-gliders (mainly the 120cc version though)
[20:51:07] <JymmmEMC> SWPadnos: ping
[20:51:07] <alex_joni> JymmmEMC: pong
[20:51:07] <alex_joni> he's eating
[20:51:07] <JymmmEMC> He's ALWAYS eating
[20:51:07] <alex_joni> you also noticed?
[20:51:07] <JymmmEMC> yeah, in person too.... The man loves his food =)
[20:51:07] <JymmmEMC> He can eat more than me even!!!
[20:51:11] <LawrenceG> hey guys.... hangin well?
[20:53:26] <skunkworks> LawrenceG: how is the current servo amp coming?
[20:53:34] <robin_sz> meep/
[20:53:35] <robin_sz> ?
[20:53:43] <alex_joni> dude
[20:53:49] <skunkworks> ROBIN!
[20:53:51] <alex_joni> long time no see
[20:53:52] <robin_sz> hey:)
[20:53:57] <robin_sz> heh
[20:53:58] <LawrenceG> still smokin!
[20:54:02] <robin_sz> well, yes its been a while
[20:54:17] <alex_joni> busy licking stamps?
[20:55:47] <robin_sz> been *a bit* busy :)
[20:55:47] <robin_sz> alex_joni: oh yes
[20:55:47] <robin_sz> alex_joni: had a good sale early this month,
[20:55:47] <alex_joni> glad to ehar
[20:55:47] <alex_joni> hear even
[20:55:47] <LawrenceG> JymmmEMC: did you ever decide on a watchdog cct for your machine?
[20:55:47] <robin_sz> alex_joni: 1 item, hit the 10^6 dollars indicator
[20:55:57] <robin_sz> alex_joni: and the sheetmetal busines sis very very busy too
[20:56:05] <alex_joni> good stuff then
[20:56:09] <robin_sz> yep
[20:56:15] <alex_joni> same here.. busy :)
[20:56:21] <robin_sz> excellent
[20:56:42] <robin_sz> dual wire 1200 amp robotic laser welders?
[20:56:51] <alex_joni> without the laser :D
[20:56:59] <robin_sz> now wheres the fun in that?
[20:57:00] <alex_joni> no, actually it was more thinsheet lately
[20:57:05] <alex_joni> <300 amps
[20:57:09] <robin_sz> kk
[20:57:16] <alex_joni> no fun, I know..
[20:57:23] <alex_joni> but 6-8m linear axes
[20:57:34] <robin_sz> we've been doing oodles of tig recently
[20:57:42] <robin_sz> bought another tig set
[20:57:56] <alex_joni> ac/dc I hope
[20:58:01] <robin_sz> of course
[20:58:11] <robin_sz> not sqaure wave though
[20:58:13] <robin_sz> sadly
[20:58:28] <alex_joni> coo
[20:58:35] <alex_joni> I just finished another wig-drive unit
[20:58:49] <alex_joni> for filler wire
[20:58:55] <robin_sz> ah, kk
[20:58:57] <robin_sz> gecko?
[20:59:31] <alex_joni> nah, this was with the original card from teh mig-drive
[20:59:39] <robin_sz> right
[20:59:40] <alex_joni> some simple 1-Q drive
[20:59:43] <alex_joni> no feedback :)
[20:59:50] <robin_sz> right
[20:59:57] <alex_joni> it's pretty ok though
[20:59:59] <robin_sz> is it slow? robotic tig?
[21:00:06] <alex_joni> yeah.. automatic
[21:00:13] <alex_joni> for linear welding tubes
[21:00:21] <robin_sz> ahh
[21:00:42] <robin_sz> looks like we got new factory
[21:00:52] <robin_sz> 800m^2
[21:01:03] <alex_joni> http://www.scheld-gmbh.de/images/sal1.gif
[21:01:06] <alex_joni> something like that
[21:01:34] <robin_sz> * robin_sz nods
[21:01:37] <robin_sz> seam welder
[21:03:07] <CIA-8> 03alex_joni 07TRUNK * 10emc2/src/emc/task/emccanon.cc: add circular vel constraint (expressed as time)
[21:03:24] <robin_sz> so, anyting new happen
[21:03:27] <robin_sz> wait
[21:03:29] <robin_sz> let me guess
[21:03:37] <robin_sz> ive been away for what, months ...
[21:03:41] <robin_sz> hmmm tricky ...
[21:03:57] <CIA-8> 03alex_joni 07TRUNK * 10emc2/src/emc/kinematics/tp.c: remove borken test. it's beeing tested in canon now
[21:04:01] <alex_joni> we had a vote sometime lately
[21:04:04] <alex_joni> new board, etc
[21:04:16] <alex_joni> emc2.2.0 is soon to be released
[21:04:17] <robin_sz> ok, heres my first guess ... Jymmmmmm got his PSU and drives hoooked up and his router going?
[21:04:26] <alex_joni> bad guess :P
[21:04:32] <robin_sz> damn ;)
[21:04:38] <robin_sz> whos the new board?
[21:04:44] <robin_sz> (briefly)
[21:04:49] <alex_joni> actually I don't know.. JymmmEMC was pretty quiet about his router lately
[21:04:58] <alex_joni> jmk, cradek, jepler, SWPadnos and me
[21:05:03] <robin_sz> nice
[21:05:32] <robin_sz> any amazing new developments?
[21:05:41] <alex_joni> http://www.linuxcnc.org/content/view/12/10/lang,en/
[21:05:42] <robin_sz> like emc for palm?
[21:06:11] <skunkworks> doublefreq for stepping machines. (close to doubles the steprate out of emc2)
[21:06:14] <alex_joni> http://cvs.linuxcnc.org/cvs/emc2/debian/changelog?rev=1.24;content-type=text%2Fplain
[21:06:22] <ds2> Mmmmm stepper drives on the serial port
[21:06:33] <alex_joni> ds2: ethercat all the way :D
[21:06:52] <jlmjvm> alex_joni:is 2.2 about ready?
[21:06:54] <alex_joni> robin_sz: 9 axes now
[21:07:02] <alex_joni> jlmjvm: err..
[21:07:01] <robin_sz> 9 ?
[21:07:01] <ds2> * ds2 whispers EMC4Gumstix over and over
[21:07:03] <robin_sz> coo
[21:07:11] <alex_joni> robin_sz: XYZABCUVW
[21:07:14] <skunkworks> alex_joni: what is the story on rt ethernet so far?
[21:07:23] <alex_joni> * alex_joni wouldn't know..
[21:07:32] <alex_joni> but ethercat is somethign I find *very* promising
[21:07:43] <skunkworks> * skunkworks thought someone was working on that..
[21:07:51] <alex_joni> it's some RT ethernet, but way more defined as a protocol
[21:07:55] <SWPadnos> one problem with ethercat is that the drives must have custom ethernet hardware
[21:07:55] <robin_sz> have sub-axes got any further?
[21:08:10] <alex_joni> SWPadnos: yeah, but there are a couple already out there
[21:08:20] <alex_joni> robin_sz: sub-axes?
[21:08:34] <robin_sz> big x gantry .. slooooow
[21:08:50] <alex_joni> ah.. you can probably do that with U (paralel to X)
[21:08:50] <robin_sz> small x axeis, for doing smalll, fast motions
[21:08:53] <SWPadnos> oh - yes, parallel / slaved axes are better now
[21:08:57] <robin_sz> big x tries to catch up
[21:09:15] <alex_joni> hmm.. don't think you can do that now :P
[21:09:25] <SWPadnos> at least, there's a "gantrykins" that makes things like XXYZ much easier
[21:09:27] <robin_sz> someone did
[21:09:37] <skunkworks> someone did with kins.
[21:09:41] <ds2> bit bang ethernet on the parallel port should give you all the customization you need ;)
[21:09:47] <alex_joni> ds2: lol
[21:09:52] <robin_sz> this is more X(+x)YZ
[21:10:28] <robin_sz> alex_joni: ordered another laser
[21:10:41] <alex_joni> trumpf 10kW+ ?
[21:10:47] <robin_sz> just a 2kw
[21:10:49] <alex_joni> :P
[21:10:51] <robin_sz> trumpf
[21:10:51] <jepler> the planner has no idea about things like "X can go 10ips for 1in, and 2ips for >1in", which I'm sure is in itself a gross oversimplification of what's actually required for this kind of double axes
[21:11:17] <alex_joni> * alex_joni agrees with jepler
[21:11:21] <jepler> the best you can do without a big rework is specify cartesian constraints that can be met no matter where in the work envelope you are and no matter the recent history of movements
[21:11:39] <robin_sz> you could do it in post process
[21:11:42] <alex_joni> you can also define the secondary x as U
[21:11:47] <alex_joni> and let CAM take care of it
[21:11:54] <robin_sz> * robin_sz nods
[21:11:54] <jepler> yes, in 2.2 you can program in UVW coordinates and do this manually
[21:12:09] <alex_joni> although I'm not sure you can G03UW without XY moves
[21:12:14] <robin_sz> the latest trumpf stuff does exactly this
[21:12:23] <jepler> there are no UVW arcs
[21:12:32] <alex_joni> jepler: that's what I remembered
[21:12:43] <jepler> bbl
[21:12:51] <robin_sz> it would need to be UY arcs or XY arcs, deneding
[21:12:54] <robin_sz> depending
[21:12:56] <alex_joni> CAM again.. and small lines
[21:14:08] <alex_joni> hmm.. there's one thing missing from that list
[21:14:14] <alex_joni> wait-on-input G-codes
[21:14:20] <robin_sz> ooh
[21:14:22] <robin_sz> nice
[21:15:01] <alex_joni> we also have nicer html docs
[21:15:02] <alex_joni> http://www.linuxcnc.org/docview/devel/html/gcode_main.html
[21:15:44] <alex_joni> http://www.linuxcnc.org/docview/devel/html/gcode_main.html#sec:M66:
[21:16:48] <SkinnyPuppy34> Looks nice alex !
[21:16:59] <robin_sz> * robin_sz is looking
[21:18:31] <CIA-8> 03alex_joni 07TRUNK * 10emc2/debian/changelog: mention wait-on-input
[21:18:34] <jlmjvm> looks great
[21:19:09] <skunkworks> jlmjvm: how is the stepper-encoder project working?
[21:19:46] <jlmjvm> doing good,just need to mount my x and y motors
[21:20:12] <jlmjvm> cut metal the first time yesterday with emc
[21:21:19] <jlmjvm> trying to figure out how to use the lube level input
[21:23:00] <jlmjvm> alex_joni:have you ever used the lube sensor with emc?
[21:23:07] <alex_joni> jlmjvm: nope
[21:23:41] <alex_joni> but I wrote the stuff.. and I must say.. it doesn't work :P
[21:24:19] <alex_joni> err.. wait.. reading my code now :)
[21:24:55] <alex_joni> well.. the iocontroller reads the lube-level input
[21:24:59] <alex_joni> and puts that into status
[21:25:10] <alex_joni> not sure if any of the GUI's do anything with it though..
[21:25:37] <alex_joni> (you could have a pyvcp extension to AXIS to display a warning LED when it's low..)
[21:26:51] <jlmjvm> k,so it doesnt stop the machine or anything,would just give an led if you did the pyvcp?
[21:27:21] <SWPadnos> there's no specific definition of what a low lube level should do, so it doesn't automatically do anything
[21:27:31] <SWPadnos> you can set it up to do whatever you want though
[21:27:41] <skunkworks> in hal - couldn't you hook it to the estop circuti if you wanted
[21:28:01] <jlmjvm> think im gonna run mine thru my estop loop
[21:28:21] <jlmjvm> when it has oil the switch is closed
[21:44:36] <CIA-8> 03alex_joni 07TRUNK * 10emc2/src/emc/task/emccanon.cc: silence velacc debug messages
[21:52:13] <cradek> jlmjvm: I think I said this before - on my bridgeport, low lube doesn't allow you to turn on the spindle, but it won't stop it if it's on. this lets a job finish.
[21:52:59] <cradek> causing immediate estop seems like it could cost you a part for no good reason
[21:53:32] <cradek> but that's the beauty of hal - we can both do what we want.
[21:54:37] <alex_joni> * alex_joni also does what he wants
[21:55:08] <alex_joni> in this case going to sleep
[21:55:08] <alex_joni> good night all
[21:55:08] <cradek> goodnight
[21:55:08] <alex_joni> good thing HAL lets me do this
[22:08:39] <jlmjvm> cradek:are you saying that that hal would be able to do basically what your bridgeport control does,or would that be something for classic ladder?
[22:32:54] <cradek> jlmjvm: with hal/cl you could do whatever you want. whether you write it in terms of hal logic blocks or ladder is just a matter of which you like better
[22:43:33] <[1]a-l-p-h-a> [1]a-l-p-h-a is now known as a-l-p-h-a
[22:44:30] <jlmjvm> cradek:thanks for the info,gonna go the ladder route for now
[22:45:53] <jlmjvm> probably gonna start taking some plc classes at nite when the next semester starts up
[22:48:26] <dmess> ladder is quite easy to read and pick up
[22:49:24] <jlmjvm> im hoping i get the hang of it
[22:50:02] <jlmjvm> im playing around with the demo step cl setup right now