#emc | Logs for 2008-09-13

[00:07:33] <SWPadnos> it should, if you use CVS trunk
[00:08:27] <SWPadnos> actually, you may not need to - the velocity estimate has been pretty good for a while (threading uses "position" of the spindle, so a position estimate had to be added)
[00:17:37] <dareposte> yeah i'm using the trunk version
[00:17:43] <dareposte> (thanks to your help compiling it..)
[00:17:59] <SWPadnos> cool then :)
[00:18:57] <dareposte> this is getting really exciting
[00:19:15] <SWPadnos> so exciting I should convert my mill soon :)
[00:19:21] <dareposte> i agree what type is it
[00:19:36] <dareposte> i tried to explain how exciting this is getting to someone at work and they just looked at me
[00:20:02] <dareposte> maybe not universally exciting :-/
[00:20:30] <SWPadnos> Bridgeport
[00:20:37] <SWPadnos> manual, series 1
[00:20:49] <dareposte> oh that's a large fella
[00:21:16] <SWPadnos> heh, for some :)
[00:21:23] <SWPadnos> it only weighs a ton
[00:21:26] <dareposte> my mill is probably about the same size as the vice on that beast
[00:21:33] <dareposte> vise
[00:21:44] <SWPadnos> or the rotary table - that's 130 pounds. the vise is only 80 (each, I have tw)
[00:21:47] <SWPadnos> teo
[00:21:48] <SWPadnos> two
[00:22:00] <dareposte> i carried my mill in from the trunk of my car and assembled it on the kitchen counter
[00:22:14] <dareposte> then carried it over to the workbench from there
[00:22:27] <SWPadnos> hmmm. I dont think it would be my wife that would stop me from doing that :)
[00:22:42] <dareposte> it only weighs 0.075 tons
[00:23:07] <SWPadnos> yep, about like a 6" vise
[00:23:48] <dareposte> but for what it is it has served me well. i do get machine envy a lot
[00:24:50] <SWPadnos> well, yours is a CNC, so that makes it better than mine for a lot of things
[00:25:24] <dareposte> no my mill is not yet
[00:25:31] <dareposte> it's a pain in the rear to use in fact
[00:25:46] <dareposte> i have the parts to convert it but am doing the lathe first, because i just like lathes more
[00:26:17] <SWPadnos> oh, in that case get to work!
[00:26:21] <dareposte> in hindsight if I had done the mill first then i could have made the brackets for the lathe a lot prettier and more accurate
[00:27:09] <dareposte> the purpose in all of this is to in fact build an extremely large machine, about 86" long
[00:27:55] <dareposte> but it takes a number of parts which i'm way too lazy to machine by hand, and probably would screw up
[00:27:57] <SWPadnos> that's good size
[00:28:06] <dareposte> so i figured make these cnc first then use them to make the parts to make the monster
[00:28:28] <SWPadnos> yep
[00:29:15] <dareposte> the monster machine is a giant drill press with a tapering attachment and servo spindle
[00:29:53] <dareposte> i don't think stepconf will work for it :(
[00:30:41] <SWPadnos> depends. if you expect to use a parport and run stepper drivers, then maybe
[00:31:03] <SWPadnos> there is support in stepconf for using a PWM output for spindle speed control
[00:31:09] <dareposte> yeah, steppers for the drill feed, AC servo for the spindle
[00:31:13] <dareposte> stepper for the taper feed
[00:31:24] <dareposte> yeah i saw that
[00:31:42] <dareposte> on my lathe i have a 0-10v speed control input
[00:31:55] <dareposte> i've been puzzlin on a way to convert the PWM to 0-10v somehow
[00:31:58] <SWPadnos> it's not too expensive to get real interface hardware though, a Mesa board and a couple of interface cards totals $400 or so
[00:32:10] <SWPadnos> op-amp circuit
[00:32:18] <dareposte> yeah thats what i was thinking
[00:32:27] <dareposte> an RC filter with op-amp buffer maybe
[00:32:32] <SWPadnos> yep
[00:32:39] <dareposte> i was just worried bout the linearity of it
[00:32:58] <SWPadnos> I think the PWM can be set for PDM mode, so you can use pretty small value caps and resistors
[00:32:58] <dareposte> so then i considered a uC with DAC
[00:33:06] <SWPadnos> I wouldn't bother with that
[00:33:09] <SWPadnos> though it is possible
[00:33:16] <dareposte> what is PDM mode
[00:33:41] <dareposte> pulse duty?
[00:33:47] <SWPadnos> pulse density modulation - it basically toggles the output as fast as possible for the requested duty cycle
[00:34:01] <dareposte> ahh
[00:34:02] <SWPadnos> rather than turning it on at the beginning of a (long) cycle, then turning it off xx% later
[00:34:17] <dareposte> that would probably be more linear
[00:34:23] <SWPadnos> so 50% is on-off-on-off-on-off
[00:34:29] <SWPadnos> rather than on-on-on-off-off-off
[00:35:15] <dareposte> how do i set that up
[00:36:56] <SWPadnos> I'm not sure with stepconf. I think maybe you set the period to 0
[00:37:45] <dareposte> i'll have to investigate when I get it all hooked together
[00:39:13] <SWPadnos> sounds like a plan
[00:41:41] <dareposte> the real question is can i control my lathe AND the 6-dof robot i plan on adding to change the tooling?
[00:41:47] <dareposte> ;-)
[00:42:26] <SWPLinux> sure, just get 2 PCs
[00:42:59] <dareposte> what about add another parallel port
[00:43:50] <SWPLinux> sure, if you want to write your own interpreter and kinematics module
[00:44:10] <dareposte> sounds dangerous
[00:44:15] <jmkasunich> do you want to control the robot with g-code?
[00:44:36] <dareposte> i want to control it with a toolchange command
[00:44:46] <jmkasunich> that is not "control"
[00:45:03] <jmkasunich> "change tool" is just a way to start the toolchange
[00:45:06] <dareposte> yeah
[00:45:09] <dareposte> good point
[00:45:30] <jmkasunich> "control" is "move the arm to here, then to there, then rotate 30 deg, then release tool, then open robot hand,,,,,,,"
[00:45:32] <dareposte> in that case i guess i would like to control it by recording joint moves and replaying them, triggered with a change tool command
[00:45:54] <dareposte> so no gcode
[00:46:02] <jmkasunich> if it truly is as simple as replaying recorded moves, you could use halstreamer
[00:46:22] <dareposte> yeah it would be that simple
[00:46:44] <jmkasunich> heaven help you if something doesn't go as planned though
[00:47:10] <dareposte> just record a path for removing each tool from the toolpost and putting it away, then another path for getting each tool out of the tool holder and sending it to the tool post
[00:47:15] <jmkasunich> "latch tool into spindle, release tool from robot" - if the tool doesn't latch for some reason, when the robot opens its hand tool falls on floor
[00:47:23] <dareposte> halstreamer huh
[00:47:29] <jmkasunich> usually there would be some kind of feedback from sensors to tell you that it latched
[00:47:40] <dareposte> yeah there would have to be IO involved
[00:47:47] <dareposte> sensors and whatnot
[00:49:36] <dareposte> maybe that project will be after the monster drill
[00:49:43] <dareposte> waay after
[00:51:25] <dareposte> do you know if anybody has tried running emc on a PC/104 form board?
[00:51:51] <dareposte> it might be handy for reducing the controller size, and i know they can run some variants of linux on those things
[00:52:10] <SWPLinux> they'll all run Linux, more or less
[00:52:21] <dareposte> get the pc/104 processor and a parallel port io module and they're barely bigger than a box of cigars
[00:53:25] <dareposte> i may have one around here i can try it on, but it would be a slow processor for sure
[00:56:06] <SWPLinux> they're small and they run Linux, but they're not necessarily great for realtime
[01:00:33] <dareposte> i can see that
[01:00:40] <dareposte> there are some pretty speedy ones out there though
[01:01:21] <dareposte> i think real time kernel extensions are pretty common for PC/104's too, but probably not at the speeds you work at here
[01:24:31] <dareposte> if you CNC your bridgeport will you try to move the knee or just the quill?
[01:32:36] <JymmmEMC> SWPadnos: SWPLinux
[01:32:46] <SWPLinux> what what
[01:32:48] <JymmmEMC> SWPadnos: SWPLinux http://www.arduino.cc/
[01:32:50] <JymmmEMC> SWPadnos: SWPLinux http://www.arduino.cc/
[01:32:51] <JymmmEMC> SWPadnos: SWPLinux http://www.arduino.cc/
[01:32:58] <SWPLinux> yeah, what about it?
[01:33:04] <SWPLinux> jeff has one I think
[01:33:07] <JymmmEMC> *SMACK*
[01:33:37] <JymmmEMC> been trying to show that to you for three days ya goober!
[01:33:50] <SWPLinux> I've seen it you - uh - peanuy
[01:33:51] <SWPLinux> t
[01:34:23] <JymmmEMC> * JymmmEMC just shakes his head
[01:34:54] <JymmmEMC> SWPLinux: you awake (mentally) ?
[01:35:04] <SWPLinux> yes
[01:35:30] <SWPLinux> I know about the arduino board(s) - what did you want me to see?
[01:35:36] <JymmmEMC> SWPLinux: (not a trick question) How do I uniquely identify a perosn globally
[01:36:08] <jmkasunich> tattoo a 25 digit number on their forehead
[01:36:24] <JymmmEMC> jmkasunich: that be easy
[01:36:36] <jmkasunich> glad I could help
[01:36:43] <SWPLinux> DNA
[01:36:46] <JymmmEMC> cannot be based upon geographical location
[01:37:07] <JymmmEMC> SWPLinux: Ok, now to put that in digital form
[01:37:09] <jmkasunich> SWPLinux: DNA won't work for twins
[01:37:10] <JymmmEMC> ?
[01:37:16] <JymmmEMC> jmkasunich: really?
[01:37:27] <JymmmEMC> fingerprints?
[01:37:31] <jmkasunich> identical twins anyway....
[01:37:32] <SWPLinux> not unique
[01:37:50] <JymmmEMC> twins have identical fingerprints?
[01:37:55] <jmkasunich> JymmmEMC: as phrased the question is impossible
[01:38:37] <JymmmEMC> I thought that NOBODY has the same fingerprints, not even twins
[01:38:55] <jmkasunich> does globally mean you want to be able to identify individuals in a rural villiage in africa that hasn't even heard of computers, from your computer?
[01:39:02] <SWPLinux> JymmmEMC: no, not necessarily, but I recall a segment (on That's Incredible! maybe) where there was an entire family with no fingerprints
[01:39:07] <JymmmEMC> jmkasunich: yes
[01:39:27] <JymmmEMC> SWPLinux: cool, I'd like to have no fingerprints
[01:39:35] <jmkasunich> sandpaper
[01:39:43] <JymmmEMC> =)
[01:40:02] <SWPLinux> yes - they ended the segment by saying "I told my kids if you want to get ahead, steal. Steel, that is." :)
[01:40:20] <JymmmEMC> lol
[01:54:37] <cradek> finally solved my persistent Z oscillation problem
[01:54:50] <cradek> after much studying, I soldered the right resistor in the right place
[01:55:43] <SWPLinux> with the right solder!
[01:56:04] <Guest316> it was a drive issue?
[01:57:15] <cradek> velocity loop tuning problem, which is in the servo drive/tach
[01:57:46] <cradek> it's very responsive but doesn't oscillate now
[01:57:55] <cradek> I even put the covers on the box.
[02:00:55] <stustev> hey - very good :)
[02:01:37] <cradek> still could use new balls though. I have not tried ordering them yet.
[02:01:50] <SWPLinux> uh-oh. cover on the box. that's a bad omen
[02:02:13] <stustev> no hurry now - that lathe could make bushings with the current ball screw
[02:02:34] <cradek> no, bad omen would be moving it so I can't easily open the door on the box
[02:02:38] <SWPLinux> heh
[02:02:51] <SWPLinux> after forgetting to replace the cover :)
[02:35:02] <Guest316> Guest316 is now known as skunkworks_
[02:38:47] <SWPadnos> hmmm. there's peck drilling along W - does tapping not work that way?
[02:38:56] <skunkworks_> huh - wonder where I a logged in
[02:39:03] <skunkworks_> * am
[02:39:27] <cradek> SWPadnos: no, but SMOP
[02:39:34] <skunkworks_> I must have left my computer on at work
[02:39:46] <SWPadnos> ok, makes sense then :)
[02:39:54] <cradek> what does?
[02:39:58] <SWPadnos> oh, there's another good one :)
[02:40:05] <SWPadnos> cradek, why Stuart is asking for it :)
[02:40:07] <cradek> you're talking in code
[02:40:26] <SWPadnos> see emc-devel list with Stuart's two most recent requests
[02:40:29] <cradek> ohhh
[02:40:47] <cradek> yeah, someone ought to write that stuff
[02:40:53] <stustev> just a small request
[02:41:00] <SWPadnos> or cut off his internet access
[02:41:06] <SWPadnos> oh, hi stustev :)
[02:41:16] <stustev> my ears are burning
[02:41:20] <stustev> :)
[02:41:46] <stustev> tell me where to start - I'm on it :)
[02:42:52] <cradek> um, in the interpreter
[02:43:05] <stustev> you will notice I didn't suggest to tell me where to go to get started :)
[02:43:16] <SWPadnos> then on to NML, canon, and motion I think
[02:43:19] <cradek> then, tapping needs a bit of special help the trajectory planner
[02:43:43] <stustev> sounds like a plan
[02:43:59] <cradek> I bet tapping is easier than arcs
[02:44:11] <cradek> I don't remember if I ran into an obstacle doing arcs.
[02:44:18] <SWPadnos> there's been some work on tilted arcs, but I'm not sure where it is
[02:44:18] <fenn> what about tapping along an arc? :P
[02:44:19] <cradek> they might be done or almost done on a branch.
[02:44:39] <stustev> fenn: with a bent tap?
[02:44:55] <SWPadnos> threadmilling on an angle, along W
[02:45:08] <fenn> that's just a helical move
[02:45:09] <SWPadnos> UVW-oriented helical milling
[02:45:14] <fenn> er, but i guess emc can't do that
[02:45:25] <stustev> it will
[02:46:49] <SWPadnos> if you have a trunion-type 5-axis machine can you still tap along Z?
[02:47:08] <stustev> sure - it is just 3 axis tapping
[02:47:13] <SWPadnos> presumably you can, but if the kins is fiddling with other joints it makes my head hurt
[02:49:39] <cradek> if I ever worked on uv arcs, as I think I recall doing, I didn't check it in
[02:50:05] <SWPadnos> I remember seeing an AXIS screenshot with tilted arcs in it
[02:50:30] <SWPadnos> not UVW aligned, just in arbitrary planes I think
[02:51:34] <cradek> TP and posemath have supported arbitrary arcs for a long time. but canon (and of course interp) do not.
[02:51:48] <SWPadnos> oh. hmmm
[02:52:13] <stustev> is TP code for the trajectory planner?
[02:52:22] <cradek> yes
[02:58:30] <stustev> It was especially gratifying to change the parameters I exported out of the kins and see the results on the machine. That gave me a real sense of POWER! bwa ha ha :)
[02:58:57] <SWPadnos> if the cinci doesn't give you a sense of power, you have a problem ;)
[02:59:05] <cradek> hmm, the antidote for that sense of power would be to look at ARC_FEED in emccanon.cc
[02:59:16] <SWPadnos> heh
[02:59:27] <stustev> looking
[03:01:31] <cradek> the arc structure in posemath would need to change. currently the normal vector is a PmCartesian.
[03:02:16] <cradek> maybe the definition of ARC_FEED could stay the same
[03:02:34] <SWPadnos> is that normally (no pun intended) [0,0,1], [0,1,0], or [1,0,0] ?
[03:03:45] <cradek> yes
[03:04:34] <SWPLinux> do you think it would "just work" if that were populated with a normal vector calculated from ABC instead?
[03:04:38] <cradek> it would need to be changed to an EmcPose or similar so the normal can be in the other directions
[03:04:45] <cradek> no
[03:04:50] <SWPLinux> ok :)
[03:04:57] <cradek> you'd still have to monkey a lot of layers
[03:05:02] <SWPLinux> hmm
[03:05:12] <stustev> that is bordering on complicated
[03:05:54] <cradek> it's not a 5 minute change but neither is it a 5 week change
[03:06:43] <SWPLinux> it could be 5 weeks for me
[03:06:59] <SWPLinux> maybe 5 days
[03:08:20] <cradek> hm, I forgot you can move ABC during an arc...
[03:08:22] <SWPLinux> hmmm - I think there's an error in one of the comments in that function
[03:08:35] <cradek> comments are notoriously buggy
[03:08:47] <SWPLinux> "mapping of rotation to turns"
[03:09:19] <SWPLinux> then again, maybe not
[03:09:33] <SWPLinux> it seems weird that you'd get one extra turn for negative values
[03:09:50] <SWPLinux> or one less for positive, depending on how you look at it
[03:10:05] <cradek> it's stupid that we translate between so many arc representations
[03:10:14] <SWPLinux> but it is correct, so nevermind
[03:10:26] <cradek> gcode has one, canon has one, posemath has one
[03:10:46] <SWPLinux> you'd think a matrix multiplication or two would get rid of most of those case statements
[03:11:05] <cradek> ... now look at pmCircleInit
[03:11:14] <tomp> drool, wish for arbitrary tilt [0.707106781,0,-0.866025404] . Z is now along this line! ( waves arms in air)
[03:11:36] <SWPLinux> that would be W ;)
[03:12:21] <stustev> tomp: that is already done - emc moves along W
[03:12:53] <tomp> then circular interp on that plane is done?
[03:13:14] <fenn> no, just linear
[03:13:30] <stustev> not circle interp - linear interp - we are talking about circle interp right now
[03:15:07] <stustev> two matrix multiplications will orient the motion to anywhere
[03:16:34] <stustev> the inverse of the product will allow calculation in the three axis planes and then translate back to the tool orientation
[04:03:19] <fenn> any british people awake? what do the letter sequences like H7 F8 E9 mean? http://www.laboratoryformicroenterprise.org/lme/Drawings/2.1.2-1.gif
[04:05:51] <cradek> in the top string, for the dimensions to add up, the 10E9 must be 10
[04:06:11] <fenn> i think it has something to do with iso-2768
[04:07:55] <SWPadnos> http://www.misumi.co.th/products/technical/mold/pdf/pl1121_1122.pdf
[04:11:14] <fenn> so, am I blind or are there actually numbers in the boxes in that pdf?
[04:11:41] <SWPadnos> yes, if you look at the right hand table, there are tolerances in um
[04:12:08] <SWPadnos> but I don't know why there's a grid for the Hx numbers - maybe it's rod/sleeve combinations
[04:13:22] <SWPadnos> http://www.mwtech.demon.co.uk/MWT%20EDD/Useful%20Information/Mechanical%20Informatoin/ISO%20Limits%20And%20Fits/iso%20buttons.htm
[04:13:47] <SWPadnos> click on the various ISO Limits & Fits links
[04:14:40] <fenn> aha, much better, thanks
[04:14:55] <SWPadnos> sure. hopefully you can figure out how to read the damn thing
[04:14:59] <SWPadnos> time for bed. night
[04:15:45] <stustev> SWPadnos: good night - talk to you tomorrow :)
[04:15:53] <SWPadnos> ok :)
[04:18:12] <tomp> h7 is precision ( different than general print precision )
[04:19:42] <tomp> form the gif.. typical on bearing seats/bores, very metric
[04:53:23] <stustev> the same function multiplying two matrices could do the 3D coordinate system rotation
[07:48:16] <mumu> hello
[07:48:39] <Vq^> hiya
[07:50:00] <mumu> i have a problem, i bought a new pc, installed ubuntu 8.04, installed emc....everything worked fine....but now i found out that ubuntu doesn't detect the lpt port
[07:50:23] <mumu> there is no /dev/lp0
[07:52:24] <Vq^> emc doesn't really use the parallel port through the interface file /dev/lp0
[07:52:41] <Vq^> so it doesn't need to detect it
[07:53:18] <Vq^> but if you think that the port is inactive check out the settings in the BIOS configuration
[07:54:26] <mumu> i did check the bios configuration...it is detected there i set it to epp mode
[08:00:35] <mumu> how can i check if it works...i tested a sample configuration (stepper-xyza) and run "hal configuration". It "WATCH" the Xdir signal. It changes its value when i manually move the x axis, but the output on parport pin 2 is not changing (i checked it with the oscilloscope).
[08:03:31] <Vq^> you can set the parallel-port pins with halcmd, might be easier to test with
[08:03:57] <Vq^> are you sure you have configured the right parallel-port address?
[08:04:02] <alex_joni> mumu: look in the BIOS post messages for the parport address
[08:04:30] <alex_joni> on most PCs you can set it from the BIOS
[08:04:48] <alex_joni> make sure it matches 0x378 as stepper-xyza is uding
[08:05:10] <alex_joni> also. watch the Xstep, not Xdir.. if you have a scope
[08:05:27] <alex_joni> the Xdir will always jump back to a default value while not moving I think..
[08:05:46] <mumu> ok, i will restart my pc and check that...will be back, thank you
[08:05:53] <Vq^> alex_joni: i thought it stayed :/
[08:05:56] <mumu> no x dir stays at last value
[08:06:16] <mumu> i see this in hal configuration
[08:08:27] <alex_joni> Vq^: seems I'm wrong.. might have changed since I last tried this.. a couple years ago :D
[08:09:45] <alex_joni> sunfire is fun :)
[08:10:03] <Vq^> alex_joni: do you know if the ecp/epp settings can affect the function with emc?
[08:11:46] <alex_joni> I meant sunspider
[08:11:56] <alex_joni> Vq^: they sometimes certainly do
[08:12:15] <alex_joni> for regulat stepping the most basci mode (Bidirectional) is all there's needed
[08:26:47] <mumu> just to inform you, you were right, the lpt port address was not set to 0x378, now it works fine :)
[08:26:53] <mumu> thank you
[08:40:30] <alex_joni> mumu: great
[08:40:53] <alex_joni> mumu: where are you from?
[08:50:27] <mumu> alex_joni: slovenia
[09:00:33] <alex_joni> mumu: cool, not that far from here :)
[09:03:52] <mumu> Romania, right?
[09:04:06] <alex_joni> right
[13:18:22] <ftaburt> motd
[13:33:39] <Vq^> did someone clone Dallur? :)
[13:34:07] <alex_joni> MPD ?
[13:35:06] <alex_joni> now called DID
[13:43:50] <fragalot> crud. I just bought an L297 ,.... and a 293 instead of a 298 >.>
[13:44:17] <alex_joni> typo?
[13:44:40] <fragalot> :'(
[13:44:55] <fragalot> alex_joni: might need better glasses, lol
[13:44:59] <alex_joni> lol
[13:45:15] <fragalot> or could i somehow use the 293 ... hmm
[13:48:17] <fragalot> looks like a 293 could work too...
[13:54:59] <jepler> [86139.306140] Machine check events logged
[13:55:02] <jepler> I wonder what this means
[13:55:49] <fragalot> that some "events" got "logged"
[13:55:49] <fragalot> :p
[14:56:18] <fragalot> does EMC feed a clock line over the parallel port, or ........... :p
[14:58:03] <alex_joni> jepler: that sounds like a problem
[14:58:08] <alex_joni> http://www.advancedclustering.com/faq/im-getting-mce-machine-check-exception-errors-what-does-this-mean.html
[15:01:06] <fenn_> fenn_ is now known as fenn
[15:01:21] <JymmmEMC> fragalot: for like a charge pump, it can.
[15:03:02] <fragalot> JymmmEMC: cool
[15:03:40] <fragalot> JymmmEMC: I'm making an L297/298 circuit for it,.. but i boo-boo'd, and got a 293 instead,.. trying to figure out if i can do it with the L293B too,.. somewhat doubt it tho as it doesn't have the sens. pins
[15:04:30] <alex_joni> fragalot: you can.. but you have no current limitation
[15:05:00] <fragalot> alex_joni: that was my main concern :p need to find some .. I dunno.. lightbulbs?
[15:05:12] <alex_joni> you need some matched resistors
[15:05:14] <fragalot> or hook 8 .5W resistors in series
[15:05:18] <fragalot> parallel*
[15:05:23] <alex_joni> but you'll get loads more crappy performance
[15:05:28] <fragalot> lol, yes
[15:05:36] <fragalot> meh, i'll just get the 298's on monday
[16:01:30] <archivist_emc> any gotchas about nested loops?
[16:07:44] <archivist_emc> was a tyop
[16:14:51] <alex_joni> nope
[16:15:49] <archivist_ub> now to find next buglet in code....
[16:19:23] <archivist_emc> found :)
[16:21:21] <alex_joni> what code?
[16:22:16] <archivist_emc> writing a clock wheel crossing out program, a special pocketing
[16:23:04] <steve_stallings> steve_stallings is now known as steves_logging
[16:24:17] <archivist_emc> its scaled curves and a rotation of the A axis
[16:27:24] <archivist_emc> another docs tyop in section http://www.linuxcnc.org/docview/html//gcode_main.html#r4_2 "search through and cange numbers in a program file"
[16:31:05] <archivist_emc> http://www.linuxcnc.org/docview/html//gcode_main.html#r5_12_2 L poorly defined
[16:31:30] <BigJohnT> good news my daughters are both safe and without power in Houston :)
[16:33:05] <archivist_emc> been windy down there!
[16:33:28] <BigJohnT> starting to get windy here in swamp east missouri
[16:35:29] <BigJohnT> archivist_emc: is it the L word that is poorly defined?
[16:36:13] <archivist_emc> just says it means orientation but then nothing more
[16:36:54] <archivist_emc> also another docs tyop in section http://www.linuxcnc.org/docview/html//gcode_main.html#r4_2 "search through and cange numbers in a program file"
[16:37:10] <BigJohnT> If I didn't know what it meant it would confuse me... I'll add a link to the cutter orientation section
[16:37:32] <BigJohnT> thanks again
[16:38:25] <archivist_emc> I have a habit of going straight to the gcode to use it not reading preamble, hence I find things like that
[16:39:39] <jepler> alex_joni: thanks for the link
[16:39:59] <jepler> the items in my mcelog look harmless, though
[16:39:59] <jepler> CPU 0 THERMAL EVENT TSC fa75eb6715d
[16:40:00] <jepler> Processor core below trip temperature. Throttling disabled
[16:40:00] <jepler> STATUS 881a00c0 MCGSTATUS 0
[16:40:24] <jepler> (strangely there's not an event to notify that it was enabled, though)
[16:45:10] <alex_joni> jepler: well.. throttling might be something that happens on it's own
[18:48:26] <JymmmEMC> SWPadnos: SWPLinux ping?
[18:49:16] <alex_joni> http://www.newegg.com/Product/Product.aspx?Item=N82E16834220370
[18:52:56] <JymmmEMC> not pad
[18:52:58] <JymmmEMC> not bad
[18:53:11] <tomp> the 701 is too damn small
[18:53:18] <tomp> wish i waited :(
[18:54:14] <JymmmEMC> is that the 8" ?
[18:54:50] <tomp> 7"
[18:54:50] <JymmmEMC> I jsut returned the $400 15" amd64 dual core one yesterday
[18:54:53] <JymmmEMC> ah
[18:54:59] <JymmmEMC> compaq laptop
[18:55:09] <tomp> really, why returned? sounded good
[18:56:02] <JymmmEMC> It was nice for what it is, just 1280 wide and I usually do dual page up on screen, the battery was 2hours, and really could use longer.
[18:56:33] <JymmmEMC> two hours was it sitting on the couch next to me, not any real on-the-road stuff
[18:56:53] <JymmmEMC> got awefully hot too
[18:57:35] <JymmmEMC> and now that I'm seeing laptops with 7-8 hour battery life on them, I would just rather wait.
[18:58:32] <JymmmEMC> I've got way too many systems around here as it is. no need to collect more on a whim, but I have glad that I had a chance to play with it at least, it would be perfect for a first computer
[18:58:39] <JymmmEMC> or a backup spare
[18:59:03] <JymmmEMC> just get rid of vista first =)
[19:00:37] <tomp> i couldnt get a decent w98 on virtualBox ( to use ACAD MDT), but just got W2K up with it on VMserver. now i can CAD 'under' linux :)
[19:01:13] <JymmmEMC> I got XP under vbox going no problem
[19:01:34] <tomp> yeh XP SP2 was fine, but MDT no likey
[19:01:39] <JymmmEMC> ah
[19:02:08] <JymmmEMC> you know you're ging backwards.... a 64b box running a 16b os =)
[19:02:36] <tomp> what it takes for decent CAD (besides lotsa $$ )
[19:02:56] <JymmmEMC> heh, I understand, but was fun to point that out anyway =)
[19:08:31] <alex_joni> tomp: what CAD do you need?
[19:08:40] <alex_joni> 2D, 3D, any special features?
[19:12:32] <tomp> right now laying out Z & W drill slides and drill guide slides for (yet to be done) 7 axis edm drill (XYZWABC)
[19:12:40] <tomp> solid to print capability is nice
[19:13:10] <tomp> so extrude 3d, get print section automagicly is being used on <DT
[19:13:18] <tomp> MDT
[19:14:11] <tomp> btw: the LM97 monorail slides are nice
[19:15:08] <tomp> and MDT power pack drag n drop bolts threads pins is really a time saver
[19:16:53] <alex_joni> tomp: try alibre xpress
[19:17:20] <tomp> thx, googlin now
[19:17:29] <alex_joni> only doze.. but free
[19:19:26] <tomp> ah, was lookin for platform. thx
[19:22:02] <tomp> i got enuf invested in MDT to keep it rather than learn new stuff ( as one of the older dogs, i can plead that case ;)
[20:51:36] <alex_joni> good night all
[20:53:32] <micges> good night alex
[20:59:16] <toastydeath> i just spent 240 on tea and tea related supplied
[20:59:20] <toastydeath> *supplies
[20:59:28] <toastydeath> american dollars
[21:03:54] <tomp> gnite alex
[21:04:25] <tomp> ali shan green buds?
[21:14:05] <toastydeath> tomp: i don't know the country of origin
[21:14:19] <toastydeath> i bought a ton of samplers and a tea brewing thing
[21:14:32] <toastydeath> and a pound of the tea i know i like
[21:17:33] <tomp> good on you! a pound, wow!
[21:28:10] <toastydeath> i can't tell if you are making fun of me =(
[21:28:35] <toastydeath> but it is good tea, it's a jasmine oolong :D
[21:28:53] <toastydeath> http://www.adagio.com/oolong/dragon_pearl.html?SID=5bd2f8448e7b6c70bab5b4dd47e7f923
[21:29:05] <toastydeath> i rly like adagio teas, it is like the amazon.com of tea
[21:29:32] <toastydeath> and it just asked me if i want to set up an automatic shipment since i order frequently =(
[21:29:45] <toastydeath> i think that means i need more friends
[21:30:35] <tomp> go to a tea house ( like coffee house ) ( i should talk workin 7days a week ;)
[21:31:05] <toastydeath> hahah
[21:31:06] <tomp> http://www.teafromtaiwan.com/Alishan_area.htm
[21:31:33] <toastydeath> neat!
[22:03:06] <mshaver> Anyone using the 5i20 (or other Mesa board) for steppers?
[22:04:05] <mshaver> Anyone got a stepper machine running with hostmot2?
[22:07:56] <mshaver> Don't all start typing at once!
[22:08:24] <SWPadnos> I don't know of anyone who's actually running a stepper machine with hostmot2
[22:08:41] <mshaver> OK, that's good to know....
[22:08:46] <SWPadnos> it's "supposed to work" , though there may still be some bugs lurking
[22:09:43] <mshaver> Just to start testing hardware it would be OK to just have 72 gpio pins.
[22:10:16] <mshaver> I'm sure my io map doesn't correspond to any of the pre-built .BIT files.
[22:10:17] <SWPadnos> you won't be able to run steppers in the same way as you would with parport
[22:10:24] <SWPadnos> the update cycle takes too long
[22:10:36] <SWPadnos> so you can't use the GPIO in the fast thread
[22:10:38] <mshaver> You mean I couldn't just toggle the bits?
[22:10:49] <SWPadnos> not anywhere near as fast, AFAIK
[22:11:02] <mshaver> wow
[22:11:53] <mshaver> well, that's somewhat discouraging
[22:12:18] <SWPadnos> heh
[22:12:36] <SWPadnos> the better thing to do is see if the stepper config works, and submit bug reports if it doesn't ...
[22:12:46] <SWPadnos> you know - the whole open source thing ;)
[22:12:55] <mshaver> I think I'll eat dinner, ruminate (pun), and then do it the right way!
[22:13:05] <SWPadnos> rum-i-ate ;)
[22:13:21] <mshaver> yea, I guess (kicks up dirt)...
[22:13:27] <SWPadnos> sounds good to me. maybe it's time for me to stop tearing up a retaining wall and eat something myself
[22:14:02] <mshaver> I'll try back here in about an hour or so & report my (non)progress!
[22:14:07] <SWPadnos> heh
[22:14:34] <SWPadnos> fel free to ask questions and/or email the list. seb_kuzminsky is around from time to time, especially when there are problems with hostmot2 :)
[22:15:24] <mshaver> The real problem is: I need to move an axis by 9:00am tomorrow morning in order to please management (dog & pony show)...
[22:15:31] <SWPadnos> he
[22:15:32] <SWPadnos> h
[22:15:48] <mshaver> thanks for your support!
[22:15:50] <SWPadnos> I don't know of any specific problems with it, so it may "just work"
[22:16:01] <SWPadnos> any time! (if only it were always this easy)
[22:16:04] <mshaver> I'll let you know!
[22:16:08] <SWPadnos> ok
[22:27:01] <JymmmEMC> What? You don't have an EASY button? http://blog.tmcnet.com/blog/tom-keating/images/easy-button.jpg
[22:51:18] <fenn> jymmm going to use that as your e-stop? (e is for easy)
[22:59:32] <JymmmEMC> fenn++
[22:59:33] <JymmmEMC> lol
[23:41:39] <KimK_> mshaver: I am interested in the Mesa boards and hostmot2 (mostly for servos, but maybe steppers too) but I don't have anything running yet
[23:47:51] <cradek> I think according to the mailing list, a few people are running steppers on mesa
[23:47:57] <cradek> I haven't watched closely though
[23:50:47] <skunkworks_> cradek: did you have the balls to fix your lathe?
[23:51:01] <skunkworks_> well - that sounds odd..
[23:55:08] <cradek> ha
[23:55:23] <cradek> I had the ... guts to take apart the ballnut
[23:55:49] <toastydeath> brass cohones of the incas
[23:55:52] <cradek> I have not ordered new balls yet.
[23:56:21] <skunkworks_> are you going with the same size?
[23:56:24] <cradek> I will likely need to try a lot of sizes to get it right. I am not really looking forward to that. I assume they will be a bit expensive.
[23:56:31] <cradek> no they will have to be quite a bit bigger.
[23:56:37] <skunkworks_> wow
[23:56:41] <skunkworks_> scary