#emc | Logs for 2007-03-09

[00:03:31] <a-l-p-h-a> hey, what's the cheapie stepper drivers called? xylotex?
[00:03:31] <a-l-p-h-a> or something like that?
[00:03:31] <a-l-p-h-a> what's their website...
[00:08:06] <cradek> finally timezone fixes for ubuntu...
[00:15:47] <ds3> on a barely fast enough machine, is having software opengl a safe thing wrt to EMC?
[00:47:18] <jepler> ds3: it will be slow, but it shold not cause the dreaded "unexpected realtime delays".
[00:47:31] <jepler> ds3: if you're on really constrained hardware you may find it's just too unresponsive
[00:54:00] <skunkworksemc> jepler: the 1024 line encoder seems to spin just fine at 6500 rpm without losing it.
[00:54:25] <jepler> skunkworksemc: whee, how many edges per second is that?
[00:54:28] <skunkworksemc> (didn't want to try any faster ;))
[00:54:46] <jepler> 443733.33 Hz? whew
[00:55:11] <skunkworksemc> 443733
[00:55:13] <skunkworksemc> right
[00:56:17] <jepler> a bit more than software could do
[00:56:34] <skunkworksemc> a bit ;)
[01:29:53] <Jymmmm> yo
[01:37:30] <a-l-p-h-a> Jymmmm, what were those drivers you have called? who are they made by?
[01:42:21] <fremder> er -- I've got a kinematics question re emc 2.1.1: I have a 3axis stepper type machine but (since I built it) the x and y axes are not perpendicular. The docs mention that kins can correct this. I looked at the source and the mail list archive and didn't see an implementation. I can knock one out (yay, simple geometry) but don't want to reinvent the wheel.
[01:42:27] <a-l-p-h-a> http://www.xylotex.com/
[01:43:33] <jmkasunich> fremder: how bad out of square is it?
[01:44:09] <fremder> not too bad, but more than I can correct mechanically. I'd guess about 1deg
[01:45:40] <jmkasunich> is the error uniform?
[01:46:01] <jmkasunich> in other words, the axes are straight, just not at 90 degrees to each other
[01:46:27] <fremder> jmkasunich: as far as I can measure, it is uniform
[01:46:51] <Jymmmm> a-l-p-h-a: Xylotex?
[01:47:00] <jmkasunich> so what you really need to do is for every inch of travel along X, add or subtract some small amount from the Y coordinate, right?
[01:47:12] <jmkasunich> like 0.008 per inch, or whatever
[01:47:17] <fremder> yep -- that would do it
[01:47:37] <jmkasunich> do you have home switches?
[01:47:49] <fremder> not yet (they're in the mail)
[01:47:58] <jmkasunich> ok, you will
[01:48:27] <jmkasunich> what I have in mind wouldn't work if you planned to just touch off on a part and zero the machine there - it needs a consistent home
[01:48:41] <jmkasunich> are you comfortable working with HAL? read the HAL manual yet?
[01:49:04] <fremder> yeah - I read the Hal manual, I think I can puzzle my way through most things.
[01:49:08] <jmkasunich> ok
[01:49:24] <jmkasunich> if you start your emc, and run halcmd show pin axis
[01:49:28] <jmkasunich> you will get a list of pins
[01:49:39] <jmkasunich> one of them is the axis position, not sure what it is at the moment
[01:50:13] <fremder> sure, I can find it
[01:50:28] <jmkasunich> the X axis position could be scaled down and added to the Y axis command
[01:51:04] <jmkasunich> are you on your EMC computer now?
[01:51:29] <fremder> I'm afraid not -- I'm shirking my responsibilities at work right now.
[01:51:33] <jmkasunich> ok
[01:51:38] <fremder> I see where you're going with that, though
[01:51:42] <jmkasunich> when you get home ;-) do "man offset"
[01:51:51] <jmkasunich> offset is a component that can do the summing
[01:52:11] <jmkasunich> and also removes the offset from the feedback so it doesn't throw EMC off
[01:52:44] <jmkasunich> "man scale" will tell you about a block that can be used to scale down the X position before you add it to Y
[01:53:34] <jmkasunich> if you had large angles you'd probably want to use kinematics, but for small corrections the HAL trick will work fine, and is easy
[01:54:13] <jmkasunich> what kind of machine is this? router, mill?
[01:54:37] <fremder> but you said it is limited to using a consistent home? I don't get that, the offsets should be linear no matter what the origin is.
[01:54:50] <fremder> its mostly a router.
[01:55:16] <jmkasunich> yeah, but if one day the middle of the table is X=10, you'll be adding 0.100 there (assuming the scale is 0.01 per inch)
[01:55:38] <jmkasunich> another day you home in a differnt place, the middle of the table is 13, and you are adding 0.130 there
[01:56:02] <jmkasunich> it will still be square, but off by 0.030 from day to day
[01:56:12] <jmkasunich> of course, if you are just touching off, I guess that doesn't matter
[01:56:30] <jmkasunich> so forget I said that
[01:56:31] <jmkasunich> ;-)
[01:56:46] <Jymmmm> forget you said what?
[01:56:58] <jmkasunich> do you have a nice big accurate square? do you have a dial indicator?
[01:57:25] <fremder> got a dial indicator. My big square is cheap so I don't trust it too much.
[01:57:41] <jmkasunich> I'd clamp the square down to the table, attach the indicator where the router goes, and get one leg of the square parallel to X
[01:58:02] <jmkasunich> just jog x back and forth with the indicator dragging on the square
[01:58:21] <jmkasunich> then lock the square down tight, and move the indicator to the other leg
[01:58:32] <jmkasunich> that will tell you how much per inch you need to tweak it
[01:58:43] <jmkasunich> and it will tell you when you have it right
[01:59:10] <fremder> yeah, good idea.
[01:59:26] <jmkasunich> I hope you don't have much backlash
[01:59:47] <fremder> I used pretty good hardware, so there's not much backlash
[01:59:52] <jmkasunich> because when you make a long move one way, the correction will be making small moves the other way to correct the out of square
[01:59:59] <jmkasunich> lash could make a mess of those small moves
[02:03:32] <fremder> thanks for the help. I may still bang up that kinematics module, just to try it out. Do you think the project is interested in having it?
[02:03:50] <Jymmmm> Is there anythign I could read up on for routing/maching where the finished piece does even have swirl marks?
[02:04:00] <Jymmmm> swirl == tooling marks
[02:04:01] <jmkasunich> we're alway interested in good code
[02:08:14] <Jymmmm> * Jymmmm nudges jmkasunich
[02:11:13] <jmkasunich> waddaya want
[02:11:44] <Jymmmm> To take over the world in 15 minutes
[02:12:40] <Jymmmm> jmkasunich I need to learn to machine with no tooling marks, any suggestions on what to read?
[02:12:49] <jmkasunich> no
[02:12:53] <jmkasunich> what are you machining?
[02:13:00] <jmkasunich> steel, plastic, wood?
[02:13:40] <Jymmmm> jmkasunich: I want to do some vacuum forming of plastics. So I was considering making the dies from machinable wax.
[02:14:06] <Jymmmm> (disposable dies that is)
[02:14:11] <jmkasunich> doesn't vacuum forming involve getting the plastic hot?
[02:14:29] <jmkasunich> * jmkasunich sees melting wax forms
[02:14:54] <skunkworksemc> * skunkworksemc does also
[02:15:22] <Jymmmm> I'm hoping machinable wax needs higher temps (not parafin wax), as it is designined to be machined and would heat up
[02:15:24] <skunkworksemc> * skunkworksemc is cutting a sign on the gantry while irc'ing ;)
[02:17:48] <Jymmmm> jmkasunich http://www.use-enco.com/CGI/INPDFF?PMPAGE=683&PARTPG=INLMK3&PMITEM=505-3627
[02:18:35] <jmkasunich> well, I guess it depends on how hot the plastic needs to be
[02:18:55] <jmkasunich> as for machine marks, you could probably flame polish after machining
[02:19:40] <Jymmmm> Is there a way to even avoid having todo that?
[02:19:46] <jmkasunich> no
[02:19:50] <jmkasunich> machining makes marks
[02:19:54] <jmkasunich> its the nature of the beast
[02:19:59] <Jymmmm> ok, lets say al
[02:20:07] <jmkasunich> its the nature of the beast
[02:20:14] <jmkasunich> machining makes marks
[02:20:41] <jmkasunich> sanding, grinding, polishing are the usuall ways to get rid of the marks
[02:20:58] <jmkasunich> some machining operations make more marks than others
[02:21:05] <jmkasunich> and some marks are more obvious than others
[02:21:06] <Jymmmm> what about something like cyc walls of an engine block?
[02:21:12] <jmkasunich> honed usually
[02:21:48] <Jymmmm> Ok, so anytime I've ever seen something that's been machined, it's NEVER come off the cutting machine "clean"?
[02:22:08] <jmkasunich> depends on how you define clean
[02:22:28] <Jymmmm> Well, not mirror finish, but no tooling marks.
[02:22:40] <jmkasunich> lol
[02:22:46] <Jymmmm> machinable wax melting point = 226 F
[02:23:07] <jmkasunich> if its not a mirror finish, anything less is a tooling mark
[02:23:24] <jmkasunich> thats not very hot
[02:23:30] <jmkasunich> barely over boiling water
[02:23:57] <Jymmmm> to reclaim the max, you have to heat it to 290F
[02:24:00] <Jymmmm> wax
[02:24:14] <Jymmmm> http://www.freemansupply.com/instruction/Reclaiming.pdf
[02:28:40] <a-l-p-h-a> 290F, is less than my what I use to make toast. :D
[02:37:27] <ejholmgren> anyne seen the unix rap?
[02:42:33] <ejholmgren> http://www.youtube.com/watch?v=Fow7iUaKrq4
[02:43:24] <ejholmgren> kill dash nine baby!
[02:43:35] <a-l-p-h-a> a-l-p-h-a is now known as kill-9
[02:43:42] <kill-9> XD
[02:44:17] <ejholmgren> ohnoes!
[02:44:44] <kill-9> I'm a gigabit lan!
[02:44:45] <kill-9> hahaha
[02:44:50] <kill-9> LOL
[02:44:56] <kill-9> <--
[02:44:56] <kill-9> <--
[02:45:01] <kill-9> <--
[02:45:06] <kill-9> kill dash nine babe!
[02:46:10] <kill-9> kill-9 is now known as a-l-p-h-a
[03:40:22] <tomp> hello
[04:11:52] <SWPadnos> hi
[04:17:36] <ejholmgren> hi
[09:40:32] <anonimasu> I just got the toech for the table..
[09:40:51] <anonimasu> torch..
[09:41:04] <anonimasu> it'll cut up to 300mm
[09:51:59] <Dallur> nice, :D
[09:52:12] <Dallur> anonimasu: we want pictures :D
[09:52:32] <anonimasu> Dallur: also got valves and the 2nd driv..
[09:52:35] <anonimasu> drive..
[09:52:49] <Dallur> anonimasu: working on it over the weekend ?
[09:52:49] <anonimasu> Dallur: when it's all painted and nice ,)
[09:52:51] <anonimasu> yeah
[09:53:01] <anonimasu> going to try and get it all togther so we can start cutting parts at monday
[09:53:37] <Dallur> anonimasu: I'll be working on my own table, testing plasma tonight, all movements seem to work great
[09:53:48] <anonimasu> got like 10 parts that weigh 90 kg I need drilled and reamed at the mill over the weekend.. also :)
[09:53:57] <Dallur> anonimasu: :D
[09:54:00] <jlmjvm> good morning
[09:54:01] <anonimasu> so I'll have the mill running while working on the table..
[09:54:12] <Dallur> morning jlmjvm
[09:54:19] <anonimasu> need to get the mill free so I can machine the rest of my torch mount :)
[09:55:25] <anonimasu> Dallur: I hate parts I cant lift..
[09:55:29] <anonimasu> clamping them down scared me as hell
[09:55:41] <jlmjvm> do yall know how to get involved in the emc discussions
[09:55:56] <jlmjvm> that are emailed every day
[09:56:12] <Dallur> jlmjvm: you need to subscribe to the mailing list
[09:56:12] <anonimasu> join the list at sourceforge
[09:56:36] <jlmjvm> k,will check on that
[09:56:51] <anonimasu> Dallur: feels like if a clamp breaks you'll die
[09:57:01] <jlmjvm> i noticed they were talking about a jog wheel
[09:58:43] <jlmjvm> this is a great jog wheel
[09:58:54] <jlmjvm> http://www.contourdesign.com/shuttlepro/
[09:59:15] <Dallur> anonimasu: I don't think I have ever machined anything that big but I bet it's pretty intense :D
[10:01:07] <jlmjvm> http://www.contourdesign.com/shuttlepro/spv2_combo_image.htm
[10:03:02] <anonimasu> it's not much machining
[10:03:09] <anonimasu> it's drill and ream..
[10:03:12] <anonimasu> drill 24 ream 25
[10:03:26] <anonimasu> err 44 and 45
[10:03:32] <anonimasu> or was it 24 and 25..
[10:03:33] <anonimasu> I cant remember
[10:04:05] <anonimasu> err it was 24..
[10:04:17] <anonimasu> but the parts are big :/
[10:05:33] <anonimasu> brb
[10:05:36] <anonimasu> luncyh
[10:05:40] <anonimasu> :)
[10:05:43] <anonimasu> Dallur: nice that your machine works
[10:05:54] <anonimasu> Dallur: it's a nice feeling
[10:07:15] <Dallur> anonimasu: I'm not opening the bottle of champagne until I am able to cut with HF on without any problems, got my fingers crossed
[10:17:43] <alex_joni> hello
[10:18:12] <Dallur> hey alex
[10:18:54] <Dallur> alex_joni: I'm hoping to give your gantry homing a try this weekend
[10:19:59] <alex_joni> not mine
[10:21:33] <Dallur> alex_joni: hmm I though you did the gantry homing, probably cradek then
[10:21:40] <alex_joni> jepler did
[10:23:17] <Dallur> alex_joni: hmmm could have been something else you did which I was also going to test (I have the memory span equal to that of mold)
[10:23:34] <alex_joni> well.. I guess it depends on the mold
[10:25:34] <Dallur> alex_joni: green is better than red, and red is better than black, but then again mold ain't that great imho
[10:45:31] <alex_joni> http://en.wikipedia.org/wiki/Image:Blue_Stilton_Quarter_Front.jpg
[10:52:07] <anonimasu> Dallur: hf isnt a big problem
[10:52:13] <anonimasu> Dallur: seriously
[10:52:38] <Dallur> anonimasu: last time I caused me no end of problems
[10:52:43] <anonimasu> Dallur: my machine ran without missing steps or drifting..
[10:52:50] <anonimasu> that's with a good ground to my motor cabling..
[10:53:00] <anonimasu> and the pc a bit away from the plasma..
[10:53:06] <anonimasu> and drives very close to the pc..
[10:53:20] <Dallur> anonimasu: I had issues with comp freezing up, drivers going offline
[10:53:22] <anonimasu> :/
[10:53:23] <anonimasu> argh..
[10:53:31] <anonimasu> that sucks..
[10:53:37] <anonimasu> though im using a diff supply for the comp..
[10:53:39] <Dallur> anonimasu: but this time around I rewired everything
[10:54:02] <anonimasu> Dallur: odd.. it didnt cause me anything at all
[10:54:03] <anonimasu> :)
[10:54:27] <Dallur> anonimasu: might also be related to the fact that HF in thermal dynamics is 20.000V
[10:54:35] <Dallur> anonimasu: Which I think is a big on the high side
[10:55:17] <anonimasu> I have no idea how high the hf goes on this
[10:57:31] <anonimasu> :)
[10:57:46] <anonimasu> Dallur: well I guess somone's got to be lucky..
[10:57:54] <anonimasu> it's like running canbus without termination.
[10:58:00] <Dallur> anonimasu: :D
[10:58:04] <anonimasu> it works great..
[10:58:14] <anonimasu> until you cut 2dm off your cabling.
[10:58:31] <anonimasu> ^_^
[10:58:38] <alex_joni> Dallur: I think any HF is above 10kV
[10:59:46] <anonimasu> hm
[10:59:53] <Dallur> i'm pretty sure the problem in my case had to do with ground loops
[11:00:30] <Dallur> since my table is all metalic the motors were being grounded in the driver box and through the motor mounts
[11:01:04] <anonimasu> oh, didnt you have separate grounds for your motors?
[11:01:20] <anonimasu> or just the ground loops..
[11:01:27] <anonimasu> hm, I should have been having that too..
[11:03:37] <anonimasu> a2ah
[11:03:47] <anonimasu> Zip 3.0 is insulated.. apparently
[11:04:23] <anonimasu> atleast the torch..
[11:05:05] <anonimasu> but it may be a marketing gimmick.
[13:49:15] <CIA-6> 03jepler 07TRUNK * 10emc2/src/hal/user_comps/hal_input.py: add #! line so it's easier to copy this file into place
[14:14:14] <xemet> hi
[14:16:11] <xemet> jepler: do you remember the functions needed to calc the nurbs, yout old me put them in another file, add it to LIBRS274SRCS in src/emc/rs274ngc/Submakefile and then they will be available to anything using librs274...question: How do I know if a file uses librs274?
[14:36:31] <skunkworks> touchoff does work nice. Good work
[14:39:34] <jepler> xemet: anything that is implementing the canon calls like STRAIGHT_FEED or NURBS_xxx will luse librs274
[14:40:17] <xemet> ok
[14:40:40] <jepler> xemet: in src/emc/sai/Submakefile you see that bin/rs274, the standalone interpreter, is using librs274 by looking at its rule, which begins
[14:40:43] <jepler> ../bin/rs274: $(call TOOBJS, $(SAISRCS)) ../lib/librs274.so.0 ../lib/libemcini.so.0
[14:40:54] <jepler> rs274 is made from the things nmaed in SAISRCS, librs274, and libemcini
[14:41:27] <jepler> named
[14:41:43] <xemet> ok, understood
[14:42:09] <xemet> I have to try
[14:42:50] <xemet> jepler: question on the fly, what should I use in c++ for abs()?
[14:42:56] <SWPadnos> http://youtube.com/watch?v=nuzRsXKm_NQ
[14:43:00] <SWPadnos> fabs?
[14:43:16] <xemet> I tried abs but it says that I'm passing a double to a int function
[14:43:20] <SWPadnos> it
[14:43:22] <SWPadnos> it'
[14:43:24] <SWPadnos> gah
[14:43:24] <xemet> fabs is for double?
[14:43:28] <SWPadnos> it's fabs for floats
[14:43:30] <xemet> ok
[14:43:35] <xemet> :) thanks!
[14:43:36] <SWPadnos> or doubles, I guess
[14:43:55] <SWPadnos> you'd think they'd overload it in C++, but it's probably a compatibility issue or something
[14:49:56] <jepler> std::abs() is overloaded for other data types, but abs is not (for compatability with C, presumably)
[14:50:11] <SWPadnos> right - no mangled names needed :)
[14:50:21] <xemet> uhm...used fabs and everything works...
[14:50:59] <SWPadnos> actually, it may matter whether you include <math.h> or <math>
[14:51:17] <xemet> jepler: maybe I could include in estern file also the arc and biarc functions
[14:52:05] <jepler> xemet: if you believe they might be used elsewhere -- but I don't think the standalone interpreter or the gcodemodule will use them.
[14:53:06] <xemet> uhm...yes, now that I think about it, you have not used arcs and biarcs for the AXIS preview
[14:53:19] <xemet> just straight line, so no need for arcs and biarcs, you're right
[14:58:30] <skunkworks> SWPadnos: nice video.
[14:58:39] <SWPadnos> yeah - cool stuff
[14:59:22] <skunkworks> <microsoft>there is no way you can boot an os in 2mb</microsoft>
[14:59:41] <SWPadnos> <microsoft>there is no way you can boot an os in 8 sec</microsoft>
[15:00:06] <cradek> my junk laptops that I use as X servers boot from a floppy
[15:00:17] <cradek> err X stations
[15:00:19] <SWPadnos> heh
[15:00:30] <SWPadnos> I did that for a while with the firewall at my last company
[15:00:39] <SWPadnos> still rand from HD, but had to boot from floppy
[15:00:53] <SWPadnos> it definitely took longer than 8 seconds from powerup though
[15:00:59] <cradek> no I mean only a floppy, they have no hard disks, cds, batteries
[15:01:00] <skunkworks> SWPadnos: did you ever get the live cd working on a usb drive?
[15:01:23] <SWPadnos> skunkworks, no - I haven't even looked at that since I first mentioned it
[15:01:33] <SWPadnos> I think I went on vacation or something ;)
[15:01:39] <skunkworks> loser
[15:01:48] <SWPadnos> at least I had a vacation :P
[15:02:16] <SWPadnos> and now I have this: http://www.wunderground.com/weatherstation/WXDailyHistory.asp?ID=KVTESSEX1
[15:02:53] <skunkworks> we should be in the 50's by next week :)
[15:03:05] <SWPadnos> heh - we should go from ~-30 last night to 40 tomorrow
[15:03:15] <SWPadnos> global warming! global warming!
[15:03:18] <a-l-p-h-a> SWPadnos, do you get enough sleep?
[15:03:22] <SWPadnos> rarely
[15:03:24] <SWPadnos> you?
[15:04:30] <SWPadnos> you know - it figures that none of my motherboards would support LinuxBIOS
[15:04:38] <SWPadnos> err - be supported by
[15:05:01] <SWPadnos> skunkworks, oh yeah - that was 2 mB, not 2mb. still impressive though
[15:05:20] <skunkworks> right
[15:20:19] <xemet> jepler: great it worked.
[15:22:01] <xemet> jepler: just another question...I've added the functions prototypes in canon.hh because I'm sure it is shared between emccanon.cc and gcodemodule.cc, you told me to add them in rs274ngc.hh, is it shared between the two files?
[15:22:35] <SWPadnos> did you add canon functions or interpreter functions? (or both)
[15:22:49] <SWPadnos> I guess it would be both, huh
[15:23:03] <alex_joni> lol
[15:23:05] <alex_joni> http://youtube.com/watch?v=nuzRsXKm_NQ
[15:23:12] <alex_joni> things are getting even more useless
[15:23:28] <jepler> xemet: I believe it is
[15:23:41] <SWPadnos> I don't like the audio, but I like the fast boot :)
[15:23:44] <CIA-6> 03cradek 07v2_1_branch * 10emc2/src/emc/rs274ngc/ (interp_arc.cc interp_convert.cc rs274ngc.hh): backport single-arc semicircular comp entry fix
[15:24:57] <xemet> uhm, looking at this CRAZY thing http://www.mattshaver.com/include_graphs/emc2/src/emc/rs274ngc/gcodemodule.cc.png seems like the two files don't share rs274ngc.hh
[15:25:24] <xemet> http://www.mattshaver.com/include_graphs/emc2/src/emc/task/emccanon.cc.png or I cannot see it...
[15:27:59] <xemet> SWPadnos: those functions are needed to calculate nurbs points, nurbs points are calculated by the canon function NURBS_FEED, but I need them also for the gcode module in order to show the preview of the path in AXIS.
[15:28:17] <SWPadnos> ok
[15:28:24] <alex_joni> xemet: those two files only care about canon.hh
[15:28:27] <alex_joni> that is the canonical interface
[15:28:28] <a-l-p-h-a> SWPadnos, I get enough sleep... sometimes.
[15:28:32] <a-l-p-h-a> hey alex_joni.
[15:28:36] <a-l-p-h-a> alex_joni, you use django?
[15:28:39] <SWPadnos> canon shouldn't need the rs274ngc.hh include because it doesn't deal with G-code
[15:28:48] <alex_joni> that's the separation point where you could ("theoretically") plug in another interpreter
[15:29:06] <xemet> ok, so I add them in canon.hh, and everything works.
[15:29:07] <SWPadnos> both modules (canon and gocdemodule) include canon.hh
[15:29:13] <SWPadnos> seems that way to me
[15:29:36] <xemet> ok, infact at the moment I've them in canon.hh and everything seems to work
[15:29:37] <alex_joni> xemet: yes
[15:29:42] <SWPadnos> but the G-code related functions should be in rs274ngc.hh
[15:29:51] <SWPadnos> prototypes that is
[15:29:52] <alex_joni> SWPadnos: if there are any
[15:30:06] <SWPadnos> right
[15:30:09] <xemet> yes...but mine are not related to gcode in any way
[15:30:12] <alex_joni> there are/might be things in canon which aren't in g-code at all
[15:30:16] <SWPadnos> yep
[15:30:28] <alex_joni> xemet: then there's no need to add them to the g-code interpreter rs274ngc
[15:30:30] <SWPadnos> or things in canon that aren't actually there, but are related to gcode
[15:30:33] <jepler> they are convenience functions for finding points and tangents to NURBS, which might come from emc's extended rs274 or from another interpreter
[15:30:44] <SWPadnos> (see the COMP related functions :) )
[15:30:45] <alex_joni> right
[15:30:58] <SWPadnos> ok
[15:31:04] <CIA-6> 03cradek 07v2_1_branch * 10emc2/debian/changelog: document various backported fixes
[15:31:55] <jepler> if you're implementing NURBS_FEED (prototype in canon.hh) you're very likely to use these functions -- so canon.hh seems a good place
[15:32:02] <jepler> probably better than the rs274ngc.hh I suggested at first
[15:32:22] <xemet> ok, thanks
[15:32:50] <xemet> but the file I created is in the rs274 library
[15:33:19] <alex_joni> xemet: the CANON funtions are in emc/task/emccanon.cc
[15:33:29] <xemet> alex, I know
[15:33:45] <alex_joni> but that's not necessarely a better place than rs274
[15:34:05] <alex_joni> oh, you said library.. (need more caffeine)
[15:40:32] <alex_joni> hi plattschnauze
[15:43:03] <xemet> maybe there is no need to put the file I created in the rs274 library...
[15:54:52] <alex_joni> xemet: if it's only in CANON, then I don't see a need
[16:32:36] <pier> jepler: could I use open function in python to open /dev/rtf0?
[16:34:22] <SWPLinux> pier: if you have sufficient rights, you should be able to open it. that doesn't mean that you'll be able to do anything useful with the open file
[16:34:49] <SWPLinux> I haven't looked at the RTAI docs to see what's supposed to be there, so I'm of no help other than that last little bit of wisdom ;)
[16:36:15] <pier> SWPadnos: I was trying to see if python could be used to replace a piece of C code
[16:36:48] <pier> that sends a structure of data to a rt module via FIFO
[16:38:03] <SWPLinux> if all it does is fopen / fprintf (and friends) then I see no reason why not
[16:38:10] <jepler> pier: I have never used realtime FIFOs from python
[16:38:14] <jepler> (at all, actually)
[16:39:15] <pier> jepler: I am reading the tkinter and python tutorial but this latter talks about files
[16:39:34] <pier> not device... I'll try...
[16:39:47] <SWPLinux> they're "device files" ...
[16:40:00] <pier> SWPadnos: yes...
[16:40:33] <pier> sorry SWPLinux
[16:40:46] <SWPLinux> either way
[16:40:57] <SWPLinux> I only hae to move 3 feet to change my name ;)
[16:41:00] <SWPLinux> have
[16:41:08] <lerman> A FIFO (aka a named pipe) can be treated just like any other file (with the exception that you can't seek on them).
[16:41:53] <pier> lerman: ok ... I imagined that but I wanted to make sure
[16:42:02] <CIA-6> 03jepler 07TRUNK * 10emc2/src/ (configure configure.in Makefile Makefile.inc.in depend.sh.in): depend.sh and associated code was only needed for pre-gcc3 systems
[16:42:57] <lerman> There are issues concerning blocking on open. The easiest way to avoid problems is to open them read/write (even if you are only going to read). That way you won't have the open block because there is not writer.
[16:43:17] <lerman> I forget the exact rules (maybe that's why I use the easy way).
[16:43:52] <lerman> 'not writer' s/b 'no writer'
[16:43:56] <pier> lerman: I thought of using different dev for read/write flow
[16:45:13] <lerman> If you need bidirectional communication, use two fifos -- one in each direction. Be careful, though. A fifo has a limited buffer size. It's easy to accidently deadlock yourself.
[16:47:39] <a-l-p-h-a> hey... anyone know how I can word this less harsh?
[16:47:41] <a-l-p-h-a> "Items two and three (2 & 3), will be done immediately when your account balance becomes current."
[16:48:12] <a-l-p-h-a> they owe me $$$$... so I don't want to sound like a terrorist holding things as hostage, until I get paid... but I also don't wnat to do it, until I get paid!
[16:48:51] <SWPadnos> "pay me you bastards, or no code for you!"
[16:49:03] <a-l-p-h-a> SWPadnos, oh... you have no clue how tempted I am.
[16:49:12] <a-l-p-h-a> she wrote me a bad check, for the wrong year... so it's useles.s
[16:49:18] <a-l-p-h-a> she told me to just cross it out, and cash it...
[16:49:20] <a-l-p-h-a> HELL no...
[16:49:26] <SWPadnos> something like "Accounts must be current before additional work can be performed"
[16:49:37] <a-l-p-h-a> SWPadnos, that's GOOD.
[16:50:07] <pier> lerman: yes... but now it's just a struct with 4 step(ul) dir(uchar) pair, enable/disable(char), start/stop(uchar)
[16:50:12] <SWPadnos> "and old invalid checks don't count :P "
[16:51:26] <a-l-p-h-a> SWPadnos,... let me gather some emails up... you'll get a laugh, or high blood pressure if you are me.
[16:51:28] <a-l-p-h-a> sec.
[16:52:06] <SWPadnos> heh - I've seen it. I should compose a similar mail to one customer who owes me $5602.10 (not that it bothers me or anything :) )
[16:52:13] <lerman> Tear the check in half and send it back to her with a stamped, self addressed, priority mail envelop and a note that you'ld be more comfortable with a replacement check.
[16:53:04] <SWPadnos> yeah - that's a better approach. you have two problems: one is that AP made a mistake (hopefully) and sent you invalid payment. The other is that they want additional work to be done.
[16:53:14] <SWPadnos> which you don't want to do until you're paid
[16:53:57] <SWPadnos> man - "Scent of a Mule" is a great song
[16:54:16] <lerman> It should only take a few days to get the replacement check. In the mean time, just ignore the additional work. Unless she is really stupid, she will understand.
[16:56:33] <a-l-p-h-a> I got the replacement check... 2 weeks LATER!
[16:56:42] <a-l-p-h-a> it's in my hand, got it an hour ago from the mailman
[16:56:58] <SWPadnos> then cash it and be done with it
[16:58:52] <lerman> If she is doing it on purpose, she will stiff you on the final check. I would suggest that you make sure that you can afford to not get the final payment. As an alternative, have your credit deparment send a note saying that based on past performance, you are no longer able to extend credit.
[16:59:22] <lerman> All future work will be done on a COD basis.
[16:59:51] <lerman> (Credit Department means that you put on your other hat.)
[17:00:15] <a-l-p-h-a> lerman, :D
[17:00:24] <a-l-p-h-a> lerman, I like that...
[17:00:55] <lerman> My wife, Laura Lerman, uses the name Laura Martin when I need someone with a different hat.
[17:01:07] <xemet> great, now I've also the AXIS preview for nurbs: http://imagebin.org/7561
[17:01:14] <lerman> That way I don't look like a one many company.
[17:01:45] <a-l-p-h-a> lerman, in Canada, we are able to sue based on a "check". So if I have a post dated check, I can sue for that amount.
[17:02:22] <a-l-p-h-a> lerman, and easily win... as I have supporting documentation, showing the work done, and the amounts owed and to be paid to me.
[17:02:34] <lerman> If it's small dollars, you can't afford to sue. If it's big dollars, the lawyer takes a big hunk.
[17:02:52] <a-l-p-h-a> Never ever pay with cash... as there would be no records of the transaction from a third party.
[17:03:11] <a-l-p-h-a> a bank would be a neutral third party.
[17:09:38] <a-l-p-h-a> lerman, small claims court
[17:16:09] <duerz> I have a question concerning the VTI cards. I have the VTI and Io expander card installed on a system. And I am using the VTI sim which came with emc2. Within the vti.motion.hal file as have setup loadrt hal_vti num_chan=4 dio="IIIOOO". I am using western reserve card - with dc input modules installed. I have my ribbon cable plugged into j2 on the expander card. I look in hal show under pins/vti/in-00 - and see nothing.?
[17:17:19] <duerz> The western reserve cards have 5 vdc power and dc com hooked up also
[17:19:24] <duerz> i am forcing 24 vdc on the western reserve card and watching hal show
[17:19:57] <SWPadnos> I'm not sure anyone here has heard ofa "western reserve card"
[17:20:12] <jepler> duerz: the most recent developer on the vti driver is <ejohnson AT aaainc.com> -- maybe you should contact him
[17:20:25] <jepler> I don't believe anyone on the channel owns the vti card
[17:20:29] <jepler> I mean, here on the #emc channel
[17:23:57] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/ini/ (iniaxis.cc initool.cc initraj.cc): decrease number of indirect header file inclusions
[17:23:58] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/iotask/ioControl.cc: decrease number of indirect header file inclusions
[17:24:00] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/nml_intf/ (emc.cc emc.hh emcargs.cc emcglb.h interpl.cc interpl.hh): decrease number of indirect header file inclusions
[17:24:01] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/rs274ngc/ (gcodemodule.cc rs274ngc.hh rs274ngc_pre.cc): decrease number of indirect header file inclusions
[17:24:02] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/task/ (5 files): decrease number of indirect header file inclusions
[17:24:06] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/ (8 files): decrease number of indirect header file inclusions
[17:24:07] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/axis/extensions/emcmodule.cc: decrease number of indirect header file inclusions
[17:24:07] <CIA-6> 03jepler 07TRUNK * 10emc2/src/libnml/rcs/rcs.hh: decrease number of indirect header file inclusions
[17:24:14] <CIA-6> 03jepler 07TRUNK * 10emc2/src/libnml/nml/ (nml_type.hh nml.hh nmlmsg.hh): decrease number of indirect header file inclusions
[17:27:04] <kwajpol> hello
[17:27:11] <jepler> welcome
[17:28:39] <kwajpol> i have a problem with latency test, i removed all hardware except ethened pci card and graphical card, i tried to enebla/dissable acpi, but all i get is four zeros, ad than 1, 2, 3, etc
[17:28:43] <a-l-p-h-a> all hail kwajpol.
[17:29:34] <kwajpol> what can i do, change part by part of hardware?
[17:31:49] <jepler> kwajpol: do you mean you get increasing numbers in the "overruns" column? in that case you will not be able to run emc on that hardware.
[17:32:22] <kwajpol> yes, sory about my english/bad explonation
[17:32:25] <jepler> kwajpol: if you mean it shows 0 in all columns, then a 1 in all columns, etc., then you are running the overrun test incorrectly.
[17:33:46] <kwajpol> no, i am talking only about averrun column, i get 0,0,0,1,2,3,4...
[17:35:11] <jepler> kwajpol: if you have an intel CPU, maybe you need to try the suggestions here: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?FixingDapperSMIIssues
[17:38:11] <kwajpol> will try this, thank you!
[17:53:00] <kwajpol> i tred to apt-get the rtai3.3 source, but i get an error: unable to find a source package for rtai-3.3
[17:54:05] <kwajpol> rtai
[17:54:34] <SWPadnos> kwajpol, what distribution are you using?
[17:54:50] <kwajpol> of linux? ubuntu 6.06
[17:55:06] <SWPadnos> is it from the EMC2 liveCD?
[17:55:10] <kwajpol> no
[17:55:56] <SWPadnos> ok, you should install emc2 from these instructions:
[17:56:24] <SWPadnos> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#On_Ubuntu_5_10_or_6_06_using_precompiled_EMC2_packages
[17:56:48] <SWPadnos> unless you've already done that :)
[17:57:19] <kwajpol> i did that
[17:57:40] <kwajpol> and i alos updated the emc to latest version using automatic update
[17:57:50] <SWPadnos> ok, then you shouldn't need any RTAI source packages for this troubleshooting
[17:57:56] <alex_joni> kwajpol: what kind of errors are you seeing?
[17:58:16] <kwajpol> none, except latency test fails
[17:58:42] <alex_joni> kwajpol: seen this page? http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TroubleShooting
[17:59:03] <alex_joni> for emc2 to run correctly you need overruns to stay at 0
[17:59:05] <kwajpol> did that
[17:59:24] <kwajpol> i removed all the hardware not needed
[17:59:36] <kwajpol> the wondows are opening/closing smoothley
[17:59:40] <alex_joni> do you have onboard graphics ?
[17:59:51] <kwajpol> no
[18:00:23] <alex_joni> what kind of graphics?
[18:01:25] <kwajpol> i think nvidea 32MB....will check...
[18:01:35] <alex_joni> make sure yu are running the OSS driver
[18:01:45] <alex_joni> not the NVIDIA accelerated one
[18:03:39] <kwajpol> OSS?, i am running the default one.
[18:05:06] <kwajpol> the device manager showed me: NV28 [geforce4 ti 4200 agp 8x]
[18:07:13] <kwajpol> i see...i shoild have open sorce software drivers...will check...
[18:08:01] <SWPadnos> look in /etc/X11R6/xorg.conf - if the driver is "nv" you're OK, but if it's "nvidia" then you're not
[18:10:23] <kwajpol> yes, the driver is "nv"
[18:11:13] <kwajpol> (but i looked at /etc/X11/xorg.conf - dont have X11R6)
[18:13:57] <SWPadnos> ok - my memory isn't perfect
[18:14:07] <xemet> jepler: just to let you know (I don't think it is so important), I've a max speed set to 90 mm/min (1.5 in the .ini file), when I put the jog speed slider to the max in AXIS this is what it shows: http://imagebin.org/7551 a lot of 999999
[18:47:24] <kwajpol> what else can i do with my latenc problem...i also don't have the intel motherboard (i have QDI), so SMI (described in wiki) also can't be a problem
[18:49:35] <SWPadnos> I think SMI can be an issue on any motherboard for an Intel-compatible CPU, not just those manufactured by Intel (ie, any PC motherboard)
[18:52:07] <kwajpol> how can i check which version of rtai i have
[18:52:46] <rayh> cat /proc/version
[18:53:27] <SWPadnos> uname -a
[18:57:50] <kwajpol> iname -a
[18:59:00] <kwajpol> why wiki taks about apt-get source rtai-3.3, if i have Linux version 2.6.15-magma (root@ubuntu) (gcc version 4.0.3 (Ubuntu 4.0.3-1ubuntu5)) #1 Fri Jun 9 20:51:19 EEST 2006
[19:01:51] <SWPadnos> the wiki talks about several installation methods, including from source
[19:01:59] <SWPadnos> you only need to follow one of them
[19:05:02] <a-l-p-h-a> SWPadnos, people suck.
[19:05:37] <SWPadnos> indeed
[19:18:58] <alex_joni> http://www.resursadefun.ro/s-a-stricat-trenul.htm
[19:34:57] <CIA-6> 03jmkasunich 07TRUNK * 10emc2/src/hal/components/xor2.comp: added 2 input XOR gate (HAL component)
[19:41:45] <duerz> hello rayh
[19:43:18] <rayh> hi
[19:43:38] <duerz> did you ever get a vti card to work
[19:43:56] <duerz> i should say - the io expander
[19:45:28] <duerz> did you ever get a vti card to work
[19:45:28] <duerz> <duerz> i should say - the io expander
[19:45:49] <rayh> no
[19:46:00] <rayh> i've not tried
[19:46:05] <rayh> phone
[19:46:23] <duerz> I got an email into eric johnson - I cant get hal to see any inputs
[19:47:30] <duerz> you wany my phone #?
[19:47:42] <SWPadnos> no - he's on the phone
[19:49:19] <duerz> just wondering - what kind of support do you give the end-user of this product. phone support - or just this?
[19:49:47] <duerz> i know about the classes
[19:50:52] <cradek> since it's Free/open source software, we're all volunteers, so you get support however we have time for it
[19:51:05] <cradek> I think you'll find us the most responsive on IRC
[19:51:20] <SWPadnos> vigilant should support their hardware
[19:51:22] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: export the currently selected jog increment on a HAL pin
[19:51:32] <cradek> oh maybe I didn't understand the question
[19:51:53] <SWPadnos> I think they have recently gotten a programmer to do updates to the driver, but nothing has been done since that (in the last couple of months)
[19:52:11] <SWPadnos> there has been no functional change to the vti driver in almost a year
[19:52:25] <SWPadnos> and none of the active developers have one (purchased or provided by vigilant)
[19:52:30] <duerz> so it should work
[19:52:35] <SWPadnos> that's not what I said
[19:52:49] <SWPadnos> it should, but we don't know, and can't know, if it does
[19:53:06] <duerz> I am willing to do testing
[19:53:09] <duerz> :)
[19:53:10] <SWPadnos> if it doesn't, you should contact vigilant and tell them that you can't get their hardware to work with the driver
[19:53:16] <duerz> got it all set up
[19:53:33] <SWPadnos> that's good. if you're willing to send one of us a set of cards for development, it would help ;)
[19:53:43] <CIA-6> 03cradek 07TRUNK * 10emc2/docs/man/man9/.cvsignore: new autogenerated manpage
[19:53:54] <duerz> dont think my boss would go for that
[19:54:15] <duerz> hes a tight wad
[19:54:25] <cradek> duerz: what city are you in?
[19:54:29] <SWPadnos> you should contact Vigilant then. we can help with setup anf configuration issues, but not with hardware / driver issues
[19:54:34] <duerz> ann arbor
[19:54:38] <duerz> mi
[19:54:43] <skunkworks> duerz: http://www.cnczone.com/forums/showthread.php?t=32001
[19:54:48] <cradek> ==> PORT ... done. ==> LIST ...
[19:54:53] <cradek> oops
[19:55:07] <SWPadnos> it's not that we're unwilling to help, we just don't have the hardware to do so
[19:57:02] <duerz> interesting skunkworks - i have spoke with jon rowland before - OpenCNC supports this board as well
[19:57:16] <duerz> I currently do retrofits for them
[19:58:39] <duerz> i got a driver for windows
[19:58:52] <duerz> dont think it will work for linux though
[19:59:43] <duerz> i going to give him a call
[20:01:08] <cradek> OpenCNC Starting at $7,000.00
[20:01:10] <cradek> ??
[20:01:30] <robin_sz> coo
[20:01:31] <cradek> I don't see what makes this "open"
[20:01:36] <robin_sz> eh?
[20:01:48] <cradek> oh for 3 axes, it's $9000
[20:01:53] <duerz> yes
[20:02:07] <duerz> 7000 bucks
[20:02:07] <robin_sz> why cant it be open and expensive?
[20:02:26] <duerz> but it works!!;)
[20:02:48] <SWPadnos> if you send me $7000, I'll make it work with EMC ;)
[20:02:52] <duerz> wait till you see 5 and up
[20:03:44] <duerz> i like emc also- just wish I knew more about it
[20:04:22] <duerz> i just integrate the stuff
[20:04:37] <robin_sz> another 12 months and emc will really give those $$$ packages soemting to think about
[20:04:51] <duerz> customers buy it - emerson, big companys
[20:05:03] <duerz> ill bet
[20:05:11] <a-l-p-h-a> robin_sz, $$$... oooooooooh...
[20:05:15] <cradek> I'm not saying there's anything wrong with it, I just don't know what "open" means
[20:05:16] <duerz> thats why my boss has me looking at this
[20:05:22] <a-l-p-h-a> not CAM software, but CNC software
[20:05:40] <a-l-p-h-a> cradek, as opposed to closed.
[20:05:46] <a-l-p-h-a> cradek, or ajar.
[20:05:50] <robin_sz> i guess if you are building a 200,000 dollar mill, a few $ on the control is not an issue. you buy the best control you can find, cost doesnt come into it
[20:06:00] <duerz> OpenCNC uses isagraf - much better for plc- ladder sucks
[20:06:19] <alex_joni> classicladder can do other PLC languages too
[20:06:23] <alex_joni> not only ladder
[20:06:33] <robin_sz> french?
[20:06:41] <duerz> besides sfc - what else?
[20:06:44] <duerz> st?
[20:07:02] <alex_joni> SFC
[20:07:13] <a-l-p-h-a> well... aside from that, there's soon to be femmebot controls from alex_joni. :)
[20:09:57] <robin_sz> I suspect open cnc has lots of neat configuration and integration features, drag and drop gui creators etc and a published api
[20:15:21] <robin_sz> it seems "open" is a buzzword all cnc control manufacturers watn to add to their controls, although not all of them agree on what it shoudl actually mean!
[20:17:03] <robin_sz> I guess thats only because normal cnc controls are so crazily locked down ... like if you nopught a Bosch CNC control wouldnt you expect to get the tools to configure it how you needed it? ... nah, thats several thousand dollars more if you want to be able to set it up
[20:17:22] <duerz> just spoke with jon rowland - he was under the impression that a driver was never wrote for the io expander or indusry pack
[20:17:49] <robin_sz> well, there you go
[20:18:05] <robin_sz> I guess they cant have sold many cards if nothing can talk to it
[20:18:15] <duerz> so eric is in charge of this ?
[20:18:27] <robin_sz> who?
[20:18:35] <duerz> ejohnson?
[20:18:51] <robin_sz> in charge is not a concept I understand
[20:19:08] <duerz> trying to develop?
[20:19:42] <robin_sz> not knowing
[20:19:59] <robin_sz> has your boss ever bought opencnc?
[20:21:13] <duerz> dont know if he ever did- mdsi might have given it to him - he got all the versions. we do retrofits for them.
[20:21:20] <robin_sz> right
[20:21:32] <robin_sz> whatever
[20:22:02] <robin_sz> in business terms, the investment in a VTI card and providing it to a developer to have them make it work is tiny
[20:22:09] <robin_sz> teeny weeny
[20:22:44] <duerz> vigilant products already gave one of your developers 2 cards
[20:22:49] <duerz> 2002
[20:22:56] <robin_sz> they did?
[20:23:01] <robin_sz> whom?
[20:23:15] <duerz> jon is forwarding me an e-mail all about it
[20:23:29] <robin_sz> jon?
[20:23:31] <rayh> Paul got at least one of the cards.
[20:23:34] <duerz> rowland
[20:23:37] <robin_sz> right ...
[20:24:00] <robin_sz> so the cards are with Mr Corner then?
[20:24:08] <SWPadnos> duerz, you're treating us like acorporate interest, wnd we're not
[20:24:14] <SWPadnos> s/wnd/and/
[20:24:23] <rayh> we had a recent flap about that Paul's promises.
[20:24:25] <duerz> a corporate interest?
[20:24:31] <SWPadnos> emc is offered with no support and no warrantee, officially
[20:24:43] <SWPadnos> we do what we can because we're nice
[20:24:48] <duerz> im just making conversation with robin
[20:25:00] <robin_sz> "vigilant products already gave one of your developers 2 cards" .. they are not *OUR* developers ... they are just people ...
[20:25:11] <SWPadnos> it's up to the manufacturer of the hardware to make it work with EMC, whether they do that by writing the driver themselves or by donating equipment to developers
[20:25:32] <SWPadnos> ok - it's a tone I've noticed - not a big deal, but it could be :)
[20:25:50] <robin_sz> I suspect if they already provided paulc with a couple of cards, they presume they have "done their bit"
[20:25:53] <duerz> sorry - i guess i dont know the hierchy in here
[20:26:04] <robin_sz> there is no hierarchy
[20:26:10] <robin_sz> well, not much
[20:26:21] <duerz> whatever
[20:26:44] <SWPadnos> I'm not trying to be rude here - I just want to be sure you understand who's responsible for what
[20:26:58] <robin_sz> * robin_sz nods
[20:27:06] <robin_sz> we aitn responsible for nuttin ;)
[20:27:10] <SWPadnos> heh
[20:27:14] <duerz> i would like to see a driver made for vigilant cards - for emc - thats it. ill just try another io card.
[20:27:43] <duerz> thats why people pay 7000 for openCNC
[20:27:46] <robin_sz> well, maybe paulc did write one and it will appear in his tuxcnc thing ...
[20:27:53] <SWPadnos> Iunderstand your desire, and there is a driver. unfortunately, since none of the current emc developers have any of this hardware, we can't help much on the driver end
[20:28:16] <SWPadnos> so when there's new hardware or some problem with the driver, we can't help
[20:28:21] <SWPadnos> it's not that we don't want to
[20:28:25] <rayh> The only heirar... is if the board chooses to step in.
[20:30:02] <rayh> The company behind the vti spoke recently on another list suggesting that they might set up a test bench with Ubuntu there so that they could continue development of the EMC2 stuff for it.
[20:30:30] <SWPadnos> "I intend to follow up, I have learned the EMC does support some of the features of one of our boards, I would like to see more of our functionality supported, so I am going to ask what help they need and see what I can do to make it happen. Here, I probably need to set up a Linux platform, so that I can test and evaluate their product. Maybe we need to go ahead and do a machine retrofit...
[20:30:33] <SWPadnos> ...ourselves here in our shop, I think that may be the best way to explore what the EMC software can do with our hardware." - J. Rowland
[20:31:11] <SWPadnos> that was one month ago, and I don't know what, if anything, has happened since
[20:32:08] <robin_sz> i thought the emc board had withered away?
[20:32:14] <SWPadnos> harldy
[20:32:16] <SWPadnos> hardly
[20:32:33] <SWPadnos> though we're officially overdue for a new election, it seems
[20:32:39] <SWPadnos> (assuming a 1-year term)
[20:32:48] <robin_sz> ooh, i see tuxcnc has another outpouring of sour grapes for us
[20:32:51] <robin_sz> http://www.tuxcnc.org/pivot/entry.php?id=14#body
[20:32:56] <SWPadnos> excellent
[20:33:59] <CIA-6> 03jepler 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: When the jog increment is "continuous", put 0 on the HAL pin instead of the last-shown incremental jog distance
[20:34:11] <cradek> those posts are fascinating
[20:34:19] <SWPadnos> heh
[20:34:35] <cradek> they especially carry a lot of weight, being a place where nobody else can respond
[20:34:51] <SWPadnos> though the uninstall thing is something I've thought about a bit
[20:35:00] <SWPadnos> make uninstall, that is
[20:35:11] <cradek> yes that might be nice
[20:35:25] <cradek> I'm not sure how complex it should be - should it notice files have been changed manually for instance?
[20:35:29] <robin_sz> the random logfiles thing is a point worth noting too
[20:35:31] <jepler> remember, every individual package should always make sure to implement a shitty packaging system
[20:36:13] <cradek> robin_sz: they're not log files, they're errors that are meant to be sent to the developers
[20:36:40] <SWPadnos> hmm. what is an RA license?
[20:36:54] <cradek> robin_sz: obviously /var/log is the wrong place for that, as a user can't even write there - that suggestion makes me think he doesn't understand something that's going on
[20:37:09] <cradek> robin_sz: (/tmp is a sane possible candidate)
[20:37:12] <robin_sz> cradek : perhaps /tmp then
[20:37:15] <robin_sz> noted
[20:37:36] <xemet> jepler: I don't know if you've seen my message before, look at this image, look at the jog speed
[20:37:36] <xemet> http://imagebin.org/7551
[20:38:21] <xemet> jepler: I've ma max speed setted to 90 mm/min, 1,5 in .ini, when I put the slide at the max in AXIS it shows this value 999999
[20:38:47] <robin_sz> cradek, all you need to know is this: emc releases stuff every day, we ar yet to see a single line of code from the tuxcnc project .. nuff said
[20:39:02] <xemet> jepler: just to let you know, I don't think it is so important
[20:39:05] <jepler> robin_sz: I hear there is an SVN repository somewhere that is not publicized
[20:39:33] <robin_sz> whatever
[20:39:40] <cradek> I don't care about any comparisons or flamewar - I have an active project to help run
[20:40:06] <robin_sz> quite
[20:40:36] <cradek> but it is fascinating that there's much more discussion of emc2 than tuxcnc on tuxcnc.org
[20:40:58] <jepler> xemet: yes, I can reproduce the problem
[20:41:19] <awallin> who runs tuxcnc ?
[20:41:26] <SWPadnos> paul c
[20:41:55] <cradek> it appears everything on that site was written by just him
[20:42:02] <robin_sz> yep ...
[20:42:12] <awallin> ah, has he given up on the BDI releases?
[20:42:13] <robin_sz> as a matter of interest ... where does the "available for study but not for change or use without permission of the author."
[20:42:19] <robin_sz> thing come from
[20:42:31] <cradek> that was an unofficial comment by rayh
[20:42:40] <cradek> we do not currently have any code like that in CVS
[20:42:59] <robin_sz> right
[20:43:29] <robin_sz> difficult to see how we could, what with all that GPL stuff lying about
[20:43:42] <cradek> also I don't know if the quote is exactly right
[20:43:54] <cradek> it's in the list archives somewhere if you care
[20:44:05] <robin_sz> not sure I care enough to look
[20:44:06] <duerz> seems ray got it to work: "The driver for the PCI-ENCDAC board has been running without
[20:44:06] <duerz> problem for about 6 months [more like 10 months now]. Encoders,
[20:44:06] <duerz> analogs and the 8 on-board I/O are thoroughly tested, and
[20:44:06] <duerz> extensively used. This board also supports several expansion
[20:44:06] <duerz> modules. All digital I/O expander modules (meaning those using
[20:44:06] <duerz> the expander interface, not the PCI bus) should be supported and
[20:44:09] <duerz> working. I only have the EXPDIG-128, and only tested it early
[20:44:09] <duerz> on.
[20:44:12] <duerz>
[20:44:14] <duerz> "Expansion DIO support should be working, but it has not been
[20:44:15] <duerz> extensively tested. Only the PCI version of the board is
[20:44:18] <duerz> supported in this driver, as I do not have the ISA version, nor
[20:44:20] <duerz> do I see the need to support it.
[20:44:21] <duerz>
[20:44:24] <duerz> "The VIP-ENCDAC expander module is also not supported at this
[20:44:26] <duerz> time since I do not have that module. However the interface is
[20:44:28] <duerz> quite simple and would only require a couple minor modifications
[20:44:29] <duerz> to support it.
[20:44:31] <duerz>
[20:44:33] <duerz> Hope this helps.
[20:44:35] <duerz> Ray
[20:44:47] <duerz> sorry about the scroll
[20:45:05] <robin_sz> next time, just paste the url
[20:45:30] <duerz> it was a e-mail from Jon rowland
[20:45:53] <SWPadnos> that looks like a message from Ray which quotes someone else
[20:46:26] <duerz> if the driver spports the expander module - I cant get it to work
[20:46:33] <duerz> who's ray?
[20:46:40] <rayh> Yes it was from the comments in the source.
[20:46:54] <rayh> * rayh = ray
[20:47:05] <duerz> hehe
[20:47:07] <SWPadnos> what is the output of halcmd show pin vti ?
[20:47:30] <SWPadnos> also dmesg, which should show messages from the driver (which may include a list of detected hardware)
[20:48:33] <robin_sz> ooh ther you go .. looky .. the tuxcnc svn server :)
[20:49:18] <duerz> SWPadnos: what? dont understand
[20:49:48] <SWPadnos> if you run emc using the vti config, then you can open a terminal window and run the command "halcmd show pin vti"
[20:49:57] <SWPadnos> that will print a list of the pins the driver exports
[20:50:05] <robin_sz> http://svn.gna.org/viewcvs/tuxcnc/
[20:50:24] <LawrenceG> hey... I cant get bdi to run on dos.... I guess there is an operating system dependancy in BDI that needs fixing!
[20:50:32] <SWPadnos> if that list doesn't change when you have the expander card plugged in vs not having it plugged in, then the driver is not detecting the expansion card
[20:50:32] <duerz> have to boot the thing up again - brb
[20:50:51] <CIA-6> 03jepler 07TRUNK * 10emc2/share/axis/tcl/axis.tcl: before this change, a maximum jog speed of 1.5mm/second would show as all 9s on the readout
[20:51:14] <SWPadnos> if it does change (and show the I/Os you expect to have), then the driver is working fine, and you need help with configuration
[20:52:19] <jepler> SWPadnos: the way I read the source code, it relies on the configuration string on the 'loadrt' line to tell it how many DIO are available
[20:52:34] <duerz> i have done that
[20:52:46] <duerz> dio="IIIOOO"
[20:53:00] <SWPadnos> ok. I haven't read the code much
[20:54:49] <duerz> ok I ran halcmd and got a list
[20:55:10] <jepler> put on http://pastebin.ca/
[20:55:34] <duerz> 24 input and 24 outputs
[20:55:52] <SWPadnos> ok. it appears that the driver is doing what you want. where's the problem again? ;)
[20:56:08] <duerz> 4 encoder channels and 4 dac channels
[20:56:30] <duerz> if I force the input high I cant see it in hal watch
[20:57:05] <jepler> are you sure you identified the right hardware pin for the hal pin you are watching?
[20:57:28] <SWPadnos> use IIIIII instead of IIIOOO for the config, and look at all of them with halcmd show vti.in
[20:57:38] <SWPadnos> that will make all pins input and tell you which one changes
[20:57:59] <duerz> k
[20:58:01] <SWPadnos> if none change (when forced to 1 and to 0 - we don't know the default state of the card inputs)
[20:58:11] <SWPadnos> then there's likely a problem with the driver
[20:58:43] <SWPadnos> you should realize that TTL and CMOS inputs (likely what's on the card) default to logic 1 when unconnected
[20:59:04] <SWPadnos> the exact sequence is:
[20:59:07] <SWPadnos> halcmd show pin vti.in
[20:59:17] <SWPadnos> make sure they're all the same (they should be anyway)
[20:59:28] <SWPadnos> connect a pin to +5V
[20:59:30] <SWPadnos> halcmd show pin vti.in
[20:59:34] <SWPadnos> one pin may change state
[20:59:49] <SWPadnos> connect a pin to 0V
[20:59:50] <SWPadnos> halcmd show pin vti.in
[21:00:17] <SWPadnos> one pin should definitely be different from the original state read in the first halcmd show pin
[21:01:34] <rayh> He could also watch for pin changes using the watch tab in halshow.
[21:01:42] <SWPadnos> shhh - too easy ;)
[21:01:50] <rayh> or make a very nice pyvcp display
[21:01:57] <SWPadnos> shhh - too hard
[21:03:02] <rayh> so many ways to do things!
[21:03:55] <SWPadnos> shhh - too confusing
[21:04:53] <robin_sz> the tuxcnc svn has someintersting bits in it
[21:04:59] <Jymmmm> Finally Finished! Just need a couple of latches now
[21:05:09] <robin_sz> rabbit hutch?
[21:05:23] <Jymmmm> Yeah... mounted door last night.
[21:05:32] <robin_sz> sounds painful
[21:05:54] <Jymmmm> Only took me 10 days to build....eeeesh
[21:06:06] <jepler> 10 days to build? sounds like a piece of software I worked on once
[21:06:18] <robin_sz> whewn does the rabbit arrive?
[21:07:02] <Jymmmm> Rabiit is already in the hutch, just need to get some grease to bring it to life again (ran out last night)
[21:07:21] <robin_sz> you grease your rabbits?
[21:07:31] <Jymmmm> Don't you?
[21:07:55] <robin_sz> well, sometimes. but not when people are watching
[21:08:25] <Jymmmm> Well, grease is much cheaper than ruined pices and worse yet, ruined rail trucks.
[21:09:16] <robin_sz> this is the router you boaught as a pre-built thing?
[21:10:33] <Jymmmm> Close to mine, just not as tall... http://www.k2cnc.com/DuportalPro343Access/Pictures/detail.asp?iData=29&iCat=636&iChannel=3&nChannel=Pictures
[21:11:06] <Jymmmm> 5" Z and 24" x 24"
[21:13:13] <duerz> it didnt like all "IIIIIIIIIIIIIIII"
[21:15:06] <Jymmmm> robin_sz: You mean dust like this =) http://www.k2cnc.com/DuportalPro343Access/images/cnc-router-KG-3925-K2cnc.jpg
[21:15:06] <SWPadnos> duerz, what error(s) do you get?
[21:15:06] <SWPadnos> is the EMC computer attached to the internet?
[21:15:06] <duerz> no i can do that though
[21:15:18] <duerz> debug section?
[21:16:01] <SWPadnos> it's easier to post configuration and error information to the web when the machine is connected
[21:16:10] <SWPadnos> http://pastebin.ca is a very nice thing
[21:16:34] <SWPadnos> first, you should run emc from a terminal, not the applications menu
[21:16:52] <SWPadnos> that will provide you with some debug information that's lost from the menu
[21:17:07] <duerz> i have it being launched form desktop "in terminal" is selected
[21:17:31] <SWPadnos> second, you can run dmesg to get "debug messages" from the kernel - the last 20 lines or so of that would be helpful (right after a failed run, or other stuff may get added to the log)
[21:17:31] <jepler> then paste everything that is in the terminal into pastebin.ca
[21:17:43] <SWPadnos> ok - as long as the terminal remains open after a failed run
[21:18:42] <SWPadnos> third, two files will be placed in your home directory - these may also be needed (debug.txt and print.txt, I think they're called)
[21:20:10] <duerz> no such files
[21:20:44] <SWPadnos> in the ~/ directory?
[21:22:53] <Jymmmm> It looks like we lost robin_sz... not sure if this is a good thing or bad. LOL
[21:23:38] <Jymmmm> * Jymmmm pokes SWPadnos for no good reason
[21:23:44] <SWPadnos> I could answer that, but I'm too nice ;)
[21:23:54] <SWPadnos> * SWPadnos ignores Jymmmm
[21:24:13] <duerz> its almost beerthirty - ill keep plugging away at it monday
[21:24:15] <Jymmmm> ok, he's still alive, no need to toss in the grave just yet
[21:24:20] <SWPadnos> in the immortal words of a Peon in Warcraft II, "Stop Poking Meeee"
[21:24:41] <duerz> have a good weekend all
[21:24:42] <Jymmmm> I've never played WOW
[21:24:48] <Jymmmm> see ya duerz
[21:24:56] <SWPadnos> duerz, I may have asked you before, but have you gone through the HAL tutorial?
[21:25:10] <duerz> yes
[21:25:11] <SWPadnos> getting the skills to discover what's available in a loaded HAL is vital
[21:25:13] <duerz> several time
[21:25:14] <SWPadnos> ok
[21:25:19] <SWPadnos> (no pun intended)
[21:27:40] <jlmjvm> SWPadnos:Hello
[21:27:50] <SWPadnos> hi jlmjvm
[21:28:35] <jlmjvm> was looking over email for user group,were you the guy talking about the shuttle express?
[21:28:57] <SWPadnos> I mentioned the PowerMate
[21:29:08] <SWPadnos> someone else mentioned the Shuttle Express
[21:29:34] <SWPadnos> John Prentice
[21:30:02] <SWPadnos> I almost ordered one of each today, but managed to restrain myself ;)
[21:30:20] <jlmjvm> just wanted to say the shuttle brand has detents on the inner wheel,the powermate doesnt
[21:30:33] <SWPadnos> hmmm
[21:30:36] <jlmjvm> the shuttles work great
[21:30:41] <SWPadnos> I thought the powermate had detents
[21:30:58] <jlmjvm> i called and the guy said it didnt
[21:31:04] <SWPadnos> ok
[21:31:12] <jlmjvm> they look really cool
[21:31:13] <SWPadnos> does the shuttle have any indicator lights?
[21:31:21] <Jymmmm> url's?
[21:31:22] <SWPadnos> the powermate can glow blue
[21:31:37] <SWPadnos> google for griffin powermate and/or shuttle pro
[21:31:51] <SWPadnos> http://www.griffintechnology.com/products/powermate/
[21:31:58] <SWPadnos> http://www.contourdesign.com/shuttlepro/shuttlexpress.htm
[21:32:48] <SWPadnos> oops - that's the express
[21:32:50] <jlmjvm> just wanted to say the shuttle pro has removable buttons so it can be labeled easy
[21:32:53] <Jymmmm> it's ok
[21:32:56] <SWPadnos> oh - cool
[21:33:23] <SWPadnos> if you can control lights on/near the buttons, that would be pretty ideal
[21:33:30] <a-l-p-h-a> SWPadnos, your future machine, that's in the process of being built, but at a momentary pause, are you using steppers? or servos?
[21:33:31] <jlmjvm> i took an express to a customer earlier this week
[21:33:40] <SWPadnos> a-l-p-h-a, servos
[21:33:49] <a-l-p-h-a> SWPadnos, what's the drivers you're using? :)
[21:33:51] <a-l-p-h-a> or going to use
[21:33:59] <SWPadnos> jlmjvm, do you do video related stuff?
[21:34:09] <jlmjvm> ?
[21:34:10] <SWPadnos> a-l-p-h-a, I have geckos, but may upgrade before I install them ;)
[21:34:22] <Jymmmm> SWPadnos upgrade to what?
[21:34:50] <SWPadnos> jlmjvm, those jog/shuttles are mostly advetised for video applications - just wondering of you deal with that end or the computer hwardware end ...
[21:34:57] <a-l-p-h-a> Jymmmm, probably something that doens't use step/dir
[21:35:03] <SWPadnos> Jymmmm, unknown as yet. possibly a version of skunkworks driver
[21:35:08] <SWPadnos> yep
[21:35:14] <Jymmmm> ah
[21:35:21] <jlmjvm> naw,no video,just cnc
[21:35:24] <SWPadnos> I think analog control provides a better finish, so I may do some tests
[21:35:35] <a-l-p-h-a> damn. so hungry... pineapple doesn't really fill you up.
[21:35:40] <SWPadnos> jlmjvm, ok - just curious
[21:35:56] <a-l-p-h-a> 3 pine apples for $1.00CDN. So damn cheap... so damn tastey... so damn hungry.
[21:36:16] <SWPadnos> that's what the Tahi / Vietnamese food will be for (as soon as my wife gets home)
[21:36:20] <SWPadnos> Thai, that is
[21:36:32] <a-l-p-h-a> SWPadnos, same here tonight
[21:36:34] <Jymmmm> Basil Pork.... good stuff
[21:36:43] <a-l-p-h-a> SWPadnos, gonna have some osyters there too... :)
[21:37:01] <SWPadnos> Thai Basil anything is generally good, panang, massaman, choo-chee ...
[21:37:02] <a-l-p-h-a> SWPadnos, there's this pineapple fried rice they have... very yumm.
[21:37:05] <SWPadnos> yummy
[21:37:18] <a-l-p-h-a> I'm starving. :|
[21:37:21] <SWPadnos> I do wish there was a Korean restaurant around here
[21:37:28] <jlmjvm> what you gonna do with your geckos if you go with skunkworks
[21:37:37] <a-l-p-h-a> SWPadnos, haha... sucka... you should move to Toronto. :)
[21:37:39] <SWPadnos> probably sell them
[21:37:45] <SWPadnos> too cold there ')
[21:37:47] <SWPadnos> ;)
[21:37:49] <a-l-p-h-a> SWPadnos, I'll give you 20 cents on the dollar. :)
[21:37:56] <a-l-p-h-a> SWPadnos, no colder than buffalo NY.
[21:37:58] <jlmjvm> are they 320,s
[21:38:01] <SWPadnos> yep
[21:38:18] <SWPadnos> I may go to AC servos as well, before my DC ones are installed
[21:38:23] <jlmjvm> those are killer little drives
[21:38:29] <a-l-p-h-a> http://www.theweathernetwork.com/weather/cities/can/Pages/CAON0696.htm
[21:38:31] <SWPadnos> yep - very nice
[21:38:54] <SWPadnos> a-l-p-h-a, it got down near here -30 last night (F, without wind chill)
[21:38:57] <jlmjvm> the last mill i did has them,running 250 ipm rapids
[21:38:57] <a-l-p-h-a> I have an AC stepper driver... well... it takes AC, and controls my oriental motor, 5phase motor.
[21:39:01] <SWPadnos> I'm not sure you guys are having it any colder
[21:39:15] <a-l-p-h-a> what's -30°F in real temperatures?
[21:39:33] <SWPadnos> ~-35 to -36
[21:39:39] <a-l-p-h-a> wow. that's freak'n cold...
[21:39:41] <SWPadnos> no - -34 to -35
[21:39:45] <SWPadnos> yes
[21:39:48] <a-l-p-h-a> OH. wait. that's the wind from two days ago that drifted over To.
[21:40:05] <jlmjvm> what kinda dc motors you got
[21:40:13] <SWPadnos> jlmjvm, yep. the motors I have are only 40V/krpm, so I'd get around 175 IPM max from Geckos
[21:40:27] <SWPadnos> Baldor MTE-4070-BLBCE
[21:41:06] <a-l-p-h-a> I've seen Baldor's ads in trade magazines... I think they're from Ontario.
[21:41:13] <SWPadnos> 27 or 29 in-lb cont, ~115 peak, 150V max, 9A cont 37A peak, 1000l encoders and 7v/krpm tachs
[21:41:30] <jlmjvm> i have a motor question,your an ee
[21:41:47] <SWPadnos> I'm an ee, but not so great with motors. feel free to ask though
[21:41:52] <a-l-p-h-a> I need a snack. bbiab
[21:42:03] <SWPadnos> see ya
[21:42:07] <jlmjvm> im looking at a pacsci stepper motor
[21:42:35] <SWPadnos> ok - you've lost me ;)
[21:42:51] <SWPadnos> (kidding - go ahead :) )
[21:43:00] <a-l-p-h-a> pacsci are great.. but another alternative, check out oriental motors... they're relatively affordable, and good drives all around
[21:43:15] <jlmjvm> but instead of being 65v like the n series,its 35v,does this mean less power?
[21:43:45] <SWPadnos> lower top speed and lower power, yes
[21:44:14] <SWPadnos> err - are you talking about the same motor running at two different voltages, or two different motors?
[21:44:16] <jlmjvm> its still rated for the same rpm as the 65v version
[21:44:30] <jlmjvm> 2 different motors
[21:44:49] <SWPadnos> the specs are actually for 35V and 65V - that's exceedingly high
[21:44:51] <SWPadnos> ?
[21:46:12] <jlmjvm> let me send you a link
[21:46:14] <SWPadnos> ok
[21:48:47] <jlmjvm> http://cgi.ebay.com/POWERPAC-PACIFIC-SCIENTIFIC-STEPPER-MOTOR_W0QQitemZ170088601179QQcategoryZ78196QQssPageNameZWDVWQQrdZ1QQcmdZViewItem?hash=item170088601179
[21:48:59] <jlmjvm> i have 1 of those coming
[21:50:18] <SWPadnos> hmm
[21:50:32] <jlmjvm> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=320089396600&ssPageName=MERCOSI_VI_ROSI_PR4_PCN_BIX&refitem=130085484656&itemcount=4&refwidgetloc=closed_view_item&refwidgettype=osi_widget
[21:50:58] <jlmjvm> this is for a regular n series,same frame and stack
[21:51:44] <jlmjvm> was wondering if my 65v powersupply would hurt the motor
[21:52:21] <SWPadnos> not with a chopper driver, but you won't get very good performance (unless I'm not understanding something about the motors, which is quite possible)
[21:53:14] <jlmjvm> why not?
[21:54:00] <SWPadnos> well, the standard rule for steppers (using a chopper drive like a gecko) is to run them at 20-25x the nameplate voltage
[21:54:51] <SWPadnos> the reason is that the chopper will never allow more current than it's set for, and the extra voltage causees the motor to get up to that current faster
[21:55:19] <SWPadnos> the motor is an inductor, which resists changes in current flow. the more voltage you apply, the faster the change
[21:56:03] <pier> dV=-L di/dt
[21:56:06] <SWPadnos> rigth
[21:56:09] <SWPadnos> right
[21:56:17] <jlmjvm> yep
[21:57:01] <SWPadnos> but in any case, the 35V 5.5A 110W motor seems very inefficient (35V * 5.5A is 192.5W in, and only 100 out, so that's a lot of heat)
[21:58:43] <jlmjvm> didnt realize that
[21:59:00] <jlmjvm> the 65v is about the same way
[21:59:27] <SWPadnos> but don't take my word for it. I'd bet I'm missing something
[21:59:59] <jlmjvm> me too,cant wait till it gets here
[22:01:08] <jlmjvm> the powerpac series from pacsci is 850 oz in for the nema34 single stack
[22:01:30] <SWPadnos> that's pretty good
[22:01:58] <jlmjvm> it is for something that little
[22:02:35] <SWPadnos> this thread implies that pacsci rates motors on max V, not coil V: http://www.cnczone.com/forums/showthread.php?p=228404
[22:03:13] <CIA-6> 03cradek 07TRUNK * 10emc2/src/emc/rs274ngc/interp_convert.cc: fix single-arc entry on lathe
[22:04:02] <CIA-6> 03cradek 07v2_1_branch * 10emc2/src/emc/rs274ngc/interp_convert.cc: fix single-arc entry on lathe
[22:07:11] <jlmjvm> thanks for that link
[22:07:24] <SWPadnos> sure
[22:07:31] <SWPadnos> just found it ;)
[22:07:58] <SWPadnos> I did a google search for the first word in the motor model number
[22:10:22] <jlmjvm> i actually called pacsci earlier this week
[22:10:29] <jlmjvm> that was fun
[22:11:14] <a-l-p-h-a> mini croissants (sp), three bean salad, and what else did I have? more pine apples.
[22:12:09] <jlmjvm> by the way,do you think this motor should work
[22:13:49] <jlmjvm> sounds like i will need to drop my power supply
[22:15:24] <SWPadnos> I'm not qualified to answer :)
[22:15:36] <SWPadnos> I'm sure it will work, for some value of "work"
[22:16:49] <jlmjvm> my other 2 motors are working good with my 65v power supply
[22:18:10] <pier> night all and good week-end
[22:19:35] <a-l-p-h-a> later folks.
[22:31:28] <cradek> this thing paul fox recommended looks cool: http://www.newegg.com/Product/Product.asp?Item=N82E16833180021
[22:31:36] <cradek> wireless-to-ethernet-port thingy
[22:34:22] <Rugludallur> Im getting problems running my config now under 2.1.1 -> maximum velocity of 30066 & maximum number of steps possible is 10111 errors, is there a hard limit enforced now (BASE_PERIOD/VELOCITY settings don't seem to haver any effect)
[22:35:14] <cradek> check BASE_PERIOD vs INPUT_SCALE and STEPGEN_MAXVEL
[22:35:44] <jepler> Rugludallur: the calculation is the same as it's always been, but now you're told if that calculation gives a number below what you requested
[22:35:53] <Rugludallur> cradek: I tried chaning each parameter in turn but the error numbers did not change, .... ..
[22:36:22] <cradek> is STEPGEN_MAXVEL being read by your hal file? Maybe you have hard numbers in there
[22:36:27] <Rugludallur> jepler: so the system detects if It can't reach the speed requested ?
[22:36:54] <alex_joni> Rugludallur: right, it checks for the max steprate
[22:38:07] <Rugludallur> in that case the performance is way down, I'm getting ferrors at 1/3 the speeds I used to with 2.0 :(
[22:38:45] <Rugludallur> might be the new classicladder
[22:39:27] <jepler> is this the dallur config in CVS, or is your config different than what's there?
[22:39:57] <Rugludallur> jepler: the base/velocity/stepfreq is the same
[22:40:14] <Rugludallur> jepler: it's almost the same config
[22:41:01] <cradek> bbl
[22:42:07] <jepler> hm -- I get this: FREQGEN: Channel 0: The requested maximum velocity of 15000 steps per second is not attainable.
[22:42:12] <jepler> FREQGEN: The maximum number of steps possible is 12500 per second
[22:42:16] <jepler> FREQGEN: Channel 1: The requested maximum velocity of -2147483648 steps per second is not attainable.
[22:42:19] <jepler> FREQGEN: The maximum number of steps possible is 12500 per second
[22:42:34] <cradek> ouch
[22:42:38] <alex_joni> eek
[22:43:56] <jepler> BASE_PERIOD = 20000
[22:44:04] <jepler> INPUT_SCALE = 150.333
[22:44:12] <jepler> STEPGEN_MAXVEL = 300
[22:44:39] <alex_joni> jepler: that's for AXIS_1 ?
[22:45:00] <jepler> AXIS_0 and AXIS_1 look the same to me
[22:45:45] <Rugludallur> jepler: axis_1 and axis_0 have the same motors & gearing (x and y axis)
[22:45:54] <jepler> and I get 45099.9 for the requested steps per second
[22:45:57] <alex_joni> only max_vel is different
[22:46:02] <jepler> which is a different number than any I've seen so far
[22:46:12] <alex_joni> 220 vs. 240
[22:49:03] <K`zan> I knocked a hole in the bottom of a ceramic honey jar (please don't ask :-), what could I use to fill it that wouldn't put strange chems / flavors in the honey? Anyone got an idea on this? TIA!
[22:52:11] <jepler> oh oh -- those errors are from freqgen, not from stepgen
[22:52:32] <jepler> the ones I pasted anyway
[22:53:49] <jepler> it's possible that the checks on freqgen are just mistaken
[22:54:25] <jepler> dallur-advanced.hal:loadrt freqgen step_type=0
[22:54:37] <jepler> 32773 bit OUT FALSE freqgen.2.dir
[22:55:00] <jepler> how come I get 3 of them?
[22:57:10] <jepler> interesting -- it's a bug in sim!
[22:57:14] <jepler> (--enable-simulator, that is0
[22:57:15] <jepler> )
[22:58:22] <alex_joni> how about step_type="0" ?
[23:00:04] <jepler> that's a different bug
[23:00:09] <jepler> bbl
[23:00:14] <jepler> or bb tomorrow maybe
[23:00:44] <alex_joni> * alex_joni goes to bed
[23:00:57] <alex_joni> good night all
[23:02:04] <Rugludallur> night alex
[23:16:41] <skunkworks> logger_emc: bookmark
[23:16:41] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-03-09.txt
[23:29:47] <Jymmmm> SOB! Home Depot doens't have spring load cabinet latches... WTF this *IS* earthquake country!!!
[23:36:28] <ds3> try lowes?
[23:36:32] <ds3> or osh?
[23:37:47] <Jymmmm> Gonna head to OSG in a few minutes.
[23:37:49] <Jymmmm> OSH
[23:38:20] <Jymmmm> Still though... something so simple and not even anything close.
[23:39:39] <Jymmmm> wb LawrenceG
[23:41:57] <ds3> just epoxy the cabinet shut ;)
[23:46:52] <LawrenceG> hey Jymmmm
[23:47:38] <LawrenceG> I heard anbout all your greasy rabbits on the internet... any pix?