#emc | Logs for 2004-11-14

[00:23:28] <an0n__> * an0n__ sighs
[00:23:31] <an0n__> I think I just broke a servo
[00:23:34] <an0n__> :(
[00:23:34] <an0n__> an0n__ is now known as an0n
[00:24:52] <paul_c> How can you "break a servo" ??
[00:57:18] <jepler> Is anybody running emc in simulator mode on fedora core 2? I'm having some trouble getting it to work despite some help from cradek. I've never built emc before, except to run emcplot3d.
[00:59:56] <paul_c> * paul_c has only installed fedora once - Just to see what it was like....
[01:01:37] <jepler> the program builds and executing sim.run gets me the xemc interface
[01:02:23] <jepler> but for instance if I jog to about y=2 then hit home, I get a message box that says "Axis 1 following error", and the same for the other axes
[01:04:27] <jepler> switching to mdi mode and executing commands (such as "g0 z1" or "G0 Z1" etc) does nothing but print errors in the terminal
[01:04:31] <jepler> and running a program doesn't work either
[01:04:51] <paul_c> what sort of errors ?
[01:05:50] <jepler> trying to execute one of the files from programs/:
[01:05:56] <jepler> emcTaskPlanLine() returned 1
[01:05:56] <jepler> emcTaskPlanCommand() called. (line_number=1)
[01:05:56] <jepler> canterp: unrecognized canonical command
[01:05:56] <jepler> emcTaskPlanExecute(0) return 1
[01:05:56] <jepler> emcTaskPlanClose() called at emctaskmain.cc:1248
[01:05:59] <jepler> Issuing EMC_TASK_PLAN_SYNCH -- (+516,+12, +0,)
[01:07:06] <jepler> executing "g0 z1":
[01:07:06] <jepler> Issuing EMC_TASK_PLAN_EXECUTE -- (+509,+268, +4,g0\040z1,)
[01:07:06] <jepler> canterp: unrecognized canonical command
[01:07:06] <jepler> emcTaskPlanExecute(g0 z1) returned 1
[01:07:06] <jepler> emctaskmain.cc 2104: !ERROR! error executing command 509:EMC_TASK_PLAN_EXECUTE
[01:07:42] <paul_c> have you done an update from cvs ?
[01:08:04] <jepler> yes. this is a fresh checkout from this morning of the CVS HEAD of rcslib and emc (not emc2)
[01:09:36] <paul_c> OK - I recall seeing where these errors get produced....
[01:10:35] <paul_c> but the canterp interface shouldn't be linked in...
[01:11:39] <jepler> hm, I took a copy of cradek's sim.ini, and he's uncommented centerptask and commented minimilltask
[01:13:06] <jepler> OK, I blew away sim.ini and checked it out again, and now I can "g0 x1 y1 z1" in mdi
[01:13:47] <jepler> that also changed me from xemc to tkemc, I wonder which change mattered
[01:14:21] <jepler> no, it works now in xemc.
[01:14:43] <jepler> I still get the "Axis N following error" if I jog to y=1 and then hit home
[01:17:20] <paul_c> check your scales
[01:17:20] <jepler> OK, great, everything I was interested in is now working. I can run xemc and see the live plot in emcplot3d
[01:17:38] <jepler> 3D_Chips.ngc is now plotting
[01:18:46] <jepler> Is there somewhere I can look in the source tree for the simplest program that would monitor the state of the mill, just printing coordinates to stdout a few times per second?
[01:19:38] <paul_c> don't remember seeing anything to do that..
[02:15:14] <jepler> Here's a screenshot of my Python + PyOpenGL rs274ngc viewer: http://craie.unpy.net/~jepler/rs274py-gplot.png
[02:15:25] <jepler> It has no dependency on emc
[02:15:37] <jepler> just a little something I've been working on...
[02:22:15] <paul_c> For emc2 ?
[02:22:44] <jepler> I'm not sure what my goals are
[02:23:14] <jepler> I admit I started serious work on it when cradek told me that emcplot3d didn't have a clear license.
[02:24:19] <jepler> on the other hand, I doubt that python + pyopengl is something you want to make a requirement for emc2, and having a separate parser and interpreter for rs274 means the viewer may show different things from what emc will do
[02:25:35] <paul_c> It's notusing the same rs274 interp. code then ?
[02:26:15] <jepler> nope, though I am using http://www.linuxcnc.org/handbook/RS274NGC_3/RS274NGC_3TOC.html as my documentation for the language, and my "canonical form" is also inspired by that document.
[02:27:08] <jepler> sorry, I have to go.
[02:27:18] <jepler> I'll read this scrollback later and may be back tomorrow
[05:56:30] <paul_c> Hi John
[05:57:19] <logger_jmk> I'm not really here '-)
[05:59:10] <paul_c> just testing the conection ;}
[06:11:20] <an0n> * an0n broke a servo
[06:11:30] <an0n> 500$ :(
[06:11:50] <an0n> I feel like crying or somthing
[07:08:10] <asdfqwega> Don't feel bad an0n...I broke a $1000 laser - and a replacement cost $3000
[09:01:55] <paul_c> cradek: http://craie.unpy.net/~jepler/rs274py-gplot.png
[09:02:16] <cradek> hi paul, hi jepler
[09:03:09] <paul_c> Evening cradek
[09:05:34] <cradek> so what did I miss?
[09:06:07] <paul_c> not a lot
[09:06:23] <paul_c> It's been fairly quiet the last few days.
[09:06:58] <cradek> I'm going to have to complain to my sysadmin about these network outages
[09:07:17] <pfred1_> cradek yeah I bet they need a good laugh!
[09:07:44] <cradek> pfred1_: the real irony here is that I'm the sysadmin
[09:08:38] <paul_c> 10:30pm up 29 days, 8:37, 1 user, load average: 0.00, 0.00, 0.00
[09:08:51] <cradek> paul_c: ?
[09:09:41] <paul_c> 23:03:23 up 29 days, 8:32, 3 users, load average: 1.49, 1.64, 1.58
[09:09:59] <paul_c> My router & server stats ;}
[09:10:19] <cradek> but only a windows user could be proud of that
[09:10:27] <cradek> 5:00pm up 1202 days, 6:46, 1 user, load average: 0.12, 0.04, 0.02
[09:10:54] <paul_c> I had to reboot 29 days ago after a power outage
[09:11:01] <cradek> -> ls -ld /lost+found/
[09:11:01] <cradek> drwxr-xr-x 2 root root 8192 Oct 17 1991 /lost+found/
[09:11:15] <cradek> I had to shut this machine off because the office moved 4 years ago
[09:12:39] <cradek> so jepler showed off gplot
[09:13:18] <paul_c> Looks good
[09:13:29] <cradek> and it has a proper license
[09:13:41] <paul_c> Now... If we could talk him in to committing to emc2 under GPL...
[09:14:20] <paul_c> * paul_c thinks about the possibilities of using vtk
[09:14:32] <cradek> I don't see any particular reason it should be in the same repository as emc
[09:14:54] <cradek> they can read the same files
[09:14:58] <cradek> they are useful together
[09:15:09] <cradek> but they are unrelated otherwise (no code in common)
[09:17:52] <cradek> what did he say about this when you asked him?
[09:18:02] <paul_c> the main emc tree is a mess when it comes to working out what is GPL, Public Domain, non-free, and non-commercial
[09:21:45] <pfred1_> emc2 isn't GPL?
[09:22:14] <paul_c> emc2 is, but emc(1) isn't
[09:22:40] <pfred1_> emc1 is public domain?
[09:25:00] <paul_c> Parts of the source code is
[09:25:17] <paul_c> some is GPL
[09:25:36] <paul_c> a couple of files are non-commercial
[09:26:05] <paul_c> an a few of the sources in rcslib could be proprietory
[09:26:28] <pfred1_> damn what a mess!
[09:26:55] <pfred1_> what is "non-commercial" ?
[09:27:09] <paul_c> not for profit
[09:29:25] <pfred1_> paul_c so that mean you can't resell the code or cna't use it in any commercial application at all?
[09:32:09] <cradek> pfred1_: at least one file's license is "free for non-commercial use"
[09:32:14] <cradek> pfred1_: that's the whole license
[09:32:25] <paul_c> emcplot3d is the file in question - And Yes, that is how I read the response to my query
[09:45:44] <jepler> rs274py / gplot is GPL.
[09:45:57] <jepler> if you want it in the emc2 tree you're certainly welcome to it
[09:47:10] <paul_c> jepler: Do you have cvs access ?
[09:47:39] <jepler> paul_c: Nope, I sure don't.
[09:47:55] <cradek> jepler: look out, you're about to become a developer
[09:47:58] <jepler> oh hell
[09:48:06] <paul_c> Are you registered with SF ?
[09:48:12] <jepler> Yes, I'm "jepler" there
[09:49:10] <jepler> What's with the % at the top and bottom of .ngc files? It's not mentioned in the reference I'm using (http://www.linuxcnc.org/handbook/RS274NGC_3/RS274NGC_3TOC.html)
[09:49:51] <paul_c> It's an optiona marker for the start & end of a file
[09:50:24] <jepler> So if there's a line with a lone "%" on it, followed by more lines, the following lines are ignored by the interpreter?
[09:51:55] <paul_c> jepler: Yu now have full write access to CVS
[09:52:11] <cradek> jepler: sucker
[09:52:59] <paul_c> I would suggest creating emc2/src/tools/3dplot (or pick your own name).
[09:56:21] <cradek> jepler: now you need a machine
[09:56:30] <jepler> Does the emc project buy me one?
[09:56:43] <cradek> I think so - just arrange it with paul
[09:57:07] <cradek> actually I think they're about to buy me a new one, and I can pass mine on to you
[09:58:19] <jepler> let me know when that happens
[09:58:26] <cradek> right
[09:58:47] <paul_c> Sherline will do a discount for registered developers (if you ask them)
[09:59:30] <cradek> paul_c: we really need bigger machines with servos for proper testing.
[10:01:01] <paul_c> hexapods with 256 tool loaders
[10:01:22] <pfred1_> sherlines are toys
[10:01:28] <cradek> we need a grant or something don't we
[10:01:54] <paul_c> You want to know what I mount my Sherline on ?
[10:02:00] <cradek> pfred1_: you can make parts with them. just smaller parts.
[10:02:15] <cradek> pfred1_: mine's on a piece of oak
[10:02:18] <cradek> err paul:
[10:02:46] <paul_c> mine gets bolted to the table of a Bridgeport Interact II
[13:04:31] <jepler> yay, live plot display in my rs274ngc display app: http://craie.unpy.net/~jepler/rs274py-gplot-live.png
[13:44:19] <cradek> jepler: wow
[13:46:12] <cradek> jepler: check it in!
[13:53:41] <cradek> jepler: you need to come up with a name for it
[20:36:40] <paul_c> Up early John ?
[22:27:15] <Imperator_> Hi Paul, are you here ???
[22:29:50] <paul_c> Hi Imperator_
[22:30:02] <Imperator_> Hi Paul how are you
[22:31:09] <Imperator_> can you give me a short overview what was going on the last two weeks. In one sentence maybe
[22:31:27] <paul_c> Not a lot...
[22:32:18] <paul_c> Vreified the origins of emcplot3d and found it is subjuct to a non-commercial licence.
[22:32:59] <Imperator_> so it has to be rewritten
[22:33:05] <paul_c> On going discussions about where he emc2 files should go when installed
[22:33:20] <Imperator_> jup
[22:33:25] <Imperator_> I am thinking on making EMC able to drive a gantry axis, but i want to discuss the way to do that before
[22:33:50] <paul_c> jepler already has a python based replacement for emcplot3d
[22:34:11] <Imperator_> phyton
[22:34:22] <Imperator_> emc gets a multi language software
[22:34:48] <Imperator_> tcl c c++ phyton
[22:35:06] <paul_c> Python is usefull for rapid development of all sorts of stuff...
[22:35:40] <Imperator_> ok
[22:36:25] <Imperator_> do you want to say something about the gantry axis, or did i have to discuss that with john ?
[22:36:38] <Imperator_> hope he is here today
[22:37:37] <paul_c> You wanting to slave two drives ?
[22:37:42] <Imperator_> jep
[22:38:05] <Imperator_> i think this is not that big deal, but homing would be a problem
[22:38:49] <paul_c> EMC will drive an XXYZ machine
[22:38:54] <Imperator_> i have spoken with the guys of Heidenhaim, they said that not so much controllers are able to do that
[22:39:00] <paul_c> without any changes to the code
[22:39:06] <Imperator_> hm
[22:39:34] <paul_c> but you are correct in your assumptions about homing
[22:39:42] <paul_c> It would be a pain...
[22:39:49] <Imperator_> but the XX Axis must move in any case exactly the same
[22:39:56] <paul_c> yup
[22:40:06] <Imperator_> free mode kills everything :-)
[22:40:19] <paul_c> yes
[22:41:22] <Imperator_> there is no work in prozess at the time, or ??
[22:42:13] <Imperator_> or anything that could help
[22:42:24] <Imperator_> Hi les
[22:42:31] <les> hello
[22:42:47] <paul_c> Les would know about the problems of an XX axis
[22:43:00] <paul_c> * paul_c passes the buck
[22:43:17] <les> well, emc was/is not set up for it
[22:43:47] <les> but I use a jackshaft to drive the big gantry on my machine
[22:43:56] <les> twin ballscrews
[22:44:12] <Imperator_> ok, i want to change that, or better i have to because my frends are starting to build such a machine at the moment and my tast is to get it work
[22:44:25] <les> great
[22:44:35] <Imperator_> :-)
[22:44:52] <Imperator_> ok that is also our backup solutian
[22:44:56] <les> the jackshaft transmissions I made were very expensive
[22:45:01] <les> but work well
[22:45:25] <les> one problem with electronic gearing....
[22:45:53] <Imperator_> be get two linear encoders from heidenhain with 1700mm and 2200mm at ebay
[22:46:03] <les> is that the aparent mass that each end sees is constantly changing
[22:46:47] <les> so with fixed tuning some racking forces can exist due to unequal following error
[22:46:58] <Imperator_> i think that is very easy to solve with the HAL, i have only to conect two PID controller to one X output, but homing is the problem i think
[22:47:03] <les> with very tight tuning it works fine though
[22:48:11] <les> the best tandem systems would use feedforward based on the position of ANOTHER axis
[22:48:37] <les> the one(s) carried by the tandem driven axis
[22:49:01] <les> but what about homing?
[22:49:12] <les> what problem do you see there?
[22:49:36] <Imperator_> i must home that two axis the same time !!!
[22:49:45] <Imperator_> otherwise the machine is gone
[22:49:53] <les> oh...
[22:49:54] <les> hmmm
[22:50:04] <Imperator_> and then i have to deside how to synchronise them
[22:50:31] <les> well if one truly slaved off the other...
[22:50:32] <Imperator_> because they are out of sync when i switch of the machine
[22:50:55] <les> in other words commanded position of one=measured position of the other
[22:51:10] <les> rather than just independent
[22:51:50] <Imperator_> ithink commanded possision must be the input for both axis
[22:52:18] <Imperator_> that is with HAL very easy
[22:52:33] <Imperator_> but they must be synchronised at homing
[22:52:34] <les> true slaving/electronic gearing needs this and emc does not do it as far as I know
[22:52:44] <les> but right...very easy to code
[22:53:14] <les> then home the master and the slave follows
[22:53:50] <les> (with servos)
[22:55:05] <les> There will always be a group delay between then due to integrators in the control laws but it coulsd be very small
[22:55:12] <les> could
[22:55:24] <Imperator_> hope so
[22:55:42] <les> high servo updates...
[22:55:52] <Imperator_> jep
[22:56:12] <les> no big problems that I see...just has to be coded
[22:57:00] <les> If a y is riding on a gantry x...
[22:57:45] <les> the master and slave x will benefit from a zero order feedforward term based on the square of y position
[22:58:12] <les> (due to moment of inertia of y)
[22:58:27] <Imperator_> godd idea
[22:58:33] <Imperator_> there are some smal problems that make the think interesting :-)
[22:59:48] <les> in the general sense inverse kinematics need jacobians and stuff in the control laws
[23:00:02] <les> but for an xxyz not very complicated
[23:00:57] <les> I wish I could have figured out where to write this into emc1
[23:01:15] <les> math is easy but figuring out the code was not
[23:01:58] <les> so jackshaft, transmission, and mechanical pitch matching cams are what I had to use
[23:02:21] <les> BIG $$$$$ that I would have preferred not to spend
[23:03:18] <Imperator_> jup
[23:05:21] <les> well I am going to make some breakfast...back in a while
[23:05:32] <Imperator_> ok i will have a deep look into the code, see you
[23:15:25] <les> ok...one last thing...if one does true slslaving for xxyz where racking error is made closed loop...
[23:15:42] <les> he will have solved lathe threading as well.
[23:35:29] <Imperator_> hm I'm not sure, because that is another thing
[23:35:41] <Imperator_> but i don't know
[23:40:24] <jepler> good morning golks
[23:40:26] <jepler> er, folks
[23:46:33] <Imperator_> hi jepler
[23:47:06] <Imperator_> are you new here ?? I was two weeks away and i have never seen you before
[23:48:03] <Imperator_> I am a guy from germany how wants to make emc able to use ganry axis
[23:49:22] <jepler> I've been around a few times before, but not that much.
[23:49:34] <jepler> I'm a friend of cradek, and sometimes I borrow his mill to do circuit boards
[23:51:06] <jepler> I've done a little work improving emcplot3d, and most recently I've been writing a g-code interpreter and viewer in Python, which I believe Paul mentioned earlier.
[23:52:25] <Imperator_> jep I have heard about it, is it in the CVS ??
[23:52:34] <jepler> Nope, it's not in emc cvs
[23:53:01] <jepler> There's a screenshot online: http://craie.unpy.net/~jepler/rs274py-gplot-live.png
[23:53:33] <jepler> and my cvs can be viewed online: http://unpy.net/cgi-bin/viewcvs.cgi/rs274py/
[23:53:38] <Imperator_> hm, that looks nice
[23:54:13] <jepler> It only supports a small subset of rs274ngc so far
[23:56:05] <Imperator_> very good, that are the tools that will make working with emc much easyer
[23:58:14] <jepler> I was surprised that Paul immediately suggested checking it into emc2, it's so far from complete, duplicates the parser and interpreter, and is in a new language not used in emc before...