Back
[00:34:15] <rob__> hi, i added some info to the coordinatesystems page on wiki,
http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?CoordinateSystems also added some G52 info hope it helps users with there G92 questions
[00:48:27] <jepler> rob__: I'm not sure that mixing in a feature request with that page is the best thing to do
[00:49:24] <jepler> pictures help, so if that's your work then thank you
[00:49:33] <rob__> yea i did pics
[00:51:42] <jepler> I'd either move the stuff that is not in emc2 to a new page, or move it to a separate new section on that page labeled something like "for users of other controls", with a focus on how to perform the same tasks in emc as in those other controls (if possible)
[00:55:32] <rob__> ok give us sec i change it around
[00:56:53] <rob__> ok changed it around now, put it in Features from other control systems not EMC related - only reason put it in wiki is try better explane its use
[00:57:34] <jepler> thanks
[00:58:03] <rob__> if wish to use info and images any where else feel free, not the wrolds best pics
[01:03:07] <fenn> isn't "external work coordinate offset" just g92?
[01:14:20] <rob__> depends which book look at, some fanuc call it external, some call it common, that is the trimmer offset note i added refering to , its not programable though or shouldt be, its set via a menu screen effects all axes by the ammount
[02:49:38] <cradek> I programmed a part entirely in diameter mode ... it works nice
[02:50:02] <cradek> it's differently-unnatural than radius mode. I can't decide yet which is better.
[02:51:02] <SWPadnos> I guess if you have a drawing wirh diameters, diameter mode should be easier
[02:51:06] <SWPadnos> eith
[02:51:08] <SWPadnos> with
[02:53:44] <jmkasunich> I should try that
[02:54:14] <jmkasunich> I just got done programming a lot of code that looks like "G1 X[0.7505/2]"
[02:54:19] <cradek> yeah, ick
[02:54:30] <SWPadnos> heh
[02:54:34] <SWPadnos> oops :)
[02:54:42] <cradek> this part has a taper - I had to think twice when calculating the slope
[02:54:48] <cradek> but everything else was pretty natural
[02:54:58] <cradek> and touch off works in diameter mode too, now
[02:55:25] <cradek> I was always typing something like .7505/2 in the X touch off
[05:40:05] <cradek> dang time change
[05:40:13] <SWPadnos> heh
[05:40:14] <cradek> grrrr
[05:40:31] <SWPadnos> yeah. it should only be a bit late now, not freaking late
[05:41:53] <cradek> goodnight
[05:42:01] <SWPadnos> see you
[12:41:24] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/html/gcode_fr.html: french translation update
[12:41:24] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/gcode/main_fr.lyx: french translation update
[12:41:37] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/gui/axis_fr.lyx: french translation update
[13:56:28] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/nc_files/smartprobe.ngc: with the right extension this can be opened directly
[13:56:57] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/nc_files/smartprobe.ngc: curses
[13:57:21] <cradek> A G52 command will cancel a G92 in effect, while a G92 will cancel a G52 in effect.
[13:57:39] <cradek> this doesn't seem too hard -- it's not adding another kind of offset, only another way of setting an existing one.
[13:58:01] <cradek> G52 is cancelled when M30 M2 (end of program is read) or the system is reset
[13:58:11] <SWPadnos> it's basically absolute G92
[13:58:12] <cradek> ^ but this would be contrary to how G92 (now) works
[13:58:14] <jepler> cradek: no new features right now !!!!!111elventy
[13:58:19] <cradek> I know I know I know
[13:58:28] <jepler> anyway, g52 is just about the last feature we need
[13:58:42] <cradek> ok then what's the next-to-last?
[13:58:49] <SWPadnos> USB
[13:58:53] <SWPadnos> :)
[13:59:56] <jepler> fffff
[14:01:12] <SWPadnos> hmmm. I do need to commit something for halcmd completion when there's no RT loaded
[14:01:44] <SWPadnos> I think I got it working more or less correctly, but I have some extra stuff in there for detecting the thread a function is attached to for delf
[15:25:34] <greenpowerlight> hi
[15:25:50] <greenpowerlight> is here a person who can speak german?
[15:26:32] <greenpowerlight> because i've a problem with my emc2 on ubuntu
[15:26:34] <cradek> sometimes, but your english looks fine
[15:26:41] <cradek> what is the problem?
[15:26:54] <greenpowerlight> i want to drive it with a usb device
[15:27:03] <cradek> what kind of usb device?
[15:27:06] <greenpowerlight> because my machine has only a usb port
[15:27:34] <cradek> usb can be used for human control - like a joystick
[15:27:37] <greenpowerlight> is it possible?
[15:27:43] <cradek> usb cannot be used for machine control because it is not realtime
[15:28:00] <cradek> you will need to use a pci or parallel port interface to control the machine
[15:28:02] <greenpowerlight> aahhh soo there arent machines with ubs?
[15:28:04] <greenpowerlight> usb
[15:28:23] <cradek> right, emc does not support any usb interface for machine control
[15:29:01] <greenpowerlight> my teacher in school said to me that we have a usb mill
[15:29:02] <cradek> all the supported interfaces are PCI, parallel port, or ISA
[15:29:30] <cradek> then it is not running EMC
[15:29:36] <greenpowerlight> hmm
[15:29:45] <cradek> does your computer have PCI?
[15:30:19] <greenpowerlight> yes
[15:31:05] <greenpowerlight> so maybe the information from my teacher is wrong?
[15:31:15] <greenpowerlight> or are there mill's with USB?
[15:31:33] <greenpowerlight> it would be great to use EMC2 on my ubuntu
[15:31:37] <greenpowerlight> with this mill
[15:31:45] <cradek> brb
[15:32:08] <greenpowerlight> okay
[15:32:19] <SWPadnos> there are USB controllers, but none of them work with EMC
[15:32:28] <greenpowerlight> ohh bad ;)
[15:32:36] <SWPadnos> yes and no :)
[15:32:40] <greenpowerlight> why no ;)
[15:32:43] <SWPadnos> EMC does a lot that none of the USB controllers can
[15:33:39] <cradek> instead of buying a usb controller, you need to buy a pci card, and you can do what you want
[15:33:40] <greenpowerlight> hmm
[15:33:51] <cradek> usb has no benefits, it only has problems
[15:34:03] <cradek> or, you can buy a different computer
[15:34:03] <SWPadnos> if you already have a USB-based controller, then you should just use the software that came with it
[15:34:20] <greenpowerlight> i have a usb based controller mill
[15:34:34] <greenpowerlight> but there is no software
[15:34:34] <greenpowerlight> ;)
[15:34:49] <greenpowerlight> (sorry for my bad english)
[15:35:00] <SWPadnos> well in that case you're in trouble
[15:35:07] <SWPadnos> (with the mill, not your english :) )
[15:35:26] <cradek> you cannot use it with EMC. you may be able to use it with some other software but we cannot help you with that.
[15:35:29] <greenpowerlight> yeah i think so to
[15:35:36] <greenpowerlight> too
[15:35:43] <greenpowerlight> not good :(
[15:36:09] <greenpowerlight> there are no freewares like emc2
[15:37:14] <greenpowerlight> and what about a controller from USB to paralel device?
[15:37:37] <SWPadnos> no
[15:37:43] <greenpowerlight> hmm
[15:39:02] <greenpowerlight> so i can only hope that my teacher was wrong with his information ;)
[15:39:10] <greenpowerlight> or i will be in trouble ;)
[15:39:28] <greenpowerlight> hope the first one
[15:39:32] <greenpowerlight> :-D
[15:41:03] <cradek> good luck greenpowerlight
[15:41:16] <greenpowerlight> thx
[15:41:17] <greenpowerlight> ;)
[15:41:23] <greenpowerlight> thx for your help ;)
[15:42:29] <cradek> welcome
[15:43:22] <greenpowerlight> okay bye have to go ;)
[15:48:05] <SWPadnos> hmmm. I wonder what happened to Ray's email
[15:50:02] <jepler> sometime in the next week or two there'll be a disruption to the cvs server. My DSL service will be getting an upgrade (yay) -- unfortunately it only boosts incoming speed, not outgoing (boo)
[15:50:26] <cradek> IPs won't change I hope?
[15:50:37] <jepler> no
[15:50:48] <jepler> at least, nobody told me that :-P
[15:50:53] <cradek> ha
[15:50:53] <SWPadnos> heh
[15:51:12] <SWPadnos> I actually pay for a static IP, and somehow it magically got changed on me
[15:51:29] <SWPadnos> I called to ask why, and the tech more or less didn't understand the question
[15:51:32] <cradek> not incompetence, magic!
[15:51:40] <SWPadnos> exactly
[16:52:26] <micges> logger_dev: bookmark
[16:52:26] <micges> Just this once .. here's the log:
http://www.linuxcnc.org/irc/irc.freenode.net:6667/emcdevel/2009-03-10.txt
[17:37:20] <BJT-Work> thanks for the info micges
[17:40:57] <micges> BJT-Work: np
[18:01:20] <micges> jepler: about joints/axes: are there some additional debug info that could be turn on (beside DEBUG from ini file)
[18:01:23] <micges> ?
[18:07:54] <jepler> I have no special information about that
[18:51:47] <micges> how can I catch RTAPI_MSG_DBG messages ?
[19:23:35] <jepler> micges: man rtapi_set_msg_level
[19:24:27] <micges> thanks
[19:34:24] <aystarik> how to script "pause" in axis, so that tool comes up to some "safe-z" height? and "continue" moves tool back ?
[19:38:24] <cradek> you can't do that
[19:39:21] <aystarik> forbidden?
[19:39:46] <cradek> not sure that's the word I would choose - more like not possible
[19:40:52] <aystarik> or "not implemented"?
[19:41:01] <fenn> lots of hal hacking maybe
[19:41:05] <SWPadnos> not possible with EMC as it's currently written
[19:41:21] <fenn> mux commanded position and feedback position
[19:42:03] <cradek> I'm also bold enough to say it's a bad idea :-)
[19:42:07] <fenn> i think you'd want some timers too so you dont get a following error
[19:42:22] <fenn> aystarik: what sort of machine is this for?
[19:42:35] <SWPadnos> you may, and I stress may, be able to somewhat automate pausing the program, figuring out where it is, stopping it, issuing MDI moves to get to safe height, waiting for some HAL pin to change, then setting run-from-line and then starting the program again
[19:42:36] <aystarik> XYZ mill
[19:42:43] <SWPadnos> but that doesn't sound like a good idea to me
[19:43:40] <fenn> nah just feed hold and change the commanded/fb position
[19:44:29] <fenn> i dont see the point though on a mill
[19:45:09] <aystarik> "it was implemented on previous Z80 incarnation of this machine" :)
[19:45:25] <SWPadnos> well they were dumb ;)
[19:45:32] <SWPadnos> (just kidding)
[19:45:54] <SWPadnos> at the moment, there is no good way to do any motion while the machine is running a program
[19:46:02] <SWPadnos> (other than the program, of course)
[19:46:14] <aystarik> that was my first thought too.
[19:46:15] <fenn> what's wrong with my solution, just curious
[19:46:16] <SWPadnos> "paused" is still running as far as EMC2 is concernet
[19:46:45] <SWPadnos> fenn, it probably would work
[19:46:51] <SWPadnos> s/t/d
[19:46:54] <SWPadnos> heh
[19:46:56] <SWPadnos> std :)
[19:47:52] <aystarik> fenn: thanks, I'll check if it's feasible :)
[19:49:56] <aystarik> related question: it we use "run from selected line" is it hard to look over program and check if the spindle needs to be started, etc?
[19:50:17] <fenn> run from line doesn't work so well because of issues like that
[19:51:29] <aystarik> may be it's possible to create fast-forward in interpreter for this?
[19:51:45] <SWPadnos> it's there, it just doesn't save the state (unfortunately)
[19:52:02] <SWPadnos> I think this may be an area we look at for 2.3++
[19:52:38] <SWPadnos> in 2.3, you can fiddle with coolant and spindle and stuff, and there's an option to not reset those when you change from MDI/manual to run mode
[19:52:48] <SWPadnos> which is half way there (or so)
[19:53:02] <cradek> SWPadnos: there is not an option anymore in 2.3, the default and only behavior is correct
[19:53:08] <SWPadnos> oh, ok :)
[19:54:33] <aystarik> cradek: could you please elaborate? what is the state?
[19:55:40] <aystarik> everything is going to be reset to "off"?
[19:56:35] <cradek> no, nothing is turned off when you switch modes
[19:56:45] <cradek> sorry, I didn't mean to be obscure
[19:56:55] <cradek> you can turn the spindle and coolant on, then start the program from the middle just fine.
[19:57:33] <cradek> or, you can do it with MDI (S1000 M3 M8) and then go start the program
[19:58:38] <aystarik> how about correct XYZ if they are omitted from the line?
[19:59:51] <SWPadnos> the program will resume from whatever position the machine is at
[20:00:54] <aystarik> and if it never get off machine 0?
[20:05:47] <SWPadnos> ?
[20:07:47] <aystarik> if tool is still at machine/G54 x0y0z0, then the program is likely to do something not intended, or will run out of machine limits.
[20:08:04] <SWPadnos> that's definitely possible
[20:08:36] <SWPadnos> if the program is restarted in relative mode, it may stop with an error complaining about soft limits
[20:08:52] <SWPadnos> it's best to move to the right spot first, of course
[20:09:45] <cradek> if you start on a G1, it will feed in a straight line to the programmed destination
[20:09:56] <cradek> of course that can be bad if you're in the wrong place.
[20:10:17] <cradek> it's best to restart a program at a move at safety height or tool change etc.
[20:12:20] <aystarik> I know. this is basically a question, how much automation we can/should put into g-code interpreter, so that it becomes more convinient/handy...
[20:14:22] <rob__> the best program auto restart function i ever used is the one on a philips control on a lathe, it would make a very good refrence to get some ideas from if you needed some
[20:27:21] <aystarik> do you have any manual or something?
[20:27:53] <SWPadnos> "best" is relative to the task at hand
[20:28:03] <SWPadnos> best for a skilled operator/NC programmer?
[20:28:13] <SWPadnos> best for a button-pushing drone?
[20:28:21] <SWPadnos> best for an automated system?
[20:28:27] <SWPadnos> best for a remotely controlled system?
[20:28:35] <SWPadnos> I don't think they're all the same
[20:54:58] <rob__> sorry was away, for a operater that has there wits about them, and does it in a safe maner also, yes i have manual, also machine with control , in simple terms it does a fast scan from start of program, and goes through the motions etc program would if ran normaly
[20:56:15] <rob__> and as it goes setting all G codes, M codes etc in use, so when it gets to the highlited line, all needed active codes are already loaded, spindle speed set etc, but you still need to do the start location from a safe place, i cant see any simple way around that program from "button pushers" as you put it
[20:56:31] <SWPadnos> the way that I've thought about, and which I think has been bandied about a bit, is to make a list of words (like S, M3/4/5, M7/8/9) that are executed, including axis positions, and only execute the most recent one just before restarting
[20:57:26] <SWPadnos> there are things that can't be done that way though. for example custom M-codes
[20:57:29] <rob__> yes that pritty much i guess what they did on there control, it is only now most new machines are having a restart from any where function
[20:57:35] <SWPadnos> emc wouldn't know what to do with them
[20:57:57] <rob__> if you do a program check on a fanuc or any other control it always lists active gcodes and m codes some where, as it shows u what is in the buffer
[20:57:58] <SWPadnos> tickling something in a PLC/classicladder also would require operator intervention
[20:58:20] <SWPadnos> sure, we can see those in the status buffer
[20:58:39] <SWPadnos> but then again, you don't want to do this when there's a skilled operator running the machine
[20:59:01] <SWPadnos> you only want to second-guess people sometimes
[20:59:18] <rob__> humans hay
[20:59:31] <SWPadnos> and at other times, the person running the machine knows what they're doing, and the machine should just play dumb
[20:59:41] <SWPadnos> the hard part is supporting both of those cases
[21:11:13] <rob__> is there any plans to add any more info onto the big DRO tab
[21:12:22] <SWPadnos> no specific plans, but "offsets in effect" might make it there some day
[21:14:39] <rob__> offsets?
[21:15:18] <rob__> work offsets
[21:15:26] <SWPadnos> like G92 and whatnot
[21:16:53] <rob__> would be nice to see active codes on dro i think so one can qucikly see what mode you are in etc
[21:17:46] <SWPadnos> that's already on the MDI page. maybe it should be visible in run mode too
[21:18:18] <rob__> heres a photo of a fanuc program check page -
http://innovative-rc.co.uk/fanuc-prg_chk.jpg also found offsets photo i took while ago -
http://innovative-rc.co.uk/fanuc-offsets.jpg
[21:31:08] <fenn> clear as green mud on a black background
