#emc | Logs for 2009-07-12

Back
[00:06:53] <pilotltd> http://imagebin.ca/view/Ddwvn5.html
[00:07:05] <pilotltd> Does that make sense?
[00:07:45] <cradek> yes
[00:09:07] <cradek> I think it will work this way
[00:09:17] <pilotltd> Hope so :)
[00:10:20] <pilotltd> I'll leave the drives out at first and check the voltages and transformer temp <G>
[00:14:19] <pilotltd> Thanks - bed time here, I'll let you know how it goes tomorrow, gdnight :)
[00:14:38] <cradek> goodnight
[01:01:41] <Jymmm> what kind of mad scientist transfer does he have
[01:01:47] <Jymmm> transformer
[01:04:06] <Jymmm> http://imagebin.ca/view/uC-Xeqj.html
[01:09:21] <wholepair> tkinter.TclError: Togl: X server has no OpenGL GLX extension
[01:10:52] <wholepair> I am setting up a new PC for my machine - emc wont run with axis - tkemc works
[01:10:59] <wholepair> I get this error: tkinter.TclError: Togl: X server has no OpenGL GLX extension
[01:59:22] <Jymmm> a cylinder with a 4" diameter and 18" long ~= 0.13 cubic feet - Does that sound about right?
[02:04:45] <jepler> yes, that's what units gives
[02:04:50] <jepler> You have: (4 inch/2)^2 pi * 18 inch
[02:04:50] <jepler> You want: ft3
[02:04:50] <jepler> * 0.13089969
[03:18:49] <Jymmm> thanks jepler... http://www.kutv.com/content/news/local/story/Man-Builds-Pipe-Bomb-That-Nearly-Kills-Boy-On-Bike/EgLTCEekr0KISol9FNTX3A.cspx
[05:48:28] <coden4life> anyone know a good spot to put a link for a cad cam tutorial ... can I post a link in here
[05:48:44] <coden4life> just uploaded it and it is for SagCad beginners
[05:50:13] <coden4life> http://www.metacafe.com/watch/3048835/sagcad_tutorial/
[06:28:55] <Jymmm> http://wiki.linuxcnc.org/
[07:13:43] <fenn> people actually use sagcad?
[07:23:29] <alex_joni> that one sounds like donald
[12:42:47] <EbiDK> EbiDK is now known as EbiDK|AWAY
[12:48:15] <BigJohnT> Is there a way to get updates to EMC without a network connection? I seem to remember something about this but can't find anything...
[12:48:27] <BigJohnT> like from 2.3.0 to 2.3.2
[13:22:51] <mozmck> BigJohnT: I just downloaded the .debs and put them on a thumb drive.
[13:28:15] <mozmck> look here for hardy .debs http://linuxcnc.org/hardy/dists/hardy/
[13:28:22] <BigJohnT> mozmck: thanks, I thought it was something simple like that
[13:29:19] <mozmck> np
[13:33:54] <frallzor> is /A = A'
[13:34:17] <frallzor> talking about stepper connections and different schematics
[13:37:37] <mozmck> I "think" those both mean "not A", but not completely sure...
[13:38:51] <mozmck> or on a stepper is be the other side of coil "A" with A being the one side
[13:39:34] <mozmck> is be = it would be
[13:39:43] <frallzor> well thats pretty obvious =)
[13:40:22] <frallzor> but schematics says A-A but one says A' and the other says /A so i guess both are "not A" then =)
[13:40:55] <mozmck> I'm pretty sure this is correct.
[13:41:10] <frallzor> http://www.brundin.biz/images/extra/STM76-185-wire-parallell.jpg
[13:41:31] <frallzor> A C should be A then
[13:41:52] <frallzor> and B D > B preferably =)
[13:43:26] <mozmck> looks like, but I've not seen a stepper schematic like that.
[13:44:03] <mozmck> the 8-wire motors I've seen have 4 coils
[13:44:27] <frallzor> well they just made it like that so you see how it works
[13:44:40] <mozmck> so it actually has 4 coils?
[13:44:51] <frallzor> yup
[13:45:15] <frallzor> http://www.brundin.biz/images/extra/STM76-185-wire-serial.jpg
[13:46:20] <mozmck> ah, that's better. you can connect it in series for lower current or parallel for higher current.
[13:46:54] <frallzor> I know =) just the odd / ' that confused med
[13:47:06] <frallzor> shouldnt there be like a "standard" on how to mark things? =)
[13:47:35] <mozmck> I think there are 2 or 3 standards!
[13:47:38] <mozmck> :)
[13:48:14] <mozmck> where is the / ? on a different schematic?
[13:48:18] <frallzor> I wouldnt be surprised if a' was THE A
[13:48:27] <frallzor> geckodrive manual
[13:48:44] <frallzor> they call it /A and /B there =)
[13:51:08] <mozmck> I think the A' would be the /A. you can sway the A and /A you know, the motor just turns the other way
[13:52:12] <mozmck> sway = swap, bleh, I need coffee
[13:52:15] <steve_stallings> steve_stallings is now known as steves_logging
[13:52:37] <frallzor> Yeah but I want it the way its supposed to be :P
[13:59:16] <SWPadnos> there is no "the way it's supposed to be"
[13:59:35] <SWPadnos> some machines have the motor on the right side, others on the left
[13:59:47] <SWPadnos> that changes the direction of table travel for a given rotation of the motor
[14:00:03] <frallzor> well if the schematic says one thing I want it like that =)
[14:00:05] <SWPadnos> some machines move the tool (like a gantry), others move the work (like a knee mill)
[14:00:08] <frallzor> for me that is right
[14:00:14] <SWPadnos> that also changes the effective direction of motion
[14:00:42] <SWPadnos> ok. if that's what you want, then that's right for your machine
[14:01:11] <frallzor> just needed to find out if A' was /A =)
[14:01:18] <SWPadnos> probably :)
[14:01:31] <frallzor> yup
[14:02:02] <SWPadnos> the coils aren't labeled with a "plus/minus" scheme, possibly because they're driven in both directions under normal operation
[14:02:30] <frallzor> yeah I know but I still want to connect like the manual says, principle =)
[14:02:36] <SWPadnos> usually A and /A are used for the plus and minus sides of a differential pair, or for a signal and its inverse
[14:03:01] <SWPadnos> (or you might have /RESET, which says it's an active-low signal)
[14:03:22] <SWPadnos> "/" is usually read as "not": /A is "not A"
[14:03:38] <frallzor> ah
[14:03:38] <SWPadnos> (logical not, of course :)
[14:03:40] <SWPadnos> )
[14:05:51] <frallzor> finally founda nice case to order for the electronics, yeey =)
[14:06:19] <SWPadnos> make sure it's twice as big as you think it should be
[14:06:30] <SWPadnos> maybe three times
[14:06:53] <frallzor> well its roomy if that what you mean =)
[14:06:57] <SWPadnos> yes
[14:07:09] <frallzor> vents too, so I can mount a nice fan
[14:07:17] <SWPadnos> and a filter ... ;)
[14:07:30] <frallzor> of course
[14:07:31] <frallzor> http://www.conrad.se/?article=520489
[14:08:00] <frallzor> one of those =)
[14:08:25] <frallzor> Aint got much to put in, 540 and a pretty small psu
[14:10:42] <pilotltd> Hi guys
[14:14:59] <Sordna> http://www.pactecenclosures.com/ I like these guys for cases..and some you can get as a free sample here and there.
[14:16:40] <Sordna> and, morning all.
[14:19:02] <pilotltd> Is there any way to add additional keyboard shortcuts to Axis?
[14:21:50] <pilotltd> like shift+jogkeys for rapid jogging
[14:25:59] <mozmck> pilotltd: I think there is a way to change key shortcuts in the INI file, but not add new ones that I'm aware of
[14:26:35] <mozmck> but don't ask me how :)
[14:27:35] <jepler> no, there's no inifile way to change axis keyboard shortcuts
[14:27:54] <jepler> a developer has proposed a patch specifically to add fast jogging with shift. You can try it, but you have to build emc2 from source: http://mid.gmane.org/200904291608.56617.mb@bu3sch.de
[14:28:21] <jepler> (actually, a better link is http://thread.gmane.org/gmane.linux.distributions.emc.devel/2157 since it allows you to get to the patches more easily)
[14:28:48] <jepler> argh, no, that's still not the right link
[14:29:46] <jepler> http://mid.gmane.org/200904291605.48995.mb@bu3sch.de http://mid.gmane.org/200904291607.57963.mb@bu3sch.de
[14:30:26] <SWPadnos> since that's in python, couldn't it be used without recompiling?
[14:30:33] <SWPadnos> just patch axis.py
[14:31:00] <SWPadnos> (or was there some other stuff in gcodemodule.cc or somewhere else that I missed?)
[14:31:36] <jepler> no, it was .py and .tcl
[14:31:43] <jepler> so I suppose you can do the edits on your installed system
[14:31:49] <jepler> but they'll be overwritten the next time a bugfix release comes out
[14:32:26] <SWPadnos> yes, it's not perfect, but it could be easier than building EMC (assuming that -dev packagtes and whatnot need to be installed ...)
[14:35:35] <micges> you can change shortcuts in .axisrc (or any USER_COMMAND_FILE)
[14:55:19] <pilotltd> Is it going to be included? If so - I'll wait, I haven't a clue how to patch anything
[14:59:16] <SWPadnos> it was added to TRUNK a fwe months ago. I'm not sure when it will appear in a released EMC2 package
[14:59:20] <SWPadnos> few
[15:01:25] <pilotltd> If it was a few months ago - surprised it wasn't in the 2.3.2 release?
[15:02:12] <SWPadnos> it wasn't added to the release branch at that time. things like that should be tested a bit before being backported
[15:02:46] <SWPadnos> there was a note from the author that rapid jogging didn't obey axis valocity limits, which would certainly have made me concerend about adding it to the release branch :)
[15:03:37] <pilotltd> Weird, there should be no way to exceed the limits?
[15:04:14] <SWPadnos> yeah. I don't really know what he meant by that
[15:04:32] <SWPadnos> I se that the applied change uses the traverse rate (G0) as the jog speed
[15:04:34] <SWPadnos> see
[15:04:59] <cradek> obeying the gui's max velocity slider was the issue I think
[15:05:19] <pilotltd> yes - using G0 shoudl be fine (or is in all the other packages I've used)
[15:05:29] <SWPadnos> oh, that could be
[15:06:08] <pilotltd> The max velocity slider is superflous IMO
[15:06:42] <pilotltd> If your machine is tuned correctly, G0 should be max vel
[15:07:09] <SWPadnos> G0 is max vel
[15:07:15] <pilotltd> IF you want to go slower use G1 :)
[15:07:31] <SWPadnos> which may change based on the axis(es) involved in a particular move
[15:08:12] <pilotltd> Yep - makes no difference
[15:08:50] <cradek> pilotltd: it's for safely testing a program the first time, not for configuration
[15:09:32] <pilotltd> But why would that affect G0?
[15:09:45] <cradek> that's the whole point
[15:09:50] <SWPadnos> max velocity is a setting which is specifically intended to affect G0
[15:10:03] <SWPadnos> several people had asked for a separate feed override for G0
[15:10:04] <cradek> you can slow the motion way down, and watch DTG etc, and check your program as it runs
[15:10:31] <SWPadnos> it's for the "let's hope we clear the clamps" stage of program development
[15:10:41] <cradek> on my fast lathe, maxvel is controlled by the jog wheel, and it's very easy to safely test a program
[15:10:58] <cradek> SWPadnos: for me, "wonder if I got the tool offsets right?"
[15:11:07] <SWPadnos> yep. about the same stage ;)
[15:14:05] <Sordna> g-code question...can a single g-code "program" call/load/reference other g-code "programs"...example...one program moves the cutter head to a specific positions, then calls another to do a small cut on the target, then the main program moves to next location, calls sub-program, etc etc...each program being a separate g-code file...
[15:14:16] <pilotltd> 20+ years of CNC'ing and never used it. I just air run at full speed. I think I've used single step mode a handful of times debugging other peoples poor code :)
[15:14:44] <SWPadnos> Sordna, sort of :)
[15:15:07] <Sordna> hehe.
[15:15:12] <Sordna> was afraid of such a question.
[15:15:13] <SWPadnos> you can call subroutines that are in other G-code files with O<name> call
[15:15:23] <Sordna> aahh...kewl...awesome actually.
[15:15:34] <SWPadnos> the code needs to be in your nc-files directory, and has to be named <name>.ngc
[15:16:06] <Sordna> just started reading up on g-code to build(from scratch or scavenged from other code) a g-code interpreter on a microcontroller.
[15:16:27] <anonimasu> hm, g0 should be affected by FO
[15:16:46] <anonimasu> unless you specifically disable it, as far as I know
[15:16:48] <Sordna> the idea is to load the g-code files onto an SD, stick it in the box, and let it cut and run.
[15:16:49] <cradek> anonimasu: it is, and also the max velocity slider
[15:16:59] <Sordna> without a large PC needed to run the device.
[15:17:18] <anonimasu> because if you are slowing down a program for cutting air.. you want g0's not to crash your machine
[15:18:08] <cradek> Sordna: sounds like my mill (early 80s technology)
[15:18:18] <pilotltd> I know my program extents from CAM - how wouyld I crash - move far enough away not to hit anything
[15:18:39] <cradek> pilotltd: an incorrect tool offset
[15:18:40] <anonimasu> shit happens :)
[15:19:00] <Sordna> cradek, kinda building whole thing from scratch from old copier parts...and avr microcontrollers.
[15:19:33] <cradek> Sordna: it has a lot of limitations and tradeoffs because of its wimpy computer. consider that using a PC is probably a feature, not a bug
[15:19:45] <cradek> just my opinion though
[15:20:07] <Sordna> well, looking at the fact that all my new machines have no parallel or serial ports actually.
[15:20:18] <anonimasu> here's a ting ot think about
[15:20:25] <Sordna> and it's to be a simple 3 axis woodworking device.
[15:20:39] <anonimasu> do you want to mess around reinventing the wheel and spending time to make a solution that kindof works..
[15:20:42] <anonimasu> or do you want to make parts?
[15:21:26] <Sordna> it's a hobby toy actually.
[15:21:27] <anonimasu> I mean, emc is good enough to use in a production env already..
[15:21:35] <pilotltd> Thats what G40 and G49 are for in init line, but if your table is wrong, you deserve to crash :)
[15:21:54] <Sordna> being built out of free bits I get from work when we strip down old copiers.
[15:22:22] <Sordna> and a learning experience...as well as a hobby project.
[15:22:50] <anonimasu> yeah, but implementing a gcode interpreter by yourself along with trajectory planning and that stuff.
[15:23:12] <Sordna> hence the possibility of hijacking other code
[15:23:27] <Sordna> just a funzie project mostly.
[15:23:54] <anonimasu> oh, im just telling you, did you look at the gcode interpreter? for example
[15:23:56] <anonimasu> :)
[15:24:04] <cradek> I can understand wanting to do it because of historical interest
[15:24:07] <Sordna> about to.
[15:24:26] <Sordna> it'll go alongside my neuvo-retro computer I'm almost done with.
[15:24:29] <cradek> if you work very hard you may be able to come up with something as good as one of the early 80s controls
[15:24:47] <pilotltd> Seen a classic on a Milacron - operator got a 1" drill and a 3" endmill mixed up in tool cassette - it threw the pallet through the side of the machine. That piece of code had been run over 100000 times, slow G0 wouldn't have saved it :)
[15:24:56] <Sordna> using avr microcontrollers to bit-bang 640x480 vga video, and a basic interpreter/os akin to the old commodore systems.
[15:26:10] <cradek> pilotltd: I always dry-run a program, watching/controlling each entry move from the tool change location, after I mess with the tools. This practice would in fact have caught that error.
[15:26:45] <pilotltd> Production environment, no time to do that, op got the sack
[15:27:31] <Sordna> now, if I could get a decent small avr32 board life would become very interesting as it can run linux....but am doing this with 8-bit 20mips microcontrollers.
[15:27:47] <cradek> no time to prevent accidents, just fire people (if they survive) sounds like poor practice to me. but whatever.
[15:27:51] <pilotltd> He has a sheet with cassette numbers and tool types, his responsibility to check before pressing cycle start.
[15:28:26] <Sordna> lucky he wasn't in the way of ye olde flying bit..
[15:28:39] <cradek> yep that works great if he's you and never ever makes mistakes - sucks if he's anyone else including me
[15:28:43] <cradek> haha
[15:29:06] <pilotltd> He was loading the one next to it, definitely a brown underpants moment
[15:29:57] <pilotltd> Cost $50K to fix the machine!!
[15:30:27] <anonimasu> shit happens
[15:30:38] <cradek> production environment, no time to prevent $50k of damage = false economy
[15:30:44] <cradek> also yeah, shit happens
[15:31:19] <anonimasu> I have my cam program make line numbers on safe lines
[15:31:21] <pilotltd> What else can you do - he has a sheet, he's done it a thousand times before, he's supoosed to chack and sign sheet that he's checked
[15:31:55] <anonimasu> so I can just fire it up and run a single part of the op to see that nothing is messed uo
[15:31:57] <anonimasu> up
[15:31:58] <frallzor> If he is the only one doing it, its bound to happen
[15:32:20] <frallzor> get too linear and you might mistakes =)
[15:32:24] <frallzor> *make
[15:32:37] <anonimasu> well, im not into production though, but being careful usually costs less
[15:33:33] <cradek> well we were debating the max velocity slider and whether it was unnecessary - clearly from the discussion it is not unnecessary
[15:33:50] <pilotltd> HOw would that help?
[15:33:51] <cradek> people can run production however they want, I don't care
[15:34:11] <cradek> how would what help?
[15:34:27] <pilotltd> Velocity slider in that situation?
[15:34:48] <cradek> I said my practice earlier is to use maxvel to control and check the tool approach after any time I mess with the tools or tool table.
[15:35:12] <cradek> if you turn it all the way down, the tool stops - you can look at DTG and see if it is right.
[15:35:33] <cradek> in your example, DTG would have said 3", and look at the tool - 1" from something - UH OH
[15:36:22] <cradek> it works especially well when hooked to the jog wheel - very easy to control and check the entry move
[15:36:28] <pilotltd> The program is 80K lines and has 17 tool changes, it takes 2 hours to run at full speed. Take a week doing that <G>
[15:36:47] <cradek> * cradek shrugs
[15:37:04] <cradek> whatever - but the max velocity slider is useful and not unnecessary/redundant
[15:40:51] <pilotltd> OK I concede - some may find it useful :)
[15:45:51] <BigJohnT> cradek: I do the same thing on my Anilam... crank the feed override to 10% and when I'm kinda sure things are going as I expect them to do then crank it back up to 100%
[15:47:30] <anonimasu> hm, maybe im designing my lathe the wrong way.. perhaps it's a better idea to flip the x axis over instead and get rid fo the shielding issues
[15:47:31] <anonimasu> of..
[15:48:28] <anonimasu> http://www.allslidecnc.com/index.5.jpg like they did tehre
[15:48:30] <anonimasu> there..
[16:08:44] <bjt-plasma> I'm trying to read an encoder on my 5i20 and not having any luck. I have this one from AD http://web5.automationdirect.com/static/specs/encoderld.pdf
[16:09:18] <bjt-plasma> I'm pretty sure it is wired up correctly but I don't see any counts in the HAL watch window
[16:11:21] <bjt-plasma> it is a 5 volt... do I have the 5i20 set up for 3.3 or 5 let me go see :/
[16:15:18] <bjt-plasma> :? what dummy set the 5i20 for 3.3 to use with a 4.5-5v encoder
[16:15:23] <bjt-plasma> must have been me
[16:18:28] <frallzor> hmm is it possible to hook my spindle (that I cant control via vfd) to my Gecko 540 and just use a relay to let the pc turn it on/off? =)
[16:19:00] <Jymmm> sure
[16:19:13] <frallzor> so itll just cut AC when not needed
[16:20:37] <Jymmm> The G540 has two general purpose outputs called OUTPUT 1 and OUTPUT 2 on the MAIN TERMINAL BLOCK. They are at Pos 5 and Pos 6 respectively on the terminal block. These outputs may be used to drive relay coils or for any other purpose. The outputs are rated at 1A and 50VDC maximum. Connect one end of the load to the output and connect the other end of the load to a positive DC voltage. This voltage may be the G540 power supply or it may be a separate power su
[16:21:00] <Jymmm> frallzor: ^^^^ RTFM =)
[16:21:28] <frallzor> Well I know THAT =) but if its a good idea to do what I asked
[16:21:38] <frallzor> simple brute on/off
[16:22:13] <Jymmm> Why wouldn't it be?
[16:22:33] <frallzor> might not be like by the spindle etc etc? =)
[16:22:56] <Jymmm> say that again?
[16:23:03] <frallzor> *liked
[16:23:43] <Jymmm> Is the spindle sensative? Does it want it's mommy or something?
[16:23:44] <frallzor> since it isnt a simple 230V in and then nothing more, 230V in then some unit that does 40V out to the spindle
[16:24:29] <frallzor> its as simple as the G540 might break if you kill DC instead of AC first =)
[16:26:55] <Jymmm> Your spindle needs to be able to handle full power starts of course, but they gecko is just controlling a relay. Safety wise you NEED to prevent the spindle from restarting if dc power is removed replaced. Usually in the form of a safety relay.
[16:27:16] <Jymmm> removed and replaced
[16:28:22] <Jymmm> s/they/the/
[16:29:52] <frallzor> hmm the spindle itself uses 40VDC from the controlunit, maybe it would be possible to find a nice vfd-unit that can play ball with it
[16:31:06] <BigJohnT> I use a potential relay on my BP to drop out the servos in case I loose the three phase power to the spindle
[16:32:22] <Jymmm> G540 has a 0-10VDC VFD control on it.
[16:32:27] <bjt-plasma> well the 5i20 was set for 5v but looked like 3.3 when the card was upside down and in the bottom slot of the MB
[16:32:38] <Jymmm> http://www.geckodrive.com/upload/G540%20REV3%20MANUAL.pdf
[16:50:41] <skunkworks> jepler: Are you getting more comfortable with wiring?
[16:51:31] <bjt-plasma> well I have 5v at the breakout board so that much works
[16:55:39] <bjt-plasma> I guess I'll have to rig up a BOB and see if I can get the encoder to work on the parallel port
[16:56:00] <skunkworks> bjt-plasma: what is the problem?
[16:58:28] <bjt-plasma> I don't see any counts from the encoder in the HAL configuration window
[16:58:57] <SWPadnos> and num_encoders is sufficient for that encoder to be enabled?
[16:59:32] <bjt-plasma> that is a good question, but I don't see any way to enable it
[16:59:32] <skunkworks> can you see the pins change state when you slowly spin the encoder?
[16:59:48] <bjt-plasma> with my fluke I see some change
[17:00:11] <skunkworks> is this hooked into the mesa card or printer port?
[17:00:13] <SWPadnos> bjt-plasma, you set the parameter "num_encoders" to some number on the loadrt like
[17:00:19] <SWPadnos> line
[17:00:19] <bjt-plasma> to a 5i20
[17:00:22] <skunkworks> ah
[17:00:49] <bjt-plasma> I have a couple of encoders
[17:00:57] <skunkworks> can't you monitor the pin as gpio and see if it is changing state?
[17:01:04] <bjt-plasma> and I see them both in the HAL config window
[17:01:12] <bjt-plasma> let me see
[17:03:17] <bjt-plasma> all three are showing as on all the time as I spin the encoder at 1 rpm
[17:03:33] <SWPadnos> usde halscope or spin slower ;)
[17:03:37] <SWPadnos> gah. use ...
[17:03:59] <skunkworks> :)
[17:04:12] <bjt-plasma> the index stays on so suspect something is wrong
[17:04:12] <skunkworks> use halscope 'and' spin slower.
[17:04:27] <skunkworks> ;)
[17:04:35] <SWPadnos> yes. that does sound like something's wrong
[17:04:41] <SWPadnos> but it's hard to tell
[17:04:46] <SWPadnos> what's the resolution of the encoder?
[17:04:52] <bjt-plasma> 2500
[17:05:18] <SWPadnos> so that's 41.66 cycles or ~166 counts per second at 1RPM
[17:05:32] <SWPadnos> which is still too fast for userspace to reliably show you
[17:05:40] <bjt-plasma> but the index is once per min at 1 rpm
[17:05:51] <SWPadnos> and it lasts for 1/166th of a second
[17:05:58] <skunkworks> :)
[17:06:00] <SWPadnos> so it may never be noticed
[17:06:08] <SWPadnos> (by userspace)
[17:06:12] <bjt-plasma> s.001 now
[17:06:25] <Jymmm> SWPadnos: That's like stop motion man....
[17:06:25] <SWPadnos> see you in a hundred minutes ;)
[17:06:36] <SWPadnos> heh
[17:06:40] <bjt-plasma> lol cranking up halscope now
[17:06:47] <SWPadnos> err, 1000 minutes ...
[17:09:53] <bjt-plasma> even in halscope they stay on all the time as far as I can tell
[17:10:07] <SWPadnos> ok
[17:10:52] <SWPadnos> you would see some changes on the A and B lines at that slow rate of movement
[17:10:59] <skunkworks> I was able to see the 500 line encoder change state randomely in halmeter when turning the shaft by hand
[17:11:27] <skunkworks> tapping on the shaft
[17:11:56] <SWPadnos> sure, you should see something, but it's not conclusive if you're relying on X for display
[17:12:11] <SWPadnos> if you see nothing ,you have no new information. if you see something, you do
[17:12:13] <bjt-plasma> I disconnected the A and it is still on
[17:12:44] <bjt-plasma> in HAL config
[17:12:46] <SWPadnos> and if you manually twiddle that input?
[17:12:56] <SWPadnos> (assuming it's set to input ...)
[17:13:19] <SWPadnos> incidentally, is this going through a 7i(whatever) or is it direct?
[17:13:29] <bjt-plasma> direct to the 5i20
[17:13:45] <SWPadnos> ok, thought so but wanted to make sure
[17:13:47] <bjt-plasma> I wonder if I need to invert the input for the encoder
[17:14:06] <SWPadnos> if the line doesn't change, then inverting it will be an unchanging "other alue"
[17:14:10] <SWPadnos> value
[17:15:36] <bjt-plasma> this is the encoder http://web5.automationdirect.com/adc/Shopping/Catalog/Sensors_-z-_Encoders/Encoders/Light_Duty_Standard_Shaft_(TRD-S_Series)/TRD-S2500-VD
[17:15:52] <bjt-plasma> specs http://web5.automationdirect.com/static/specs/encoderld.pdf
[17:18:01] <bjt-plasma> the second encoder pins are all on as well and nothing is hooked to them
[17:22:02] <bjt-plasma> I have out A, B, and Z connected to the 5i20 and out A, B, and Z not floating
[17:22:32] <bjt-plasma> power to 5v and ground to 0v
[17:24:38] <skunkworks> the read/write functions for the mesa card are running?
[17:25:25] <bjt-plasma> I also have a stepper that I'm running from the mesa card that is working
[17:26:00] <bjt-plasma> I think I know what it is :/
[17:26:21] <skunkworks> heh
[17:26:31] <bjt-plasma> I have the breakout board plugged into the wrong cable
[17:26:48] <Jymmm> DOH
[17:26:50] <bjt-plasma> it should be on P2 not P4
[17:27:17] <bjt-plasma> * bjt-plasma smacks his forehead on the desk
[17:27:37] <Jymmm> * Jymmm moves the desk closer to bjt-plasma's forehead.
[17:29:54] <bjt-plasma> * bjt-plasma gets a marker out and scribbles on the cables which one they are
[17:31:16] <bjt-plasma> be back in a bit after some wire swapping :)
[17:33:09] <Jymmm> Man python is ugly when viewed as plain text.
[17:38:31] <Sordna> it's even uglier when you rip out all the tabs!
[17:43:09] <bjt-plasma> I think I should refrain from playing with power tools, guns, and explosives today...
[17:43:31] <bjt-plasma> and play with encoders now that I have it plugged in to the right cable
[17:44:34] <mozmck> does it work now?
[17:44:45] <bjt-plasma> yes :)
[17:44:56] <skunkworks> :)
[17:45:48] <bjt-plasma> spinning at 1500 rpm at the moment
[17:46:58] <mozmck> I did something similar the other day. had the wrong cable plugged into the computer and couldn't get anything to work - tried two different cards and wasted an hour or 2 before I finally figured it out.
[17:51:40] <bjt-plasma> 2000 rpm now
[17:51:56] <bjt-plasma> says it will do 3600 rpm with 70volts
[17:53:53] <bjt-plasma> my small stepper will do 4500 rpm at 70 volts, maybe I should make a small pcb mill...
[17:56:13] <frallzor> where do you find nice info like that?
[17:57:33] <Jymmm> frallzor: From the drunk and passed out guy at the end of your local bar or tavern.
[18:00:10] <bjt-plasma> from the place where I purchased the steppers
[18:00:35] <frallzor> lukcy you, Im happy I got a crappy schematic =)
[18:01:10] <bjt-plasma> that's why I buy the expensive $19 steppers to play with for the docs
[18:01:32] <frallzor> thats expensive?
[18:02:17] <bjt-plasma> my nema 23 triple stack cost me $39
[18:02:33] <Jymmm> bjt-plasma: url?
[18:03:17] <bjt-plasma> http://web5.automationdirect.com/adc/Home/Home
[18:03:49] <frallzor> my triplestack cost me $100 sek>usd conversion
[18:04:01] <bjt-plasma> ouch
[18:04:05] <bjt-plasma> sek?
[18:04:11] <frallzor> swedish currency
[18:04:17] <frallzor> and for me that was cheap
[18:04:26] <bjt-plasma> ok
[18:04:29] <frallzor> but thats pretty useless to compare like that
[18:04:41] <frallzor> different economies and currencies =)
[18:09:29] <Jymmm> bjt-plasma: These? http://web5.automationdirect.com/adc/Shopping/Catalog/Motion_Control/Stepper_Systems/Motors_-z-_Cables/STP-MTRH-23079
[18:10:06] <bjt-plasma> yep the 23079's I use on the plasma cutter
[18:10:45] <Jymmm> what voltage?
[18:11:14] <bjt-plasma> 63v I think on the plasma and 24v on my desk to play with
[18:12:23] <bjt-plasma> now to see what the manual says about hooking up spindle sync motion :)
[18:28:59] <anonimasu> frallzor: 80 usd
[18:29:46] <frallzor> ? =)
[18:31:02] <bjt-plasma> Sweet! my mockup lathe just ran the g76.ngc example program
[18:31:50] <frallzor> is there some calibrationsetting in emc that lets you jog each a
[18:32:01] <frallzor> axis to its limit to get the values right
[18:32:29] <bjt-plasma> to get which values right?
[18:32:48] <frallzor> travel
[18:32:58] <frallzor> from A to B
[18:33:56] <SWPadnos> no
[18:33:56] <bjt-plasma> for setting your scale?
[18:34:56] <frallzor> nope, real values in mm, bit SWPadnos answered
[18:35:00] <frallzor> *but
[18:35:10] <SWPadnos> a ruler will tell you that
[18:36:02] <frallzor> yup, but I just wondered if there was a function like that =)
[18:36:14] <SWPadnos> it's not possible to have a function that does that
[18:36:43] <SWPadnos> you have one variable (the number of steps/encoder ticks moved) and two unknowns (the distance moved and the counts per unit of distance)
[18:37:20] <frallzor> isnt it possible to make a motor complete a revolution exact?
[18:37:24] <SWPadnos> so you have to enter at least one of the numbers (scale or distance traveled) for it to even be possible
[18:37:42] <SWPadnos> well, with encoders that have index it could be done
[18:37:46] <SWPadnos> probably
[18:38:25] <SWPadnos> but one revolution of the motor could be almost any distance
[18:38:42] <frallzor> not if you know how much the table travels for 1 rev
[18:38:43] <SWPadnos> it's dependent on gearing and the screw (or pinion ...)
[18:38:58] <SWPadnos> as I said, you have to enter one of those values ;)
[18:39:10] <frallzor> but it could work right? =)
[18:39:20] <SWPadnos> yes, it's possible
[18:39:24] <SWPadnos> not implemented though
[18:39:57] <frallzor> but on the other hand, it wouldnt know when to stop
[18:40:05] <frallzor> so Id be screwed anyway =)
[18:40:07] <frallzor> (clear
[18:40:16] <SWPadnos> you would have to jog some distance and tell EMC how far that distance was
[18:40:19] <frallzor> clearly that would be an issue for me
[18:40:22] <SWPadnos> then EMC could calculate the scale
[18:40:52] <frallzor> im just brainstormning, not much to do until tomorrow =)
[18:41:08] <SWPadnos> you could potentially then jog to the axis limits and tell EMC "here's the minus limit, which I want to call 0", and "here's the positive limit"
[18:41:10] <SWPadnos> sure
[18:41:27] <frallzor> with some luck I have a cnc-converted little proxxon then :P
[18:42:38] <Jymmm> SWPadnos: Then hope that the temp doesn't go from -40 to 120 anything soon =)
[19:18:39] <L84Supper_> L84Supper_ is now known as L84Supper
[19:23:22] <bjt-plasma> hmmm, G33 seems to work so I think I have all the spindle feedback hooked up ok but the G76.ngc just skips by the threading part
[19:23:36] <bjt-plasma> in dev
[19:29:38] <bjt-plasma> # Spindle Feedback
[19:29:40] <bjt-plasma> net spindle-position hm2_5i20.0.encoder.01.position => motion.spindle-revs
[19:29:41] <bjt-plasma> net spindle-velocity hm2_5i20.0.encoder.01.velocity => motion.spindle-speed-in
[19:29:43] <bjt-plasma> net spindle-index-enable hm2_5i20.0.encoder.01.index-enable <=> motion.spindle-index-enable
[19:30:13] <bjt-plasma> this is what I added to get my spindle feedback is there anything else?
[19:39:14] <bjt-plasma> also in the g76.ngc if you have any tool diameter in tool 1 you get an error
[19:46:50] <jepler> bjt-plasma: I agree, g76.ngc doesn't seem to actually cut a thread
[19:47:39] <bjt-plasma> do you think the g76 line is not correct?
[19:49:23] <bjt-plasma> if you comment out line 26 g0z0 then tool 1 can have a nose radius/diameter in the tool table
[19:50:55] <jepler> yeah I think that's expected (I dunno what tool dia it's written for)
[19:51:47] <jepler> not sure though
[19:52:25] <bjt-plasma> was that for the old tool compensation where you needed a straight move to enter the tool comp
[19:54:05] <jepler> looks like I'm the one who porked it up.
[19:54:07] <jepler> 5e1a5fb05ca774723ab9ed475e0c66147ed076cc is first bad commit
[19:54:07] <jepler> commit 5e1a5fb05ca774723ab9ed475e0c66147ed076cc
[19:54:07] <jepler> Author: Jeff Epler <jepler@unpythonic.net>
[19:54:09] <jepler> Date: Sun Jul 5 18:10:39 2009 -0500
[19:54:12] <jepler> reject invalid I, J, K values in G76
[20:00:50] <jepler> and it's in 2.3.2 as well :-/
[20:02:50] <bjt-plasma> so g76 is broken for now?
[20:03:20] <jepler> yes, unfortunately
[20:03:40] <bjt-plasma> ok
[20:06:03] <jepler> you can 'git revert 5e1a5fb' to fix it, but please don't push the result -- I need to fix it properly
[20:12:16] <jepler> aha!
[20:17:00] <cradek> oh no
[20:19:19] <jepler> oh yes
[20:19:45] <cradek> sorry, I suck at testing.
[20:19:56] <jepler> - CHKS((block->i_number <= 0),
[20:19:57] <jepler> + CHKS((block->i_number == 0),
[20:19:57] <jepler> (_("In G76, I must not be 0")));
[20:20:07] <jepler> I suck at writing what I mean
[20:20:07] <cradek> sure, I saw it right away (now)
[20:20:58] <cradek> "just skips the threading part"? it should have errored.
[20:21:19] <jepler> yeah there's that
[20:21:24] <jepler> (I think I've fixed both :-/)
[20:57:15] <bjt-plasma> jepler: G76 works for me now in dev :)
[21:05:52] <bjt-plasma> If I have a 2500 cpr encoder what scale do I use for the encoder component for spindle feedback?
[21:06:18] <cradek> you want spindle-revs to increase by 1.0 for each full turn
[21:06:32] <cradek> did you get a lathe?
[21:07:01] <bjt-plasma> not yet but I have a simulated spindle driven by a stepper with an encoder attached to it :)
[21:09:09] <bjt-plasma> right now my scale is set at 2500 and motion.spindle.speed-in is 4 times motion.spindle.speed-out-rps
[21:09:45] <cradek> that doesn't sound right
[21:09:59] <bjt-plasma> hmmm
[21:11:21] <bjt-plasma> it is connected to hm2_5i20.0.encoder.01.velocity
[21:12:48] <skunkworks> is it 2500 line so should it be 10000?
[21:13:09] <bjt-plasma> yes, I think you have it
[21:13:18] <skunkworks> * skunkworks doubts if it is 625 line ;)
[21:13:33] <skunkworks> -if
[21:14:07] <bjt-plasma> so if it is 2500 lines you get 4x that?
[21:14:11] <skunkworks> yes
[21:14:14] <bjt-plasma> I suck at encoders
[21:14:29] <bjt-plasma> but I do have the numbers right now :)
[21:14:29] <skunkworks> for each trasision
[21:14:39] <skunkworks> transision
[21:15:00] <bjt-plasma> for A on, A off, B on, B off?
[21:15:06] <skunkworks> exactly
[21:15:18] <bjt-plasma> I got it now boss :)
[21:16:54] <bjt-plasma> now to make motion.spindle-at-speed functional I use something like the near component?
[21:17:34] <bjt-plasma> comparing motion.spindle-speed-in and motion.spindle-speed-out-rps?
[21:22:47] <bjt-plasma> boy at low rpm the encoder.nn.velocity bounces around a lot...
[21:31:06] <Jymmm> noise?
[21:44:50] <bjt-plasma> could be Jymmm
[21:45:29] <bjt-plasma> I don't have anything in a proper enclosure so the shield is not grounded to anything
[22:03:52] <skunkworks> I noticed that when I was playing around with this http://imagebin.ca/img/xCPTcJ.jpg
[22:06:13] <skunkworks> i figured it was a granularity(sp) problem at low rpm
[22:06:21] <Sordna> is the encoder debounced? Seems that if the encoder is not debounced you could get some noise and jitter.
[22:06:58] <bjt-plasma> yea, hostmot2 encoder has a filter
[22:07:22] <Sordna> well, looks like my $0.02USD is worth about half that then. [:
[22:09:32] <bjt-plasma> :)
[22:11:55] <bjt-plasma> that's kind of what I thought skunkworks
[22:12:23] <bjt-plasma> * bjt-plasma looks at wires jammed into the parallel port plug of skunkworks computer
[22:13:05] <Sordna> something wrong with that?
[22:13:40] <bjt-plasma> if you live in an earthquake zone yes
[22:14:57] <Sordna> bah.
[22:15:09] <Sordna> earthquake emergency disconnect system!
[22:18:26] <skunkworks> http://imagebin.ca/img/hR9ceTb7.jpg
[22:26:03] <bjt-plasma> * bjt-plasma can't wait for the high speed wireless string to be installed
[22:26:40] <bjt-plasma> skunkworks: what am I looking for in the pic?
[22:27:31] <bjt-plasma> ahh it's done downloading
[22:29:23] <skunkworks> just rpm (I like hal)
[22:29:59] <bjt-plasma> ok
[22:30:31] <bjt-plasma> good to see you again Dr. Chandler
[22:36:13] <cradek> jepler: the shop is cool if you want to come out
[22:42:45] <BigJohnT> cradek: do you have some beer and shrimp too?
[22:43:08] <SWPadnos> or tofu
[22:48:56] <BigJohnT> I'd rather have jello
[22:49:17] <SWPadnos> also animal-derived
[22:49:48] <BigJohnT> I love animals
[22:49:57] <SWPadnos> yes. they're tasty
[22:50:04] <BigJohnT> :)
[22:54:53] <cradek> beer yes, shrimp no
[22:55:08] <BigJohnT> almost time to cook some Korean Meatballs
[22:55:28] <BigJohnT> cradek: you don't eat anything with eyes excepts potatoes?
[22:55:58] <SWPadnos> heh
[22:56:15] <SWPadnos> as one riend of mine says, "if it has a face or a mother, I don't eat it"
[22:56:17] <Sordna> shrimp and lobsters...scavengers of the deep.
[22:56:18] <SWPadnos> friend
[22:56:23] <BigJohnT> * BigJohnT looks for his "heh" key on this old keyboard
[22:56:23] <Sordna> specially lobsters
[22:56:39] <Sordna> lobsters are the vultures of the sea.
[22:56:48] <SWPadnos> and crabs
[22:56:53] <BigJohnT> lobsters are only one branch from a roach
[22:56:56] <Sordna> true too.
[22:56:58] <dmess> i love them
[22:57:05] <Sordna> I won't eat them.
[22:57:12] <Sordna> mainly because I'm allergic to shellfish.
[22:57:17] <dmess> i do... with butter
[22:57:23] <SWPadnos> that would put a damper on it
[22:57:30] <SWPadnos> but I sure do like the taste
[22:57:37] <Sordna> I have the same reaction to shellfish that some people have to beestings..puts me into anaphalactic shock.
[22:57:40] <SWPadnos> shellfish are probably my favorites
[22:57:47] <BigJohnT> I prefer king crab the best
[22:57:54] <SWPadnos> yep, that's actualyl a fairly copmmon allergy
[22:57:57] <SWPadnos> gah
[22:57:58] <SWPadnos> common
[22:58:04] <Sordna> ayup.
[22:58:16] <Sordna> has put me in the hospital twice now.
[22:58:25] <BigJohnT> that sucks
[22:58:42] <Sordna> ayup.
[22:58:44] <BigJohnT> I also cook crawfish
[22:58:59] <Sordna> now, interestingly enough, I can eat those.
[22:59:04] <Sordna> fresh water lobsters..
[22:59:13] <Sordna> but still don't like the flavor.
[22:59:14] <dmess> not for me though.. i spent 2 weeks in Prince edward Island a few yrs ago and ate about 100 lbs of lobster those 2 weeks...
[22:59:28] <BigJohnT> then you have not had them cooked correctly
[22:59:32] <dmess> crawdads
[22:59:40] <Sordna> I've had them cooked many different ways.
[22:59:48] <BigJohnT> I cook them cajun style
[22:59:52] <Sordna> both by people who grew up living on them and a cajun chef.
[23:00:00] <dmess> that a way BJ
[23:00:03] <SWPadnos> ugh
[23:00:14] <Sordna> now frog legs!
[23:00:17] <Sordna> them's good eatin.
[23:00:27] <BigJohnT> if you don't like the taste you just don't like them
[23:00:28] <SWPadnos> anyone who has sat at the same table as my mother eating a plate of cajun crawdads would never eat one again
[23:00:35] <dmess> we got big bull frogs as a kid
[23:00:43] <SWPadnos> and might not eat anything else for a while either
[23:00:50] <BigJohnT> I don't like frog legs but my wife does
[23:02:14] <BigJohnT> if your standing next to a table full of crawfish your only watching your fingers so you don't bite them off :)
[23:02:55] <SWPadnos> not me
[23:03:04] <SWPadnos> I'm not too partial to them
[23:03:07] <Sordna> I grew up eating all kinds of seafood...fresh and salt water...
[23:03:08] <SWPadnos> too muck work ;)
[23:03:13] <SWPadnos> much
[23:03:19] <SWPadnos> too much muck too
[23:03:24] <BigJohnT> crabs are too much work for me
[23:03:46] <SWPadnos> not the way I eat them
[23:03:58] <BigJohnT> soft shell?
[23:04:01] <Sordna> now, when I was in highschool I worked for a fancy restaurant as a waiter...bought a bushel of king crab legs and took them home and cooked them up for my family before they got home that night.
[23:04:09] <SWPadnos> there's a restaurant here that makes "Eggs Arcadia", which is basically eggs benedict with crabcakes instead of ham
[23:04:09] <Sordna> my father made himself sick eating them things.
[23:04:19] <SWPadnos> BigJohnT, restaurant ;)
[23:04:32] <BigJohnT> LOL
[23:05:09] <Sordna> I got good ham steaks in the oven and the wifey and her mother are making salmon patties!
[23:05:12] <Sordna> EEEWWWW!
[23:05:37] <SWPadnos> salmon is my favorite fish
[23:05:45] <Sordna> smells like,,,
[23:05:47] <Sordna> FISH
[23:05:49] <SWPadnos> except for sushi, where I prefer mackerel
[23:05:51] <Sordna> eeww.
[23:05:57] <SWPadnos> and tuna is rigth up there too
[23:06:01] <SWPadnos> right
[23:06:02] <Sordna> sorry, only thing I eat rare is good beef.
[23:06:11] <Sordna> no way I eat fish raw.
[23:06:18] <SWPadnos> oh. well a good Carpaccio is also tasty
[23:06:27] <SWPadnos> or steak tartare
[23:06:37] <Sordna> just a good rare steak.
[23:06:48] <SWPadnos> (I'm relatively omnivorous, if you haven't guessed)
[23:06:58] <Sordna> carnivore here...land based.
[23:07:31] <Sordna> eat the occasional potato and corn and tomato.
[23:07:40] <BigJohnT> omnivore here
[23:07:46] <Sordna> once in a great while I'll eat lettuce
[23:07:52] <Sordna> though I do enjoy spinache...
[23:07:58] <Sordna> from it's natural habitat,,,,a can.
[23:08:43] <BigJohnT> hmmm, what was the git command to delete local files?
[23:09:46] <Sordna> no idea
[23:09:51] <Sordna> have yet to play with git.
[23:12:22] <SWPadnos> git rm, I believe