#linuxcnc | Logs for 2012-10-06

[00:00:19] -!- Jim_ [Jim_!~jsr@in-184-1-142-50.dhcp.embarqhsd.net] has joined #linuxcnc
[00:01:47] -!- Jim_ has quit [Client Quit]
[00:01:48] -!- logger[mah] has quit [Remote host closed the connection]
[00:01:55] -!- logger[mah] [logger[mah]!~loggermah@mail.mah.priv.at] has joined #linuxcnc
[00:01:56] -!- logger[mah] has quit [Remote host closed the connection]
[00:02:02] -!- logger[mah] [logger[mah]!~loggermah@mail.mah.priv.at] has joined #linuxcnc
[00:03:49] <skunkworks__> why oh why do people have so may problems burning/booting linuxcnc?
[00:04:17] <skunkworks__> I think I have had 2 or 3 computers that I just could not get to work out of hundreds
[00:05:50] -!- Nick001-Shop has quit [Quit: ChatZilla 0.9.89 [Firefox 15.0.1/20120905151427]]
[00:05:55] <PCW> Also with linux you can usually just move a hard drive around and it will just work on whatever hardware try that with windows
[00:09:27] <Tom_itx> skunkworks__, there was an issue with either too much ram or being connected to the interweb during install
[00:09:46] <Tom_itx> PCW, i've done that with windows a number of times
[00:10:14] -!- micges has quit [Quit: Leaving]
[00:11:13] <PCW> My experience is that if you change major hardware under windows it blue screens at boot
[00:11:15] <andypugh> PCW: Do you think that a multi-turn resolver should modify the scale?
[00:11:30] -!- morfic has quit [Ping timeout: 264 seconds]
[00:11:31] <skunkworks__> pcw: sure - done that...
[00:12:08] -!- asdfasd has quit [Ping timeout: 240 seconds]
[00:12:09] <skunkworks__> Tom_itx: I have had a XP move work about 20% of the time
[00:12:44] <andypugh> I am going to add a cycles-per-rev parameter to the resolver driver, so it only gives one index per rev. I am trying to decide whether I should incorporate that into the scale calcs, or leave that to the integrator?
[00:12:50] <skunkworks__> Tom_itx: the current livecd was fixed. (and 8.04 didn't have that issue iirc)
[00:13:08] <PCW> andypugh: Im not sure. if this is for LAIR his situation is that his resolver is geared up from the spindle so threading index does not work
[00:13:47] <andypugh> So presumably his scale is not the normal 65535.
[00:13:58] <andypugh> 65536, I mean
[00:14:50] <andypugh> But for a generic multi-turn resolver it may want to be an additional factor on top of the scale.
[00:15:40] <PCW> I woulds just have it patch the index (with the warning that since the index is no longer absolute dot try re starting a treading job after reloading Linxcnc)
[00:15:55] <PCW> s/dot/dont/
[00:17:03] <PCW> he claims the resolver is 1:1 with spindle and that it is 1x speed but I think it must be geared up 1:2
[00:17:30] <andypugh> 4 pole motors often have 2-cycle resolvers.
[00:17:40] <PCW> Yes
[00:17:51] <andypugh> I think cradek had a 256 cycle one.
[00:18:28] -!- gmagno has quit [Ping timeout: 244 seconds]
[00:18:40] <PCW> I have a I think a 256 cycle 3 phase resolver from a robot arm
[00:18:40] <andypugh> The "index" in the resolver driver just spots the wrap.
[00:19:17] <andypugh> So a 2-cycle resolver would show two indexes (indices?)
[00:19:38] <PCW> Yes (or one geared up 1:2)
[00:20:49] <PCW> this is the same issue with simulated index on encoders (its hard to patch around in HAL because index-enable is a bi-directional signal)
[00:23:04] <PCW> so you can fix this but you cannot guarantee that the threading wild start at the same place if you exit/re-enter linuxcnc
[00:23:29] <PCW> s/wild/will/
[00:23:46] <andypugh> No, but then you would deserve all you got if you tried :-)
[00:25:03] <PCW> maybe a more general solution would be a index mask input bit
[00:25:44] <skunkworks__> things like kflop don't use index - (he doesn't think it is necessary...)
[00:26:03] <PCW> (so for LAIRs case a external HAL flipflop would do)
[00:26:25] <skunkworks__> so everytime the machine is restarted - the index is in a different place.
[00:26:33] <andypugh> PCW: I don't think you can use a flip-flop, because of the bidirectionality.
[00:26:35] <Tom_itx> skunkworks__ i wasn't aware the 10.04 cd was fixed
[00:26:46] <skunkworks__> Yes - cradek did
[00:27:03] <PCW> thats OK for some things disasterous for others
[00:27:13] <skunkworks__> I don't like it.
[00:27:23] <skunkworks__> but it is a mach thing
[00:27:34] <skunkworks__> for the most part
[00:27:53] <PCW> andypugh the ff goes to the new index mask pin on the resolver driver
[00:28:22] <PCW> so a external hall or opto could be used as well
[00:28:56] <PCW> index mask is just an input
[00:31:06] <PCW> I guess it requires an index out bit as well for the FF clock
[00:31:07] <PCW> just though unbundling it would allow hardware index mask inputs
[00:31:09] <PCW> in addition to div by n
[00:32:29] <PCW> but it would require 2 new pins on the resolver hal interface (index-out, index-mask)
[00:34:48] <PCW> andypugh has fallen asleep...
[00:34:54] <skunkworks__> heh
[00:35:36] <PCW> I'm just about to fall asleep and he's 8 hours or so later
[00:35:51] <gmouer> hi Sam, any experience with remapping a unused g code? one is giving me fits, big John couldn't get it to work either
[00:36:30] <andypugh> Trying to get my new component to compile in both simulator and useful versions.
[00:36:32] <skunkworks__> gmouer: have not done that.. Might be just a little above my pay grade...
[00:37:24] <gmouer> I got the subroutine for the remap gcode so it runs from the mdi line, but the remap in the ini file apparently isn't calling it for some reason
[00:38:05] <PCW> does the simulator build have access to rtai includes?
[00:38:16] <andypugh> I am really not sure
[00:38:52] <andypugh> Well, I passed the tests, lets try building the debs...
[00:39:27] <PCW> you fixed the simulator build?
[00:40:41] <skunkworks__> gmouer: wait for mhaberler to be on - he is the one that did the remapping mod to linuxcnc
[00:41:32] <gmouer> thanks sam! I am writing that down LOL
[00:42:29] <skunkworks__> :)
[00:43:54] <andypugh> PCW: I can fix the sim build with an include that breaks the useful build
[00:45:18] <andypugh> I have tried to bodge round that with a #ifndef but it's a guess how that is meant to work
[00:49:31] <PCW> yuck. is it new includes that are causing the problem?
[00:50:12] <andypugh> It's the 64-bit division library that I am using to handle more than 8 digits of float
[00:50:32] <andypugh> It was a really ewasy solution in realtime...
[00:51:07] <andypugh> I think it just comes in as a side-effect of one of the standard realtime includes.
[00:53:22] <PCW> you would think that lots of things depend on "ifdef simulator" includes
[00:53:24] -!- gmagno [gmagno!~gmagno@a213-22-30-249.cpe.netcabo.pt] has joined #linuxcnc
[00:54:11] <PCW> not sure what a simulator build is good for...
[00:54:33] <andypugh> Making life really difficult.
[00:54:59] <andypugh> But it lets people run linuxCNC as an ordinarly application on a normal kernel
[00:55:28] <andypugh> Whether enough people who would want to do that know that they can do that and so do do that is a different question.
[00:55:32] -!- sumpfralle1 [sumpfralle1!~lars@c.mail.systemausfall.org] has joined #linuxcnc
[00:55:32] <PCW> is there a userland library that includes similar math functions?
[00:55:48] <andypugh> How should I know? Do I look like a geek?
[00:56:18] <PCW> well...
[00:57:08] <andypugh> I certainly don't really know my way round the linux standard libraries
[00:59:30] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[01:00:31] -!- morfic [morfic!~morfic@unaffiliated/morfic] has joined #linuxcnc
[01:02:30] <PCW> bbl ttgh
[01:03:30] -!- kmiyashiro has quit [Quit: kmiyashiro]
[01:03:31] -!- PCW has quit [Quit: ChatZilla 0.9.89 [Firefox 15.0.1/20120905151427]]
[01:04:10] -!- broofa has quit [Quit: Computer has gone to sleep.]
[01:10:35] -!- jthornton_ has quit [Read error: Connection timed out]
[01:11:57] -!- jthornton_ [jthornton_!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc
[01:12:55] <gmouer> Hi John, got the sub to run
[01:17:20] <gmouer> also think I figured out why a remap of G68 to a G10 command is not possible
[01:18:13] <andypugh> If it is G-code, I thought it was meant to be possible.
[01:18:35] <gmouer> nobody has had much luck on this one andy
[01:18:55] <andypugh> I admit I have not tried remapping anything.
[01:19:29] <gmouer> according to the remap doc, only gcodes in modalgroup 1 are supported, so I think that is why this one is giving us fits
[01:20:07] <gmouer> remap doc is a bit unclear on some of the finer points
[01:22:57] <andypugh> "Users familiar with Python will likely find it easier to write remapped codes, glue, O-word procedures etc in pure Python, without resorting to the somewhat cumbersome RS274NGC language at all."
[01:23:48] <gmouer> funny thing is that this is a real simple one-liner remap
[01:24:12] <andypugh> What modal group did you put your G68 into?
[01:25:13] -!- syyl [syyl!~syyl@p4FD15886.dip.t-dialin.net] has joined #linuxcnc
[01:25:23] <gmouer> bingo, that is my present concern, I used group 1 because that is all remap supports according to the doc, but.. a unassigned gcode does not really belong to a modalgroup at that point, does it?
[01:26:03] <andypugh> G68 is a version of G10, so I think it should be in Group 0
[01:26:06] -!- sumpfralle [sumpfralle!~lars@31-16-20-95-dynip.superkabel.de] has joined #linuxcnc
[01:26:49] <gmouer> I hope not... G10 is indeed group 0 I read, and only group 1 is supported by remap for g codes
[01:27:04] <gmouer> I think that is why it refuses to work
[01:27:41] <Tom_itx> what's another tooling fixture supplier aside from mcmaster?
[01:29:03] <gmouer> clear something up for me Andy, Why would unassigned Gcodes belong to any modalgroup? they are unassigned afterall????
[01:29:24] <gmouer> or does it go by the target codes it is being reassigned to?
[01:29:24] <Tom_itx> ahh, that's it.. carr lane
[01:29:30] <andypugh> The question is what group the _result_ belongs in.
[01:29:53] -!- sumpfralle1 has quit [Ping timeout: 255 seconds]
[01:29:57] <gmouer> I had a hunch that was the case
[01:30:16] <gmouer> its only been a few weeks since switching to linuxcnc but I am learning
[01:30:20] <andypugh> gmouer: Which version of LinuxCNC are you running?
[01:30:50] <gmouer> 2.6.0 pre0
[01:31:19] <gmouer> got that because of fixes Chris Morley did to pncconf
[01:31:20] <andypugh> That would have been too easy, I guess
[01:32:11] <gmouer> close reading of the remap doc gives the answer I think, only modalgroup 1 gcodes are supported
[01:32:40] <Tom_itx> any of you guys ever use these: http://www.carrlane.com/catalog/index.cfm/27025071F0B221118070C1C512D020609090C0015482013180B041D1E173C3B28535B465D
[01:32:41] <Tom_itx> ?
[01:32:47] <gmouer> there is some common belief that any unused gcode can be remapped, its looking like that is not necessarily true
[01:34:07] -!- gmagno has quit [Quit: Leaving]
[01:35:27] -!- jpk has quit [Ping timeout: 240 seconds]
[01:36:41] -!- FinboySlick [FinboySlick!~shark@squal.net] has joined #linuxcnc
[01:38:45] <Tom_itx> I saw similar of these today on a clamping fixture and i must say they're pretty cool: http://www.carrlane.com/catalog/index.cfm/28025071F0B221118070C1C513E111D081B0006280B1713050245221E0107070F1A3C3B285352425A56
[01:38:53] <Tom_itx> various types
[01:39:57] -!- maximilian_h has quit [Quit: Leaving.]
[01:39:59] <Tom_itx> these appear to be hydraulic though, what i saw wasn't but i can't seem to find em
[01:40:04] <skunkworks__> gmouer: be sure to ask mhaberler
[01:40:39] -!- jthornton_ has quit [Read error: Connection reset by peer]
[01:41:07] <gmouer> yea I will Sam, but is now looks like its just going to be a final comfirmation of no-can-do
[01:41:26] <andypugh> gmouer: I think you need some Python Glue, do you have that?
[01:41:33] <gmouer> I learned a lot more in attempting though.
[01:41:48] <skunkworks__> (he also likes a Challenge..) ;)
[01:42:08] -!- JT-Shop-2 [JT-Shop-2!~John@] has joined #linuxcnc
[01:42:08] -!- jthornton [jthornton!~john@] has joined #linuxcnc
[01:42:08] <gmouer> andy, I am real new to linuxcnc, don't even know what python glue is just yet, seen the term in my ventures though
[01:42:32] -!- JT-Shop has quit [Read error: Connection reset by peer]
[01:42:33] <andypugh> I am just quoting from MAH's docs :-0
[01:45:17] <gmouer> with luck, maybe I can catch him and get some confirmations about the situation
[01:47:05] <mozmck> Maybe it's like Gorilla Glue?
[01:47:59] <andypugh> But longer?
[01:49:51] <Tom_itx> andypugh do you have a link to your retention knob drawbar?
[01:49:57] <Tom_itx> or drawing of it
[01:51:17] <andypugh> https://picasaweb.google.com/lh/photo/Tad4D5FLiM1goCdswmXtddMTjNZETYmyPJy0liipFm0?feat=directlink
[01:52:45] <Tom_itx> thanks
[01:55:13] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[02:01:44] -!- jpk has quit [Ping timeout: 260 seconds]
[02:02:42] -!- Thetawaves [Thetawaves!~Thetawave@7-139-42-72.gci.net] has joined #linuxcnc
[02:09:26] -!- jthornton_ [jthornton_!~john@] has joined #linuxcnc
[02:09:27] -!- jthornton has quit [Read error: Connection reset by peer]
[02:09:27] -!- JT-Shop-2 has quit [Read error: Connection reset by peer]
[02:09:29] -!- JT-Shop-3 [JT-Shop-3!~John@] has joined #linuxcnc
[02:11:02] <r00t4rd3d> just won 25 bucks on a dollar scratch off :D
[02:12:34] <ReadError> yea but how many did you lose before that? ;p
[02:12:44] <r00t4rd3d> alot but that dont matter
[02:13:47] <r00t4rd3d> I got invited to a craft fair, lady says I can have my own table lol.
[02:13:57] -!- ktchk [ktchk!~eddie6929@n219078190004.netvigator.com] has joined #linuxcnc
[02:14:32] <r00t4rd3d> fucking craigslist.
[02:16:16] <r00t4rd3d> I think I am going to email her back and tell her, sorry but I am smoking some crack that day.
[02:19:32] <ReadError> set up a table with a wooden gun arsenal
[02:19:54] <r00t4rd3d> lol
[02:20:15] <r00t4rd3d> around here i would probably sell them
[02:21:58] <r00t4rd3d> I gave a hunting freak friend the 9mm I made and I guess people ask about it all the time.
[02:22:31] <r00t4rd3d> he hung it on his wall
[02:23:44] <r00t4rd3d> I could make a awesome one now
[02:23:51] <r00t4rd3d> fully detailed
[02:23:59] <r00t4rd3d> both sides
[02:25:17] <r00t4rd3d> Im gonna go fire up the machine and start another paradise box
[02:25:47] <r00t4rd3d> I got another one sold for 75
[02:25:55] <ReadError> a waaaah?
[02:25:56] -!- theorbtwo has quit [Ping timeout: 246 seconds]
[02:26:03] <r00t4rd3d> a stupid little box
[02:26:18] <ReadError> whats it do?
[02:26:22] <r00t4rd3d> nothing
[02:26:27] <r00t4rd3d> sits therer
[02:26:32] <r00t4rd3d> -r
[02:26:38] <r00t4rd3d> http://i.imgur.com/S47Q5.jpg
[02:26:56] <ReadError> fancy
[02:27:05] <ReadError> im glad to see you stop using particle board
[02:27:10] <ReadError> make me proud, sir
[02:27:17] <r00t4rd3d> http://www.vectric.com/WebSite/Vectric/projects/march-2011/paradise-box.htm
[02:27:23] <r00t4rd3d> those plans
[02:27:34] <r00t4rd3d> I am cutting this next one in oak.
[02:27:44] <r00t4rd3d> walnuts stain
[02:29:24] -!- syyl has quit [Quit: Leaving]
[02:29:35] <r00t4rd3d> this next one is getting some fish in the top lol
[02:31:16] <r00t4rd3d> like this:
[02:31:17] <r00t4rd3d> http://i.imgur.com/wmrQ8.jpg
[02:31:18] <ktchk> home position. Is it possible to home at the middle of the xy travel?
[02:31:29] <r00t4rd3d> sure
[02:31:34] <r00t4rd3d> i never home
[02:32:19] <ReadError> ktchk, you can setup switches
[02:32:21] <ReadError> and autohome
[02:32:25] <r00t4rd3d> ktchk, its in the menu, like File:Homing:Home X Axis
[02:32:44] <ktchk> soft home no switch
[02:33:39] <ReadError> uhhh
[02:33:44] <ReadError> impossibru
[02:33:49] <r00t4rd3d> i just zero everything to my material
[02:33:59] <r00t4rd3d> how ever the hell i clamp it in
[02:34:18] <r00t4rd3d> sometimes is a challenge
[02:34:25] <ktchk> that is g43
[02:34:56] -!- andypugh has quit [Quit: andypugh]
[02:35:10] <r00t4rd3d> off to the batcave!
[02:39:47] -!- gmouer has quit []
[02:40:56] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc
[02:43:45] -!- Eartaker_ [Eartaker_!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #linuxcnc
[02:43:45] -!- Eartaker_ has quit [Changing host]
[02:43:46] -!- Eartaker_ [Eartaker_!~Eartaker@unaffiliated/eartaker] has joined #linuxcnc
[02:43:51] -!- Eartaker_ has quit [Read error: Connection reset by peer]
[02:44:42] -!- andypugh has quit [Client Quit]
[02:44:57] -!- r00t-Shed [r00t-Shed!~CNC@cpe-67-249-21-218.twcny.res.rr.com] has joined #linuxcnc
[02:45:35] -!- JT-Shop-3 has quit [Read error: Connection timed out]
[02:46:49] -!- JT-Shop-3 [JT-Shop-3!~John@] has joined #linuxcnc
[02:57:25] -!- Keknom has quit [Quit: Leaving.]
[03:03:54] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[03:15:53] -!- ktchk [ktchk!~eddie6929@n219078190004.netvigator.com] has parted #linuxcnc
[03:25:00] <r00t-Shed> i hate when i use a ramp toolpath and forget to set the plunge rate higher.
[03:31:22] <r00t-Shed> now i sit here cutting at half speed on a large part with no way to over ride
[03:33:17] -!- bradsimantel has quit [Client Quit]
[03:44:39] -!- icarsufactor [icarsufactor!~factor@r74-195-220-46.msk1cmtc02.mskgok.ok.dh.suddenlink.net] has joined #linuxcnc
[03:45:06] -!- factor has quit [Ping timeout: 264 seconds]
[03:48:06] -!- icarsufactor has quit [Client Quit]
[03:48:28] -!- factor [factor!~factor@r74-195-220-46.msk1cmtc02.mskgok.ok.dh.suddenlink.net] has joined #linuxcnc
[03:52:54] -!- r00t-Shed has quit [Ping timeout: 240 seconds]
[03:58:12] -!- broofa has quit [Quit: Computer has gone to sleep.]
[04:10:33] -!- pjm__ has quit [Ping timeout: 245 seconds]
[04:23:47] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc
[04:27:52] <r00t4rd3d> lol, now even hulk hogan has a sex tape.
[04:33:15] -!- IAmWill [IAmWill!~roberts@cpe-72-130-208-210.hawaii.res.rr.com] has joined #linuxcnc
[04:35:56] <IAmWill> i'm not sure if this is the right place, but here goes: i am looking to purchase a mini cnc laser cutter. I currently have a cnc table that cuts 12"x18" and I am looking for the same rough size, and I want it to be compatible with emc. I do most of my designs in google sketchup, export to stl, and then use pycam to get my gcode. does anyone know of a linux compatible laser cutter? looking to spend around $2200.
[04:37:48] -!- charlie [charlie!4a6b7c66@gateway/web/freenode/ip.] has joined #linuxcnc
[04:38:12] charlie is now known as Guest85559
[04:38:31] Guest85559 is now known as charlieMiller
[04:39:16] <r00t4rd3d> just add a laser to your current cnc setup
[04:39:34] <charlieMiller> two lasers
[04:39:57] <charlieMiller> two beam lasers at 90 degree angles ot each other, that will give you dead center :)
[04:40:14] <charlieMiller> One on the front/back of the mill and one on the left/right side
[04:40:45] <r00t4rd3d> he wants to cut with the laser not position with it
[04:40:55] <charlieMiller> doh
[04:41:34] <r00t4rd3d> IAmWill, also lookup phlatscript if you use Sketchup
[04:41:55] <IAmWill> okay... thanks
[04:42:04] <charlieMiller> I hope you enclose your cnc setup for that
[04:42:06] <r00t4rd3d> it converts your models to gcode so you could skip the pycam part
[04:42:11] <charlieMiller> and be careful what you cut
[04:42:34] <charlieMiller> i.e. I don't think pcv is too healthy for you
[04:42:49] <r00t4rd3d> IAmWill, on your cnc machine you have now, does it have a 4 axis controller?
[04:43:15] <IAmWill> no
[04:44:06] <IAmWill> i am using the fireball v90, i built the controller myself with the kl4030
[04:44:22] <IAmWill> i am almost positive it only has 3 axis
[04:44:55] <r00t4rd3d> you could use a 4th axis to turn the laser on and off
[04:45:03] <r00t4rd3d> i think
[04:45:06] <charlieMiller> Do you have a c10 card?
[04:45:29] <IAmWill> no
[04:45:50] <charlieMiller> Another parallel port breakout setup?
[04:45:57] <IAmWill> yeah
[04:46:22] <r00t4rd3d> http://www.microcarve.com/mcMV1.html
[04:46:35] <charlieMiller> The lasers I have used had a ttl trigger 0-5v.. I know you can control a spindle that way. I don't know why you would need a 4th axis?
[04:46:37] <r00t4rd3d> if you got 2k , buy that and slap a laser on it
[04:47:31] <r00t4rd3d> charlieMiller, just so he could use his current cnc machine with out too much work
[04:47:32] <IAmWill> http://www.probotix.com/FireBall_v90_cnc_router_kit/ <-- already have this one... shouldn't be to hard to strap laser to it
[04:52:16] <IAmWill> actually.. i do the C10 board
[04:52:27] <IAmWill> http://www.kelinginc.net/ThreeXCNCPackage.html
[04:53:38] <charlieMiller> So of the lasers I have seen, you will have a driver board
[04:53:50] <charlieMiller> that driver board will have a power connector (AC or to a DC power supply)
[04:53:58] <charlieMiller> and a ttl interface (0-5v)
[04:54:10] <charlieMiller> you would want to connect that to one of the free output pins of the c10 board
[04:54:31] <charlieMiller> I haven't used linuxcnc for laser stuff, so I can't say what you would need to do for software configuration
[04:54:56] <charlieMiller> be careful with the laser too.. you should enclosure your machine, and don't look at the beam, yada yada
[04:55:35] <IAmWill> yeah... that is why i was thinking about just buying one.... plus, i move to japan in 6 months and it looks like i might be taking my projects inside since i wont have a garage
[04:56:08] <charlieMiller> They are pretty stinky to operate indoors
[04:56:13] <IAmWill> with it being enclosed, i can just hook a vacuum system to it
[04:56:26] <charlieMiller> Yea, it does help, but you will still smell it
[04:56:36] <charlieMiller> I knew someone who had one in his apartment
[04:56:51] <charlieMiller> would cut cardboard and oak plywood
[04:56:56] <IAmWill> is it the CO2? or the laser itself
[04:57:07] <charlieMiller> You won't smell the co2, it's whatever you are cutting
[04:58:14] <charlieMiller> and again, if you didn't know, you have to be careful what you cut
[04:58:15] <charlieMiller> pvc = bad as well as some other plastics
[04:58:16] <IAmWill> mostly going to be working with woods
[04:58:16] <charlieMiller> How thick?
[04:58:25] <IAmWill> but i will check the materials before i cut them... i didnt some plastics are bad.. (but then again, burning most plastics isn't healthy either)
[04:58:34] <IAmWill> at most, 1/4inch
[04:58:35] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[04:58:46] <IAmWill> some etching, some cutting
[04:59:04] <charlieMiller> That is pretty thick. So you also have to remember that at that thickness, you may have to do multiple passes, and pick up or drop the laser head
[04:59:13] <charlieMiller> the beam isn't perfect, it's v-shaped
[04:59:26] <charlieMiller> so if you don't have it adjusted properly, it won't cut all the way through because it isn't focused
[04:59:31] <charlieMiller> especially at 1/4
[04:59:46] <charlieMiller> 1/8" is ok, I don't know about double that.. the lasers I am familar with are 40w
[04:59:53] <charlieMiller> co2
[05:00:04] <IAmWill> so it would be good to have an automatic z axis as well...
[05:00:13] <IAmWill> yeah.. the ones i was looking at were 40w as well
[05:00:14] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:00:40] <charlieMiller> you will need to be able to focus regardless of what you cut too
[05:01:03] <charlieMiller> epilogue has this little dangly thing you drop down and then adjust the surface until it just barely touches the workpiece
[05:01:19] <IAmWill> http://www.emco.co.uk/product/Mini18
[05:01:50] <charlieMiller> I don't know about that one
[05:02:09] <charlieMiller> I used the zing24
[05:02:10] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:02:36] <charlieMiller> a buddy has one of these: http://fslaser.com/
[05:02:59] <IAmWill> yeah i was thinking about one of the FS ones too
[05:03:09] -!- Fox_Muldr has quit [Ping timeout: 260 seconds]
[05:03:16] <charlieMiller> So they work, but you get what you pay for
[05:03:36] <charlieMiller> I don't think they have a motorized z-axis
[05:03:41] <charlieMiller> you have to crank a wheel under the table
[05:03:44] -!- mutil [mutil!muti@] has joined #linuxcnc
[05:04:28] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-5f747d18.pool.mediaWays.net] has joined #linuxcnc
[05:04:29] <IAmWill> then i have this option.... http://www.buildlog.net/cnc_laser/index.php
[05:04:48] <IAmWill> since i will be in japan for 3 years... i will have plenty of time to build this damn thing and make it work
[05:05:17] <charlieMiller> There's a point where tinkering just becomes.. not worth it
[05:05:17] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:05:28] <IAmWill> haha... that is the truth
[05:05:29] <charlieMiller> and then there's the safety issue if you aren't familiar with electronics and lasers
[05:05:42] <IAmWill> electronics yes... lasers, no
[05:06:25] -!- BHSPiMonkey has quit [Read error: Operation timed out]
[05:06:25] -!- mutilator has quit [Ping timeout: 246 seconds]
[05:06:51] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:06:57] <charlieMiller> Have you seen: http://labs.nortd.com/lasersaur/
[05:07:56] -!- FinboySlick has quit [Quit: Leaving.]
[05:08:22] <IAmWill> no.. i need to do some research on this one
[05:08:59] -!- BHSPiMonkey [BHSPiMonkey!~BHSPitMon@68-185-203-185.dhcp.dntn.tx.charter.com] has joined #linuxcnc
[05:09:55] <IAmWill> brb
[05:10:21] <charlieMiller> So, I almost forgot the reason I came in here. I got some quick change tooling (tormach) for my mill. I've loaded the collets with a few tools and (offline) measured the tool heights and entered them into the tool table.
[05:10:47] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:11:05] <charlieMiller> When I do M06 T1, touch off the workpiece and then manually switch to another tool, I noticed that the tool height offset seems to be ignored
[05:11:14] <charlieMiller> This is also the case when running jobs and switching between tools
[05:11:37] <charlieMiller> How can I make linuxcnc recognize these tool lengths from the tool table?
[05:12:23] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:15:29] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:16:39] -!- plushy has quit [Ping timeout: 256 seconds]
[05:16:48] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:31:07] -!- IAmWill has quit [Ping timeout: 240 seconds]
[05:44:05] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:44:30] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:50:23] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:51:39] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:53:46] -!- dhoovie has quit [Read error: Connection reset by peer]
[05:54:05] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[05:59:31] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:01:17] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[06:18:47] -!- theos has quit [Ping timeout: 255 seconds]
[06:18:47] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:20:18] -!- charlieMiller has quit [Quit: Page closed]
[06:20:35] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[06:28:06] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:29:43] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[06:31:41] -!- vladimirek [vladimirek!~vladimire@] has joined #linuxcnc
[06:33:03] -!- theos [theos!~theos@unaffiliated/theos] has joined #linuxcnc
[06:40:58] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:42:25] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[06:47:18] -!- DJ9DJ [DJ9DJ!~Deejay@unaffiliated/dj9dj] has joined #linuxcnc
[06:47:26] <DJ9DJ> moin
[07:00:35] -!- plushy has quit [Quit: Leaving.]
[07:10:09] -!- dhoovie has quit [Read error: Connection reset by peer]
[07:11:27] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[07:30:40] -!- dhoovie has quit [Read error: Connection reset by peer]
[07:31:09] -!- IAmWill [IAmWill!~roberts@cpe-72-130-208-210.hawaii.res.rr.com] has joined #linuxcnc
[07:32:43] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[07:36:11] -!- dhoovie has quit [Read error: Connection reset by peer]
[07:36:29] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[07:39:58] -!- mk0 [mk0!~Geissler@] has joined #linuxcnc
[07:40:27] -!- dhoovie has quit [Read error: Connection reset by peer]
[07:41:58] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[07:44:40] -!- dhoovie has quit [Read error: Connection reset by peer]
[07:44:48] -!- Vq [Vq!~vq@90-227-210-195-no122.tbcn.telia.com] has joined #linuxcnc
[07:46:31] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[07:48:26] -!- Kent_ [Kent_!~kent@90-227-210-195-no122.tbcn.telia.com] has joined #linuxcnc
[08:00:15] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:02:09] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:06:12] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:07:50] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:08:10] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:10:13] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:12:58] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:14:54] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:15:21] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:16:51] -!- IAmWill has quit [Ping timeout: 256 seconds]
[08:17:18] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:18:11] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:20:04] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:20:18] -!- asdfasd [asdfasd!grgrgrgrg@] has joined #linuxcnc
[08:22:21] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:23:54] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:26:16] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:27:46] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:29:51] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:30:13] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:35:51] -!- mk0 has quit [Quit: Leaving]
[08:40:31] -!- vladimirek has quit [Ping timeout: 260 seconds]
[08:40:50] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:42:19] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:45:22] -!- dhoovie has quit [Read error: Connection reset by peer]
[08:46:57] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[08:49:36] -!- rob_h [rob_h!~rob_h@5e040606.bb.sky.com] has joined #linuxcnc
[09:03:31] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:05:10] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:13:13] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:15:04] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:20:42] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:20:47] -!- syyl [syyl!~syyl@p4FD15886.dip.t-dialin.net] has joined #linuxcnc
[09:22:37] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:27:48] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc
[09:31:16] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-101-95-dynip.superkabel.de] has joined #linuxcnc
[09:31:16] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:31:44] <IchGuckLive> pcw_home: ? are you oline
[09:32:19] Cylly2 is now known as Loetmichel
[09:32:50] <Loetmichel> mornin'
[09:33:02] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:33:04] <IchGuckLive> B)
[09:33:12] -!- IchGuckLive has quit [Client Quit]
[09:37:20] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:37:40] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:40:12] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:40:32] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[09:58:34] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:58:58] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:03:46] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:05:29] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:10:40] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:11:07] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:14:10] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:15:55] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:16:27] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:18:19] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:19:08] -!- Loetmichel has quit []
[10:21:21] -!- andypugh has quit [Quit: andypugh]
[10:21:39] jthornton_ is now known as jthornton
[10:24:05] -!- syyl_ [syyl_!~syyl@p4FD15886.dip.t-dialin.net] has joined #linuxcnc
[10:24:43] -!- syyl has quit [Ping timeout: 256 seconds]
[10:25:33] -!- Loetmichel [Loetmichel!cylly@p54B12ABA.dip.t-dialin.net] has joined #linuxcnc
[10:32:42] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:33:08] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[10:37:33] -!- pjm [pjm!~pjm@uhfsatcom.plus.com] has joined #linuxcnc
[10:39:53] -!- vladimirek [vladimirek!~vladimire@] has joined #linuxcnc
[10:42:59] -!- kmiyashiro has quit [Quit: kmiyashiro]
[10:46:58] <jthornton> dang he left...
[10:50:44] -!- dhoovie has quit [Read error: Connection reset by peer]
[10:52:34] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[11:02:34] -!- micges [micges!~toudi@edy132.neoplus.adsl.tpnet.pl] has joined #linuxcnc
[11:02:57] -!- dhoovie has quit [Read error: Connection reset by peer]
[11:04:55] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[11:07:24] -!- vladimirek has quit [Ping timeout: 244 seconds]
[11:16:47] -!- dhoovie has quit [Read error: Connection reset by peer]
[11:17:31] -!- sumpfralle [sumpfralle!~lars@31-16-20-95-dynip.superkabel.de] has joined #linuxcnc
[11:18:44] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[11:20:14] -!- dhoovie has quit [Read error: Connection reset by peer]
[11:22:01] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[11:22:36] -!- dhoovie has quit [Read error: Connection reset by peer]
[11:24:32] -!- dhoovie [dhoovie!~kvirc-aye@] has joined #linuxcnc
[11:39:17] -!- dhoovie has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
[11:53:53] -!- syyl__ [syyl__!~syyl@] has joined #linuxcnc
[11:56:37] -!- syyl_ has quit [Ping timeout: 245 seconds]
[11:59:45] -!- ybon has quit [Quit: WeeChat 0.3.8]
[12:00:07] -!- jpk has quit [Ping timeout: 246 seconds]
[12:11:01] -!- toudi_ [toudi_!~toudi@aedz251.neoplus.adsl.tpnet.pl] has joined #linuxcnc
[12:13:07] -!- syyl_ws [syyl_ws!~sg@p4FD1438A.dip.t-dialin.net] has joined #linuxcnc
[12:14:18] -!- micges has quit [Ping timeout: 245 seconds]
[12:17:53] toudi_ is now known as micges
[12:48:52] -!- gmagno [gmagno!~gmagno@a213-22-30-249.cpe.netcabo.pt] has joined #linuxcnc
[12:51:56] -!- BenceKovi11119 [BenceKovi11119!~BenceKovi@C2B0E442.dialup.pool.telekom.hu] has joined #linuxcnc
[13:15:04] -!- davec_ has quit [Quit: Leaving]
[13:42:38] -!- gmagno has quit [Ping timeout: 252 seconds]
[13:43:32] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #linuxcnc
[14:02:42] <Tom_itx> jthornton, you ever used those pull fasteners?
[14:05:23] <Tom_itx> http://www.carrlane.com/catalog/index.cfm/28025071F0B221118070C1C513E111D081B0006280B1713050245221E0107070F1A3C3B285352425A56
[14:05:30] <Tom_itx> mount em to a workplate
[14:05:35] <Tom_itx> in a grid pattern
[14:05:51] <Tom_itx> those are hydraulic but the ones i ran across were manual
[14:07:12] <awallin> is this for fixing a vise or rotary to a t-slot table?
[14:07:43] <Tom_itx> no, they're for a production workplate on a mill etc
[14:07:46] <Tom_itx> tooling fixture
[14:07:53] <Tom_itx> you can mount all sorts of things on them
[14:08:10] <Tom_itx> the OD thread is left and the stud is right hand thread
[14:09:12] <Tom_itx> make a grid on a plate ~2" centers then put them where required for each job
[14:09:30] <Tom_itx> on the odd 2" centers drill similar holes for solid pin mounts
[14:12:06] <Tom_itx> handier than a tee slot setup
[14:18:23] -!- broofa has quit [Quit: Computer has gone to sleep.]
[14:19:05] -!- mozmck has quit [Ping timeout: 244 seconds]
[14:21:38] -!- mozmck [mozmck!~moses@client-] has joined #linuxcnc
[14:26:06] <awallin> Tom_itx: on a big machine, just face-mill the whole table, then drill and tap directly the table? :)
[14:26:48] <Tom_itx> that would ruin the table
[14:27:03] <Tom_itx> we typically would bold aluminum workplates to them all
[14:27:54] <awallin> ok, what's the advantage of that fitting compared to a tapped whole in the workplate?
[14:27:59] <awallin> hole
[14:28:13] <Tom_itx> then if you _do_ have an oops it won't gouge the table either
[14:28:43] <Tom_itx> we used to do the tapped holes in the workplate as well
[14:28:53] <Tom_itx> sometimes you don't want that big a hole
[14:29:16] <Tom_itx> these smaller grippers allow you to use holes in the workpiece directly
[14:29:33] <Tom_itx> also, fixture clamps are made to use with these
[14:29:38] <Tom_itx> all types
[14:30:11] <Tom_itx> and you can use a stud or thread type screw mount
[14:32:30] <Tom_itx> on the odd grid, steel stud mounts are driven into the plate
[14:39:31] -!- gmouer [gmouer!~gmouer@cpe-74-65-14-255.rochester.res.rr.com] has joined #linuxcnc
[14:49:55] -!- BenceKovi11119 has quit [Ping timeout: 256 seconds]
[14:55:12] -!- FinboySlick [FinboySlick!~shark@squal.net] has joined #linuxcnc
[14:59:09] -!- ink_ has quit [Remote host closed the connection]
[15:17:36] -!- syyl_ws has quit [Quit: Verlassend]
[15:28:36] -!- dimas_ [dimas_!~dimas@] has joined #linuxcnc
[15:40:34] -!- micges has quit [Quit: Leaving]
[15:42:32] -!- gmagno [gmagno!~gmagno@a213-22-30-249.cpe.netcabo.pt] has joined #linuxcnc
[15:46:56] -!- BenceKovi11119 [BenceKovi11119!~BenceKovi@C2B0E4B8.dialup.pool.telekom.hu] has joined #linuxcnc
[15:48:38] -!- gmouer has quit []
[15:54:24] -!- theorbtwo [theorbtwo!~theorb@] has joined #linuxcnc
[16:22:25] * Loetmichel does a bit of aluminium drilling/milling for the company... (24 pieces) ... bad that i have broken my last 2,5mm drill...-> http://www.youtube.com/watch?v=nDFLvk_0UJ0 ... will look like this when done: http://www.cyrom.org/palbum/main.php?g2_itemId=13555
[16:23:26] <awallin> that looks like milling, not drilling?
[16:25:25] -!- factor has quit [Read error: Connection reset by peer]
[16:27:15] -!- theorbtwo has quit [Remote host closed the connection]
[16:27:32] -!- vladimirek [vladimirek!~vladimire@] has joined #linuxcnc
[16:29:52] -!- factor [factor!~factor@r74-195-220-46.msk1cmtc02.mskgok.ok.dh.suddenlink.net] has joined #linuxcnc
[16:34:54] -!- theorbtwo [theorbtwo!~theorb@] has joined #linuxcnc
[16:41:34] <FinboySlick> Loetmichel: Are you really doing 40k rpm in that one? http://www.youtube.com/watch?v=v59vyLhvj78
[16:41:59] <Loetmichel> oh, thats a mistype.
[16:42:07] <Loetmichel> 400Hz= 24kRPM
[16:42:40] -!- racycle has quit [Quit: racycle]
[16:42:52] <Loetmichel> awallin: if you have no 2.5mm drill, one takes a 2mm mill bit and "milldrilling" it to fit ;-)
[16:42:59] <FinboySlick> Loetmichel: OK.
[16:44:18] <Loetmichel> FinboySlick: better?
[16:44:53] <FinboySlick> Loetmichel: I wasn't complaining!
[16:45:08] <Loetmichel> no, but you were right to ask
[16:45:10] <Loetmichel> ;-)
[16:45:39] <FinboySlick> It's a puff of my inexperience floating away...
[16:46:01] -!- automata_ [automata_!~automata@triband-mum-] has joined #linuxcnc
[16:46:19] <Loetmichel> hrhr, have to make 24 of the aluminium sheets... *rangrangrangrangrang* *wiiip* *rangrangrangrangrangrang* ...
[16:46:22] <FinboySlick> Can't wait to get that mill head trammed so I can do some real cutting.
[16:46:26] <Loetmichel> that will take a while....
[16:46:59] <FinboySlick> Loetmichel: Bolt two more spindles on the sides of that one?
[16:47:13] <Loetmichel> not enough space on the table
[16:47:45] <Loetmichel> it would be cheaper to buy some TC drill bits in 2.5mm
[16:48:03] <Loetmichel> but boss wont do that fpr my private machine...
[16:48:29] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-101-95-dynip.superkabel.de] has joined #linuxcnc
[16:48:29] <Loetmichel> and i am not in the mood to donate the money to the company
[16:48:35] <IchGuckLive> hi all ?
[16:49:18] <IchGuckLive> pcw_home: ? online
[16:49:30] <FinboySlick> Loetmichel: Makes sense.
[16:49:50] mutil is now known as mutilator
[16:51:11] <Loetmichel> ... as i am already donating my free time and the electricity... and our Case builder/supplyer / sheet metal bender has donated ME the 1,5mm aluminium sheet ;-)
[16:51:32] <automata_> anyone trying xenomai on raspberry pi? I read that mhaberler was trying it out...
[16:51:57] <automata_> i would like to contribute to that effort..
[16:52:57] <mhaberler> well, a linuxcnc rtapi adaptation to xenomai would be real helpful; there's also two examples out already but they need to be brought back into the main code
[16:53:30] <automata_> I read that miniemc2 got xenomai going for mini2440
[16:53:49] <mhaberler> right, and then there's this french guy, let me look that up
[16:54:11] <mhaberler> unfortunately the miniemc2 code lacks commit history
[16:55:38] <automata_> if i understand correctly, rtapi is only a wrapper sitting on top of any realtime kernel. Also, I get the impression that jmkanusich is still around to answer quietions incase we get stuck!!
[16:56:44] <mhaberler> right
[16:57:20] <mhaberler> well given the examples it wouldnt be terribly hard I guess; for starters it could be based on linux-sim (I think)
[16:57:27] <automata_> well, miniemc2 is still GPL V2 so we can start with just the latest commit and see if we can bring it into our mainline
[16:57:49] <automata_> what is linux-sim?
[16:57:58] <mhaberler> the simulator mode
[16:58:05] <mhaberler> basically all userland
[16:58:08] <automata_> I just got the xenomai kernel to boot on my rpi...
[16:58:12] <automata_> oh ok...
[16:58:53] <awallin> is xenomai performance much better than rt-preempt ?
[16:59:00] <mhaberler> rtapi_app needs to become xenomai-rt aware (scheduling priority etc)
[16:59:09] <automata_> the xenomai kernel is from powet.eu/2012/07/25/raspberry-pi-xenomai/
[16:59:13] <mhaberler> in my experience, yes but I didnt push very hard
[16:59:36] <mhaberler> right, well I tried building my own, that recipe is ok
[16:59:42] <automata_> awallin: definitely xenomai out-performs rt_preempt..
[17:00:21] <mhaberler> but then - if you're running servo-only with outboard hw, it begs the question what for
[17:01:24] <automata_> I remember reading a white paper by some commercial company in boston on the comparison between the two and at-least lastyear xenomai beat the pants of rt_preempt under moderate to heavy load on the kernel
[17:01:27] <mhaberler> would you be willing to start a xenomai-integration branch off master?
[17:01:32] <awallin> so while you're at it, just write adapter libs from RTAPI to both xenomai and rt-preempt? :)
[17:01:54] <mhaberler> the rt-preempt case is almost identical
[17:02:07] <mhaberler> (provided you use only userland xenomai features)
[17:02:20] <automata_> raspberry pi may not be able to do stepper pulsing or even encoder counting...
[17:02:22] <mhaberler> I havent looked into xenomai kernel rt
[17:02:38] <mhaberler> no, the rpi isnt a great platform for hal
[17:02:56] <mhaberler> there's this issue with USB ports - need 8000 ints/sec to keep them alive
[17:03:04] <mhaberler> a each int eats 4-5us
[17:03:18] <automata_> Oh..
[17:03:33] <mhaberler> if rpi is the goal with stepgen: dont waste your time
[17:03:35] <automata_> that is a lot...
[17:03:51] <mhaberler> if xenomai support for linuxcnc is the goal: great - happy to contribute
[17:04:00] <automata_> stepgen on rpi is a waste of time.. i agree...
[17:05:21] <automata_> but getting xenomai support will open the door to many arm based boards (beagleboard/bone, cubieboard which is not yet released, wandboard, etc)
[17:06:13] -!- mhaberler has quit [Read error: Connection reset by peer]
[17:06:17] <automata_> first we should get xenomai with rtapi on x86...
[17:06:32] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc
[17:06:40] -!- mk0 [mk0!~Geissler@] has joined #linuxcnc
[17:07:34] <automata_> awallin: if we are opening up rtapi, getting both xenomai and rt_preempt shouldbe goal...
[17:08:21] <mhaberler> just reading back - sorry, my mac copped out
[17:08:30] <awallin> right. I understood that rtai isn't really developed that much anymore, so it seems good to future-proof linuxcnc in this way. I'm afraid I don't have much time to help though...
[17:08:31] <mhaberler> yes, I agree
[17:09:48] <mhaberler> I think what should be done is somebody takes the initiative and starts rt-preempt and xenomai branches off master, and goes ahead until we have something tangible - these splintered efforts here and there are pretty much wasted as far as the linuxcnc project is concerned
[17:11:26] <mhaberler> since sim_parport now exists, you can do basic parport I/O from the simulator mode already, and beef that up with rt-preempt/xenomai specific code
[17:12:17] <mhaberler> btw I dont think hal/rtapi for rpi is useless - it's just useless for stepper cnc and I dont see anybody connecting servo hw to an rpi just yet
[17:12:23] -!- gmouer [gmouer!~gmouer@cpe-74-65-14-255.rochester.res.rr.com] has joined #linuxcnc
[17:12:35] <automata_> I guess we should branch off 2.5.1 and start the out-moding rtai effort...
[17:13:21] * FinboySlick votes for that.
[17:13:36] <mhaberler> fine as well, I dont think there were many rtapi-related changes from 2.5 to master, but you'd have to backport sim_parport, but thats easy
[17:13:56] <mhaberler> actually I think it should run unchanged on 2.5
[17:13:57] <automata_> mhaberler: I would want to try to connect the Rpi with a FPGA card and then to servos!!!
[17:14:35] <mhaberler> so you'll have a €35 cpu driving a €350 FPGA card ;-?
[17:15:07] <automata_> well the FPGA card is really only $35!!
[17:15:47] <mhaberler> ok fine, but before you do this you might want to do want to check motion fp needs are in line with what the rpi cpu can offer; I would guess yes, but measuring is better
[17:15:56] <automata_> I mean the mesa parallelport 7i43's cards can drive 6 PWM + encoders...
[17:16:06] <mhaberler> (which fpga card, btw?)
[17:16:16] <automata_> I read about fp in kernel ...
[17:16:27] <mhaberler> fpga tail wagging the cpu dog
[17:16:44] <mhaberler> yes, but you want to make sure it has enough umph
[17:17:33] <mhaberler> fp performance of the rpi is about 1/7 of a atom core
[17:17:36] <automata_> Let me explain my system a bit: I use a 2 PC model for linuxcnc... One is the GUI and the other is the real-time stuff.
[17:18:42] <automata_> I have modified emcrsh and halrmt and added a halmonitor and an emcmonitor program which communicate with a GUI PC over ethernet.
[17:19:13] <mhaberler> oh man
[17:19:15] <automata_> the realtime PC is runnig headless with an FPGA tail...
[17:19:44] -!- toastydeath has quit [Read error: Connection reset by peer]
[17:20:16] <mhaberler> in linuxcnc3 I hope we shall return to a usable distributed model - the current scheme is.. well, leaves to be desired
[17:20:35] <automata_> Well my client does not want to sell software for linux!!! so I have developed a new GIU for him on windows!!!
[17:21:10] -!- toastydeath [toastydeath!~toast@c-69-244-86-195.hsd1.md.comcast.net] has joined #linuxcnc
[17:21:13] <mhaberler> paying clients .. 'Jump!' 'How high?'
[17:21:17] <automata_> I actually think that the current scheme is capable of running super distributed!!!
[17:22:06] <mhaberler> watch out - halrmt is super-broken, and you are heading seriously for race-condition land due to two independent comms mechanisms (NML AND halrmt)
[17:22:54] <automata_> We did a couple of machine for this client with a linux PC and his customers kept breaking the OS by loosing the shortcuts and basically being asses about runninglinux...
[17:22:58] -!- Kent__ [Kent__!~kent@90-227-210-195-no122.tbcn.telia.com] has joined #linuxcnc
[17:23:07] -!- Vq_ [Vq_!~vq@90-227-210-195-no122.tbcn.telia.com] has joined #linuxcnc
[17:23:13] <mhaberler> you have some stuff from hal reported in nml, some in halrmt - who was first, and what gives?
[17:23:31] <automata_> I actually have the whole solution under control!! and we have sold over 20 machines over the past 6 months and no complaints yet!!!
[17:23:46] <mhaberler> good for you - you are very brave
[17:24:04] -!- Vq has quit [Ping timeout: 260 seconds]
[17:24:10] -!- Kent_ has quit [Ping timeout: 246 seconds]
[17:24:30] <mhaberler> anyway - xenomai for 2.5.. how do we get going?
[17:25:35] <FinboySlick> automata_: What type of machines are they?
[17:25:35] -!- automata_ has quit [Read error: Connection reset by peer]
[17:25:44] -!- automata_ [automata_!~automata@triband-mum-] has joined #linuxcnc
[17:25:51] <FinboySlick> automata_: What type of machines are they? (re)
[17:25:54] <automata_> router crapped out...
[17:26:10] <IchGuckLive> Autsch O.O
[17:26:52] <gmouer> mhaberler: I have some remap questions that other said were best answered by you, no hurry though, I can wait
[17:26:58] <automata_> these are jewellary cutting machines...
[17:27:18] <automata_> we are cutitng gold bangles...
[17:27:23] <mhaberler> gmouer: sure
[17:28:46] <gmouer> myself and others tried a remap, G68 to G10 L2 P0 R180, the app was for a lathe rear toolpost, I believe its a illegal map, G10 being modal group zero, is that correct? no can do?
[17:29:28] <automata_> Mainly I use only emcrsh + emcmonitor for operation...
[17:30:17] <automata_> emcmonitor reads the status messages and just pushes any changes over ethernet...
[17:30:38] <mhaberler> G68 is a new code, right?
[17:30:54] <gmouer> yes, its a unused code presently
[17:31:16] <mhaberler> fine - well that should work - post setup files and error messages
[17:31:29] <mhaberler> or are you saying
[17:31:41] <mhaberler> 'g68 shall be modal group zero'?
[17:32:02] <automata_> xenomai + 2.5 :I think we should start with a beer with jmkanusich...
[17:32:25] <mhaberler> long flight for me
[17:32:34] <gmouer> well, it will do the same function as a G10 L2, and the G10 L2 is modalgroup zero, so I would assume the new G68 would be too????
[17:32:42] <automata_> Actually even longer for me... (I am in India)!!
[17:33:21] <automata_> mharerler: how do we start??
[17:33:23] <mhaberler> no, it is in the modalgroup specifiied as per http://www.linuxcnc.org/docs/devel/html/remap/structure.html#_the_remap_statement
[17:33:46] <mhaberler> for G codes, only mgroup 1 is possible
[17:34:12] <mhaberler> for M codes its one of 5,6,7,8,9,10
[17:34:33] <mhaberler> but eventually that doesnt matter if you put the code on a block of its own
[17:35:02] <mhaberler> it makes a difference only if you have multiple blocks on a line, then order of execution as per mgroup kicks in
[17:35:05] -!- motioncontrol [motioncontrol!~io@] has joined #linuxcnc
[17:35:12] -!- motioncontrol has quit [Client Quit]
[17:35:12] <gmouer> that is the part that confused me, for Gcodes, group1 only, G10 is existing/group0 G68 is the new code being remapped
[17:35:42] <gmouer> it wouldn't work with the G10 in the remap sub called, several other code lines worked ok
[17:35:48] <mhaberler> you arent remapping G10, you are using G10 in a remap procedure which is triggered by G68, which will be in mgroup 1
[17:36:19] <mhaberler> 'wouldn't work with the G10 in the remap sub called' isnt good enough an error description
[17:36:27] <gmouer> gottcha !!! that is what I thought originally, but others told be the opposite
[17:36:29] <mhaberler> what happens
[17:36:38] <mhaberler> where was that told?
[17:36:43] <gmouer> here last night
[17:36:48] <automata_> I have one fulltime engineer and myself available for the task... and we do take directions...
[17:37:14] <mhaberler> automata: hold on for a minute, need to make gmouer happy first..
[17:37:20] <gmouer> let me get into the gararge and I can post the code here, its very shot
[17:37:29] <gmouer> I will brb
[17:37:31] <mhaberler> excellent
[17:38:16] <mhaberler> automata: first the goal - is 'linuxcnc on a PC running xenomai' agreeable for now?
[17:38:31] <mhaberler> the rpi stuff must come after that
[17:38:46] <mhaberler> and you dont want to develop on the rpi anyway
[17:39:15] <automata_> In my current setup(s), we are putting a big-honking intelatom which is really only runningthe 1ms servo thread and some fp calculations ...
[17:39:42] <automata_> agreed: rpi comes after xenomai...
[17:41:19] <automata_> rpi is a secondary goal... ensuring linuxcnc's survival beyond intel atom and RTAI will make many ppl happy!!
[17:41:30] <mhaberler> ok, so a) need to pick or build a xenomai kernel for 10.04 - there's some stuff in the package stream but I have no idea how old that is; I would think the latest combination of 2.6.* and xenomai patches would be the starting point
[17:41:40] <mhaberler> actually I did this a while ago, let me see
[17:43:03] <mhaberler> right, the last combo of the 2.6 series I found was v2.6.35.9, and that built
[17:43:57] <mhaberler> I think I got it to boot and run, but I didnt probe much further than that; anyway lets assume you have a kernel
[17:44:26] <automata_> so versionlinux kernel
[17:45:02] <mhaberler> well yes, that was the latest 2.6 series code I found xenomai patches for, later stuff is for linux3.x
[17:45:22] <mhaberler> feel ok with kernel builds and such?
[17:46:58] <automata_> yup... done my fair share of 2.2 and 2.4 version kernel builds... not much should have changed since right?
[17:47:20] <mhaberler> the xenomai build docs and wiki n stuff are way better than any rtai arcana
[17:47:51] <mhaberler> no, not really, getting the config right might need some experimentation
[17:48:27] <mhaberler> anyway, now the tough part ;) adapting the linuxcnc build environement so xenomai is a recognized build option (bonus for autodetection ;)
[17:48:32] <automata_> I am going to use single core intel atom D425KT motherboard
[17:48:39] <mhaberler> fine
[17:49:16] <gmouer> ok, the RS274 line in the ini file is REMAP=G68 modalgroup=1 ngc=g68 will post the sub in a bit
[17:49:58] <automata_> I see.. so linuxcnc should automatically recognize if the current kernel is RTAI or xenomai?
[17:50:02] <mhaberler> what I'm not totally sure about (not enough xenomai source reading, me) is whether kernel RT comps are a viable option for xenomai; just steal from the miniemc2 folks and the french fellow to start with
[17:50:05] <mhaberler> ideally yes
[17:50:13] <mhaberler> but forcing it to start with is fine
[17:50:44] <mhaberler> I mean './configure --with-rt=<magic>
[17:51:01] <automata_> got it...
[17:51:14] <mhaberler> maybe it's in there anyway.. let me see
[17:52:20] <mhaberler> no, isnt
[17:54:18] <mhaberler> so it would be looking into src/configure.in section 2.1, it has some deprecated stuff for RTlinux which is history but no xenomai yet
[17:54:20] <automata_> ok.. so we create an option in the configure script..
[17:55:16] <mhaberler> right, make it either detect xenomai (say through a test program in the config script which uses xenomai libs and does some xenomai system call, like setpriority or so)
[17:55:19] <gmouer> here is the sub file http://pastebin.com/FsW2fftJ
[17:55:52] <mhaberler> gmouer: and what's wrong with that?
[17:56:14] <gmouer> it gives a "unknown gcode" error message
[17:56:38] <mhaberler> at the g10 line?
[17:57:11] <gmouer> typing g68 in the mdi window
[17:57:36] <mhaberler> let me verify that
[17:58:28] Vq_ is now known as Vq
[17:59:47] <gmouer> if I enter o<g68> call on the mdi line , it executes just fine
[18:00:10] <gmouer> so I would guess the sub file code is ok
[18:03:53] -!- automata_ has quit [Quit: Leaving]
[18:04:02] <gmouer> I am new to linux and linuxcnc, coming over from the mach3 crowd a few weeks ago, so I may not comprehend a lot
[18:04:09] -!- automata__ [automata__!~automata@triband-mum-] has joined #linuxcnc
[18:04:47] <gmouer> fwiw I am in lathe mode
[18:10:19] -!- automata__ has quit [Ping timeout: 260 seconds]
[18:10:51] -!- automata- [automata-!~automata@triband-mum-] has joined #linuxcnc
[18:12:44] <automata-> so we create a new configure option for xenomai...
[18:13:41] -!- cncbasher_ has quit [Remote host closed the connection]
[18:15:06] <automata-> mhaberler??
[18:15:23] <mhaberler> I am busy with gmouer, brb
[18:15:43] <gmouer> sorry for interrupting, it could have waited
[18:17:09] <automata-> no problem.. I will sort out the configure options in the mean time..
[18:19:52] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 14.0.1/20120713225625]]
[18:22:44] -!- Nick001-Shop [Nick001-Shop!~chatzilla@] has joined #linuxcnc
[18:25:24] <Loetmichel> sooo, 5 of 24 sheets done... end of work for today. my wife would kill if i dared to saw the aluminium sheets to fit the mill NOW (20:25 over here) ;-)
[18:26:45] -!- crib has quit [Remote host closed the connection]
[18:27:42] <mrsun> moving house ... "fun" =)
[18:28:00] <mhaberler> automata: probably you need to do this on a xenomai kernel and support libs built, do you have that already?
[18:28:05] <mrsun> alot of grease and steel to steel seems to be the melody
[18:28:10] <mrsun> alot of work to move it tho :P
[18:30:11] -!- BenceKovi11119 has quit [Ping timeout: 244 seconds]
[18:30:11] -!- crib [crib!~chris@port-92-202-73-189.dynamic.qsc.de] has joined #linuxcnc
[18:30:31] -!- crib has quit [Remote host closed the connection]
[18:35:05] <automata-> nope not yet...
[18:35:43] <automata-> will get that going on monday morning...no ;inux desktop at home...
[18:42:35] -!- ktchk [ktchk!~eddie6929@n219079226180.netvigator.com] has joined #linuxcnc
[18:44:13] <FinboySlick> automata-: Blasphemy.
[18:45:00] <automata-> :-)
[18:46:27] <mhaberler> gmouer: you found an embarrassing bug - fix forthcoming
[18:47:08] <gmouer> lol leave it to me !!
[18:47:53] <mhaberler> try REMAP=G68.1 for now and it should work (mehopes)
[18:49:02] <gmouer> ok, try right now
[18:49:58] -!- ktchk [ktchk!~eddie6929@n219079226180.netvigator.com] has parted #linuxcnc
[18:50:12] <mhaberler> I mean 'REMAP=G68.1 modalgroup=1 ngc=g68 '
[18:51:39] <gmouer> testing right now
[18:52:55] <gmouer> error msg all axis missing with motion code let me recheck my typing
[18:53:22] <mhaberler> no, that is correct.. G's need axes
[18:53:46] <mhaberler> what you want is some M200 or so, M's dont require axes
[18:53:56] -!- zzolo has quit [Quit: zzolo]
[18:54:11] <gmouer> I am on a lther, in lathe mode, does the G68 need axis? or in the remap?
[18:54:22] <gmouer> on a lathe
[18:54:23] -!- b_b has quit [Excess Flood]
[18:54:47] <mhaberler> all group 1 codes need axes; and sorry, we can currently only remap group 1 gcodes
[18:55:01] <gmouer> I wanted to use G68 and G69 because that are the common codes for lathes to select between the front and rear turrets, but if it does not work then a M code could certainly be used.
[18:55:57] <gmouer> can you give me a sample syntax for the G68.1 in the mdi ?
[19:00:19] <gmouer> nevermind, I got it G68.1 x0 z0 did the trick
[19:01:15] <mhaberler> let me see if that can be worked around
[19:01:30] <gmouer> sorry for being a PITA
[19:01:59] <mhaberler> no, very valuable - thanks. the fix is in master, pull and rebuild: http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commit;h=238a94b222d85668fc3c2928b8003367ee7ec156
[19:08:34] -!- vladimirek has quit [Remote host closed the connection]
[19:09:27] -!- crib [crib!~chris@port-92-202-73-189.dynamic.qsc.de] has joined #linuxcnc
[19:09:31] -!- crib has quit [Read error: Connection reset by peer]
[19:10:17] <mhaberler> gmouer: I fixed this 'all axes missing with motion code' if it's a remapped G code: http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commit;h=c24c98d58a491538ddf28e92e8dd26170ebd492c
[19:11:04] <gmouer> so that means axis are not required, just a plain G68 ??
[19:11:34] -!- automata- has quit [Ping timeout: 245 seconds]
[19:11:52] <gmouer> I am new to linux and linuxcnc, came over from mach3 only a few weeks ago, so I am still a dummy
[19:11:57] <mhaberler> yes
[19:12:04] <gmouer> wow fantastic !
[19:13:03] <gmouer> now for a dumb question, the fix, would that also be on the buildbot if I download the master deb? thats the only way I have updated thus far
[19:13:41] <gmouer> I am running 2.6.0 pre0 right now
[19:14:08] <mhaberler> yes, but it might take a bit of time, it's still building: http://buildbot.linuxcnc.org/buildbot/waterfall
[19:14:30] <mhaberler> did you pull a buildbot package?
[19:14:32] <gmouer> great, I will get it later
[19:15:11] <gmouer> can't thank you enough for your help!! this will make things simpler for front/rear turret lathes
[19:15:22] <mhaberler> sure, you're welcome
[19:15:40] <mhaberler> automata: still around?
[19:15:44] -!- micges [micges!~toudi@aedz251.neoplus.adsl.tpnet.pl] has joined #linuxcnc
[19:23:12] -!- phantoneD [phantoneD!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[19:25:39] -!- servos4ever [servos4ever!~chatzilla@173-87-49-187.dr01.hnvr.mi.frontiernet.net] has joined #linuxcnc
[19:26:12] -!- phantoxeD has quit [Ping timeout: 248 seconds]
[19:33:56] -!- mhaberler has quit [Ping timeout: 255 seconds]
[19:34:24] -!- crib [crib!~chris@port-92-202-73-189.dynamic.qsc.de] has joined #linuxcnc
[19:40:15] <skunkworks__> gmouer: isn't the support awesome? ;)
[19:40:23] <skunkworks__> I mean :)
[19:42:06] -!- ink [ink!ink@katsu.triplehelix.org] has joined #linuxcnc
[19:47:34] -!- MarkusBec has quit [Read error: Operation timed out]
[19:50:41] -!- MarkusBec [MarkusBec!znc@ist.deswahnsinns.de] has joined #linuxcnc
[19:52:03] -!- toastyde1th [toastyde1th!~toast@c-69-244-86-195.hsd1.md.comcast.net] has joined #linuxcnc
[19:53:33] <gmouer> hi Sam, the support is well beyond outstanding
[19:54:44] -!- mhaberler [mhaberler!~mhaberler@extern-183.stiwoll.mah.priv.at] has joined #linuxcnc
[19:55:18] -!- mrsun has quit [*.net *.split]
[19:55:18] -!- Spida has quit [*.net *.split]
[19:55:18] -!- theorbtwo has quit [*.net *.split]
[19:55:19] -!- L84Supper has quit [*.net *.split]
[19:55:19] -!- _ink has quit [*.net *.split]
[19:55:19] -!- t12 has quit [*.net *.split]
[19:55:19] -!- cpresser has quit [*.net *.split]
[19:55:20] -!- toastydeath has quit [Ping timeout: 265 seconds]
[19:55:59] -!- Spida_ [Spida_!~timo@ns1.spinnennetz.org] has joined #linuxcnc
[19:56:15] -!- theorbtwo [theorbtwo!~theorb@] has joined #linuxcnc
[19:56:15] -!- L84Supper [L84Supper!~ly@unaffiliated/l84supper] has joined #linuxcnc
[19:56:15] -!- _ink [_ink!~ink@c-67-170-200-44.hsd1.ca.comcast.net] has joined #linuxcnc
[19:56:15] -!- t12 [t12!~t12@dodeca-t.com] has joined #linuxcnc
[19:56:15] -!- cpresser [cpresser!~cpresser@rstenpresser.de] has joined #linuxcnc
[19:56:17] -!- Geissler [Geissler!~Geissler@] has joined #linuxcnc
[19:56:30] -!- r00t4rd3d_ [r00t4rd3d_!~r00t4rd3d@cpe-67-249-21-218.twcny.res.rr.com] has joined #linuxcnc
[19:59:03] -!- r00t4rd3d has quit [Ping timeout: 244 seconds]
[19:59:34] -!- mk0 has quit [Ping timeout: 244 seconds]
[20:00:39] -!- Spida_ has quit [*.net *.split]
[20:01:03] -!- Spida_ [Spida_!~timo@ns1.spinnennetz.org] has joined #linuxcnc
[20:02:20] -!- mrsun [mrsun!~mrsun@c-5f3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #linuxcnc
[20:11:58] Jim is now known as Guest86879
[20:35:10] -!- phantoneD has quit []
[20:35:26] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc
[20:35:29] -!- phantoxeD [phantoxeD!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[20:40:02] -!- charlesmiller [charlesmiller!4a6b7c66@gateway/web/freenode/ip.] has joined #linuxcnc
[20:41:13] JT-Shop-3 is now known as JT-Shop
[20:42:02] <charlesmiller> I recently purchased some tormach quick change tooling for my mill. I will be changing this tooling out manually. Offline, I have probed the length of the tooling and have entered it into the tooling. In my CAM program for each tool change I run T# M6; G43 H#
[20:42:32] <charlesmiller> It is definitely doing something, but not what I expect, something is happening too that causes updates to the tool table in linuxcnc which I don't think should be happening
[20:43:30] <charlesmiller> Assuming I have a valid tool table, what's the proper way to work with this? What is the proper way to touch off at the start of this setup?
[20:44:13] <charlesmiller> And should I be doing a G43 H0 to clear the current tool offset before swapping tools and doing the next G43 command?
[20:48:11] <andypugh> What you describe really ought to work.
[20:49:02] <charlesmiller> I assure you it's not, and I also assure you it is operator error :)
[20:49:45] <andypugh> I tend not to bother with the H number, but it is probably good practice to use it.
[20:50:25] <charlesmiller> I did see it was not necessary in linuxcnc, but it seems necessary for other setups so I included it.
[20:50:44] <andypugh> So, the tool table contains the length from a reference of each tool?
[20:51:17] <andypugh> Are there any G10 commands in the code that your CAM generates?
[20:53:28] <charlesmiller> Here is the code:
[20:53:30] <charlesmiller> http://pastebin.com/GiuCWQhS
[20:54:49] <charlesmiller> I have a touch probe that I ran before running that (I entered M6 T4; G43 H4)... h4 is the probe
[20:55:08] <charlesmiller> All of these were measured on the granite block and height gauge
[20:55:38] <charlesmiller> it's almost as if (without the G43 0) that the entire tool length is being added/subtracted from the curren tool height
[20:55:44] <charlesmiller> So I get drastic changes where I shouldn't
[20:55:47] <andypugh> And the tool table contains offsets from the length of the probe, or absolute lengths?
[20:56:33] <charlesmiller> Absolute lengths
[20:56:48] <andypugh> (It should be fine either way, as long as you have G43 H4 in place before touching-off to the probe.
[20:57:01] -!- gmagno has quit [Ping timeout: 246 seconds]
[20:57:02] <charlesmiller> I measured the value off-line on a piece of granite and using a height gauge for each of the tools
[20:58:32] <charlesmiller> I suppose it would be useful to know I'm using linuxcnc 2.5
[21:01:04] Geissler is now known as mk0
[21:01:06] <andypugh> What _should_ happen is this. Imagine you have a 20mm tool and a 10mm tool and a 50mm probe. You use the probe to set workpiece top to zero (using the probe offset). Your absolute machine position is now 50mm above the surface.
[21:02:42] <andypugh> If you then move the head to +50 then the absolute position is 100. If you then insert the 20mm tool and G43 then the DRO should change to 130. If you then G0 Z0 the tool would stop at the surface.
[21:03:32] <andypugh> It might be worth selecting "reload tool table" just to be sure.
[21:03:55] -!- factor has quit [Read error: Connection reset by peer]
[21:05:04] <charlesmiller> Let me re try this. I did the tool table stuff last night, turned off the machine and have tried this twice... lets see if the 3rd time is a charm
[21:06:46] <DJ9DJ> gn8
[21:07:50] -!- DJ9DJ has quit [Quit: by(t)e]
[21:08:52] <andypugh> Can you do it with 2 tools, and write down the lengths, and what the DRO values change to after the G43?
[21:11:01] <charlesmiller> Ok, so I ran it (with the G43 H0 if that makes any difference) and it worked)... although I do need to remember to add a command to lift the head so I can actually change the tooling haha
[21:11:16] <charlesmiller> I will try that manually now
[21:11:21] <charlesmiller> (the 2 tools thing
[21:11:41] -!- zzolo has quit [Quit: zzolo]
[21:15:15] <andypugh> You can predefine a safe tool change position.
[21:15:57] <charlesmiller> Yes, it is working now... I wonder what mode I got in that caused the tool table to change?
[21:16:00] -!- dscsoft has quit [Quit: Page closed]
[21:16:43] <charlesmiller> m06 t1;g43 t1; (zeroed in g54); m06 t5;g43 t5 and it did exactly what I expected
[21:16:51] <andypugh> See G30 for automatic head-raise: http://www.linuxcnc.org/docs/html/gcode/gcode.html#sec:G30-G30.1
[21:16:56] -!- theorbtwo has quit [Ping timeout: 252 seconds]
[21:17:02] <charlesmiller> g0 z0 went to the work surface
[21:17:31] <andypugh> Good, I am glad it works a bit.
[21:17:57] <andypugh> I suppose the next thing to do might be to touch-off to some air well above the table and run the code?
[21:18:29] <charlesmiller> I actually did that just now. So I have both manually changed this, and run the code and it worked as expected
[21:19:03] -!- factor [factor!~factor@] has joined #linuxcnc
[21:19:25] <andypugh> This is rather fun: http://youtu.be/5JCm5FY-dEY
[21:19:26] <charlesmiller> I don't know what happened compared to the earlier two runs, but those were vasty different. The best I could describe it is that instead of shifting the difference between a 1" tool and a 2" tool by moving 1", it was starting at 0" and shifting all the way up to the 2 inch mark
[21:20:19] <andypugh> I wonder if you had an incremental mode in?
[21:21:20] <charlesmiller> I suppose I could have fat fingered that
[21:21:52] <charlesmiller> That domino video reminds me of the variable piston/fluid setup where one can lift crazy forces
[21:23:04] <andypugh> Well, even in incremental move mode tool-length stuff should work the same.
[21:23:49] -!- micges has quit [Quit: Leaving]
[21:24:35] <charlesmiller> I've found another wierd issue after I installed my probe, but I think it's a hardware vs a linuxcnc issue. Basically if the probe touches during a non-probe move, it will freeze my machine
[21:25:20] <charlesmiller> That's ok, but what's scary is if you press an arrow key to move the machine in continuous mode, it can go really far (even after you stopped pressing the key) before stopping
[21:25:44] <charlesmiller> enough to destroy a bit despite only tapping a key
[21:26:00] <charlesmiller> I want to say it's because my HD is dying
[21:26:08] <charlesmiller> but I intend to investigate that more
[21:26:43] <charlesmiller> I.e. normally, a short button press would cause the machine to move a cm tops, but when this even occurs, it will move an inch
[21:27:57] <charlesmiller> er.. and this behavior happens after a "freeze event"
[21:28:19] <charlesmiller> like.. after the machine regains it's composure after the jog touch, it then has a little spazz moment
[21:29:11] <charlesmiller> I'll get a video of it later if I can
[21:30:54] <andypugh> I wonder if your RT kernel is working properly. Do you get any latency error messages?
[21:31:35] <andypugh> USB keyboard?
[21:32:30] <charlesmiller> I did not, I do have a usb keyboard. I didn't encounter this type of issue until I started using a probe.
[21:33:24] <charlesmiller> It only happens after "probe tripped during non-probe mdi command" and naturally I can't recreate the phenomenon now
[21:33:55] -!- chillly has quit [Quit: Leaving]
[21:34:02] <charlesmiller> oh, I can stimulate it to occur if I am ssh-ed into the machine
[21:34:32] <charlesmiller> perhaps I need to drop the base rate
[21:34:40] <charlesmiller> .. or increase it rather :)
[21:34:52] <charlesmiller> This machine doesn't need to move at 180ipm anyhow
[21:35:06] <andypugh> What is the base thread?
[21:35:50] <andypugh> I don't have a probe and don't know that part of the code, so can't help much.
[21:38:30] -!- asdfasd has quit [Ping timeout: 264 seconds]
[21:38:50] <charlesmiller> That's ok, I can figure that out. I just wanted to mention it. I do intend to dig deeper into it.. I would hate to find out it was a real bug vs operator error and endanger some of those x ton machines I see running this great software :)
[21:39:41] <charlesmiller> base time is 41599... max time for today is 27288
[21:40:14] <charlesmiller> Thanks for your help with the G43 stuff by the way
[21:42:26] <andypugh> If you can find a sequence that repeatedly causes the problem, then there is a bug tracker linked from the "community" page of the website.
[21:42:53] <jthornton> I use a probe on my plasma all the time with no problem... iirc a plug in keyboard
[21:43:33] <jthornton> cpu too busy to stop jogging?
[21:44:32] <charlesmiller> As andy said, I really need to do a better job re-creating the bug before saying more
[21:45:01] <jthornton> I've see the jog issue with a base thread too low before
[21:45:30] <jthornton> basically it just ties up the cpu and the keyboard inputs have to wait
[21:45:47] <charlesmiller> does linuxcnc pop up a warning if this happens? Or is the only way to open up the hal viewer/etc?
[21:46:13] <charlesmiller> Oh, and that reminds me, for debug/info messages that pop up in the lower left hand corner.. is there a keyboard shortcut to clear those?
[21:46:45] <charlesmiller> I've looked for a while for that one :)
[21:46:56] <jthornton> usually you will get a latency error when you start up Axis, but you only get one notification (if you got one for every occurrence you would get flooded perhaps)
[21:47:11] <jthornton> lower right corner?
[21:47:48] <jthornton> in Axis Help > Quick Reference
[21:48:10] -!- r00t4rd3d_ has quit [Quit: Leaving]
[21:48:36] -!- r00t4rd3d [r00t4rd3d!~r00t4rd3d@unaffiliated/r00t4rd3d] has joined #linuxcnc
[21:48:37] <charlesmiller> haha, I've been using linuxcnc for a year and never noticed that
[21:48:54] <charlesmiller> This is the first useful help menu item I have ever clicked
[21:49:01] <r00t4rd3d> lol
[21:49:10] <jthornton> that's why I sent you there :)
[21:49:29] <andypugh> charlesmiller: Ctrl-Space
[21:49:40] <r00t4rd3d> alt-f4
[21:49:48] <r00t4rd3d> fixes all your problems
[21:50:18] <jthornton> I gotta run now... and 327 spammers have registered
[21:50:35] <andypugh> Let me get them.
[21:50:39] <r00t4rd3d> more like 1 or 2 spammers created 327 accounts
[21:50:42] <jthornton> thanks
[21:50:46] <r00t4rd3d> with automated bots
[21:51:20] <charlesmiller> Have you tried adding a hidden field with a dumb text field in it? Like "what's 2+2"
[21:51:30] <charlesmiller> Users don't see it so they never enter the answer
[21:51:33] <charlesmiller> robots will
[21:51:45] <r00t4rd3d> charles, they use really bad forum software. Problem #1.
[21:52:26] <r00t4rd3d> really bad AND out of date. Its like, HEY SPAMMERS, OVER HERE!
[21:52:33] <charlesmiller> hmm
[21:53:15] <andypugh> I almost want pussy27 lesbo36D to be real
[21:53:21] <r00t4rd3d> keeps jt and andy buzy though
[21:55:05] <andypugh> 6 real humans, I think
[22:05:27] <r00t4rd3d> should just switch to phpBB or vbulletin
[22:06:44] <r00t4rd3d> https://www.phpbb.com/customise/db/modifications/anti-spam-9/
[22:07:09] <r00t4rd3d> would make the forums better and more secure.
[22:12:21] <r00t4rd3d> you can even bridge joomla to phpbb
[22:13:21] <r00t4rd3d> retain joomlas looks and features but have phpBB's backend and mod support.
[22:14:23] -!- jepler has quit [Quit: hardware maint.]
[22:15:57] <r00t4rd3d> not that I think that would be a good idea.
[22:16:24] <andypugh> "Should" and "Can" are different things. The chaps inconvenienced are not those with the admin rights.
[22:16:48] <andypugh> And I would be very nervous of my ability to effect a clean changeover too.
[22:17:00] <r00t4rd3d> I could do it
[22:17:23] <r00t4rd3d> and save your database so all the posts and users would be there
[22:17:39] -!- IAmWill [IAmWill!~roberts@cpe-72-130-208-210.hawaii.res.rr.com] has joined #linuxcnc
[22:17:50] <r00t4rd3d> but I would need root access
[22:19:04] -!- servos4ever has quit [Remote host closed the connection]
[22:19:12] <andypugh> And there be the problem. I don't even know who has root access. It's one of the comittee, you know those guys who we see so little of now?
[22:19:31] <r00t4rd3d> Ray Henry or Steve Stallings
[22:20:07] <andypugh> Possibly Alex.
[22:20:19] <andypugh> But the point is, we don't know.
[22:20:47] <r00t4rd3d> does cradek ?
[22:21:51] <r00t4rd3d> send a ping out on the mailing list.
[22:24:16] <r00t4rd3d> "hey can anyone move us over to phpBB?"
[22:24:46] <r00t4rd3d> then more people would probably use the forums
[22:27:52] <r00t4rd3d> you guys have the registration restrictions set so high its hard for a real user to register
[22:28:34] <r00t4rd3d> yet 327 bots can :/
[22:28:42] <r00t4rd3d> in one day
[22:28:59] -!- mk0 has quit [Quit: Leaving]
[22:29:21] <andypugh> Yeah. I am not sure how they get through the Captcha
[22:30:29] <andypugh> Luckily (and bizarrely) every single spam ID is something like r00t4rd3d r00t4rdly DW. It's really easy to scan through looking for names without a DW at the end.
[22:38:05] -!- jepler [jepler!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc
[22:41:17] -!- sumpfralle has quit [Ping timeout: 244 seconds]
[22:55:32] <gmouer> I had no problem registering at all
[22:56:55] <andypugh> There seems to be a problem with the confirmation emails, if you get that, then it's easy.
[22:59:43] -!- gmouer has quit []
[23:05:16] <Jymmm> archivist: You might appreciate this... http://www.youtube.com/watch?v=-S6gXmElHoI&feature=related
[23:08:39] <r00t4rd3d> http://i.imgur.com/yogcN.jpg
[23:16:08] <andypugh> You could make a great servo from that alternator :-)
[23:18:39] -!- Cylly2 [Cylly2!cylly@p54B132CC.dip.t-dialin.net] has joined #linuxcnc
[23:18:44] -!- Loetmichel has quit [Ping timeout: 248 seconds]
[23:19:22] <andypugh> Jymmm: Your next project, completley mechanical HAM radio
[23:23:13] <r00t4rd3d> thats gotta hurt: http://i.imgur.com/oySDv.jpg
[23:24:41] -!- theorbtwo [theorbtwo!~theorb@] has joined #linuxcnc
[23:26:54] <andypugh> You have to hope that the lever has snapped off.
[23:27:34] <r00t4rd3d> it didnt
[23:27:59] <r00t4rd3d> unrelated and very cool - http://www.mimoa.eu/images/12806_l.jpg
[23:28:25] <r00t4rd3d> unless there is a fire
[23:28:39] <andypugh> Where is that?
[23:28:46] <r00t4rd3d> no clue
[23:28:59] <r00t4rd3d> Curvaceous wooden room in the Eco Villa on the Thai island of Koh Kut
[23:29:01] <r00t4rd3d> ok, clue.
[23:29:03] <andypugh> It is particularly nice.
[23:33:47] <r00t4rd3d> I helped clean out a warehouse today, the people were like anything you find and you want, take it. I got like 50lbs of aluminum.
[23:34:27] <charlesmiller> nice
[23:35:00] <r00t4rd3d> i was like oh , paycheck #2
[23:36:18] -!- charlesmiller has quit [Quit: Page closed]
[23:48:03] <IAmWill> r00t4rd3d: is that the bicycle brake handle IN the knee?
[23:48:27] -!- asdfasd [asdfasd!grgrgrgrg@] has joined #linuxcnc
[23:48:50] <r00t4rd3d> yeah, not my photo though, reddit.
[23:52:45] -!- skunkworks__ has quit [Quit: ChatZilla 0.9.89 [Firefox 13.0.1/20120615040358]]
[23:53:05] -!- Cylly2 has quit []
[23:53:14] -!- Loetmichel [Loetmichel!cylly@p54B132CC.dip.t-dialin.net] has joined #linuxcnc
[23:54:09] -!- skunkworks__ [skunkworks__!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc
[23:58:25] -!- ybon has quit [Quit: WeeChat 0.3.8]