#emc | Logs for 2011-05-09

[00:03:24] -!- theorb [theorb!~theorb@] has joined #emc
[00:04:26] -!- theorbtwo has quit [Ping timeout: 258 seconds]
[00:04:30] theorb is now known as theorbtwo
[00:07:33] -!- mshaver [mshaver!~mshaver@c-76-21-166-2.hsd1.md.comcast.net] has parted #emc
[00:12:59] -!- vladimirek has quit [Remote host closed the connection]
[00:18:10] -!- robh__ has quit [Ping timeout: 240 seconds]
[00:22:56] -!- tom3p [tom3p!~tomp@ip-216-36-91-237.chi.megapath.net] has joined #emc
[00:54:53] -!- Birdman3131 [Birdman3131!~quassel@ip70-178-238-227.ks.ks.cox.net] has joined #emc
[01:30:18] <tom3p> anonimasu, hello
[01:32:40] -!- Aggrav9d [Aggrav9d!~aggrav8d@S0106000b6a15cda1.vc.shawcable.net] has joined #emc
[01:35:00] -!- aggrav8d has quit [Ping timeout: 240 seconds]
[01:37:21] -!- tom3p has quit [Quit: Ex-Chat]
[01:44:34] -!- Techrat has quit [Remote host closed the connection]
[01:46:33] -!- Techrat [Techrat!~TRat@c-24-2-169-60.hsd1.ct.comcast.net] has joined #emc
[01:46:33] -!- Techrat has quit [Changing host]
[01:46:33] -!- Techrat [Techrat!~TRat@unaffiliated/techrat] has joined #emc
[02:47:01] -!- rooks has quit [Read error: Operation timed out]
[02:56:41] -!- ries has quit [Quit: ries]
[03:00:05] -!- ries [ries!~ries@] has joined #emc
[03:00:07] -!- ries has quit [Client Quit]
[03:03:45] -!- ries [ries!~ries@] has joined #emc
[03:04:11] -!- ve7it has quit [Remote host closed the connection]
[03:06:23] -!- awallin [awallin!~quassel@cs27057046.pp.htv.fi] has joined #emc
[03:10:39] -!- awallin has quit [Remote host closed the connection]
[03:15:19] <The_Ball> Anybody made their own tool height touch off?
[03:36:11] <toastydeath> ?
[03:37:03] -!- awallin [awallin!~quassel@cs27057046.pp.htv.fi] has joined #emc
[03:44:15] <Valen> The_Ball: in the process of
[03:49:46] <Valen> The_Ball: http://www.cnczone.com/forums/calibration_measurement/109146-contact_probe_build.html
[03:55:58] <The_Ball> Valen, yes exactly
[03:56:02] <The_Ball> cheers
[04:07:57] -!- WalterN has quit [Quit: Leaving]
[04:15:32] <Valen> hard part is making it so there is enough spring force to close the contacts again but not so much that i'll break a .8mm tool
[04:22:01] <Valen> anybody have a link to the guy who made a hex type socket on his lathe with EMC?
[04:22:41] <Valen> nvm http://www.youtube.com/watch?v=T4q8gCpeY1A
[04:25:06] <The_Ball> Valen, oh wow that's impressive
[04:26:12] <The_Ball> Valen, is he turning the spindle by hand?
[04:27:37] <The_Ball> doh, rtc
[04:32:08] <Valen> yeah
[04:32:20] <Valen> emc = tha bomb
[04:34:37] <toastydeath> that is called slow slide servo or asymmetric turning
[04:37:58] <toastydeath> http://www.youtube.com/watch?v=t8-soGNPRS4
[04:41:59] <toastydeath> usually seen on air bearing + linear servo lathes
[04:44:03] <Valen> cool
[04:44:51] <toastydeath> i think there are a few roller bearing lathes that have it as well
[05:02:56] -!- WesBaker has quit [Ping timeout: 248 seconds]
[05:22:20] -!- awallin_ [awallin_!~quassel@2001:708:110:1020:224:7eff:feda:7c7d] has joined #emc
[05:27:21] -!- aggrav8d [aggrav8d!~aggrav8d@S0106000b6a15cda1.vc.shawcable.net] has joined #emc
[05:28:59] -!- Aggrav9d has quit [Ping timeout: 276 seconds]
[05:40:11] -!- izua has quit [Ping timeout: 240 seconds]
[05:43:35] -!- izua [izua!~izua@] has joined #emc
[05:43:35] -!- izua has quit [Changing host]
[05:43:35] -!- izua [izua!~izua@unaffiliated/izua] has joined #emc
[05:47:52] -!- izua_ [izua_!~izua@unaffiliated/izua] has joined #emc
[06:13:05] -!- e-jones [e-jones!~jkastner@nat/redhat/x-qrydyxqeiazghmib] has joined #emc
[06:17:42] -!- vladimirek [vladimirek!~vladimire@bband-dyn166.178-41-220.t-com.sk] has joined #emc
[06:55:32] -!- factor has quit [Quit: Leaving]
[06:58:24] -!- capricorn_one has quit [Remote host closed the connection]
[07:22:26] -!- awallin_ has quit [Read error: Operation timed out]
[07:23:31] -!- awallin_ [awallin_!~quassel@2001:708:110:1020:224:7eff:feda:7c7d] has joined #emc
[07:29:40] -!- awallin has quit [Read error: Operation timed out]
[07:42:47] -!- robh__ [robh__!~robert@5ace706f.bb.sky.com] has joined #emc
[08:20:07] -!- cradek has quit [Read error: Connection reset by peer]
[08:20:13] -!- cradek [cradek!~chris@outpost.timeguy.com] has joined #emc
[08:43:49] -!- pjm [pjm!~pjm@uhfsatcom.plus.com] has joined #emc
[08:47:10] -!- pjm__ has quit [Ping timeout: 252 seconds]
[09:24:03] -!- mhaberler [mhaberler!~mhaberler@imac.stiwoll.mah.priv.at] has joined #emc
[09:48:39] -!- yangliang [yangliang!~yangliang@] has joined #emc
[09:49:12] <yangliang> hello
[09:55:45] -!- yangliang has quit [Quit: Visitor from www.linuxcnc.org]
[10:01:20] -!- Surfcam123 [Surfcam123!EMCIRCUser@] has joined #emc
[10:01:50] -!- surfcam1234 [surfcam1234!EMCIRCUser@] has joined #emc
[10:01:58] -!- surfcam1234 has quit [Client Quit]
[10:02:29] -!- surfcam1234 [surfcam1234!EMCIRCUser@] has joined #emc
[10:02:35] -!- surfcam1234 has quit [Client Quit]
[10:11:47] -!- nicko [nicko!~nicko@125-237-79-227.jetstream.xtra.co.nz] has joined #emc
[10:14:21] <nicko> *ello all
[10:18:38] <Surfcam123> Hi
[10:20:49] <nicko> Hi, I'm wondering if anyone here is knowledgeable about the Mesa range of products ?
[10:26:03] <jthornton> a little
[10:26:54] <nicko> I'm new to EMC2 and motion control in general
[10:27:25] <nicko> I've just run EMC2 in an open loop capacity sending out pulses to drive that close their own loop
[10:28:14] <nicko> I want to use EMC to close the loop and understand that the mesa cards can afford me more speed
[10:28:37] <nicko> they can also allow me to operate two drives I have here that only take analog input
[10:29:05] <nicko> I also have another non-CNC application where I;m not sure if they would be helpful or not
[10:30:01] <nicko> the mesa site says "so it is truly an Anything I/O card"
[10:30:21] <nicko> how do you access the cards outside of EMC ?
[10:30:41] <nicko> a .dll or ?
[10:36:29] <awallin_> the emc2 driver is probably open source. as well as the fpga vhdl. if you want to take a look at them and modify them yourself
[10:36:50] <nicko> vhdl ?
[10:37:08] <awallin_> the cards have an fpga, a programmable logic chip. vhdl code is for that chip
[10:37:20] <awallin_> the vhdl defines encoder counters, pwm generators and such
[10:39:14] <nicko> ha - yet another language to learn !
[10:39:23] <nicko> is it like ladder logic ?
[10:39:46] <awallin_> you probably don't want/need to do vhdl for normal use of the cards
[10:40:20] <nicko> well, maybe I have an abnormal use...
[10:41:06] <nicko> Maya, the 3D program - usually used to render imagery for 3D movies - has a an almost perfectly built user interface for some projects
[10:42:48] <nicko> I want to build a program/plug in for it that pulls info from the current maya scene at time T then outputs it as either positional data from which velovity data can be derived from comparison with time T+1
[10:43:30] <nicko> then gets this out into the real world via 'something' - to some motors that move
[10:43:43] <nicko> the 'something' could be a mesa card
[10:44:50] <nicko> so I write something in C++/whatever - how do I access the cards/VHDL from the program ?
[10:45:17] <archivist> if using emc the just a series of g1 moves
[10:51:04] <nicko> 240 times a second ?
[10:51:25] <archivist> that slow :)
[10:51:43] <nicko> good to hear!
[10:51:50] <Vq> nicko: I don't know anything about the mesa products but i found this piece of documentation in the emc2 tree
[10:51:53] <Vq> http://git.linuxcnc.org/gitweb?p=emc2.git;a=blob;f=src/hal/drivers/mesa-hostmot2/README;h=d5cc16bff4e43de21c8c34dde90c7ef258bc947f;hb=HEAD
[10:52:07] <archivist> are you thinking in time with an external sync
[10:53:22] <TekniQue> now you're thinking in portals
[10:54:00] <nicko> external sync is something I need to look into yes
[10:54:07] <nicko> so much dang stuff to think about
[10:54:18] <nicko> and I'm learning
[10:54:56] <nicko> I have an old PC at the moment, running out of its parallel port directly
[10:55:48] <nicko> As I understand it most newer PC's have issues with their parallel ports and latency ?
[10:56:01] <archivist> 240 a sec is probably more than needed anyway, as the trajectory planner can help/hinder the smoothing of motion
[10:56:37] <nicko> once a Mesa card 5I22 / 7I43 is used are those latency issues redundant and any PC can be used ?
[10:56:41] <archivist> latency is not the par port its other stuff in the pc
[10:57:11] <nicko> ok, yup - yes I should understand that
[10:57:49] <archivist> the PC is in the loop, so bad video card interrupts can be problematic, and power management
[10:58:30] <nicko> so even with the Mesa cards I'm using old PC's ?
[10:59:18] <archivist> the proper answer is depends, boot the cd and do a latency test
[10:59:50] <nicko> Is there any database/wiki of newer systems that people have found appropriate ?
[11:00:29] <jthornton> a popular motherboard for use with a 5i20 is the D510MO and cheap
[11:00:34] <jthornton> and small
[11:00:49] <nicko> heh, its hard to load the CD on a box of bits that aren't bought or put together yet
[11:01:08] <archivist> some carry cd to shop :)
[11:02:55] <nicko> I can see myself doing that soon ;)
[11:03:24] <nicko> thing is I'm keen on getting a hackintosh (bunch of stuff tested by others that can run OS X)
[11:03:41] <nicko> its never all in any one room at the same time that I'm around
[11:04:18] <nicko> the Atom (D510MO) has mini-PCI express
[11:04:24] -!- WesBaker [WesBaker!~wes@] has joined #emc
[11:04:35] <nicko> how do the larger boards fit on ? 5I22 ?
[11:05:29] <jthornton> it has a PCI slot too
[11:08:10] <nicko> ok, that helps ! what sort of numbers are people getting from it in terms of pulse rates ? and are there any implications of what other parts are used with it ?
[11:08:44] <jthornton> pulse rates with a 5i20?
[11:11:30] <nicko> yes - and also for comparison, just the pport only
[11:11:40] <archivist> are you thinking servo motors
[11:11:41] <nicko> (thanks all for your help by the way)
[11:11:51] <nicko> servo motors
[11:11:56] <nicko> well, yes
[11:12:19] <jthornton> velocity input drives?
[11:13:19] <jthornton> max step rate with a 10,000ns jitter is 40,000Hz
[11:13:21] <nicko> 3x Granite VSD-XE drives (throw almost whatever you want at them in terms of input)
[11:13:46] <nicko> 1 x Kollmorgen servostar DDR - analog
[11:14:20] <nicko> 2 x AB Ultra3000 - analog or step/dir - position/velo/current options
[11:15:50] <nicko> 40,000 ok
[11:16:04] <nicko> better than what I have at the mo'
[11:16:23] <jthornton> if you have +-10vdc drives with encoder feedback you plug a 7i33ta into the 5i20
[11:18:35] <jthornton> then your limit is how fast you can read the endcoders
[11:18:39] <jthornton> In TTL mode, the input RC filter limits the maximum encoder input frequency to
[11:18:40] <jthornton> approximately 1 MHz. This corresponds to 4 million counts per second with most
[11:18:42] <jthornton> quadrature counters (4X mode).
[11:18:48] <jthornton> In TTL mode, the input RC filter limits the maximum encoder input frequency to
[11:18:49] <jthornton> approximately 1 MHz. This corresponds to 4 million counts per second with most
[11:18:51] <jthornton> quadrature counters (4X mode).
[11:19:05] <jthornton> In TTL mode, the input RC filter limits the maximum encoder input frequency to approximately 1 MHz. This corresponds to 4 million counts per second with most quadrature counters (4X mode).
[11:19:21] <jthornton> ^ 7i33ta manual
[11:19:46] <jthornton> oh crap double posted
[11:20:42] <jthornton> triple posted :/
[11:23:18] <nicko> All the drives can run analog
[11:23:24] <nicko> one has to
[11:23:35] <nicko> so it sounds like the thing to do
[11:23:58] <nicko> the BLDCs are 4000 ppr
[11:24:05] <nicko> the AC servos ...
[11:24:06] <nicko> um
[11:24:15] <nicko> at least that or higher
[11:24:15] <jthornton> closed loop is always better to do
[11:24:23] <nicko> agree
[11:24:30] <nicko> I want to get into it for sure
[11:24:54] <nicko> I need a heap of I/O for some other stuff
[11:25:04] <nicko> so the 5I22 looks interesting
[11:25:22] <nicko> I'll email Mesa and get the juice
[11:25:34] <nicko> I heard they hang here on occasion ...
[11:25:48] <jthornton> yea pcw_home is on here a lot
[11:26:10] <nicko> he's on right now!
[11:26:12] <nicko> hah
[11:26:28] <jthornton> it's a bit early for left coasters
[11:26:39] <nicko> its very early in CA right now I guess
[11:26:53] <nicko> I'm in New Zealand
[11:28:45] -!- servos4ever [servos4ever!~chatzilla@173-87-55-88.dr01.hnvr.mi.frontiernet.net] has joined #emc
[11:32:25] <jthornton> so it is tomorrow for you or yesterday?
[11:32:39] <nicko> its today
[11:32:44] <nicko> ;)
[11:32:59] <nicko> its almost midnight - still monday
[11:33:04] <nicko> we are ahead
[11:33:13] <jthornton> ah
[11:42:38] <jthornton> I see some EMC'ers on the map in NZ
[11:43:06] -!- rooks [rooks!~rooks@102-bem-18.acn.waw.pl] has joined #emc
[11:43:42] <nicko> I am yet to find any myself
[11:44:12] <jthornton> you looking at the user map?
[11:44:15] <nicko> I've come across Mach3 and Fagor thus far
[11:44:27] <nicko> not really in the scene so to speak
[11:44:42] <jthornton> on linuxcnc.org is a user map
[11:44:43] <nicko> not the user map
[11:44:46] <nicko> yes
[11:46:27] <nicko> time for bed though
[11:46:33] <nicko> goodnight and cheers
[11:46:34] <jthornton> ok
[12:02:02] -!- nicko has quit [Ping timeout: 276 seconds]
[12:06:32] -!- Surfcam123 has quit [Quit: Visitor from www.linuxcnc.org]
[12:06:48] -!- skunkworks has quit [Ping timeout: 240 seconds]
[12:37:24] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.] has joined #emc
[12:55:57] -!- acemi [acemi!~acemi@unaffiliated/acemi] has joined #emc
[13:02:21] -!- Valen has quit [Quit: Leaving.]
[13:14:14] -!- FinboySlick [FinboySlick!~shark@] has joined #emc
[13:23:52] -!- nullie [nullie!~nullie@nullie.telenet.ru] has joined #emc
[13:42:34] -!- mhaberler has quit [Quit: mhaberler]
[13:46:49] -!- nullie has quit [Quit: Ex-Chat]
[14:02:43] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[14:21:07] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[14:21:33] -!- servos4ever [servos4ever!~chatzilla@173-87-55-88.dr01.hnvr.mi.frontiernet.net] has joined #emc
[14:28:49] <aggrav8d> i'm losing steps on my machine. power levels are good, wiring is fine. Is my acceleration too high?
[14:30:52] <ssi> quit
[14:30:54] <ssi> er
[14:30:55] -!- ssi has quit [Quit: leaving]
[14:31:29] ssi_ is now known as ssi
[14:31:31] <ssi> :D
[14:38:30] <ssi> aggrav8d: turning down accel would be a reasonable place to start
[14:38:33] <archivist> aggrav8d, yes, or friction too high, or loose couplings
[14:40:02] -!- grommit [grommit!~chatzilla@] has joined #emc
[14:41:50] <grommit> I am getting Unexpected Realtime Delay message on startup of a new-ish EMC install. The system is an Intel D510 motherboard. I have disabled Hyperthreading, and installed the grub2 script that addes isolcpus=1 to the end of the kernel line. This fixed this problem on my other system.
[14:42:24] <grommit> I have a base thread set to 15000. The latency test has quite low numbers (6100, 9300).
[14:42:31] <grommit> What else can cause this?
[14:43:33] <grommit> I have also tried changing the base thread to 17000 and 20000 but still get the message everytime on startup...
[14:48:30] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[14:49:32] <aggrav8d> archivist - couplings are new, and tight. friction is at a minimum. I'll try 10% accel reduction and see how it goes.
[14:49:55] -!- acemi has quit [Quit: WeeChat 0.3.2]
[14:50:01] <archivist> aggrav8d, or more you need headroom
[14:51:05] -!- icarusfactor [icarusfactor!~factor@r74-195-188-223.msk1cmtc01.mskgok.ok.dh.suddenlink.net] has joined #emc
[14:51:12] <aggrav8d> headroom?
[14:51:15] <ssi> voltage headroom?
[14:51:52] <archivist> no force headroom, cutting load, etc
[14:52:16] <ssi> are you losing steps only while cutting, or in rapids too
[14:52:19] <aggrav8d> well atm i'm not cutting anything, i'm just running tort.nc
[14:52:36] <archivist> then reduce till its ok
[14:53:19] -!- toastydeath has quit [Read error: Connection reset by peer]
[14:53:48] -!- toastydeath [toastydeath!~toast@c-69-140-223-139.hsd1.md.comcast.net] has joined #emc
[14:56:42] <aggrav8d> reduce accel, or something else?
[14:56:51] <ssi> start with accell
[14:56:59] <aggrav8d> ok.
[14:57:07] <ssi> may need to reduce velocity as well
[14:57:20] <ssi> the faster you turn a stepper, the less torque it makes
[14:57:40] <ssi> so you need enough torque at your max rapid speed to overcome all the frictions in your system
[14:57:44] <aggrav8d> i'll try one axis first and see if it improves. if so then I'll know i'm on the right path.
[14:57:51] <ssi> and you need enough torque at max cutting speed to overcome all the frictions plus cutting load
[14:58:04] <ssi> then you need additional torque to accelerate
[14:58:34] -!- maximilian_h1 has quit [Quit: Leaving.]
[14:59:04] <aggrav8d> well everything is rated for the speed i'm running at. I'm thinking accel is too high so some step commands are getting "lost"?
[15:08:21] -!- Connor has quit [Read error: No route to host]
[15:08:58] -!- Connor [Connor!~Connor@] has joined #emc
[15:13:18] -!- e-jones has quit [Quit: Ex-Chat]
[15:25:14] -!- Athlocatle [Athlocatle!~vip0@] has joined #emc
[15:26:02] <Jymmm> grommit: Did you run the latency test for 24 hours by chance?
[15:28:49] <Jymmm> aggrav8d: Just curious, what driver are you using?
[15:31:15] <grommit> jymmm: yes, it ran it for a very long time.
[15:32:17] <Jymmm> grommit: Does "long time" mean 24 hours or more?
[15:32:23] <grommit> yes
[15:32:25] <Jymmm> k
[15:32:26] <grommit> ore
[15:32:29] <grommit> more
[15:33:18] <Jymmm> which atom board ?
[15:33:27] <grommit> The realtime error happens the moment I start up axis too, so it seems strange.
[15:33:36] <grommit> D510MO (I think that is the one)...
[15:34:19] <Jymmm> This one? http://www.newegg.com/Product/Product.aspx?Item=N82E16813121399
[15:35:13] <grommit> to be specific, this one: http://www.mini-box.com/D510MO-mini-ITX-Intel
[15:36:22] <grommit> mini-box is $0.04 cheaper. that's why I bought it there ;-)
[15:36:45] <ssi> sound reasoning
[15:37:06] <Jymmm> both have the same chipset
[15:38:44] -!- capricorn_one [capricorn_one!~capricorn@zima.linwin.com] has joined #emc
[15:39:35] <Jymmm> Both being compared to this one which I believe jthornton has used succesfully http://www.newegg.com/Product/Product.aspx?Item=N82E16813121442
[15:40:33] <grommit> We have the D510MO in another device that does not have this issue...
[15:40:47] <Jymmm> what BIOS versions?
[15:40:53] <grommit> Also, my Atom 310 board does not either
[15:41:12] <grommit> Hang on, will have to boot into bios to see...
[15:41:31] <Jymmm> and is it the latest bios version?
[15:41:32] -!- ve7it [ve7it!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[15:41:44] <Jymmm> he's here.... shhhhhhhhhhhhhh
[15:42:49] <grommit> I think it is the latest. It is: MOPNV10N.86A.0509.2011.0127.1000
[15:47:51] <grommit> Looks like it is 2 revisions back from the latest. Mine being from Jan 2011, the newest from March.
[15:48:52] <grommit> Nothing in the fixes seems relevant but this is: Increased Setup time and Data range.
[15:49:02] <grommit> whatever that means...
[15:49:15] <Connor> This is the Intel board I used for my EMC machine. http://www.newegg.com/Product/Product.aspx?Item=N82E16813121399&nm_mc=TEMC-RMA-Approvel&cm_mmc=TEMC-RMA-Approvel-_-Content-_-text-_-
[15:49:25] <Connor> it's not available anymore.. but.. works really well.
[15:50:30] <grommit> I can update the BIOS but I am not confident this will change anything
[15:51:17] <Jymmm> grommit: Be sure to clear the CMOS after you've updated the BIOS (record any settings you need to retain before hand)
[15:51:42] <grommit> how do I do that? and what settings do I care about
[15:52:04] -!- awallin_ has quit [Remote host closed the connection]
[15:52:49] <Jymmm> grommit: Well, did you make any changes to the BIOS?
[15:53:06] <grommit> yeah, but I can make them again, only Hyperthreading I think
[15:53:22] <Jymmm> Then that's all you need to write down =)
[15:53:28] <grommit> done
[15:54:13] <grommit> So, aside from this (which will be a miracle if it works) any other reason to get Unexpected Realtime Delay on startup of Axis?
[15:54:51] <Jymmm> grommit: Clearing the CMOS, is moving that jumper over for 60+ seconds, powering off, move the jumper back, then powering on.
[15:55:21] <grommit> Do I need to do that? (if I don't care about settings)?
[15:55:29] <Connor> Umm.. I had that happen when I had some settings changed in EMC
[15:56:05] <Jymmm> Well, when updating BIOS, sometimes residual things will remain if you don't. It's the safe bet in doing so.
[15:56:41] <Connor> BASE_PERIOD was the problem.. I had it too low.
[15:57:08] <Jymmm> He's increased his base perios to 17000
[15:57:24] <grommit> Yes, I have played with the base period setting it from 15000 up to 20000. The latency is 9300.
[15:57:50] <Jymmm> grommit: did you customize emc beyond the norms?
[15:58:30] <grommit> Hmm, not sure what the "norms" are, but I am experimenting with a new gantry config, so this is a new config, not something that worked and then broke/
[15:58:44] <Connor> my base is 30684
[15:58:45] <skunkworks> 20000 might still be too low when everything is running...
[15:58:58] <Connor> sorry.. 30864
[15:59:08] <Jymmm> grommit: So, if you use one of the included configs, does it work?
[15:59:24] <skunkworks> grommit: are you not using mesa hardware?
[15:59:37] <grommit> no, not on this box. parallel only
[15:59:48] <Connor> and my servo is 1000000
[15:59:52] <grommit> (unfortunately!)
[16:00:00] <skunkworks> grommit: heh
[16:00:26] <grommit> I shouldn't have let myself be talked into doing it the "simple" way :-)
[16:00:38] <Connor> What's the simple way?
[16:00:50] <grommit> EXACTLY!
[16:01:24] <Connor> What values do you have for the base and servo now?
[16:01:39] <skunkworks> grommit: could you up the base period some more - justt to see?
[16:01:48] <grommit> Yes, I will do that
[16:01:51] <Jymmm> grommit: like 30000
[16:01:53] <Connor> and, hyperthreading ? Turn it off.
[16:02:03] <grommit> Way easier than updating the bios.
[16:02:05] <Jymmm> Connor: he has it off
[16:02:11] <Connor> Jymmm: OKay, missed that part.
[16:02:11] <Jymmm> grommit: wuss
[16:02:38] <grommit> :-)
[16:02:41] <Connor> Yea.. my base is fine at 30864
[16:02:55] <grommit> i will update the bios if it doesn't work. I promise :-)
[16:03:05] <Connor> Much lower and I was getting realtime errors.
[16:03:14] <Jymmm> grommit: Yeah, but only becasue you've run out of ideas =)
[16:03:34] <Connor> my original config had it at 55555
[16:03:38] <Connor> hehe 5 5's. :)
[16:03:39] <grommit> But given my latency test numbers, shouldn't 15000 or at least 20000 be more than enough
[16:04:00] <Jymmm> grommit: see what 30000 does for you first
[16:04:11] <Connor> You have to take into account CPU maxxing out.. Did you run all kinds of stuff while checking the latency ?
[16:04:19] <grommit> A pragmatist I see...
[16:04:30] <Connor> Axis has a high impact on CPU when it first starts up..
[16:05:09] <grommit> Yes, I had a billion Mozilla windows open, terminal windows open, glxgears running. I also surfed the web for part of the day it was running. It ran for a couple days at least.
[16:05:31] <Connor> something open doesn't mean it's using CPU.
[16:05:33] <Connor> or much.
[16:05:52] <Connor> my fav test is a python script in a while loop with no sleep statement.. chews up CPU.
[16:06:03] <Jymmm> Have it calc Pi to a million digits =)
[16:06:47] <Jymmm> (I have that as a file now btw)
[16:07:01] <Jymmm> take forever to load
[16:08:43] <Jymmm> I want one of these, this thing is just sexy.... http://www.newegg.com/Product/Product.aspx?Item=N82E16813131697
[16:09:55] <aggrav8d> Jymmm - i ran the latency test for maybe 10 minutes.
[16:10:34] <Jymmm> aggrav8d: I was asking what driver you are using
[16:10:43] <Jymmm> motor driver
[16:11:45] <skunkworks> Jymmm: I got a few of these - really like them so far. http://www.newegg.com/Product/Product.aspx?Item=N82E16813182238&cm_re=mbd-x7spe-hf-_-13-182-238-_-Product
[16:11:50] <grommit> So the magic number seems to be 21000.
[16:12:22] <grommit> Anything over 21000, 21500 for example, and it gives me the error.
[16:12:30] <grommit> Odd
[16:12:36] <Connor> OVER ?
[16:13:09] <grommit> Good point.
[16:13:14] <grommit> Hang on....
[16:14:18] <Jymmm> skunkworks: I'm not the biggest fan of supermicro.
[16:14:21] <grommit> 22000
[16:14:48] <Connor> it could be that when you tried 21000 it worked.. you need to fire it up again and again to be sure..
[16:15:01] <Connor> again.. My 1.8 CPU works fine @ 30864.
[16:15:02] <grommit> it doesn't always give me the error at 21000, or 21500 - but sometimes. So this must be the hairy edge.
[16:15:25] <grommit> 22000 doesn't seem to be a problem
[16:15:29] <Connor> Yea. I didn't want to be close to it because of other things the computer could be doing.
[16:15:55] <grommit> Well, I want it as low as possible since I am generating step/dir on the cpu directly.
[16:16:13] <Connor> as am I.
[16:16:28] <Connor> two axis @ 1/8 stepping, 1 at 1/16
[16:16:46] <skunkworks> Jymmm: I have had really good luck with supermicro so far.
[16:16:51] <Connor> I'm able to do around 300 IPM on the 1/8's (with a 1 turn to 10mm lead screw)
[16:17:19] <Jymmm> skunkworks: Feel lucky, it's a mixed bag with them much of the time.
[16:17:56] <Connor> grommit: Start with 30864 and get the machine up and working.. then walk the setting down after..
[16:17:59] <grommit> This is a four axis (2 motors on Y) gantry with fairly large servos. We want to do 1000 or better ipm rapids.
[16:18:26] <Jymmm> grommit: get a REAL cpu =)
[16:18:34] <Connor> What pitch?
[16:18:39] <Connor> for lead screw?
[16:18:45] <grommit> Rack and pinion
[16:19:00] <grommit> please define REAL cpu ;-)
[16:19:00] <Connor> okay. you doing microstepping ?
[16:19:05] <Connor> Non Atom.
[16:19:10] <Jymmm> grommit: You need 16 cores to get 1000ipm ;)
[16:19:12] <Connor> i5 or i7 core
[16:19:16] <Jymmm> grommit: and 128GB ram
[16:19:31] <grommit> Then why did I turn off 3 of my 4? :-)
[16:19:45] <Jymmm> grommit: CUZ YER DUMB?
[16:19:50] <Jymmm> <snickers>
[16:19:53] <grommit> Perhaps!
[16:19:58] <Connor> You didn't.. Hyper threading isn't a real core.
[16:20:05] <Connor> Dual Core is fine.
[16:20:24] <Jymmm> grommit: and you need tokenring too!
[16:20:26] <Connor> hyper threading borrows from idle CPU to make "virtual" core.
[16:20:40] <grommit> Sorry, yes, 1 of my 2 - I should have said.
[16:20:44] <Connor> it can cause issues for Real Time.
[16:20:50] <Jymmm> grommit: and you need a 300BAUD modem
[16:21:13] <Jymmm> grommit: and a MONOCHROME crt
[16:21:19] <grommit> I don't have even dual core anymore because I just pegged emc to one.
[16:21:52] <Connor> Turn the 2nd core back on.. leave hyper threading off, go with 30000 base and give it a try.. You may need a different driver board, or use 1/2 stepping or something.
[16:22:14] <Jymmm> You would think that you could split the GUI/AXIS on one core, and the machinery on another
[16:22:22] <grommit> Or, I could have just done what I knew would work and mitigate all this - Mesa cards!!
[16:22:49] <grommit> I don't care what my latency is on my mill with my 7i43 card in it!
[16:22:55] <grommit> Live and learns :-)
[16:22:58] <Connor> That too.
[16:23:51] <Connor> but, I'm just giving you info based on what you have.. not what you think you need in hindsight
[16:23:56] <Connor> :)
[16:24:03] <Jymmm> skunkworks: Love the " RAID 0, 1, 5, 10 support (Windows Only)" NOT!!! ;)
[16:24:44] <skunkworks> right - I have it running a raid box through a esata card
[16:24:49] <skunkworks> linux
[16:25:06] <Jymmm> skunkworks: what raid box?
[16:27:27] <skunkworks> I had bought a few of these when they where on super sale http://www.newegg.com/Product/Product.aspx?Item=N82E16816132016&cm_re=8_bay_sata-_-16-132-016-_-Product
[16:27:49] <Connor> Nice looking box.
[16:27:59] <Connor> linux doing the RAID though right ?
[16:28:03] <Connor> no raid card ?
[16:29:20] <skunkworks> the raid card that comes with it is a little anemic - bought a but more expensive one (pci-e 4x)
[16:29:34] <skunkworks> still not hardware raid though - seems to be pretty fast.
[16:30:20] <Connor> I built a 4 Drive NAS box (1U Rack mount) for my backup server.. it's 1.5TB drivers, and runs soft raid..
[16:30:23] <Jymmm> skunkworks: how long have you had it running, and does it have 8 hdd's?
[16:30:23] <Connor> it's nice.
[16:30:47] <skunkworks> Jymmm: one has been running for a few years. full of 1tb drives
[16:31:06] <Jymmm> any issues?
[16:31:25] <Jymmm> skunkworks: RAID5 ?
[16:32:09] <ssi> heh my machines run on 10 year old 1GHz Via C3 ITX boards
[16:32:16] <ssi> "Mesa makes it happen!"
[16:32:17] <ssi> :D
[16:33:52] <skunkworks> Jymmm: yes
[16:35:36] <Jymmm> skunkworks: I'm contemplating using ZFS instead
[16:39:50] -!- JT-Shop [JT-Shop!~chatzilla@216-41-156-49.semo.net] has joined #emc
[16:41:45] <skunkworks> I use mdadm for raid. Remember to get the latest stable release of it.. On ubuntu the version in synaptic is quite a few versions behind
[16:41:49] <bzzzz> for mother's day, i cnc-ed a greeting card into 0.5" thick aluminum. without breaking the cnc. epic.
[16:46:47] -!- awallin [awallin!~quassel@cs27057046.pp.htv.fi] has joined #emc
[16:48:46] -!- izua_ has quit [Quit: leaving]
[16:49:01] -!- izua has quit [Quit: leaving]
[16:49:47] -!- izua [izua!~izua@] has joined #emc
[16:49:48] -!- izua has quit [Changing host]
[16:49:48] -!- izua [izua!~izua@unaffiliated/izua] has joined #emc
[16:50:11] -!- ve7it has quit [Ping timeout: 240 seconds]
[16:53:30] -!- newbynobi [newbynobi!~norbert@p4FC9A31A.dip.t-dialin.net] has joined #emc
[17:10:02] -!- ve7it [ve7it!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[17:10:51] -!- motioncontrol [motioncontrol!~i@host216-76-dynamic.21-87-r.retail.telecomitalia.it] has joined #emc
[17:11:29] -!- newbynobi has quit [Quit: Verlassend]
[17:25:08] <Connor> My MPG came.. and to my delight I now know the exact part number and china company who makes it.
[17:25:29] <archivist> sexy ebay toy item 160585795097
[17:26:20] <bzzzz> ugh, it's all from china now
[17:26:48] <Connor> Oh yea.. where else? :)
[17:27:15] <bzzzz> someplace closer where the shipping time isn't 5-10 days
[17:27:35] <Connor> When ordering 1 or 2 yea..
[17:29:07] -!- psha [psha!~psha@] has joined #emc
[17:29:21] <bzzzz> yeah, but only the big guys order 1k+
[17:31:23] -!- marcin_ose [marcin_ose!~marcin_os@207-119-215-41.dyn.centurytel.net] has joined #emc
[17:43:30] -!- motioncontrol has quit [Quit: Sto andando via]
[17:51:23] -!- psha has quit [Quit: Lost terminal]
[17:53:36] -!- acemi [acemi!~acemi@unaffiliated/acemi] has joined #emc
[18:03:12] -!- motioncontrol [motioncontrol!~motioncon@host216-76-dynamic.21-87-r.retail.telecomitalia.it] has joined #emc
[18:13:43] -!- mhaberler has quit [Quit: mhaberler]
[18:14:27] <JT-Shop> neat http://www.linuxcnc.org/component/option,com_comprofiler/task,userProfile/Itemid,17/user,3068/lang,english/
[18:15:46] -!- jtektool [jtektool!~jtektool@rrcs-70-63-38-234.central.biz.rr.com] has joined #emc
[18:16:49] <jtektool> hey guys need help wiring a senc 125 glass scale using for an e stop if following erro gets to high
[18:16:57] <jtektool> already established i have enough hz
[18:17:02] <jtektool> 44000
[18:17:12] <jtektool> 5 micron glass scales
[18:17:47] <jtektool> so that makes 5079.9999922784 lines per inch quarature x 4
[18:17:51] <JT-Shop> just set the following error you want for the axis in your ini file
[18:18:37] <jtektool> yeah dont know about wiring though im using parallel port preakout
[18:18:55] <jtektool> I have senc 125 glass scales
[18:19:11] <jtektool> it sez pin 1 - not used
[18:19:21] <jtektool> pin 2 - Channel A +
[18:19:35] <jtektool> pin 3 - not used
[18:19:48] <jtektool> pin 4 - B +
[18:19:59] <jtektool> pin 5 - not used
[18:20:08] <jtektool> pin 6 - ground
[18:20:39] <jtektool> pin 7 - vcc + 5.1 v + or - .1
[18:21:00] <jtektool> pin 8 - channel R +
[18:21:37] <jtektool> in my .hal i got pin 12 and 13 set as inputs
[18:21:55] <jtektool> and it works cuz i get joint following error without the scales hooked
[18:22:17] <jtektool> as soon as i move
[18:22:56] <JT-Shop> servo driven axis?
[18:23:06] -!- acemi has quit [Quit: WeeChat 0.3.2]
[18:23:49] <jtektool> it is servo but i have a wierd setup as far as my config is concerned its actually a stepper
[18:24:22] <jtektool> i have encoders going back to servo drives and want to use emc as DRO
[18:24:46] <jtektool> glass scales as DRO
[18:26:12] -!- marcin_ose has quit [Ping timeout: 240 seconds]
[18:26:34] <jtektool> what is channel R +????
[18:26:41] <jtektool> im so confused
[18:26:44] <JT-Shop> index I assume
[18:26:55] <jtektool> oh ok
[18:26:57] <cradek> yes some scales have periodic index
[18:27:03] <cradek> use a scope to find out for sure
[18:27:09] <JT-Shop> how can you have quadrature with only two inputs?
[18:27:10] <jtektool> gotcha
[18:27:21] <cradek> JT-Shop: ?
[18:27:37] <JT-Shop> I thought you needed A /A B and /B
[18:27:41] <ssi> nah
[18:27:52] <ssi> A and B only... /A and /B are for differential signals
[18:28:05] <jtektool> it sez channels a and b in 90 degree quadrature relationship
[18:28:05] <JT-Shop> ah ok
[18:28:05] <cradek> A + /A is a system that is more immune to noise, called differential
[18:28:29] <bzzzz> is that related to lvds?
[18:28:38] <ssi> yeah, low voltage differential
[18:29:00] * JT-Shop goes back to work now
[18:29:33] <ssi> idea is that any noise picked up on the cable will be the same on both lines, and at the other end the /A gets inverted and summed with the A, and so the noise is inverted as well and summed, canceling out
[18:29:52] <jtektool> it sez square-wave pulse TTL-level
[18:30:24] <bzzzz> cool
[18:30:25] <jtektool> all cables are shielded and grounded at the cnc box plate
[18:30:28] <JT-Shop> ssi: thanks
[18:30:32] <cradek> JT-Shop: if you're brave, TTL compatible signals can probably be hooked directly to the parallel port
[18:30:38] <cradek> er, too many JTs
[18:30:43] <JT-Shop> lol
[18:31:12] <jtektool> sorry yeah i was talking to you cradek friday but i was on antibiotics my brain is much more with it today
[18:32:01] <jtektool> no opto isolation???
[18:32:18] <cradek> isolation is always better
[18:32:50] <cradek> but if you run the scale on the pc supply anyway, isolating the encoder signals is not going to gain you much
[18:33:08] <jtektool> so i would isolate at both ends?
[18:34:02] <cradek> isolate what and how?
[18:34:05] <jtektool> so maybe i input to my parallel card instead of main board
[18:34:49] <jtektool> i mean isolate CPU PSU >> ENCODER >> isolate >> parallel
[18:35:17] <jtektool> i have 2 parallel ports a netmos and main board
[18:35:55] <jtektool> so your saying its safe to power from CPU PSU >> encoder and back to parallel
[18:36:07] <jtektool> or not so much
[18:36:29] <jtektool> << computer programmer not EE
[18:36:35] <cradek> I'm hesitant to give you advice one way or the other
[18:37:28] <jtektool> gotcha
[18:38:00] -!- psha [psha!~psha@] has joined #emc
[18:38:47] <jtektool> like on my break out board therez in puts i get to put A + to pin 12 and B + to pin 13
[18:39:14] <jtektool> but therez also a pin 12 and 13 + what wire goes there from my encoder
[18:40:32] <jtektool> im going to pull out my multimeter hold on
[18:41:04] -!- jrobson [jrobson!EMCIRCUser@iburst-41-213-44-254.iburst.co.za] has joined #emc
[18:45:20] <jtektool> ok pins on my break out board beside input pin is + 5v
[18:45:32] <jtektool> my output pins is - 5V
[18:45:41] <jtektool> COM
[18:49:56] <jtektool> OFF TOPIC:does anybody know how long before theres a servo wizard i heard that was comming soon
[18:50:10] <jtektool> *DOES
[18:52:24] <JT-Shop> you can use it now
[18:53:44] <JT-Shop> jtektool: http://www.linuxcnc.org/docview/html/config_pncconf.html
[18:54:35] -!- awallin has quit [Read error: Operation timed out]
[18:56:02] <jtektool> oh thanks
[18:57:53] <jrobson> Hi
[18:58:16] <jrobson> Is it possible to edit directly inside emc2 rather than with an external program?
[18:58:30] <JT-Shop> edit what?
[18:58:57] <jrobson> g-code
[18:59:47] <JT-Shop> you can open an editor from Axis, it will be the one you specify in your ini file
[19:00:27] -!- awallin [awallin!~quassel@cs78199127.pp.htv.fi] has joined #emc
[19:00:36] <JT-Shop> jrobson: http://www.linuxcnc.org/docview/html/config_ini_config.html#sub:[DISPLAY]-section
[19:01:16] <jrobson> sure, but in the case of giving the control over to someone that has never used a pc before, only industrial controls it is better if the editing is direct rather than an external editor, I'm not sure if you are familiar with fanuc controls for instance
[19:01:20] -!- mhaberler [mhaberler!~mhaberler@imac.stiwoll.mah.priv.at] has joined #emc
[19:01:23] <jrobson> thanks, I'll have a look.
[19:01:49] <JT-Shop> once you edit you have to reload
[19:02:21] <JT-Shop> I'm familiar with DX-32 controls :)
[19:02:30] <JT-Shop> and Anilam
[19:03:06] <jrobson> Okay well in a Fanuc, you can edit and run straight away if the program you were editing was the one selected to run.
[19:03:40] <jrobson> Even in the older ones you can stop the program with single block, edit the program and continue running
[19:06:08] <JT-Shop> can't do that in EMC at this moment
[19:07:18] <cradek> that can only make sense for controls that handle one line at a time (like one that reads from paper tape)
[19:07:43] <jrobson> no it works for controls that read from memory too
[19:08:02] <jrobson> well I've never used tape
[19:08:21] <cradek> sorry I was not clear. you can remove the tape reader and still have a control that works the same way (read one line, execute it, read the next line)
[19:08:36] <cradek> on that machine too, you could edit later lines in the program while the execution is paused
[19:08:51] <jrobson> I don't think I am either, bare with me, it might take a few lines to explain.
[19:08:53] <cradek> this is not at all like EMC, which blends moves together and has subroutines and loops and what all else
[19:09:22] <cradek> EMC reads far ahead in the program to give you smooth motion that is not possible if you read one line at a time, like very old controls had to
[19:10:51] <jrobson> This is the same, they also read far ahead, however when you switch to single block and go into the program even though let's say line N010 was executed the cursor now sits at line N015, if you move it back to N010 it will execute from there again, if you go straight back to the run mode without editing anything it will carry on.
[19:11:10] <jrobson> this doesn't work for feedhold because of the same reason you just posted
[19:11:14] <jrobson> I mean wrote
[19:11:58] <jrobson> it's specefically in single block mode, because when in single block you are only executing one block at a time, no need for read-ahead
[19:12:41] <cradek> I see what you are saying - EMC doesn't do what you say
[19:13:15] <jrobson> Okay thanks, not a big deal
[19:13:17] <cradek> I think it is very hard or impossible to do what you say 100% correctly
[19:13:48] <jrobson> oh
[19:13:51] <jrobson> okay
[19:14:06] <jrobson> not that big of a deal
[19:16:34] <jrobson> any way to get the gui to integrate an editor though?
[19:16:58] <jrobson> just for the sake of creating new programs or editing others
[19:17:08] <cradek> with open source any modification is possible
[19:17:38] <cradek> but consider that you have a windowing system and running more than one application, each suited for a different task, is one of its strengths
[19:21:00] <JT-Shop> by integrate do you mean an editor pops up when you select File/Edit?
[19:21:52] -!- motioncontrol has quit [Quit: Sto andando via]
[19:23:31] -!- marcin_ose [marcin_ose!~marcin_os@72-161-236-253.dyn.centurytel.net] has joined #emc
[19:23:45] <jrobson> I'll try and explain from the beginning. Walk into the shop and turn on the control, it goes to the message screen. I then select memmory and program it shows a program folder with sub folders and files, I then navigate via the cursor arrows to the program I want to run, I select D-SET.
[19:24:00] <jrobson> I now press the edit button and I can edit the program.
[19:24:25] <jrobson> when done I can press save, click the run button and press start
[19:24:27] <jrobson> program runs
[19:25:25] <jrobson> on emc2/mach3 you have to click the edit button and the operating system dialog box opens up, then you click on the file and it says file loaded you can now run it, if you want to edit you have to start the editor, edit the file, save it, then reload it, then run it.
[19:25:49] <jrobson> sorry I mean the open file instead of "edit button"
[19:27:10] <jrobson> It is nicer when you can edit it directly within emc2, when saving or going back to run mode it automatically reloads it, so there are no operating system dialog boxes and you can then have the files specifically on only one directory without the user accidently opening the wrong directory and getting lost.
[19:28:11] <cradek> there is clearly not consensus about which system is better
[19:28:41] <cradek> on my mill, there is no keyboard at all. I edit programs at a nearby desk, not standing uncomfortably in front of the mill
[19:28:44] <marcin_ose> I got my Gecko 540 working with the torch table. I can move the individual axes, but when I try to slave 2 axes into one (I have 2 steppers driving the x axis from opposite sides), only one of the motors moves. I selected parallel port pin 2,3 for x axis, and 4,5 for x axis. Any suggestions?
[19:29:18] <jrobson> I agree, for a lot of people they are happy with the operating system being "visible" but I'm curious if it's possible to do it the way we are used to.
[19:30:10] <cradek> like I said before, any modification is possible. but what you are asking for isn't currently in AXIS and we have no plans to put it in. if you want to modify it in this way, you can do that.
[19:30:26] <jrobson> when you have a few production lathe's you might prefer to program right by the machine, the programs are short, with the mill I exclusively use my notebook, but that's because I don't write any code manually (cam) with the lathe's it's different, we only code manually.
[19:30:43] <jrobson> okay it's good to know that it's possible to make it work
[19:30:49] <jrobson> would it require a whole new gui?
[19:30:57] <JT-Shop> on my lathe I use ngcgui it's much faster than writing a program
[19:31:20] <JT-Shop> and it is built into the gui
[19:31:21] <cradek> if all the existing guis are not what you want, then yes you could write a new one
[19:31:41] <cradek> if one of the existing guis is close to what you want, you could modify it
[19:32:29] <jrobson> Hi JT I'll look into NGC gui, the biggest problem I have is that I need to also have someone that has never used a pc program it.
[19:33:21] <cradek> maybe an intro-to-pc session would be better than writing a new gui specifically for this uncommon person
[19:33:50] <cradek> 15 years ago at my work we used to get people in for training (on our software product) who had never used a computer before. we don't get them anymore.
[19:34:09] <JT-Shop> na, you can program it and the button pusher just has to fill in the blanks
[19:35:14] <cradek> a new gui that you have just written will be MUCH more buggy than one that is years old and currently used by thousands of people. you may be trading one perceived problem for many more real problems for your new-to-computers guy.
[19:36:17] <jrobson> Perhaps
[19:36:53] <cradek> I will stop trying to tell you what to do now.
[19:38:23] <JT-Shop> marcin_ose: what did you do to connect the two steppers to the stepgen?
[19:39:40] <jrobson> no problem, I'm just trying to find out which is easier to do what I want it to, mach3 has an easy interface customising system but other things it does poorly and it's not very future proof (axis limited)... but contrary to what you guys think is a non-issue an integrated system is not a bad thing to have.
[19:41:31] <jrobson> but you are right to write a whole new gui one must first have a very hard look at what is available and make sure you can't use it instead
[19:42:49] <jtektool> Sorry to chime in with my *nix arrogance but come on if someone either cant use an editor or run a program like cimco editor or mastercam on windows and send the files via ethernet then dont bother using linux EMC2
[19:43:29] <marcin_ose> JT-Shop - the steppers are connected to the x and y channels of the Gecko. Is that what I'm supposed to do?
[19:43:50] <cradek> I see two kinds of new users - some have a thing in mind, often being experienced using another kind of control, and want to make emc like that. Others see being on a pc as a new freedom and they embrace it fully - like jtektool says, maybe set up a wireless network with the office where the cam guys work, etc. It's a different world and it's a very powerful one. running more than one app simultaneously is just a tiny part of that world.
[19:44:18] <marcin_ose> JT-Shop - mean x-axis steppers (2 of them) are connected to the x and y Gecko driver channels.
[19:44:55] <cradek> in my opinion, it is 100% impossible that by showing a guy how a modern computer's windowing system works, you are wasting his time.
[19:45:32] <JT-Shop> marcin_ose: I mean in your config?
[19:45:57] <Vq> cradek: modern = tiling?
[19:46:16] <jtektool> yeah we got one of those guys here he sez prototrak prototrak prototrak
[19:46:22] <jtektool> its 2011 not 1989
[19:47:16] <jtektool> use a widescreeen and open the text editor next to axis that works awesome here
[19:48:02] <cradek> I edit at my desk, and run a sim version of EMC2/AXIS to test my programs
[19:48:21] <cradek> then I copy them over the network to the machine, running EMC/Touchy, which needs no keyboard/mouse
[19:48:30] <jtektool> awesome
[19:48:51] <cradek> it's not like two copies of EMC cost any more than one copy
[19:48:57] <jtektool> right
[19:49:17] <jtektool> cradek your the man
[19:49:21] <cradek> haha
[19:49:33] <jrobson> What kind of setups?
[19:49:42] -!- psha has quit [Read error: Operation timed out]
[19:49:43] -!- jtektool [jtektool!~jtektool@rrcs-70-63-38-234.central.biz.rr.com] has parted #emc
[19:50:20] <cradek> simple stuff - 1 to 2 fixtures
[19:50:35] <cradek> the machine has a probe for setting origins etc
[19:51:17] <jrobson> okay, I understand of course where you are coming from
[19:51:45] <jrobson> we run production, lathe's run 20 hours a day, the only thing I want to get away from is the high cost when one of the electronic components take a dive
[19:52:14] <jrobson> not particularly looking to reinvent "our wheel" if you get what I mean :)
[19:52:30] <jrobson> I don't write any programs on the mill manually either
[19:52:33] <jrobson> all via cam
[19:52:35] <cradek> I'm surprised that you have operators editing at the control in a production shop like that
[19:52:46] <jrobson> but with the lathe's it's different.
[19:53:03] <marcin_ose> JT-Shop - In Parallel Port Setup - I have pin 2 = x step, pin 3= x direction, pin 4 = x step, pin 5 = x direction.
[19:53:05] <jrobson> we'll it's barfed so you program today and run the same 30 line program for a wheek
[19:53:34] <cradek> marcin_ose: check that the generated hal is right, if you're not getting what you expect
[19:53:38] <jrobson> I can write a lathe program much faster directly on the machine than if I first had to write it on a external pc
[19:53:42] <cradek> marcin_ose: that might not be something stepconf handles
[19:54:38] <cradek> jrobson: hm, for my lathe I still write/test at the desk. I have to stand too much as it is.
[19:56:06] <cradek> the production shop I'm familiar with is all mill, however. maybe lathes are different (parts sure are usually simpler)
[19:56:08] <jrobson> I suppose it's just being used to a different way of doing things, even with the mill I prefer to sit by it when programming.
[19:56:10] <JT-Shop> marcin_ose: let me look at stepconf for a second
[19:56:48] <jrobson> yes lathe parts are significantly simpler of course, for me a big program has 50 lines, on the mill, 20 000 lines no problem
[19:56:50] <cradek> (however, using two windows seems like a very small modification to make to your work style)
[19:57:17] <cradek> anyway I hope you are successful using EMC at your shop
[19:58:16] <jrobson> I think for the lathe's I'll go with mach3 and a custom interface, I've been quoted around $1000-$2000 which isn't bad when fitting to 4 machines, if I ever get something with more than 3-axis I would do emc2 no problem with the interface then.
[19:58:40] <jrobson> thanks for the advice, much appreciated
[19:58:59] <JT-Shop> do you do any threading on your lathes?
[19:59:01] <cradek> hint: try threading first, before picking mach for a bunch of machines
[19:59:05] <cradek> haha
[19:59:08] <JT-Shop> lol
[19:59:09] <jrobson> Yes, kflop handles the threading
[19:59:10] <archivist> jrobson, mach is a bit flaky threading on a lathe
[19:59:29] <archivist> bit/lot
[19:59:31] <jrobson> I know about the mach limitations, also with feedhold, kflop handles that as well
[19:59:51] <JT-Shop> marcin_ose: seems like stepconf handles the hal connections properly for a gantry
[19:59:53] <cradek> no matter what you pick, retrofit ONE machine, not four, at first - yikes
[19:59:54] <jrobson> threading is geared, you can see video's of rigid tapping with that board
[20:00:01] <cradek> I have no clue what kflop is
[20:00:09] <jrobson> yeah for sure I meant that if I ended up doing all 4 the price is spread
[20:00:13] <jrobson> www.dynomotion.com
[20:00:37] <jrobson> it's a board that handles the mach3 io, closes the loop etc
[20:01:21] <marcin_ose> JT-Shop and Cradek - The slaving of two motors worked for me with the former Xylotex controller. Any ideas? Bad cable or something?
[20:01:23] <cradek> ok, well, we'll see you again in 6 months then I bet :-)
[20:02:04] <JT-Shop> marcin_ose: could be time to swap parts to see what is broke and what works
[20:02:29] <JT-Shop> pretty simple with a G540 as I recall
[20:03:01] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #emc
[20:03:49] <skunkworks> iirc a lot of the emc guts where ported into the k-flop.
[20:03:49] <jrobson> hahaha maybe! btw, just curious but hours wise how much to build a new gui?
[20:04:09] <jrobson> that is correct
[20:04:22] <marcin_ose> JT-Shop - What do you suggest for swap procedure? Swap cables around?
[20:04:23] <cradek> that's like asking how long is a stick
[20:04:32] <skunkworks> so you get none of the flexability of emc2 - but you get to pay for it.. ;)
[20:05:01] <jrobson> what can emc2 do on a lathe that mach can't?
[20:05:10] <cradek> cut good threads
[20:05:11] <JT-Shop> and have to buy special hardware to use it
[20:05:15] <cradek> rigid tap
[20:05:32] <jrobson> k-flop takes care of rigid tapping
[20:05:38] <jrobson> and good threads
[20:05:41] <marcin_ose> JT-Shop - I swapped motors to see if another stepper may work on the x axis that is not spinning. Each works, but only one at a time, not when slaved together.
[20:05:51] <archivist> geared axes IN emc
[20:06:32] <cradek> all the benefits of free/open source software
[20:06:43] <JT-Shop> marcin_ose: each stepper runs on both outputs but only if one is plugged in?
[20:06:53] <jrobson> yeah but a closed gui :p
[20:06:55] <archivist> friendly support in here :)
[20:07:09] <jrobson> for sure archivist! thanks for that btw!
[20:07:16] <cradek> jrobson: what do you mean closed gui?
[20:07:20] <jrobson> I appreciate the effort!
[20:07:46] <jrobson> well you just said to make some modifications you need to create your own gui and it sounds like a lot of work
[20:08:02] <cradek> that's not what I said
[20:08:05] <jrobson> mach has a fast system to create a new gui, and if you want someone complete then you can do it in flash
[20:08:36] <JT-Shop> does the gui do a good job of running your machine and threading?
[20:08:40] <cradek> you are confusing mach's screen designer with having the ability to make any change to the program you want.
[20:08:50] * JT-Shop shuts up now
[20:09:02] <jrobson> yes JT it does.
[20:09:11] <cradek> you said you were considering EMC because you wanted to get away from expensive proprietary controls and (I'm assuming) how you're beholden to the company that makes them
[20:09:32] <cradek> IMO you are about to make the mistake of NOT actually fixing that problem
[20:09:39] <jrobson> no no, I said I wanted to get away from the expensive hardware, if a fanuc axis board goes it's $2000 to fix
[20:10:08] <cradek> and if your kflop fails in 2018 then what happens?
[20:10:29] <archivist> machine flops
[20:10:31] <jrobson> use emc2 :p
[20:11:49] <JT-Shop> jrobson: no, the underlying code runs your machine not the gui
[20:13:30] <JT-Shop> the thing that seems to be consistent with someone that wants to run mack and comes here they end up purchasing mach anyway for some reason...
[20:13:50] <jrobson> JT switch off the screen then see how it goes ;)
[20:14:10] <jrobson> people ask questions to see what the limitations are of both
[20:14:17] <jrobson> don't take it personally
[20:15:13] <jrobson> nobody says it's wrong or bad, just seeing if it's close to what one is used to
[20:15:27] <JT-Shop> lol
[20:16:40] <JT-Shop> seems like the last line says it all you want a Fanuc controller but don't want to pay 2k for a repair board
[20:17:29] <jrobson> yes something that works as well as a fanuc or siemens or heidenhein
[20:17:30] -!- skunkworks has quit [Ping timeout: 252 seconds]
[20:18:58] <jrobson> *heidenhain
[20:20:14] <JT-Shop> marcin_ose: you still around?
[20:21:36] <marcin_ose> JT-Shop - I tried troubleshooting by Setting up an xyz coordinate system where the x motor on the torch table is connected to the x Gecko output, the second x motor on the torch table is connected to y Gecko output, and the y axis on the torch table gantry is connected to the z Gecko output. These all worked in 'test this axis' and in a run with the EMC interface. I could only move the axis slightly, but these were all working as ind
[20:21:37] <marcin_ose> ividual channels.
[20:22:14] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.] has joined #emc
[20:22:55] <marcin_ose> JT-Shop - when I tried x and y Gecko outputs running the two sides of the x axis gantry of the torch table, only one motor moves.
[20:23:32] <JT-Shop> marcin_ose: can you pastbin your config?
[20:23:45] -!- vladimirek has quit [Remote host closed the connection]
[20:23:48] <JT-Shop> the one with both motors slaved to the X axis
[20:24:14] <JT-Shop> the myconfig.hal file
[20:24:16] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[20:24:25] <marcin_ose> jt-shop - You want me to pastebin my torchtable.ini file?
[20:24:55] <JT-Shop> no, the torchtable.hal file
[20:28:21] <marcin_ose> JT-Shop: yes, just a sec.
[20:29:32] -!- billykid [billykid!~billykid@host140-28-dynamic.2-87-r.retail.telecomitalia.it] has joined #emc
[20:30:47] <billykid> hello
[20:38:52] <Tom_itx> i smell a genuine troll
[20:42:37] <billykid> :-)
[20:43:04] <JT-Shop> lol
[20:43:46] <JT-Shop> marcin_ose: I have to run to town for a bit, what version of EMC are you using?
[20:44:36] <billykid> can anyone help me with the probe g38?
[20:56:39] <cradek> billykid: you can always just ask a question, without asking permission first. then anyone who knows the answer can help
[20:56:55] <marcin_ose> JT-Shop - torchtable.hal file - http://pastebin.com/u02D6Uvn
[20:58:58] <billykid> o<auto-tool-change> sub
[20:59:00] <billykid> G90 G53 G0 Z0
[20:59:02] <billykid> G90 G53 G0 X0 Y0
[20:59:04] <billykid> G91 G38.3 Z-100 F500
[20:59:06] <billykid> G91 G38.5 Z10 F15
[20:59:08] <billykid> G43.1 k[#5063-#1000]
[20:59:10] <billykid> G90 G53 G0 Z0
[20:59:12] <billykid> G90 G53 G0 X50 Y50
[20:59:14] <billykid> o<auto-tool-change> endsub
[20:59:15] <Connor> Dude
[20:59:16] <billykid> M02
[20:59:17] <Connor> stop
[20:59:26] <Connor> use pastebin
[20:59:50] <billykid> I'm trying to clear that tools
[21:00:46] <cradek> I don't understand what you mean by clear that tools - can you ask another way
[21:01:36] <billykid> but it gives me error: K word without G2 G3
[21:02:26] <billykid> sorry Cradek
[21:03:08] <cradek> http://www.linuxcnc.org/docs/2.4/html/gcode_main.html#sub:G43,-G49:-Tool
[21:03:38] <billykid> is the translator
[21:03:55] <cradek> I'm guessing you are using a newer version that your code was written for. G43.1 with K is invalid now (but it used to be valid). with G43.1 you should specify the axis you want to offset. with a mill that's usually Z.
[21:05:04] <marcin_ose> JT-Shop - EMC Version 2.3.0
[21:07:05] <billykid> I basically do this: http://www.linuxcnc.org/component/option, com_kunena / Itemid, 20/func, view / id, 5596/catid, 10/limit, 6/limitstart, 30/lang, Inglese /
[21:07:53] <cradek> that url did not paste correctly.
[21:07:57] <cradek> did you understand my answer?
[21:08:19] <billykid> yes
[21:09:44] -!- skunkworks has quit [Ping timeout: 252 seconds]
[21:10:08] <billykid> I find it difficult to express myself properly
[21:10:40] <cradek> you are doing fine
[21:10:54] <billykid> http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/id,5596/catid,10/limit,6/limitstart,30/lang,english/
[21:11:32] -!- FinboySlick has quit [Quit: Leaving.]
[21:12:10] <cradek> I have to go - I hope someone else can help if you have a second question
[21:13:02] <billykid> thanks Cradek
[21:17:00] -!- billykid [billykid!~billykid@host140-28-dynamic.2-87-r.retail.telecomitalia.it] has parted #emc
[21:22:39] -!- grommit has quit [Ping timeout: 252 seconds]
[21:38:41] -!- crazy_imp has quit [Ping timeout: 246 seconds]
[21:39:53] <ssi> I'm having a hard time getting spindle speed control working
[21:40:09] <ssi> I'm using a pwmgen on the 7i43 out to a cnc4pc C41 pwm to analog board
[21:40:36] <ssi> my analog out won't go higher than 2.1v
[21:40:52] -!- crazy_imp [crazy_imp!~mj@a89-183-31-199.net-htp.de] has joined #emc
[21:44:12] <JT-Shop> marcin_ose: The config looks just fine...
[21:44:27] <cpresser> ssi: but the pwm.dudy-cycle scales from 0-100%?
[21:45:01] <ssi> I'm not sure... I haven't put a scope on the pwm signal yet
[21:45:06] <ssi> the only parameter that is being set is output scale
[21:45:17] <ssi> and it's set to 3000, so I'd expect that S3000 would give me 100% duty
[21:47:32] <cpresser> i would check the duty-cycle. just to make sure.
[21:47:42] <ssi> how?
[21:47:44] <ssi> other than scope
[21:48:03] <cpresser> on the other hand, did what voltage does the pwm-to-analog board create wen you use DC-voltage as input?
[21:48:31] <cpresser> try starting in software with hal-scope
[21:48:47] <cpresser> use a regular multimeter to measure the AC-value. it should max at 50% DC
[21:50:16] <ssi> I don't know thath I'll be able to see the pwm with hal-scope
[21:50:21] <ssi> since it's a hardware pwmgen
[21:51:07] <ssi> when there's no spindle command, the C41 board shows up as .4V
[21:51:14] <cpresser> right, mesa board. sorry for that
[21:51:27] <ssi> hm I should try changing the 0-10v/4-20ma switch on the vfd and see if it makes a difference
[21:51:31] <ssi> could be an impedance mismatchi ssue
[21:51:44] -!- jtektool [jtektool!~jtektool@rrcs-70-63-38-234.central.biz.rr.com] has joined #emc
[21:51:49] <jtektool> hello all
[21:52:33] <jtektool> so i hooked up this old glass scale encoder to my tektronics scope and is producing a good square wave
[21:53:05] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[21:53:55] <cpresser> ssi: do you own a 'good' digital multimeter? in any case, i would try to measure the output of the board with DC-input (0 and 5V)
[21:54:18] <cpresser> it should give you 0 and 10Volts output
[21:54:24] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-4d01f5ee.pool.mediaWays.net] has joined #emc
[21:54:34] <jtektool> im afraid to connect it to parallel port though cuz i get the square wave when connecting a+ to one side of probe and other side to neg
[21:55:08] <ssi> i have a fluke 189
[21:55:09] <jtektool> my parallel port bob has "inputs" with +5v signals i just dont get im not an EE
[21:55:16] <ssi> and a scope, which I think I'm about to have to drag downstairs :P
[21:55:47] <jtektool> same with b+ and neg
[21:56:08] <cpresser> ssi: the fluke can measure duty-cycle and pwm-frequency!
[21:56:30] <ssi> lemme see if I can figure it out :)
[21:57:08] <jtektool> how to hook encoder to BOB?
[21:57:42] -!- jrobson has quit [Quit: Visitor from www.linuxcnc.org]
[21:57:47] <jtektool> ohhhh sorry not trying to hijack your guys thread right now
[21:57:53] <cpresser> ssi: set to Voltage, AC; press "Hz % ms"
[21:58:05] <cpresser> jtektool: not a problem, its an open irc channel
[21:58:36] <ssi> yeah, that's what I've been trying
[21:58:43] <ssi> it claims 0hz, "OL" %
[21:59:00] <ssi> and 2.7V DC at the current setting
[21:59:13] <cpresser> try measuring your AC-Line. it should be 50Hz (or 60 in US?)
[21:59:15] <ssi> .39V AC
[21:59:42] <cpresser> how fast is the pwm? the fluke can do 1Mhz max
[21:59:57] <ssi> oh you know what, I'm a moron
[22:00:06] <ssi> I was trying to measure pwm on the dc setting
[22:00:09] <ssi> on ac setting, I get something
[22:00:14] <ssi> not what I expect, but somethng
[22:01:20] <ssi> ~300hz, 0.11% duty
[22:01:29] <ssi> something's definitely not right
[22:01:43] <cpresser> its time to fiddle with hal, i guess :)
[22:01:58] <cpresser> tray
[22:01:59] <cpresser> y
[22:02:02] <cpresser> argh. sorry
[22:02:47] <cpresser> try to load a pwm-gen, but dont ook it up to emc2. just set the values using "halcmd setp" and try to debug that way
[22:02:54] <ssi> lemme get the scope so I can see wtf I'm accomplishing
[22:06:25] -!- jtektool has quit [Quit: Visitor from www.linuxcnc.org]
[22:10:21] <ssi> ok so the scale of my pwmgen is 3000
[22:10:41] <ssi> if do an M3 S3000, I get 100% duty cycle
[22:10:48] <ssi> if I do S1000, I get 33% duty cycle
[22:10:53] <ssi> 100khz, which is what I asked for
[22:11:28] <ssi> S300 is 10% duty
[22:11:28] -!- MattyMatt_ [MattyMatt_!~matt@cpc4-birk6-0-0-cust799.1-3.cable.virginmedia.com] has joined #emc
[22:11:33] <ssi> S30 is 1%
[22:11:39] <ssi> looks like that part is working Just Fine
[22:11:41] -!- MattyMatt has quit [Ping timeout: 246 seconds]
[22:12:53] <marcin_ose> JT-Shop - Any suggestions? Can you run my file with a system that you use to see if you get predicted results?
[22:15:46] <JT-Shop> yes, I can look at that, but I'm really confused as why it won't work... can you pastbin the XYZ hal file that moves all steppers?
[22:19:31] <ssi> I'm confused about what this C41 board is expecting as input
[22:19:51] <ssi> the manual talks about it being a PWM input
[22:19:55] <ssi> 2-5V dc, 20-100khz
[22:20:08] <ssi> but then it goes on to explain how to setup mach3 to drive it as though it were a stepper
[22:21:57] <JT-Shop> ssi e-mail him, he is quick about answering your questions
[22:23:08] <ssi> mk
[22:23:23] <ssi> also the vfd is pumping a lot of noise into the system :P
[22:25:17] <JT-Shop> marcin_ose: just to recap you can use an XYZ config and move all the steppers but with your XXYZ only one stepper moves on the X axis???
[22:25:30] <JT-Shop> without changing anything but the config...
[22:26:12] -!- lilalinux has quit [Ping timeout: 240 seconds]
[22:27:01] -!- lilalinux [lilalinux!znc@] has joined #emc
[22:36:44] <Jymmm> how do you calibrate something like that?
[22:37:31] <Jymmm> each power up is gonna be slightly different I'd suspect
[22:38:20] <JT-Shop> Jymmm: some push the gantry against a stop then power up lol
[22:39:10] <JT-Shop> but if you don't have it adjusted for a full step on each stepper you will be out a smidgen but to a plasma cutter it don't matter that much
[22:39:12] <Jymmm> Ok, but what if MotA goes -2 steps and MotB goes +2 steps?
[22:39:25] <JT-Shop> it won't do it for long lol
[22:39:25] <Jymmm> at power up
[22:39:37] <JT-Shop> mine don't move at power up
[22:39:39] <Jymmm> smartass, lol
[22:39:56] <Jymmm> well the motors have to lockup at pwr up
[22:40:00] <JT-Shop> they might move a micro step to a full step position
[22:40:27] <Jymmm> but cant they start tweeking things over time?
[22:40:30] <JT-Shop> again for a plasma it don't matter that you be within 0.0001"
[22:40:32] <Jymmm> s/they/that/
[22:41:06] <Jymmm> bbl, gotta pickup an mri
[22:44:30] -!- jtektool [jtektool!~jtektool@rrcs-70-63-38-234.central.biz.rr.com] has joined #emc
[22:45:29] <jtektool> hey guys getting a nice square wave on encoa and enc0b halscope 90 degrees out of phase perfectly from each other but my dro is not showing movement??
[22:45:49] <jtektool> (dro in axis)
[22:46:07] <jtektool> any answers???
[22:46:41] <jtektool> where do i go next
[22:46:45] <jtektool> anybody???
[22:47:45] <JT-Shop> what are you doing to connect your scale to the dro? can you pastebin the code?
[22:47:46] <ssi> JT-Shop: yep, he responded quickly alright
[22:48:01] <ssi> turns out I misread the datasheet by a few orders of magnitude
[22:48:03] <JT-Shop> I've had him respond on Sundays
[22:48:05] <ssi> it wants 20-100hz
[22:48:07] <ssi> not 20k-100k
[22:48:37] <jtektool> aha probably not connected to dro in hal file right?
[22:48:39] <ssi> seems obscenely low, but I'll give it a shot
[22:48:48] <JT-Shop> could be...
[22:49:17] <jtektool> whats the code to do that give me a sec ill get hal file its on another machine
[22:49:44] <jtektool> sorry for beeing a noob
[22:52:03] <JT-Shop> everyone has to be a noob once
[22:54:01] <jtektool> should i post the whole hal (not sure if that'll work on irc)
[22:54:12] <Tom_itx> pastebin
[22:54:32] <JT-Shop> pastebin ^^ what Tom_itx said
[22:54:44] <JT-Shop> usually pastebin.com is up
[22:54:56] * Tom_itx shoots a gun in a barrel to see if it echos too
[22:55:09] <Tom_itx> if not try .ca
[22:56:59] <ssi> 100hz behaves exactly the same :(
[22:57:39] <jtektool> ok i put it up
[22:57:47] <jtektool> but how do you name a post???
[22:57:48] <Tom_itx> so paste the link here
[22:57:55] <Tom_itx> it gives you a link
[22:58:03] <jtektool> http://pastebin.com/ewC5nbDi
[23:00:43] <JT-Shop> when you watch the encoder counter in the Show HAL Configuration window of Axis can you see it change?
[23:01:00] <jtektool> hold on
[23:03:16] <jtektool> encoder.o.counts changes numbers when i move yes
[23:04:00] <jtektool> if thats the right one
[23:04:09] <JT-Shop> what were you expecting to see on the DRO?
[23:04:52] <JT-Shop> in another way what did you connect the encoder to ?
[23:05:45] <jtektool> well DRO (in axis) sits at 0.000
[23:05:56] <jtektool> no matter when i move
[23:06:02] <JT-Shop> yes, it would if nothing is connected to it
[23:06:48] <JT-Shop> net xpos-fb <= encoder.0.position is not connected to anything...
[23:07:09] <JT-Shop> read this page http://www.linuxcnc.org/docview/html/hal_basic_hal.html
[23:08:11] <JT-Shop> hint: xpos-fb is only a signal name
[23:08:16] -!- Valen [Valen!~Valen@ppp121-44-231-147.lns20.syd7.internode.on.net] has joined #emc
[23:09:43] <JT-Shop> jtektool: that was a good guess by the way
[23:13:16] <jtektool> net xpos-fb <= encoder.0.position => axis.0.motor-pos-fb ????
[23:13:40] <jtektool> how to output to DRO???
[23:14:17] <jtektool> i mean whats the correct code?
[23:16:25] <jtektool> oh that worked im a big moron
[23:16:35] <jtektool> JT-Shop thanks for the help
[23:17:15] <jtektool> if you need a shell script to convert pdfs to test in *nix i would love to share it with the world as a thanx
[23:17:18] <JT-Shop> you got it working?
[23:17:25] <jtektool> yeah above code works
[23:17:37] <jtektool> i just have to test with indicators
[23:17:40] <JT-Shop> cool, now your not a noob any more
[23:18:13] <JT-Shop> convert pdfs to what?
[23:18:22] <jtektool> anyways ive been writing shell scripts since 1982 if you need to hack pdfs into real text i got your back
[23:18:32] <JT-Shop> ok, thanks
[23:18:49] <jtektool> thanks!!!
[23:19:22] <JT-Shop> our documents are now in asciidoc and run through the doc chain to get pdf's and html
[23:19:35] <JT-Shop> the pdf images don't look the best :/
[23:19:38] -!- robh__ has quit [Ping timeout: 240 seconds]
[23:20:36] <jtektool> my code splits it into pdfs out into jpeg tiff and text
[23:21:08] <jtektool> im going to post this to that site too but i have to cut some code out cuz it was written as a hack for this lawyer in columbus
[23:21:47] <jtektool> they get dockets from the court house with peoples addresses and he extracts the addresses out into mailers as advertising
[23:21:53] <jtektool> im a horrible guy i know
[23:23:19] -!- mhaberler has quit [Quit: mhaberler]
[23:23:34] <jtektool> but hey whatever being a hacker is fun
[23:24:07] <jtektool> I actually need my ego taken down a few notches...and EMC is doing well at that
[23:25:58] <JT-Shop> lol
[23:41:58] <Tom_itx> just alot to absorb in a short time
[23:46:14] <jtektool> later dudes that made my day
[23:46:20] -!- jtektool has quit [Quit: Visitor from www.linuxcnc.org]
[23:47:02] <Tom_itx> mmm does linuxcnc.org have an irc client?
[23:47:16] <Tom_itx> noticed his part message
[23:55:40] <ssi> aoeu
[23:56:22] <KimK> Tom_itx: Yes, nice parting message from jtektool. On your Q, there's the java IRC client at: http://www.linuxcnc.org/component/option,com_mospjirc/Itemid,8/lang,english/ if that helps. I prefer the non-gnome Xchat, myself. (available in Synaptic, although the gnome version is the one with the "Ubuntu recommended" logo.)
[23:57:37] <Tom_itx> i wasn't really looking for one. some host sites provide a web interface to irc and i'd never bothered looking for it on linuxcnc
[23:59:08] <Tom_itx> yeah, i've got xchat installed on the emc box too