Back
[00:57:42] <jmkasunich> SWPadnos: motion is unaware of offsets, TLO, etc - it works strictly in machine coords
[01:36:45] <SWPadnos> control.c is where I found the screw comp code
[01:37:01] <SWPadnos> I thought that got linked in with motmod
[02:01:35] <jmkasunich> ?
[02:01:43] <jmkasunich> control.c is one of the main parts of motmod
[02:35:55] <SWPadnos> that's what I thought :)
[02:36:02] <SWPadnos> being in the motion subdir and all
[02:38:04] <jmkasunich> I guess I'm not sure how (or if) your statement about screw comp code ties to the prior discussion
[02:39:41] <SWPadnos> ray had mentioned that it was in userspace, I remembered a conversation mostly between you and Alex about getting comp data into HAL
[02:40:03] <SWPadnos> so I looked for the code, and found it in motion, which I assume is RT code
[02:41:20] <jmkasunich> ok, I thought ray's stuff was DRO related, not screw comp
[02:41:30] <SWPadnos> the reason it came up at all was because I don't know if rotary axes (joints?) can use modulo positions to get comp info - ie, can you comp an A table that might do 10 revolutions, but has comp data for 0-360 degrees which should be repeated 10 times
[02:41:46] <SWPadnos> oh, I brought it up in relation to Stuart's getting comp data for the cinci
[02:42:10] <jmkasunich> thats an interesting point about the rotaries
[02:43:28] <SWPadnos> yeah, keep it in mind while working on joints/axes :)
[02:44:12] <jmkasunich> comp is farther to the right on the drawing than any of the joints/axis stuff that I expect to touch
[02:45:26] <SWPadnos> heh
[02:45:47] <SWPadnos> I figured the comp stuff would be in RT code since it's got to work on joints to be useful, which prompted me to look
[02:46:37] <SWPadnos> I wonder what happens if the screw error is > PID deadband
[02:46:51] <jmkasunich> nothing in particular
[02:47:00] <jmkasunich> screw error could be a foot
[02:47:15] <SWPadnos> unless it's different depending on direction (backlash)
[02:47:32] <jmkasunich> then it acts just like backlash comp
[02:48:20] <SWPadnos> where is the feedback correction going to be applied?
[02:48:36] <SWPadnos> I guess it's in motion, since that's where the comp code is
[02:49:00] <jmkasunich> http://jmkasunich.com/pics/emc2-motion-dataflow.pdf
[02:49:05] <jmkasunich> all of that will be in motion
[02:49:41] <jmkasunich> most if not all of the right hand side (right of kins) already is
[02:50:06] <jmkasunich> comp is added to outgoing position, and subtracted from incoming feedback
[02:52:58] <SWPadnos> I guess if motion won't oscillate on its own, then PID probably won't help it do so
[02:53:17] <jmkasunich> motion can't oscillate
[02:53:28] <SWPadnos> my concern is for locations where comp is dependent on direction
[02:54:39] <SWPadnos> oh right, the compensation numbers will be generated with the actual drive mechanism, so the difference between feedback and command will have been taken into account
[02:54:41] <SWPadnos> duh
[11:09:26] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Master_User.lyx: Removed install and configure topics covered in Integrator Manual
[11:10:41] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/Master_User.lyx: Removed install and configure topics covered in Integrator Manual
[14:03:24] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Master_Integrator.lyx: Added MPG example
[14:03:25] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/examples/mpg.lyx: Added MPG example
[14:04:36] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/Master_Integrator.lyx: Added MPG example
[14:05:56] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/examples/mpg.lyx: Added MPG example
[14:29:25] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/user_intro.lyx: Trimmed operator intro
[14:33:36] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/common/user_intro.lyx: Trimmed operator intro
[15:46:14] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/user_intro.lyx: A little clean up
[15:47:12] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/common/user_intro.lyx: A little clean up
[17:00:23] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Submakefile: Add reference in G92
[17:00:26] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gcode/ (coordinates.lyx main.lyx): Add reference in G92
[17:02:52] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/Submakefile: Add reference in G92
[17:02:57] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/gcode/ (coordinates.lyx main.lyx): Add reference in G92
[17:13:52] <Roguish> morning all. is there a pin or something hal for the tool radius or diameter?
[17:17:14] <BigJohnT> what are you doing?
[17:19:09] <Roguish> trying to get the diameter into vismach.
[17:19:34] <Roguish> via a program variable of some sort, rather than hard code.
[17:19:48] <BigJohnT> can you get that from the tool table?
[17:20:00] <Roguish> just did a hal show, and did not see anything.
[17:20:18] <Roguish> i guess that's what i'm looking for. someway to get it from the tooltable.
[17:21:14] <BigJohnT> did you look in halui? I'm just guessing I have not looked
[17:22:07] <Roguish> halui has
[17:22:20] <Roguish> 'tool_offset' and 'tool_number'
[17:22:39] <BigJohnT> can you use the tool number to look in the tool file?
[17:25:16] <Roguish> i have no idea. i'm not really a programmer and have no idea how to access that data.
[18:01:13] <alex_joni> tool radius and offset should be the same.. right?
[18:04:49] <Roguish> just got back, had to boot. i don't know. sorta makes sense. but i don't know how it's programmed.
[18:06:47] <Roguish> there is motion.tooloffset.x but i'm watching in the show window, and it doesnot change with tool change. stays at 0. i do not have tool comp turned on.
[18:52:01] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/ (Master_User.lyx Submakefile): Add User foreword
[18:53:56] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/userforeword.lyx: added file
[20:17:40] <fenn_> fenn_ is now known as fenn
[20:22:20] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/ (Master_User.lyx Submakefile): added user foreword
[20:22:21] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/common/userforeword.lyx: added user foreword
[21:26:18] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Master_Integrator.lyx: Some spell checking and put missing section back in
[21:26:19] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/install/installing_emc2.lyx: Some spell checking and put missing section back in
[21:26:22] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/config/ini_config.lyx: Some spell checking and put missing section back in
[21:41:31] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/config/ini_config.lyx: Some spell checking and put missing section back in
[21:41:38] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/Master_Integrator.lyx: Some spell checking and put missing section back in
[21:41:38] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/install/installing_emc2.lyx: Some spell checking and put missing section back in
[21:59:56] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/motion/tweaking_steppers.lyx: Add reference to latency test
[22:01:17] <SWPadnos> Roguish, why do you need tool offset in vismach?
[22:01:42] <SWPadnos> just to make the image of the tool more accurate?
[22:02:50] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/motion/tweaking_steppers.lyx: Add reference to latency test
[22:30:34] <SWPadnos> wow. what a stupid bug I've found
[22:31:19] <SWPadnos> if you have a USB thumb drive with the U3 launcher on it, and you decide to remove U3 (to get back that couple hundred meg), there's a nice option in the U3 software to help you
[22:31:31] <SWPadnos> it even offers to back up your data and restore it after formatting the drive
[22:32:14] <SWPadnos> unfortunately, if you realize thast you've plugged it into a USB1.1 port and you decide to plug it into a USB2.0 port instead, and you hit cancel during the backup phase, it only cancels the backup - but it still formats the drive
[22:32:19] <SWPadnos> and therefore all your data
[23:01:52] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/user_intro.lyx: added more spindle info
[23:01:55] <CIA-34> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/examples/spindle.lyx: added more spindle info
[23:02:33] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/common/user_intro.lyx: added more spindle info
[23:02:33] <CIA-34> EMC: 03bigjohnt 07v2_2_branch * 10emc2/docs/src/examples/spindle.lyx: added more spindle info
[23:02:48] <SWPadnos> heh - how timely :)
[23:03:53] <BigJohnT> lol I was typing as fast as I could
[23:13:30] <jmkasunich> BigJohnT: many thanks for the manual work - you're committing up a storm
[23:15:29] <BigJohnT> yea, sometimes twice to get the desired results :)
[23:16:26] <SWPadnos> heh
[23:22:46] <cradek> you're my hero too BigJohnT
[23:22:58] <SWPadnos> me too, me too! :)
[23:23:02] <BigJohnT> * BigJohnT turns red in the face
[23:23:51] <BigJohnT> I have learned a lot about cvs and lyx
[23:25:15] <BigJohnT> I did notice this during make
[23:25:17] <BigJohnT> Depending ../docs/src/Master_Integrator.lyx
[23:25:19] <BigJohnT> make: Failed to remake makefile `Makefile'.
[23:25:20] <BigJohnT> make: Failed to remake makefile `Makefile'.
[23:25:24] <BigJohnT> is that important?
[23:25:48] <cradek> yeah it means something is screwy. sometimes make --debug=b,m will let you divine what
[23:26:20] <BigJohnT> I d/l fresh from cvs just to make sure
[23:26:34] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?FailedToRemakeMakefile
[23:27:15] <BigJohnT> ok thanks cradek your my hero too
[23:42:02] <BigJohnT> the french manual is still looking for a file...