#emc | Logs for 2005-03-06

Back
[00:00:01] <cnc_wright_> yes I can...Where do I get an md5sum checker
[00:00:16] <pfred1> cnc_wright_ which md5sum
[00:00:30] <pfred1> /usr/bin/md5sum
[00:00:35] <paul_c> lemme guess - You downloaded & burt with M$ ?
[00:00:43] <pfred1> /exec -o which md5sum
[00:01:05] <cnc_wright_> ummmmmm aaaaaaaaaaaa yeah
[00:03:48] <paul_c> http://www.irnis.net/gloss/md5sum-windows.shtml
[00:04:18] <paul_c> http://www.openoffice.org/dev_docs/using_md5sums.html#links
[00:04:35] <anonimasu> hm..
[00:04:37] <anonimasu> nml playtime.
[00:07:13] <cnc_wright_> md5sum is d29f3bb5bf3361aac0818fa384e62f0a
[00:07:43] <pfred1> cnc_wright_ go back to wherever yo udownloaded it from get the md5sum there
[00:07:58] <pfred1> they should have one same place as the file was
[00:09:09] <cnc_wright_> d29f3bb5bf3361aac0818fa384e62f0a
[00:09:11] <paul_c> d29f3bb5bf3361aac0818fa384e62f0a - The original...
[00:09:12] <cnc_wright_> they match
[00:09:22] <pfred1> then you're all good
[00:09:34] <paul_c> unless it was a bad burn...
[00:11:24] <anonimasu> paul_c: do you have that page about NML somwehre?
[00:11:34] <anonimasu> somwhere I cant find where I put the bookmark
[00:11:39] <paul_c> How many times have you tried installing from the CD ?
[00:11:57] <pfred1> paul_c til smoke started pouring out of the HDD!
[00:12:23] <paul_c> http://www.isd.mel.nist.gov/projects/rcslib/
[00:12:31] <anonimasu> thanks
[00:12:45] <paul_c> google "nist nml tutorial"
[00:13:21] <cnc_wright_> once on the PII machine. I previously had linux 2.4.27 on it. I nuked the partitions on bdi install. I also tried the cd on a PII I810 mobo but the onboard graphics have issues.
[00:14:32] <cnc_wright_> What is good el-cheapo pci graphics card for linux/emc/xfree86?
[00:14:38] <paul_c> 'bout all I can suggest is burning another CD, but select a low speed...
[00:14:50] <paul_c> * paul_c likes ATI cards.
[00:15:50] <cnc_wright_> I will give that a try tomorrow. Wife is dragging me to dinner at friends house...no more linux hacking tonight :(
[00:16:40] <paul_c> Are you on the east coast ?
[00:17:01] <cnc_wright_> nope...Portland, Or area
[00:17:18] <paul_c> fsckit - Was over there last month.
[00:17:47] <paul_c> Tigard & Beaverton to be exact.
[00:19:46] <cnc_wright_> I am across the Columbia river in Camas, WA. I try to stay away from Beaverberg as much has I can...way too many behemoth SUV's with clueless drivers yaking on their cell phones.
[00:20:59] <pfred1> hmm I wonder if when I record I am overdriving the mic?
[00:21:02] <paul_c> OK.. North of the boarder, you have a couple of fellow EMC users nearby
[00:21:26] <cnc_wright_> Really? This is good to hear.
[00:21:48] <anonimasu> bleh I am too stupid to understand how to compile stuff that uses nml
[00:21:51] <anonimasu> :)
[00:22:08] <pfred1> anonimasu never know
[00:22:21] <paul_c> If you take a look at the map at www.linuxcnc.org, you can get an idea of where some of them are.
[00:22:42] <cnc_wright_> I will do that.
[00:23:17] <paul_c> anonimasu: Which source tree are you using ?
[00:23:46] <anonimasu> emc1
[00:24:27] <paul_c> Switch to the emc2 tree - You have the bdi-4 branch....
[00:24:37] <paul_c> the API is the same.
[00:25:03] <anonimasu> hm, but I cant run segmentqueue with emc2..
[00:25:12] <anonimasu> or can you?
[00:25:49] <paul_c> * paul_c could probably hack segmentqueue to work with a 2.6 kernel...
[00:26:10] <anonimasu> hm, maybe in a bit :)
[00:26:26] <anonimasu> I am trying to send a nml message/read one..
[00:26:44] <anonimasu> but I dont quite get how to compile the examples..
[00:27:17] <pfred1> anonimasu when all else fails read the directions!
[00:27:33] <anonimasu> pfred1: where are they?
[00:27:58] <pfred1> give me link to this stuff i want to see how hard it is
[00:28:14] <anonimasu> http://www.isd.mel.nist.gov/projects/rcslib/NMLcpp.html
[00:28:21] <paul_c> cd rcslib/src/test
[00:28:29] <paul_c> make PLAT=nonrealtime
[00:28:42] <pfred1> man couldn't give me file so i could wget it? :(
[00:28:49] <pfred1> making it hard already!
[00:29:00] <paul_c> cvs co rcslib
[00:29:14] <paul_c> fskit - Make fails...
[00:29:47] <anonimasu> ah.
[00:30:19] <anonimasu> thanks.
[00:31:34] <paul_c> darn - Don't have any of the nml test code that I hacked out on this box...
[00:31:44] <anonimasu> :/
[00:31:57] <Jymmm> Is a ShopSmith worth anything?
[00:32:09] <anonimasu> do you need java or somthing to compile it?
[00:33:58] <paul_c> java is only needed for generating nmlex1.cc
[00:34:23] <anonimasu> ah ok
[00:34:36] <paul_c> http://www.isd.mel.nist.gov/projects/rcslib/nml_ex1.cc
[00:34:44] <anonimasu> make PLAT=nonrealtime -i
[00:35:57] <anonimasu> * anonimasu pokes java with a stick
[00:36:58] <paul_c> eeuww... You don't want to do that... It might bite
[00:37:30] <anonimasu> yeah
[00:37:46] <anonimasu> I think java is a fashion language.. :)
[00:38:54] <CIA-4> 03paul_c * 10rcslib/src/test/nml_ex1.cc: Committed nml_ex1.cc so that some of the tests can be compiled without the java stuff being in place.
[00:39:08] <anonimasu> there's a define for that..
[00:39:17] <anonimasu> NO_JAVA
[00:43:17] <pfred1> Jymmm nickel a pound scrap
[00:44:39] <pfred1> I donno i have to try miccing this further away i guess
[00:55:14] <anonimasu> paul_c: that change is just in the non anonymous checkout right now is it?
[00:56:53] <paul_c> it will take a few hours to hit the public servers... But you can still get it from the NIST link...
[00:58:09] <anonimasu> nist link
[00:58:37] <paul_c> scroll up a few lines.
[00:58:40] <anonimasu> ah that one
[00:58:44] <paul_c> ;)
[01:03:56] <anonimasu> hm..
[01:05:34] <anonimasu> what do I do after I've compiled it?
[01:07:45] <anonimasu> it goes into a object..
[01:08:18] <paul_c> link it to librcs
[01:08:30] <anonimasu> ah..
[01:14:31] <anonimasu> I have no idea on how to do that really :/
[01:14:45] <anonimasu> ld -o nml_ex1 nml_ex1.o -lnml ?
[01:20:38] <paul_c> You need a little more than that I'm afraid...
[01:22:04] <paul_c> ld -o nml_ex2 nml_ex1.o nml_ex2.o -Lplat/nonrealtime/lib -lnml
[01:22:26] <paul_c> nml_ex1.o is only part of the equation...
[01:22:57] <anonimasu> hm ok
[01:25:09] <paul_c> lemme see if I can find the samples I hacked.
[01:25:17] <anonimasu> yeah
[01:25:21] <anonimasu> that'd be the best
[01:28:00] <paul_c> first, locate which box they are on...
[01:30:02] <anonimasu> I am sorry for being a pita :/
[01:32:22] <paul_c> if it wern't you, it would be someone else ;-p
[01:32:26] <anonimasu> lol
[01:33:00] <anonimasu> if I just find out how to send and to receive a nml message to a c++ program then the rest isnt a problem :)
[01:33:37] <paul_c> DCC OK for you ?
[01:33:41] <anonimasu> yes
[01:34:22] <paul_c> Unpack that tarball in the bdi-4 tree
[01:34:50] <paul_c> in the src/ dir
[01:36:07] <paul_c> It may not compile...
[01:38:49] <anonimasu> is the bdi-4 tree the one that will compile on 2.6.x
[01:39:24] <anonimasu> err nm
[01:39:58] <anonimasu> I was just being stupid
[01:39:59] <anonimasu> :D
[01:40:05] <anonimasu> it does compile
[01:41:38] <paul_c> The incoming has been tested and compiles..
[01:42:35] <anonimasu> ok
[01:42:40] <anonimasu> I'll try it now
[01:42:48] <anonimasu> just need to wait for zsh to install
[01:42:58] <anonimasu> I dont match with the tab completion of bash..
[01:50:44] <anonimasu> works and compiles
[01:50:59] <anonimasu> (NML_INVALID_CONFIGURATION)
[01:51:32] <anonimasu> when running the nml_ex2 ?
[01:54:37] <paul_c> OK... The files probably need debugging - I pulled the from an experimental src tree...
[01:55:40] <anonimasu> I am trying to run with the generic.nml file..
[01:55:44] <anonimasu> instead ;)
[01:57:02] <anonimasu> cms_cfg.cc 947: cms_config: invalid buffer_type (192.168.0.3:5000)
[01:57:03] <anonimasu> cms_cfg.cc 949: cms_config: buffer_line = (B ex_buf1 192.168.0.3:5000 512 0 1 11 101
[01:57:03] <anonimasu> )
[01:57:10] <anonimasu> cms_cfg.cc 947: cms_config: invalid buffer_type (192.168.0.3:5000)
[01:57:11] <anonimasu> cms_cfg.cc 949: cms_config: buffer_line = (B ex_buf1 192.168.0.3:5000 512 0 1 11 101
[01:57:11] <anonimasu> )
[01:57:14] <anonimasu> err
[01:57:14] <anonimasu> nml.cc 1066: NML::read: CMS not configured.
[01:57:15] <anonimasu> (time=1110073847.069649,pid=6551): deleting NML (0)
[01:57:15] <anonimasu> (time=1110073847.070845,pid=6551): Leaving ~NML()
[02:00:45] <anonimasu> sint it because I lack a nml server?
[02:00:47] <anonimasu> isnt..
[02:10:39] <paul_c> I don't know at the moment - It's been a while since I messed with that code...
[02:10:52] <paul_c> Need to look it over again.
[02:10:55] <anonimasu> ok
[02:10:59] <anonimasu> it's no hurry
[02:12:07] <anonimasu> it's pretty late so I'll be heading to bed in a bit
[02:12:25] <paul_c> We will look it over tomorrow sometime.
[02:12:35] <anonimasu> yeah I'll be here most of the day/night
[02:13:10] <paul_c> time for bed.
[02:13:40] <anonimasu> night paul
[02:13:45] <anonimasu> thanks for the help
[02:13:47] <anonimasu> :)
[03:09:19] <jmkasunich> night all
[06:25:58] <rob_melb> hi all
[06:27:46] <rob_melb> I installed the emc from the BDI deb repository. but I can't get the simulation to go I'm getting errors when emc isrtai_alloc(100 (0x64), 688624 (0xA81F0) ) failed starting about
[06:28:26] <rob_melb> I installed the emc from the BDI deb repository. but I can't get the simulation to go I'm getting errors when emc is starting like this: rtai_alloc(100 (0x64), 688624 (0xA81F0) ) failed
[06:28:37] <rob_melb> after I run /usr/local/emc/generic.run
[09:41:15] <robin_sz> meep
[09:45:22] <nevyn> ok. the readme claims that rtai is not required to run generic. but when I try running generic after installing from the bdi deb repository I get errors about freqmod any pointers?
[09:45:53] <nevyn> starting EMC MOTION PROGRAM -- freqmod...
[09:56:12] <nevyn> and then it dies.
[09:56:49] <robin_z> does it say anything else? or just die?
[09:57:10] <nevyn> ok. the readme claims that rtai is not required to run generic. but when I try running generic after installing from the bdi deb repository I get errors about freqmod any pointers?
[09:57:20] <nevyn> starting EMC MOTION PROGRAM -- freqmod...
[09:57:52] <nevyn> and then it dies.
[09:58:00] <robin_z> no ... you DO need a realtime system to run generic ... AFAIK. its sim.run that needs no realtime kernel
[09:58:24] <nevyn> that's not what the README says :)
[09:58:30] <robin_z> try sim.run
[09:58:37] <nevyn> will do
[09:59:32] <nevyn> ok the other thing is on a system with the realtime modules and kernel I get rtai_alloc(100 (0x64), 688624 (0xA81F0) ) failed
[10:00:20] <robin_z> this is debian with whatever rt kernel is in the repository?
[10:00:26] <nevyn> yeah
[10:00:32] <robin_z> an you are root?
[10:01:10] <nevyn> hrm. no. but I did sudo bash -c /usr/local/emc/generic.run
[10:01:17] <robin_z> k
[10:01:18] <nevyn> robin_z: ok sim.run I don't seem to have..
[10:01:46] <robin_z> bah. lots must have changed since I last looked at it.
[10:02:33] <robin_z> to the best of my knowledge generic uses freqmod and freqmod is an RT module,
[10:02:48] <nevyn> right.
[10:04:03] <anonimasu> good morning
[10:04:11] <robin_z> nevyn: uname -a
[10:04:41] <nevyn> robin_z: I'm not on that box. but it loads the modules.. I see rtai init in kernelspace
[10:04:53] <robin_z> k
[10:05:10] <nevyn> robin_z: I'd install it here but that kernel doesn't boot my (slightly esoteric) configuration
[10:05:25] <anonimasu> did you compile emc or did you run the default?
[10:05:26] <nevyn> root is on md ;)
[10:05:31] <nevyn> run the default.
[10:05:46] <nevyn> this is from the bdi deb repo
[10:05:54] <robin_z> paul_c is the keeper of the .debs, he usually appears about now
[10:06:07] <A-L-P-H-A> A-L-P-H-A has changed the topic to: Just testing Welcome to the Enhanced Machine Control forum | Regular Developer's meetings every Sunday between 14:00 & 18:00 GMT
[10:06:07] <ChanServ> ChanServ has changed the topic to: Welcome to the Enhanced Machine Control forum | Regular Developer's meetings every Sunday between 14:00 & 18:00 GMT
[10:06:11] <A-L-P-H-A> heheh.
[10:06:25] <nevyn> I've already run in with paul_c :)
[10:06:33] <nevyn> so has that meeting been already?
[10:06:41] <robin_z> nah
[10:06:41] <A-L-P-H-A> was it a good run in? or a wanna beat him up run in?
[10:07:18] <nevyn> umm I know debian policy and these break really fundamental ones kinda runin ;:)
[10:07:28] <robin_z> well, yes,
[10:07:37] <robin_z> but some bits of debian policy are mad
[10:07:47] <nevyn> some bits are. the bit about /usr/local isn't
[10:08:01] <nevyn> if you really want to be out of the normal tree use /opt it's what it's there for.
[10:08:17] <robin_z> no that bit is probably sane, i forget exactly what happens in usr/local for emc
[10:08:42] <robin_z> nevyn: can I hire you to explain that to DJB too?
[10:09:16] <nevyn> the fact I had to make a simlink in my homedirectory to /usr/local/emc/plat before things would even attempt to run was very dodgy too
[10:09:19] <nevyn> :)
[10:09:21] <robin_z> as you say, /opt is probably the place for it
[10:09:29] <nevyn> hi paul_c :)
[10:09:56] <nevyn> robin_z: djb is beyond anybody's reasoning.
[10:09:57] <robin_z> nevyn: for various reasons I run qmail on debian, the madness of DJB's use of dirs always iritates me
[10:09:57] <paul_c> Morning
[10:10:13] <robin_z> nevyn: too true.
[10:10:14] <nevyn> robin_z: we're mid migration from qmail to postfix
[10:10:38] <nevyn> mainly because of djb licencing issues.
[10:10:46] <robin_z> nevyn: sadly, I dont think I can, as I use vpopmail/qmail/sqwebmail for virtual domains, i dont think postfix/exim are up to that
[10:11:06] <nevyn> umm I'm sure postfix or courier can do it.
[10:11:19] <nevyn> courier for a vhosted type setup seems very elegant
[10:11:23] <robin_z> djb's licensing is nutty, but there again, some of the debian idiocy is just as bad
[10:11:32] <nevyn> robin_z: eg?
[10:12:22] <robin_z> leaving out important packages because the wording of the licenses has an "a" rather than "the" in a cerrtain place
[10:13:43] <nevyn> the problem is the licence is the licence and a small technicality like that should be fixed in the licence. something as small as that can change the entire meaning of a passage/clause
[10:13:52] <nevyn> eg:
[10:14:41] <robin_z> php
[10:14:45] <anonimasu> morning paul
[10:15:08] <nevyn> php's licence is broken take it up with rasmus :)
[10:15:38] <A-L-P-H-A> what's borked with php's license?
[10:15:41] <nevyn> is there any active development on CNCMILL by Phil Cobbin ?
[10:15:52] <A-L-P-H-A> oh, anyone tried out lighttpd?
[10:16:18] <robin_z> A-L-P-H-A: sane people us apache :)
[10:16:30] <nevyn> well....
[10:16:33] <A-L-P-H-A> i am. just wondering... hearing good things about it.
[10:16:34] <anonimasu> paul_c: wouldnt the handwheel.cc be great to look at?
[10:17:25] <nevyn> so deskcnc what's the bad and ugly of it? anyone used it?
[10:17:51] <anonimasu> nope
[10:18:18] <A-L-P-H-A> don't know about that one. I tried kellycam to do my excellon to gcode. Gave up on it, and wrote my own php script to do it for me.
[10:19:35] <nevyn> have a friend in my model engineering club who just bought a "solution" from cnctecnik that included it.
[10:19:43] <robin_z> looks interesting
[10:20:02] <robin_z> looks like a xilinx fpga and a serial link :)
[10:20:18] <robin_z> and it runs under 'doze ...
[10:20:26] <robin_z> looks neat.
[10:20:56] <nevyn> robin_z: check their bargin bin.
[10:21:06] <nevyn> there are fpga's going cheap ;)
[10:22:04] <robin_z> uri?
[10:22:22] <robin_z> can I take a gues sfirst?
[10:22:26] <A-L-P-H-A> ebay!
[10:22:40] <robin_z> xilinx spartan3 in a tq100 package ;)
[10:23:09] <A-L-P-H-A> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=4660&item=3877599057&rd=1&ssPageName=WDVW ??
[10:23:35] <anonimasu> hm..
[10:23:54] <anonimasu> neat
[10:24:08] <A-L-P-H-A> I don't touch, nor know what to do with an FPGA.
[10:24:15] <robin_z> the more I see of the way cnc is going, the more I become convinced that off-board pulse generation is the way to go
[10:24:18] <A-L-P-H-A> I just started tinkering with atmels. :)
[10:24:37] <A-L-P-H-A> robin_z, like g2002?
[10:24:39] <robin_z> tried the xilinx software yet?
[10:24:54] <anonimasu> * anonimasu is too
[10:25:02] <A-L-P-H-A> anonimasu is what?
[10:25:04] <anonimasu> although that requires very good algorithms..
[10:25:14] <robin_z> A-L-P-H-A: possibly the G2002, or something similar
[10:25:18] <anonimasu> A-L-P-H-A: convinced that you need a external pulse generator
[10:25:33] <A-L-P-H-A> yes. Parallel ports are going the way of the dodo
[10:25:45] <robin_z> A-L-P-H-A: try the xilinx 'webpack' ISE .. its free and VERY nice to use.
[10:25:55] <anonimasu> it's got to be loads smoother with a external one
[10:26:20] <A-L-P-H-A> well, I'm sure someone could do one for the atmel, an external pulse generator.
[10:26:30] <A-L-P-H-A> swpandos has one made.
[10:26:32] <robin_z> A-L-P-H-A: its not the // port issue .. its the "its not worth task swapping an entire 686 machine just to do a pulse edge" issue
[10:26:48] <anonimasu> A-L-P-H-A: Oh I could write one for atmel..
[10:27:01] <anonimasu> A-L-P-H-A: but I think real hardware is the way to go..
[10:27:05] <robin_z> A-L-P-H-A: you do have win XP online?
[10:27:11] <A-L-P-H-A> heck! I'm still having issues writing that tachometer.
[10:27:45] <A-L-P-H-A> i'm on winxp here (workstation), and on the laptop. The CNC machine is linux
[10:27:48] <A-L-P-H-A> well, the BDI.
[10:27:53] <nevyn> robin_z: but hard realtime OS's are cool
[10:27:56] <nevyn> ;)
[10:28:01] <anonimasu> if you can get servos to work with g2002 I want it..
[10:28:16] <anonimasu> * anonimasu is getting unsure if a vital card is better then a g2002
[10:28:17] <A-L-P-H-A> anonimasu, don't they just interface with Step and direction?
[10:28:18] <anonimasu> is it?
[10:28:24] <robin_z> A-L-P-H-A: well, go to xilinx and try the webpack ISE software, its awesome .. truly.
[10:28:29] <anonimasu> A-L-P-H-A: I have g340's right now..
[10:28:42] <A-L-P-H-A> rob_melb, what's it do? it maybe over my head.
[10:28:43] <anonimasu> so it probably should..
[10:28:45] <robin_z> anonimasu: the new G200X will do real servos, if you want ti to
[10:29:00] <anonimasu> ok..
[10:29:06] <robin_z> anonimasu: it has 6 encoder inputs and 4 a-d outputs
[10:29:10] <anonimasu> I guess I'll take my time before deciding..
[10:29:15] <anonimasu> I dont have drives like that anyway..
[10:29:19] <anonimasu> yet
[10:29:30] <robin_z> well it has step/dir already
[10:29:44] <robin_z> the hardware is there to implenent real pid on the rabbit
[10:29:50] <robin_z> the ideal for me would be
[10:29:53] <anonimasu> I coulb buy new drives but $650 per drives is expensive
[10:30:00] <anonimasu> drive..
[10:30:04] <A-L-P-H-A> so, what's the design process for programming FPGAs? Is is like ASM? C? even higher levels?
[10:30:08] <anonimasu> or well, I should buy new drives.
[10:30:08] <robin_z> PID on the rabbit, plan the moves onthe PC
[10:30:29] <robin_z> A-L-P-H-A: drag logic gates onto a diagram, wire them up
[10:30:45] <robin_z> A-L-P-H-A: even complex gartes, like 16bit latch,
[10:30:49] <A-L-P-H-A> how many times can you reprogram the chip?
[10:30:59] <robin_z> as many as you like, if its a flash one
[10:31:10] <A-L-P-H-A> I see 100K gates... $2 for 500K units.
[10:31:41] <A-L-P-H-A> so, spend a million bonkers, for 1/2 mill of chips.
[10:31:49] <robin_z> just download the webpack ISE and play with it ... I reproduced the entire G2002 circuit of ~100 LS ttl chips in an evening
[10:32:04] <nevyn> robin_z: cnctechnix site is completely un navigateable without FLASH!
[10:32:08] <A-L-P-H-A> there's 100 LS ttl chips in the G2002?
[10:32:20] <A-L-P-H-A> dang.
[10:32:20] <robin_z> the spartan3 small chip is like less than 10 dollars
[10:32:26] <nevyn> I'm sitting here going wtf this isn't what I looked at at rob's place EVIL
[10:32:36] <robin_z> A-L-P-H-A: noit any more .. replaced with G200X .. 1 fpga
[10:32:53] <A-L-P-H-A> not that's impressive.
[10:32:59] <anonimasu> yep
[10:33:15] <A-L-P-H-A> now... q. could you do that FPGA as an atmel? [just for the educational purpose]
[10:33:28] <A-L-P-H-A> not=now
[10:33:41] <robin_z> I couldnt ... because I dont have the atmel dev stuff
[10:33:56] <robin_z> and when I looked at it it required me to know verilog/vhdl
[10:34:13] <A-L-P-H-A> robin_z, got the flow chart of how the g2002 is supposed to do stuff?
[10:34:30] <robin_z> the xilinx stuff is ueable by anyone that can draw a circuit
[10:35:26] <robin_z> A-L-P-H-A: there are different ways ... its all open .. it goes from marisss original concept (just tell it the *ABSOLUTE* position to go to, and it goes)
[10:35:42] <A-L-P-H-A> my understanding is that the G2002 is just a black box. Send in Gcode over a usb bus, and it decyphers, and implements the pulses for me.
[10:35:47] <robin_z> to arts segments ...(just takes emc like segments and runs them)
[10:36:01] <robin_z> or steve hardys EMC embedded interp
[10:36:11] <robin_z> or write your own :)
[10:36:31] <robin_z> the original concept was to send positions
[10:36:38] <A-L-P-H-A> so it's a black box, where you program the code, and it generates the pulse for me.
[10:36:52] <robin_z> its gens the pulses yes
[10:37:05] <robin_z> but you can choose how much or little is done on the PC
[10:37:07] <A-L-P-H-A> so, computer -> g2002 -> stepper drivers
[10:37:10] <robin_z> yes
[10:37:18] <A-L-P-H-A> 'ight.
[10:37:29] <robin_z> heres the normal CNC cycle
[10:37:41] <A-L-P-H-A> so what use it is, if I have only 3axis, and my computer can send more pulses than my steppers can handle already?
[10:37:50] <robin_z> GCODE->interp->motion_planner->pulse gen
[10:37:56] <anonimasu> A-L-P-H-A: precision...
[10:38:05] <anonimasu> smooth pulsetrains..
[10:38:06] <anonimasu> :)
[10:38:11] <A-L-P-H-A> ahh.
[10:38:13] <robin_z> A-L-P-H-A: 1 sommth pulse trains
[10:38:31] <robin_z> 2 you cant send more pulses than your steppers can handle already
[10:38:44] <A-L-P-H-A> well, I get some resonance happening.
[10:38:53] <robin_z> EMC is not capable of driving a microsteppign drive to its potential.
[10:39:06] <robin_z> sad fact, but its a fact
[10:39:14] <A-L-P-H-A> thanks! No you made me feel that my machine is enimic.
[10:39:18] <A-L-P-H-A> no=now
[10:39:23] <anonimasu> haha
[10:39:24] <anonimasu> :)
[10:39:54] <A-L-P-H-A> so where do I go get a g2002? they arne' tfor sale yet... are they?
[10:40:04] <robin_z> you really need to see how SMOOTH pulses can spin big steppers really fast and smooth
[10:40:13] <robin_z> ive had my G2002 over a year
[10:40:22] <A-L-P-H-A> I thought you were testing them though.
[10:40:24] <A-L-P-H-A> like berta
[10:40:25] <A-L-P-H-A> beta
[10:40:30] <robin_z> the original is now withdrawn from sale, the 200X is weeks away
[10:40:39] <robin_z> life is a beta test :)
[10:40:50] <A-L-P-H-A> what's the asking price?
[10:41:02] <robin_z> dunno .. not announced
[10:41:15] <A-L-P-H-A> I stopped recieving the geckodrive email list on yahoo.
[10:41:25] <A-L-P-H-A> just never read it... I had like 1500 emails in there, which I never read
[10:41:33] <A-L-P-H-A> so I realize at that point, to just leave it be.
[10:41:40] <robin_z> it would be nice to see it as off-baord pulse gen for emc
[10:41:43] <A-L-P-H-A> otherwise I'd be just reading it all day, and never get anything done.
[10:41:46] <anonimasu> yep
[10:42:04] <robin_z> it could flug in just where freqmod plugs in now ...
[10:42:36] <anonimasu> yeah that's what I think would be the best..
[10:43:02] <robin_z> hard realtime kernels are fun, but trying to generate SMOOTH pulse trains at 10 or 20 khz is not their forte im afraid
[10:44:08] <robin_z> with encoder inputs and analogue ouptuts, theres some VERY interesting potential for the G200X
[10:45:01] <A-L-P-H-A> So, essentially, if I were to write say, turboCNC's motion routines, into ASM on an atmel, I'd have the same thing as a gecko 200x, except in a single chip. So I'd just send Computer -> TTL -> Atmel -> pulses -> stepper drivers
[10:45:13] <anonimasu> there's a bit more I am afraid..
[10:45:40] <A-L-P-H-A> anonimasu? like?
[10:46:23] <A-L-P-H-A> brb. need water.
[10:46:25] <anonimasu> A-L-P-H-A: RTOS ;)
[10:46:47] <A-L-P-H-A> anonimasu, so I'd use an atmega128, and dump freeRTOS onto one.
[10:46:58] <robin_z> the way its done on the g2002 ..
[10:47:06] <robin_z> have an external clock divider
[10:47:13] <robin_z> generate interrupts
[10:47:18] <robin_z> yeah, I know
[10:47:21] <robin_z> but it works
[10:47:23] <anonimasu> I'd buy a g200x ;)
[10:47:38] <robin_z> theres pics of the board inthe files section
[10:47:44] <anonimasu> or build one off a fpga if I had time but I dont..
[10:48:14] <robin_z> quite
[10:48:28] <anonimasu> or just buy a vital card..
[10:48:29] <anonimasu> :)
[10:48:41] <robin_z> well ... thats possible
[10:48:47] <robin_z> this should be cheapoer
[10:48:59] <anonimasu> yep
[10:49:11] <robin_z> im hopingless than $200
[10:49:43] <anonimasu> that's a nice price
[10:51:01] <A-L-P-H-A> k, my electronics knowledge is limited [above basic]. What's a clock divider do?
[10:51:02] <roel> hi
[10:51:06] <robin_z> we'll see
[10:51:09] <robin_z> hi roel
[10:51:18] <robin_z> etched anything nice today?
[10:51:47] <roel> nope nothing 'm waiting on your order robin :)
[10:51:54] <robin_z> heh ;)
[10:53:08] <robin_z> theres always the lamp-dimmer boards ... we get them far east at the moment
[10:53:56] <robin_z> but you wouldnt want to compete agaisnt far-east pricing
[10:54:39] <nevyn_> nevyn_ is now known as nevyn
[10:54:47] <roel> for mega orders thats the place
[10:55:00] <robin_z> yeah
[10:55:13] <robin_z> we do 25,000 boards a month from there
[10:55:33] <A-L-P-H-A> lamp dimmers?
[10:55:36] <robin_z> yeah
[10:55:39] <A-L-P-H-A> robin_z, what do you make?
[10:55:41] <A-L-P-H-A> lamps?
[10:55:49] <A-L-P-H-A> :D
[10:56:03] <robin_z> lamp dimmers, its not me .. its .. well. complicated.
[10:56:12] <robin_z> somehow I end buying the components
[10:56:14] <A-L-P-H-A> as is life.
[10:57:06] <A-L-P-H-A> freak'n hell. FF1.01 still has rendering issues with slashdot. Lets see if the newest nightly will as well
[10:57:54] <A-L-P-H-A> chocolate truffles aren't keeping me awake anymore.
[10:58:21] <robin_z> I do enjoy buying components though ... theres something nice about asking a price and then as they are working it out saying "I'll need 100,000 pieces" .. and listening to them panic
[10:58:51] <A-L-P-H-A> yah! the newest nightly works.
[10:59:25] <A-L-P-H-A> robin_z, I'm curious. what do you guys make? I'll stop asking after this.
[10:59:34] <robin_z> I told you, lamp dimmers
[11:00:04] <robin_z> mains in ... pot .. variable mains out
[11:00:05] <A-L-P-H-A> oh, I thought you bought lamp dimmers.
[11:00:17] <A-L-P-H-A> just the boards then.
[11:00:24] <robin_z> no .. we make them .. PCB, components .. bunch of girls
[11:00:36] <A-L-P-H-A> you make girls? can I have one?
[11:00:43] <roel> i make only the boards (pcb's)
[11:00:46] <A-L-P-H-A> take custom orders?
[11:00:47] <A-L-P-H-A> :D
[11:00:48] <robin_z> you wouldnt want the ones we have in stock
[11:01:14] <A-L-P-H-A> so they sit there, solder everything together.
[11:01:41] <robin_z> most goes on SMD, Zevatech pick and place machines, reflow oven
[11:02:14] <robin_z> the triac, coil and pot, and class X cap go on by hand
[11:02:48] <robin_z> 25,000 units in a normal month ... 45k in a busy one
[11:03:00] <A-L-P-H-A> that's alot.
[11:03:12] <robin_z> theres wave solder for the hand placed components
[11:03:13] <A-L-P-H-A> is that world wide sales? or just local?
[11:03:23] <robin_z> .eu
[11:03:24] <A-L-P-H-A> local as in domestic
[11:03:28] <A-L-P-H-A> ahh
[11:03:44] <A-L-P-H-A> what's the CSA , UL , equivalent in .eu?
[11:03:52] <robin_z> CE marked
[11:04:05] <A-L-P-H-A> ahh
[11:04:43] <robin_z> that covers everyhing
[11:04:51] <robin_z> from EMC emissins
[11:04:55] <robin_z> to paint toxcity
[11:04:58] <robin_z> and lead content
[11:05:23] <robin_z> if you CE mark it .. you are saying "I guarantee it complies with all the regs"
[11:05:53] <A-L-P-H-A> neat.
[11:06:00] <A-L-P-H-A> I know gecko's don't have the CE mark.
[11:06:07] <robin_z> correct
[11:06:09] <A-L-P-H-A> nor any mark. :D
[11:06:17] <A-L-P-H-A> potential firehazard. heh
[11:06:19] <robin_z> thats why no dealers in the eu stock them
[11:06:35] <A-L-P-H-A> I'm sure the dealers don't want their asses sued to oblivion.
[11:06:41] <robin_z> exactly
[11:06:42] <A-L-P-H-A> or confescated by customs.
[11:06:49] <robin_z> you can bring them in
[11:07:03] <A-L-P-H-A> why won't mariss get them certed?
[11:07:07] <A-L-P-H-A> what's the cost?
[11:07:18] <robin_z> if you resell as a "single merchantabel unit" it must be tested and marked
[11:07:40] <robin_z> well, its tricky
[11:07:42] <A-L-P-H-A> each and every individual unit? that seems a little extreme to me.
[11:08:08] <robin_z> ANYTHING sold in the eu MUST be marked
[11:08:11] <robin_z> anything
[11:08:15] <robin_z> pencils ..
[11:08:17] <robin_z> paper ...
[11:08:25] <robin_z> resistors
[11:09:12] <A-L-P-H-A> what's the neat think about eu paper? a4 specifically?
[11:09:22] <A-L-P-H-A> how's it directly proportional to the other sizes?
[11:09:46] <robin_z> well, CE marked paper will not contain toxins,
[11:10:08] <robin_z> if it passes that (probably all pape does) CE mark it and sell ti
[11:10:26] <A-L-P-H-A> think=thing
[11:10:29] <A-L-P-H-A> I'm so tired.
[11:10:29] <robin_z> of the A4 thing ... thats not mandatory ...
[11:10:31] <A-L-P-H-A> Time: 06:14:33 -0500 GMT, Windows XP Professional, Service Pack 2 (5.1 - 2600), AMD XP 2800+ running at (1-AMD , 2123MHz, 512KB (16% Load)), DDR400 RAM Usage: 427/1024MB (41.70%), C: 50.73gb of 75.42gb free, D: 28.29gb of 39.05gb free, E: 0gb of 0gb free, Current Uptime: 3days 15hrs 59mins 34secs, Record Uptime: 3wks 12hrs 48mins 50secs, 3 Samsung 19" flat monitors (1 Trinitron, 2 CRT).
[11:10:48] <robin_z> A-L-P-H-A: take a piece of A4 ... portrait
[11:10:51] <A-L-P-H-A> no, but a4 is like directly half of b4
[11:11:01] <roel> nop
[11:11:03] <robin_z> A-L-P-H-A: fold it in half horiz
[11:11:07] <robin_z> thats A5
[11:11:21] <A-L-P-H-A> a5 < a4??
[11:11:28] <robin_z> Stick two bits of A4 together olong the long edge ... thats A3
[11:11:49] <robin_z> you start with A0 which is fscking huge
[11:11:56] <robin_z> cut it in half .. A1
[11:11:58] <robin_z> etc
[11:12:01] <A-L-P-H-A> I see.
[11:12:14] <A-L-P-H-A> I thought that was the other way around.
[11:12:21] <robin_z> no A0 is huge
[11:12:28] <robin_z> A5 is notepad size
[11:12:35] <robin_z> A5 is half A4
[11:12:53] <robin_z> A3 A4 and A5 are the only normal ones you see
[11:13:01] <A-L-P-H-A> a4 is the most common, isn't it? like for printing from a printer and such.
[11:13:06] <robin_z> yeah
[11:13:29] <A-L-P-H-A> wonder why we have this stupid 8.5x10 or 8.5x11 sheets here. Seems like odd sizes to me.
[11:13:40] <robin_z> dont ask me
[11:13:55] <A-L-P-H-A> well... gotta ask someone old.
[11:13:57] <A-L-P-H-A> :D
[11:14:04] <robin_z> thers all that old "letter" "foolscap" etc nonesense
[11:14:05] <A-L-P-H-A> gezr may know. :D hehe
[11:14:29] <robin_z> anyway .. back to motion control :)
[11:14:33] <roel> b sizes are for offset printers to cut the etge's clean afther printing
[11:14:37] <A-L-P-H-A> hey, I grew up in a metric world, that uses imperial for everything, except for food.
[11:14:52] <A-L-P-H-A> food weights.
[11:15:00] <A-L-P-H-A> horray for Canada.
[11:15:01] <robin_z> really?
[11:15:09] <A-L-P-H-A> damn yanks.
[11:15:13] <A-L-P-H-A> well, speeds too.
[11:15:22] <robin_z> is canada part of the US?
[11:15:29] <A-L-P-H-A> just when you want to buy carpet... it's in sq ft.
[11:15:37] <A-L-P-H-A> real estate is in imperial.
[11:15:50] <A-L-P-H-A> paper is here too.
[11:16:05] <A-L-P-H-A> umm... nuts are bolts are all instant metric units.
[11:16:13] <robin_z> I thought the US built a highway all the way to Alaska or someting??
[11:16:14] <A-L-P-H-A> i.e. 3/4" -> 19mm.
[11:16:16] <A-L-P-H-A> stuff like that.
[11:16:43] <A-L-P-H-A> robin_z, uh... probably, but the section that runs through Canada, is CANADA! :P
[11:16:48] <robin_z> heh,
[11:16:54] <robin_z> only winding you up ;0
[11:16:55] <A-L-P-H-A> if we want to bash...
[11:17:08] <A-L-P-H-A> so how's the poodle tony blair? Bush leading you guys around some more?
[11:17:15] <A-L-P-H-A> :D
[11:17:16] <robin_z> bah
[11:17:20] <robin_z> you want him?
[11:17:25] <robin_z> we'll ship!
[11:17:31] <A-L-P-H-A> uh. no.
[11:17:43] <A-L-P-H-A> At least canada stuck to it's gun, and no bothered with this Iraq war BS.
[11:17:56] <A-L-P-H-A> I think we sent humanitarian aid... but minimal.
[11:18:09] <A-L-P-H-A> we did however participate in the Afghan war.
[11:18:13] <robin_z> well, whatever, I couldnt hear it from my place
[11:18:50] <robin_z> as ever, religion tries iots best to fuck the planet, lets move on before this gets ugly
[11:19:16] <A-L-P-H-A> I didn't start it! :) hehe
[11:20:10] <A-L-P-H-A> sleep time... my eyelids are heavy now.
[11:25:27] <anonimasu> I'll be back later need to go weld a pressure tank for coolant
[11:25:28] <anonimasu> :
[11:25:30] <anonimasu> :)
[11:32:01] <nevyn> but what's cool about metric paper is the 1-> sqrt(2) ness of it
[11:42:08] <robin_z> oh, deskcnc ..
[11:42:23] <robin_z> it appears it can only do arcs if your X and Y axes have the same pitch
[11:42:55] <robin_z> hahaha.
[11:57:38] <nevyn> oh.
[11:57:47] <nevyn> this restriction does not apply to emc?
[12:27:01] <robin_z> no, emc can use any X and Y scales you like .. infact almost ALL cnc progs can
[12:27:09] <robin_z> except ... deskcnc ...
[12:27:17] <robin_z> it seems. that must be VERY basic
[12:27:51] <robin_z> it is extremely UNLIKEY that you would have the exact same scale on X and Y
[12:38:44] <Imperator_> Imperator_ is now known as Imperator_away
[12:45:03] <nevyn> yeah that's what I would think
[12:45:19] <nevyn> even if you used the same motors and threadcount?
[13:10:28] <paul_c> Morning Ray
[13:13:20] <rayh> Hi Paul.
[13:16:16] <rayh> "flexible as Galil's" is an interesting concept.
[13:18:27] <paul_c> well... You can either run the motion code on the PC, or off load it on to a DSP
[13:20:41] <rayh> Yep.
[13:21:10] <rayh> I'm thinking that this is sort of an explorer v settler issue.
[13:21:27] <paul_c> having said that - The Galil cards aren't too expensive.
[13:21:36] <rayh> EMC is much more flexible than any hardware solution.
[13:22:06] <rayh> But you've got to be willing to make it work for you.
[13:36:20] <paul_c> I look forward to reading your reply..
[13:39:06] <rayh> Yea, right!
[13:39:40] <paul_c> Hrmmm... I shouldn't need to flip FAULT_POLARITY with the sherline ini....
[13:40:49] <rayh> What's happening with it.
[13:42:42] <paul_c> got the driver box plugged in & powered up... take EMC out of E-Stop, and the amp fault is triggered.
[13:43:19] <robin_z> rayh: although EMC is flexible, you really need to see what external hardware can do, its just so smooth ... offloading pulse generation to hardware is the way to go, I'm conviced of it. the trick is to keep the hardware simple and cheap, and make it readily available .. and as an ALTERNATIVE to pure PC based pulse gen
[13:44:52] <rayh> Absolutely.
[13:45:08] <robin_z> and .. the new G200X is looking very interesting .. if the price is right
[13:45:25] <robin_z> it has encoder inputs and D->A outputs
[13:46:02] <robin_z> ive yet to see the full spec, but it means real servos might be possible
[13:46:58] <robin_z> several interesting options there, from PID loop on the PC, pid loop on the rabbit ... allsorts
[13:47:53] <rayh> All true, but I was tripping over the term flexible.
[13:48:37] <rayh> If you've got a place where dsp or hardware pulse generation is what you need...
[13:48:59] <rayh> and the hardware you select is able to handle that...
[13:49:09] <rayh> then you are in good shape.
[13:49:23] <robin_z> well, any modern stepper motor and a uStepping drive needs LOTS of pulses
[13:49:43] <robin_z> EMC is ok for old style steppers on half or full step drives
[13:49:43] <rayh> But smart hardware always seemed to me to be more constrained
[13:50:11] <robin_z> but technology has moved on, and at high pulse rates, emc struggles in afraid
[13:50:25] <rayh> That use of terms ruffles just a bit.
[13:50:41] <robin_z> well, im putting it on an old bridgeport and itr will be fine
[13:50:46] <robin_z> perfect infact ...
[13:51:02] <rayh> It's like calling tkemc or mini old by comparison with modern.
[13:51:24] <robin_z> shrug, they work just as well as anything else .. a display is a display
[13:51:32] <robin_z> but steppers HAVE moved on
[13:51:53] <robin_z> theres a LOT of people using 10 uStep geckos ...
[13:52:10] <rayh> And low end machine tools are the ONLY ones using them.
[13:52:41] <robin_z> but you cant drive them to anywahere near their potential with emc ... you end up gearing too low
[13:52:53] <rayh> I guess I've got no problem with dedicated purpose step boards
[13:53:10] <robin_z> * robin_z nods
[13:53:15] <rayh> but if you purchase a step board, that's all you can do with it.
[13:53:22] <robin_z> is it?
[13:53:41] <rayh> You have limited the flexibility which is the term I was trying to explore.
[13:54:02] <robin_z> liek I said .. the G200X is looking *very* interesting ... 6 axes of step/dir ... six encoder inputs
[13:54:08] <robin_z> 4 channels of anaolue IO
[13:54:10] <rayh> I spent a day with Hugh Jack. He is part of the MatPLC group.
[13:54:47] <robin_z> right ...
[13:54:55] <rayh> He uses Linux throughout a six year engineering degree program.
[13:55:08] <robin_z> good choice
[13:55:26] <rayh> Senior projects are a year long application of various softwares to
[13:55:39] <rayh> a small 12 station manufacturing line.
[13:55:42] <robin_z> yep, no arguments so far
[13:56:11] <rayh> Database is the core program
[13:56:17] <robin_z> linux offers that flexibility .. inter application comms via a variety of means
[13:56:46] <rayh> Where you need pick and place, which is a good part of the
[13:56:59] <rayh> description of what juan wants to do.
[13:57:04] <robin_z> right
[13:57:19] <robin_z> even some visual feedback via cameras?
[13:57:26] <rayh> I had not thought of db as a good pick and place driver.
[13:57:46] <robin_z> guess so ... for components on a board
[13:58:17] <rayh> It does, along with some dumb io, or cameras as you were thinking,
[13:58:23] <robin_z> right
[13:58:26] <rayh> make a very flexible system.
[13:58:34] <robin_z> * robin_z nods
[13:58:43] <anonimasu> iab
[13:58:43] <robin_z> thats all high-level stuff
[13:58:55] <robin_z> thats where linux excels
[13:58:58] <rayh> In PLC, one of our issues is upgrading the running software,
[13:59:07] <rayh> while the system is running.
[13:59:13] <anonimasu> hm, does anyone have the ratios for a ejector pump?
[13:59:34] <robin_z> upgrading while running? ... sounds dodgy
[13:59:43] <rayh> With a db that is almost trivial.
[13:59:53] <anonimasu> :)
[14:00:00] <robin_z> well, with suitable locking I guess
[14:00:29] <rayh> My only experience with ejector pumps was a paired solution with one backup.
[14:00:44] <robin_z> I can think of almost no applications where shutting down to upgrade is a problem
[14:00:44] <rayh> The backup pump was installed backwards.
[14:00:54] <anonimasu> ok
[14:00:58] <anonimasu> I built a ejector pump..
[14:01:03] <anonimasu> but I have the ratios a bit wrong..
[14:01:10] <anonimasu> air/fluid
[14:01:18] <anonimasu> so I dont get as much vaccum as I'd like..
[14:01:24] <rayh> When they switched it in, all the sewage came up into the room drain.
[14:01:49] <anonimasu> :)
[14:01:51] <robin_z> rayh: I agree, linux can be very flexible ... im not sure that task-swapping a realtime OS is the most efficient way to generate pulses thats all
[14:01:52] <rayh> Coated several suits who were standing round it.
[14:01:59] <robin_z> nice
[14:02:12] <rayh> No robin I agree completely.
[14:02:20] <robin_z> 'k
[14:02:21] <robin_z> :)
[14:02:47] <rayh> So in those specific places around that 12 station cell,
[14:03:16] <rayh> where high speed pulse train is required, you better grab a
[14:03:24] <rayh> board to do the job.
[14:03:39] <rayh> I don't believe that I'd even look at a rabbit for that.
[14:03:52] <rayh> When there are dedicated, dumber solutions.
[14:04:10] <robin_z> I dont care how dumb or smart it is so long as it does the job
[14:04:13] <rayh> pico, out to 250k
[14:04:37] <rayh> or one of the sine wave stepper systems.
[14:04:49] <anonimasu> yep
[14:05:09] <rayh> But then steppers don't make a whole lot of sense to me in a
[14:05:14] <rayh> critical app.
[14:05:15] <robin_z> pico is a good solution, I cant see haow it differs from a G2002 though
[14:05:29] <robin_z> except one is // port, the other USB
[14:05:47] <anonimasu> I like the idea of a ejector since it has no moving parts..
[14:06:49] <rayh> I guess I'm thinking of the glue tub at Mercury where you have
[14:07:05] <rayh> ten seconds before a major explosion
[14:07:27] <rayh> if the chemistry and temp is not controlled.
[14:08:02] <rayh> We use several PLC's and do voting
[14:08:10] <anonimasu> yep
[14:08:27] <anonimasu> * anonimasu would use several plc's and failover
[14:08:32] <rayh> but it seems to me that the less complicated the solution
[14:08:35] <rayh> the better.
[14:08:55] <robin_z> cheaper, more flexible ...
[14:08:56] <rayh> at each stage in the process.
[14:09:06] <robin_z> simpler means many things to many people
[14:09:15] <rayh> Yes it does.
[14:09:40] <anonimasu> * anonimasu dosent see how it can get easier then with PLC's ;)
[14:09:47] <anonimasu> but I am biased
[14:10:24] <robin_z> what emc/CNC users will want is something simple to buy and hookup
[14:11:08] <robin_z> I dont really think they will care that much HOW it works, so long as it works
[14:11:08] <anonimasu> yep..
[14:11:18] <rayh> There are places where the multitasking os like Linux is dead wrong.
[14:11:18] <robin_z> and they can fiddle with it to suit their needs
[14:11:50] <anonimasu> linux isnt stable, enough to do some things :)
[14:12:10] <rayh> Yep and if the hobby leve guy blows up the basement or shop, he knows where the blame lies.
[14:12:12] <roel> i got a question can emc handle one or more 8255 controllers
[14:12:29] <anonimasu> rayh: I doubt hobby guys do that kind of stuff
[14:12:36] <robin_z> quite
[14:12:36] <rayh> Yes.
[14:13:24] <anonimasu> do that?
[14:13:27] <anonimasu> they?
[14:13:40] <nevyn> ?
[14:13:47] <rayh> roel: What bus is the 5255 connected to?
[14:14:03] <roel> pci
[14:14:29] <paul_c> roel: I have a 3x8255 card here...
[14:14:44] <robin_z> anonimasu: I very much doubt it .. and even if they do, then its irrelevant
[14:14:46] <roel> working ! paul
[14:14:50] <rayh> Okay. There are drivers out there for most of the common 8255 emulation boards like MC.
[14:15:40] <rayh> There is a DIO_Exercise.tcl in tcl/scripts that you can play with once you see the address
[14:16:02] <robin_z> yeah, handy script that
[14:16:09] <rayh> of the card but it may well lock your computer when you try to config port D
[14:16:32] <paul_c> rayh: With a 433MHz cpu, what would you suggest as a minimum for PERIOD ?
[14:16:45] <rayh> I found with the MC pci board that if I configed with the ordinary driver first
[14:16:57] <rayh> and then used the DIO script
[14:17:08] <rayh> It was able to handle the tasks.
[14:17:23] <rayh> But a full page pci solution would be better yet.
[14:18:33] <rayh> paul_c: I seem to remember something about 40 or a little bit larger.
[14:19:09] <rayh> nevyn: ??
[14:19:55] <robin_z> * robin_z is having fun generating rack profiles and gear pinions
[14:20:03] <paul_c> rayh: down to 15�Sec now...
[14:20:13] <robin_z> more complex than you might think
[14:20:28] <roel> somthing else any of us using VME bus
[14:21:17] <rayh> If I remember right, split VME is where EMC started.
[14:21:45] <rayh> Motion on one side Task and IO on the other
[14:22:06] <rayh> * rayh goes for some coffee.
[14:22:23] <robin_z> * robin_z goes off to play with a LASER :)
[14:23:38] <alex_joni> greetings
[14:23:42] <anonimasu> hello alex
[14:23:45] <roel> hi alex
[14:23:54] <alex_joni> hey an0n
[14:24:01] <alex_joni> hello roel
[14:25:33] <paul_c> 15�Sec is producing some "odd" results with jog.
[14:26:22] <alex_joni> YO paul_c
[14:28:57] <paul_c> Hey Ray...
[14:29:19] <paul_c> Getting an indicated 32ipm out of the Sherline h/w
[14:30:01] <anonimasu> hm..
[14:30:14] <rayh> paul_c: With a 433 box?
[14:30:20] <paul_c> yup
[14:30:33] <alex_joni> hey John
[14:30:38] <jmkasunich> hi alex
[14:30:38] <alex_joni> hey ray
[14:30:39] <rayh> What kind of "odd"?
[14:30:44] <anonimasu> hm..
[14:30:46] <rayh> Hi Alex.
[14:30:51] <anonimasu> handwheel.cc 234: !ERROR! can't get emcCommand buffer
[14:31:03] <anonimasu> shouldnt that work while I have emc running?
[14:31:14] <alex_joni> an0n: standard emc.nml ?
[14:31:30] <paul_c> With PERIOD<16�Sec, cont. Jog is somewhat erratic
[14:31:35] <anonimasu> alex_joni: generic.nml
[14:31:54] <alex_joni> an0n: emc1?
[14:31:57] <anonimasu> but I moved it and changed the name to emc.nml
[14:31:57] <rayh> paul_c: Does it loose steps?
[14:31:57] <anonimasu> yes
[14:32:20] <rayh> Morning John.
[14:32:22] <paul_c> rayh: It didn't appear to.
[14:33:19] <alex_joni> an0n: sounds about right ;)
[14:33:32] <alex_joni> shouldn't be able to work :)
[14:33:40] <anonimasu> I dont have a gameport ;)
[14:33:54] <alex_joni> nevermind the gameport
[14:34:00] <anonimasu> ok?
[14:34:17] <alex_joni> handwheel is starting and searching the nml you specified for a handwheel section
[14:34:21] <alex_joni> which.. doesn't exit :D
[14:34:27] <anonimasu> oh ok..
[14:34:54] <alex_joni> try renaming keystick to handwheel (all 5 appearances)
[14:35:08] <anonimasu> in handwheel.cc ?
[14:35:14] <alex_joni> no
[14:35:18] <anonimasu> or in the nml..
[14:35:21] <alex_joni> in emc.nml (former generic.nml)
[14:36:24] <anonimasu> neat.
[14:36:25] <anonimasu> :]
[14:36:33] <alex_joni> does it work?
[14:36:36] <anonimasu> yes..
[14:36:46] <alex_joni> well.. at least you should have gotten rid of that problem
[14:36:55] <alex_joni> the gameport would be required ;)
[14:37:01] <anonimasu> oh, I removed that part.
[14:37:05] <alex_joni> cool
[14:37:21] <alex_joni> so it works...
[14:37:28] <jmkasunich> morning Ray
[14:38:51] <anonimasu> I am going to write some routines to get data off a plc later tonight
[14:40:13] <roel> have to reboot Back in a wimp
[14:41:56] <nevyn> is there anywhere I can get the sim files? sim.run sim.ini etc?
[14:42:05] <alex_joni> should be in CVS
[14:42:08] <anonimasu> nevyn: cvs
[14:42:39] <nevyn> does emc2 work with sims?
[14:42:51] <jmkasunich> nope
[14:42:56] <alex_joni> nevyn: not really
[14:43:08] <alex_joni> you can work with emc2 without motors, but you still need RT
[14:43:12] <jmkasunich> "sims" are for systems with no realtime OS
[14:43:22] <nevyn> right.
[14:43:35] <paul_c> rayh: Running 16.5�Sec on a 433MHz box - Jogging @ 30ipm with 110% override. No lost steps.
[14:43:45] <nevyn> * nevyn thinks I should do stuff due tomorrow not compile a kernel tonight ;)
[14:44:12] <anonimasu> I dont think the problem is about lost steps..
[14:44:21] <anonimasu> it's polling the ui..
[14:44:21] <alex_joni> * alex_joni sets to compile a 2.6 right now
[14:44:29] <anonimasu> or somthing like it
[14:44:50] <alex_joni> ui is not RT.. so it will get slow
[14:44:55] <nevyn> so steppers or servos? which are the bomb?
[14:45:24] <anonimasu> I think I need another box for running the UI on.. it feels so slow :)
[14:45:44] <anonimasu> I love running it on my windows box when playing around :)
[14:46:16] <jmkasunich> nevyn: steppers vs. servos depends on your situation
[14:46:34] <alex_joni> the bomb: AC drives with resolvers
[14:46:46] <jmkasunich> size of machine, desired performance, available money
[14:46:51] <alex_joni> will hit your pockets like hiroshima
[14:47:10] <anonimasu> :)
[14:47:15] <anonimasu> haha
[14:47:21] <nevyn> heh
[14:48:10] <nevyn> jmkasunich: I have a friend who converted a small mill from steppers to servo's to get more linear motion. does this actually work?
[14:48:39] <nevyn> 4"x10" table 6" z
[14:48:40] <jmkasunich> again, it depends. no simple answers
[14:48:51] <nevyn> sigh. I like simple answers.
[14:48:59] <jmkasunich> bad servos can be worse than good steppers
[14:49:00] <alex_joni> nevyn: you might get more speed from servo's
[14:49:09] <anonimasu> and acceleration..
[14:49:14] <alex_joni> err.. from good servos
[14:49:55] <jmkasunich> right - good servos are usually better than good steppers (unless the machine is very small, where the additional performance that servos _might_ provide isn't really usable anyway)
[14:50:42] <anonimasu> hm, how would I change the feedrate in nml
[14:50:49] <anonimasu> are there any list of possible parameters?
[14:51:39] <jmkasunich> cradek: you here, or just logging?
[14:52:10] <alex_joni> an0n: feedrate in nml?
[14:52:13] <anonimasu> yeah
[14:52:16] <anonimasu> changing it..
[14:52:21] <alex_joni> how do you mean that?
[14:52:35] <anonimasu> emcCommandBuffer->write (emc_axis_incr_jog_msg);
[14:52:50] <anonimasu> feedrate override that is
[14:52:51] <anonimasu> I mean
[14:53:58] <alex_joni> look at feedslider.cc
[14:54:02] <alex_joni> does exactly that
[14:54:13] <alex_joni> src/emctask/feedslider.cc
[14:54:13] <anonimasu> ah.
[14:54:14] <anonimasu> :D
[14:54:37] <alex_joni> emc_cmd_buffer->write(feed_override_msg);
[14:54:56] <jmkasunich> jepler: you here or just logging?
[14:57:00] <anonimasu> yeah it works nicely
[14:59:09] <anonimasu> that's nice to look at
[15:02:03] <alex_joni> an0n: you're getting good at rcs-stuff ;)
[15:02:11] <anonimasu> alex_joni: lol no
[15:02:16] <anonimasu> alex_joni: I dont understand anything yet
[15:02:23] <anonimasu> alex_joni: but well, maybe someday
[15:04:19] <jmkasunich> The Programmer's Fest is getting closer, and only five people are "officially" registered... at one time we had nearly a dozen folks who said they were interested (mostly here on IRC)
[15:04:42] <anonimasu> I'd come if there was less water in the way
[15:04:51] <jmkasunich> where are you?
[15:05:00] <alex_joni> I'd come if I had a bigger boat ;)
[15:05:03] <anonimasu> sweden..
[15:05:06] <alex_joni> ahhh.. and a faster one too
[15:05:07] <jmkasunich> wide amounts of water are a problem...
[15:05:18] <jmkasunich> alex_joni: you have a sailboat?
[15:05:25] <alex_joni> rowboat
[15:05:30] <jmkasunich> :-)
[15:05:32] <alex_joni> 3-seater
[15:05:37] <alex_joni> would take a long time
[15:05:58] <jmkasunich> room for you, anonimasu, and Imperator
[15:06:13] <alex_joni> yeah.. Imperator bugged me to go...
[15:06:38] <anonimasu> :D
[15:06:48] <jmkasunich> somebody on that side of the pond should organize an EU-EMC-Fest
[15:07:06] <paul_c> and you would come of course.
[15:07:26] <alex_joni> I could optimise the track (take the Danube to Germany, pickup Martin on the way, switch to the Rhein, then up to the East-Sea.. get an0n somehow on the way)
[15:07:29] <jmkasunich> probably not, for the same reason that Imperator/Alex/anon can't come here... $$$
[15:07:44] <alex_joni> * alex_joni also needs a visum ;)
[15:08:03] <anonimasu> getting to germany isnt that hard..
[15:08:04] <anonimasu> :)
[15:08:18] <anonimasu> I think you can go to denmark and take a train..
[15:08:19] <paul_c> US Embassy takes up to 6 months to issue visas
[15:08:38] <jmkasunich> just to visit? wow... I had no idea
[15:08:42] <alex_joni> it's less down here (probably 1-2 weeks)
[15:08:57] <alex_joni> but it's usually a 10-years visum
[15:09:17] <paul_c> Ah... You mean passport ?
[15:09:21] <rayh> * rayh would be up for a eurofest.
[15:09:22] <anonimasu> hillarious..
[15:09:24] <alex_joni> nope... I mean US visum
[15:09:26] <anonimasu> feedslider works..
[15:09:35] <anonimasu> now I just got to get stuff from somthing other then the keyb ;)
[15:09:52] <alex_joni> right... use the mouse an0n
[15:10:02] <anonimasu> nah
[15:10:07] <anonimasu> a rotary encoder
[15:10:30] <alex_joni> the mouse (at least the old ones) has two of those
[15:10:37] <anonimasu> I'll order ones at monday
[15:10:49] <anonimasu> they are like 7$
[15:10:53] <anonimasu> :)
[15:11:23] <anonimasu> I got a order ready for more things at work anyway
[15:11:47] <jmkasunich> alex_joni: did you mean that you can get a visa that lets you stay in the US for 10 years? in only 1-2 weeks?
[15:12:26] <alex_joni> no.. I meant you are allowed to visit US for 10 years (but not at a time)
[15:12:41] <alex_joni> you probably can stay upt o a few months at once (don't remember how many)
[15:13:03] <jmkasunich> I see.. handy if you have business that takes you there frequently - don't need a new visa every time
[15:13:22] <jmkasunich> * jmkasunich knows nothing about international travel
[15:13:37] <alex_joni> * alex_joni would think jmk is from the US
[15:13:40] <jmkasunich> I've been to Canada several times and Mexico once
[15:13:55] <jmkasunich> alex_joni would be correct
[15:14:03] <alex_joni> it was a joke ;)
[15:14:12] <paul_c> jmkasunich: You sticking around for long ?
[15:14:25] <jmkasunich> on and off all day I think
[15:14:42] <jmkasunich> here for at least the next hour or so, until my wife gets home from church
[15:15:02] <paul_c> OK.. I need a coffee, then I'll talk about codeFest trips.
[15:15:20] <alex_joni> how about tee paul?
[15:15:30] <alex_joni> switched to bad habits?
[15:21:44] <alex_joni> * alex_joni will be right back, booting a brand new 2.6
[15:34:00] <paul_c> hrmmm... GUI is certainly sluggish @ 16.5�Sec
[15:35:30] <jmkasunich> that's the nature of GUIs
[15:36:24] <alex_joni> darn.. gotta rebuild that kernel
[15:36:46] <jmkasunich> what happened?
[15:36:57] <alex_joni> the tty's hang
[15:37:09] <alex_joni> ssh works though :-?
[15:38:33] <alex_joni> think it's a modprobe thing (I get a lot of char_major_21, 21_0 - 21_15 not found)
[15:40:59] <paul_c> alex_joni: Did you enable /dev/pts in your config ?
[15:41:16] <alex_joni> dunno ;)
[15:41:20] <alex_joni> gotta check
[15:41:52] <paul_c> and what kernel ver. .config did you use for a baseline ?
[15:42:06] <alex_joni> 2.6.10 and it's config (no rtai patch yet)
[15:43:37] <alex_joni> paul_c: didn't use /dev/pts.. should I?
[15:43:57] <paul_c> incoming .config that I used.
[15:44:18] <alex_joni> thx.. but I still need to patch it with adeos (was trying to get 2.6 running first)
[15:44:29] <paul_c> Which version of RTAI do you plan to use ?
[15:44:42] <alex_joni> 3.1 ?
[15:45:26] <paul_c> 3.1 (a.k.a Vesuvio) won't compile...
[15:45:40] <paul_c> not on a 2.6.10 kernel at least...
[15:45:50] <alex_joni> any ideea what I should try?
[15:46:24] <paul_c> vesuvio and 2.6.9 compiles & works.
[15:48:35] <alex_joni> any rtai working on 2.6.10 ?
[15:49:21] <CIA-4> 03paul_c 07bdi-4 * 10emc2/src/ (11 files in 3 dirs): Needed to make a few changes with those __attribute_used__ defines to get the code to compile on a 2.6.10 kernel.... Darn those usr space includes - They are a pain in the...
[15:49:23] <nevyn> does emc work with fusion?
[15:49:26] <paul_c> fusion should compile
[15:49:48] <alex_joni> rtai-3.2test2 ?
[15:50:01] <paul_c> BUT... Fusion is very experimental and currently under heavy development.
[15:51:57] <paul_c> alex_joni: Use CVS - You'll get all the latest bug fixes that didn't make it in to the tarballs.
[15:52:28] <paul_c> * paul_c hides
[15:52:28] <nevyn> * nevyn waits for very slow cvs checkout..
[15:52:36] <dave-e> too late
[15:53:05] <alex_joni> nevyn.. you could get only the sims using the web-interface from sourceforge
[15:53:09] <alex_joni> if that's all you need
[15:54:02] <nevyn> alex_joni: I did a partial cvs checkout and got errors about emcsimot being missing.. shrug I'll build it in my homedir and have a play
[15:54:27] <alex_joni> right
[15:54:35] <nevyn> I copied sim* from the partial cvs tree to my binary tree but it didn't work so I'll just build it.
[15:55:17] <paul_c> On a 2.6 kernel ?
[15:55:57] <alex_joni> * alex_joni was talking about sim.ini, nml, run, var, tbl
[15:57:43] <anonimasu> iab
[15:58:19] <anonimasu> hm, I'd rather have a console gui if it'll lag less ;)
[15:58:23] <alex_joni> * alex_joni is checking out fusion from CVS
[15:58:38] <anonimasu> * anonimasu has never seen/heard about fusion
[15:58:49] <alex_joni> an0n: you can adjust your PERIOD.. but that'll make the machine run slower
[15:59:04] <anonimasu> alex_joni: oh it's still sluggish ;)
[16:00:07] <alex_joni> well... run it remote
[16:00:13] <alex_joni> the GUI I mean
[16:00:17] <anonimasu> yeah that's probably the only good solution
[16:00:32] <alex_joni> or use a better machine ;)
[16:00:44] <alex_joni> or oc it
[16:00:56] <anonimasu> I have a celeron 300a somwhere..
[16:01:24] <anonimasu> it'll boot at 800mhz
[16:01:25] <anonimasu> :)
[16:02:00] <jmkasunich> hey paul... you were gonna talk about Fest
[16:02:52] <paul_c> Yes - Put my name down.
[16:04:05] <jmkasunich> ok
[16:05:36] <paul_c> It will give the MIB's to do their checks.
[16:05:42] <rayh> Paul gets a better rate if he flies in midweek.
[16:05:46] <paul_c> ^time
[16:06:13] <rayh> We were talking about getting together the weekend before we get to NIST.
[16:07:00] <paul_c> Did you speak to Matt about that at all ?
[16:07:14] <rayh> I should have Matt's Hardinge ready for threading by then.
[16:07:18] <rayh> Yes.
[16:07:56] <jmkasunich> I'm planning on driving in Sunday night
[16:08:03] <jmkasunich> when/where is the get-together
[16:08:22] <rayh> Not a good clue yet.
[16:08:41] <rayh> Perhaps Matt's house
[16:10:02] <alex_joni> rayh: any new thoughts on the config stuff ?
[16:10:28] <rayh> I'll try to get all of the before NIST details worked out this week.
[16:10:47] <rayh> SMP put some qconfig stuff in emc2
[16:11:00] <alex_joni> SWP ;)
[16:11:06] <rayh> It seems to work well and uses a very simple sort of text file.
[16:11:09] <alex_joni> simmetric wulti processing ;)
[16:11:19] <rayh> Yea. That's him.
[16:11:31] <nevyn> symetric ?
[16:11:40] <rayh> Fingers are lagging badly here.
[16:11:46] <alex_joni> nevyn: right...
[16:11:51] <jmkasunich> webpage updated with Paul's name... I'm still short several folks who indicated that they were interested
[16:11:53] <alex_joni> rayh: ditto
[16:12:10] <rayh> The gui for qconfig allows for rtf and some html in the help section.
[16:12:46] <rayh> I'm thinking that if we use a config directory, we can put visuals
[16:13:06] <rayh> in there. Someone mentioned the gear train as a possible visual
[16:13:30] <alex_joni> what's that?
[16:13:35] <rayh> leadscrew pitch, belt or gear ratio, stepper or encoder ppr
[16:13:49] <alex_joni> ahhh.. yes, I remember that
[16:14:02] <paul_c> OK... Got a number of arrival times to choose from... 14:40, 18:15, or 22:35 for Dulles, or just one @ 16:00 for Baltmore.
[16:14:03] <rayh> Then I'm thinking of a graphic for each hal module with named pins
[16:14:04] <alex_joni> btw... just a thought .. I think the IRC logs need to get automated
[16:14:13] <rayh> something like real IC's
[16:14:19] <alex_joni> don't think anyone will be able to maintain those
[16:14:52] <rayh> Baltimore is closer, I think, and not nearly so busy.
[16:15:43] <paul_c> Went through distances with Steve last week - IAD is closer to NIST by about 15 miles.
[16:17:42] <rayh> If you want to look over the qconfig stuff you will need 4.14 or better
[16:17:50] <rayh> and add in all the compile stuff including qt3.
[16:18:05] <rayh> Or I can post images.
[16:18:33] <alex_joni> * alex_joni tries to get his 2.6 working
[16:18:45] <SteveStallings> Baltimore is closer as the crow flies, but no good connecting roads.
[16:19:16] <jmkasunich> alex_joni: your
[16:19:18] <jmkasunich> oops
[16:19:35] <jmkasunich> you're experiencing IRC log burnout? ;-)
[16:20:20] <alex_joni> I did a while ago
[16:20:28] <alex_joni> just checked.. last log is december 2004 .(
[16:20:43] <jmkasunich> SteveStallings: is it possible to see how often (if at all) the IRC logs on linuxcnc.org are accessed... if nobody actually reads them, then we're just wasting time and energy putting them up there
[16:20:45] <alex_joni> and we're almost done with 1/4 of the year
[16:21:28] <alex_joni> jmk: I used to read those.. before the wiki it was one of the valuable sources of information
[16:21:34] <alex_joni> not sure how many stil use it
[16:21:49] <rayh> somebody asked a while back about why the logs were not up to date.
[16:21:59] <jmkasunich> the information density is pretty low..
[16:22:17] <jmkasunich> cause it's a drag doing them (as both Alex and I can testify)
[16:22:28] <SteveStallings> ....looking........
[16:22:44] <alex_joni> It has been used in applications with up to 13 controlled motors, with a sampling period of the position loop as frequent as every 100 microseconds, featuring less than 10 micrometers of maximum following error at speeds of up to 36 m/min and with accelerations up to 7g.
[16:22:57] <alex_joni> right.....
[16:23:08] <jmkasunich> huh?
[16:23:22] <jmkasunich> they talking about emc?
[16:23:33] <alex_joni> nah.. MaVcnc
[16:23:42] <jmkasunich> perhaps it has indeed been used for all those things, but not likely all at the same time
[16:23:54] <alex_joni> lol.. right
[16:25:05] <jmkasunich> hmmm... at 36m/min, that is 60micrometers in 100uS... so they can't claim 10umeter following error...
[16:25:53] <jmkasunich> show me a salesman, I'll show you a liar
[16:26:58] <alex_joni> yay..2.6 started
[16:27:07] <alex_joni> a lot of errors .. but it's useable
[16:27:47] <paul_c> what sort of errors ?
[16:28:07] <SteveStallings> IRC chat logs had about 600 hits in January.
[16:28:28] <alex_joni> paul_c: insmod errors complaining about missing modules
[16:28:40] <alex_joni> I gotta check if I updated modutils...
[16:28:46] <paul_c> USB modules ?
[16:30:06] <paul_c> and you want module-init-tools, not modutils installed.
[16:30:14] <alex_joni> right
[16:30:41] <alex_joni> example: modprobe: FATAL: Module char-major_4_72 not found.
[16:31:36] <paul_c> Have you added a mount line for sysfs to fstab ?
[16:31:53] <alex_joni> nope :)
[16:32:12] <paul_c> none /sys sysfs defaults 0 0
[16:32:34] <paul_c> (the typical stanza for sysfs)
[16:33:03] <alex_joni> thx
[16:37:47] <rayh> Any thoughts on configuration before I have to run and make lunch for the fam.
[16:38:27] <paul_c> A little concerned about dependencies on the Qt dev libs
[16:38:36] <alex_joni> no new thoughts here...
[16:38:54] <jmkasunich> second pauls comments.... I don't like dependencies in general
[16:38:56] <paul_c> Will not have space on the BDI-4 disk for them...
[16:39:26] <alex_joni> how about keeping it text mode?
[16:39:35] <jmkasunich> I'm all for htat
[16:39:37] <jmkasunich> that
[16:39:38] <alex_joni> make menuconfig ?
[16:39:46] <rayh> Right. We could write a similar looking sort of thing with tickle.
[16:39:57] <rayh> Text mode does not meet the aunt tillie at all.
[16:40:14] <jmkasunich> what, she can't read?
[16:40:17] <rayh> With the kernel configuration stuff there are three possibles.
[16:40:42] <rayh> text curses and qt
[16:40:47] <nevyn> tickle ungood. really.. 1984 called and want's it's widgets back.
[16:41:04] <nevyn> bleh my grammar bad.
[16:41:19] <rayh> once qconfig is built it can run without the dev stuff.
[16:41:25] <paul_c> The ncurses version would be the only one to compile from an "out of the box" BDI install.
[16:41:52] <rayh> nevyn: Haven't looked at some of the widgets lately?
[16:42:18] <rayh> * rayh baits nevyn...
[16:42:40] <jmkasunich> I'm a little disturbed by the fact that BDI-4 can't compile the head of CVS... anything that continues the path of "you must use what is already compiled" isn't good IMO
[16:43:15] <paul_c> All the BDI-4 sources are in CVS
[16:43:24] <rayh> jmk baits paul_c
[16:43:39] <nevyn> is emc1 autoconfified? or only emc2?
[16:43:42] <jmkasunich> but can someone who just installed BDI-4 actually compile them?
[16:43:57] <paul_c> I can do a merge of the BDI-4 branch and HEAD...
[16:44:04] <nevyn> rayh: look tcl/tk was good for it's day but it's honestly had it's day.
[16:44:05] <rayh> I've done it here with the help of the wiki page.
[16:44:11] <paul_c> * paul_c throws the bait back
[16:44:34] <jmkasunich> not talking about head vs. branch... talking about lack of packages/dependencies
[16:45:10] <alex_joni> * alex_joni wonders about 2.6
[16:45:25] <alex_joni> how would Makefiles look for both 2.4 and 2.6 ?
[16:45:53] <rayh> nevyn: And what package would you use instead?
[16:46:06] <paul_c> All the basic tools required to compile from CVS are on the latest CD - Had to move things around and delete some bits to do it.
[16:46:24] <jmkasunich> ok, that is news to me... good news
[16:46:31] <nevyn> rayh: qt, wx, gtk any of the above.
[16:46:36] <SteveStallings> time for the BDI-DVD ?
[16:46:42] <jmkasunich> * jmkasunich likes GTK
[16:46:57] <paul_c> Only if PMDX want's to host a DVD
[16:47:03] <rayh> paul and alex just complained about qt bloat
[16:47:08] <rayh> I don't write c.
[16:47:39] <jmkasunich> not to be nasty, but few of the rest of us write tcl
[16:47:43] <rayh> c & j announced that they are not pushing forward the dev of axis
[16:47:58] <jmkasunich> why not?
[16:48:22] <rayh> jmk baits rayh
[16:48:33] <rayh> This could get to be an interesting morning after all.
[16:49:07] <jmkasunich> not baiting...
[16:49:56] <nevyn> rayh: indeed why not? it looked really good from the screenshots I've seen
[16:50:18] <jmkasunich> C and C++ are "mainstream" as are Python and such... I never heard of tck before EMC, I know nobody except Ray who programs in Tcl... (granted, I don't get out much ;-)
[16:50:28] <rayh> We could drop the tickle dependency from the package but we need a working replacement.
[16:50:32] <alex_joni> I know a few people who do tcl/tk stuff
[16:50:52] <SteveStallings> Their web pages says it works well enough for "amateurs" and they are happy with it. Further development left to others. It is in CVS on their site.
[16:50:54] <alex_joni> I have some friends working @ Alcatel
[16:50:59] <nevyn> rayh: tcl/gtk exists I think
[16:51:03] <jmkasunich> not suggesting dropping Tcl, at least not soon
[16:51:08] <alex_joni> right
[16:51:22] <alex_joni> paul_c: I could host a DVD ;)
[16:51:33] <alex_joni> but I'm not sure how many would dl/burn it
[16:51:41] <alex_joni> maybe in a year or so ;)
[16:51:56] <jmkasunich> true... I can't burn (or even read) DVDs on any of my boxes
[16:52:10] <paul_c> same here.
[16:52:17] <jmkasunich> I tend to have trailing edge technology, but still.....
[16:52:32] <rayh> I a whole lot rather see a user disk and a developer add on disk.
[16:52:52] <rayh> Than abandwidth sucking dvd.
[16:52:57] <alex_joni> right
[16:52:59] <jmkasunich> agreed
[16:53:08] <alex_joni> let's forget DVD for a while
[16:53:12] <alex_joni> how about 2.6 ?
[16:53:24] <alex_joni> a lot of people are switching to the 2.6 train
[16:53:26] <SteveStallings> come on Ray, get a bigger straw... 8-)
[16:53:36] <alex_joni> and neither emc is working on 2.6
[16:53:40] <alex_joni> emc1 isn't
[16:53:42] <nevyn> point is not so much the language sucks (I mean it does...) but the toolkit lacks modern functionality
[16:53:42] <nevyn> does tk even use fontconfig etc?
[16:53:45] <alex_joni> emc2 isn't
[16:53:50] <alex_joni> bdi-4 does, but only with rtai
[16:54:02] <jmkasunich> when you are d/l-ing ISOs and DVDs, you need a firehose, not a straw (of any size)
[16:54:22] <paul_c> alex_joni: 2.6 support for RTLinux is at a price.
[16:54:30] <alex_joni> paul_c: right
[16:54:43] <alex_joni> but 2.6 support for emc shouldn't be
[16:54:49] <paul_c> See no point in lining Yodakin's pockets.
[16:55:06] <alex_joni> I agree on that... but HEAD should work on 2.6
[16:55:27] <SteveStallings> there is no RTLinux free for 2.6 ?
[16:55:30] <jmkasunich> the way I see it EMC2 should be able to compile on 2.2/RTL, 2.2/RTAI, 2.4/RTL, 2.4/RTAI, and 2.6/RTAI
[16:55:33] <paul_c> head of emc910 or emc2 ?
[16:55:46] <alex_joni> emc911 ;)
[16:55:51] <paul_c> s\910/(1)
[16:55:56] <alex_joni> any HEAD
[16:56:02] <alex_joni> both
[16:56:11] <paul_c> No.
[16:56:14] <alex_joni> no?
[16:56:40] <paul_c> To get rcslib/emc trees to compile with 2.6 is a dead end path.
[16:57:06] <jmkasunich> focus effort on emc2
[16:57:29] <paul_c> There comes a point when it makes sense to concentrate on a single source tree.
[16:57:50] <alex_joni> I agree that rcslib would be a pita
[16:58:07] <alex_joni> right .. that leads the discussion to interesting paths
[16:58:11] <alex_joni> emc2
[16:58:27] <alex_joni> libnml, emc2, hal, bdi-4
[16:58:35] <paul_c> It was agreed last year that emc1 would receive bug fixes, but any further development would take place in the emc2 tree.
[16:58:46] <alex_joni> not necessarely in that order
[16:58:51] <alex_joni> I agree
[16:59:18] <alex_joni> lets take libnml
[16:59:23] <paul_c> So, leave rcslib/emc1 with 2.4 and earlier kernels
[16:59:36] <alex_joni> libnml compiles in 2.6 (right?), bdi-4 branch
[16:59:37] <paul_c> and move 2.6 dev to the emc2 tree
[16:59:53] <paul_c> libnml is kernel independant
[17:00:24] <jmkasunich> RTAPI was intended to hide RTOS dependencies... kernel dependencies should be hidden there too where practical
[17:01:07] <jmkasunich> s/dependencies/differences/
[17:01:22] <rayh> http://www.linuxcnc.org/dropbox/emc-qconfig.png
[17:01:43] <alex_joni> paul_c: regarding includes problems you have (on the bdi-4), wouldn't a configure.h be usefull?
[17:01:52] <paul_c> no
[17:02:42] <jmkasunich> rayh: your sample image proves my point!
[17:02:58] <jmkasunich> the format of the information is less important than the content
[17:03:18] <jmkasunich> "I have no idea how bridgeport differs from minimill task controller"
[17:03:41] <jmkasunich> _that_ is what needs fixed if you're gonna help Aunt Tilly
[17:03:45] <alex_joni> lol
[17:03:51] <alex_joni> that's only temporary
[17:04:24] <jmkasunich> I understand.... but it's there because those differences _really_ are poorly understood by many folks
[17:04:36] <jmkasunich> including me
[17:05:22] <alex_joni> there shouldn't be a difference
[17:05:51] <jmkasunich> if that's true, then there shouldn't be "bridgeport" and "minimill", there should be only "mill"
[17:06:16] <jmkasunich> * jmkasunich thinks there _should_ be only "mill"
[17:06:34] <alex_joni> right
[17:06:47] <alex_joni> rayh: what was the branch where SWP did his magic?
[17:07:13] <rayh> Your's I think
[17:07:26] <alex_joni> one of them ;)
[17:07:53] <alex_joni> * alex_joni thinks it's auto_configure_0_2
[17:08:00] <alex_joni> right
[17:08:04] <paul_c> beat me to it...
[17:08:20] <alex_joni> I never used that branch.. made it in error
[17:08:43] <rayh> my tag says Tbdi4
[17:08:53] <alex_joni> I did make autoconf_install_0_1 though.. and did some work there.. but no-one wanted to see it :(
[17:09:43] <rayh> oops Tbdi-4
[17:09:44] <paul_c> you need to keep on committing and attracting attention to it...
[17:10:10] <alex_joni> well.. it was a while ago
[17:10:16] <alex_joni> the discussion was about make install
[17:10:22] <alex_joni> if anyone remembers that ;)
[17:11:01] <alex_joni> but it seems that it's not an actual target that's needed
[17:11:04] <rayh> Now that I'm compiling with 4.6 I worry about overwriting a good install
[17:11:24] <rayh> I think that was a part of that discussion.
[17:11:31] <paul_c> We will use an install target at some point - That much is enevitable.
[17:11:43] <jmkasunich> "at some point" being the key words
[17:12:00] <rayh> * rayh must whip up some soup for the fam.
[17:12:12] <rayh> Nice talking with you all.
[17:12:14] <jmkasunich> gonna be lunch time here soon too
[17:12:16] <jmkasunich> bye ray
[17:12:20] <alex_joni> bye ray
[17:12:23] <paul_c> mmm... food
[17:12:33] <alex_joni> * alex_joni had food a while ago..
[17:12:43] <alex_joni> I could repeat that experience soon
[17:12:50] <jmkasunich> ;-)
[17:13:12] <alex_joni> paul_c: when that "some point" will come.. check out autoconf_install_0_1
[17:13:22] <paul_c> Introducing a new installed layout requires one of two approaches...
[17:13:23] <alex_joni> probably a lot won't be up to date by then ..
[17:13:57] <paul_c> Either a clean break from the old way, and dump it on everyone in a single shot, or...
[17:14:20] <paul_c> a gradual change, a few files at a time.
[17:14:36] <jmkasunich> I usually prefer the former
[17:14:58] <jmkasunich> the latter often involves some ugly compromizes needed to keep things working
[17:16:59] <paul_c> If you use the BDI-4 install, you will find things are slowly moving towards an "install" target.
[17:17:25] <paul_c> for example - Kernel modules are in /lib/modules/ tree
[17:17:41] <paul_c> and the ngc samples are in ~/gcode
[17:19:27] <alex_joni> right
[17:26:20] <paul_c> Back in a bit - Time to eat.
[17:38:23] <jmkasunich> time for lunch
[17:38:30] <jmkasunich> jmkasunich is now known as jmk-away
[18:02:30] <CIA-4> 03alex_joni * 10emc2/ (configure configure.in): fixed a bug that prevented rtai-24.1.x to be used. rtai-24.1.x usually gets installed in /usr/realtime (by running make install_rt_devel), if that wasn't the case usually configure would have failed. this is fixed now.
[18:19:18] <Jymmm> Isn't this 'useful' content and navigation --> http://www.cadem.com/cademdoor/cnc-plan.htm
[18:21:37] <alex_joni> Jymmm: lol
[18:21:47] <Jymmm> :)
[18:22:52] <Jymmm> * Jymmm is STILL looking for cnc router plans... need more ideas
[18:24:10] <anonimasu> hello
[18:24:18] <Jymmm> Morning anonimasu
[18:24:33] <anonimasu> heh
[18:24:34] <anonimasu> it's 19:28
[18:24:34] <anonimasu> :D
[18:24:42] <Jymmm> [10:28:59] <anonimasu> it's 19:28
[18:24:55] <anonimasu> I am having dinner right now
[18:25:17] <anonimasu> :)
[18:25:32] <Jymmm> anonimasu: well you going to share with the rest of us?
[18:25:51] <anonimasu> no way it's my fish soup
[18:26:01] <Jymmm> yuck!
[18:26:02] <anonimasu> and it is delicious
[18:27:25] <Jymmm> I'm sorry, but this is NOT something I would design... The powersupply is on the x axis.
[18:27:32] <Jymmm> http://www.torchmate.com/box.htm
[18:28:16] <anonimasu> * anonimasu ponders cranking up the rapids a bit more
[18:31:58] <anonimasu> looking at it now
[18:32:32] <anonimasu> yeah
[18:33:45] <Jymmm> I can see the laser itself on the gantry, less mirrors. But not the PS.
[18:34:12] <anonimasu> agreed
[18:35:02] <Jymmm> Is it just me, or does it seem that the way the laser itself is mounted drmatically reduces the Y axis travel in that pic?
[18:35:52] <anonimasu> yeah that's true
[18:36:03] <anonimasu> but it might be on top
[18:36:09] <Imperator_away> anonimasu: about a EU-EMC-Fest: I suggested to make a meeting here in Germany a few month ago
[18:36:35] <alex_joni> hello Martin
[18:36:36] <anonimasu> oh nice
[18:36:41] <alex_joni> change your nick ;)
[18:36:43] <Jymmm> anonimasu: Look at the bracket on the gantry, then follow it to the front on tha machine next to the PS.
[18:36:56] <Imperator_away> Hi Alex
[18:36:58] <Jymmm> s/on/of/
[18:37:32] <anonimasu> I wpuyeah
[18:37:32] <anonimasu> that's true
[18:38:05] <Imperator_away> have started a new project with a other german guy: EMC servo card, based mainly on on FPGA and a DAC chip
[18:38:12] <alex_joni> coo
[18:38:22] <Imperator_away> a ISA card
[18:38:32] <alex_joni> ISA?
[18:38:36] <Jymmm> ISA?
[18:38:39] <anonimasu> I'll have this working in a couple of days
[18:38:50] <anonimasu> plc > a nml wrapper
[18:38:51] <Imperator_away> Imperator_away is now known as Imperator_
[18:39:01] <anonimasu> ?
[18:39:07] <Jymmm> * Jymmm MIGHT have an old machine around here with an ISA slot, but seriously doubts it
[18:39:13] <anonimasu> Imperator_: why dont you build pci card?
[18:39:32] <Imperator_> have a PCI prototype board here
[18:39:39] <Jymmm> oh wait... I do have one it's a P120 64MB 3GB
[18:39:50] <Imperator_> but we want something that can be builded quickly
[18:39:56] <CIA-4> 03alex_joni 07auto_configure_0_2 * 10emc2/Kconfig: added some more options to Kconfig
[18:40:08] <anonimasu> alex_joni: and automatic feed override adjustment..
[18:40:08] <anonimasu> alex_joni: but that comes a bit later :)
[18:40:12] <alex_joni> * alex_joni goes to grab a byte
[18:40:19] <anonimasu> alex_joni: did you see the page I pasted about it
[18:40:19] <anonimasu> ?
[18:40:21] <alex_joni> an0n: nice to hear that
[18:40:24] <alex_joni> nope
[18:41:15] <Imperator_> anonimasu: do you have experience with PCI designs ?
[18:41:26] <anonimasu> wait a sec :)
[18:41:26] <anonimasu> I'll find it
[18:42:49] <Jymmm> Imperator_: any reason not par/ser/usb ?
[18:43:06] <anonimasu> Imperator_: no
[18:43:06] <anonimasu> I wish I had
[18:43:06] <anonimasu> http://www.mmsonline.com/articles/079802.html
[18:43:12] <Imperator_> par/ser is no option, i think
[18:43:30] <Imperator_> and USB <-> Realtime ??? I don't know
[18:43:48] <Jymmm> yeah USB RT bad thing, as USB polls devices in the USB bus
[18:44:01] <Jymmm> FireWire =)
[18:44:17] <Jymmm> 800Mb/s
[18:44:42] <Imperator_> I think ISA is ok for the first step
[18:44:59] <alex_joni> Jymmm: USB goes high on transfer rates too
[18:45:07] <alex_joni> but like Firewire it sucks for packets
[18:45:13] <alex_joni> for small packets
[18:45:16] <Jymmm> Imperator_: Personally I would always keep in mind it'll be PCI eventually.
[18:45:16] <Imperator_> we want to give the design and everything free, so then somebody can implement the PCI stuff
[18:45:20] <alex_joni> on a high rate
[18:45:31] <alex_joni> Imperator_: cool ;)
[18:45:36] <Jymmm> alex_joni Ah ok, didnt know that. Only use it for video
[18:45:42] <alex_joni> bbl
[18:46:12] <Imperator_> at the moment I don't know so much about FPGAs but the other one how helps :-)
[18:46:31] <Imperator_> there are some free PCI cors on the web
[18:46:50] <Jymmm> Imperator_: Just think about it this way... you are 500% ahead of me =)
[18:47:00] <anonimasu> alex_joni: might be overkill though..
[18:47:00] <anonimasu> ;)
[18:47:02] <Imperator_> :-)
[18:48:01] <Jymmm> isn't there a 16bit half on a pci slot?
[18:48:11] <Jymmm> * Jymmm can't recall
[18:48:31] <Imperator_> half of what ???
[18:48:38] <Jymmm> ISA
[18:48:41] <Imperator_> nope
[18:48:46] <Jymmm> oh, ok
[18:49:16] <Jymmm> DO you have any idea how long it's been since I've even seen a signal map of a pci/isa slot?!?!?! lol
[18:50:35] <Imperator_> on that PCI proto board is a CPLD that does the PCI stuff, I think in such a CPLD can't be so much stuff, maybe what you can read on www.fpga4fun.com is nearly everything
[18:51:49] <Imperator_> http://www.fpga4fun.com/PCI.html
[19:00:11] <alex_joni> back
[19:06:03] <Jymmm> every time I see pic of the k2 machines, I just start to foam att the mouth
[19:06:19] <alex_joni> k2?
[19:06:28] <Jymmm> k2cnc.com
[19:08:13] <Imperator_> hm, a lot of aluminium
[19:09:09] <Imperator_> not the material of first choice for a frame of a mil
[19:09:33] <Jymmm> but it s a router, not a mill =)
[19:09:46] <Imperator_> ah, ok
[19:10:13] <Jymmm> and it beats MDF =)
[19:10:58] <Jymmm> and the stupid thing is I can't figure out a nice way to design a cnc router with mdf -
[19:11:33] <Imperator_> :-)
[19:12:26] <Jymmm> I don't see anythign wrong with MDF, but would like the movement a lil cleaner.
[19:13:29] <gezr> what is MDF?
[19:13:46] <Jymmm> Medium Density Fiberboard....
[19:13:50] <Imperator_> mittel dichte Fasterplatte :-)
[19:13:59] <Imperator_> you are faster :-)
[19:14:13] <Jymmm> http://images.google.com/images?q=MDF&num=100&hl=en&lr=&safe=off&sa=N&tab=wi
[19:14:15] <alex_joni> looks likewood to me
[19:14:34] <Jymmm> MDF == sawdust + glue
[19:15:04] <Jymmm> VERY easy to mill/machine, pretty ridged too.
[19:16:08] <Jymmm> http://www.design-technology.org/mdf.htm
[19:17:03] <gezr> ah okay, yeah I know that stuff, like masonite
[19:17:20] <Jymmm> This I didn't know though.... "MDF contains a substance called urea formaldehyde, which may be released from the material through cutting and sanding. Urea formaldehyde may cause irritation to the eyes and lungs. "
[19:18:19] <Jymmm> I know floral foam uses formaldehyde, didn't know MDF did too.
[19:19:51] <alex_joni> * alex_joni wonders why it's so quiet in here
[19:19:59] <Jymmm> hang overs?
[19:20:01] <alex_joni> where did everybody go?
[19:20:14] <alex_joni> it was a lot more animated sooner
[19:20:20] <paul_c> <lurk>
[19:20:53] <alex_joni> paul_c: found a bug on configure.in
[19:20:54] <Imperator_> * Imperator_ prepairs something to eat
[19:21:18] <paul_c> Yes - I was just looking at that....
[19:21:38] <alex_joni> got the mail to the dev list?
[19:21:51] <paul_c> Of course.
[19:22:27] <alex_joni> damn typo ;)
[19:25:00] <paul_c> damned typos of alex's - Have to check all the other branches & modules now....
[19:30:26] <CIA-4> 03paul_c * 10rcslib/etc/ (configure configure.in): Fix a bug that Alex committed a change for (in the emc2 tree).
[19:31:49] <CIA-4> 03paul_c 07bdi-4 * 10emc2/src/ (configure configure.in): Those bugs of Alex's get everywhere...;)
[19:32:04] <Jymmm> lol, that's cold.
[19:35:02] <paul_c> You don't see M$ fixing bugs that quick do you !
[19:35:40] <Jymmm> sure i do. Oh, you said FIX.... my bad.
[19:41:32] <Jymmm> anyone know what this rail material is called:
[19:41:34] <Jymmm> o
[19:41:34] <Jymmm> _/ \_
[19:42:33] <alex_joni> paul_c: guess who started the bug ;)
[19:42:43] <alex_joni> http://cvs.sourceforge.net/viewcvs.py/emc/emc2/configure.in?r1=1.1.2.20&r2=1.1.2.21
[19:43:43] <alex_joni> you should rename the commit message now ;)
[19:44:06] <alex_joni> "Those bugfixes of Alex's get everywhere Paul's bugs get... ;)"
[19:45:16] <paul_c> That's right - Blame me 'cos you didn't check it.... Wot happened to peer review ??
[19:45:37] <alex_joni> well.. I took the shame on me ...
[19:45:45] <alex_joni> but then you had to point it out
[19:46:09] <paul_c> * paul_c doesn't hold a grudge.
[19:46:41] <alex_joni> whats a grudge?
[19:46:51] <Jymmm> * Jymmm chants "throwning stones for sale"
[19:47:29] <paul_c> * paul_c spies some low cost lasers...
[19:47:48] <Jymmm> paul_c: usually cheap crap
[19:47:59] <paul_c> 20mW
[19:48:13] <Jymmm> now if you said 20MW now were talking!
[19:48:26] <alex_joni> lol
[19:48:44] <Jymmm> * Jymmm starts ebaying for a 50KW generator
[19:48:57] <Jymmm> 440 3 phase
[19:49:05] <alex_joni> coo
[19:49:29] <alex_joni> paul_c: was just kidding about that bug.. bottom line: it's fixed ;) (and fast too)
[19:49:48] <Jymmm> I can get a 50W CO2 with PS for under $1000 brand new including shipping. But I'm waiting till someone else receives theirs first.
[19:50:15] <paul_c> alex_joni: (hopefully) Like you, I was just pulling your leg..
[19:50:33] <alex_joni> I know...
[19:50:38] <Jymmm> paul_c: what would you do with a 20nW laser?
[19:50:44] <Jymmm> just curious
[19:51:00] <alex_joni> but in the process I increased my CIA stats ;)
[19:51:17] <alex_joni> you were faster on rcslib and bdi-4... I gotta blame my connection for that
[19:52:10] <paul_c> To be honest, I'm of the opinion that 24.1.xx & RTLinux should be dropped from the bdi4 branch
[19:53:19] <alex_joni> any reasons for dropping 24.1.x?
[19:53:31] <paul_c> obsolete
[19:53:41] <alex_joni> stated by rtai?
[19:53:46] <paul_c> Yes.
[19:56:02] <alex_joni> rtai.24.1.13 is still listed under stable
[19:56:08] <alex_joni> at least on www.rtai.org
[19:56:58] <paul_c> You file a bug report against 2.4.xx, and you will be told to upgrade.
[19:57:17] <alex_joni> I know.. and I agree
[19:58:43] <alex_joni> what I would do ... add a few comments to ./configure (if it's rtai-24.1.x or rtlinux -> upgrade to a newer RTAI)
[20:01:19] <Jymmm> dumb question... what is RTAI? Does it provide a API for hard/software?
[20:01:33] <alex_joni> rtai is a RT-extension for linux
[20:01:44] <alex_joni> one of the two that are more common
[20:01:45] <Jymmm> I read the website, just dont get it.
[20:01:51] <alex_joni> rtlinux is the other one
[20:02:05] <paul_c> * paul_c finds bug/problem with Fusion's configure
[20:02:07] <alex_joni> but nowadays rtlinux is starting to get very commercial
[20:02:10] <Jymmm> ok, either one... what their purpose/meaning in life?
[20:02:25] <alex_joni> well.. they create an RealTime environment
[20:02:39] <alex_joni> if you add a RT task it will run in that environment
[20:02:41] <Jymmm> wher the RT kernel doesn't do this?
[20:02:57] <paul_c> alex_joni: RTLinux has been commercial from day one.
[20:02:58] <alex_joni> the RT kernel is part of RTAI or RTLinux
[20:03:14] <alex_joni> paul_c: right, but you had a free version available
[20:03:23] <Jymmm> alex_joni: Sorry, I'm trying to understand the difference between the two.
[20:03:30] <alex_joni> now I don't consider you still do.. if it doesn't work on 2.6
[20:03:41] <alex_joni> Jymmm: standard linux doesn't have RT
[20:03:53] <alex_joni> you need to patch the linux kernel with a RT system
[20:03:59] <alex_joni> either RTAI or RTLinux
[20:04:06] <Jymmm> alex_joni: Oh, ok.
[20:04:30] <alex_joni> after that you'll have RT tasks
[20:04:33] <alex_joni> and normal tasks
[20:04:47] <Jymmm> Can you only have one RT task at a time?
[20:04:57] <gezr> paul_c : is stromboli outdated?
[20:04:59] <alex_joni> you can define RT tasks as you like
[20:05:17] <paul_c> As many RT tasks as memory will allow.
[20:05:27] <alex_joni> Jymmm: the point is, RT will run at a preestablished rate
[20:05:36] <alex_joni> during RT time the RT tasks will run
[20:05:52] <alex_joni> when they finish, context is switched back to normal linux operation
[20:06:11] <Jymmm> alex_joni: I understand the RT concept, just wans't sure if more than one could be in operation concurrently or not.
[20:06:17] <alex_joni> where normal stuff is done (memory management, swapping, user-interface, drivers, etc.)
[20:06:27] <alex_joni> you have a RT-scheduler
[20:06:34] <alex_joni> that schedules RT tasks
[20:06:42] <Jymmm> FIFO ?
[20:06:50] <alex_joni> actually you have more to choose from
[20:06:54] <paul_c> gezr: Stromboli is the stable tree... Let me dig out the statement from RTAI covering the status/naming of the trees.
[20:07:05] <alex_joni> at least that's the way with rtai (not sure about rtlinux, never tried it)
[20:07:06] <gezr> paul_c : hold up
[20:07:31] <gezr> paul_c : I guess my question was totally wrong, since its not rt-linux, its the rtai, which is the newer stuff I think
[20:07:32] <Jymmm> alex_joni: Ok, thanks. I better back out of this now =)
[20:07:34] <alex_joni> Jymmm: RT tasks usually are periodic
[20:08:21] <alex_joni> you usually need RT to do repetive tasks at a given frequency (or at given moments in time)
[20:08:33] <alex_joni> like ADC conversions, sampling, etc
[20:08:43] <alex_joni> motor outputs (like EMC does)
[20:09:49] <paul_c> Stromboli is/was 24.1.12/13
[20:09:58] <paul_c> Etna, 24.1.11
[20:10:10] <alex_joni> * alex_joni thinks etna is cool
[20:10:46] <paul_c> RTAI-3.0 is Kilauea, 3.1, Vesuvio
[20:11:30] <paul_c> RTAI-3.2 will be Magma after a namechange (probably).
[20:12:34] <paul_c> In the words of the RTAI team....
[20:12:38] <paul_c> To sum up, here are the relationships between the CVS tree names and
[20:12:38] <paul_c> RTAI release numbers:
[20:12:39] <paul_c> Etna = 24.1.11
[20:12:39] <paul_c> Stromboli = 24.1.12, 24.1.13
[20:12:40] <paul_c> Kilauea = 3.0, 3.0r2, 3.0r3, 3.0r4, 3.0r5
[20:12:40] <paul_c> Vesuvio = 3.1
[20:12:41] <paul_c> Magma = currently aiming at the 3.2 series
[20:13:18] <alex_joni> and fusion
[20:13:36] <paul_c> Experimental branch
[20:13:56] <alex_joni> yup
[20:14:11] <paul_c> To quote:
[20:14:15] <paul_c> One may have noticed that RTAI/fusion has not been cited in the above
[20:14:15] <paul_c> explanations. As a matter of fact, "fusion" is the experimental branch
[20:14:15] <paul_c> of the RTAI project, and as such, it does not fit yet in the regular
[20:14:16] <paul_c> development organization. For this reason, it is currently maintained
[20:14:16] <paul_c> aside of the usual process, and evolves in a completely unrelated
[20:14:17] <paul_c> manner with respect to the other development/testing/stable branches.
[20:14:34] <alex_joni> I see
[20:18:17] <alex_joni> * alex_joni will be back later
[20:18:29] <alex_joni> alex_joni is now known as alex_joni_away
[20:19:05] <jmk-away> jmk-away is now known as jmkasunich
[20:19:32] <Jymmm> * Jymmm appreciates alex_joni warning =)
[20:21:10] <paul_c> Fusion's configure is also broken... It won't accept --with-cc or envion settings
[20:22:47] <Jymmm> paul_c: If you don't mind me asking, what do you do for a living?
[20:23:34] <paul_c> today ?
[20:23:39] <Jymmm> sure =)
[20:23:45] <paul_c> Nothing.
[20:24:16] <Jymmm> paul_c: how about M-f ?
[20:24:48] <paul_c> Open to offers.
[20:25:02] <Jymmm> paul_c: heh, fair enough =)
[20:25:45] <jmkasunich> paul is a "man of leisure"
[20:25:53] <paul_c> You had something in mind ?
[20:25:55] <jmkasunich> ;-)
[20:26:25] <Jymmm> jmkasunich: Ah.... "Prefessionally Unemployed" == Consultant
[20:26:56] <Jymmm> paul_c: Nah, just beign nosey... especially since you do dev and machine work.
[20:27:49] <Jymmm> rockwell60 or ss 303 ???
[20:30:57] <Jymmm> which is harder?
[20:32:04] <jmkasunich> depends... rockwell 60 is a measure of hardness... 303 is an alloy, that can have varying hardness depending on processing, heat-treat, etc
[20:32:26] <jmkasunich> is that rockwell B 60 or C 60
[20:32:38] <jmkasunich> C 60 is pretty hard, I doubt 303 would be that hard
[20:32:46] <Jymmm> Not sure, they only said "Rockwell 60"
[20:33:24] <Jymmm> * Jymmm feels so clueless with this stuff much fo the time.
[20:33:37] <jmkasunich> who is "they"?
[20:33:57] <Jymmm> http://www.vxb.com/page/bearings/PROD/LM/Kit659/
[20:34:13] <Jymmm> the rail is chrome plated
[20:34:31] <gezr> 303 i dont think can become that hard, its not a very high carbon stainless, a 416 can be hardened to that level,
[20:34:45] <gezr> the chrome is what makes it have a high outside hardness, chrome is hard stuff
[20:35:43] <Jymmm> the metal supply house offers: 303 304 316L and 17-4 SS rod
[20:36:09] <jmkasunich> you need stainless because?
[20:36:14] <gezr> 17-4 is a ph stainless, its probably the hardest of the bunch
[20:36:32] <pfred1> isn't stainless stainless because it doesn't have carbon in it?
[20:36:58] <jmkasunich> nope... it's usually stainless because it has lots of chromium in it (and often nickel too)
[20:36:59] <gezr> but if its chrome plated the chrome gives it the outside hardness, its not a complete hardness
[20:36:59] <Jymmm> jmkasunich: just looking for local source of materials
[20:37:10] <pfred1> I know you can't burn stainless
[20:37:17] <pfred1> it simply won't burn
[20:37:27] <jmkasunich> if you don't _need_ stainless, you don't want to use stainless... expensive, and can be a bitch to machine
[20:37:40] <gezr> Jymmm : ground 4140 or ground 4340 would suit your needs just as well
[20:37:56] <jmkasunich> that's what I was thinking, 4140 TGP
[20:37:56] <Jymmm> jmkasunich: Was looking at it for y and z rails.
[20:38:03] <gezr> that stuff starts out of the mill at around 26 rhc
[20:38:12] <jmkasunich> unless you need corrosion resistance, skip the stainless
[20:38:24] <Jymmm> what is the difference between B and C ?
[20:38:30] <gezr> quite a bit
[20:38:34] <jmkasunich> you mean hardness?
[20:38:39] <Jymmm> in general
[20:38:40] <jmkasunich> B and C are two different scales
[20:38:55] <jmkasunich> high on the B scale is low on the C scale
[20:38:59] <Jymmm> so 10b might equal 50b (as example) ?
[20:39:04] <Jymmm> 50c
[20:39:09] <jmkasunich> other way around
[20:39:16] <Jymmm> oh, ok.
[20:39:16] <jmkasunich> 10C might equal 50B
[20:39:29] <Jymmm> so basically diff between K and M ?
[20:40:07] <jmkasunich> http://www.gordonengland.co.uk/hardness/hardness_conversion_1m.htm
[20:40:37] <jmkasunich> 10C is 91B
[20:40:58] <Jymmm> oh, big difference.
[20:41:25] <paul_c> Stainless is also easy to dint unless it has been heat treated.
[20:41:43] <gezr> Jymmm : you need hardness or a case hardness so that the roller's balls dont divit the rod they roll on, much like a bearing casing, chrome and a good case hardening would provide about the same external hardness, even a nitrotec process would give a good case hardening.
[20:41:48] <jmkasunich> is "dint" a type, or is it British for "dent"?
[20:41:53] <Jymmm> I tried looking for PGS in my area, but no luck.
[20:42:11] <jmkasunich> Jymmm: what is your area?
[20:42:17] <Jymmm> San Jose Calif
[20:42:27] <paul_c> dint - Small nicks or dents..
[20:42:27] <gezr> Jymmm : just find a shop with a centerless grinder
[20:43:24] <Jymmm> So PGS isn't available "off the shelf" ?
[20:43:34] <gezr> sure it is, but its going to cost you
[20:43:38] <jmkasunich> what is PGS?
[20:43:40] <Jymmm> oh, heh
[20:43:46] <Jymmm> Precision Ground Steel
[20:43:47] <gezr> and stainless isnt that hard off the shelf
[20:44:11] <gezr> you need to look for it designated as TGP turned ground and polished
[20:44:14] <jmkasunich> oh... I've usually heard it refered to as TGP - turned, ground, and polished
[20:44:16] <Jymmm> You know, I'm probably overthing this whole thing.
[20:44:33] <gezr> jmkasunich : /hi5
[20:44:39] <pfred1> you need to look for a winning lottery ticket
[20:45:05] <jmkasunich> 4140 TGP isn't very pricy
[20:45:13] <gezr> no its not, good stuff,
[20:45:28] <gezr> Jymmm : you can get that pre hardened to 36rhc
[20:46:08] <gezr> or harder
[20:46:12] <Jymmm> What I'm trying todo is make an inexpensive cnc router to sell (retail < $1000 less computer), but with enough quality in it to say I'd personally use it for small jobs.
[20:46:56] <pfred1> Jymmm with a table of about 2x3 inches right?
[20:47:04] <jmkasunich> you and about 100 other folks ;-)
[20:47:10] <Jymmm> pfred1: 2x3 or maybe 1x2
[20:47:27] <jmkasunich> inches?!?
[20:47:29] <Jymmm> jmkasunich: The only way I know I can do this is in qty.
[20:47:30] <pfred1> upgrade to 10x14 for only $2,999 !
[20:47:31] <Jymmm> feet
[20:48:01] <gezr> cost goes up squared for each aditional inch of travel or there abouts I think
[20:48:20] <Jymmm> gezr: Yeah, that's what I've come across as well.
[20:48:21] <jmkasunich> actually, one-offs can be cheaper, cause you can scrounge many of the parts... when you have to buy everything from repeatable sources, the $1000 number isn't very high
[20:48:43] <Jymmm> jmkasunich: Thus why I'm looking for alternative materials.
[20:48:54] <pfred1> you can't get much non consumer material for $1000
[20:49:02] <jmkasunich> I could build _one_ unit for well under $1000... but to be able to produce them.... no way
[20:49:16] <gezr> not at one unit at a time
[20:49:26] <gezr> if you get enough to build 50 of them maybe
[20:49:41] <paul_c> What kind of accuracy & repeatability are you aiming for ?
[20:49:49] <jmkasunich> and rigidity?
[20:50:26] <Jymmm> Well..... MDF for the base material, aluminum channel/angle for the rails (unless I can find alternative material)
[20:50:29] <paul_c> also, the cutter size range & material thicknesses ?
[20:50:37] <pfred1> I've seen MDF with drawer slide machines that seem pretty accurate
[20:50:40] <Jymmm> HF trim router.
[20:51:39] <Jymmm> just think "one step above using 1/4-20 threaded rod"
[20:51:56] <jmkasunich> oh... then you can do it for less than $1000 ;-)
[20:51:57] <les> hi all.
[20:51:58] <pfred1> 5/16-18 ?
[20:52:07] <jmkasunich> I was thinking several steps above all-thread
[20:52:09] <gezr> les : how are things
[20:52:13] <Jymmm> pfred1 smartass.... acme rod... ok =)
[20:52:29] <jmkasunich> how about bronze bushings on TGP?
[20:52:37] <pfred1> Jymmm if you polish all thread I think it's nice
[20:52:41] <paul_c> You'd be muscling in on Les' market at that price range.
[20:52:47] <Jymmm> jmkasunich: I was thinking bronz bushing but wasn't sure about them.
[20:52:55] <les> ok just in a paint program drwing in the stone fireplace to see how it would look
[20:53:00] <les> drawing
[20:53:10] <Jymmm> pfred1: all-thread == threaded rod from home depot?
[20:53:13] <paul_c> Jymmm: ptfe bushes..
[20:53:22] <les> What's this allthread stuff? good acme rod is awfully cheap
[20:53:28] <pfred1> Jymmm yeah regular threaded rod
[20:53:28] <Jymmm> paul_c: would you use plastic bushing?
[20:53:38] <jmkasunich> right - all-thread is crap
[20:53:50] <gezr> so is the cheap acme rod
[20:54:00] <pfred1> jmkasunich well there's lots of the "crap" around and it seems to work
[20:54:09] <Jymmm> No, I want to use cheap acme rod. np there.
[20:54:20] <les> the nook stuff is not very expensive
[20:54:33] <les> A nice dinner would cost more than a set
[20:54:40] <Jymmm> les neither is the ENCO stuff
[20:54:55] <Jymmm> $6
[20:55:10] <jmkasunich> 5/8 dia, 5tpi, heat-treated 4140 ACMS screws from McMaster Carr - $30 for 3 feet
[20:55:33] <jmkasunich> I've used the stuff, it's pretty nice - very smooth black oxide finish
[20:55:50] <jmkasunich> I made a tap out of a length of it and cut matching nuts in bronze
[20:55:59] <jmkasunich> very low backlash, smooth running
[20:56:13] <jmkasunich> (my application was very low speed tho, no wear worries)
[20:56:19] <Jymmm> $6.99 USD 1/2-10 6'LENGTH ACME THREADED ROD
[20:56:23] <jmkasunich> based on the surface finish, I thin it would wear well
[20:56:52] <pfred1> I took 1/2-13 and spun it and ran a thread with valve griding compound in it over the whole thing than used UHMW block tapped
[20:56:53] <jmkasunich> what's the accuracy on that $6.99 rod?
[20:56:54] <Jymmm> http://www.use-enco.com/CGI/INSRIT?PMAKA=408-0220&PMPXNO=942694&PARTPG=INLMK32
[20:57:07] <Jymmm> jmkasunich link ^^^^^
[20:57:50] <Jymmm> they have 304SS acme too
[20:58:03] <Jymmm> http://www.use-enco.com/CGI/INPDFF?PMPAGE=621
[20:58:06] <jmkasunich> can't connect
[20:58:06] <Jymmm> PDF
[20:58:51] <Jymmm> http://www.use-enco.com/CGI/INSRIT?PMAKA=408-0220
[20:59:27] <Jymmm> gawd I hate FRAMES
[20:59:32] <jmkasunich> a pox on enco... their website doesn't work with konqueror or mozilla, just with IE
[20:59:45] <Jymmm> * Jymmm is using Moz
[21:00:45] <Jymmm> jmkasunich they only say 126Hb, no precision
[21:01:02] <jmkasunich> both browsers on my linux box barf at it, but IE on the W95 box is happy
[21:01:12] <alex_joni_away> alex_joni_away is now known as alex_joni
[21:01:25] <alex_joni> * alex_joni is back
[21:01:30] <pfred1> jmkasunich on http://www.use-enco.com/ ?
[21:01:37] <jmkasunich> that stuff is intended for static applications like bolts, not for leadscrews
[21:01:40] <K`zan> jmkasunich: No one has a site that makes it worth having to use ie...
[21:01:56] <jmkasunich> especially not Enco - yuck
[21:02:01] <pfred1> that and the fact that it works fine in Mozilla
[21:02:44] <Jymmm> jmkasunich: well we are talking < $1k too now
[21:02:47] <paul_c> Morons !!!
[21:03:03] <paul_c> accept cookies from enco, and it still refuses...
[21:03:19] <jmkasunich> I'm running moz 0.9.2.1, and get "the connection was refused when attempting to contact www.use-enco.com"
[21:03:38] <paul_c> Your browser is not set to allow cookies
[21:03:58] <narnia> jmkasunich, works for me. ;-)
[21:04:03] <jmkasunich> the moz is - I use it with ebay, which needs cookies (and for the mailman admin pages, and etc, etc)
[21:04:07] <Jymmm> it's a PITA when sites require sessions/cookies to view
[21:04:28] <alex_joni> does work here...
[21:04:31] <jmkasunich> maybe it's sessions... doesn't matter... I used IE to look at the threaded rod
[21:04:32] <alex_joni> on Firefox
[21:04:59] <jmkasunich> and I still think that the diff between $6.99/6ft of bolt grade rod and $30/3ft of leadscrew grade rod is money well spend
[21:05:00] <Jymmm> jmkasunich: yeah, which doesn't block shit.
[21:05:13] <pfred1> enco website on Linux in mozilla http://68.84.51.85:10000/new/enco.jpg
[21:05:37] <pfred1> quit spreading FUD!
[21:06:04] <jmkasunich> huh?
[21:06:49] <jmkasunich> it doesn't work on my box... and I'm not fudding about Mozilla or Konqueror... I'm pissed at enco for using stuff on their website that plain vanilla browsers can't read
[21:07:11] <pfred1> jmkasunich all i have is a "plain" Mozilla browser
[21:07:12] <Jymmm> jmkasunich: try links ?
[21:07:30] <K`zan> LOL
[21:07:32] <Jymmm> just out of curiosity
[21:08:02] <jmkasunich> lynx: unable to connect to remote host
[21:08:09] <K`zan> http://www.use-enco.com/CGI/INSRHM looks fine in firefox here...
[21:08:14] <pfred1> jmkasunich BTW you may just want to think about upgrading to a version of Mozilla that was made in the past year or so it may help
[21:08:14] <K`zan> http://www.use-enco.com/CGI/INSRHM looks fine in firefox here...
[21:08:23] <K`zan> Oops
[21:08:27] <K`zan> Sorry about that.
[21:08:37] <K`zan> Firefox 1.0 here...
[21:08:50] <pfred1> Mozilla 1.7.5
[21:08:54] <jmkasunich> pfred1: I run old software on old computers
[21:09:04] <pfred1> I haven't run .97 in probably over 2 years
[21:09:05] <jmkasunich> newer is _not_ always better
[21:09:10] <K`zan> Firefox ain't heavy :).
[21:09:20] <K`zan> jmkasunich: So very true...
[21:09:27] <pfred1> jmkasunich well with Mozilla newer stuff does run better even on older hardware!
[21:09:41] <narnia> jmkasunich, do you have javascript enabled? i just tried elinks and www.use-enco.com complains about no javascript.
[21:09:57] <K`zan> Yep, the open source stuff does improve and is generally worth upgrading...
[21:10:22] <pfred1> yeah lose the newer needs more horsepower theory
[21:10:38] <K`zan> I really need to build an NC drill setup, I hate drilling PCBs...
[21:10:43] <jmkasunich> http://www.mozilla.org/products/firefox/system-requirements.html
[21:10:45] <K`zan> Not always, but yes.
[21:10:56] <Jymmm> it works in NS4.8 =)
[21:11:02] <jmkasunich> I'm running BDI-TNG, AKA RH7.2 with RTAI patches
[21:11:07] <pfred1> K`zan it's not that bad
[21:11:09] <jmkasunich> they want RH8 or higher
[21:11:35] <jmkasunich> anyway, this isn't about browsers or bloat, it's about leadscrews and linear rails and routers ;-)
[21:11:48] <Jymmm> Yeah! what jmkasunich said
[21:12:31] <K`zan> pfred1: I assume you mean the NC drill machine, yes, once you get it all done, the second one is indeed trivial ;-) LOL
[21:12:48] <K`zan> However, I haven't built the first one yet :).
[21:12:48] <pfred1> K`zan no drilling out PCBs
[21:13:19] <K`zan> 400+ holes sucks with the dremil on a drill stand, or perhaps I am just lazy :)
[21:13:31] <Jymmm> jmkasunich: you think using .250 aluminum angle for the rail would be THAT bad (abec-7 bearings) ?
[21:13:39] <Jymmm> .250"
[21:13:43] <jmkasunich> k'zan: you need a cnc drill!
[21:13:51] <pfred1> K`zan or undertooled http://68.84.51.85:10000/Electronics/Projects/Overdrive/MakeOverdrive/drillboard.jpg
[21:13:56] <K`zan> jmkasunich: No kidding :)
[21:14:14] <jmkasunich> Jymm - I wouldn't use aluminum for rails
[21:14:25] <jmkasunich> it is _very_ soft
[21:14:56] <Jymmm> jmkasunich: cheap ass scrape yard angle iron?
[21:15:08] <K`zan> pfred1: Better than the dremel I think but the foam for a surface is prolly better than the board I use now, thanks.
[21:15:09] <pfred1> which happens to be steel
[21:15:14] <jmkasunich> are you aiming for under $1000 or under $100?
[21:15:19] <Jymmm> what pfred1 said
[21:15:24] <pfred1> K`zan heh yeah foam nicer backer
[21:15:54] <Jymmm> jmkasunich Ratail price would be $1000, my cost < $500
[21:16:05] <pfred1> K`zan that drill press I'm using I got from Harbor Freight for $29.99 too bad they don't offer it anymore
[21:16:09] <K`zan> pfred1: Helping my roomie put together a display for the upcoming lapidary show and have a PILE of that stuff around now :).
[21:16:23] <paul_c> Yay... One of the RTAI team has realised the CC flags may be borked in Fusion..
[21:16:36] <K`zan> pfred1: Yes it is, HF is a staple place, I am surprised I never saw that one :-(.
[21:16:46] <pfred1> K`zan I been using that once piece of foam for a while
[21:16:53] <jmkasunich> so that comes to about $125 per axis and $125 for the spindle... that is pretty cheap
[21:17:10] <K`zan> pfred1: Prolly saves a few drill bits :-).
[21:17:22] <pfred1> K`zan I have never broken one with that rig
[21:17:28] <jmkasunich> I don't know who your customers are, but hell would freeze over before I spent $1000 on something using aluminum slides or old bedrails
[21:17:44] <Jymmm> jmkasunich: It was just a thought =)
[21:18:03] <Jymmm> jmkasunich: but it's gotta be better than drawer slides.
[21:18:25] <gezr> * gezr is proud of his single axis drawer slide thing
[21:18:34] <Jymmm> gezr: url?
[21:18:37] <pfred1> hell yeah drawer slides rule!
[21:18:43] <jmkasunich> well, folks who spend $1000 for a $500 machine are fools
[21:18:58] <Jymmm> jmkasunich: one word.... aol.com
[21:19:03] <gezr> http://24.144.24.182/gallery/cutorbecut/aae
[21:19:25] <Jymmm> gezr: how is it?
[21:19:31] <gezr> it sucks
[21:19:35] <Jymmm> lol
[21:19:40] <Jymmm> bbiab
[21:19:41] <pfred1> gezr nice!
[21:19:42] <gezr> but its better then just watching a motor turn in the past
[21:19:44] <paul_c> <wooppiee.... fun'n'games> Fusion segfaults on removal of modules....
[21:19:47] <Jymmm> * Jymmm nust have coffee
[21:19:59] <paul_c> But hey, this is experimental code.
[21:19:59] <jmkasunich> not saying that there is no place for cheap, simple machines like that... but if you expect sane people to pay $1000 for one......
[21:20:04] <K`zan> pfred1: Heh, I keep the cheap drill bit store ( http://www.store.yahoo.com/drillcity/newtuncardri.html ) in business :-(.
[21:20:30] <gezr> I have a machine I want to build in my head, its going to cost me a lot to build, so Im just learning. right now the project im working on is restoring an old bike
[21:20:37] <jmkasunich> keep in mind that they can buy a grizzly mini-mill for less than 1000
[21:20:53] <gezr> i want to make machines so I can use them :)
[21:20:59] <Jymmm> jmkasunich: there is this, but thats $140 an axis plus screw + motors http://www.vxb.com/page/bearings/PROD/LM/Kit659/
[21:21:01] <pfred1> jmkasunich a rung-fu ?
[21:21:10] <gezr> they wont be of any value to a home shop guy
[21:21:35] <pfred1> jmkasunich keep in mind that they'll pay through the lungs for steppers strong enough to run a grizzly mini-mill
[21:21:58] <gezr> pfred1 : 200oz-inch enough for that you figure?
[21:22:07] <pfred1> gezr 250
[21:22:13] <pfred1> bare minimum
[21:22:18] <gezr> all be darned
[21:22:20] <pfred1> I have an RF-31
[21:22:31] <pfred1> or mini-mill
[21:22:32] <gezr> I hjope to have enough to use servos
[21:22:48] <pfred1> and it's 700 pounds not everyone is up for moving that around!
[21:22:54] <gezr> I want to have essentially a mini version of what ive ran with big machine power
[21:22:54] <jmkasunich> Jymmm - if you are trying to keep costs down, forget ball bearings
[21:23:28] <gezr> if I can snap a 1/2 end mill without bogging down the spindle ive made my goal
[21:23:28] <pfred1> Jymmm automotive sized bearings are very reasonable
[21:23:29] <jmkasunich> bronze or PTFE bushings on drill-rod or TGP rod is cheaper, and can actually perform better in gritty environments
[21:24:02] <pfred1> gezr it's the chatter that'll blow the bit apart :)
[21:24:03] <les> An Abec 1 6204 is what...$300?
[21:24:07] <les> 3.00
[21:24:12] <les> heh
[21:24:30] <gezr> pfred1 : :)
[21:24:33] <jmkasunich> les: I was referring to linear bearings like in the URL you posted
[21:24:36] <alex_joni> hey les
[21:24:40] <les> oh
[21:24:42] <jmkasunich> sory, the url Jymm posted
[21:24:43] <les> hi alex
[21:24:44] <pfred1> gezr and i got a pile of bits to back that statement up!
[21:25:30] <paul_c> Flippin #eck....
[21:25:49] <gezr> pfred1 : I just want something fairly strong you know, im a machinist by trade, I probably wont get anywhere near my goals but ill get close
[21:25:50] <paul_c> Two answers to my RTAI bug report already.
[21:25:55] <les> Perhaps cam ball bearing on homade rail is a good low cost way to go
[21:26:14] <les> Like Bishop Wisecarver ( they are $$)
[21:26:25] <gezr> pfred1 : im wanting 6x6x6 with 4th axis ability
[21:26:31] <jmkasunich> high point stresses need hard rod (and are sensitive to small particles)
[21:26:41] <jmkasunich> bushings, I say! ;-)
[21:26:42] <gezr> no tool changer, but a simple tool exchange method
[21:26:46] <les> yes
[21:26:55] <pfred1> gezr well there goes the mini mill it only goes up 5 inches
[21:28:43] <cnc_wright> paul_c: I checked the BDI4.18 disk I was using be ripping it back into an ISO image. The MD5sum on the new image checked out.
[21:28:47] <pfred1> and with them mini-mills you have to change out the lead screwd for ballscrews
[21:28:58] <pfred1> it's not an option to use what's in there
[21:29:03] <jmkasunich> why not?
[21:29:09] <pfred1> too much backlash
[21:29:12] <jmkasunich> gotta be better than all-thread
[21:29:35] <paul_c> cnc_wright: Long shot - Try another CDROM drive.
[21:29:36] <Jymmm> jmkasunich: ok drillrod is cheap enough... bronze or PTFE ?
[21:30:19] <jmkasunich> could be either
[21:30:21] <pfred1> cnc_wright is the CD drive on cable select jumper?
[21:30:26] <jmkasunich> * jmkasunich looks in mcmaster carr catalog
[21:30:34] <cnc_wright> paul_c: I'll let you know...
[21:30:39] <Jymmm> * Jymmm isn't familuree with PTFE in a high motion enviroment.
[21:31:03] <jmkasunich> 3/4" dia x 1" long bronze $1.07
[21:31:06] <gezr> i got a flyer in from MSC today, they are selling a small cnc machine for 10k, its not quit a mini but not quite bridgeport size
[21:31:18] <les> really
[21:31:34] <Jymmm> I just got my MSC catalog, that thing is HUGE!
[21:31:38] <les> is it a knee type?
[21:31:43] <pfred1> gezr lets just say that i have a mini-mill and i did a little looking and it's way cheaper to build a whole new machine
[21:31:53] <gezr> les : no its the non knee,
[21:32:06] <gezr> pfred1 : oh yeah, im just mentioning things
[21:32:19] <jmkasunich> PTFE same size $3.xx
[21:32:25] <jmkasunich> Rulon - $6-9
[21:32:27] <gezr> les : I think 9x24 table top mill/drill thats been cnced
[21:32:35] <pfred1> people do CNC them but it's not just drop some steppers on and go
[21:32:54] <gezr> pfred1 : what im wanting to do will involve me getting casting made
[21:32:56] <les> I use the bondable teflon sheet for saddles and stuff
[21:33:00] <les> pretty neat
[21:33:05] <Jymmm> jmkasunich: I'm not familure with either one.
[21:33:09] <A-L-P-H-A> I have a 8x24, and I made some brackets, changed it to ball screws, and it works fine.
[21:33:10] <gezr> les : trucite?
[21:33:22] <pfred1> gezr people are doing nice stuff with extrusions I think
[21:33:26] <les> similar to turcite
[21:33:33] <gezr> pfred1 : yeah they are
[21:33:50] <gezr> pfred1 : ive even considered trying an epoxy type base
[21:34:02] <gezr> pfred1 : I dont have much money so I can dream a bit
[21:34:07] <jmkasunich> Jymm... the choice of bushing material isn't so critical
[21:34:30] <jmkasunich> most important thing is being able to get the two rods exactly parallel without lots of precision machining or tweaking
[21:34:40] <jmkasunich> labor will cost more than parts in most cases
[21:35:02] <Jymmm> jmkasunich : Yeah, I'm still working on some sort of "alignment" process.
[21:35:04] <gezr> pfred1 : linear bearing slides, ball screws, table only moves on the X, y and z are flying
[21:35:04] <cnc_wright> pfred1: was on other comptuer. cd drive is on scsi.
[21:35:15] <thrag> hello
[21:35:34] <alex_joni> hello thrag
[21:35:38] <jmkasunich> gezr and Jymmm are talking two different levels
[21:35:40] <les> hi
[21:35:41] <pfred1> cnc_wright then you need the right driver in the boot disk
[21:35:47] <thrag> hi alex. hi les.
[21:35:58] <jmkasunich> gezr: linear bearings, ballscrews jymmm: bushings, acme screws
[21:36:02] <pfred1> you probably using wrong boot image
[21:36:13] <gezr> jmkasunich : ive been thinking about this for 6 years :)
[21:36:22] <thrag> am an experienced linux person, absolutely no idea what i'm doing with cnc, about to find out, can anyone help answer a couple of questions?
[21:36:34] <gezr> jmkasunich : my want hand is still empty, the other I leanred to just wear a glove :)
[21:36:35] <thrag> about linuxcnc.org software?
[21:36:37] <jmkasunich> sure, ask away
[21:36:37] <alex_joni> thrag: shoot
[21:36:40] <thrag> okay.
[21:36:46] <Jymmm> jmkasunich: no no no.... wax coated dowels and nylon string =)
[21:36:49] <alex_joni> a lot of skilled people around (not including me ;)
[21:36:52] <thrag> i just found a sherline 8-axis mill on ebay
[21:37:12] <cnc_wright> BDI boots off the scsi fine and seems to install fine. The scsi driver loads and detects both scsi cd roms during boot. it hangs a little while later in the boot process
[21:37:13] <thrag> and a ... [just looking it up...]
[21:37:30] <pfred1> not for nothing but sherlines are toys
[21:37:53] <jmkasunich> pfred1: depends on what you are cutting... some folks think Bridgeports are toys
[21:37:54] <thrag> mycncmill acumotion kit - http://www.acumotion.com/cnckit3.shtml
[21:38:03] <pfred1> jmkasunich yall
[21:38:24] <thrag> well, i'm happy to spend $200 at the moment, to make something "work".
[21:38:45] <gezr> thrag : thats not a bad investment to get going
[21:38:49] <pfred1> thrag $200 in the world of CNC won't cover the morning cup of coffee
[21:38:58] <thrag> my first pair of skates were ?50. the last pair i bought were ?400, and one pair i bought the _boots_ were ?800. anyway :)
[21:39:14] <jmkasunich> how much is the acumotion kit?
[21:39:21] <thrag> on ebay? $450.
[21:39:29] <thrag> new? $900.
[21:39:44] <thrag> oh - the _kit_ - new about $1800
[21:39:47] <gezr> does sherline do 8 axis?
[21:39:49] <jmkasunich> 3 motors, probably a Xylotex or equiv drive
[21:39:50] <pfred1> that thing 2000chevy used that board from cnckits is pretty cool for the price
[21:40:00] <jmkasunich> sherline's "8 axis" mill has three real axis
[21:40:01] <thrag> on ebay ?200-ish.
[21:40:18] <jmkasunich> they count anything that you can adjust as an axis
[21:40:45] <jmkasunich> if it was a B-port, they would count turret rotation, ram travel, and head tilt and swing as 4 more axis
[21:40:59] <thrag> well, as long as it cuts things - one-offs - i don't really mind. this isn't for serious day-in, day-out usage
[21:41:02] <K`zan> Sheesh, try to find a mini-drill press...
[21:41:30] <pfred1> this board here http://www.hobbycnc.com/driverboards/4aupc/4aupc.htm
[21:41:37] <jmkasunich> thrag - are you looking for more-or-less turnkey, or are you willing to invest sweat equity?
[21:41:55] <pfred1> K`zan yeah i got lucky with mine now they like $130 or so
[21:42:33] <pfred1> thrag this board is pretty cool as a cheap driver board http://www.hobbycnc.com/driverboards/4aupc/4aupc.htm
[21:42:39] <thrag> DRAT, sorry guys, it looks like i just pressed clearscreen.
[21:43:00] <thrag> pfred1: thanks for the link - stored it in bookmarks, really appreciated.
[21:43:14] <pfred1> thrag bout as cheap as it gets for a driver
[21:43:21] <thrag> someone asked a question (which clearing screen just deleted) about turnkey?
[21:43:26] <pfred1> $79 for 3 axis
[21:43:27] <jmkasunich> thrag... alot depends on whether you want to "do-it-yourself" on the cheap, or want something more-or-less complete
[21:43:47] <thrag> 127oz-in motors - not bad for $239.
[21:43:56] <pfred1> thrag yeah turnkey should cost you about 4X what do it yourself will
[21:44:06] <paul_c> thrag: Another Brit ?
[21:44:16] <pfred1> and you're accepting their design considerations
[21:44:19] <jmkasunich> that $79 thing is a kit - solder it yourself
[21:44:20] <thrag> well basically i want to make myself a mini tesla turbine or two, as experiments. obviously i can't _buy_ tesla turbines
[21:44:30] <thrag> hey, i _like_ soldering things together :)
[21:44:43] <pfred1> thrag so you're talking machining metal?
[21:44:58] <thrag> pfred1: yeh.
[21:45:00] <jmkasunich> seems like tesla turbines are mostly lathe work, not mill
[21:45:14] <pfred1> they CNC lathes
[21:45:50] <jmkasunich> yeah, but if he wants to make "one or two tesla turbines", he probably needs a manual lathe, not a CNC mill
[21:45:57] <thrag> jmkasunich: well, i like making life difficult for myself :) there's some really tricky grooves [that most people miss out by not reading the patent properly] that need to provide a path between in and out.
[21:46:05] <pfred1> thrag if all you want it for is a couple of jobs it may pay to just farm out the machining
[21:46:06] <jmkasunich> and certainly not a sherline - too small, unless the turbines are _VERY_ small
[21:46:22] <thrag> yes really small, jmk.
[21:46:31] <pfred1> yeah sherlines good for machining jewelerys
[21:46:41] <jmkasunich> how small? 1" dia?
[21:46:55] <thrag> 3in across, no more.
[21:47:00] <thrag> 3in diam
[21:47:13] <jmkasunich> 3" turbine diameter, so 4" or more for the housing?
[21:47:21] <thrag> yeh, about that.
[21:47:54] <thrag> i need to experiment, plus i want to keep it confidential, hence i'm preferring to do the work myself.
[21:48:14] <gezr> you can enter into an agreement with any shop
[21:48:21] <pfred1> heck their rotary table is a monster!
[21:48:29] <thrag> even if it means breaking things :)
[21:48:51] <thrag> ... mmmm, i'd rather learn to make things! i wouldn't be _here_ if i wasn't!!
[21:49:39] <paul_c> well... This is the place to learn about the software at least.
[21:49:44] <jmkasunich> yep
[21:50:08] <alex_joni> * alex_joni goes to bed
[21:50:13] <A-L-P-H-A> later aj
[21:50:14] <alex_joni> too much learning for me
[21:50:14] <thrag> so anyway - assuming i haven't spent my hard-earned credit card unwisely, the linuxcnc.org CD is basically, just... install it and off i go?
[21:50:15] <jmkasunich> if the groove geometry you are trying to make is too difficult for manual machines, then CNC might make sense
[21:50:25] <thrag> alex_joni: he he.
[21:50:32] <pfred1> thrag not really
[21:50:34] <alex_joni> was kidding ;)
[21:50:38] <thrag> :)
[21:50:49] <thrag> the grooves are described as "labrynthine"....
[21:50:57] <alex_joni> thrag: the BDI is what it's called (a brain-dead install)
[21:51:04] <jmkasunich> you are talking about the sherline 8-axis mill and that conversion kit (on your credit card)?
[21:51:05] <alex_joni> you should be ok with it
[21:51:10] <thrag> alex_joni: gooood.
[21:51:12] <pfred1> yeah the install is brain dead the configuring isn't
[21:51:18] <alex_joni> but you'll need a little configuring to get the machine going right
[21:51:23] <pfred1> a little?
[21:51:33] <alex_joni> number of steps/inch, speeds, etc
[21:51:42] <thrag> jmkasunich: second-hand off ebay, yes. i'd _never_ in my right mind spend $2000 on an experiment!!!!
[21:51:46] <alex_joni> pfred1: compared to compiling emc/rcslib .. yeah
[21:51:48] <jmkasunich> it's a stepper machine, shouldn't be too terribly hard
[21:52:14] <thrag> basically i have to cut out some extremely thin concentric rings in the outer plate and the inner plate
[21:52:26] <alex_joni> later guys
[21:52:29] <thrag> such that they will match up
[21:52:29] <K`zan> pfred1: Up to about $180 for what essentialy looks like yours :-(. Guess I am going to have to be happy with the dremel for a while...
[21:52:30] <gezr> thrag : ive probaly spent that much on playing with things reguarding cnc control of stuff over the years
[21:52:39] <paul_c> The BDI has configs for Sherline
[21:52:45] <thrag> paul_c: GREAT.
[21:52:47] <thrag> thanks
[21:52:50] <K`zan> Off to brew up my Octoberfest homebrew :). BBL
[21:52:50] <pfred1> K`zan try woodcraft they have one similar to mine
[21:52:55] <jmkasunich> concentric rings - concentric to the overall part?
[21:53:05] <jmkasunich> sounds like a job for a manual lathe, not a mill
[21:53:08] <paul_c> And the BDI has a UK support number.
[21:53:27] <pfred1> K`zan but yeah sux harbor doesn't offer mine anymore it's the snizzle for drilling boards
[21:53:28] <alex_joni> thrag: yeah.. bug paul when in doubt
[21:53:40] <thrag> well the gas needs to be able to flow from in to out, even if the turbine is stationary.
[21:53:46] <paul_c> * paul_c forwards all calls to .ro
[21:53:56] <alex_joni> * alex_joni wouldn't mind
[21:54:12] <paul_c> reverse charges.
[21:54:14] <alex_joni> * alex_joni sets up a toll-number (1.5 GBP/minute)
[21:54:21] <alex_joni> too late ;)
[21:54:23] <jmkasunich> thrag: what machining experience do you have?
[21:54:32] <thrag> so around the edge, if you take a cross-section head-on, it looks like several "E"s stacked up so that the gas goes wiggling between them (like a maze)
[21:54:44] <thrag> jmkasunich: ABSOLUTELY NONE :) *ROTFL*
[21:54:54] <thrag> *cackle*
[21:54:59] <jmkasunich> first thing to realize - CNC will NOT make you a machinist
[21:55:10] <thrag> lots of experience of hitting things with a hammer though.
[21:55:19] <thrag> i built my own electric car, does that count ? :)
[21:55:22] <gezr> I dont even own a hamer
[21:55:29] <jmkasunich> in fact IMHO, if you don't know how you would go about making the part with manual machines, then CNC is not going to help and might be a hinderance
[21:55:30] <A-L-P-H-A> * A-L-P-H-A gives thrag a square peg, and round hole.
[21:55:31] <alex_joni> gezr: chips has one
[21:55:33] <jmkasunich> (my $0.02)
[21:55:56] <thrag> jmkasunich: your advice really appreciated.
[21:56:25] <alex_joni> right..
[21:56:27] <alex_joni> good night
[21:56:28] <thrag> yes i realise there are things about speeds, oil, different metals. gonna find out aaaaall about it
[21:56:29] <jmkasunich> others may feel differently, but based on what you seem to want to do, I'd look for a small to medium manual lathe
[21:56:33] <thrag> night alex
[21:57:00] <A-L-P-H-A> jmkasunich, 4" is asking ALOT from a sherline lathe.
[21:57:02] <A-L-P-H-A> ALOT
[21:57:11] <thrag> *sigh* okay. yeh.
[21:57:27] <A-L-P-H-A> I don't think it even can do 4" without a headstock spacer
[21:57:37] <thrag> well... there are bits that would be better off cast rather than machined, and a lathe certainly wouldn't do that.
[21:57:39] <jmkasunich> right - when I said small to medium manual lathe, I was thinking 7-10" swing, like a small south bend, or the harbor freight 7x20, or something like that
[21:57:39] <A-L-P-H-A> Ithought the max size was like 3" dia, or 2.5" dia
[21:58:12] <jmkasunich> sherline isn't small, it's miniature
[21:58:25] <thrag> the exhaust ports on a tesla turbine come out sideways around the bearings.
[21:58:32] <A-L-P-H-A> I've got an 6" dia, bye 3ft bed. Schaublin 102. excellent machine for precision work.
[21:58:32] <paul_c> 3.5" for the Sherline
[21:58:37] <thrag> well, i can always stuff it back on e-bay and find something else :)
[21:58:53] <A-L-P-H-A> paul_c, thank you. howmany sherline machines do youown?
[21:59:15] <paul_c> 1.2533
[21:59:21] <A-L-P-H-A> 1.2533?
[21:59:23] <jmkasunich> that many?
[21:59:31] <Jymmm> jmkasunich: if sherline is minature, then whats this? =) http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=40101
[21:59:34] <paul_c> One mill plus some spares.
[21:59:46] <A-L-P-H-A> sherline is micro.
[21:59:56] <paul_c> Jymmm: Toy
[22:00:00] <A-L-P-H-A> paul_c, ahh. :)
[22:00:21] <A-L-P-H-A> hahaha. I saw that in the store. I was like wTF?
[22:00:23] <jmkasunich> Jymmm: toy
[22:00:24] <Jymmm> paul_c: "That is a precision instrument"
[22:00:27] <paul_c> 6W motor isn't going to able to do much is it...
[22:00:28] <A-L-P-H-A> people buy that stuff?
[22:00:44] <thrag> http://www.sherline.com/specs.htm
[22:00:53] <paul_c> Loads of mugs out there
[22:01:00] <A-L-P-H-A> I'll give it that it has an 18,000 rpm motor. :D
[22:01:08] <jmkasunich> http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=46199
[22:01:28] <jmkasunich> thrag: that might work for you
[22:01:39] <A-L-P-H-A> heh. thatmodel, the combo. People buy it, and a few months later, go back into the store, and buy the real things.
[22:01:43] <thrag> so the 2000 has 8in x 7in x 5.38in travel.
[22:01:59] <jmkasunich> alpha: you mean the 3-in-1?
[22:02:00] <A-L-P-H-A> but for the price, it's not bad at all.
[22:02:03] <A-L-P-H-A> jmkasunich, yes.
[22:02:30] <jmkasunich> I have a shoptask 3-in-1 (bigger than that thing, about $1500 and 700 lbs) that I bought in 1998
[22:02:37] <Jymmm> in the manual for the 3-in-1 they mention the use of a crane for placement! lol
[22:02:51] <Jymmm> hoist
[22:03:04] <thrag> ta jmk. it's nice to know that what i _already_ spent my money on might actually do the job :) teehee. in feet first oh dear me you'd think i'd learn after the first time.
[22:03:12] <jmkasunich> the 3-in-1s are standard chinese crap, but you can do a lot with them in not much space
[22:03:19] <les> whoops
[22:03:22] <A-L-P-H-A> I was so interested in buying one, a few years back... I asked him, so what's the opinion on thi smachine... "oh, awesome" "except people come in and buy bigger machines a few months later"
[22:03:32] <jmkasunich> mind you, I've since added a Van Norman #12 mill and South Bend 13" lathe
[22:03:38] <jmkasunich> but I still use the 3-in-1 a lot
[22:03:47] <Jymmm> jmkasunich how come?
[22:03:53] <jmkasunich> convenient
[22:04:00] <Jymmm> ah, ok.
[22:04:09] <jmkasunich> I was able to get the 3-in-1 into the basement... the larger machines are in the garage
[22:04:14] <jmkasunich> the _unheated_ garage
[22:04:28] <pfred1> jmkasunich don't that suck?
[22:04:32] <jmkasunich> yep
[22:04:38] <gezr> I have a space heater in my garage
[22:04:46] <pfred1> yeah i haven't been able to do anything because of the weather lately
[22:04:47] <gezr> winter is way more manageable then the summers
[22:04:55] <jmkasunich> gezr: where are you?
[22:05:04] <gezr> jmkasunich : central Arkansas
[22:05:06] <pfred1> somewhere warm
[22:05:08] <Jymmm> http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=39743
[22:05:12] <jmkasunich> you don't get winter there
[22:05:31] <gezr> heck no, not a frozen for weeks type winter
[22:05:33] <pfred1> Jymmm just get a lathe
[22:05:49] <pfred1> you can get a lot of lathe for $409
[22:06:12] <pfred1> a lathe that'll do more than that head floting over this thing will
[22:06:13] <Jymmm> * Jymmm wonders if anyone would get upset if I mention that I'm in shorts and actually have the fan running to get some fresh air in the house, plus can actually get a lil tan outside today.
[22:06:30] <pfred1> hey it hit like 40 here today
[22:06:39] <pfred1> tho it was just too windy
[22:06:45] <jmkasunich> yeah, here too... a warm spell, stuff is melting all over
[22:06:46] <paul_c> * paul_c doesn't beleive a word Jymmm says
[22:06:56] <pfred1> the weather is going to break anytime now i can feel it!
[22:07:08] <thrag> jymmm: that's a pretty cool machine,
[22:07:19] <Jymmm> thrag: it only has like 3" travel
[22:07:26] <jmkasunich> thrag: dunno how serious you are, but given some learning this could do what you want pretty well: http://www.shoptask.com
[22:07:29] <pfred1> thrag it's great until you have a job to do on it
[22:07:39] <jmkasunich> I've had an earlier version of it for 7 years now
[22:07:48] <Jymmm> paul_c: http://www.google.com/search?q=weather+san+jose+california&sourceid=mozilla&start=0&start=0&ie=utf-8&oe=utf-8
[22:07:51] <thrag> :)
[22:08:12] <thrag> [off on a browser trip...]
[22:08:23] <pfred1> jmkasunich what's that puppy cost?
[22:08:35] <jmkasunich> dunno what it is now, I paid $1500
[22:08:46] <jmkasunich> (for the manual version)
[22:08:50] <pfred1> jmkasunich you got one of those for $1500?
[22:09:17] <jmkasunich> in 1998, they didn't have the raise/lower head, but other than that, pretty much the same machine
[22:09:24] <thrag> 900lbs in weight good GRIEF
[22:09:27] <pfred1> jmkasunich good deal
[22:09:33] <jmkasunich> light for a real machine
[22:09:39] <jmkasunich> heavy for a toy
[22:09:41] <cradek> haha, it's 75(39C) degrees today, forecast says cold tomorrow, snow tuesday.
[22:09:44] <pfred1> thrag simple fact mass = stability
[22:09:57] <pfred1> especially when you try to work metal!
[22:09:59] <jmkasunich> my #12 Van Norman weighs 1800 lbs
[22:10:00] <gezr> yummy mass
[22:10:32] <pfred1> you'll be casting a concrete base for it after your first chatter experience
[22:10:33] <jmkasunich> thrag: where are you?
[22:10:36] <thrag> ST98W CNC SYSTEM WINDOWS 2495.00 *LOL*
[22:10:40] <paul_c> BP Interact - 1.5 metric tonnes
[22:11:00] <thrag> pfred1: duh, course it does. brain. work.
[22:11:04] <jmkasunich> thrag: their CNC is crap and overpriced...
[22:11:09] <thrag> jmkasunich: banbury, uk
[22:11:23] <paul_c> Oxfordshire
[22:11:31] <jmkasunich> oh... then you don't want the shoptask... shipping would be a bitch ;-)
[22:11:35] <thrag> ah well i won't spend any money on windows then :)
[22:11:43] <pfred1> it seems like there's more people into hobby machining in the UK
[22:11:47] <thrag> :) unless someone's already brought one over.
[22:11:48] <jmkasunich> made in china, imported into the US, then shipped to England... not a good plan
[22:11:59] <pfred1> which makes equipment more expensive and harder to get
[22:12:08] <paul_c> Buy the shoptask type machines over here...
[22:12:33] <paul_c> thrag: google for Chester machine tools.
[22:12:37] <thrag> yeh, for some bizarre reason. the entire south coastline (ship maintenance) is littered with machine shops, people doing fibreglass construction (boat repairs)
[22:12:41] <thrag> paulc: ack
[22:13:55] <jmkasunich> the mill head on that "model B super" is a little....dissapointing?
[22:14:15] <jmkasunich> the shoptask mill isn't the most rigid, but far better than that I expect
[22:14:19] <pfred1> jmkasunich RETAIL PRICE 2995.00 I guess the price has gone up a little since you bought?
[22:14:26] <jmkasunich> wow
[22:14:39] <jmkasunich> yes... is that for the manual machine, or the CNC version?
[22:14:52] <pfred1> Manual Labor
[22:15:03] <pfred1> http://www.shoptask.com/shoptaskwebsite/machines.htm
[22:15:54] <jmkasunich> that's ridiculous
[22:16:19] <jmkasunich> it's a chinese machine for gawd sake!
[22:16:36] <thrag> okay, people - gotta get on. i'm sure i'll be back, though... :)
[22:16:41] <pfred1> jmkasunich well then you make something better for cheaper and drive them out of business!
[22:16:49] <gezr> thrag : take it easy
[22:16:50] <thrag> advice much appreciated
[22:16:55] <thrag> too gezr.
[22:17:00] <jmkasunich> bye thrag
[22:17:09] <thrag> byeall.
[22:17:44] <jmkasunich> pfred1: no way... I'm not an importer, and have no desire whatsoever to deal with chinese tool factories
[22:18:07] <robin_sz> right .. what did I miss?
[22:18:12] <pfred1> jmkasunich allergic to cosmoline are you?
[22:18:19] <jmkasunich> wonder how smithy compares
[22:18:47] <pfred1> the thing is to know your machines and buy quality used there's plenty around
[22:19:08] <jmkasunich> right - that's what I did _after_ I used the shoptask for a couple years
[22:19:20] <jmkasunich> got my Van Norman for $500, and the SouthBend for $800
[22:19:29] <jmkasunich> total $1300, less than I paid for the Shoptask
[22:19:30] <pfred1> guy i know has one of them CNC bridgeport machines I think he got it for nothing
[22:19:42] <jmkasunich> for 3400 lbs of quality American made machinery
[22:19:54] <pfred1> and his lathe i htink he paid like $200 for it it's got an 8 foot bed i htink
[22:19:56] <robin_sz> pah, lightweight :)
[22:20:14] <jmkasunich> too heavy to get in the basement
[22:20:23] <robin_sz> pfred1: in machinery price is inversely proportional to size
[22:20:28] <pfred1> oh that lathe must weigh 4000
[22:20:30] <jmkasunich> I almost cried when I had to pass up a $1500 Monarch 10EE
[22:20:35] <jmkasunich> 3000 lbs there
[22:20:43] <pfred1> and the bridgeport CNC maybe the same
[22:20:46] <gezr> oh man, thats not a bad price either
[22:21:05] <jmkasunich> one of the finest lathes ever made, IMHO
[22:21:07] <Jymmm> jmkasunich: Come on.... a $20 hand truck from home depot and you could of got that in your basemetn NP.
[22:21:07] <pfred1> you hire a towing company with a flatbed
[22:21:17] <pfred1> they'll do it!
[22:21:39] <robin_sz> jmkasunich: better than the Hardinge HLV
[22:21:40] <robin_sz> ??
[22:21:43] <jmkasunich> 1927 house, stairwell in the middle, no direct access to any door
[22:21:51] <pfred1> that's how Larry got the stuff he has home
[22:22:03] <jmkasunich> robin_sz: in the same class as the HLV
[22:22:11] <robin_sz> coo
[22:22:48] <robin_sz> now .. on the subject of which ...
[22:22:50] <Jymmm> jmkasunich: Got sawsall?
[22:22:56] <robin_sz> please someone take my CNC punch!
[22:23:19] <Jymmm> robin_sz: you fly, I'll buy
[22:23:36] <jmkasunich> CNC punch $20... freight $20,000
[22:24:22] <robin_sz> ok, I'll pay the $20 .. you pay the freight ..
[22:25:31] <jmkasunich> lhttp://www.lathes.co.uk/monarch/page2.html
[22:25:48] <Jymmm> robin_sz I said YOU fly, I'll buy =)
[22:26:26] <pfred1> Jymmm monarchs are the real deal
[22:26:35] <gezr> the cadalic
[22:27:00] <gezr> however you spell it
[22:27:21] <pfred1> gezr buick man huh?
[22:27:27] <jmkasunich> * jmkasunich drools at the lathes.co.uk photos
[22:27:31] <gezr> pfred1 : ford
[22:28:00] <gezr> I would love to own a nice 10EE
[22:28:09] <pfred1> yeah I need a good lathe
[22:28:21] <pfred1> when I'm permanently in my new location
[22:28:55] <gezr> I would just as much love to have a myford super7 as much though
[22:29:10] <pfred1> heck I'm figuring on a southbend 9x20
[22:29:10] <jmkasunich> easier to move around, that's for sure!
[22:29:15] <gezr> yep
[22:29:20] <robin_sz> Myford?
[22:29:24] <gezr> I have a southbend 9x36 im rebuilding
[22:29:38] <jmkasunich> my brother has a SB 9 he wants to sell (western PA)
[22:29:42] <robin_sz> myford are not "real" lathes ...
[22:29:53] <gezr> robin_sz : for real?
[22:30:09] <gezr> robin_sz : compared to say a hf mini lathe?
[22:30:31] <robin_sz> but compared to a decent Colchester or Boxford ...
[22:30:35] <pfred1> gezr that'll do it
[22:30:50] <gezr> yeah I see the point
[22:30:51] <robin_sz> you pay more $$$ for a myford than a similar industrial lathe ..
[22:30:52] <jmkasunich> three classes of machines: 1) top quality industrial toolroom machines 2) light industrial/hobby machines 3) chinese crap
[22:31:03] <robin_sz> yeah
[22:31:16] <jmkasunich> Myford is class 2, I think
[22:31:24] <jmkasunich> as is the SB 9"
[22:31:26] <robin_sz> the only reason the hobby machines exist is because toolroom lathes where too much $$$
[22:31:40] <robin_sz> now, class 1 is cheap
[22:31:41] <gezr> either a monarch or a hardinge would be good class 1 machines, but new they are astronomical
[22:31:47] <robin_sz> nah
[22:31:50] <robin_sz> cheap
[22:31:54] <gezr> I mean new
[22:31:59] <robin_sz> well new yes
[22:32:13] <robin_sz> but you'll buy an HLV for less than a myford super 7
[22:32:26] <pfred1> yeah at my age I only need maybe 20 years out of a machine
[22:33:11] <gezr> I can't, wait my floor cant support a machine of that weight
[22:33:45] <jmkasunich> gezr: how heavy can you take? 300 lbs? 500?
[22:34:01] <robin_sz> myfords still command big $$$ from the steam-engine making fraternity that want something light enough to go in the back bedroom on the first floor
[22:34:02] <gezr> a south bend heavy 10 starts to cross the line from class2 to 1
[22:34:02] <Jymmm> 40#'s max
[22:34:16] <Jymmm> 12 ounces at a time
[22:34:19] <gezr> I think I have 4" of concrete
[22:34:33] <Jymmm> gezr: got jackhammer?
[22:34:36] <jmkasunich> gezr: yea, IMO the heavy 10 is the ideal SB
[22:34:38] <gezr> Ill get a bridgeport sooner then later
[22:34:44] <gezr> jmkasunich : I have a friend with one
[22:34:46] <robin_sz> a 12x20 toolroom lathe is easy on 4" conc
[22:34:50] <pfred1> heh you dopes you pour a pad on top of the existing
[22:34:55] <jmkasunich> large enough to do real work (5C collets, 1-3/8 thru hole) but not too large
[22:35:12] <Jymmm> gezr: I have never even mounted a safe on 4"
[22:35:12] <gezr> jmkasunich : I didnt have the money to buy it, he may sell it to me in time, who knows :)
[22:35:14] <jmkasunich> My 13" SB has the same spindle as the heavy 10
[22:35:16] <pfred1> unless you're planning on working on the floor
[22:35:37] <gezr> I dont work on the floor, rule 2
[22:35:37] <pfred1> that's how machines are put into commercial places
[22:35:51] <robin_sz> so where were we, .. oh yes .. shipping address for this punch ;)
[22:36:19] <jmkasunich> 100 meters down a 50 meter pier
[22:36:22] <pfred1> gezr yeah i tend to avoud those floor jobs now myself too sometimes though still there
[22:36:22] <gezr> pfred1 : we put down 16 inches of base for the 36klb machine i run at work :)
[22:36:46] <Jymmm> robin_sz: Jymmm c/o Port of Stockton, Pier 12, Doc 49, Stockton california USA
[22:36:55] <pfred1> gezr I've poured a smokestack anchor was 22 foot cube
[22:37:03] <robin_sz> digging around in a cupboard I found 2 nice steppers and 8 geckos I;d forgotten about :)
[22:37:05] <gezr> huge
[22:37:08] <pfred1> was somehting like 143 yards
[22:37:59] <robin_sz> yards?
[22:38:00] <gezr> my shaper needs to be anchored to the floor, that damn thing giddie up and go
[22:38:18] <jmkasunich> cubic yards of concrete (we use strange units here)
[22:38:24] <robin_sz> heh,
[22:38:25] <gezr> I am the atlas man
[22:38:28] <robin_sz> you do for sure!
[22:39:07] <jmkasunich> talk about machines where price = 1/size... shapers
[22:39:20] <robin_sz> yeah
[22:39:25] <jmkasunich> 7" south bend shapers go for $500-800
[22:39:33] <gezr> I have a 7" atlas
[22:39:34] <robin_sz> shapers are a little used tool nowadays
[22:39:38] <jmkasunich> 30" shapers pretty much get given away
[22:40:31] <Jymmm> http://csdl.computer.org/comp/proceedings/rtss/1996/7689/00/76890300abs.htm
[22:40:50] <robin_sz> I went to a sale were they had some horizontal borers, like 10m high, 6m^2 table .. they went for �1000 or less
[22:41:23] <robin_sz> about the same price as some pillar drills in the same sale ...
[22:41:52] <gezr> big iron isnt very expensive
[22:41:57] <Jymmm> robin_sz: Dang, you need to get into the recycling business, or buy a steel/iron mill/forge
[22:41:58] <gezr> but its just big
[22:42:02] <robin_sz> right.
[22:42:27] <robin_sz> theres very little new big iron ... its all old.
[22:42:56] <Jymmm> did they have a choice back in the day?
[22:43:15] <robin_sz> apart from that 5 axis machine at Scaled Composites, obviously :)
[22:43:29] <gezr> cincanaiti, monarch, hardinge, lots of builders back in the day
[22:43:59] <A-L-P-H-A> gezr, how can a man not own a hammer?
[22:44:22] <gezr> A-L-P-H-A : my wife has one we use in the house, I dont knwo where it is
[22:44:26] <Jymmm> A-L-P-H-A: I didn't own a hammer till someone gave me one last year.
[22:44:34] <robin_sz> but they are so handy ...
[22:44:39] <jmkasunich> robin_sz: that machine at Scaled isn't big iron, it's probably big steel, or even big aluminum
[22:44:44] <pfred1> yeah I must own 30 or so
[22:44:44] <robin_sz> for "adjusting" things
[22:44:53] <robin_sz> jmkasunich: true, true.
[22:45:07] <A-L-P-H-A> well, my house probably has like 5-15 of them... dunno where they all are...
[22:45:08] <gezr> all i meant was that machining doesnt involve deforming the face of a part with a hammer
[22:45:08] <Jymmm> robin_sz: Oh, you mean a 4' crowbar.... those I own =)
[22:45:36] <robin_sz> jmkasunich: I presume steel. if it was ally and that size, youd have LOTS of problems with explansion
[22:45:43] <pfred1> gezr actually the only way to position things is by hammering them
[22:45:55] <robin_sz> * robin_sz nods
[22:45:57] <pfred1> you can't move it with your hand
[22:45:58] <gezr> yeah
[22:46:13] <gezr> I use a deadblow
[22:46:14] <A-L-P-H-A> I own an arbor as well.
[22:46:15] <A-L-P-H-A> :D
[22:46:19] <pfred1> move it .0005 right!
[22:46:23] <gezr> not a ballpeen
[22:46:34] <A-L-P-H-A> arbors are sooo useful.
[22:46:45] <robin_sz> as hammers?
[22:46:50] <Jymmm> * Jymmm hammer has 4 flat head screwdrivers in the handle =)
[22:47:03] <A-L-P-H-A> you can kill a man, and use it to weigh them down in the lake.
[22:47:15] <robin_sz> * robin_sz adds that to his list
[22:47:17] <A-L-P-H-A> Jymmm? you don't have real screw drivers?
[22:47:36] <A-L-P-H-A> scary.
[22:47:42] <Jymmm> A-L-P-H-A: I have a $50 manual screwdriver, but no hammers
[22:47:59] <pfred1> Jymmm a $50 screwdriver?
[22:48:04] <Jymmm> snap-on
[22:48:14] <robin_sz> I think i traded most of my screwdrivers for hex balldrivers.
[22:48:26] <Jymmm> robin_sz those I have
[22:48:35] <A-L-P-H-A> a real tool box needs: real screw drivers, wrenches (preferrable wracheting ones), hammers, files, socet wrenches. and DUCK TAPE.
[22:48:43] <pfred1> they wear out awfully fast don't they?
[22:48:45] <robin_sz> DUCT tape
[22:48:49] <Jymmm> A-L-P-H-A you forgot the bailing wire
[22:49:01] <pfred1> A-L-P-H-A keeps the quacks down huh?
[22:49:04] <Jymmm> just no hammers
[22:49:21] <pfred1> I have to take a pic of my hammer basket
[22:49:36] <robin_sz> DUCT tape (or Gaffa tape) originally came from the ventilation industry ..
[22:49:43] <robin_sz> for sealing ductwork
[22:50:07] <jmkasunich> DUCK tape on the other hand, is for sealing ducks...
[22:50:15] <robin_sz> fair point.
[22:50:26] <jmkasunich> if they're not sealed, the water leaks in and they sink
[22:50:36] <robin_sz> ah yes.
[22:50:38] <SteveStallings> not to mention what leaks out
[22:50:45] <A-L-P-H-A> it was called duck tape, cause in wwi or wwii, they used it to keep ammo boxes dry. And when they noticed the water repealled off like water off a ducks back. [that was from a TV show on how duck/duct tape is made]
[22:50:46] <robin_sz> they have a hole at the back, from what I remember.
[22:51:13] <jmkasunich> at least one
[22:51:26] <A-L-P-H-A> http://en.wikipedia.org/wiki/Duck_tape
[22:51:33] <Jymmm> pfred1: best damn screwdriver I ever bought... http://buy1.snapon.com/catalog/item.asp?P65=&tool=hand&item_ID=6447&group_ID=702&store=snapon-store&dir=catalog
[22:52:10] <A-L-P-H-A> damn my computer is loud
[22:52:20] <pfred1> Jymmm it's pretty fancy
[22:52:29] <pfred1> A-L-P-H-A can you hear it out on the street?
[22:52:33] <A-L-P-H-A> Jymmm, not a real mans screw driver.
[22:52:33] <Jymmm> pfred1: has a very nice ratcher to it.
[22:52:51] <A-L-P-H-A> I'm talking manual traditional ones, with nice good solid handles.
[22:53:06] <A-L-P-H-A> I prefer the ones with the rubber coating, with lifetime warrantees.
[22:53:08] <Jymmm> A-L-P-H-A this has the best damn handle I've ever seen
[22:53:21] <Jymmm> A-L-P-H-A snapon is lifetime warratanted
[22:53:26] <pfred1> man buy craftsman when it's beat you bring it back
[22:53:27] <A-L-P-H-A> yeah, my good screw drivers are like that one, but manual.
[22:53:30] <pfred1> get a new one
[22:53:38] <jmkasunich> $50 for a screwdriver, it better be warranteed
[22:53:48] <pfred1> snap-on lifetime
[22:53:49] <A-L-P-H-A> I have craftsman, and mastercraft. both have lifetime warrantees.
[22:54:17] <A-L-P-H-A> mastercraft is Canadian Tire brand... a national chain, that's been around for 40 years.
[22:54:17] <pfred1> crapman quick tools!
[22:55:35] <robin_sz> I prefer the Stanley rubber handled screwdrivers wiuth a triangular grip
[22:55:42] <pfred1> in this pic the sockets in the red box are snap-on http://68.84.51.85:10000/wood/cshop/tools/p9290015.jpg
[22:55:45] <Jymmm> A-L-P-H-A it has a great feel in your hand. plus you can really put some torue on it and not feel that the shank is twisting on you or the tip is abut to bend.
[22:55:48] <pfred1> I wonder what that set cost?
[22:56:28] <A-L-P-H-A> grey tools is another good company.
[22:56:30] <robin_sz> the rep does a good demo, he has a trinagular handle coupled to a normal screwdriver handle .. you can try and twist it agianst hime either way around .. the triangle always wins ...
[22:56:35] <Jymmm> pfred1: that + small ratchet == $149 and gawd I love those sockets.
[22:56:38] <A-L-P-H-A> used to be nice highend metal in all their tools.
[22:56:46] <A-L-P-H-A> now they're just imports with their names on it
[22:56:50] <robin_sz> yep
[22:56:57] <A-L-P-H-A> same warrantee... but shittier tooling
[22:57:09] <pfred1> heh snap-on doesn't evne have this set on their website it's 8-27mm
[22:57:14] <pfred1> I mean 6
[22:57:15] <robin_sz> I prefer "Beta Untensli" from italy, pricey, but very nice
[22:57:25] <A-L-P-H-A> craftsman in Canada is not hte smae Craftsman in the states... we get the shitty imports, where you guys get made in USA
[22:57:34] <pfred1> snap-on best sockets made
[22:57:40] <robin_sz> nah
[22:57:46] <Jymmm> pfred1 mine is 9/16 to 3/16 in ten steps
[22:57:48] <robin_sz> snap-on is all marketing
[22:57:57] <pfred1> snap-on invented sockets
[22:58:01] <robin_sz> nah
[22:58:02] <A-L-P-H-A> I like the maclearan F1 titanium tool set that comes with the car. :D
[22:58:08] <pfred1> they're the mercedes
[22:58:11] <A-L-P-H-A> 100% titanium.
[22:58:26] <robin_sz> there were sockets around a LONG time before snapon started
[22:58:33] <robin_sz> and there are better ones too.
[22:58:41] <pfred1> no snap-on invented sockets
[22:58:44] <robin_sz> snap-on just market them very well
[22:58:57] <anonimasu> hello everyone
[22:59:24] <Jymmm> the screwdriver and the deep well socket plus a snapper wrench is snap-on, else it's sears or whatever.
[23:00:21] <A-L-P-H-A> I like wrachet wrenches... they're so nice.
[23:00:43] <pfred1> A-L-P-H-A usually they don't have the clearance you need
[23:01:09] <A-L-P-H-A> http://en.wikipedia.org/wiki/Socket_wrench
[23:01:21] <A-L-P-H-A> pfred1, you can get bent arm ones now too.
[23:01:33] <A-L-P-H-A> with like an inch clearance.
[23:01:59] <pfred1> A-L-P-H-A you mean like these? http://68.84.51.85:10000/wood/cshop/tools/p9290018.jpg in the back of the drawer
[23:02:33] <A-L-P-H-A> yeah. I have the flat ones, which work great for me. Though I probably have 75K work of tools in the garage.
[23:02:37] <pfred1> so I guess snap-on did invent sockets
[23:02:41] <A-L-P-H-A> $75K worth
[23:03:04] <pfred1> BTW that 24" 3/8 extension is a snap-on
[23:03:06] <A-L-P-H-A> no. American tools did. But the inventor and co-worker started snapon tools.
[23:03:07] <robin_sz> well well, I stnad corrected
[23:03:29] <pfred1> OK the guys that formed snap-on invented the socket
[23:03:46] <robin_sz> my favourite socekt wrench has a sprag clutch rather than a ratchet
[23:03:56] <A-L-P-H-A> nm, I read that wrong
[23:03:59] <robin_sz> no slip at all
[23:04:15] <robin_sz> wonderful in really tight spaces, you waste no movement at all
[23:04:18] <A-L-P-H-A> my favouring tool is still the hammer.
[23:04:28] <robin_sz> mmm .. hammer.
[23:04:42] <robin_sz> not the "blue tipped spanner"?
[23:05:31] <A-L-P-H-A> favourite style of head on a screw/bolt. Hex, or Robertson. Both can be operated with one hand while theother hand aligns, with no slippage.
[23:05:40] <A-L-P-H-A> philips slips.
[23:05:48] <A-L-P-H-A> flathead just suck.
[23:05:51] <robin_sz> yeah
[23:06:00] <robin_sz> philps and pzidrive are common here
[23:06:04] <robin_sz> pozidrive
[23:06:13] <A-L-P-H-A> torques is too uncommon. except for small electronics. like cellphones, and pdas
[23:06:18] <A-L-P-H-A> never seen a pozidrive
[23:06:36] <A-L-P-H-A> not on wikipedia.
[23:06:48] <A-L-P-H-A> robin_sz!!! make a page. for pozidrive. :)
[23:08:14] <A-L-P-H-A> google isn't helping
[23:08:37] <jmkasunich> pozidrive is just a slighly modified philips
[23:09:56] <A-L-P-H-A> what? so it looks like a nazi symbol?
[23:10:30] <paul_c> OK... Not going to get anywhere with fusion tonight..
[23:10:36] <jmkasunich> no.. the geometry is just a tiny bit different so it's less likely to slop
[23:10:39] <jmkasunich> slip
[23:10:42] <A-L-P-H-A> paul_c, heat? or cold?
[23:10:56] <paul_c> seqfaults in kernel space.
[23:11:08] <A-L-P-H-A> cold fusion... when I was a kid, in 1986... it was such a BIG comotion.
[23:11:19] <A-L-P-H-A> I was 7 at that time probably. 6 or 7...
[23:11:22] <jmkasunich> they add four non-functional little lines to make it look like a star, so you can tell its a posidrive, otherwise you can't tell by eye
[23:11:29] <A-L-P-H-A> I think I still have that popular science mag somewhere
[23:12:04] <A-L-P-H-A> jmkasunich, what makes the head different then?
[23:12:13] <A-L-P-H-A> could I use a philips in a posidrive thing?
[23:12:15] <jmkasunich> don't know the details
[23:12:25] <jmkasunich> you can use a phillips screwdriver in a posidrive
[23:12:38] <jmkasunich> you lose the anti-slip properties
[23:12:54] <A-L-P-H-A> must be a flat base then or something.
[23:13:07] <jmkasunich> many, many screws in consumer products are posi-drive and most folks don't even realise it
[23:13:34] <A-L-P-H-A> I've never heard of them before today
[23:13:50] <A-L-P-H-A> must be some newfangled yankee thing
[23:14:01] <jmkasunich> http://www.shadetreemg.com/posidrive.htm
[23:14:41] <cnc_wright> paul_c: I finally got BDI4.18 working! I reinstalled...still hung at boot. Changed some bios settings and it booted all the way.
[23:15:30] <A-L-P-H-A> oh I have seen those before