#emc | Logs for 2007-06-01

Back
[00:00:03] <jmkasunich> * jmkasunich gives cradek a roll of electrical tape
[00:02:16] <CIA-8> 03cradek 07TRUNK * 10emc2/nc_files/g76.ngc: cut the end spherical before threading
[00:02:50] <cradek> I think this program uses pretty much all of gcode now
[00:05:01] <cradek> yay, I have usb
[00:05:17] <jmkasunich> got the newegg order?
[00:05:40] <cradek> I 2/3 of the shipments my 4 items inexplicably are coming in
[00:07:40] <CIA-8> 03jepler 07TRUNK * 10emc2/src/rtapi/rtapi_bitops.h:
[00:07:40] <CIA-8> for userspace, always use the bundled version of bitops -- the current compromise caused problems on x86 2.6.17-rtai3.5, which passed the current configure test but didn't actually have a usable asm/bitops.h
[00:07:40] <CIA-8> augment the x86-64 set of bitops because a few more are used in rtai_ulapi that aren't in sim_ulapi
[00:19:25] <CIA-8> 03compile-farm 07BDI-4.51 (2.6.16.20-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot6_log.txt
[00:38:53] <CIA-8> 03jepler 07TRUNK * 10emc2/src/emc/rs274ngc/rs274ngc_pre.cc: the gettext convenience macro '_' is now defined in interp_internal.hh
[00:44:33] <CIA-8> 03compile-farm 07BDI-4.51 (2.6.16.20-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot6_log.txt
[00:45:40] <cradek> crap, got the limit error again, it's not chips
[00:45:49] <cradek> maybe noise
[00:47:04] <skunkworks> didn't you run the signal with the motor wires ;)
[00:48:46] <cradek> um yep
[00:49:32] <skunkworks> have you scoped the line yet? at the pluto?
[00:49:45] <cradek> nope
[00:49:55] <cradek> whine first, debug later
[00:50:12] <skunkworks> be interesting to see how much noise is on the line
[00:50:20] <cradek> for some crazy reason I keep putting the screws back in the box
[00:51:50] <skunkworks> I don't have that problem
[00:52:15] <cradek> apparently I really want this to be done
[00:59:33] <CIA-8> 03compile-farm 07BDI-4.51 (2.6.16.20-rtai) * 10emc2head/: build PASSED
[01:00:29] <tomp2> can halscope trigger on an event ( like this error)? I guess some amount of relevant signal history could be stored in a buffer and viewed after the trigger.
[01:01:12] <cradek> sure it could easily trigger on the limit switch
[01:01:44] <jepler> by setting the trigger position, you can control how much of the data collected is from before the triggering event
[01:01:56] <tomp2> oh, limit >switch<... i heard a limit, didnt suspect switch.
[01:04:09] <CIA-8> 03jepler 07TRUNK * 10emc2/src/ (Makefile.inc.in Makefile.modinc.in configure configure.in):
[01:04:09] <CIA-8> * fix JOYSTICK_H_USABLE test
[01:04:09] <CIA-8> * remove traces of ASM_BITOPS_H_USABLE (we always assume it's not)
[01:05:24] <jepler> man CRTs are fuzzy
[01:05:37] <jepler> clearly I should replace this otherwise just fine monitor with a new LCD
[01:06:02] <SWPadnos> clearly
[01:07:09] <SWPadnos> newegg had some you'd hate on sale recently - 22" for $300
[01:07:23] <cradek> it's probably motor noise
[01:07:38] <jepler> hate, eh?
[01:07:50] <SWPadnos> 22", 1680x1050 ...
[01:08:17] <jepler> oh yeah -- I'm afraid I'd have to hold out for 1920x1200
[01:08:24] <SWPadnos> yep, it's much nice
[01:08:25] <SWPadnos> t
[01:08:27] <SWPadnos> r
[01:08:45] <jepler> anything with fewer pixels than my laptop is out of the question
[01:08:51] <SWPadnos> heh
[01:10:01] <SWPadnos> looks like $500-ish is the minimum on taht res, in a 24" variety
[01:10:07] <SWPadnos> (at least at newegg)
[01:11:29] <tomp2> maybe some torroids or shielding on the motor lines
[01:13:14] <jepler> SWPadnos: when you tried chris's SMP RTAI kernel, did it not boot at all, or did it die when you tried RT?
[01:13:26] <SWPadnos> it didn't boot at all
[01:14:12] <SWPadnos> I don't remember exactly where it hung, but it was pretty early on - there wasn
[01:14:24] <SWPadnos> 't even a screen full of boot messages
[01:15:16] <jepler> cradek: it looks like tests/ccomp/lathe-comp now fails
[01:15:29] <cradek> oops
[01:15:39] <jepler> it may be an inconsequential change...
[01:15:53] <jepler> 6 N..... USE_LENGTH_UNITS(CANON_UNITS_INCHES)
[01:15:53] <jepler> 7 N..... USE_TOOL_LENGTH_OFFSET(0.0000 0.0000)
[01:15:53] <jepler> 8 N..... SELECT_TOOL(0)
[01:16:03] <jepler> vs
[01:16:03] <jepler> 6 N..... USE_LENGTH_UNITS(CANON_UNITS_INCHES)
[01:16:04] <jepler> 7 N..... USE_TOOL_LENGTH_OFFSET(0.0000 0.0000)
[01:16:04] <jepler> 8 N..... USE_TOOL_LENGTH_OFFSET(0.0000 0.0000)
[01:16:04] <jepler> 9 N..... SELECT_TOOL(0)
[01:16:13] <cradek> hmm, strange
[01:16:52] <jepler> yep the calls to USE_TOOL_LENGTH_OFFSET are doubled both places they occur
[01:17:32] <cradek> I didn't do that on purpose - wonder what change it was
[01:18:08] <cradek> a well-placed capacitor fixed my problem
[01:19:07] <SWPadnos> somewhere very near the switch input wires? :)
[01:19:14] <cradek> yes
[01:19:29] <cradek> time constant of about .1 servo cycles
[01:19:33] <cradek> shouldn't hurt anything
[01:20:06] <skunkworks> using the internal resistance of the pluto? or do you have a resister in series?
[01:20:27] <cradek> it has a pullup, the switch grounds
[01:20:51] <skunkworks> ah
[01:21:19] <cradek> crappy fix for a crappy problem due to crappy construction: perfect
[01:27:28] <Ziegler> Linux guys... anyone have a decent wireless pci card that they would recomend? (for cheap)
[01:28:40] <cradek> no, I hear you should use the wireless->ethernet bridge thingies instead
[01:28:56] <Ziegler> hmmm
[01:30:20] <cradek> wireless cards rarely work right with free software - ethernet cards almost always do
[01:31:06] <Ziegler> yeah... never had a problem getting an eth0 card to work
[01:31:18] <Ziegler> just never tried a wireless adapter.
[01:32:43] <cradek> if you can help it, don't (seriously)
[01:33:34] <Ziegler> LOL
[02:10:57] <dmwaters> {global notice} Good day all. In about an hour i'm going to begin maintenence on 3 main rotation servers that have been out of rotation for a while. These servers have been out of rotation for a while, but the impact will still be a bit rough. About 2200 users will be effected.
[02:58:22] <toastyde1th> toastyde1th is now known as toastydeath
[03:09:51] <ds2> Hmmmm
[03:17:20] <CIA-10> 03jepler 07TRUNK * 10emc2/scripts/halrun.in: make sure hal threads are stopped before trying to unload. doing otherwise causes kernel panics on x86-64
[03:20:24] <eric_u> http://www.youtube.com/watch?v=a14zELKPw8M Hp printer ripped apart and made to work again
[05:17:50] <CIA-10> 03jmkasunich 07TRUNK * 10emc2/src/hal/drivers/mesa_5i2x/firmware/ (core.mspec core.spec Makefile spec2vhdl.py stepgen.mspec): work in progress
[13:46:09] <CIA-10> 03jmkasunich 07TRUNK * 10emc2/src/hal/utils/comp.g: all comps use 'memset', this header avoids warnings on non-RT builds
[13:54:21] <maddash> :O emc on non-RT kernel?!
[13:54:56] <SWPadnos> simulator mode ...
[13:56:00] <CIA-10> 03cradek 07TRUNK * 10emc2/tests/ccomp/lathe-comp/ (expected test.sh): remove line numbers from output to make diff work
[13:56:01] <CIA-10> 03cradek 07TRUNK * 10emc2/tests/ccomp/mill-g90g91g92/ (expected test.sh): remove line numbers from output to make diff work
[13:56:02] <CIA-10> 03cradek 07TRUNK * 10emc2/tests/ccomp/mill-line-arc-entry/ (expected test.sh): remove line numbers from output to make diff work
[13:56:02] <CIA-10> 03cradek 07TRUNK * 10emc2/tests/ccomp/mill-zchanges/ (expected test.sh): remove line numbers from output to make diff work
[14:04:50] <CIA-10> 03cradek 07TRUNK * 10emc2/tests/ccomp/mill-g90g91g92/ (expected test.ngc): fix missing units specification
[14:17:47] <CIA-10> 03jepler 07TRUNK * 10emc2/src/Makefile: when doing a 'clean', don't re-run configure even if the timestamps tell us to
[15:06:20] <CIA-10> 03cradek 07TRUNK * 10emc2/src/emc/rs274ngc/interp_convert.cc: remove redundant TLO call for g49
[16:16:43] <JymmmEMC> http://origin.mercurynews.com/ci_6035915
[16:22:11] <lerneaen_hydra> anyone know where to find geckos in sweden?
[16:22:20] <lerneaen_hydra> or some other good stepper driver?
[16:24:38] <JymmmEMC> define "find"
[16:25:20] <Unit41> how did that guy with the xubuntu make out with his 64 mb ram ?
[16:25:45] <JymmmEMC> as in fell off a truck and you find them, or find them being in operation and you "borrow" them, or ???
[16:26:33] <Unit41> pet store, deffinatly
[16:27:42] <Unit41> get a couple and feed them fly's
[16:28:00] <Unit41> duno how much gecko's eat though
[16:28:41] <jepler> "Try and offer more smaller prey items, than fewer larger ones. The smaller ones tend to be more nutritious and easier to digest." -- http://www.drgecko.com/dietselection.htm
[16:29:25] <JymmmEMC> lerneaen_hydra: Any other questions?
[16:29:42] <lerneaen_hydra> oh, back now
[16:30:08] <lerneaen_hydra> JymmmEMC: in a retail store or ordered or something :p
[16:30:09] <Dallur> do gecko's make good drivers ?
[16:30:24] <JymmmEMC> I found a great use for paraport external disk enclosures.... rip out the internal 5/12v PS =)
[16:30:24] <lerneaen_hydra> or are there others worth looking intp?
[16:30:35] <lerneaen_hydra> yep, they're really nice :)
[16:31:07] <JymmmEMC> lerneaen_hydra: ebay.se maybe?
[16:31:36] <lerneaen_hydra> the swedish variateys are very. very lacking
[16:31:49] <lerneaen_hydra> you'll just find common consumer products there
[16:31:56] <lerneaen_hydra> which is a pity
[16:32:34] <JymmmEMC> What about denmark or Germnay?
[16:32:59] <lerneaen_hydra> hmm, I'm not sure if the ebay-esque sites they have are any good
[16:33:04] <lerneaen_hydra> shipping may be a pita
[16:33:17] <lerneaen_hydra> hmm..
[16:33:20] <JymmmEMC> where are you in SE? North South?
[16:33:26] <lerneaen_hydra> * lerneaen_hydra pokes anonimas1
[16:33:30] <lerneaen_hydra> south
[16:33:38] <lerneaen_hydra> on the west coast
[16:33:39] <JymmmEMC> lerneaen_hydra: ROAD TRIP!
[16:33:40] <lerneaen_hydra> under norway
[16:33:44] <lerneaen_hydra> hehe :D
[16:34:28] <JymmmEMC> Bought a brand new ext paraport case yesterday for $5 and even came with a DB25 cable.
[16:34:45] <lerneaen_hydra> ... man I wish I still lived in the US sometimes
[16:35:03] <lerneaen_hydra> sweden is such a backwater country
[16:35:12] <Dallur> JymmmEMC: I buy external USB hdd enclosures just to get the 12/5v psu
[16:35:28] <Dallur> JymmEMC: cheaper than buying the psu from a store
[16:36:18] <JymmmEMC> Dallur: Tell me about it... For YEARS I've been looking at products that I think I could strip out for parts.... mostly XMFR's (battery chargers, etc)
[16:36:56] <JymmmEMC> This one is 1A for % & 12 V
[16:36:57] <JymmmEMC> 5
[16:37:12] <JymmmEMC> 1" x 4"
[16:37:59] <JymmmEMC> even has a ferrite bead on it
[16:38:05] <JymmmEMC> (surprisingly)
[16:38:25] <Dallur> JymmEMC: the ones I'm buying are 2A 12 2A 5v
[16:38:40] <Dallur> JymmEMC: I use them to drive AMD Geode based machines
[16:39:11] <JymmmEMC> what are you paying for them?
[16:39:37] <JymmmEMC> and can you get away with 1A@5 and 1A@12
[16:40:24] <Dallur> JymmEMC: probably about 40$ but then again Iceland is expensive as hell so ....
[16:41:57] <JymmmEMC> Dallur: http://www.weirdstuff.com/cgi-bin/item/20033
[16:42:31] <JymmmEMC> If you need alot of them, could but locally, strip out the PS and ship to iceland
[16:42:40] <Dallur> :) thx
[16:42:42] <JymmmEMC> they have 1000 of them
[16:42:58] <Dallur> JymmEMC: this is just for prototype work though, getting them from China for the actual product stuff
[16:43:12] <JymmmEMC> ah gotcha
[16:43:39] <JymmmEMC> do you have an enclosure already.... these might do good for both
[16:44:56] <JymmmEMC> I just got done stripping this one apart... no wire cutters necessary,
[16:46:16] <Dallur> JymmEMC: it's all custom plastic molding for the product :P
[16:46:38] <JymmmEMC> fancy ;)
[16:46:57] <Dallur> JymmEMC: heading out to Taiwan tonight to inspect the factories and present the product at Computex
[16:47:16] <JymmmEMC> good luck
[16:47:28] <JymmmEMC> your project or your emploeers?
[16:47:30] <Dallur> JymmEMC: thx, i'm just a contracted techie though, but I hope it all goes well
[16:47:50] <Dallur> JymmEMC: my company is doing the software side of the project and technical consulting
[16:47:58] <JymmmEMC> ah, ok
[16:50:50] <JymmmEMC> LOL it even has dos/win/os2/NT4 drivers
[16:51:07] <JymmmEMC> win3 and win98
[16:51:45] <JymmmEMC> Gawd I wish I could find a board that had IDE and ethernet on it
[16:56:21] <Dallur> JymmEMC: There are plenty of Geode boards that have that
[16:57:19] <Dallur> JymmEMC: mine has dual nic 2 sata, 2 ide, plenty of usb, genio, parallel serial and more :)
[16:58:45] <JymmmEMC> url to pic?
[17:00:19] <Dallur> just a sec
[17:02:46] <Dallur> http://www.ieiworld.com/files/IEINET/takucho/2006/11/28/17/11/12/WAFER-LX2-front_500.jpg
[17:04:40] <JymmmEMC> apx dimensions?
[17:05:56] <Dallur> Jymmm: sbc format if I remember correctly, just using for development
[17:06:35] <Dallur> 145mm x 102mm
[17:06:38] <JymmmEMC> Dallur: What I was kinda looking for is something the size of a 3.5" HDD
[17:07:08] <Dallur> JymmmEMC: this is just about the same size ad a HDD
[17:07:40] <JymmmEMC> The idea being is that once you plug in the ethernet cable there could be a 2x16 LCD that would disoplay the IP (dhcp) and it's your own lil portable computer
[17:08:07] <JymmmEMC> pocket sized
[17:08:49] <Dallur> JymmmEMC: I see, hmm well the project I'm working on has a motherboard based on the Geode LX and a 7" color monitor and it's $200 retail, uses 12V only psu
[17:09:01] <JymmmEMC> ethernet, power, and maybe serial and/or usb
[17:09:18] <JymmmEMC> Dallur: purpose?
[17:09:42] <Dallur> JymmmEMC: i'm bound by NDA
[17:09:56] <Dallur> JymmEMC: but it's not cnc related or anything like that
[17:10:01] <JymmmEMC> dal hypethical purpose for an imaginary project
[17:10:51] <Dallur> JymmmEMC: but I have thought about making one and using it as an ultra slim laptop style computer but solid state only
[17:11:05] <Dallur> JymmmEMC: since I'm booting off flash and there are no fans
[17:11:40] <JymmmEMC> I use thumbstick now, but I'd link to have something that is VERY easily portable and just plugin and go.
[17:12:06] <JymmmEMC> power and network
[17:12:06] <Dallur> JymmmEMC: the real positive side is that 12 plugs right into your car :)
[17:12:24] <Dallur> JymmmEMC: i'm also booting X in less than 10 sec
[17:12:44] <JymmmEMC> Well, that's easy enough to add in a couple extra connector for power 120/12
[17:13:41] <JymmmEMC> CF ?
[17:14:02] <Dallur> JymmmEMC: it ships with a 100-250v adapter with adapters for every country
[17:14:18] <Dallur> JymmEMC: not on the one i'm working on, uses NAND onboard
[17:14:20] <JymmmEMC> Yeah, I HATE ext PS, just a PITA
[17:14:52] <Dallur> JymmmEMC: ;)
[17:14:55] <JymmmEMC> too many freaking cable and other crap you have to carry around.
[17:15:43] <JymmmEMC> At work, we already have power cords and ethernet cables plugged in and ready to go.
[17:16:33] <Dallur> would be interesting to have a machine with retractable cables built in
[17:16:34] <JymmmEMC> I could do the same at home, then just carry it back and forth and transfer what I need. plugin an ext cd/dvd via USB as needed.
[17:16:56] <Dallur> JymmmEMC: modular computing :)
[17:17:00] <Unit41> SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS
[17:17:14] <JymmmEMC> Dallur: and stick in it a 750GB hdd =)
[17:17:28] <JymmmEMC> Dallur: We have obscene bandwidth at work =)
[17:18:31] <JymmmEMC> You gotta love it when you can dl an iso in 4 minutes =)
[17:19:18] <Dallur> JymmmEMC: yeah, that's university network speed ;)
[17:19:37] <JymmmEMC> Dallur: 10GigE pipes globally =)
[17:19:59] <Dallur> JymmmEMC: where do you work ?
[17:20:12] <JymmmEMC> Dallur: a data center =)
[17:20:26] <Dallur> JymmmEMC: ahh figures ;)
[17:20:34] <JymmmEMC> Dallur: Do you know who Akami is?
[17:20:38] <Dallur> JymmmEMC: yup
[17:20:54] <JymmmEMC> Dallur: Lets put it this way... they buy bandwidth from us =)
[17:22:10] <Dallur> JymmmEMC: nice, I used to do some work for Teleglobe a couple of years ago, also UUnet but nothing recently
[17:22:52] <JymmmEMC> oh man, haven't hear the name UUnet in decades =)
[17:23:32] <Dallur> JymmmEMC: nah they went under pretty spectacularly, I remember the most annoying thing about them is that they had active sound suppression in their work spaces, used to drive me mad
[17:24:10] <JymmmEMC> heh
[17:25:34] <Dallur> JymmmEMC: are you a networking guy then or ?
[17:33:00] <Dallur> JymmmEMC: cisco or hw or co-location
[18:30:15] <JymmmEMC> Dallur: working on my CCNA
[18:30:46] <skunkworks> woo hoo. My goto mount for my telescope is on its way.
[18:50:45] <JymmmEMC> http://www.cnczone.com/forums/showpost.php?p=167950&postcount=30
[18:51:28] <JymmmEMC> http://www.cnczone.com/forums/showthread.php?t=24699
[18:51:31] <JymmmEMC> ~~~~~
[18:52:35] <ds2> Jymmm: what was that thing you were trying to show me the other night at weird stuff? all I got was search expired on the link
[18:52:41] <JymmmEMC> Has anyone put a controller setup inside a PC case with a computer? Very much like sherline does iirc.?
[18:53:12] <JymmmEMC> ds2: Oh, it's a adapter cbale to give you that power connection you were looking for.
[18:53:32] <ds2> let try searching for 'power connection' then
[18:53:53] <lerneaen_hydra> which gecko drive is most commonly used?
[18:53:53] <JymmmEMC> what was the link I gave you?
[18:54:02] <lerneaen_hydra> 202?
[18:54:22] <ds2> something with cgi-bin in there I think
[18:54:39] <JymmmEMC> ds2: the full url
[18:54:51] <ds2> I don't have it anymore...scrolled off
[18:56:16] <ds2> bookmark
[18:56:36] <ds2> what's the thing to ask the bot for the log URL?
[18:56:54] <JymmmEMC> logger_emc: bookmark
[18:56:54] <JymmmEMC> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-06-01.txt
[18:56:59] <ds2> Oh like that
[18:57:57] <ds2> http://www.weirdstuff.com/cgi-bin/item/12067
[18:58:00] <ds2> that was it
[18:58:27] <ds2> looks like it is sold out, not timed out... N/a for $4.95
[18:58:41] <JymmmEMC> ah
[18:59:50] <JymmmEMC> ds2: Well, basically what it did was grab 12V from a drive connector and made it into a the square connector
[19:00:18] <ds2> that'd be perfect
[19:00:28] <ds2> but alas, like all useful things...they are no longer available
[19:03:01] <Guest460> Guest460 is now known as PasstScho
[19:04:41] <JymmmEMC> ds2: http://search.ebay.com/search/search.dll?from=R40&_trksid=m37&satitle=pentium+4+power+adapter&category0=
[19:04:48] <jepler> welcome, PasstScho
[19:06:14] <JymmmEMC> ds2: I suspect that you could hit frys, pixelusa, or centralcomputer and find one.
[19:06:40] <PasstScho> hi all
[19:07:32] <JymmmEMC> ds2: Now you know why I clip off all connectors from PC PS's =)
[19:09:00] <skunkworks> PasstScho: whats up?
[19:09:07] <JymmmEMC> or at least the shells, as I have some pins now
[19:09:12] <PasstScho> mom... brb ;)
[19:09:22] <cradek> haha
[19:09:26] <ChecKeR> ChecKeR is now known as PasstScho
[19:09:29] <PasstScho> :)
[19:09:33] <JymmmEMC> BUSTED! lol
[19:10:16] <Dallur> My mom says I can't play with 2 ton metal robots which spew fire and cut metal, no fair
[19:10:16] <PasstScho> i just installed emc and now i'll have a look at what is possible
[19:11:05] <skunkworks> PasstScho: what do you hope to do with it?
[19:11:18] <ds2> Jymmm: where is pixelusa?
[19:11:30] <PasstScho> i don't hope, i know i'll drive a CNC mill (and a lathe later..) with it :)
[19:11:37] <ds2> don't do ebay so...
[19:12:00] <PasstScho> i already got my steppers running just as a test and i wondered how easy it was
[19:12:22] <PasstScho> atm i am googlin' around for connection of jog wheel and stuff
[19:12:34] <PasstScho> and how to use the rs232 ports my computer has
[19:13:50] <jepler> PasstScho: emc doesn't have much included support for rs232 devices -- there is "serport", which allows the rs232 control lines to be used as digital I/O, but that's it
[19:14:44] <PasstScho> that is already better than nothing
[19:15:24] <jepler> PasstScho: so you've already used emc to spin your steppers?
[19:15:45] <PasstScho> i gues writing an own driver to support... like an AVR on the serial port, with ios,pwms... and other stuff is a hell of a project?
[19:15:50] <PasstScho> jepler: yep, i did
[19:17:10] <jepler> PasstScho: and you have some programming skills, it sounds like?
[19:17:42] <PasstScho> yep, but all my linux-experience base one some hours of using suse...
[19:17:55] <PasstScho> i've grown up with windows ;)
[19:18:06] <jepler> PasstScho: it would certainly be possible to make emc communicate with a microcontroller on the serial port, but I'd consider it an advanced project
[19:18:57] <PasstScho> that is what i fear... i wouldn't want to spend too much time, as far as my other projects aren't finished
[19:18:57] <skunkworks> PasstScho: I am not a linux person - but have grown to like it over the last year or so.
[19:21:42] <PasstScho> so far, i had no problems with linux - and i hope i won't have any ;)
[19:23:04] <rob-h> Hello
[19:23:35] <PasstScho> hi rob-h
[19:24:56] <rob-h> just wundering if there any one who can help me out with a HAL problem for PWM?
[19:25:13] <cradek> don't ask to ask, just ask
[19:26:14] <rob-h> simply, problem is that i am not seeing a PWM signal out on my PIN9, im using the nist-lath.hal as a starting block
[19:26:31] <rob-h> you can see a copy of my hal here, saves sending the file, http://innovative-rc.co.uk/nist-lathe.hal
[19:27:43] <cradek> looks like you have to turn on the spindle and set a nonzero speed to get output - are you doing that? (like S500 M3)
[19:28:32] <rob-h> yes, i have put in S500 M03 and HAL meter says PWM enable is on, put a scope on pin9 and im not seeing a PWM
[19:29:01] <jepler> rob-h: what happens if you use halscope to show parport.0.pin-09-out ?
[19:29:01] <cradek> does halmeter/halscope show the pwmgen output when you look at the output pin?
[19:30:05] <rob-h> which thread should i be picking on hal scope? servo or base
[19:30:09] <jepler> PasstScho: what kind of devices do you want to connect, using the I/O and PWM on the microcontroller?
[19:30:13] <jepler> rob-h: "base"
[19:31:38] <PasstScho> jepler: so far i don't really have a device to connect, but i'd like to learn what is possible and how it would work
[19:31:39] <rob-h> what i get is a pulse now and then, not a PWM
[19:31:55] <PasstScho> maybe connect a button-matrix
[19:32:03] <PasstScho> or some potentiometers
[19:32:15] <PasstScho> so nothing special
[19:32:26] <jepler> PasstScho: I ask because in emc there are two kinds of drivers: realtime and non-realtime. non-realtime drivers are easier, because you can take advantage of linux device drivers; realtime drivers are required for things like limit switches or controlling motors
[19:33:07] <PasstScho> yep, i thought such an different must exist
[19:33:13] <PasstScho> a non-realtime driver would be enought
[19:33:37] <PasstScho> like just sending some "text" via rs232 that the avr understands and executes
[19:34:07] <rob-h> jepler, see here for a sample of what im getting on pin9 http://innovative-rc.co.uk/scope1.JPG
[19:35:19] <PasstScho> atm i read the HAL documentation, seems to be nice things in there
[19:35:21] <jepler> so in HAL it appears there are pulses, but they are far apart and short?
[19:35:43] <jepler> rob-h: so in HAL it appears there are pulses, but they are far apart and short?
[19:35:53] <rob-h> yes correct
[19:36:24] <jepler> rob-h: does that change if you issue a higher S-number, like S5000 M03 ?
[19:37:31] <jepler> PasstScho: there are a number of people on this channel who can help you when you are ready to write your own HAL driver. Just drop by again and ask for help.
[19:37:33] <cradek> the scale is 1400 - it should be about half on at S500 I think
[19:37:34] <rob-h> no they stay same spacing
[19:37:51] <rob-h> at 500 or s1000
[19:37:52] <jepler> If you look at the examples and manual first you'll be in good shape to ask smart questions
[19:38:11] <PasstScho> jepler: that's fine, but i'll still have to read a lot
[19:38:15] <Dallur> l8r, see you guys in 2 weeks
[19:38:28] <jepler> see you Dallur
[19:38:30] <jepler> going on vacation or something?
[19:38:34] <PasstScho> and the rs232 thing isn't my most important project atm
[19:38:38] <Dallur> jepler: computex
[19:38:41] <cradek> rob-h: you're scaling spindle-speed-out by .00089 and then setting pwm so full scale is 1400
[19:39:09] <cradek> so full on would be something like S1500000
[19:39:17] <cradek> (I think)
[19:39:30] <cradek> your scaling is wrong - check your numbers/math
[19:39:59] <jepler> cradek: yeah you may have hit the nail on the head
[19:40:22] <cradek> I bet if you specify a million RPM you'll see better output
[19:40:32] <rob-h> like i said i was trying to use nist exaple to start the ball rolling, in HAL manual i cant track down the scale and offset meanings,
[19:40:36] <jepler> setp scale.0.gain .00089
[19:40:36] <jepler> setp scale.0.offset .07
[19:40:56] <jepler> in that configuration, these two values ^^^ are used to set the scale, assuming that pwmgen.0.scale is left at the default which is 1
[19:41:07] <rob-h> yes you are right, at s100000 now im seeing what should be PWM
[19:41:27] <cradek> ok, it is working right, you just have a scaling problem
[19:42:02] <cradek> if you know what rpm you get when your signal is full on, you can set the scaling correctly
[19:42:30] <rob-h> the lath can do 1400RPM max
[19:43:31] <cradek> try not setting pwmgen.0.scale, and set that scale.0 block to a scale of 1/1400 (.0007)
[19:43:50] <cradek> or, don't use the scale block, and set pwmgen.0.scale to 1400 like you have it
[19:43:57] <cradek> just don't do both
[19:44:57] <cradek> ah, someone used scale because pwmgen doesn't/didn't have an "offset"
[19:45:11] <jepler> cradek: yes, that's the reason
[19:45:25] <jepler> (and "someone" was me)
[19:45:30] <cradek> you may have to monkey with offset in order to get your low speeds right (like S50)
[19:47:19] <cradek> if it's fairly linear response, you should be able to measure two speeds (near the max and min speeds you will use) and calculate slope/offset to plug into the scale block
[19:48:07] <skunkworks> * skunkworks thinks rob is gigling like a school girl watching emc control the spindle.
[19:49:12] <skunkworks> giggling
[19:49:24] <rob-h> lol
[19:49:45] <rob-h> its all clear to me now, and yea, s1000 is showing up nicly on scop, nice PWM
[19:50:25] <cradek> I love to hear things like "it's all clear to me now"
[19:51:07] <rob-h> now iv just got it get it to keep putting PWM out with spindle reverse
[19:51:11] <skunkworks> is your board outputting a voltage now?
[19:51:57] <cradek> rob-h: hint: abs block
[19:52:05] <rob-h> hal scope is showing PWM, i will go back to the lath in abit and check real world values, im on remote right now onto EMC machine
[19:53:00] <rob-h> ABS?
[19:53:14] <cradek> abs - Compute the absolute value and sign of the input signal
[19:54:07] <cradek> just curious, what do you do on the lathe with the spindle going backward?
[19:54:24] <rob-h> left hand threading
[19:54:30] <rob-h> right hand threading etc
[19:54:57] <rob-h> its nice and handy to have
[19:54:58] <cradek> with cnc you can just move the tool the other way...
[19:55:34] <cradek> wait - wouldn't you have to turn the tool upside-down? I guess I don't follow
[19:55:56] <skunkworks> or put the tool on the other side of the part
[19:56:03] <cradek> true
[19:57:02] <rob-h> or use backwards for tapping also
[19:57:12] <skunkworks> that is how we do left-handed threads on the old manual lathe
[19:57:17] <cradek> yeah I can sure see that
[19:57:57] <cradek> skunkworks: really? can't you just add another change gear to get it to move the other way?
[19:58:02] <rob-h> we use tapping on our big CNC hardings lathes
[19:58:34] <skunkworks> it is hard to start the treads going out from the spindle - as it is ending the treads going into the spindle
[19:58:45] <cradek> rob-h: I hope to do rigid tapping soon - but since I don't have a motorized tailstock I won't be able to use it myself
[19:59:47] <lerneaen_hydra> does anyone have the link to the eagle to gcode app?
[19:59:47] <cradek> skunkworks: ok I see, you don't want to cut a relief in the middle of the part so you can sit and wait there to start a pass
[19:59:49] <skunkworks> cradek: is it is just one more motor ;)
[19:59:58] <skunkworks> cradek: exactly
[20:00:29] <jepler> lerneaen_hydra: http://unpy.net/cgi-bin/viewcvs.cgi/eagle/ulp/
[20:00:33] <cradek> what a pain - cnc threading is so much better
[20:00:50] <lerneaen_hydra> ah jepler had it :)
[20:01:10] <jepler> lerneaen_hydra: you may want to get rev1.5 of gcode.ulp (http://unpy.net/cgi-bin/viewcvs.cgi/~checkout~/eagle/ulp/gcode.ulp?rev=1.5;content-type=text%2Fplain) as 1.6 has some stuff very specific to chris's setup
[20:01:18] <jepler> (tool length sensing, which is very cool)
[20:01:47] <lerneaen_hydra> ah. so it autoprobes and finds length?
[20:02:17] <jepler> yes, there's a switch at a defined location on the table
[20:02:32] <jepler> I don't think chris has put anything online about it yet ..
[20:02:43] <lerneaen_hydra> cool
[20:03:12] <lerneaen_hydra> jepler: is the drill ulp also modified?
[20:03:24] <rob-h> im going to go play with the lath, thank alot you for your quick help
[20:03:32] <cradek> rob-h: come back anytime
[20:04:08] <jepler> lerneaen_hydra: gcode.ulp does trace isolation and drilling both
[20:04:32] <cradek> ... and copper clearing and board cutout
[20:04:38] <lerneaen_hydra> oh so if I only want drilling I can disable trace milling?
[20:04:48] <cradek> yes
[20:04:56] <cradek> isolation passes = 0
[20:05:02] <cradek> etc
[20:05:11] <lerneaen_hydra> ah right
[20:05:42] <lerneaen_hydra> how's the "stability" of the conversion? has is ever done anything stupid? like g0 into the table or something?
[20:05:57] <cradek> conversion?
[20:06:16] <lerneaen_hydra> err, code generation
[20:06:27] <lerneaen_hydra> * lerneaen_hydra is rather tired and incoherent
[20:07:12] <cradek> This program is distributed in the hope that it will be useful,
[20:07:12] <cradek> but WITHOUT ANY WARRANTY; without even the implied warranty of
[20:07:13] <cradek> MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
[20:07:14] <CIA-10> 03jepler 07TRUNK * 10emc2/src/hal/components/pwmgen.c: the General Reference says that offset is added, not subtracted
[20:07:32] <cradek> lerneaen_hydra: "works great for me" is all I can say
[20:07:40] <skunkworks> I have only started using recently - but all my test boards (read mistakes) have cut correctly
[20:07:52] <lerneaen_hydra> cradek: ah, perfect
[20:08:13] <cradek> a long hard look at any generated gcode is always smart
[20:08:18] <lerneaen_hydra> yeah
[20:08:18] <cradek> AXIS shows obvious problems pretty well
[20:08:23] <lerneaen_hydra> that's very true
[20:08:35] <jepler> AXIS shows less well "the spindle isn't turned on" which is a mistake I made at least once
[20:08:44] <lerneaen_hydra> hehe
[20:08:45] <lerneaen_hydra> true
[20:09:22] <jepler> maybe it should squawk if any "feed" move is done without spindle on
[20:09:31] <jepler> that's usually a mistake
[20:11:10] <cradek> jepler: there's a lot of pain down that road isn't there?
[20:11:20] <jepler> cradek: yes
[20:11:28] <jepler> "axis didn't detect that my gcode did xxx"
[20:11:32] <cradek> ok, just checking
[20:11:56] <jepler> "I am doing CNC spray-painting and there is no spindle, dorkwad"
[20:12:16] <cradek> ok, just checking
[20:12:28] <cradek> dorkwad?
[20:12:54] <cradek> obligatory http://www.theonion.com/content/node/29317
[20:15:09] <jepler> haha http://www.theonion.com/content/opinion/i_believe_in_evolution_except
[20:15:23] <jepler> "So, if you follow my reasoning to its logical end, the only sound conclusion is that, at some point, God paused evolution and stepped in, made a few modifications, and boom! Pterosaurs. There is simply no way evolution alone could be responsible for the giant leap between archosaurs and other, different archosaurs with better developed hip joints and slightly differently shaped teeth."
[20:15:59] <cradek> that's a truly nuanced position I hadn't considered before
[20:16:07] <maddash> in mode x, i noticed that both parport.*.pin-[1,14,16,17]-in/out exist. shouldn't there only be -in, and not -out?
[20:16:24] <cradek> maddash: that depends who you ask, and what version of emc you're using
[20:17:00] <cradek> maddash: since those are open collector, it does, in my opinion, make some sense to have both in and out pins.
[20:17:18] <jepler> on the other hand, it's confusing because most users of "x-mode" simply want more inputs
[20:17:31] <jepler> and don't want to have to configure the -out ports properly (logic TRUE in HAL)
[20:17:38] <cradek> maddash: yes, if you want to read them as inputs, you have to set the output the right way
[20:18:35] <cradek> the question is really whether the added flexibility of open collector is worth enough to require the user to understand open collector
[20:19:06] <maddash> cradek/jepler: how do I "set the output"? and what is the wrong way of doing so?
[20:19:16] <jepler> halcmd "setp"
[20:19:38] <maddash> man halcmd
[20:19:43] <maddash> whoops
[20:20:09] <jepler> it's also the case that "x-mode" simply does not work on some parports
[20:22:31] <jepler> the documentation suggests a method for determining if x-mode works on your port: http://linuxcnc.org/docs/2.1/html/hal/drivers/drivers.html#foot616
[20:22:58] <jepler> To determine whether your port has ``open collector'' pins, load hal_parport in ``x'' mode, output a HIGH value on the pin. HAL should read the pin as TRUE. Next, insert a 470$ \Omega$ resistor from one of the control pins to GND.
[20:23:05] <jepler> If the resulting voltage on the control pin is close to 0V, and HAL now reads the pin as FALSE, then you have an OC port. If the resulting voltage is far from 0V, or HAL does not read the pin as FALSE, then your port cannot be used in ``x'' mode.
[20:23:08] <jepler> The external hardware that drives the control pins should also use open collector gates (e.g., 74LS05). Generally, the -out HAL pins should be set to TRUE when the physical pin is being used as an input.
[20:23:12] <jepler> On some machines, BIOS settings may affect whether ``x'' mode can be used. ``SPP'' mode is most most likely to work.
[20:23:13] <maddash> never saw that - thanks
[20:23:50] <jepler> if you're unfamiliar with what an "open collector" I/O pin is, google is your friend
[20:27:32] <maddash> brb, checking my bios
[20:38:20] <PasstScho> someone told me there is an serport driver available, but i don't find a hal_serport entry in the user manual - did you mean the linux-serport-driver?
[20:41:27] <SWPadnos> it may just be called "serport"
[20:41:34] <SWPadnos> not hal_serport
[20:41:41] <cradek> man serport
[20:41:49] <jepler> http://linuxcnc.org/docs/2.1/html/man/man9/serport.9.html
[20:41:52] <SWPadnos> yep
[20:43:57] <PasstScho> hm, ok - i'll have a look at it and try later.... now i read on the documentation...
[21:01:51] <maddash> feck, either my parport is damaged, or it isn't "open controlled." damn.
[21:07:32] <CIA-10> 03jepler 07TRUNK * 10emc2/src/hal/halmodule.cc: attempt to improve support for python2.5 + 64 bit machines
[21:31:40] <maddash_> hm, is there a doc that covers serport?
[21:33:09] <PasstScho> lol
[21:33:36] <PasstScho> dunno, but i guess it works like parport, right?
[21:34:05] <PasstScho> just connect Xdir & Xstep to it and steppers or whatever should run...?
[21:36:18] <mschuhmacher> maddash_ :the doc for serport is man serport or http://linuxcnc.org/docs/2.1/html/man/man9/serport.9.html
[21:37:33] <maddash_> mschuhmacher: cool, thanks
[21:40:44] <maddash_> PasstScho: seems like you're right, based on the doc mschuhmacher posted
[21:41:14] <PasstScho> cool hehe
[21:51:41] <jepler> PasstScho: rs232 voltages are different than regular ttl logic voltages
[21:51:42] <jepler> exercise caution
[21:51:59] <PasstScho> uh, right i almost forget it...
[21:52:16] <PasstScho> usually i only use the rs232 with AVR and max232 so i never really had to care about it
[21:52:37] <jepler> yes, you could use max232 for this purpose as well
[21:52:48] <PasstScho> maddash_: you shouldn't forget that too
[21:53:37] <PasstScho> hehe mschuhmacher
[21:53:58] <mschuhmacher> http://blog.der-link.de/archives/1305-Passt-scho.html
[21:55:20] <maddash_> ^^^ that's what happens when your estop is broken
[21:56:51] <CIA-10> 03jepler 07TRUNK * 10emc2/src/rtapi/rtai_rtapi.c:
[21:56:51] <CIA-10> * keep track of actual timer counts from rtai. it turns out
[21:56:51] <CIA-10> nano2count(count2nano(x)) != x under some circumstances, which led to rtai
[21:56:51] <CIA-10> erroneously detecting overruns, which lead to a thread occasionally being
[21:56:51] <CIA-10> executed back-to-back.
[21:56:52] <CIA-10> * check the return value from rt_task_wait_period() and log the first ten overruns in dmesg.
[21:59:17] <xantipe> hello! I am very new to EMC. I want to build my own CNC mill and control it with EMC. Can somebody, please, give me some information about what is required?
[22:00:11] <PasstScho> hum, a mill, some time, a pc with ~1ghz and 256mb ram + emc2 :)
[22:00:32] <CIA-10> 03compile-farm 07Ubuntu 5.10 (breezy) realtime (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot2_log.txt
[22:00:33] <PasstScho> then some servo or stepper control and you are ready to start
[22:00:57] <PasstScho> CIA-10: Ubuntu 5.10? why you use that old one?
[22:01:33] <xantipe> Ok. That's clear. I mean the hardware part. Does EMC do the stepping on a parallel port (2 pins/motor), or does it need some special hardware?)
[22:02:19] <xantipe> PasstScho: Is there a very detailled PDF available for download?
[22:02:52] <PasstScho> i only know a very detailed pdf about emc
[22:03:17] <PasstScho> starting to tell about mills and stepper/servo drives wouldn't fit into a pdf ;)
[22:03:31] <xantipe> Sounds good. Does it also explain the interfacing to the steppers a bit?
[22:03:47] <mschuhmacher> xantipe: http://www.linuxcnc.org/docs/EMC2_Integrator_Manual.pdf
[22:03:51] <PasstScho> what would you like to mill? which material? which size? how much money do you have?
[22:05:01] <xantipe> I want to mill plastic parts (enclosures), PCBs, and such stuff...
[22:05:31] <xantipe> Money: The cheaper the better! A4 size (working area) is minimum.
[22:05:50] <PasstScho> cheap projects are always... hum... hard to say
[22:06:08] <PasstScho> you can take what you get and make a mill out of it
[22:06:47] <PasstScho> i would say you should google a bit to see different types of machines...
[22:06:50] <PasstScho> there is too much to tell
[22:07:10] <PasstScho> you should enter some forums and read and stuff
[22:07:12] <xantipe> I already built a stepper driver board which can operate 3 steppers with 0.75A/coil. It needs 2 bits per stepper for control. Was originally thought to be used on a parallel port...
[22:07:41] <PasstScho> 0.75A isn't the world, but it should work fine with emc :D
[22:08:21] <xantipe> What I couldn't find any info about how EMC controls the mill-electronics.
[22:08:58] <PasstScho> it outputs step/direction signals over the LPT port (or some other ways...)
[22:09:17] <PasstScho> just connect your two wires or bits as you called them to your lpt and it should run
[22:09:24] <PasstScho> i did the same today - worked fine
[22:09:56] <xantipe> Oh, oh... So it doesn't send 00 01 11 10 .... out for every stepper to control the half bridge?
[22:10:49] <PasstScho> hm, i bet it could, but that is not the best way
[22:11:08] <xantipe> Why?
[22:11:25] <PasstScho> cuz i wouldn't know how it works then... ;)
[22:11:43] <xantipe> ? ;-D
[22:12:10] <PasstScho> today i installed emc - the first time... so i guess you aren't really talking to the right person ;)
[22:12:24] <PasstScho> have a look at a L297-Like IC.... those can be easily connected to emc
[22:12:25] <xantipe> Oh! Also new with it?
[22:12:30] <PasstScho> yep
[22:12:33] <PasstScho> ;)
[22:12:55] <PasstScho> atm i read the manual... and read... and read... and read... and sleep... huuuuh and read...
[22:13:14] <xantipe> The manual I was pointed to?
[22:13:31] <PasstScho> EMC2_User_Manual.pdf
[22:13:32] <mschuhmacher> xantipe: take a look at chapter 14 of the Integrator Manual
[22:13:44] <mschuhmacher> or the user manual
[22:13:54] <mschuhmacher> http://www.linuxcnc.org/docs/EMC2_Integrator_Manual.pdf
[22:14:33] <mschuhmacher> Stepgen
[22:14:41] <xantipe> mschuhmacher: I downloaded it. Will read it tomorrow. Where is the User manual?
[22:15:06] <xantipe> Is something in >German< available, too?
[22:15:33] <mschuhmacher> The User manual is basically the same as the Integrator manual, but with less details
[22:15:56] <xantipe> Ok, so I will read the integrator manual.
[22:16:31] <xantipe> Only Q of interest: Do you know WinPCNC?
[22:17:26] <xantipe> I borrowed a CNC mill from a friend. Works nice, and understands .PLT files. Does EMC also understand them?
[22:17:50] <PasstScho> xantipe: german - cncecke.de ;)
[22:18:11] <xantipe> PasstScho: ... for LINUX?
[22:18:36] <PasstScho> for everything that has to do with cnc
[22:18:46] <xantipe> (Only asked for german docs...) ;)
[22:18:54] <PasstScho> i can remember some german emc wiki link hmmmmmmmm
[22:19:27] <mschuhmacher> there isn´t much in german available
[22:19:29] <PasstScho> http://linuxwiki.de/EMC/EMC2_Installation
[22:19:32] <xantipe> Wahh - Please, no Wikis. I prefer something to print and read then.
[22:19:51] <PasstScho> hum, that is bad...
[22:20:03] <xantipe> (get a stiff neck from reading from CRT)
[22:21:11] <PasstScho> i didn't really search for such a documentation and don't know if one is out there
[22:21:24] <PasstScho> german AND printable is a bit too much i guess ;)
[22:21:42] <mschuhmacher> but most User Interfaces have german translations
[22:21:49] <skunkworksemc> wondering - if you start at a certain line number - It would not read any of the variables defined in the beginning of the program - will it?
[22:21:56] <xantipe> Anyway, thanks for your kindly help. I will read now the manual and hope to know then what I must do. I have an option for a dead CNC mill (electronics dead but mechanics excellent), so this will be my target.
[22:22:21] <maddash_> wtf? parport.0.pin-14-in works. but pin-1 doesn't
[22:23:10] <maddash_> meh, i have mode x after all.
[22:26:56] <maddash_> this is weird -- while testing mode x, pin-[1,14]-in give the correct readings when I alternately connect and disconnect the pins from the ground (pin 18-25), but pin-[16,17]-in don't. wtf?
[22:28:12] <maddash_> jepler: ^^
[22:28:46] <jepler> maddash_: no idea -- I don't have experience with enough parallel ports to know what is typical and what is atypical
[22:32:42] <Ziegler> who maintains the joomla based website?
[22:33:05] <Ziegler> http://www.linuxcnc.org/
[22:34:20] <skunkworksemc> alex_joni: does it iirc
[22:35:11] <Ziegler> alex_joni: how do you like it... compared to some of the other content managements out there?
[22:35:30] <PasstScho> is there any function on parport1 pin??
[22:36:26] <Ziegler> PasstScho: do you have it set to do something?
[22:36:39] <PasstScho> (00:37:03) • maddash_ has left IRC () <== that guy had problems with it
[22:36:46] <PasstScho> (00:28:43) (maddash_) wtf? parport.0.pin-14-in works. but pin-1 doesn't
[22:36:56] <PasstScho> i am not sure if parport starts at 1 or at 2
[22:37:22] <Ziegler> ahhI see
[22:37:43] <Ziegler> I have signal going to pin one in a number of my setups
[22:38:04] <PasstScho> ah ok
[22:38:42] <Ziegler> I think the example stepper hal configs have it set that way too
[22:39:01] <Ziegler> * Ziegler looking
[22:39:42] <PasstScho> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?action=browse&diff=1&id=Xylotex_With_EMC2
[22:39:44] <PasstScho> starting with 2...
[22:40:11] <CIA-10> 03jepler 07TRUNK * 10emc2/src/rtapi/rtai_rtapi.c: further improve rt_task_wait_period error detection
[22:40:13] <Ziegler> yeah the default
[22:40:33] <Ziegler> however... I had a couple home brew circuits that used 1 iirc
[22:40:50] <PasstScho> hmm, 1 was CLOCK, might that be?
[22:40:56] <Ziegler> its can be
[22:41:23] <Ziegler> you could easily say: linksp Xstep => parport.0.pin-01-out
[22:41:30] <Ziegler> or something similar
[22:41:51] <PasstScho> ok
[22:42:21] <Ziegler> linkps parport.0.pin-01-out XbitA
[22:42:22] <Ziegler> linkps parport.0.pin-02-out XbitB
[22:42:22] <Ziegler> linkps parport.0.pin-03-out XbitC
[22:42:35] <Ziegler> (for example)
[22:43:37] <PasstScho> my steppers are running... his made problems ;D
[22:43:51] <Ziegler> xylotec board?
[22:44:10] <PasstScho> hm?
[22:44:33] <Ziegler> What board was he using?
[22:44:40] <PasstScho> dunno
[22:44:46] <Ziegler> hehe
[22:46:42] <CIA-10> 03compile-farm 07Ubuntu 5.10 (breezy) realtime (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot2_log.txt
[22:54:46] <CIA-10> 03jepler 07TRUNK * 10emc2/src/rtapi/rtai_rtapi.c: apparently these defines did not exist on older rtai versions
[22:55:32] <skunkworksemc> wow - for 80 minutes a side (5ipm) milling this circuit board... I need a more stable machine.
[22:57:37] <Ziegler> hows it look
[22:58:16] <Ziegler> I went looking for 2.5" aluminum square tubing today to create stepper mounts
[22:58:20] <Ziegler> no luck yet
[22:58:21] <skunkworksemc> ony 1/2 done - so far so good
[23:03:43] <CIA-10> 03compile-farm 07Ubuntu 5.10 (breezy) realtime (2.6.12-magma) * 10emc2head/: build PASSED
[23:13:33] <Ziegler> hey... does accel settings affect "de-accel"
[23:14:04] <Ziegler> if I am at 0,0 and I g1 x10
[23:14:21] <Ziegler> do it accelerate, and then start to de-accelerate when it gets close to 10?
[23:18:07] <mschuhmacher> Ziegler: It should
[23:20:10] <skunkworksemc> oops
[23:20:10] <mschuhmacher> Ziegler: The acceleration is proportional to the force and therefore the same force would apply for acceleration and de-acceleration except you use a brake
[23:31:31] <Ziegler> ok good
[23:31:52] <Ziegler> units in inches / sec / sec?