#emc | Logs for 2007-07-17

[11:03:24] <Dallur> * Dallur is idling
[11:04:12] <alex_joni> hi Dallur
[11:06:20] <Dallur> hey alex
[11:07:31] <Skullworks-PGAB> alex - I watched zeitgeist its "old news"
[11:07:48] <alex_joni> Skullworks-PGAB: I didn't :P
[11:07:52] <Dallur> hmm I think I didn't mention before but I finished interfacing with a serial port via python / userspace
[11:08:14] <alex_joni> cool..
[11:08:18] <alex_joni> * alex_joni goes back to his vacation
[11:08:18] <Dallur> easier than drinking water btw
[11:08:29] <alex_joni> see you later guys
[11:08:34] <Dallur> alex_joni: :) later
[11:09:31] <Skullworks-PGAB> other than the first section - its largely dealing with U.S. Gov history and stuff under the carpet.
[12:09:04] <sebjames> Hi Guys. Well, I set motion.feedhold high, but I seem to have a slight residual output through to my servos. I wonder why that is...
[12:09:46] <sebjames> Also, I'd like to set DISPLAY in emc.ini from a cmd line parameter. I guess I just need to edit /usr/bin/emc suitably
[12:16:15] <cradek> sebjames: feedhold stops the commanded position from moving. It does not turn off power to the servos (because they can't hold that position then)
[12:26:03] <Guest683> Guest683 is now known as skunkworks_
[12:56:57] <hcseb> hcseb is now known as sebjames
[12:57:27] <sebjames> We're churning out the plates here.. It's great
[12:57:43] <sebjames> I've got some video, which I'll post later on
[12:58:27] <archivist> oo cant wait to watch
[12:58:27] <skunkworks_> Great - nice job.
[13:00:35] <skunkworks_> can't wait to see the videos.
[13:08:52] <anonimasu> awe
[13:08:57] <anonimasu> sebjames: residual?
[13:09:11] <anonimasu> sebjames: you mran they still hunt for position? or drift?
[13:25:50] <sebjames> anonimasu: Drift
[13:25:57] <sebjames> Very slowly
[14:18:30] <xemet> hello
[14:18:56] <xemet> how much is the max number of steps per second I could generate with a P4 1.5 GHz?
[14:19:47] <skunkworks_> your not really going to know until you try it.
[14:19:52] <skunkworks_> you're
[14:19:59] <xemet> I've a friend that hasabout...
[14:20:02] <xemet> sorry
[14:20:12] <xemet> I've a frind that uses microstep
[14:20:27] <xemet> and has a problem with EMC2,
[14:20:49] <xemet> I've tried his configuration and I can generate max about 35000 steps per second
[14:21:11] <xemet> so the max speed, with hiss value of INPUT_SCALE is of about 450 mm/min
[14:21:31] <xemet> he says that using Mach he was going faster...is it possible??
[14:21:40] <xemet> Mach3 on windows...
[14:23:51] <maddash_> have you tried lowering your base-thread timeout?
[14:24:00] <maddash_> maddash_ is now known as sudo_maddash
[14:24:11] <skunkworks_> have you guys been lowering the base_period setting? It is defaulted to 50000. you can try lowering it until the gui becomes sluggish.
[14:24:18] <skunkworks_> what maddash said ;)
[14:24:22] <sudo_maddash> heh.
[14:25:07] <xemet> I've it to 15000
[14:25:26] <xemet> I've tried 5000 :) but everything freezes...
[14:25:32] <skunkworks_> a base period of 20000 (20us) gives you a 10000 steps per second.
[14:25:48] <xemet> the input_scale value is about 3000!
[14:26:10] <xemet> 3000 step per mm
[14:26:20] <sudo_maddash> er.
[14:26:21] <sudo_maddash> wow?
[14:26:23] <xemet> it is very high I think...
[14:26:28] <skunkworks_> that should give 166k pulses per second.
[14:26:31] <sudo_maddash> the standard is 1000steps/inch
[14:26:40] <cradek> that's too high for software step generation
[14:26:44] <cradek> can he turn down the microstepping?
[14:26:58] <sudo_maddash> eheh "turn down"
[14:27:34] <cradek> sure a lot of drives let you choose the microsteps
[14:27:34] <sudo_maddash> he could always grab an fpga board and move base-thread to hardware
[14:27:55] <sudo_maddash> cradek: no, I was remarking at the irony
[14:28:04] <xemet> yes I think he can turn down the micro step and I will suggest it to him
[14:28:25] <xemet> remain my curiosity to know hom much fast was he going with Mach3!!
[14:29:06] <xemet> I've read that the Pluto now has a step generation firmware
[14:29:11] <xemet> is it ready to be used?
[14:29:24] <xemet> maybe a day I can try, I've a Pluto
[14:29:28] <cradek> it is probably ready to be *tested*
[14:29:47] <xemet> well...I like to test out new stuff!
[14:30:08] <cradek> I think it's in cvs
[14:30:59] <sudo_maddash> I get, "Error opening /dev/rtf3" everytime I try running an rtai latency test. `mknod /dev/rtf3 c 150 3` solvees this, but /dev/rtf3 disappears during the next reboot. How can I make it stay?
[14:31:05] <xemet> ok, and how much steps per second could it generate?
[14:31:26] <xemet> ah! how can I run the rtai latency test?
[14:31:27] <cradek> sudo_maddash: there's no good way to do that
[14:32:14] <sudo_maddash> cradek: so what's going on? my other rtai/emc2 setup never had this problem
[14:32:33] <cradek> xemet: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TroubleShooting#RTAI_Latency_test
[14:32:59] <cradek> sudo_maddash: it's a udev/rtai issue - the easy way to deal with that is to create it when you need it. you could put it in rc.local or something if you really want.
[14:33:04] <tomp> maybe a shell script to run the latency test with a peek at the existance of /dev/rtf3 before tring to create it
[14:33:30] <cradek> xemet: have you seen http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TweakingSoftwareStepGeneration
[14:36:32] <skunkworks_> I think the max that mach can do now is 100k steps'
[14:37:01] <xemet> 100000
[14:37:17] <xemet> so more than emc2?
[14:37:19] <skunkworks_> yes 100000 steps per second.
[14:37:29] <skunkworks_> depends on the computer.
[14:38:42] <cradek> I'd be surprised to see 100kHz steps out of EMC2 (but, I'd also be surprised to see it out of mach)
[14:38:56] <cradek> (I have never used mach)
[14:39:07] <sudo_maddash> "RTD| -818| -834| 1070| 3311| 13219| 0" -- is it just me, or are these latencies horrible?
[14:39:18] <cradek> it's just you
[14:39:44] <sudo_maddash> the test was done on an idle cpu...
[14:41:29] <cradek> 13000 is decent. to get reliably under 10000 you need SMP with CPU isolation.
[14:41:46] <cradek> many machines are around 20000
[14:41:53] <cradek> (yours probably will be if you load it some)
[14:42:33] <sudo_maddash> ovlmax was 25000 while compiling a kernel.
[14:44:00] <sudo_maddash> what are the timings for your dual proc machine?
[14:44:37] <xemet> my max was about 12000
[14:44:49] <cradek> with isolation I think it was about 5000
[14:45:36] <xemet> what's isolation?
[14:45:50] <xemet> how to do that?
[14:46:43] <cradek> it's a special rtai setup
[14:47:58] <sudo_maddash> mind telling me how to do it? i've got a second p3 cpu doing nothing
[14:49:01] <cradek> sudo_maddash: build kernel and rtai for smp, boot with isolcups=1, rebuild emc2
[14:49:06] <cradek> isolcpus
[14:51:32] <sudo_maddash> awesome.
[14:51:34] <sudo_maddash> thanks, cradek
[14:53:01] <sudo_maddash> brb, reboot
[15:00:04] <alex_joni> hi
[15:08:41] <Dallur> alex_joni: go back on vacation ¨
[15:08:50] <Dallur> alex_joni: that's an order :)
[15:09:12] <alex_joni> * alex_joni is quite bad at following orders
[15:09:39] <Dallur> alex_joni: hmm that works out pretty good, i'm bad at giving them
[15:12:11] <maddash> shouldn't fixed-point be enough for trajectory calcs?
[15:22:28] <maddash> I don't get it -- why aren't these /dev/rt* devices automatically created?
[15:22:49] <cradek> because rtai didn't (doesn't?) support udev very well yet
[15:23:02] <cradek> it might be better in the very latest rtai versions
[15:23:34] <maddash> this is rtai-3.5
[15:23:55] <cradek> maddash: I've put my play kernels at http://linuxcnc.org/experimental/
[15:24:07] <cradek> the works ok except there's no bootsplash (which I don't care about)
[15:24:15] <maddash> the troubleshooting page is very confusing -- is /dev/rtai_shm supposed to a symlink or its own device?
[15:24:20] <cradek> I'm uploading the source for it now
[15:25:01] <cradek> feel free to try these but be aware they come with no support :-)
[15:26:07] <maddash> heh
[15:28:06] <alex_joni> maddash: some versions of rtai insist of creating /dev/RTAI_SHM so /dev/rtai_shm is usually a symlink to the first one
[15:29:56] <maddash> alex_joni: indeed...
[15:30:04] <maddash> finally fixed the problem by adding a boot script.
[15:38:08] <alex_joni> bbl
[16:36:31] <anonimasu> chi
[16:38:02] <anonimasu> sebperhaps you need to adjust the offset..
[16:38:09] <anonimasu> sebjames
[16:38:28] <anonimasu> /whosis awallin
[16:59:21] <maddash> er
[16:59:25] <maddash> wtf?
[17:46:40] <CIA-8> 03jmkasunich 07TRUNK * 10emc2/src/hal/classicladder/module_hal.c: fixed a couple minor warnings
[17:46:40] <CIA-8> 03jmkasunich 07TRUNK * 10emc2/src/hal/drivers/pluto_common.h: fixed a couple minor warnings
[18:21:20] <skunkworks_> * skunkworks_ tripped a breaker.
[18:21:55] <skunkworks_> * skunkworks_ found out the ups isn't working anymore.
[18:24:04] <alex_joni> ouch
[18:25:06] <skunkworks_> ah - its xp. it recovered ;)\
[18:25:40] <alex_joni> * alex_joni needs an idea for carving a walking stick
[18:32:06] <bill2or3> something with metal spikes.
[18:35:12] <skunkworks_> alex_joni: by hand or machine?
[18:35:42] <alex_joni> hand
[18:36:10] <alex_joni> and I'm not really skilled .. so something simple :P
[18:36:16] <alex_joni> maybe something like this: http://images.statelibrary.tas.gov.au/tasimg/jun2001/normal/AUTAS001127117463.jpg
[18:36:29] <skunkworks_> someething that looks like a chain would be cool
[18:36:43] <skunkworks_> Neat
[18:37:30] <skunkworks_> could carve it so it looked like a circuit board wrapped around the stick...
[18:37:49] <skunkworks_> * skunkworks_ is just being goofy
[18:39:00] <martin_lundstrom> Hello people
[18:39:12] <skunkworks_> Hi
[18:39:20] <martin_lundstrom> Dallur: are you there?
[18:39:30] <Dallur> yup i'm here
[18:39:37] <Dallur> but i'm heading home in a couple of minutes
[18:39:39] <martin_lundstrom> how are you?
[18:40:03] <martin_lundstrom> ok, maybe we continue when you get home
[18:41:21] <Dallur> martin: talk to you in a bit
[19:20:42] <`\x90> `\x90 is now known as \x90
[19:52:15] <Jimmybondi> hi
[19:52:36] <Jimmybondi> anybodi out there?
[19:53:06] <JymmmEMC> no. leave a msd at the beeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeep
[19:53:22] <Jimmybondi> .lol
[19:54:14] <skunkworks_> Jimmybondi: Hi
[19:54:24] <Jimmybondi> can anyone gimme a hint where to search how to add a 2nd par-port within emc?
[19:54:39] <cradek> wiki.linuxcnc.org
[19:54:41] <JymmmEMC> ini
[19:55:14] <Jimmybondi> of course - i added port at addf
[19:55:20] <Jimmybondi> ups
[19:55:22] <Jimmybondi> not
[19:55:36] <Jimmybondi> hal_parport cfg
[19:55:58] <Jimmybondi> when trying to add addf i receive only errors
[19:56:55] <cradek> ok, then tell what you did, and tell what error you got
[19:57:32] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?NetMos
[19:57:38] <cradek> ^^ some nice information here
[19:58:00] <Jimmybondi> i added the adress of second LPT (loadrt ... cfg="0x0378 0x0278"
[19:58:15] <Jimmybondi> one moment - lets me take a look
[19:58:22] <Jimmybondi> -s
[19:58:48] <cradek> unless both your parports are on ISA those are unlikely to be the correct addresses
[19:59:01] <cradek> PCI cards always end up somewhere else
[20:00:26] <Jimmybondi> ok - when adding only port and start emc it works
[20:01:09] <Jimmybondi> but when adding addf parport.1.read (write) ... i got error
[20:01:43] <cradek> ok again, tell us the line you added, and the error
[20:01:44] <Jimmybondi> base adress should be ok - both are ISA and work well among DOS
[20:02:13] <cradek> ok
[20:02:19] <Jimmybondi> loadrt hal_parport cfg="0x0378 0x0278"
[20:02:38] <Jimmybondi> addf parport.1.read base-thread 1
[20:02:50] <Jimmybondi> addf parport.1.write base-thread -1
[20:02:54] <Jimmybondi> thats all
[20:03:11] <Jimmybondi> and the errors:
[20:03:11] <cradek> and the error?
[20:03:39] <Jimmybondi> can not find -sec HAL -var HALUI -num 1
[20:06:14] <cradek> that's not an error, it's a normal warning
[20:06:55] <Jimmybondi> ok - but i'm not able to start emc
[20:07:12] <cradek> there must be another error - paste the entire output on http://pastebin.ca please
[20:07:47] <Jimmybondi> one moment - its another PC
[20:08:14] <cradek> ok
[20:09:37] <Jimmybondi> it's done
[20:10:56] <skunkworks_> hmm digikey just informed me that the optical isolators that I am using in my h-bridge are going to be obsolete.
[20:11:07] <Jimmybondi> without the 2 added lines emc works - for first steps :- )
[20:11:44] <cradek> HAL:1: Unknown command '.lol'
[20:11:52] <cradek> what's on line 1?
[20:12:46] <Jimmybondi> wtf
[20:12:51] <JymmmEMC> skunkworks they just want you to buy out the remaining 21,000,000 they have in stock is all
[20:13:04] <Jimmybondi> i'm not mutliprocessor-able x.D
[20:13:12] <skunkworks_> they only have 527 in stock..
[20:13:26] <cradek> Jimmybondi: ??
[20:13:32] <JymmmEMC> skunkworks did they tell you a replacement?
[20:13:38] <Jimmybondi> one moment
[20:13:55] <skunkworks_> no direct replacement.
[20:14:00] <Jimmybondi> can't believe - no error ...
[20:14:05] <JymmmEMC> Jimmybondi: EMC can't run SMP, though you can install /enable the SMP by adding the kernel
[20:14:26] <maddash> Jimmybondi: does emc start or not?
[20:14:28] <cradek> EMC2 can definitely run on SMP
[20:14:39] <JymmmEMC> cradek: not out of the box
[20:14:40] <cradek> maddash: no, there's a syntax error in a hal file
[20:14:44] <Jimmybondi> i'v tried it several times- the .lol cames within the last minutes as i want wite here and choose wron keyboard ....
[20:14:56] <Jimmybondi> it's starting
[20:15:44] <maddash> cradek: smp ftw!
[20:16:06] <maddash> JymmmEMC: it's a kernel thing.
[20:16:21] <JymmmEMC> maddash: No, it's a make thing
[20:18:05] <maddash> JymmmEMC: i'm running smp now. it's a kernel thing. a kernel parameter thing. brb.
[20:19:26] <JymmmEMC> At least he's livign up to his nick
[20:21:09] <JymmmEMC> Question... on the drive having the following disignations: A, A#, B, B# Which is the return? The A or the A# ?
[20:33:13] <JymmmEMC> cradek: Has something changed since jmkasunich told me I needed to add the SMP kernel that doens't allow EMC to run?
[20:35:45] <cradek> I don't understand the question
[20:37:15] <JymmmEMC> cradek: When I installed ubuntu/emc on this SMP box, I didn't have dual spu's nor did it see the 2gb. jmkasunich said I needed to add/enable the SMP kernel, which I did, and it saw both cpu's and the 2gb, BUT EMC couldn't run in that environment/kernel.
[20:37:41] <cradek> EMC can only run (in realtime) with a realtime kernel
[20:37:49] <cradek> the realtime kernel most people are using isn't smp
[20:38:35] <JymmmEMC> cradek: ok, so what RT kernel and you using that is SMP ?
[20:38:45] <JymmmEMC> s/and/are/
[20:38:52] <cradek> unfortunately it's impossible to make a SMP realtime kernel that runs on all machines
[20:39:07] <cradek> so, the one I put on the CD and build packages for can't be SMP
[20:40:15] <JymmmEMC> is this strictly a make thing, or a little more than just that?
[20:40:31] <cradek> I don't know what you mean by "make thing"
[20:40:41] <JymmmEMC> cvs, make, make install
[20:40:53] <cradek> everything has to be compiled as SMP: kernel, rtai, emc2
[20:41:42] <JymmmEMC> so basically from scrat for each machien that wants rt SMP?
[20:41:46] <JymmmEMC> scratch
[20:41:46] <cradek> yes
[20:41:52] <cradek> unfortunately
[20:42:36] <Rugludallur> cradek: Would it be possible to have two distributions, one smp and one single ?
[20:42:55] <JymmmEMC> cradek: That's what I was trying to say to Jimmybondi, as if he's having issues adding a 2nd paraport, I'd suspect that he'd have issues RT SMP as well.
[20:43:10] <cradek> Rugludallur: maybe, with only twice the work :-/
[20:43:35] <Rugludallur> cradek: Yeah, and four times the number of bugs and then there are threading issues and ......
[20:43:43] <JymmmEMC> * JymmmEMC has a stack of dual cpu mobo's here
[20:43:43] <cradek> Rugludallur: SMP is not important for people using EMC in production - they need reliability
[20:44:08] <cradek> (IMO)
[20:44:14] <JymmmEMC> cradek: So, EMC in an SMP env doesn't really benefit the EMC aspects?
[20:44:48] <cradek> it's a tradeoff
[20:45:11] <Rugludallur> cradek: true, but there are definite advantages to multiple cores, being able to distribute rt threads and make userspace apps more responsive under load
[20:45:52] <Rugludallur> cradek: but for now I think 99% of the userbase does not need smp
[20:46:03] <tomp> what determines the velocity update rate for servos ? ( how often is a new voltage created for the amplifier)
[20:46:09] <JymmmEMC> IMNSHO... a controller should be dedicated... PC HW is cheap enough these days to have another computer for your everyday usage - side by each
[20:46:16] <cradek> tomp: the servo cycle
[20:46:43] <cradek> Rugludallur: someone who needs SMP can definitely compile it, nothing stopping him
[20:46:51] <Rugludallur> JymmmEMC: yeah, but UI things are userspace :)
[20:46:59] <Rugludallur> cradek: true
[20:47:00] <cradek> Rugludallur: he can't use our packages then, though
[20:47:31] <JymmmEMC> Rugludallur: Play your music or p0rn on another computer =)
[20:47:41] <tomp> cradek: is servo cycle an ini param, a hal param? ( what determines the servo cycle )
[20:48:03] <JymmmEMC> Rugludallur: or are you speaking of GUI interface?
[20:48:11] <cradek> tomp: [EMCMOT]SERVO_PERIOD
[20:48:16] <Rugludallur> JymmmEMC: I do, but I preferr to keep Axis on the same machine, the wifi is a bit unstable when I start cutting with the plasma :)
[20:48:15] <tomp> thanks
[20:48:25] <cradek> sorry tomp I did not understand your question
[20:48:39] <tomp> my bad, not yours
[20:48:42] <JymmmEMC> Rugludallur: CAT5e or CAT6 STP =)
[20:49:02] <JymmmEMC> Rugludallur: just get your overalls on and start pulling that cable!!!
[20:49:04] <Rugludallur> JymmmEMC: yeah, I use that for my limit/estop switches,
[20:49:35] <JymmmEMC> Rugludallur: I mean forget the wifi!
[20:49:49] <JymmmEMC> gigabit baby!
[20:50:23] <Rugludallur> JymmmEMC: I would but I need to be able to move the whole thing around so Wifi is just easier (although I hate Wifi for most parts)
[20:51:00] <JymmmEMC> Rugludallur: What, you have 120/220VAC wireless too?
[20:51:25] <Rugludallur> JymmmEMC: nahh more like 63A 400V wired
[20:51:44] <tomp> wireless 120Vac ? hurray tesla!
[20:52:08] <JymmmEMC> Rugludallur: There ya go, run the CAT6 STP parallel to the 400V =)
[20:52:19] <Rugludallur> JymmmEMC: http://www.dallur.com/typo3temp/pics/b5a6a779a8.jpg
[20:52:40] <Rugludallur> JymmmEMC: http://www.dallur.com/typo3temp/pics/b87094f954.jpg
[20:52:59] <JymmmEMC> wimpy connector
[20:53:39] <Rugludallur> JymmmEMC: yeah, :)
[20:56:12] <JymmmEMC> Hmmmm, the brand new battery charger I bought just turned GRN within 2hrs on a BRAND new battery. I wonder if it has a trickle charge now (to top off the battery), or will it prevent from overcharging. Guess I'll wait to unplug it, the see what the battery temperature is in 30-45 minutes from now.
[21:01:45] <archivist> Rugludallur, those connectors are fun to pull out when welded together from "use"
[21:07:10] <JymmmEMC> brb
[21:08:24] <Jimmybondi> puh - 5 axis working :- )
[21:08:36] <cradek> yay
[21:08:43] <Jimmybondi> small problem with preview - but not so urgent ...
[21:08:44] <cradek> what's your project?
[21:09:14] <Jimmybondi> standard bed mill with 2 add angular axis
[21:09:29] <Jimmybondi> lets see if i have a link for picture
[21:09:36] <cradek> sounds fun
[21:09:38] <skunkworks_> Jimmybondi: pictures?
[21:09:41] <skunkworks_> oh
[21:10:13] <Jimmybondi> avi well too?
[21:10:20] <skunkworks_> sure
[21:11:00] <Jimmybondi> http://irc.global-radio.ch/img/5Achsen.avi
[21:11:21] <Jimmybondi> driven by Mach3
[21:12:00] <Jimmybondi> but i get a cast-iron working plate at te weekend - custom made and very heavy ...
[21:12:23] <cradek> I've never seen a dremel spindle on such a machine :-)
[21:12:32] <Jimmybondi> .lol
[21:12:46] <Jimmybondi> atm there is a Isel MA2.05 0,66kW mounted
[21:13:11] <Jimmybondi> but not ready at all
[21:13:35] <Jimmybondi> the machine is 16 years old - i build it in 1991
[21:13:44] <skunkworks_> That is cool - that is exactly how I want to make a 4,5 axis setup
[21:13:50] <Jimmybondi> at start only 110mm in Z axis
[21:14:01] <robin_sz> meep?
[21:14:12] <robin_sz> * robin_sz has been having microcontroller fun today
[21:14:24] <Jimmybondi> does anyone know turbo-cnc?
[21:14:41] <Jimmybondi> there is a picture of a wooden clock i made in 1994
[21:14:46] <cradek> sorry no
[21:15:45] <Jimmybondi> scroll down at http://www.dakeng.com/gallery.html - 4th pic from bottom
[21:15:53] <Jimmybondi> thats mine
[21:16:36] <robin_sz> nice machine!
[21:16:39] <cradek> neat
[21:16:44] <skunkworks_> Cool - so you went from turbocnc - to mach - then to emc2?
[21:16:52] <robin_sz> now all you need is the nice software to drive it ;)
[21:17:15] <Jimmybondi> i would go to emc because of planning a atc
[21:17:24] <robin_sz> * robin_sz nods
[21:17:35] <robin_sz> emc is the right choice for you I think
[21:17:40] <Jimmybondi> and mach only support Gxxx-shit and ModIO
[21:17:50] <skunkworks_> What do you thing of emc2 compared to mach so far.
[21:17:59] <tomp> Jimmybondi: The AB table/cradle is nice. Is it open loop? What spindle will you use with atc?
[21:18:00] <Jimmybondi> hmmm
[21:18:05] <cradek> Jimmybondi: http://www.youtube.com/watch?v=vNn-Yr7it5s
[21:18:20] <Jimmybondi> mach is easy to config
[21:18:23] <Jimmybondi> very easy
[21:18:32] <robin_sz> for some people, with simple needs, and limited technical ablity, Mach is a very good choice, but that machine needs a good control, and by the standard fo work on it, I;d say you are up to the task of installing EMC, it will be worht it.
[21:18:36] <Jimmybondi> emc difficult but much more smotth on axis
[21:19:09] <robin_sz> yes, its nto as easy, but, once you have set it up, the result is worth the work
[21:19:58] <Jimmybondi> @ tomp: open loop? nope - only stepper driven
[21:20:08] <Jimmybondi> the hole machine is step
[21:20:13] <robin_sz> coo
[21:20:50] <Jimmybondi> the first step-amp i created myself was a LMD18245 an 1/8 step
[21:21:09] <robin_sz> coo.
[21:21:21] <Jimmybondi> the whole rest in te world (hobbiests) run at half-step *xd*
[21:21:29] <robin_sz> na
[21:21:32] <Jimmybondi> at that time
[21:21:37] <robin_sz> the rest of the world uses geckos ;)
[21:21:41] <Jimmybondi> its 15 years ago
[21:21:56] <Jimmybondi> oh my satan - never again Gxxx
[21:22:16] <skunkworks_> what does Gxxx mean?
[21:22:17] <Jimmybondi> i've imported over 50 drives to germany
[21:22:18] <robin_sz> I used quite a lot of them, never had a problem
[21:22:38] <Jimmybondi> maris knew my name up to today
[21:23:03] <Jimmybondi> but 90% of these drives died after 3 or4 years
[21:23:21] <ds2> is it safe to wire up a machine designed for use in the UK (H/N/G 220V configuration) to use in the US by just doing a US 220V (H/H/G configuration)?
[21:23:27] <Jimmybondi> i 've 2 in use until today - G210
[21:23:33] <robin_sz> weird, I have probably 30 or 40 in daily use on various plasma machines, mills etc, no problems so far
[21:24:08] <robin_sz> mostly on about 50 to 60V
[21:24:19] <robin_sz> plenty of external capacitor
[21:24:23] <Jimmybondi> atm i'm using RTA amps at 85V DC
[21:25:07] <Jimmybondi> the geckos got 70V at my machine
[21:25:23] <robin_sz> thats on the top of the range
[21:25:28] <Jimmybondi> si
[21:25:48] <robin_sz> ive never had motors that would take that much voltage
[21:25:57] <Jimmybondi> but when using max vel you have to use such high voltage
[21:26:03] <Jimmybondi> independent
[21:26:18] <robin_sz> mine wont take that much
[21:26:29] <Jimmybondi> the voltage to the drive speeds up the motor at high frequency
[21:26:36] <robin_sz> yes, I know
[21:26:42] <Jimmybondi> because the current can rise very fast
[21:27:06] <robin_sz> but I use low R motors, dc drop perhaps 2.2V at rated current
[21:27:23] <robin_sz> 50 to 60V is max for those, or they suffer from heat
[21:27:24] <Jimmybondi> my motors have had 4A at 3,85V at that time i think
[21:27:30] <robin_sz> right
[21:27:42] <Jimmybondi> the gecko itsself don't heat
[21:27:54] <Jimmybondi> the problem is the build-in cap
[21:27:55] <robin_sz> no, but at 70V the motor would
[21:28:00] <robin_sz> * robin_sz nods
[21:28:11] <robin_sz> I always use external cap,
[21:28:16] <robin_sz> and very short leads
[21:28:23] <Jimmybondi> all below 85°C at the motor would be ok
[21:28:32] <Jimmybondi> if you don't touch *xD*
[21:28:42] <robin_sz> heh
[21:30:41] <robin_sz> today, I went out getting new work, nice order from a company that polishes lenses
[21:31:47] <JymmmEMC> what are you doing for them?
[21:32:31] <robin_sz> making the cabinets, the fabrication for the base of the machine
[21:32:41] <JymmmEMC> ah
[21:32:54] <robin_sz> just 20 to start off with
[21:33:01] <robin_sz> but a nice order anyway
[21:33:29] <JymmmEMC> new vendor, can't trust em
[21:34:30] <JymmmEMC> robin_sz: that wasn't againest you, could have been any new vendor.
[21:36:03] <Jimmybondi> another question:
[21:36:15] <Jimmybondi> does emc support a PIO48?
[21:36:31] <Jimmybondi> PIO48 = 2x 8255 on an ISA-Board
[21:39:54] <Jimmybondi> BTW: my machine in total view: http://irc.global-radio.ch/img/fraese.jpg
[21:40:12] <Jimmybondi> the white plate u see iss a styrodur-form
[21:40:29] <Jimmybondi> cutted for cast form
[21:40:52] <JymmmEMC> nice
[21:41:03] <JymmmEMC> make the machien yourself?
[21:41:15] <Jimmybondi> yes
[21:41:39] <JymmmEMC> Jimmybondi: how does the rail work in the top-bottom config for your Y axis?
[21:41:55] <Jimmybondi> i don't buy anything if i'm able to build myself
[21:42:11] <JymmmEMC> no binding?
[21:42:26] <Jimmybondi> its a linear-drive from bosch-rexroth
[21:42:34] <Jimmybondi> or what do you mean?
[21:43:03] <Jimmybondi> every side has its own ball-
[21:43:05] <tomp> Jimmybondi: emc used to have a pci 8255 driver, (hey it seems gone in 2.1.6, )
[21:43:07] <Jimmybondi> how to call
[21:43:08] <JymmmEMC> On Y axis, I see linear rail mounted on extruction, one on top and the other on the bottom
[21:43:21] <Jimmybondi> oh yeah
[21:43:26] <Jimmybondi> that one
[21:43:42] <Jimmybondi> the lower one is a 20mm
[21:44:04] <Jimmybondi> and the upper iss only 12mm to eliminate torsion
[21:44:24] <JymmmEMC> Jimmybondi: no twosting or binding under load?
[21:44:38] <JymmmEMC> Jimmybondi: no twisting or binding under load?
[21:45:21] <Jimmybondi> nope - both sliders are mounted via the steelplates: 15 x 100mm
[21:45:52] <Jimmybondi> the slider is a ring which enclosure the whole axis
[21:46:13] <Jimmybondi> 4 plate mounted together
[21:46:16] <JymmmEMC> Jimmybondi: On your Z, is there another leadscrew on the backside?
[21:46:25] <Jimmybondi> yes
[21:46:35] <JymmmEMC> so two motors on Z ?
[21:46:43] <Jimmybondi> Y and Z have 2 leadscrews coupled via belt
[21:46:48] <Jimmybondi> yes
[21:46:50] <JymmmEMC> ah, gitcha
[21:46:54] <JymmmEMC> gotcha
[21:46:58] <Jimmybondi> every spindle an own motor
[21:47:27] <Jimmybondi> and both motor coupled in series on one amp
[21:47:52] <JymmmEMC> Jimmybondi: how often do you have to check and make sure their aligned?
[21:48:01] <Jimmybondi> never
[21:48:21] <Jimmybondi> the belt has teeth
[21:48:33] <Jimmybondi> don't how it's called in english
[21:48:43] <JymmmEMC> I understand, say in english
[21:48:48] <Jimmybondi> syncron-belt?
[21:48:49] <JymmmEMC> same in english
[21:50:48] <Jimmybondi> when i've mounted the t-groove-table i made some pics from other views ....
[21:51:18] <Jimmybondi> t-groove - t-slot - i don't know
[21:51:33] <JymmmEMC> bbiab, starting bbq
[21:51:40] <bill2or3> "t-slot" I think.
[21:54:48] <Jimmybondi> ahh - a nice picture i have found:
[21:54:55] <Jimmybondi> selfmade worm-gear
[21:54:56] <Jimmybondi> http://irc.global-radio.ch/img/Schnecke.jpg
[22:00:32] <JymmmEMC> is there a way to test the signal going to the drives?
[22:00:39] <JymmmEMC> from the paraport/emc?
[22:03:16] <Jimmybondi> really what to get out or see what happens?
[22:03:36] <Jimmybondi> i've read something of a osciliskope within HAL
[22:03:58] <JymmmEMC> Sorry, I meant electrically. I only have a DMM
[22:04:04] <JymmmEMC> no oscope
[22:05:18] <Jimmybondi> with a multimeter you can see if anything happens:
[22:05:54] <Jimmybondi> when jogging any axis the result voltage woud be anything between 1,5 and 3,5V
[22:06:01] <Jimmybondi> depends on speed
[22:06:28] <JymmmEMC> Doh, didn't think of that. TY
[22:07:41] <Jimmybondi> are all located in US here 4?14¿4?15
[22:08:45] <archivist> no uk and eu as well
[22:08:53] <Jimmybondi> ah
[22:08:54] <Jimmybondi> fine
[22:09:26] <Jimmybondi> i jog through EU but oversea far enough :- (
[22:09:56] <Jimmybondi> -far +rare
[22:10:21] <JymmmEMC> argh.... Join 0 following error
[22:10:24] <JymmmEMC> joint
[22:12:45] <Jimmybondi> found the mistake?
[22:13:22] <Jimmybondi> i have had a error - seems to be very similar - but not sure - as i made a mistake in core_..hal
[22:13:41] <JymmmEMC> Jimmybondi: Yeah, forgot to plugin the db25 into the drive itself =)
[22:14:06] <Jimmybondi> you drive closed-loop?
[22:14:09] <robin_sz> JymmmEMC, that wont make ANY difference to the followign error
[22:14:21] <robin_sz> Jimmybondi, no, hes not, hes just confused :)
[22:14:29] <Jimmybondi> k
[22:14:50] <JymmmEMC> robin_sz: two seperate issues
[22:15:10] <Jimmybondi> ok - i will take some sleep
[22:15:15] <JymmmEMC> gnight
[22:15:29] <Jimmybondi> and test emc under real cond when new cpu arrived :- )
[22:15:31] <robin_sz> Jimmybondi, you'll get used to JymmmEMC eventually
[22:15:31] <robin_sz> probably a year or two ;)
[22:15:58] <Jimmybondi> he he he
[22:16:09] <Jimmybondi> ok - 4T14hanks15
[22:16:11] <Jimmybondi> gn8
[22:16:12] <JymmmEMC> Jimmybondi: but you'll NEVER get used to robin_sz
[22:16:53] <Jimmybondi> i carry my nick since more than 10 years - i'ld never drop or change :- )
[22:17:14] <Jimmybondi> so long
[22:17:49] <JymmmEMC> robin_sz: Hey bitch....
[22:17:53] <JymmmEMC> Just had to say that
[22:18:13] <robin_sz> blow me :)
[22:18:28] <JymmmEMC> robin_sz: These drives are 10x not 8x stepping, so the BP I guess is still too fscking high
[22:19:07] <robin_sz> JymmmEMC, yeah, thats the trouble, you do need high step rates with modern drives
[22:19:16] <Jimmybondi> short return - have to save server and chan - maybe to forget ....
[22:19:28] <Jimmybondi> bye
[22:19:56] <toast> toast is now known as toastydeath
[22:20:23] <robin_sz> JymmmEMC, I still think to be effective, you need to off-load the pusle generation
[22:20:33] <robin_sz> liek in Ray Henrys boards
[22:20:44] <JymmmEMC> robin_sz: they go from 4x to 10x (and beyond), but no 8x. So now that I have the power to drive, I have nothing to drive it with.
[22:20:54] <JymmmEMC> robin_sz: ray's boards? I know about USC
[22:21:01] <robin_sz> * robin_sz nods
[22:21:03] <robin_sz> that
[22:21:27] <robin_sz> I'll make a video of my grex spining motrs some day
[22:21:32] <JymmmEMC> ah, Jon Elson's.
[22:21:36] <robin_sz> oh, yeah
[22:21:39] <robin_sz> him
[22:21:44] <JymmmEMC> it's all good
[22:22:44] <robin_sz> wired an LCD up to a rabbit processor today
[22:22:58] <robin_sz> worked straight out of the tin
[22:23:16] <JymmmEMC> serial backpack?
[22:23:21] <robin_sz> ?
[22:23:29] <JymmmEMC> 2x16 LCD ?
[22:23:34] <robin_sz> 20x4
[22:23:52] <robin_sz> / interface
[22:23:54] <JymmmEMC> ok, same diff.... they use a serial backpack based off hitachi chipset
[22:24:00] <robin_sz> yeah
[22:24:02] <robin_sz> not that
[22:24:08] <robin_sz> ive used them before
[22:24:11] <robin_sz> this is //
[22:24:12] <JymmmEMC> k
[22:24:28] <robin_sz> using it in 4 bit // mode
[22:24:41] <robin_sz> 4 lines for control, 4 for data
[22:25:01] <JymmmEMC> gotcha
[22:25:30] <robin_sz> very easy to program
[22:25:41] <JymmmEMC> robin_sz: PM?
[22:25:50] <robin_sz> PM?
[22:26:38] <JymmmEMC> robin_sz: Pvt MSg
[22:30:51] <JymmmEMC> robin_sz: check your private messages
[22:31:14] <robin_sz> i have
[22:31:21] <robin_sz> i even replied :)
[22:31:50] <JymmmEMC> Oh hell, it's this dead server.... switching
[22:32:30] <robin_sz> oh wait, maybe I need to identify
[22:37:24] <ds2> no opinions on running machines for 220V H/N/G with 220V H/H/G?
[22:46:57] <toastydeath> what
[22:47:01] <toastydeath> what is h/h/g
[22:47:12] <toastydeath> oh
[22:47:13] <toastydeath> i get it
[22:47:30] <toastydeath> you have a 220 h/n/g line?
[22:47:58] <archivist> ds2 with a good input transformer no difference
[22:50:01] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[23:05:17] <ds2> archivist: I know about the transformer route... was wonder if I can just use the other flavor of 220V directly
[23:06:10] <archivist> no I mean internal to equipment (most normal equipment should be safe anyway)
[23:07:41] <ds2> Oh... I am looking at a lathe designed for use in the UK and they were using it in the US with a 1:2 transformer... but they guy lost the transformer; from what I can tell the input is not really isolated (it is a VFD)
[23:08:09] <ds2> my options appear to be - 1. Get another transformer; 2. Get a 110V input VFD, 3. Directly drive it with a H/H/G 220V
[23:08:22] <robin_sz> 3 shoudl be fine
[23:09:05] <ds2> so they do not normally connect N to the chassis in the UK? (apparently that was the case with radios in the US from the 20's)
[23:09:10] <robin_sz> no
[23:09:25] <robin_sz> G is connected to N at the fuseboard
[23:09:28] <toastydeath> is this a cnc lathe
[23:09:29] <robin_sz> inthe house
[23:09:31] <robin_sz> BUT
[23:09:34] <ds2> toastydeath: yes.
[23:09:38] <robin_sz> NEVER onthe machine
[23:09:54] <toastydeath> drive it with a vfd
[23:10:00] <ds2> toastydeath: it has a VFD!
[23:10:02] <toastydeath> oh
[23:10:04] <toastydeath> good
[23:10:06] <toastydeath> then it doesn't matter?
[23:10:15] <ds2> robin_sz: are the UK power plugs polarized?
[23:10:20] <robin_sz> you always have full isolation between the N and chassis, L and chassis
[23:10:25] <ds2> toastydeath: the VFD wants 220V
[23:10:30] <robin_sz> ds2, yes, but it doesnt matter
[23:10:37] <robin_sz> just give it HHG, it will be fine
[23:10:42] <toastydeath> any 220 will be great
[23:10:48] <ds2> 'k
[23:10:50] <toastydeath> the vfd will take care of any problems
[23:11:00] <robin_sz> * robin_sz puts his fingers in his ears
[23:11:06] <toastydeath> ?
[23:11:12] <robin_sz> for the bang
[23:11:14] <toastydeath> oh
[23:11:15] <ds2> toastydeath: except for the one involving a backhoe, a slab of granite, and a pine box :P
[23:11:15] <toastydeath> hahaha
[23:11:24] <toastydeath> i don't get it
[23:11:31] <toastydeath> slab of granite?
[23:11:36] <toastydeath> pine box?
[23:11:40] <archivist> gravestone
[23:11:42] <toastydeath> oh
[23:11:48] <ds2> going to rework it to use EMC eventually but I would like to test it out first
[23:12:02] <robin_sz> what control doe sit have?
[23:12:12] <ds2> robin_sz: It is a Denford ORAC
[23:12:16] <robin_sz> ahh.
[23:12:20] <toastydeath> i have never heard of that
[23:12:21] <robin_sz> theres a mod for them
[23:12:23] <ds2> apparently something based on a 6502
[23:12:27] <ds2> there are?
[23:12:32] <robin_sz> yeah
[23:12:36] <archivist> plug it in Gungadin
[23:12:47] <robin_sz> you have a concrete floor slab?
[23:12:48] <ds2> got a site besides the denford one?
[23:12:53] <ds2> robin_sz: yep
[23:13:01] <robin_sz> ds2, how thick?
[23:13:13] <ds2> robin_sz: probally 4-6 inches
[23:13:17] <robin_sz> ahh.
[23:13:20] <robin_sz> best advice:
[23:13:26] <ds2> it isn't heavy, AFAIK
[23:13:33] <robin_sz> cut a hole in it about 2 feet by 4 feet
[23:13:46] <robin_sz> dig out to about 4 feet deep
[23:13:56] <toastydeath> what
[23:14:03] <robin_sz> toss it in, backfill and level it off :0
[23:14:11] <ds2> :P
[23:14:11] <toastydeath> looool
[23:14:22] <robin_sz> actually yhey aint that bad
[23:14:29] <ds2> what's your complaint about the Denfords other then being EDU lathes?
[23:14:34] <robin_sz> but it was not ever intended as a "real" lathe
[23:14:44] <toastydeath> what kind of flaws does it have
[23:14:44] <robin_sz> its sold as a "training lathe"
[23:14:49] <robin_sz> its OK on plastic
[23:14:51] <ds2> from what I found, they are a lot more solid then say the import 7x10's
[23:14:53] <toastydeath> how do you have a training cnc lathe
[23:14:56] <toastydeath> oh
[23:15:01] <toastydeath> it's a teeny bugger
[23:15:13] <ds2> toastydeath: bench top unit
[23:15:13] <robin_sz> and maybe on aluminium, light cuts
[23:15:34] <robin_sz> styrofoam is within its reach too
[23:15:36] <toastydeath> you could probably use ceramic
[23:15:39] <ds2> robin_sz: most references I found say otherwise... can you elaborateon your opinions?
[23:15:39] <toastydeath> and cut steels
[23:15:43] <toastydeath> at high speeds
[23:15:54] <robin_sz> ds2, a freind had one
[23:15:55] <ds2> toastydeath: I use cut steel just fine with HSS on the 7x10's
[23:16:01] <toastydeath> i'm partially kidding
[23:16:10] <toastydeath> i think i understand what robin_sz is saying though
[23:16:23] <toastydeath> it's very light duty
[23:16:34] <ds2> it is no hardinage
[23:16:34] <robin_sz> ds2, it really was a bit weak on the spindle ... and not rigid
[23:16:54] <robin_sz> well, it would be a greta little lathe to run with emc
[23:16:57] <ds2> weak as in low HP? aware of that
[23:17:09] <robin_sz> to debug the code and that
[23:17:19] <ds2> people are saying they are more rigid then the 7x10's (it weights in at around 300lb)
[23:17:23] <robin_sz> tried cutting steel?
[23:17:31] <ds2> I cut steel on the 7x10
[23:17:40] <toastydeath> ds2: it depends on what you call "cutting"
[23:17:41] <robin_sz> 7x10?
[23:17:45] <toastydeath> most devices can take a .01 pass
[23:17:49] <toastydeath> in steel
[23:17:54] <robin_sz> .01?
[23:17:58] <robin_sz> .1 more like
[23:17:58] <ds2> import lathe, aka the "mini lathe"
[23:18:03] <toastydeath> robin_sz: inches
[23:18:05] <toastydeath> not mm
[23:18:07] <robin_sz> yeah
[23:18:19] <toastydeath> well i was saying you aren't going to get .1 out of it
[23:18:23] <ds2> www.mini-lathe.com
[23:18:24] <toastydeath> more than likely
[23:18:29] <robin_sz> I can take .1 off a big bar on the colchester
[23:18:46] <toastydeath> a colchester can take .5 or 1
[23:18:51] <ds2> toastydeath: I have roughed at about 0.050 (depth of cut, not diameter change) on the 7x10
[23:18:51] <toastydeath> or should be able to
[23:18:58] <ds2> so not sure why you can it can't do it
[23:19:04] <robin_sz> .5?
[23:19:07] <toastydeath> half an inch
[23:19:11] <robin_sz> coo
[23:19:16] <toastydeath> we are talking about a clausing colchester, right
[23:19:17] <robin_sz> never tried that much
[23:19:25] <toastydeath> 7.5ish hp
[23:19:34] <robin_sz> mm, 3hp
[23:19:35] <toastydeath> oh
[23:19:41] <ds2> toastydeath: is that 0.5" change in diameter or 0.5" depth of cut?
[23:19:45] <toastydeath> .5 doc
[23:19:46] <robin_sz> 6x24
[23:19:59] <robin_sz> ah wait, 12x24 to you americans
[23:20:17] <ds2> a 6x24? that's almost a watch maker lathe ;)
[23:20:23] <toastydeath> a 3 hp clausing will PROBABLY take a .5 doc
[23:20:27] <toastydeath> with a light feed
[23:20:36] <toastydeath> .3-.5 in/rev
[23:20:39] <toastydeath> er, .003
[23:20:41] <toastydeath> etc
[23:21:07] <ds2> it seems for the lower power machines, you can do a bigger DoC w/carbide + higher speeds (the PWM controllers seems to have morepower at higher speeds)
[23:21:27] <toastydeath> well, carbide grades you see today can't handle a heavy cut
[23:21:29] <toastydeath> insert tools, etc
[23:21:44] <toastydeath> but yes, they usually don't have constant torque VFDs
[23:21:54] <archivist> if the swarf does not fly over the tailstock, your not trying
[23:22:14] <toastydeath> so when you turn the motor up to a decent clip, you have more power
[23:22:34] <toastydeath> but .1 doc is not a heavy cut
[23:22:46] <ds2> it is for a puny desktop lathe
[23:22:50] <toastydeath> sure
[23:22:57] <ds2> esp. if the stock is 1" dia to begin with ;)
[23:23:11] <toastydeath> but in the range of "turning stuff in general"
[23:23:19] <toastydeath> it's not heavy, and doesn't overwhelm the tool
[23:23:36] <ds2> shouldn't that be qualified with the size of the inserts?
[23:23:38] <toastydeath> it may overwhelm the lathe, depending =)
[23:23:41] <toastydeath> no
[23:23:46] <toastydeath> the carbide grade, first and foremost
[23:23:59] <ds2> I can't imagine doing anything that much deeper with a 1/4" IC triangular insert
[23:24:06] <toastydeath> most carbides like .05 to .075
[23:24:26] <toastydeath> any more than that and you start to risk failure, depending on how brittle the grade is
[23:24:42] <toastydeath> even with a huge trianglular insert
[23:24:49] <ds2> Oh
[23:25:07] <toastydeath> you can get a roughing insert, and that will handle more feed and heavier cuts
[23:25:08] <ds2> but what's the min HP to even begin to consider that DoC?
[23:25:14] <toastydeath> not much?
[23:25:19] <toastydeath> 1-2 hp
[23:25:22] <toastydeath> .1 doc isn't too bad
[23:25:33] <ds2> no, the 0.5" DOC
[23:25:35] <toastydeath> ohh
[23:25:45] <toastydeath> well, on a geared head lathe, around 3 hp
[23:25:50] <tomp> any probs with carbide on stepper machines? ( my experience was that carbide wanted/insisted very smooth tool pressure on mills and lathes )
[23:26:02] <ds2> I can do the 0.1" on the manual lathes with 1.5HP spindles except the chips hurt then they hit you :(
[23:26:03] <toastydeath> but that's on a gear head lathe, with a "true" 3 hp motor
[23:26:16] <robin_sz> weeeeeel, the original spec is 150 lbs, complete
[23:26:18] <toastydeath> where you can spin the motor slow enough to not burn the tool out
[23:26:19] <robin_sz> not 300
[23:26:39] <ds2> robin_sz: maybe it is the stand? I'll go look at it in this weekend
[23:26:56] <toastydeath> i would not try a .5 doc on a cnc lathe, by the way
[23:27:06] <ds2> toastydeath: why not?
[23:27:07] <toastydeath> they are not built as rigid as a manual machine of the same size is
[23:27:16] <toastydeath> because the CNC paradigm is "fast and light"
[23:27:20] <ds2> Ohh I see
[23:27:48] <toastydeath> the manual machine works by slow turning with huge feeds and huge depths of cut
[23:28:28] <toastydeath> .075 in/rev at 2" depths of cut
[23:28:29] <ds2> robin_sz: I would give you the URL I found it in but hte entire lathe sale has dragged out to about 4 years :(
[23:28:43] <robin_sz> heh
[23:28:51] <toastydeath> versus the cnc would do that at 10k rpm, with light feed and .1 or .075
[23:29:27] <ds2> toastydeath: Hmm...this is a CNC lathe for one and only has a 3/4" pass through spindle so the chances of me wanting to do more then 0.050 DoC per pass is slim
[23:29:35] <toastydeath> yar
[23:29:44] <ds2> and it has the steppers, drivers, spindle encoder all installed
[23:29:46] <toastydeath> but even other lathes in that size category
[23:29:52] <toastydeath> that envelope, that is
[23:30:10] <toastydeath> is what i believe robin_sz is comparing it to
[23:30:26] <toastydeath> but i mean for a benchtop lathe, cnc turning, you're clearly not doing production
[23:30:28] <ds2> what would be another lathe in that class?
[23:30:36] <toastydeath> the Magnum T-8 we have at work
[23:30:39] <ds2> benchtop lathes are not that common
[23:30:44] <toastydeath> it's like a 6x10 lathe
[23:30:46] <toastydeath> (american)
[23:30:58] <toastydeath> i think it weighs in at uh, 12000 lbs
[23:31:10] <toastydeath> 10k or 12k lbs
[23:31:11] <ds2> why does a 6x10 weigh 6tons?
[23:31:14] <robin_sz> * robin_sz cant think of a althe in that class
[23:31:18] <toastydeath> it's a big cnc lathe!
[23:31:26] <toastydeath> it's built exactly the same as a larger envelope machine
[23:31:29] <robin_sz> wait, did amstrad make a lathe?
[23:31:31] <toastydeath> 12 station turrent
[23:31:33] <toastydeath> etc
[23:31:50] <ds2> Oh nice tool changers and stuff
[23:31:53] <toastydeath> the spindle is actually the same size as the total envelope
[23:32:18] <ds2> the only other ones in that range seems to be all hobby lathes - Unimat, Emco stuff, Taig (tiny), Sherline (tiny), imports
[23:32:36] <ds2> 6tons would disqualify it as a table top lathe :)
[23:32:48] <toastydeath> haha, 6 tons disqualifies it as many things
[23:33:09] <toastydeath> it's too bad birmingham doesn't make a cnc teach lathe
[23:33:08] <toastydeath> or something
[23:33:17] <ds2> hmmm no matches for magnum lathe t-8
[23:33:19] <toastydeath> that would be a cute little unit
[23:33:27] <toastydeath> yeah i've NEVER seen this brand before
[23:33:29] <ds2> I take that back... Haas does make one now
[23:33:31] <toastydeath> the magnum
[23:33:57] <toastydeath> but it's very similar to other commercial lathes in that size cateogry
[23:34:02] <toastydeath> for like, production use
[23:34:11] <toastydeath> obv not commercial prototyping, etc
[23:34:34] <ds2> I don't think an SL-10 weights 6T
[23:34:53] <toastydeath> i've used an SL-10
[23:34:56] <toastydeath> this lathe is bigger
[23:35:00] <toastydeath> frame-wise
[23:35:10] <toastydeath> haas makes kind of lightweight machines
[23:35:23] <toastydeath> very awesome for shops that require a bunch of machines
[23:35:43] <ds2> have you heard of something called a "Feeler" CNC lathe? apparently based on clones of a hardinage
[23:35:56] <toastydeath> the hardinge hsv?
[23:36:11] <toastydeath> er
[23:36:11] <ds2> donno, never seen the HSV...just the Feeler and it was dying when I used it
[23:36:11] <toastydeath> hlv
[23:36:19] <toastydeath> nar
[23:36:23] <toastydeath> is it a little benchtop lathe
[23:36:39] <toastydeath> 3k or 4k rpm spindle, collet closer
[23:36:45] <ds2> something like the HLV headstock but takes a bigger collets
[23:36:47] <toastydeath> ah
[23:37:14] <toastydeath> those are very cool lathes though
[23:37:22] <toastydeath> i have used a knockoff, but not the feeler knockoff
[23:37:24] <ds2> 16C collets, a FUNUC T0 control with a punch tape reader
[23:38:02] <toastydeath> haha punch tape?
[23:38:05] <toastydeath> amazing
[23:38:14] <ds2> yep
[23:38:15] <toastydeath> and is that fanuc or really funuc
[23:38:18] <ds2> local JC had that
[23:38:25] <toastydeath> (maybe a fanuc knockoff as well1)
[23:38:26] <ds2> FANUC i think
[23:38:30] <toastydeath> yar
[23:38:34] <toastydeath> that's like, the big big name in controllers
[23:38:46] <ds2> when it works, it is a great lathe but it is tempermental
[23:38:49] <toastydeath> kind of thin on features unless you spend big dollars
[23:38:57] <toastydeath> but it's a beautiful control to use as a set up guy
[23:38:57] <ds2> heard they may be getting rid of it and replace it with another haas
[23:39:11] <toastydeath> ha
[23:39:13] <ds2> eh? it is a horrible control (IMO)
[23:39:29] <toastydeath> what was wrong with it
[23:39:47] <ds2> the air hoses are all rotted through
[23:40:01] <ds2> there is a set screw stuck in the collet changer so you need to disassemble it to change collets
[23:40:03] <toastydeath> well no, i mean the control
[23:40:11] <ds2> Oh hehe
[23:40:40] <ds2> well, as far as I can tell, to setup the lathe in inch mode, I had to manually write down the tool offsets by hand, convert to MM then type it in
[23:40:49] <toastydeath> oh
[23:40:59] <toastydeath> why did you write them down
[23:41:02] <toastydeath> MZ-0
[23:41:05] <toastydeath> MX-0
[23:41:07] <ds2> and if you simulate things (something lock), the control gets confused
[23:41:18] <toastydeath> lock?
[23:41:25] <ds2> oh, I did not know about those codes... remember, I started on a haas
[23:41:30] <toastydeath> ah
[23:41:40] <toastydeath> yeah fanuc is a bit arcane but it has a lot of hidden crap that makes it very fast
[23:41:47] <ds2> yeah... you press this button so the tool won't move so you can get a similation
[23:41:50] <toastydeath> i think there's an option in the setup menu for MM values
[23:41:58] <toastydeath> in the offesets, etc
[23:42:04] <toastydeath> but again
[23:42:08] <toastydeath> kind of obscure
[23:42:14] <toastydeath> ohhhh
[23:42:25] <toastydeath> yeah, graph/graphics mode
[23:42:29] <toastydeath> whatever that thing is called
[23:42:34] <ds2> I was looking for a button or at least it willing to take input in the units I selected, not always in MM!!$#@$@$@#$#
[23:42:34] <toastydeath> we don't have that on our fanuc controls
[23:43:00] <ds2> after similation it seems the -/+ sense of the lathe gets swapped around
[23:43:23] <toastydeath> that would suck
[23:43:35] <toastydeath> yeah, i've never ever used the simulation or graph more
[23:43:36] <toastydeath> *mode
[23:43:44] <ds2> yes, it was rather exciting... fingers always on feed hold!
[23:43:58] <toastydeath> but that's the good thing about a lot of fanuc controls
[23:44:10] <ds2> so, it left a very bad taste for all fanuc controls
[23:44:14] <toastydeath> the control is set up in such a way that you can test the program without doing a vr run
[23:44:29] <toastydeath> just testing it live
[23:44:53] <toastydeath> it's a very nice unit if you set up lathes all day long
[23:45:03] <ds2> more then the haas?
[23:45:07] <toastydeath> waaaay more than haas
[23:45:20] <toastydeath> haas is annoying as piss to live-proof programs on
[23:45:20] <toastydeath> compared to fanuc
[23:45:29] <toastydeath> but haas has that really nice simulator
[23:45:38] <toastydeath> haas is easier to program on than fanuc
[23:45:44] <ds2> seems the opposite but then I donno the fanuc that well
[23:46:12] <ds2> the part-diameter measure and part-Z zero buttons are really nice
[23:46:16] <ds2> (wish EMC had that ;))
[23:46:19] <toastydeath> ?
[23:46:27] <toastydeath> fanuc has those, too
[23:46:29] <toastydeath> but they're hidden =)
[23:46:36] <ds2> on the haas, to setup a tool, I can face and hit a button then you get Z setup
[23:46:42] <ds2> bah
[23:46:46] <toastydeath> on fanuc, you face and hit mz 0
[23:46:52] <toastydeath> or mz-whatever value
[23:47:11] <toastydeath> same with the x axis
[23:47:20] <toastydeath> the good part about fanuc is all the knobs and buttons right there
[23:47:33] <ds2> don't think there were MZ buttons
[23:47:39] <toastydeath> no, you have to hit the M key
[23:47:40] <toastydeath> then the Z key
[23:47:43] <toastydeath> then 0
[23:47:46] <toastydeath> on the keypad
[23:47:47] <ds2> Ohhhh
[23:48:02] <ds2> other complaint is the lack of a qwerty KB
[23:48:10] <toastydeath> yar, it sucks to program on somewhat
[23:48:46] <toastydeath> but haas does not have the feed/speed/spindle override knobs
[23:48:49] <toastydeath> right there
[23:49:05] <toastydeath> and fanuc has a mode to link your rapid rate to your feed override
[23:49:06] <ds2> I do think that would be a good machine to put EMC as a community relations thing
[23:49:22] <toastydeath> so like you can set feed to 0 and everything stops
[23:49:32] <ds2> toastydeath: eh? the haas does
[23:49:42] <toastydeath> the haas has all those buttons on it
[23:49:49] <toastydeath> and doesn't have knobs you can grab in an emergency
[23:49:50] <ds2> there is a way to tie it to the knob
[23:49:52] <toastydeath> i know
[23:50:06] <toastydeath> but as a person who has crashed both haas and fanuc controls
[23:50:13] <ds2> that's what the red button(s) are for... E-stop and feedhold ;)
[23:50:16] <toastydeath> i would crash a fanuc any day
[23:50:28] <toastydeath> you just have more control over the machine
[23:50:42] <toastydeath> but the control itself is primitive
[23:50:45] <toastydeath> compared to a haas
[23:50:52] <toastydeath> so it's a big trade off
[23:51:07] <ds2> a large part of is knowing they exists
[23:51:36] <toastydeath> yar
[23:51:40] <toastydeath> that's a big part of anything cnc related =)
[23:51:52] <toastydeath> i have been drooling over the Centroid controls lately
[23:51:54] <toastydeath> i'd like to use one
[23:51:58] <ds2> the haas seems a lot more intuitive
[23:52:05] <toastydeath> the hass is very intuitive
[23:52:09] <ds2> the manuals actually made sense!
[23:52:13] <toastydeath> hahahaha
[23:53:29] <toastydeath> my school has a hass sl-10
[23:53:33] <toastydeath> with an unbalanced spindle?
[23:53:36] <toastydeath> i don't know what's wrong with it
[23:53:44] <toastydeath> but it gets goddamned scary past 2k
[23:53:52] <toastydeath> i think it was either damaged on installation
[23:53:51] <ds2> Oh hehe
[23:53:54] <toastydeath> or somebody crashed it really really bad
[23:54:04] <toastydeath> def. not haas's fault
[23:54:12] <ds2> school bought it outright?
[23:54:16] <toastydeath> yep
[23:54:19] <toastydeath> brand new
[23:54:22] <toastydeath> well, it's not new NOW
[23:54:28] <toastydeath> but they are the only owner
[23:54:33] <ds2> the one the JC had is an extended loan unit from Haas
[23:54:54] <toastydeath> nice
[23:55:00] <toastydeath> did they get a price break?
[23:55:03] <toastydeath> i don't think my place did
[23:55:08] <ds2> yep
[23:55:14] <toastydeath> bastards!
[23:55:20] <ds2> they had some special relations with them
[23:55:25] <toastydeath> well, i guess it doesn't matter too much, the lathe and mill never get any use
[23:55:30] <ds2> and Haas puts a ton of ads/banners all over
[23:55:46] <toastydeath> the CNC prof is a regular professor, he isn't someone who came from the industry
[23:55:59] <toastydeath> so nobody uses it, as they can't set it up fast enough to make any use of it
[23:56:03] <toastydeath> it's sad
[23:56:35] <ds2> get the ME dept to step in
[23:56:39] <toastydeath> it IS in the ME department
[23:56:46] <toastydeath> they actually have a machinist training program there!
[23:56:53] <ds2> Ohhhh
[23:57:06] <ds2> the JC has it as part of the machinist vocational program
[23:57:25] <toastydeath> we only get like 4 people through the machinist program at any one time, tops
[23:57:33] <toastydeath> so the cnc class runs once every year and a half or so
[23:58:43] <toastydeath> and really the problem is that nobody from industry works there who used cnc day in and day out
[23:59:12] <toastydeath> i'm the only guy in the ME department or the machinist training thing who works in a shop
[23:59:20] <toastydeath> and deals with cnc daily
[23:59:30] <toastydeath> which means the mill is always open if i need it =)
[23:59:35] <ds2> heh
[23:59:43] <toastydeath> it's a tiny college though
[23:59:49] <ds2> the CNC lathe should be useful for making furniture legs ;)
[23:59:49] <toastydeath> so i guess it's understandable
[23:59:52] <toastydeath> hahaha