#emc | Logs for 2011-01-20

Back
[00:04:10] -!- theorb [theorb!~theorb@91.84.53.6] has joined #emc
[00:04:15] -!- theorbtwo has quit [Ping timeout: 240 seconds]
[00:04:21] theorb is now known as theorbtwo
[00:08:56] -!- morfic- has quit [Changing host]
[00:08:56] -!- morfic- [morfic-!~morfic@unaffiliated/morfic] has joined #emc
[00:09:01] morfic- is now known as morfic
[00:24:45] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[00:26:11] -!- mhaberler has quit [Ping timeout: 255 seconds]
[00:32:43] -!- kilian1 [kilian1!~kilian@c-24-245-57-37.hsd1.mn.comcast.net] has parted #emc
[00:45:53] -!- fatpandas [fatpandas!~fatpandas@c-71-56-124-145.hsd1.ga.comcast.net] has joined #emc
[00:46:39] -!- mikeggg has quit [Ping timeout: 240 seconds]
[00:46:52] -!- fatpandas has quit [Client Quit]
[00:52:33] -!- n2diy has quit [Remote host closed the connection]
[01:23:27] -!- tlab has quit [Ping timeout: 240 seconds]
[01:23:43] -!- JT-Shop has quit [Remote host closed the connection]
[01:29:13] -!- Valen has quit [Quit: Leaving.]
[01:30:42] -!- Dannyboy has quit [Remote host closed the connection]
[01:31:11] -!- Valen [Valen!~Valen@ppp226-148.static.internode.on.net] has joined #emc
[01:31:59] -!- tlab [tlab!~tlab@c-69-243-187-136.hsd1.in.comcast.net] has joined #emc
[01:32:39] -!- mendelbuild [mendelbuild!~mndl@77-95-78-113.bb.cust.hknett.no] has joined #emc
[01:36:02] -!- jthornton_ [jthornton_!~chatzilla@216-41-143-50.semo.net] has joined #emc
[01:38:41] -!- jthornton has quit [Ping timeout: 276 seconds]
[01:38:45] jthornton_ is now known as jthornton
[01:39:18] -!- frallzoran has quit []
[01:39:28] -!- Tom_L [Tom_L!~Tl@unaffiliated/toml/x-013812] has joined #emc
[01:41:23] -!- fenn has quit [Quit: Lost terminal]
[01:43:17] -!- fenn [fenn!~fenn@131.252.130.248] has joined #emc
[01:54:02] -!- Emcrules_Laptop [Emcrules_Laptop!~Jean-Paul@CPE001310e6c94f-CM001bd71cb794.cpe.net.cable.rogers.com] has joined #emc
[01:54:24] -!- Emcrules_Laptop has quit [Client Quit]
[01:54:26] <Jymmm> http://www.flixxy.com/vintage-technologies.htm
[01:54:29] -!- Emcrules_Laptop [Emcrules_Laptop!~Jean-Paul@CPE001310e6c94f-CM001bd71cb794.cpe.net.cable.rogers.com] has joined #emc
[01:57:13] -!- Valen has quit [Quit: Leaving.]
[01:58:58] <Emcrules_Laptop> is happy got my D510MO today!!!!! kit should be ready soon!!!
[02:07:23] -!- Techrat has quit [Quit: Leaving]
[02:16:09] -!- ries has quit [Quit: ries]
[02:25:10] <mendelbuild> sweet :D
[02:26:48] -!- PCW has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]]
[02:27:03] -!- tlab has quit [Ping timeout: 240 seconds]
[02:44:52] -!- Birdman3131 has quit [Ping timeout: 255 seconds]
[02:46:22] -!- dgarr [dgarr!~dgarrett@adsl-76-246-16-82.dsl.pltn13.sbcglobal.net] has joined #emc
[03:03:11] -!- izua has quit [Ping timeout: 276 seconds]
[03:04:07] -!- izua [izua!~izua@188.27.168.131] has joined #emc
[03:04:07] -!- izua has quit [Changing host]
[03:04:07] -!- izua [izua!~izua@unaffiliated/izua] has joined #emc
[03:06:35] -!- Tom_L has quit []
[03:11:27] -!- izua_ [izua_!~izua@unaffiliated/izua] has joined #emc
[03:12:01] <elmo40> Emcrules_Laptop: nice! can't wait for the pics of the completed project :)
[03:12:09] -!- mendelbuild has quit [Ping timeout: 276 seconds]
[03:14:14] -!- izua has quit [Ping timeout: 276 seconds]
[03:14:48] <emcrules_> elmo actually its going to be a kit i plan to sell
[03:28:30] <Emcrules_Laptop> Im working on putting together prefabed emc2 kits based on mesa hardware. Simple stepper kits to full blow bridgeport retro kits
[03:52:35] -!- dgarr has quit [Read error: Connection reset by peer]
[03:58:09] -!- Techrat [Techrat!~TRat@c-71-232-246-244.hsd1.ct.comcast.net] has joined #emc
[03:58:09] -!- Techrat has quit [Changing host]
[03:58:09] -!- Techrat [Techrat!~TRat@unaffiliated/techrat] has joined #emc
[04:16:29] -!- Dallur [Dallur!~root@157.157.81.3] has joined #emc
[04:22:12] -!- EDocToor has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]]
[04:28:40] -!- Dallur has quit [Quit: Leaving.]
[04:54:01] -!- Emcrules_Laptop has quit [Ping timeout: 255 seconds]
[05:07:35] -!- tom3p [tom3p!~tomp@WNDSON12-1178070722.sdsl.bell.ca] has joined #emc
[05:09:23] -!- nullie [nullie!~nullie@dhcp249-250.yandex.net] has joined #emc
[05:16:42] <tom3p> ? no irc logs for 16th & 17th?
[05:19:01] -!- rooks [rooks!~rooks@102-bem-18.acn.waw.pl] has joined #emc
[05:24:52] -!- Dannyboy [Dannyboy!~Dan@ip68-101-102-133.oc.oc.cox.net] has joined #emc
[05:54:51] -!- mhaberler [mhaberler!~chatzilla@62-47-136-255.adsl.highway.telekom.at] has joined #emc
[06:05:52] -!- psha [psha!~psha@213.208.162.69] has joined #emc
[06:09:03] -!- nullie has quit [Quit: Ex-Chat]
[06:11:14] -!- nullie [nullie!~nullie@dhcp249-250.yandex.net] has joined #emc
[06:17:16] -!- dimas_ has quit [Ping timeout: 250 seconds]
[06:22:16] -!- ewidance has quit [Quit: ewidance]
[06:24:09] -!- dimas_ [dimas_!~dimas@91.192.32.177] has joined #emc
[06:55:19] -!- LawrenceG has quit [Remote host closed the connection]
[07:20:30] -!- mhaberler has quit [Ping timeout: 240 seconds]
[07:21:22] -!- mhaberler [mhaberler!~chatzilla@62-47-136-255.adsl.highway.telekom.at] has joined #emc
[07:23:55] -!- rooks has quit [Quit: So long, and thanks for all the fish.]
[07:32:46] -!- jstenback [jstenback!~jstenback@64.151.104.172] has joined #emc
[07:36:02] -!- Dannyboy has quit [Ping timeout: 240 seconds]
[07:40:24] -!- midgaze [midgaze!~mreid@155.229.21.75] has joined #emc
[07:40:49] -!- midgaze [midgaze!~mreid@155.229.21.75] has parted #emc
[08:00:45] -!- mhaberler has quit [Ping timeout: 276 seconds]
[08:15:44] -!- celeron55 has quit [Read error: Connection reset by peer]
[08:25:33] -!- celeron55 [celeron55!~perttu@a91-155-42-12.elisa-laajakaista.fi] has joined #emc
[08:29:36] -!- mhaberler [mhaberler!~chatzilla@62-47-136-255.adsl.highway.telekom.at] has joined #emc
[08:38:58] -!- Valen [Valen!~Valen@ppp226-148.static.internode.on.net] has joined #emc
[08:49:35] -!- mhaberler has quit [Ping timeout: 260 seconds]
[08:53:08] -!- robh__ [robh__!~robert@5ace70af.bb.sky.com] has joined #emc
[08:56:01] -!- mhaberler [mhaberler!~chatzilla@91-113-80-191.adsl.highway.telekom.at] has joined #emc
[09:04:24] -!- Dallur [Dallur!~root@157.157.81.3] has joined #emc
[09:08:29] -!- micges_work [micges_work!~x@aejj63.neoplus.adsl.tpnet.pl] has joined #emc
[09:22:14] <mrsunshine_> is there any log over this channel?
[09:23:26] <psha> yes
[09:23:28] <psha> two
[09:23:35] <psha> hm
[09:23:36] <psha> one :)
[09:23:38] <psha> logger[psha]: ping
[09:23:46] <psha> another one is 'logger_emc'
[09:23:49] <psha> but it's online
[09:23:55] <psha> s/online/offline
[09:24:46] <psha> mrsunshine_: time is UTC
[09:26:34] <mrsunshine_> Valen, it is you that recomended that board to me :P
[09:26:49] <Valen> probably
[09:26:50] <mrsunshine_> not so important, just looking for a link =)
[09:26:56] <Valen> the dual core atoms
[09:27:03] <Valen> D510M or something i think
[09:27:11] <Valen> check the latency test in the wiki
[09:30:38] -!- micges_work [micges_work!~x@aejj63.neoplus.adsl.tpnet.pl] has parted #emc
[09:31:31] <Valen> I wonder if any of the other mini-itx boards have low latencies when used with core2 or so CPU's
[09:31:45] <Valen> the atoms get a bit laboured
[09:36:05] <mrsunshine_> hows the heat on them ?
[09:36:35] <mrsunshine_> heatsink just got a little bit "hot" yesterday while i installed linux on the machine
[09:36:55] <psha> place board verticaly
[09:37:05] <psha> that's enought
[09:37:49] <psha> http://psha.org.ru/cgi-bin/collection.cgi?action=show_graph;plugin=sensors;type=temperature;timespan=year;plugin_instance=w83627thf-isa-0290;type_instance=temp1;host=twa.psha.org.ru
[09:38:03] <Valen> never had a problem
[09:38:04] <psha> that's D510MO board working as home server
[09:38:07] <Valen> ours is vertical
[09:38:21] <psha> with removed case fan
[09:38:23] <Valen> pulls 100% cpu though atm when running axis with a large file
[09:38:29] <Valen> ours has no side on it
[09:39:26] <psha> thing i don't like it's no heatsink on south bridge (nm10 chip)
[09:40:09] <psha> but i've bought 16x50mm heatsink and cut it into 3 parts :) one spare left :)
[09:41:09] <Valen> lol
[09:41:56] <mrsunshine_> but a little fan is not a problem to have realy
[09:42:04] <Valen> put a big fan on
[09:42:05] <mrsunshine_> just to keep the air moving =)
[09:42:09] <Valen> they are quieter
[09:42:13] <mrsunshine_> true :P
[09:42:18] <Valen> we have a 120mm one on our driver boards
[09:42:35] <mrsunshine_> driver boards?
[09:42:43] <mrsunshine_> my stepper drivers doesnt even get hot =)
[09:42:51] <mrsunshine_> and they give out 4A now =)
[09:43:13] -!- rooks [rooks!~rooks@102-bem-18.acn.waw.pl] has joined #emc
[09:43:35] <Valen> what you using?
[09:45:06] <mrsunshine_> well i guess its some chinease stuff realy, but its from motioncontrolproducts MSD something, can give out 7.something A maximum
[09:45:10] <mrsunshine_> microstepping drive
[09:45:22] <mrsunshine_> real nice things imo =)
[09:45:47] <mrsunshine_> tho they are also mounted on a big steel backplate that draws away heat, but they do not even get worm to the touch when running for hours so =)
[09:46:41] <mrsunshine_> gave my mill like 6x as much power and speed compared to my budget driver i had before this one =)
[09:46:50] <mrsunshine_> tho it cost me .... cost me alot to buy them :P
[09:47:46] <mrsunshine_> http://www.motioncontrolproducts.co.uk/products/10/36/msd880_high_performance_microstepping_drive/
[09:47:48] <mrsunshine_> can be that one
[09:48:03] <mrsunshine_> tho i have the version they had before that one
[09:48:06] <mrsunshine_> that that one replaces
[09:52:25] <mrsunshine_> i wonder if atom has thermal downclock or shutdown
[09:53:35] <Valen> probably
[09:53:49] <mrsunshine_> so it doesnt overheat and get destroyed on me :P
[09:54:03] <Valen> spendy
[09:55:15] <Valen> mesa has 3A 36V drivers, 2 drivers on the one board for $90
[09:55:43] -!- bootnecklad [bootnecklad!~bootneckl@92.17.178.78] has joined #emc
[09:56:38] <mrsunshine_> Valen, well these are 7.2A 80V :P
[09:56:44] <mrsunshine_> i was going for upgradability =)
[09:56:49] <mrsunshine_> so i can change motors etc at will =)
[09:57:20] <mrsunshine_> my motors i have now wants 4A in parallel so anything less then 4A isnt good enough =)
[09:57:44] <Valen> I thought you had 4A drivers
[09:57:56] <mrsunshine_> no
[09:58:03] <mrsunshine_> sorry, iots the motors that want 4A =)
[09:58:08] <mrsunshine_> so the driver is set to about that now
[09:58:12] <mrsunshine_> but it can take 7.2
[09:58:13] -!- bootnecklad_ has quit [Ping timeout: 255 seconds]
[09:58:19] <mrsunshine_> or 7.8
[09:58:22] <mrsunshine_> or whatever it was :P
[09:58:40] <Valen> we use servos
[09:59:01] <mrsunshine_> ahh =)
[09:59:06] <mrsunshine_> i want servos but that is spendy :P
[09:59:10] <Valen> but we are looking at a new mill and I think it might start off cheap
[09:59:16] <Valen> not too bad
[09:59:26] <Valen> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?JakeAndRussells
[10:00:05] <mrsunshine_> got an extra driver that can take 1.8A laying around, 4 axis i think it is but .. thinking of making myself a pcb milling/drilling machine with that one =)
[10:34:34] -!- JustinXJS [JustinXJS!~Justin@c-98-232-176-63.hsd1.or.comcast.net] has joined #emc
[10:39:23] <TekniQue> Valen: that your machine?
[10:39:42] <TekniQue> is the Z axis lifting the entire mill head assembly?
[10:39:46] <Valen> yes
[10:39:48] <Valen> and yes
[10:40:04] <Valen> but the head sucked so we replaced it www.cnczone.com/forums/general_metal_working_machines/118358-phenolic_basalt_head_our_hm45.html
[10:40:05] <TekniQue> ok, any reason not to use the built in feed?
[10:40:09] <TekniQue> not accurate enough?
[10:40:30] <Valen> lots of backlash and hard to grab hold of
[10:40:36] <Valen> no way to fix the backlash either
[10:40:51] <Valen> when the quill isn't locked it can float around alot
[10:42:18] <TekniQue> basalt fibres are looking good
[10:42:43] <TekniQue> my cousin was doing some research using basalt fibres in concrete
[10:43:57] -!- psha has quit [Quit: Lost terminal]
[10:44:11] -!- Techrat has quit [Quit: Leaving]
[10:53:52] -!- mhaberler has quit [Ping timeout: 272 seconds]
[10:55:58] <TekniQue> Valen: haha, wife pleaser 9000
[10:59:48] <TekniQue> also, nice shower curtains
[11:00:44] -!- mhaberler [mhaberler!~chatzilla@62-47-140-13.adsl.highway.telekom.at] has joined #emc
[11:19:33] -!- mhaberler_ [mhaberler_!~chatzilla@62-47-137-253.adsl.highway.telekom.at] has joined #emc
[11:20:06] -!- mhaberler has quit [Read error: Operation timed out]
[11:20:09] mhaberler_ is now known as mhaberler
[11:30:46] -!- logger[psha] [logger[psha]!~loggerpsh@195.135.238.205] has joined #emc
[11:30:48] <jthornton> Valen: how do you like the spindle so far?
[11:43:51] <Valen> JustinXJS: thanks ;->
[11:43:56] <Valen> TekniQue: thanks as well
[11:44:06] <Valen> its using just basalt rocks
[11:44:11] <Valen> jthornton: well it hasn't exploded
[11:44:15] -!- mhaberler has quit [Ping timeout: 240 seconds]
[11:44:30] <Valen> its a little noiser than we were expecting but its much quieter than the old head
[11:44:39] <Valen> you can talk over it
[11:45:15] <JustinXJS> I have seen those spindles on ebay, they look like a good deal
[11:45:32] <JustinXJS> have you cut steel with it yet?
[11:48:02] <TekniQue> Valen: how powerful is that spindle?
[11:48:48] <TekniQue> is this the one? http://cgi.ebay.com/water-cooled-spindle-motor-0-8kw-VFD-set-/250694854579?pt=LH_DefaultDomain_0&hash=item3a5e93e7b3
[11:49:20] <TekniQue> or this one? http://cgi.ebay.com/water-cooled-spindle-motor-3-kw-VFD-set-/250628538544?pt=LH_DefaultDomain_0&hash=item3a5aa000b0
[11:50:01] <JustinXJS> close, Valen has the 2.2kW from his cnczone post
[11:50:08] <JustinXJS> same seller
[11:50:27] <Valen> sorry guys fixing mums tv
[11:50:31] <Valen> no steel yet
[11:50:38] <Valen> need to watch the minimum speeds
[11:50:47] <Valen> only meant to run at over 6000 RPM
[11:50:55] <Valen> which with carbide is borderline on steel
[11:51:16] <Valen> we have the 2.2kw one from linearmotion
[11:51:37] <Valen> havent pushed it over 2 amps yet
[11:51:56] <JustinXJS> That is the only reason I have not bought one yet, RPMs are a bit high
[11:58:46] -!- ries [ries!~ries@200.125.129.54] has joined #emc
[12:09:13] -!- Dallur has quit [Quit: Leaving.]
[12:21:00] -!- psha[work] [psha[work]!~psha@qwe.s2s.msu.ru] has joined #emc
[12:29:14] <MrSunshine> well heat doesnt seem to be a rpboelm som far atleast
[12:29:25] <MrSunshine> like 40 - 45 deg and has been compiling kernel for like 2 hours :P
[12:40:17] -!- skunkworks has quit [Remote host closed the connection]
[12:43:43] <psha[work]> MrSunshine: i'm using gzip -c < /dev/urandom > /dev/null for heating :)
[12:53:21] <TekniQue> kernel compiles are good for stressing the memory bus
[12:56:46] -!- mhaberler [mhaberler!~chatzilla@62-47-137-253.adsl.highway.telekom.at] has joined #emc
[13:00:08] -!- grandrew_ [grandrew_!~Andrew@95-28-110-128.broadband.corbina.ru] has joined #emc
[13:01:42] <grandrew_> Hi! Is it possible to make the system generate step pulses sequentially for each motor instead of pulsing them simultaneously? Via LPT port. I have a driver that can only drive one stepper at once
[13:11:00] <SWPadnos> I think that would be difficult
[13:11:26] <SWPadnos> plus, it's kind of the opposite of coordinated motion
[13:14:23] <Valen> JustinXJS: I havent seen any spindles at all that seem to do the lower RPM range
[13:15:11] <Valen> Ideally I'd like 500-25K rpm with an ATC
[13:15:19] <Valen> but nothing seems to exist to do that
[13:15:35] <Valen> at any price
[13:24:35] -!- grandrew__ [grandrew__!~Andrew@95-28-110-128.broadband.corbina.ru] has joined #emc
[13:24:35] -!- grandrew_ has quit [Read error: Connection reset by peer]
[13:25:49] <grandrew__> So I'm just introducing delays in hal rtapi driver if the system tries to send step pulses to either two motors simultaneously
[13:26:43] <atmega> does that do anything to accuracy?
[13:28:37] <grandrew__> I'm not sure about accuracy (have not measured yet) but the system keeps sending errors about unexpexted delays in task1
[13:28:38] <Valen> why are you trying to do this?
[13:28:59] <atmega> I'd suggest a new stepper driver instead
[13:29:07] <SWPadnos> the unexpected delay errors are showing you another problem
[13:29:09] <grandrew__> :-\
[13:29:31] <SWPadnos> ie, your system has bad realtime perfromance
[13:29:36] <SWPadnos> performance
[13:29:44] <SWPadnos> it's not a laptop, is it?
[13:30:02] <Valen> might his delay be causing it?
[13:30:09] <SWPadnos> could be
[13:30:14] <grandrew__> SWPadnos, no, it is definitely because I am using rtapi delays in hal driver..
[13:30:24] <SWPadnos> ok, then I'd say you should stop doing that :)
[13:30:28] <Valen> grandrew__: why are you trying to do this?
[13:30:40] <Valen> oh the stepper driver
[13:30:47] <Valen> yeah, bad stepper driver
[13:31:06] <Valen> what will your delays do when your stepping at max rate?
[13:31:07] <grandrew__> Valen, because I already have this driver that does additional demultiplexing thus requireing sequential access
[13:31:23] <grandrew__> I will have to step 1/3 max rate
[13:32:14] <psha[work]> SWPadnos: maybe it's possible to hook something like 'oneshot' comps with different delays?
[13:32:33] <SWPadnos> I'm not sure how
[13:32:33] <Valen> have you looked at doing it outside emc? perhaps adding a fixed delay line to the lines?
[13:32:39] <grandrew__> Maybe I should reorganize HAL module somehow since I don't think that demultiplexing is an issue that can prevent the motor driver from being used
[13:32:46] <SWPadnos> the motion controller expects that all motors can move at the same time
[13:33:33] -!- Valen has quit [Quit: Leaving.]
[13:33:44] <SWPadnos> you can't enable and disable stepgens, since they will stop moving and set the internal position to the command position when disabled (thus losing the rest of the move)
[13:34:21] <psha[work]> SWPadnos: stepgen don't need to be disabled - it's output has to be delayed a bit
[13:34:33] <psha[work]> and then three of them OR'ed on one pin
[13:34:51] <psha[work]> as i understand what's multiplexing is
[13:35:07] <SWPadnos> it's not that simple
[13:35:27] <psha[work]> heh, if it's simple grandrew won't ask others :)
[13:35:29] <grandrew__> Okay. So logically the problem is not that I cannot step simultaneously - but that my driver is demultiplexing commands. Maybe I can somehow increase the realtime clock tick interval that the system will receive no 'unexpected delays' ?
[13:36:13] <SWPadnos> that could work
[13:36:31] <SWPadnos> better would be to make a multiplexed version of the parport driver or something
[13:36:32] <psha[work]> SWPadnos: what's problem with using 'delay' comps?
[13:36:49] <psha[work]> except there is no such comp :)
[13:37:10] <SWPadnos> it would need several step inputs, and would output to the parport multiple times (assuming that the driver can handle that)
[13:37:44] <SWPadnos> psha[work], delays would screw up the step timing, and you still don't guarantee that the steps happen at different times
[13:37:55] <SWPadnos> you also still need to deal with setting up the multiplexing address pins
[13:37:57] <psha[work]> ah, got it
[13:38:15] <atmega> add a uC in the middle to serialize output pulses?
[13:38:30] <atmega> or just get a non-sucking stepper driver
[13:38:43] <SWPadnos> non-sucking driver would be the best :)
[13:38:51] <SWPadnos> that would increase speed quite a bit
[13:38:54] <grandrew__> :-P
[13:39:14] <SWPadnos> grandrew__, is there a microcontroller on the driver?
[13:39:18] <SWPadnos> board
[13:39:27] <grandrew__> Okay I need this driver to work in order to mill the non-sucking driver PCB :-)
[13:40:10] <SWPadnos> you can use only straight axis-aligned moves for now :)
[13:40:23] <SWPadnos> but you still need to handle multiplexing
[13:40:25] <grandrew__> No, I've just built the driver from three l293's that are not full featured stepper motor drivers
[13:40:41] <atmega> you built the sucky stepper driver?
[13:40:47] <SWPadnos> ok, so you have a single step+dir pair and then some select lines?
[13:41:03] <grandrew__> Yep
[13:41:17] <SWPadnos> is there a separate demux chip?
[13:41:35] <atmega> buy a cheap chinese driver board, mill your board, sell the cheap chinese one.
[13:41:47] <SWPadnos> nah, I think we can repair this one :)
[13:42:03] <grandrew__> No, demuxing is done using connection logic. It's very simple
[13:42:07] <atmega> right, throw more more hardware at it!
[13:42:34] <SWPadnos> excellent. cut all those traces and run ECO wires from the L293's to the parport. problem solved!
[13:42:46] <grandrew__> Like I set the pin 1 to high when I want to move X, pin 3 to high for Y and both low for Z
[13:43:06] <SWPadnos> have you used all the parport pins?
[13:43:13] <grandrew__> Yes
[13:43:26] <SWPadnos> all 11 or 12 outputs?
[13:43:29] <grandrew__> Pins 0-4 are for multiplexing and 5-7 are for steps
[13:43:33] <grandrew__> Hmm
[13:43:36] <grandrew__> 8
[13:43:39] <grandrew__> :-)
[13:43:58] <grandrew__> I actually need 4 x 3 = 12
[13:44:07] <SWPadnos> look at the parport wiring diagram - the control and status pins are also available as outputs/inputs respectively
[13:44:13] <SWPadnos> why 3?
[13:44:17] <SWPadnos> you can use one pin for all enables
[13:44:25] <grandrew__> For 3 steppers: X, Y and Z
[13:44:33] <SWPadnos> ok, then why 4 pins per?
[13:44:49] <grandrew__> To drive the stepper actually
[13:44:57] <grandrew__> L293 is just an H-bridge
[13:44:58] <SWPadnos> ah, you're using phase drive
[13:45:03] <SWPadnos> oh, I see
[13:45:03] <grandrew__> Right
[13:45:14] <SWPadnos> well then you're screwed :)
[13:45:20] <grandrew__> =)
[13:45:24] <grandrew__> I know
[13:45:56] <grandrew__> But I guess the trick with rt clock would do
[13:46:29] <SWPadnos> so you hacked the parport driver to output the steps at different times, is that right?
[13:46:31] <grandrew__> There is no problem - logically - in doing multiplexing in RT system
[13:46:42] <psha[work]> SWPadnos: maybe multiplexing comp would not be hard...
[13:46:56] <psha[work]> 3 inputs for steppers
[13:47:12] <grandrew__> I actually hacked the parport driver to do the multiplexing
[13:47:15] <SWPadnos> there is a problem in that you can't control whether the inputs to the multiplexer will be activated at the same time
[13:47:16] <SWPadnos> ok, good
[13:47:25] <psha[work]> then align them propely (first input in 1 timeframe, second in 2, etc) and output 0,1,2,0,1,2 on 'selector' pin
[13:48:11] <SWPadnos> psha[work], sure, that could work
[13:48:22] <SWPadnos> but you have to be careful with threads and velocity limits
[13:48:28] <SWPadnos> you'd need 3 threads I think
[13:48:30] <grandrew__> The problem is that the software multiplexer has to do rtapi delays and the system now says they are unexpected and that is an error etc.
[13:48:50] <grandrew__> I can't write from threads actually
[13:48:51] <psha[work]> grandrew__: why to do delays?
[13:49:03] <SWPadnos> the parport/multiplexer in the fastest thread, the stepgen in the next fastest (running at least 3x slower), and then the servo thread at "normal" speed
[13:49:05] <psha[work]> just store state and pop it on next invocation
[13:49:21] <SWPadnos> you shouldn't use delays
[13:49:35] <grandrew__> Because I'm not only multiplexing - I am actually setting the phase so I need to wait enough time for the motor to revolve
[13:50:11] <psha[work]> pick proper thread interval
[13:50:23] <psha[work]> or you have feedback from driver?
[13:50:42] <SWPadnos> make a multiplexer as psha said - all it does is accept 3 inputs and have one set of outputs. all it does per invocation is increment a counter, clip that to the range [0-2], and output whichever set of inputs the counter (index) points to
[13:51:07] <grandrew__> Like if I get 'step X and step Y' the multiplexer will do "First, setp X - wait for completion - now step Y"
[13:51:20] <psha[work]> grandrew__: nearly
[13:51:21] <SWPadnos> that function gets added to a fast thread along with the parport driver
[13:52:07] <SWPadnos> stepgen.make_pulses gets put into a thread that's 1/3 that speed, so that the fast thread always has the time to output all three step states between stepgen.make_pulses runs
[13:52:43] <SWPadnos> the fast thread perod needs to be slow enough that the steppers can always move to the new state between runs (that's the delay - the thread period)
[13:53:12] <grandrew__> Okay I got it, but can't I just decrease the RT frequency down to 1/3 of max step rate?
[13:53:33] <SWPadnos> no, because the parport and multiplexer need to run 3x as fast as the stepgen
[13:53:41] <grandrew__> Ok right
[13:53:47] <SWPadnos> so you need an extra thread for stepgen
[13:54:05] <SWPadnos> and the vel limit needs to be based on the stepgen thread of course
[13:54:38] <SWPadnos> you would also use the standard parport driver, the multiplexer should be a separate component (a pretty simple one really)
[13:55:06] <psha[work]> SWPadnos: just load fast-thread before loading motmod
[13:55:13] <SWPadnos> yes
[13:55:24] <psha[work]> and incread your BASE PERIOD 3 times
[13:55:38] <psha[work]> everything except parport and multiplexer goes as usual to base-thread
[13:55:53] <grandrew__> Hmm
[13:56:17] <SWPadnos> yep. fast thread needs to be slow enough for the motors to move (ie, the BASE_PERIOD you'd use if there were only one motor), then BASE_PERIOD is 3x as long
[13:56:19] <grandrew__> Ok I got it. Thanks.
[13:57:02] <grandrew__> I'll have to dig in the userspace rtapi programming now :-\
[13:57:17] <SWPadnos> no, it still needs to be a realtime component
[13:57:32] <SWPadnos> look at comp though, it makes writing components like this very easy
[13:57:37] <psha[work]> grandrew__: see docs for 'comp' - it simplify process a lot
[13:57:43] -!- mhaberler has quit [Ping timeout: 240 seconds]
[13:57:44] * psha[work] too slow
[13:57:45] <psha[work]> :)
[13:57:51] <grandrew__> :)
[13:58:12] <psha[work]> all you need is one counter
[13:58:24] <psha[work]> that indicates current line
[14:00:50] <psha[work]> you even may further simplify process by processing only one set of pins in one multiplexing comp
[14:01:04] <psha[work]> and then load 4 comps for 4 pins
[14:01:14] <psha[work]> but then you need common 'index' pulse for them
[14:01:15] <grandrew__> I also need 3 phase varaibles
[14:02:23] <psha[work]> phase var is needed to split step pulse into two lines depending on where motor is?
[14:02:56] <SWPadnos> you could be crazy and write a generic component similar to "debounce", which lets the user specify multiple groups of pins and how many pins are in each group
[14:03:02] <SWPadnos> but that would be crazy
[14:03:15] <SWPadnos> phase = index variable in that context, I think
[14:04:35] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.68.115.41.210] has joined #emc
[14:06:20] <grandrew__> Ok can I just run the modified parport driver 3 times faster than stepgen thread and be ok with what I have already?
[14:07:04] <grandrew__> And also 3 times faster than task1 thread
[14:09:12] <SWPadnos> no. you need to get rid of the delays
[14:09:27] <SWPadnos> the thread period should act as your delay
[14:13:50] <grandrew__> Can I somehow use the thread period in my parport driver?
[14:14:23] <SWPadnos> well, you could if you needed to, it's passed to all HAL functions as "long period"
[14:14:27] <SWPadnos> but you shouldn't need to
[14:14:38] <SWPadnos> you really don't want to use delays, they're evil
[14:15:00] <grandrew__> I mean, I will not use the delay but will subscribe to periodic calls
[14:15:25] <psha[work]> you are already subscribed with 'addf function thread' :)
[14:15:52] <grandrew__> So when the command is sent - I will exec for the first motor, then exit and wait for second pulse - check if I have some commands left - and execute them
[14:15:54] <SWPadnos> well, I'd leave the parport driver alone and do it in a separate component. consider what would happen if you decided to add another parport (a $15 or less PCI card)
[14:15:59] <psha[work]> i bet writing separate component and leaving parport driver as is is better solution :)
[14:15:59] <SWPadnos> no
[14:16:08] <psha[work]> #$%#!
[14:16:09] <psha[work]> :)
[14:16:34] <grandrew__> Why not?
[14:16:53] * grandrew__ does not like many components
[14:17:08] <psha[work]> grandrew__: one Large Complex comp is evil
[14:17:11] <SWPadnos> you must have done a lot of things to the parport driver, because it shouldn't have enough inputs
[14:17:20] <SWPadnos> heh, the unix way :)
[14:17:58] <psha[work]> catching bugs in 3 parts each of 10 line is much easier then in one large comp
[14:18:28] <grandrew__> But having everything done in a single source file is anyway better
[14:18:36] -!- EDocToor has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]]
[14:19:03] <SWPadnos> in any case, you wouldn't delay ever. you should add a counter/index variable to the parport driver and output whichever set of signals is selected by that variable
[14:19:22] <SWPadnos> (if you insist on not doing it the right way :) )
[14:20:36] -!- grandrew_ [grandrew_!~Andrew@95-28-110-128.broadband.corbina.ru] has joined #emc
[14:20:36] -!- grandrew__ has quit [Read error: Connection reset by peer]
[14:21:14] <grandrew_> Ok now I *only* need to understand how to run components in threads and how to sync them :)
[14:21:25] <SWPadnos> threads are synchronous
[14:21:54] <SWPadnos> the fastest thread determines the system "tick", and all slower thread periods are a multiple of that "tick"
[14:22:33] <psha[work]> grandrew_: heh, it's not true :) one single file may bloat when you add features :)
[14:22:43] -!- Dallur [Dallur!~root@157.157.81.3] has joined #emc
[14:22:53] <psha[work]> multiplexing comp may be done in one file - it's simple enough
[14:22:59] <SWPadnos> the way you "run" a component is to load it (loadrt mycomponent), and add its function to a thread (addf mucomponent.function fastest-thread)
[14:23:00] <psha[work]> but adding it into parport driver is not good idea
[14:23:17] <grandrew_> I mean how to run task1 and parport threads in order. Assuming that task1 seems to be created by emc automatically
[14:23:35] <psha[work]> first you create fast-thread
[14:23:37] <SWPadnos> those commands going into a HAL file, specified in the ini file, or in HALCMD statements in the INI file itself
[14:23:47] <psha[work]> as a first line of hal file
[14:24:19] <SWPadnos> in your ini file, you can add a line in the HAL section that does "loadrt threads name1=fast period1=(whatever it needs to be, in nanoseconds)"
[14:24:33] <SWPadnos> that line should be before the first HALFILE= line
[14:25:04] <SWPadnos> or you can add the same loadrt threads... line to the top of the HAL file that loads emcmot (also specified in the ini file)
[14:26:00] <grandrew_> Ok so I have the base-thread and servo-thread by default
[14:26:16] <grandrew_> I now add the 'fast' thread
[14:28:00] <grandrew_> And just add the parport.0.write not to the base-thread but to fast-thread - and everything should be okay assuming that base-thread is 3x slower than my max step rate wich is fast-thread
[14:28:06] <SWPadnos> like I just explained above. just make sure the fast thread is created before the motion controller is loaded - the motion controller creates the base and servo threads for you, and HAL doesn't let you create a faster thread once the system tick has been defined
[14:28:15] <SWPadnos> yes, pretty much
[14:28:19] <psha[work]> man threads
[14:28:20] <grandrew_> Ok right
[14:28:21] <psha[work]> or
[14:28:22] <psha[work]> man thread
[14:28:25] <SWPadnos> as long as you get rid of the delays
[14:28:28] <psha[work]> don't remember exactly
[14:28:32] <SWPadnos> threads
[14:28:50] <psha[work]> something like loadrt threads param-param-param
[14:28:57] <psha[work]> consult man page for details
[14:30:30] <grandrew_> Hmm I can't see where the base-thread ans servo-thread are created :-\
[14:30:52] <psha[work]> grandrew_: as SWP mentioned they are created by motmod for you
[14:30:55] <SWPadnos> the motion controller does it (kind of a bad idea, but there it is)
[14:31:19] <grandrew_> ok
[14:31:21] <psha[work]> SWPadnos: maybe some flag 'don't load threads for me' will be usefil?
[14:31:23] <psha[work]> useful
[14:31:44] <psha[work]> or just check for existing threads and don't create them if there are already created?
[14:31:59] <SWPadnos> it would be better for motmod to just not bother creating threads at all, but that's not how we do it at the moment
[14:32:14] <SWPadnos> bbl
[14:35:21] <grandrew_> So do I need to alter base_period or only the servo_period ?
[14:35:33] <psha[work]> base
[14:36:18] <grandrew_> hmm so servo_period is the one that I need actually?
[14:36:18] <psha[work]> surely if you current servy period is larger then 3 * base :)
[14:36:25] <psha[work]> no
[14:36:39] <psha[work]> you need to alter only base thread
[14:36:44] <grandrew_> Base_period = 3 * servo_period ? No ?
[14:36:53] <psha[work]> base period = 3 * fast periods
[14:37:00] <psha[work]> servo period = 1ms
[14:37:37] <grandrew_> So servo period is some special?
[14:38:25] <psha[work]> base thread is for stepper control and IO
[14:38:51] <psha[work]> servo is for things like planning etc
[14:39:21] <grandrew_> In the default config servo period is 1 ms and is 10x the base period, while in my resulting config I will have servo period 1/10 of base - is that ok?
[14:39:24] <grandrew_> Ok
[14:39:56] <psha[work]> no
[14:40:03] <psha[work]> servo is slower then base
[14:40:17] <grandrew_> 10x?
[14:40:18] <psha[work]> you base period will increase x100?!
[14:40:26] <psha[work]> leave it at 1ms value
[14:40:51] <psha[work]> if you have base 100us now you'll get fast=100us, base=300us, servo=1000us
[14:41:57] <grandrew_> Base period is 24ms - 3x times slower than my steppers max step rate which is the fast-thread
[14:42:42] <psha[work]> 24ms?! or us?
[14:43:06] <grandrew_> Millisecond
[14:43:14] <psha[work]> not microseconds?
[14:43:20] <psha[work]> that's slow!
[14:43:25] <grandrew_> My motors cannot operate at microsecond speeds :-)
[14:43:49] <grandrew_> That's about 3mm a second in my cnc
[14:44:06] <grandrew_> Maybe 1mm
[14:44:08] <grandrew_> :-)
[14:44:08] <psha[work]> so you'll have 8ms fast thread, 24 base and something like 25 for servo
[14:44:22] <grandrew_> Ok
[14:44:33] <psha[work]> i really don't know enought about servo thread
[14:44:57] <psha[work]> except it has floating point and base - don't
[14:47:44] -!- grandrew_ has quit [Read error: Connection reset by peer]
[14:48:55] -!- TekniQue has quit [Ping timeout: 255 seconds]
[14:49:56] -!- TekniQue [TekniQue!baldur@unaffiliated/teknique] has joined #emc
[14:51:48] <psha[work]> bbl
[14:51:49] -!- psha[work] has quit [Quit: leaving]
[14:53:51] -!- grandrew_ [grandrew_!~Andrew@95-28-110-128.broadband.corbina.ru] has joined #emc
[14:54:11] <grandrew_> Hmm it complains about not defined base-thread
[14:54:54] <grandrew_> Seems like if the threads is already loaded, the emcmot does not define any thread by its own?
[14:57:36] <grandrew_> Ok now it says error inserting motmod operation not permitted :-
[14:59:19] -!- grandrew__ [grandrew__!~Andrew@95-28-110-128.broadband.corbina.ru] has joined #emc
[14:59:25] -!- grandrew_ has quit [Read error: Connection reset by peer]
[14:59:52] -!- tom3p has quit [Quit: Ex-Chat]
[15:02:50] -!- jstenback has quit [Ping timeout: 240 seconds]
[15:04:17] -!- jstenback [jstenback!~jstenback@64.151.104.172] has joined #emc
[15:04:49] <grandrew__> Hmm it keeps saying operation not permitted on loadrt threads or loadrt motmod whichever is the second
[15:05:07] -!- adb has quit [Ping timeout: 255 seconds]
[15:05:16] <SWPadnos> look at dmesg to see the real error
[15:08:26] -!- jstenback has quit [Ping timeout: 240 seconds]
[15:09:02] <grandrew__> It says duplicate thread name base-thread if I define that - and fails to start with 'base-thread not defined' if I dont :-\
[15:09:48] <grandrew__> Seems like motmod is unable to create its thread if I create something in front of it
[15:11:10] <grandrew__> But still tries to and receives another error, duplicate name, if I do define its threads alongside with mine
[15:11:36] <SWPadnos> ok, so maybe motmod checks to see if any threads have been created and doesn't do it if they have
[15:11:43] <SWPadnos> you can define all three threads on your own
[15:12:40] <grandrew__> When I define the three threads I get the 'duplicate thread name' while inserting the motmod
[15:13:22] <SWPadnos> loadrt threads name1=super-fast period1=<a small number of nanoseconds> fp1=0 name2=base-thread period2=<whatever the stuff on the motmod line is[MOTION]BASE_PERIOD or something> fp2=0 name3=servo-thread period3=<same deal as base> fp3=1
[15:13:25] <SWPadnos> err, hmmm
[15:14:00] <grandrew__> Hmm fp3?
[15:14:29] <grandrew__> ok trying with fp3
[15:14:30] <SWPadnos> yeah, the servo thread has to handle floating point for the motion controller and whatnot
[15:14:34] <SWPadnos> the faster ones probably don't
[15:14:42] -!- adb [adb!~Moldovean@178-211-232-101.dhcp.voenergies.net] has joined #emc
[15:15:29] <grandrew__> Still the same: motmod is trying to create its own copy of servo-thread
[15:15:59] <SWPadnos> you should not have removed the base-period= and servo-period = bit sfrom the motmod line, did you do that?
[15:16:16] <grandrew__> I didnt remo them
[15:16:19] <SWPadnos> ok
[15:16:31] <grandrew__> Trying with removed..
[15:16:37] <SWPadnos> no, don't do that
[15:16:52] <grandrew__> Why? Maybe it is why it is trying to create them still
[15:17:20] <SWPadnos> nope. I'm looking at the code, and you need the values to be specified if you don't want to end up with the default values
[15:17:42] <grandrew__> Ok I'm getting the same error
[15:18:11] <SWPadnos> what values did you use for the fast thread period and the base/servo periods?
[15:18:53] <grandrew__> Fast = 8000000 base=24000000 servo=25000000
[15:19:31] <SWPadnos> 8 milliseconds for the fast thread? that's unbelievably slow
[15:19:56] <grandrew__> Right but it is what I need
[15:19:59] <SWPadnos> the base and servo threads can't be those numbers, servo will end up being the same as base in that case
[15:20:18] <SWPadnos> since 25 isn't a multiple of 24 (million)
[15:20:39] <SWPadnos> so you'll end up with 8ms fast, and 24 ms base and servo threads
[15:20:46] <grandrew__> Ok it worked with 48mln
[15:20:52] <SWPadnos> correct
[15:21:38] <SWPadnos> the HAL files have references to a base thread, so when the motion controller doesn't create one (since the base and servo periods end up the same), you get errors when there are commands to add things to the base thread
[15:21:54] <grandrew__> Gtg now, thanks, will continue it tomorrow
[15:21:57] <SWPadnos> ok
[15:22:12] <grandrew__> I got the joint error while running the test btw...
[15:22:27] <grandrew__> Joint following error
[15:22:48] <SWPadnos> you'll have to mess with max velocities
[15:23:04] <SWPadnos> one step every 8 ms means 125 steps/second, which isn't all that much
[15:23:46] -!- jstenback [jstenback!~jstenback@64.151.104.172] has joined #emc
[15:25:31] * MrSunshine is trying to make a ubuntu usb installstick for linuxcnc cause he doesnt have a cdrom drive...
[15:25:32] <MrSunshine> sigh
[15:25:41] <SWPadnos> oh. it's actually one step every 24 ms, so only 41.666 steps/second
[15:26:40] -!- grandrew__ has quit [Ping timeout: 245 seconds]
[15:33:43] <MrSunshine> wth
[15:33:50] <MrSunshine> first boot it worked but then i didnt have a harddrive in
[15:33:52] <MrSunshine> now i get "boot erro"
[15:33:54] <MrSunshine> error
[15:38:42] <SWPadnos> the BIOS is probably set to boot from hard disk before USB
[15:38:58] -!- Dallur1 [Dallur1!~root@157.157.81.3] has joined #emc
[15:39:03] <SWPadnos> or something like that
[15:39:22] <skunkworks> SWPadnos: home again?
[15:39:27] <SWPadnos> yep
[15:39:36] <skunkworks> bet that is nice :)
[15:39:47] <SWPadnos> yep, except for the snow ;)
[15:39:52] <MrSunshine> SWPadnos, nop
[15:40:00] <MrSunshine> even removed the hd and its still giving same error =)
[15:40:11] <SWPadnos> oh. then you have a different problem
[15:40:53] -!- Dallur has quit [Ping timeout: 264 seconds]
[15:45:39] -!- acemi [acemi!~acemi@unaffiliated/acemi] has joined #emc
[15:48:13] -!- Dallur [Dallur!~root@157.157.81.3] has joined #emc
[15:49:45] -!- Dallur1 has quit [Ping timeout: 246 seconds]
[15:49:49] -!- Dannyboy [Dannyboy!~Dan@12.192.111.2] has joined #emc
[15:50:44] -!- Birdman3131 [Birdman3131!~quassel@76-220-98-18.lightspeed.fyvlar.sbcglobal.net] has joined #emc
[15:53:07] <skunkworks> logger[psha]: ping
[15:53:39] -!- psha [psha!~psha@213.208.162.69] has joined #emc
[16:01:48] -!- logger_dev has quit [Remote host closed the connection]
[16:02:04] -!- logger_emc [logger_emc!~emc-logge@vox.dreamhost.com] has joined #emc
[16:02:04] * logger_emc is logging
[16:11:39] -!- Dallur has quit [Quit: Leaving.]
[16:16:09] -!- servos4ever [servos4ever!~chatzilla@173-87-48-6.dr01.hnvr.mi.frontiernet.net] has joined #emc
[16:27:15] -!- bootnecklad_ [bootnecklad_!~bootneckl@2.97.37.191] has joined #emc
[16:29:22] -!- bootnecklad has quit [Ping timeout: 240 seconds]
[16:43:04] <skunkworks> http://www.cnczone.com/forums/general_cnc_mill_lathe_control/119567-programmable_logic_ideas-2.html
[16:53:27] -!- nullie has quit [Quit: Ex-Chat]
[16:58:40] -!- mhaberler [mhaberler!~chatzilla@91-113-84-224.adsl.highway.telekom.at] has joined #emc
[17:00:18] <skunkworks> http://www.youtube.com/watch?v=MuOdItpZntk
[17:00:30] -!- Connor [Connor!~Connor@75.76.30.113] has parted #emc
[17:03:59] -!- Connor [Connor!~Connor@75.76.30.113] has joined #emc
[17:14:21] -!- micges [micges!~micges@accg233.neoplus.adsl.tpnet.pl] has joined #emc
[17:19:50] -!- ries has quit [Quit: ries]
[17:20:44] -!- ries [ries!~ries@200.125.129.54] has joined #emc
[17:30:33] -!- vezza [vezza!~andrea__@host169-39-dynamic.7-87-r.retail.telecomitalia.it] has joined #emc
[17:44:20] <MrSunshine> anyone know of a reliable way to get the ubuntu emc live cd to an usb stick ?
[17:44:25] <MrSunshine> nothing ive tried has worked relay :/
[17:44:28] <MrSunshine> realy
[17:47:44] <MrSunshine> skunkworks, what is that used for ?
[17:47:50] <psha> MrSunshine: unetbootin
[17:48:00] <psha> or something similar
[17:48:05] -!- adb has quit [Ping timeout: 264 seconds]
[17:48:13] <MrSunshine> psha, tried :/
[17:48:18] <MrSunshine> failed misserably
[17:48:26] <psha> you may use small portion of Grub Magic ;)
[17:49:06] <psha> what's wrong with unetbootin?
[17:49:28] <psha> i've not used it though...
[17:49:42] <psha> but i may give you small partition with grub on it :)
[17:50:30] <Jymmm> I have
[17:51:00] <Jymmm> MrSunshine: Yeah, boot the LiveCD and select Make bootable USB Stick
[17:51:14] <MrSunshine> Jymmm, hmm, i could do that on this computer i guess =)
[17:52:02] <MrSunshine> would switch over the cd if it wasnt cramed into such a small space and hard to get the cables where i want :P
[17:52:58] <Jymmm> Yes, because Cd's are *SO* huge that you need a forklift to move tham and a 100,000 sf warehouse to store just one.
[17:54:07] <MrSunshine> Jymmm, takes me half an hour to get the cable back into this computer
[17:54:10] <MrSunshine> if i move it out from where it is
[17:54:34] <Jymmm> MrSunshine: Internal or external?
[17:54:46] <atmega> are all MB's with built in video bad for emc?
[17:54:57] <Jymmm> atmega: mostly
[17:55:29] <Jymmm> atmega: usually produce high latency, but you can disable internal video and add an externa video
[17:55:32] <MrSunshine> Jymmm, internal
[17:55:44] <MrSunshine> and have to remove the whole cabinet from its hole to get the sides off =)
[17:56:07] <Jymmm> MrSunshine: Then buy a another drive or an external USB one.
[17:57:17] <MrSunshine> yeah cause ive go infinite money :P
[17:57:21] <Jymmm> Having an externla USB drive comes in handy.
[17:57:25] <atmega> cool, send me some.
[17:57:28] <Jymmm> MrSunshine: you said it
[17:57:54] -!- MrSunshine has quit [Quit: Leaving]
[17:58:19] <SWPadnos> I don't think you can make a bootable stick from a liveCD boot. it needs to have an ISO available (which may not be the case on a liveCD boot)
[17:59:49] <Jymmm> https://help.ubuntu.com/community/Installation/FromUSBStick
[18:00:08] <Jymmm> Ubuntu includes the usb-creator by default on all LiveCDs and installations.
[18:00:27] <SWPadnos> note the first sentence in the Outline
[18:00:35] <psha> atmega: not true for intel graphics i think
[18:00:40] <SWPadnos> ... you point it at an ISO image ...
[18:01:01] <SWPadnos> I don't know if you have access to the raw ISO when you boot from CD. I don't think you do
[18:01:26] <atmega> psha: this looks small and cheap: http://www.newegg.com/Product/Product.aspx?Item=N82E16856119029&Tpk=N82E16856119029
[18:01:26] <psha> SWPadnos: surely you have, it's located on /dev/sdc...
[18:01:52] <psha> atmega: try to get 510, not 410
[18:01:53] <SWPadnos> I'm not sure that works with the boot disk creator. it might
[18:01:56] <SWPadnos> I've never tried
[18:02:01] <psha> it's 2-core so one may be dedicated to RT tasks
[18:02:02] -!- duend [duend!~lucass_@187.39.184.4] has joined #emc
[18:02:02] <Jymmm> SWPadnos: I have.
[18:02:08] <SWPadnos> ok
[18:02:17] <atmega> 510 is $20 more
[18:02:26] <psha> it costs that
[18:02:29] <SWPadnos> atmega, I saw that one last week or something. it looks interesting
[18:02:32] <Jymmm> I have craploads of bootable sticks
[18:02:42] <SWPadnos> it's too bad none of the Mesa cards fit in a low profile slot
[18:02:57] <SWPadnos> (and most of the small cases don't have a full height slot)
[18:03:31] <Jymmm> atmega: link to the 510 version?
[18:03:32] <psha> atmega: but if you don't need base thread difference is minor i think
[18:03:38] <SWPadnos> http://www.newegg.com/Product/Product.aspx?Item=N82E16856119023
[18:03:42] <atmega> http://www.newegg.com/Product/Product.aspx?Item=N82E16856119024
[18:04:15] <atmega> wonder what's different on those two
[18:04:17] <SWPadnos> huh. there are two
[18:04:30] <SWPadnos> different cases maybe
[18:04:33] <SWPadnos> R10 vs R20
[18:05:18] <duend> Hello all! Is there anyone who uses emc for electrophysiology data acquisition?
[18:05:22] <SWPadnos> and they screwed up the "features", they both have the same text (about the R10)
[18:05:42] <SWPadnos> never heard of anyone doing that
[18:05:52] <SWPadnos> since it's not really a machine control thing, as it were
[18:07:27] <duend> Is possible to write my own codes to control a lab-made hardware?
[18:07:31] -!- nullie [nullie!~nullie@nullie.telenet.ru] has joined #emc
[18:07:58] <atmega> do you use motion control for that?
[18:08:19] <SWPadnos> duend, you might want to take a look at scilab instead
[18:08:58] <Jymmm> The D10 deacription is used on the D20 description, very bottom of page
[18:09:09] <SWPadnos> yeah, that's what I was saying :)
[18:10:09] <duend> i dont use motion control. and i think scilab is not the best choice for my job... hard real time really matters on it.
[18:10:33] <SWPadnos> hmmm. I thought there were RTAI based hardware drivers for scilab
[18:10:41] <Jymmm> power button
[18:11:38] <SWPadnos> or reset button maybe
[18:11:40] <duend> there are, but i am a few afraid of using scilab... scilab is a really high level language
[18:12:10] <SWPadnos> what kind of acquisition rates are you looking at, and what kind of inputs (analog/digital, how many bits, etc)
[18:12:33] <SWPadnos> also, do you need to do anything based on the data, ie does the PC react to things or just acquire data?
[18:12:41] <Jymmm> I dont get it.... http://www.foxconnchannel.com/product/Barebones/detail_comparison_results.aspx?list=en-us0000068,en-us0000057&ID=en-us0000068
[18:12:50] <duend> here is the thing
[18:13:05] -!- drill [drill!~drill@46-126-159-16.dynamic.hispeed.ch] has joined #emc
[18:13:06] <drill> hi
[18:13:17] <drill> does anyone have the schematic of the pluto-p ?
[18:13:55] <duend> we need to study spikes of the neuron H1 of flyes
[18:14:18] <duend> there is an electrode inside its brain
[18:14:28] <SWPadnos> drill, this is the manufacturer of the pluto: http://www.knjn.com/FPGA-Parallel.html
[18:14:52] <SWPadnos> I don't know if they give away the schematics to anyone who asks, but maybe you should have received them with the board ??
[18:15:48] <SWPadnos> duend, ok, that's interesting but it doesn't answer my question, since I don't know anything about the H1 neiron in flies :)
[18:15:51] <duend> when it tries to fly, the electrode has to send a sign to the 'lab-made' hardware, who sends it to the PC
[18:16:07] <duend> its an analog hardware
[18:16:12] <duend> 8 bits
[18:16:14] <drill> SWPadnos but there i cant find the schematic
[18:16:32] <SWPadnos> you may have to ask them for it
[18:16:50] <SWPadnos> or are you looking for the pin header connections when using the EMC2 driver?
[18:17:28] <drill> nope, i want to add only the fpga to my own design, without the pcb the pluto is soldred on it
[18:17:58] <SWPadnos> oh. in that case you should just go to the Altera site and get the chip specs
[18:18:23] <SWPadnos> the Pluto board isn't well designed, so using it as a reference might not be a great idea
[18:18:53] <SWPadnos> duend, ok, what acquisition speed do you need? (100 samples/second, 1 million samples/second ...)
[18:19:20] <drill> if i could i would use a xilix fpga, but i think, i will need the firmware for it... or is it possible to change the pluto stepper firmware running with an xilinx fpga?
[18:19:29] <duend> 10k
[18:20:00] <SWPadnos> sure, the mesa drivers download firmware to the FPGA when they're loaded
[18:20:03] <SWPadnos> like the Pluto
[18:20:22] <drill> but the mesa are too expensive for me
[18:20:48] <Tom_itx> how do mesa compare to gecko?
[18:20:50] <SWPadnos> well, they use Xilinx chips and download to the FPGA every time they're loaded, so that shows that it's possible to do that ...
[18:21:00] -!- packrat has quit [Ping timeout: 272 seconds]
[18:21:06] <drill> i think an open source project could bi interesting
[18:21:15] <SWPadnos> Tom_itx, they don't much. gecko makes motor drivers, mesa mostly makes motion control hardware.
[18:21:18] <drill> but the pc side software is a problem for me
[18:21:44] <SWPadnos> there is a little overlap, since Mesa also makes some motor drivers as daughtercards for their FPGA cards, but there isn't a lot
[18:21:49] <skunkworks> you can get a mesa board for about the same price as the pluto.
[18:22:06] <SWPadnos> yeah, $89 for a parallel port connected one
[18:22:14] <SWPadnos> with 48 I/Os
[18:22:37] <drill> i am interested in the hardware^^ i would like to build something myself^^
[18:23:25] <Tom_itx> i haven't gotten into emc much yet. does it send out the desired clock rate to the driver board and the driver board takes over and turns it into steprates for steppers etc?
[18:23:35] <SWPadnos> depends on the hardware
[18:23:41] <SWPadnos> you can choose whatever way you want
[18:23:49] <Tom_itx> i know it's got a wide range of use
[18:24:21] <Tom_itx> you would get better performance letting the driver board do the bulk of the work though right?
[18:24:34] <Tom_itx> ie higher step rates
[18:24:49] <SWPadnos> the CPU and interrupt performance of the computer would be less likely to be the limit in that case, yes
[18:25:51] <SWPadnos> but it depends on your machine. the average PC (like that $110 one at NewEgg) can output roughly 25000-50000 steps/second (per axis) on a parallel port
[18:26:03] <Tom_itx> i've got an original flashcut system with limited steprates that i'm looking to update and i'm not looking to them to do it
[18:26:19] <duend> SWPadnos 10k
[18:26:35] <SWPadnos> for a stepper that has a 10:1 drive ratio, that's 2000 steps/inch or 20 inches/second = 1200in/minute
[18:26:44] <Tom_itx> is 200khz reasonable?
[18:27:26] <SWPadnos> duend, yep, saw that. that's possible with existing hardware. I make an expensive but good daughtercard for the Mesa boards that has 6 channels of 16-bit analog input plus 8 channels of 16-bit analog output
[18:27:50] <SWPadnos> Tom_itx, not with the CPU, but it's very feasible with just about any hardware (mesa, pluto, pico systems)
[18:27:56] <Tom_itx> yeah
[18:28:01] <Tom_itx> that's what i meant
[18:28:13] <SWPadnos> duend, that hardware is used in a system that runs a 10 kHz control loop
[18:28:46] <Tom_itx> are most of those boards fpga based?
[18:28:54] <SWPadnos> all the ones I mentioned are
[18:29:06] <duend> ok SWPadnos
[18:29:07] <duend> thanks
[18:29:07] <SWPadnos> I like the Mesa cards most, since they're the most flexible
[18:29:09] <duend> :)
[18:29:15] <Tom_itx> are there any open source projects dealing with that?
[18:29:22] <Tom_itx> say a xilinx chip of some sort
[18:29:27] -!- fatpandas [fatpandas!~fatpandas@68.171.169.100] has joined #emc
[18:29:27] <drill> SWPadnos does there an opensource fpga design exist?
[18:29:43] <SWPadnos> duend, sure. I'm not sure how else to get analog data into EMC (or just HAL actually, you don't need EMC) at the rate you want
[18:29:49] <SWPadnos> drill, not that I know of
[18:30:08] <Tom_itx> i haven't actually looked for one yet but there's probably something out there
[18:30:18] <drill> or is noone interested in it?
[18:30:36] <drill> if someone could add support to emc, i would build the hardware
[18:30:55] <SWPadnos> FPGA hardware design isn't as simple as you'd think, and the chips are very expensive in single quantity
[18:31:12] <SWPadnos> I have seen complete boards that cost less than I can buy the chip for
[18:32:47] -!- duend has quit []
[18:33:02] <drill> SWPadnos its my job @work to desing stuff like this...
[18:33:25] <SWPadnos> great. I look forward to seeing your open source design soon ;)
[18:34:22] <skunkworks> cool!
[18:35:32] <drill> SWPadnos but hardware without software isn't useable^^
[18:35:48] <SWPadnos> send me a board and we'll make sure there's software :)
[18:36:09] <drill> i mean the firmware, too^^
[18:36:13] <SWPadnos> the pluto code is GPL
[18:36:15] <drill> the vhdl design
[18:36:52] <SWPadnos> it may be possible to use some of the mesa code as well, but I'd have to check (I imagine Peter doesn't want to make it too easy for the competition)
[18:36:55] <drill> so if you have a design with an xilinx fpga and an interface to parallelport, you will do the other stuff
[18:37:26] <SWPadnos> sure. just keep in mind that I have a day job, so it might be somewhat slow going :)
[18:37:31] <drill> or would an usb interface to the pc better?
[18:37:35] <SWPadnos> nope
[18:37:41] <SWPadnos> USB is sucko for RT control
[18:37:47] <drill> i know
[18:38:01] <SWPadnos> PCI is best, but parport is probably next in line
[18:38:15] <SWPadnos> maybe PCI-E actually, but whatever
[18:38:22] <drill> i think, it could be possible to use an xilinx with an flash for the firmware
[18:38:31] <drill> that it dows not have to be downloaded every time
[18:38:52] <SWPadnos> yep, that's also doable. the Mesa 7I42 has flash onboard, and I think you can use it either way
[18:38:52] <skunkworks> that makes it less flexable though
[18:39:18] <SWPadnos> there has to be a way of programming the flash, so it's almost as flexible
[18:39:27] <drill> cause if it has a flash onboard, you dont have to write software for flashing it every time
[18:39:30] <skunkworks> ah
[18:39:47] <drill> programming with an externel jtag debugger
[18:39:50] <SWPadnos> we have code that can download a bitfile
[18:39:51] <drill> or something like this
[18:40:08] <SWPadnos> assuming that there are some control pins that can be twiddled by software
[18:40:21] <SWPadnos> but that's a little harder on the parport, since there aren't a lot of pins
[18:40:34] <SWPadnos> and you probably want the FPGA to sit right on the parport
[18:40:34] <drill> this is what i mean^^
[18:40:45] <drill> exactli
[18:41:10] <drill> but i could add a onboard jtag which is connected throu a usb port for programming on startup?
[18:41:36] <drill> and the parallelport is only for controlling the steppers and so on
[18:42:40] -!- isssy [isssy!EMCIRCUser@78-83-54-41.spectrumnet.bg] has joined #emc
[18:43:00] <SWPadnos> if you made something with a small micro on it that provided USB DFU functionality, that might make firmware easier
[18:48:13] <Tom_itx> the atmel usb chips support dfu by default
[18:48:29] <Tom_itx> but they're not full usb 2 speed
[18:50:44] -!- nullie has quit [Quit: Ex-Chat]
[18:52:17] <Jymmm> External PS http://www.newegg.com/Product/Product.aspx?Item=N82E16856167041
[18:55:58] -!- the_wench [the_wench!~the_wench@host81-149-189-98.in-addr.btopenworld.com] has joined #emc
[19:02:23] -!- archivist [archivist!~archivist@host81-149-189-98.in-addr.btopenworld.com] has joined #emc
[19:05:34] -!- adb [adb!~Moldovean@178-211-232-101.dhcp.voenergies.net] has joined #emc
[19:06:36] <skunkworks> how do you find the link for the debs?
[19:08:40] -!- micges_ [micges_!~micges@accg233.neoplus.adsl.tpnet.pl] has joined #emc
[19:09:07] <Jymmm> apt-get update;apt-cache search <blah>
[19:09:27] -!- micges has quit [Ping timeout: 240 seconds]
[19:14:05] <skunkworks> no internet
[19:14:33] <skunkworks> if the lucid live cd was used - would these debs work then? http://buildbot.linuxcnc.org/~buildmaster/dists/lucid/v2.4_branch-rt/binary-i386/
[19:15:06] -!- ewidance [ewidance!~ewidance@montpellier.civade.com] has joined #emc
[19:15:24] <skunkworks> (can't imagine not having intenet on the machine.. ) :)
[19:16:55] <skunkworks> or would there be other dependencies?
[19:20:52] -!- JT-Shop [JT-Shop!~chatzilla@216-41-143-50.semo.net] has joined #emc
[19:22:02] <L84Supper> $136.47 http://www.mouser.com/ProductDetail/Hammond-Manufacturing/R251-130-240/?qs=CYzH8SFh9nSDle6v2VPIlKOOxrLaWu9pk4h0tegZaj0%3d did the price of aluminum jump by 300% recently?
[19:22:37] -!- motioncontrol [motioncontrol!~i@host184-76-dynamic.247-95-r.retail.telecomitalia.it] has joined #emc
[19:23:14] <psha> skunkworks: they would work
[19:23:24] <psha> at least i've used them on lucid install from emc live cd
[19:23:48] <skunkworks> thanks
[19:24:01] micges_ is now known as micges
[19:24:06] <psha> skunkworks: but do you really want _2.4_ branch from buildbot?
[19:24:14] <psha> or 2.5~pre
[19:24:14] <psha> ?
[19:24:43] <Tom_itx> L84Supper, how many of those do you need?
[19:24:57] <Tom_itx> hammond is pretty good about sampling enclosures
[19:25:40] <Tom_itx> if you're developing a product that is
[19:26:00] <L84Supper> I was just surprised by the price when they add just a touch of style
[19:26:01] <skunkworks> psha: I want the latest released version. :)
[19:26:20] <Tom_itx> they gotta pay for the new tooling
[19:27:03] <Tom_itx> http://www.polycase.com/item/jb-35.html
[19:27:11] <Tom_itx> is another source for cases
[19:27:41] <psha> skunkworks: latest released is in standard repo
[19:27:58] <L84Supper> Tom_itx: hopefully thousands/yr, I was just browsing for off-the-shelf before I work on an extrusion die
[19:28:07] <psha> skunkworks: http://www.linuxcnc.org/lucid/dists/lucid/emc2.4/binary-i386/
[19:28:33] <skunkworks> psha: the computer in question doesn't have internet access. (so emc lucid livecd -> install deb if that is correct?)
[19:28:33] <psha> buildbot packages are 'latest unreleased' :)
[19:28:40] <psha> skunkworks: yes
[19:28:55] <psha> download them to some media and install with 'dpkg -i'
[19:28:58] <Tom_itx> or have one done on a 3d printer
[19:29:01] <psha> or with Double Click(tm)
[19:29:05] <skunkworks> heh
[19:29:08] <Tom_itx> i know a guy that will do that for you
[19:29:14] <skunkworks> That is what I was hoping.
[19:29:25] <skunkworks> psha: http://www.cnczone.com/forums/emc_linux_enhanced_machine_control/120166-need_bit_help_setting_up.html#post882366
[19:30:17] <L84Supper> Tom_itx: I have 3D printers here ;p
[19:31:23] <psha> skunkworks: emc 2.4.4 from livecd is not enought?
[19:31:29] <psha> or what's version is there?
[19:32:25] <psha> but yes, he may download emc2 2.4.6 deb package from repo i've posted and install it in place
[19:32:31] <psha> or at least try to do it :)
[19:32:38] <skunkworks> 2.4.3
[19:32:53] <skunkworks> heh
[19:33:08] <psha> buildmaster returned to this world only recently so i wont trust it at least for a week :)
[19:34:05] bootnecklad_ is now known as bootnecklad
[19:37:45] <Tom_itx> L84Supper, well damn.. aren't you special. i want one
[19:38:20] <Tom_itx> reprap or professional one?
[19:39:04] -!- LawrenceG [LawrenceG!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[19:40:13] <Tom_itx> what do most of you for a homebrew cnc use for a lead screw if you're not gonna get ball screws?
[19:40:26] <L84Supper> Tom_itx: all custom lab stuff, actually working on a DIY version that prints 0.5L/hr at 600 dpi (powder + binder) with one printhead
[19:41:12] <Tom_itx> this guy bought one for developing robotics stuff but he has funding backing him
[19:42:38] <bill2or3> tom, I've used acme screw + delrin nut
[19:42:41] <L84Supper> I can't bring myself to working with any of the reprap parts, reminds me too much of tinker-toys
[19:42:55] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[19:43:11] <Tom_itx> bill2or3, single or double lead? i'm guessing double lead would be harder to find
[19:43:24] <bill2or3> single, just what I had.
[19:43:24] <Tom_itx> stock drive products probably has em
[19:43:38] <skunkworks> bill2or3: Hey!
[19:43:40] <bill2or3> you can make a tap out of the acme, if you have a lathe.
[19:43:41] <Tom_itx> but they're $$
[19:43:44] * bill2or3 waves.
[19:43:53] <Tom_itx> yeah
[19:43:57] <Tom_itx> i've done that before
[19:44:49] <Tom_itx> does delrin hold up pretty well for most light work?
[19:45:04] <bill2or3> imo yea, let me find a pic.
[19:45:07] <L84Supper> unless you get the delrin too hot
[19:45:12] <Tom_itx> yeah
[19:45:19] <Tom_itx> it's higher temp by default though
[19:45:31] <bill2or3> http://protovision.com/wordpress/wp-content/uploads/P1030063.jpg
[19:45:32] <Tom_itx> not like albronze etc
[19:45:38] <bill2or3> doesn't show, but the Z-nut is delrin
[19:45:41] -!- micges has quit [Ping timeout: 264 seconds]
[19:46:13] -!- micges [micges!~micges@accg233.neoplus.adsl.tpnet.pl] has joined #emc
[19:46:26] <L84Supper> pcb router?
[19:46:43] <Tom_itx> you should have painted a barber pole on the stepper cap
[19:46:58] <bill2or3> yeah, pcb router.
[19:47:37] <Tom_itx> i didn't have much luck routing pcb but i don't have a hs spindle
[19:47:39] <bill2or3> more pics & stuff: http://protovision.com/2010/01/31/cnc-mill/
[19:47:48] <bill2or3> I used a proxxon, worked ok.
[19:48:12] <bill2or3> first try was a Bosch Colt, which didn't work so great.
[19:49:07] <skunkworks> bill2or3: http://www.youtube.com/watch?v=39q6kvrSBSk
[19:49:10] <L84Supper> there has to be a market for low cost dental cnc mill for the 3rd world (or US in 10 more years)
[19:49:14] <Tom_itx> i've got plenty of board stock though http://tom-itx.dyndns.org:81/~webpage/etching/Copper2.jpg
[19:49:56] <bill2or3> that one is a little bit bigger.
[19:50:55] <L84Supper> http://www.charlyrobot.com/2-Web_site_English/C-Applications-GB/10-Dental/Dental.htm
[19:52:02] <bill2or3> my dentist had a mill, but got rid of it.
[19:52:17] <bill2or3> I guess you pay per-use, DRM'd up the yang.
[19:52:49] <Tom_itx> bill2or3, are those steppers or servos?
[19:52:56] <bill2or3> steppers
[19:53:02] <L84Supper> how else can you bump up a simple part like a tooth to $1500 ea
[19:53:05] <bill2or3> all open-loop
[19:55:37] <Tom_itx> i'm lookin to beef this up or make a new one: http://tom-itx.dyndns.org:81/~webpage/boards/USBTiny_Mkii/Boxes/milling1.jpg
[20:05:02] -!- PCW [PCW!~chatzilla@99.88.10.65] has joined #emc
[20:14:37] -!- odiug [odiug!~guido@p57BD0D62.dip0.t-ipconnect.de] has joined #emc
[20:34:45] <Jymmm> Has anyone built a make-shift painting booth?
[20:35:06] -!- Emcrules_Laptop [Emcrules_Laptop!~Jean-Paul@CPE001310e6c94f-CM001bd71cb794.cpe.net.cable.rogers.com] has joined #emc
[20:37:26] <Jymmm> not just a structure, but something with negative airflow to it, I keep getting small dust particles on the piece while it's drying.
[20:38:50] <Tom_itx> i put a furnace blower in the celing of the garage
[20:39:09] <Tom_itx> stand them vertical when spraying
[20:39:13] <Tom_itx> if possible
[20:39:18] <Jymmm> You did, or you would?
[20:39:27] <Tom_itx> did
[20:39:57] <Jymmm> I can't really do the vertical too much, as I need to paint five sides.
[20:40:35] <Tom_itx> if you want small, get a stove vent
[20:40:46] <Tom_itx> with a fan exhaust
[20:40:55] <Tom_itx> or similar
[20:41:07] <Jymmm> I have a portable dust collector with 950 CFM =)
[20:41:44] <Tom_itx> when i spray, i wet the floor, turn the fan on and let the air clear and dive in
[20:41:59] <Tom_itx> you could use furnace filters on an inlet
[20:43:21] <Tom_itx> and wear a tyvec suit. you'll get dust off your clothes too unless you're careful
[20:43:59] <Tom_itx> but then i dunno the scale of your operation
[20:44:22] <Jymmm> Well, if I can get a negative airflow, anydust that wold settle on the workpiece hopefully would get sucked upwards instead.
[20:44:33] <Tom_itx> yeah
[20:44:37] <Jymmm> Well, if I can get a negative airflow, any dust that would settle on the workpiece hopefully would get sucked upwards instead.
[20:44:57] <Tom_itx> this thing has a 10 or 12" duct going to it
[20:45:08] <Tom_itx> vented outside
[20:45:18] <Jymmm> I'm thinking a hood of some sort, then attach the dust collector to that for the sucking action.
[20:45:41] <Tom_itx> i've seen that with a 'box' surround for small pieces
[20:46:27] <Jymmm> Yeah, nothing bigger than 24x24 inches, but might be many of them.
[20:46:53] <Jymmm> easier to paint many at a time, than just one
[20:46:53] <Tom_itx> 8" muffin fan on a cardboard box
[20:47:00] <Tom_itx> yup
[20:47:15] <Jymmm> that might work, I got a box of 4" fans
[20:47:30] <Tom_itx> if the paint goes out wet, it'll eventually melt the blades
[20:47:41] <Tom_itx> chemical reaction
[20:47:42] <skunkworks> logger[psha]: ping
[20:48:02] <Tom_itx> depending on what type of paint it is
[20:48:17] <Tom_itx> i've shot just about everything
[20:48:38] <Jymmm> eh, no biggy. I'd probably use he dust collector and a material filter, just not sure about air distribution
[20:48:45] <Jymmm> mostly clear coats
[20:49:50] -!- mhaberler has quit [Ping timeout: 276 seconds]
[20:50:05] -!- mrsunshine_ has quit [Read error: Connection reset by peer]
[21:00:35] <psha> skunkworks: may you fix link to 'latest stable' in your post on cnczone?
[21:00:43] <psha> to released versions and not buildbots
[21:05:44] <skunkworks> direct link to the download?
[21:08:51] <psha> no, to released versions
[21:09:00] <psha> to directory
[21:09:21] <psha> i think this http://www.linuxcnc.org/lucid/dists/lucid/emc2.4/binary-i386/
[21:10:52] <skunkworks> ah
[21:11:06] <skunkworks> (the latest one on the build bot isn't on that list)
[21:14:26] <skunkworks> http://www.cnczone.com/forums/882366-post6.html
[21:14:30] <skunkworks> better?
[21:15:57] -!- Valen [Valen!~Valen@ppp226-148.static.internode.on.net] has joined #emc
[21:16:45] <psha> i think it's more correct
[21:16:47] <psha> thanks
[21:21:09] -!- fatpandas has quit [Quit: leaving]
[21:22:56] -!- odiug [odiug!~guido@p57BD0D62.dip0.t-ipconnect.de] has parted #emc
[21:23:01] <skunkworks> psha: thank you - I thought there was a better location to link to.
[21:27:58] -!- markpictor [markpictor!~Mark@adsl-70-239-88-156.dsl.bltnin.sbcglobal.net] has joined #emc
[21:28:19] -!- mrsun [mrsun!~mrsun@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[21:28:38] <mrsun> hmm, when trying to associate my wlan with an ap all words work and comes out righ tin the essid, except my wlan name
[21:28:52] <mrsun> it becomes something like \12\x85\asd\!\4\afsdf\4\asfd\4 etc
[21:29:07] <mrsun> tried changing essid and it still comesout like that when i set it to the correct one :/
[21:29:15] <mrsun> are there updates of the rtai kernel somewhere?
[21:33:39] <markpictor> has anyone bought Encoder Products Company brand encoders?
[21:33:41] <markpictor> http://www.encoderoutlet.com/mall/buy-model15t.html
[21:34:20] <markpictor> there are some on ebay for $20 each (!) which sounds too good to be true
[21:34:36] <markpictor> $20 and *new*
[21:37:56] -!- Poincare has quit [Ping timeout: 276 seconds]
[21:39:59] -!- platius [platius!~platius@68.118.126.198] has joined #emc
[21:40:13] -!- platius [platius!~platius@68.118.126.198] has parted #emc
[21:42:19] -!- Valen has quit [Quit: Leaving.]
[21:43:20] -!- motioncontrol has quit [Quit: Sto andando via]
[21:45:52] <skunkworks> markpictor: looks like the 5000 line ones on ebay are also differential.
[21:45:56] <skunkworks> neat
[21:49:48] <markpictor> I don't see that one
[21:49:58] <markpictor> this is what I was looking at - http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=180604892568
[21:50:20] <skunkworks> http://cgi.ebay.com/ROTARY-ENCODER-15T-ACCU-CODER-5000-CPR-10-POLE-/180580776188?pt=BI_Control_Systems_PLCs&hash=item2a0b73f8fc
[21:51:33] <markpictor> I recently looked up BEI's price for a typical encoder. it was over $500
[21:51:35] <skunkworks> they are odd having 10 pole for brushless - but still...
[21:51:45] <skunkworks> you don't have to use that... ;)
[21:51:50] <markpictor> yea
[21:52:04] <cradek> what is a pole?
[21:52:10] <markpictor> I have a hard time believing that an $18 encoder is useable
[21:52:22] <markpictor> cradek: it's for commutation
[21:52:24] <skunkworks> the encoder has tracks for brushless comutation
[21:52:52] <cradek> so it has that AND regular quadrature?
[21:52:56] <markpictor> apparently
[21:52:59] <skunkworks> yes
[21:52:59] <cradek> wow
[21:54:50] <markpictor> reasonable shipping, unlike so many sellers - $5, $1 for additional items
[21:56:47] <skunkworks> I would be tempted...
[21:56:58] <skunkworks> they are differential (did I say that before?) ;)
[21:57:06] <markpictor> looks like 6 signals (ABZUVW) so I'm not sure how they do 10 pole commutation
[21:57:21] <markpictor> yes
[21:57:41] <markpictor> I sense a disturbance in their inventory :)
[21:57:50] -!- e3m [e3m!~IceChat7@217.96.27.164] has joined #emc
[21:58:11] <skunkworks> heh
[21:58:23] <skunkworks> 200 kHz for CPR 1 to 2540
[21:59:37] <markpictor> for some reason, the part number in the ebay pics is not in the datasheet at http://www.encoder.com/literature/datasheet-15th.pdf
[22:00:10] <skunkworks> I see that.
[22:00:31] <skunkworks> I also wonder what "This model does not have the shaft mounting screws."
[22:00:33] <skunkworks> means
[22:01:33] <skunkworks> I guess 200khz would be almost 6000rpm at 2048cpr (if I did that right)
[22:02:35] <markpictor> sounds like you have to drill and tap a hole yourself? or maybe it's meant for some kind of split clamp
[22:02:49] <skunkworks> still.... ;)
[22:02:52] <markpictor> there are slots in the id
[22:02:55] <markpictor> yea
[22:03:40] <markpictor> even if it turns out to be some strange non-grey encoding, 5k cpr is a lot
[22:05:00] <skunkworks> how many rpm would 5000cpr be at 3mhz reading?
[22:05:24] -!- Dannyboy has quit [Read error: Connection reset by peer]
[22:06:34] <markpictor> 3M/5k, right?
[22:06:39] <markpictor> 600 rpm?
[22:07:11] <skunkworks> must be more than 3mhz ;)
[22:08:06] -!- mrsunshine [mrsunshine!~mrsunshin@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[22:08:29] <skunkworks> looks like 10mhz on the 7i33 max
[22:08:36] <mrsunshine> has anyone answered on my question ? :/
[22:08:51] <skunkworks> I think 5k would be too high. the 2048 looks good.
[22:09:08] <skunkworks> I wonder if it is an encoder wheel or some sort of capasitive/magnetic encoder.
[22:09:44] <markpictor> mrsunshine: about wlan? sorry, no idea
[22:10:43] -!- Poincare [Poincare!~jefffnode@ip-188-118-11-57.reverse.destiny.be] has joined #emc
[22:10:50] <mrsunshine> hmm might it be the firmware that messes with me
[22:11:06] <mrsunshine> dont know if its a "LP-PHY" card or not
[22:13:27] -!- emcrules_mobile [emcrules_mobile!iucepj@69.41.165.101] has joined #emc
[22:14:45] <markpictor> the pdf says 8k rpm max, and 500khz max for 5000cpr encoders
[22:15:30] <markpictor> If I did the math right, that's 100 rev/s or 6000 rpm
[22:16:18] <mrsunshine> yeap might be wrong firmware
[22:16:33] <emcrules_mobile> thinking of using a D510 with a 32GB solid state hard drive. 32GB should be enough storage right?
[22:16:55] <skunkworks> markpictor: right - but you need something that will read that fast :)
[22:17:02] <markpictor> yea
[22:17:31] <markpictor> I doubt my parallel port will do it, even bathed in liquid nitrogen :)
[22:17:40] <markpictor> emcrules_mobile: that's plenty
[22:18:26] <skunkworks> 7i33 from mesa reads rs422 at 10mhz
[22:18:27] <emcrules_mobile> I thought so. thanks mark
[22:18:31] <markpictor> well, how many HD movies do you want to have on your cnc machine at any given time?
[22:18:36] <markpictor> :-P
[22:18:50] <markpictor> rs422?
[22:18:57] <skunkworks> differential
[22:19:02] <markpictor> ohhhh
[22:19:26] <markpictor> does it use rs422 for encoder interfaces?
[22:19:34] <skunkworks> or ttl
[22:19:40] <markpictor> probably ttl
[22:19:45] <skunkworks> both
[22:19:47] <emcrules_mobile> yes
[22:20:07] <markpictor> ok, when you said 422 I assumed a special chip to do it
[22:20:21] <emcrules_mobile> selectable by jumper
[22:20:22] <markpictor> I'm certain the fpga can handle far more than 10mhz
[22:20:24] <markpictor> ah
[22:20:38] <skunkworks> it has jumpers to switch each channel either ttl or differential
[22:20:45] <skunkworks> right
[22:21:05] <emcrules_mobile> I use it on my servos
[22:21:11] <skunkworks> markpictor: I think the limit of 10mhz is with sane filtering
[22:21:25] <markpictor> oh, that makes sense
[22:21:52] <markpictor> that's still *really* fast for an encoder
[22:23:45] <emcrules_mobile> yeah its nice hardware for the price
[22:26:05] <skunkworks> looks like 2 options with the 5i20 - 15 clocks or 3 - so with the 33mhz clock of the 5i20 you get a max reading of either 2.2mhz or 11mhz. if I did the math right
[22:26:58] -!- Dannyboy [Dannyboy!~Dan@12.192.111.2] has joined #emc
[22:27:00] <skunkworks> so at 2mhz and 5000cpr - does that come to 400rps or 24000rpm?
[22:27:15] <skunkworks> wow - I could really get lost ;)
[22:27:38] -!- Emcrules_Laptop has quit [Ping timeout: 240 seconds]
[22:28:06] <skunkworks> this is about when peter from mesa shows up ;)
[22:29:45] -!- tlab [tlab!~tlab@c-69-243-187-136.hsd1.in.comcast.net] has joined #emc
[22:30:31] -!- Dannyboy has quit [Remote host closed the connection]
[22:34:51] <markpictor> skunkworks: yes 24k rpm
[22:35:10] <skunkworks> wow
[22:36:01] <markpictor> If your calculator is set on hex when you enter the numbers, you get 666 rps
[22:36:11] <skunkworks> heh
[22:37:19] <markpictor> don't worry, though -- apparently historians have decided that the number of the beast was originally supposed to be 50 less than that :)
[22:40:01] <skunkworks> well - I bought a few of the 5000 line ones. If you can wait - I could let you know how they work :) (if they are still there) (it still says 'more than 10 available)
[22:40:33] <skunkworks> can't pass up something that cheap ;)(
[22:42:02] <skunkworks> even if it was 1/2 or 1/4 of that rpm - that is a minimum of 6000rpm..
[22:45:03] -!- tlab has quit [Ping timeout: 240 seconds]
[22:46:19] <markpictor> yea I bought 3
[22:46:24] <skunkworks> heh
[22:47:12] <markpictor> I was thinking that I would tell *you* how they worked :)
[22:48:06] <skunkworks> We have some older stock 2540cpr encoders that mount similar on our big machine. they where not differential though so we bought line drivers.
[22:48:12] <markpictor> I bought 2 NOS encoders several years ago. they sat until last spring, when I found out that they didn't work :/
[22:48:20] <skunkworks> it is a nice way to mount things.
[22:48:26] <markpictor> yea
[22:48:53] <skunkworks> oh well - I have some running to do..
[22:49:01] <skunkworks> markpictor: thanks for the heads up!
[22:49:13] <markpictor> no prob
[22:53:38] -!- skunkworks has quit [Ping timeout: 265 seconds]
[22:55:13] -!- emcrules_mobile has quit [Quit: Chatmosphere - IRC for Blackberry]
[22:56:47] -!- psha has quit [Quit: Lost terminal]
[23:06:26] -!- jstenback has quit [Ping timeout: 240 seconds]
[23:08:12] -!- tlab [tlab!~tlab@c-69-243-187-136.hsd1.in.comcast.net] has joined #emc
[23:14:47] <JT-Shop> anyone wanna help me shovel 6" of snow out of the new shop?
[23:15:47] <Jymmm> Got Charoal?
[23:15:52] <Jymmm> Got Charcoal?
[23:16:01] <Jymmm> or cord of wood?
[23:16:51] -!- mhaberler [mhaberler!~chatzilla@62-47-135-102.adsl.highway.telekom.at] has joined #emc
[23:18:05] -!- acemi has quit [Quit: WeeChat 0.3.2]
[23:20:43] <mrsunshine> JT-Shop, why do you have snow in the new shop ? :P
[23:21:48] <Jymmm> he was napping that day
[23:25:13] <JT-Shop> uh, no roof
[23:25:50] -!- jstenback [jstenback!~jstenback@64.151.104.172] has joined #emc
[23:28:16] -!- mrsun_ [mrsun_!~mrsun@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:29:25] -!- mrsunshine_ [mrsunshine_!~mrsunshin@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:29:53] -!- mrsunshine has quit [Ping timeout: 240 seconds]
[23:30:15] -!- mrsun has quit [Ping timeout: 276 seconds]
[23:34:08] -!- mrsunshine_ has quit [Read error: Connection reset by peer]
[23:34:09] -!- mrsun_ has quit [Read error: Connection reset by peer]
[23:35:17] -!- mrsun_ [mrsun_!~mrsun@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:35:28] -!- Dannyboy [Dannyboy!~Dan@12.192.111.2] has joined #emc
[23:36:18] -!- mrsunshine [mrsunshine!~mrsunshin@c-bc3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:38:34] -!- Valen [Valen!~Valen@ppp226-148.static.internode.on.net] has joined #emc
[23:41:05] -!- e3m has quit [Quit: The early bird may get the worm, but the second mouse gets the cheese]
[23:41:25] -!- micges has quit [Quit: Ex-Chat]
[23:46:26] -!- the_wench has quit [Remote host closed the connection]
[23:46:44] -!- the_wench [the_wench!~the_wench@host81-149-189-98.in-addr.btopenworld.com] has joined #emc
[23:46:44] markpictor is now known as markpictor-away
[23:53:19] -!- Techrat_CT [Techrat_CT!~TRat@unaffiliated/techrat] has joined #emc