#emc | Logs for 2009-02-13

[01:36:01] <Paragon27> Hello All, I have just purchased a second hand milling vice and it appears that the jaws need aligning what is the best method to achieve this?
[01:45:08] <fenn> disassemble it, remove the crud that is misaligning things, put back together
[01:49:34] <toastydeath> ^^^
[01:49:46] <toastydeath> how is it "misaligned," by the way
[02:26:13] <JymmmEMC> cradek: if you order from that place, get one of these too http://west.loadup.com/military/surplus/1570.html
[03:09:21] <jepler> if the MOSFETs are rated Vdss=30V and the driver chip is rated Vbat=50V, what other component would reduce the supply voltage range to only +24V? 25V-rated capacitors? http://www.sparkfun.com/commerce/product_info.php?products_id=9107
[03:11:44] <SWPadnos> are MOSFETS one of those devices that break in a very bad way if you exceed their maximum voltage for a few microseconds?
[03:12:41] <jepler> I dunno
[03:12:50] <jmkasunich> yes
[03:13:56] <jepler> hm, looks like sparkfun is just a reseller on that board -- this other site offers 24V and 50V versions with the same board layout (different MOSFETS) and notes that "absolute maximum voltage 30V" on the 24V one's page. http://www.pololu.com/catalog/product/755/specs http://www.pololu.com/catalog/product/756/specs
[03:14:00] <jmkasunich> at 21A, and assuming the MOSFETs switch in 50nS, you'll have 420A/uS dI/dT
[03:14:21] <jmkasunich> that will give you 4.2V across 10nanohenries of stray inductance (which isn't much)
[03:15:10] <jepler> (the 50V board's MOSFETs are 60V-rated)
[03:15:53] <jmkasunich> for reference, our 480V drives have nominal 650-680V DC busses, and trip on high bus at 815V
[03:16:19] <jmkasunich> we use 1200V IGBTs, and under short circuit conditions when we are turning off high currents, we need every bit of that 1200V
[03:16:21] <SWPadnos> for regen or just general tolerance-based overdesign?
[03:17:01] <jmkasunich> the 815V hi bus trip is to avoid tripping if you have a tiny bit of regen
[03:17:35] <jmkasunich> the 1200V IGBTs are to handle L*dI/dT voltage spikes at the IGBT terminals
[03:17:41] <SWPadnos> yep
[03:18:19] <jepler> so it sounds like using 30V parts and rating the board for 24V makes sense
[03:19:09] <SWPadnos> heh. you can either engineer and test and calculate and make complex protection circuits and all that
[03:19:21] <SWPadnos> or you can use a 20% safety margin and be done with it
[03:19:31] <SWPadnos> (I usually opt for option #2 :) )
[03:20:54] <jepler> makes perfect sense to me
[03:21:26] <SWPadnos> for small runs, especially when you're charging for your time, it also often benefits the customer
[03:21:41] <SWPadnos> for high quantity or sagety-related stuff you still have to do the math
[03:21:44] <SWPadnos> safety
[03:21:53] <SWPadnos> good thing my typing isn't a safety issue
[03:22:08] <jmkasunich> math is rarely enough
[03:22:30] <jepler> seems like math is useless without measurements
[03:22:32] <jmkasunich> gotta scope the voltages and see what you are actually getting
[03:22:33] <SWPadnos> analysis, math, testing, failure testing, lather, rinse, repeat
[03:58:26] <tomp> i bought 6 advantech panel mount computers and i find the insert key does not work. its not the kbd. i went around and tested 7 laptops and desktops here in taiwan do not change the cursor from ibeam to box, AND do not change from insert to overwrite. any ideas?
[04:05:52] <tomp> i've already asked the mfctr to respond but... chinese are always very very VERY slow to respond if they think they are wrong in any way. they begin by 'not understanding' the simple demonstration of BOX then IBEAM, then they have to find a way to say its not their fault, then make up some reason why it is impossible to fix, then i have to begin again explaining its just fskn borqd! argh!
[04:06:55] <tomp> gotta pick this up on another box, need this for W$ cad
[04:09:42] <SWPadnos> tom1, what were you using to see that the INS key wasn't working?
[04:22:21] <tom1> ah, notepad, worpad , a text input field
[04:22:30] <tom1> all of those at east
[04:22:41] <SWPadnos> ok, on Windows?
[04:22:46] <tom1> least ( and this kbd is just crap)
[04:22:47] <tom1> yep
[04:22:50] <SWPadnos> ah
[04:23:06] <SWPadnos> if there's an OpenOffice install anywhere, try that
[04:23:19] <SWPadnos> actually, I think wordpad should also show insert mode in the status bar
[04:23:26] <SWPadnos> even if the cursor doesn't change shape
[04:23:41] <SWPadnos> (unless you know for sure that the keyboards just aren't sending the insert keycode)
[04:23:46] <tom1> and to show them how it is supposed to be i used Kate on linux on my system
[04:24:03] <SWPadnos> just install Linux on the machines
[04:24:04] <SWPadnos> :)
[04:24:33] <tom1> i tried several kbds to see if it was that, then went around asking people to try it on theirs. they had no idea of what ibeam and box cursor were.
[04:24:57] <tom1> ok, off to change this country, see you a little later ;)
[04:25:16] <SWPadnos> good luck
[04:25:48] <SWPadnos> the I-beam is actually the mouse pointer shape over text areas
[04:26:03] <SWPadnos> it should just be a vertical bar cursor rather than the box
[04:26:33] <JymmmEMC> actuall it's BLOCK and underline, where BLOCK is insert mode
[04:26:55] <SWPadnos> I haven't seen an underline cursor since that last time I looked at a text terminal
[04:26:56] <tom1> no, right now in chat msg window, the mouse cursor is 2" away an a pointer, the text cursor is here and is a vertical bar ( not always i beam, agreed)
[04:27:09] <JymmmEMC> No, from DOS
[04:27:15] <JymmmEMC> or terminal, shell, etc
[04:27:19] <SWPadnos> well, you're stuck in the dark ages I guess
[04:27:30] <SWPadnos> oh, right, cmd.exe does that as well
[04:27:34] <SWPadnos> like a text terminal ;)
[04:27:40] <tom1> underline vs vert bar is according to api but thats no the issue, totally missing is wrong
[04:27:46] <JymmmEMC> SWPadnos: Nope, in the REAL Mode baby!
[04:27:58] <JymmmEMC> no drivers, virtual, or anything else to interfer
[04:28:05] <SWPadnos> real mode? well I guess nobody will ever need more than 640 k, huh
[04:28:18] <JymmmEMC> he said 512KB actually =)
[04:28:46] <JymmmEMC> SWPadnos: Hey, you know how to PXE boot ISO's?
[04:28:51] <SWPadnos> nope
[04:28:58] <tom1> yeh the idea of a driver for a kbd makes me cringe
[04:29:00] <SWPadnos> never used a amachine with PXE
[04:29:28] <JymmmEMC> SWPadnos: I'm finding it doesn't exist (sorta kinda), but is possible.
[04:29:37] <SWPadnos> tom1, ever since IBM separated keyboard reading tasks on a microcontroller, it's been necessary :)
[04:30:03] <SWPadnos> maybe the old Cherry keyboards weren't a direct matrix thing, but I don't know
[04:30:04] <JymmmEMC> eh, code pages
[04:30:20] <tom1> ah thats an idea, the code page
[04:30:36] <JymmmEMC> tom1: That's why I said DOS
[04:30:51] <SWPadnos> damn. I knew I shouldn't have eaten all the Doritos last night
[04:31:08] <tom1> send SWPadnos some salt
[04:31:21] <SWPadnos> maybe I'll have cheese and crackers instead
[04:31:30] <toastydeath> in soviet russia, cheese and crackers have YOU
[04:31:34] <SWPadnos> good Cabot private stock cheddar
[04:31:47] <SWPadnos> (as sold at Costco ;) )
[04:32:11] <tom1> beer cheese crackers & midget wrestling, thats an nite's entertainment
[04:33:01] <tom1> they have a costco here on WenXin Lu, and my card works :)
[04:33:10] <SWPadnos> cool
[04:33:15] <SWPadnos> any local stuff there?
[04:34:15] <tom1> a little, peanut butter, shampoo, no after shave (heh), mostly local, and costco for 1 guy is a it useless
[04:34:28] <SWPadnos> ah
[04:34:33] <tom1> like 50 lots of ramen noodle
[04:34:50] <tom1> big packages of lamb and veggies for hot pot...
[04:35:02] <SWPadnos> I've been to Costcos in many states (Hawai'i, Florida, California, Vermont ...) plus England, and though they have a lot of the same stuff, there are definitely differences
[04:36:14] <tom1> and really really cheap furniture (dont sit on it ) dont push the broom too hard, an dont expect it to reach to the floor (oh ys i'm having a jolly time here)
[04:36:47] <SWPadnos> get two brooms and tape them together
[04:36:54] <tom1> enuf bitchin, back to work :)
[04:36:56] <SWPadnos> heh
[04:39:11] <tom1> oh, they sell these face masks, like painters masks but with patterns/logos. needed if you walk outside in the smg (serious visible air pollution here ) all scooter drivers wear them ( and wear thier coats backwards to shed dirt)
[04:39:29] <tom1> enuf enuf bye bye
[04:40:06] <SWPadnos> see you later
[04:56:04] <toastydeath> i am trying to teach myself vector calculus
[04:56:36] <toastydeath> this has been put off for a couple weeks after i saw the first set of practice problems and realized i don't know how to deal with logarithms even algebraically, much less anything else
[04:56:42] <SWPadnos> planning of tensors some time?
[04:56:49] <SWPadnos> s/of/on/
[04:57:04] <toastydeath> seems kind of useful, yeah
[04:57:16] <SWPadnos> could be, for quantum physics
[04:57:36] <toastydeath> well there's some generalizations of stress/strain that apparently goes really nice into tensors
[04:58:07] <SWPadnos> hmm, interesting
[04:58:13] <SWPadnos> sort of ;)
[04:58:28] <toastydeath> so i'd really like to be able to solve the problems in my classes by dumping matrices into mathematica, but i can't do that if i have no idea what i'm doing.
[04:58:31] <toastydeath> that's the end goal.
[04:58:43] <SWPadnos> ah, knowing what you're doing is a good goal
[04:58:51] <SWPadnos> and/or why
[04:59:18] <toastydeath> in the mean time i've filled a notebook trying to figure out integrating and deriving logs
[04:59:27] <toastydeath> what am i doing with my life, i don't know
[04:59:35] <toastydeath> but it apparently involves a whole lot of calculus at the moment
[04:59:57] <SWPadnos> luckily, integrating and differentiating logs is trivial
[05:00:13] <SWPadnos> it's when you combine them with things that it gets hard :)
[05:00:19] <toastydeath> yeah, that's what i'm dealing with
[05:00:48] <toastydeath> recognizing all the forms is not easy for me, anyway
[05:01:13] <toastydeath> i'll probably get stuck and give up in a day or two
[05:01:19] <SWPadnos> heh
[05:13:14] <JymmmEMC> SWPadnos:
[05:14:19] <JymmmEMC> SWPadnos: I know candlepower is a measure of brightness. but is there a ref that you know of to wattage by chance?
[05:14:36] <JymmmEMC> and/or btu
[05:14:39] <SWPadnos> there are several measures of light output
[05:14:51] <SWPadnos> lumen, candlepower, lix ...
[05:14:53] <SWPadnos> lux
[05:14:55] <JymmmEMC> I'm interested in therol, not optiocal
[05:14:59] <JymmmEMC> thermal
[05:15:10] <SWPadnos> one candle is about equivalent to a 400W heater
[05:15:21] <SWPadnos> I have no idea if that's a candlepower though :)
[05:15:23] <JymmmEMC> SWPadnos: and your source of that?
[05:15:31] <SWPadnos> I read it somewhere
[05:15:32] <SWPadnos> :)
[05:15:36] <JymmmEMC> Candlepower (abbreviated as cp) is an obsolete scientific unit of luminous intensity based on the light emitted from a candle made to a specified formula
[05:15:55] <SWPadnos> yes
[05:16:20] <JymmmEMC> SWPadnos: well, if that's accurate (400W), I'm screwed then
[05:16:34] <SWPadnos> that has nothingto do with the light output though
[05:16:49] <SWPadnos> that's for actually burning a candle
[05:16:58] <JymmmEMC> I dont care about optical in the least, just it's thermal output
[05:17:01] <SWPadnos> that's why they take the chill off nicely in a cold room
[05:17:12] <SWPadnos> well candlepower has no bearing on thermals
[05:17:35] <JymmmEMC> SWPadnos: Right, and why we have 7 candles going in the living room right now
[05:17:46] <SWPadnos> an LED uses much less power (and therefore wastes/emits less heat) to make "1 candlepower" of light
[05:17:51] <SWPadnos> heh
[05:18:09] <SWPadnos> so that's about like 2 space heaters on high
[05:18:22] <SWPadnos> (at 1400W each)
[05:18:52] <JymmmEMC> SWPadnos: But, what I was considering is using the solar panel I have to do the same thing, mustly to reduce the moisture in the air in the back of the house. but it itself (full sun) produces 24W, intended to recharge gelcells.
[05:19:05] <JymmmEMC> s/mu/mo/
[05:19:14] <SWPadnos> better to open the windowshade and let the sun do it directly
[05:19:38] <SWPadnos> here are some better illumination conversions: http://en.wikipedia.org/wiki/Candela
[05:19:48] <JymmmEMC> SWPadnos: It's at night when the dew point rises.
[05:20:00] <SWPadnos> and the sun has gone down
[05:20:08] <SWPadnos> (preventing the window or the solar panels from helping)
[05:20:11] <JymmmEMC> I will NOT burn cnaldes while sleeping
[05:20:18] <SWPadnos> good plan
[05:20:34] <eric_unterhausen> curses, foiled again
[05:20:35] <tom1> why isnt the sun out at nite when it woud be useful?
[05:20:39] <JymmmEMC> So, I'm trying to think of an alternative, especially since I already have all these batteries
[05:21:21] <JymmmEMC> This is what I get for watching a damn green planet episode!
[05:21:52] <SWPadnos> tom1, that's what the moon is for
[05:21:56] <JymmmEMC> The solar panel I have is really intended to recharge battery for radio comms, not be a humidifier
[05:22:00] <SWPadnos> therefore it's the more important of the two
[05:22:21] <SWPadnos> may I suggest a temperature-controlled space heater?
[05:22:56] <JymmmEMC> SWPadnos: We thought of that, but due to being 300% over baseline had a $400 electric bill twice in a row
[05:23:07] <SWPadnos> oh
[05:23:13] <JymmmEMC> yeah... OH!
[05:23:39] <SWPadnos> do you have a gas feed, or just electric?
[05:24:01] <JymmmEMC> rated for gas heating, not electric, thus the baseline overages
[05:24:20] <SWPadnos> bu tdo you actually have gas heating equipment installed?
[05:24:27] <JymmmEMC> yes
[05:24:40] <SWPadnos> you need to do one of two things (if I'm understanding the problem right)
[05:24:58] <SWPadnos> option 1 is to add energy to your apartment at night (or evening)
[05:25:17] <SWPadnos> option 2 is to store energy during the day and release it at night (or evening)
[05:25:29] <SWPadnos> option 1 is expensive
[05:25:50] <SWPadnos> option 2 is very expensive at first, and may require structural changes to the building
[05:25:57] <JymmmEMC> we rent
[05:25:57] <SWPadnos> but then it gets less expensive
[05:26:02] <SWPadnos> exactly
[05:26:35] <SWPadnos> if you actually need 2000-3000W during the night, then gas heat is likely going to be your least expensive option
[05:26:36] <tom1> option 3, devolve to lizard cold blooded less need for heat
[05:27:04] <SWPadnos> well, if you remember the words of Mr. Tyrel, you can't fiddle with your genese that way
[05:27:12] <SWPadnos> at least not if you're a replicant
[05:27:14] <SWPadnos> :)
[05:27:24] <JymmmEMC> Not that I could do this indoors... but lighting a coleman lantern puts out a *LOT* of heat... I use to warm the tent when going camping by placing a lit lantern in the tent 30 minutes before hitting the sack. worked great.
[05:27:33] <JymmmEMC> and used very little fuel
[05:28:16] <SWPadnos> sure, if all you want is a shot of heat before going to bed, that's easy
[05:28:27] <SWPadnos> install a set-back heater control
[05:28:42] <SWPadnos> $50 plus a little more heating bill
[05:28:47] <SWPadnos> ($50 once)
[05:29:11] <JymmmEMC> these heaters are ancient, no ext thermostat on them
[05:29:22] <JymmmEMC> turn a knob on them
[05:29:34] <JymmmEMC> a very sensative knob at that
[05:29:42] <SWPadnos> ok, the portable gas things that are connected to the wall via a somewhat flexible hose?
[05:29:58] <SWPadnos> or a big ugly rectangular thing
[05:33:09] <toastydeath> never can have too many big ugly rectangular things
[05:38:59] <tom1> wodka (shot of heat)
[06:01:38] <fenn> JymmmEMC: turn the air conditioner around in the window
[06:25:30] <JymmmEMC> SWPadnos: big ugly thing
[06:25:37] <JymmmEMC> fenn: lol
[06:59:56] <tom1> looks like i just dont know how insert/overwrite works, it seesm to be application specific and is not the kbd nor computer. i had no idea that it was so not-universal, amost rare.
[07:00:38] <tom1> its those damn programmers
[10:32:30] <pjm_> archivist are u here? if so can u remember where we got them 220nM steppers, forgotten the company name
[10:33:22] <archivist> http://www.arceurotrade.co.uk/Catalogue/Stepper-Motors
[10:34:06] <pjm_> ahh nice one, thanks ;-)
[10:34:41] <archivist> getting ready for a van move today
[10:35:16] <Paragon27> Hello All, I have just purchased a second hand milling vice and it appears that the jaws need aligning what is the best method to achieve this?
[10:37:10] <pjm_> whats wrong with the jaws? are they not flat etc
[10:37:26] <archivist> re machine possibly
[10:38:19] <Paragon27> Thats right there not flat and do not close together correctly (slight gap) the jaws have two screws on them for removal.
[10:38:48] <archivist> make new jaws
[10:39:19] <Paragon27> Could I use parallels? or is this an overkill?
[10:39:46] <pjm_> probably an overkill
[10:40:03] <pjm_> use some steel plate, surface grind them flat
[10:40:10] <pjm_> and cut/drill to suite
[10:40:41] <Paragon27> right.... Dont have a service grinder though.
[10:40:54] <archivist> or softer depends on work you will hold
[10:44:01] <Paragon27> It look likes the guy that had the vice prior to me has made the draws hastily. I could attempt to machine them as they appear ti me fab'd out of mild steel. I give it a go. Thanks for your help. How have you all been? It's been a little while since Iv'e visited due to work commitments etc.
[10:45:25] <archivist> I just had 3 days in London at the British Library (conference)
[10:45:40] <Paragon27> cool...
[12:02:53] <piasdom> what module/thingy is being loaded for my video card ? http://pastebin.com/d7be7ecbb
[12:07:39] <piasdom> oops..g'mornin all
[12:47:30] <pjm_> piasdom probably #
[12:47:30] <pjm_> amd64_agp 11140 1
[12:47:30] <pjm_> #
[12:47:30] <pjm_> agpgart 31024 1 amd64_agp
[12:47:39] <pjm_> do an LSPCI and check what type of vid card u have
[12:50:21] <pjm_> ie i have here in a box;
[12:50:22] <pjm_> lspci
[12:50:23] <pjm_> 01:00.0 VGA compatible controller: Matrox Graphics, Inc. MGA G200 AGP (rev 03)
[13:14:08] <piasdom> did a lspci Nvidia Corporation NV44A [GeForce 6200]
[13:19:55] <piasdom> i'm still having realtime errors. i removed the controller and gave it it's own power, just connected to comp by printer port
[13:21:51] <piasdom> i got another video card(above) and still have the error. so i was wondering if linux is still useing the old video control
[13:55:50] <pjm_> have u run the latency checker?
[13:57:02] <archivist> pjm_, fleaybay 330306280213
[13:57:59] <pjm_> nice, i'd like the linear display plug in too
[13:58:04] <pjm_> and some probes ;-)
[13:58:35] <archivist> see sellers other items
[13:58:43] <pjm_> yeah just checking then
[13:58:45] <pjm_> them
[14:16:20] <Ghost> hey folks
[14:16:28] <Ghost> Ghost is now known as Guest15482
[14:16:38] <Guest15482> Guest15482 is now known as `Ghost
[14:16:56] <`Ghost> `Ghost is now known as shadowtec
[14:18:03] <BigJohnT> hi
[14:19:13] <shadowtec> trying to get my head around emc2 (well right now watchig youtube vids on emc2) I got my board working (stepper mm config) but it only runs in 1 direction
[14:20:43] <BigJohnT> direction not on right pin?
[14:22:39] <shadowtec> well thats what I thought but the manual says it is in the right direction (I am going to see about hooking it into a salese logic analizer) I need to get another machine so I can run emc2 on
[14:24:01] <shadowtec> or just try it on each pin till I get something :) (I know the pulse one is right)
[14:24:41] <BigJohnT> what kind of "board"?
[14:25:36] <BigJohnT> did you use the stepconf wizard?
[14:25:45] <shadowtec> something I got off ebay from HK (I got myself to blame) no manual for it I had to hunt that down
[14:26:14] <shadowtec> getting stuff in OZ is a total pain
[14:26:56] <shadowtec> still waiting on the motor coupplers from RS (1 month later)
[14:26:57] <BigJohnT> OZ?
[14:27:18] <shadowtec> australia
[14:27:28] <BigJohnT> ok
[14:27:55] <BigJohnT> so did you use the stepconf wizard to generate your config?
[14:28:36] <shadowtec> yes the board did not even work with that (give me a min and I try it again with a fresh mind) I used the steapper mm config
[14:30:14] <shadowtec> I was cranky that night and may of missed something
[14:31:32] <BigJohnT> it is easy to do
[14:31:58] <shadowtec> yeah I been very busy this week to do anything on this project (I want it to mill pcb boards)
[14:32:09] <shadowtec> brb rebooting
[14:36:55] <piasdom> pjm_: i ran the latency-test and it goes in the millions
[14:37:14] <pjm_> and u have all power management stuff disabled ?
[14:37:30] <pjm_> cpu speed throttling etc all off
[14:38:00] <piasdom> yes
[14:39:06] <pjm_> and has it always been bad or just recently been happening?
[14:39:41] <piasdom> always...have the comp about a year
[14:40:11] <piasdom> just trying to figure out what's up
[14:41:09] <piasdom> plm:thanks for your help, i'll do some more reading
[14:41:13] <eric_unterhausen> that has to be a bios issue
[14:42:13] <piasdom> i thought maybe after i installed the card, linux was still useing the old drivers
[14:43:07] <pjm_> u could boot it into txt mode only, probably runlevel2 and re-try the latency
[14:43:49] <piasdom> plm:thanks, i'll try that
[14:44:30] <piasdom> pjm_:
[14:45:17] <piasdom> bbl
[15:18:20] <Guest176> Guest176 is now known as skunkworks_
[15:44:59] <itt> why do i get this error....the gnome session manager was unable to lock the file '/home/itt/,ICEauthority. ?
[15:55:41] <cradek> gnome constantly spews all sorts of messages. is something not working?
[16:05:23] <Dallur> Anyone know when the next release will happen ?
[16:05:39] <alex_joni> Dallur: soonish ;)
[16:06:05] <alex_joni> Dallur: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Emc2.3Status
[16:06:14] <Dallur> alex_joni: c.a month ?
[16:06:30] <alex_joni> April 15: emc 2.3.0 packages on linuxcnc.org
[16:06:36] <alex_joni> at least that's the plan
[16:06:40] <Dallur> ok, great thanks
[16:06:58] <Dallur> I have a person asking me when the plasma changes will make it into Ubuntu :)
[16:07:55] <alex_joni> the config changes?
[16:08:56] <itt> cradek: it comes up after rebooting...now i can't login
[16:10:24] <Dallur> alex_joni: I also did a small code change
[16:10:35] <Dallur> alex_joni: but it's essential for the new config :)
[16:12:20] <alex_joni> humm.. now why did I miss that? :)
[16:13:29] <alex_joni> updown.comp ?
[16:14:06] <alex_joni> * alex_joni needs to bbl
[16:17:38] <itt> when i try to login i get that message....no matter what i do
[16:18:10] <itt> how do reset permission for itt ?
[16:20:26] <itt> itt is now known as piasdom
[16:20:29] <SWPadnos> itt, go to a console terminal (ctrl-alt-F1), login, then sudo chown itt:itt ~/.ICEauthority
[16:20:53] <SWPadnos> according to this thread: http://ubuntuforums.org/showthread.php?t=565554
[16:21:03] <piasdom> SWPadnos: thanks
[16:21:15] <SWPadnos> if you access the file as root, it becomes owned by root and you no longer have access to it
[16:24:02] <itt> SWPados:can you repeat that...lost my other win
[16:24:17] <SWPadnos> http://ubuntuforums.org/showthread.php?t=565554
[16:24:23] <SWPadnos> itt, go to a console terminal (ctrl-alt-F1), login, then sudo chown itt:itt ~/.ICEauthority
[16:24:35] <itt> thanks
[16:24:37] <SWPadnos> I don't know that it will work, but it seems to be the recommendation from people in that thread
[16:24:38] <SWPadnos> sure
[16:26:32] <itt> brb...(hopefully in desktop)
[16:32:51] <RonA> Good day have not done this before so here it goes - does anyone know i can determine from my hal file if my step pin is set to active high
[16:36:11] <BigJohnT> are you wanting to see it change states or how it is configured?
[16:36:59] <RonA> how it is configured
[16:38:14] <cradek> put a halmeter on it
[16:38:49] <cradek> oh, wait, that won't work
[16:38:54] <BigJohnT> you can see it with machine/show hal configuration
[16:39:05] <BigJohnT> pins/parport/0
[16:39:22] <SWPadnos> heh
[16:39:38] <SWPadnos> 1) the stepgen always outputs a high step pulse
[16:40:01] <SWPadnos> 2) you can look at the -invert parameter for the parallel port pin you're using for step output
[16:40:20] <SWPadnos> BigJohnT, should be parameters, unless those have been changed to pins recently
[16:40:51] <BigJohnT> I was just looking at the output pin...
[16:41:28] <SWPadnos> it kinda depends on doublestep though, I think
[16:41:34] <SWPadnos> -invert doesn't
[16:41:51] <BigJohnT> * BigJohnT gotta go now
[16:41:59] <SWPadnos> see you
[16:42:37] <RonA> Thanks this helps alot
[16:44:33] <SWPadnos> did you use stepconf to make this config?
[16:51:10] <RonA> My initial config was done with wizard but i found wizard to be a problem as i was making adjustments directly
[16:51:20] <SWPadnos> ok
[16:51:37] <SWPadnos> looking at the "invert" checkbox is the easiest way to check with stepconf :)
[16:51:38] <RonA> gota go now - all the best
[16:51:44] <SWPadnos> see you, good luck
[16:53:12] <piasdom> hello again
[16:53:42] <RonA> actually i quess i could use the wizard to view stap but i have deleted it from the menu. any idea how i start the wizard with out the menu
[16:55:17] <RonA> ubuntu and linux is still pretty foreign to me
[16:55:26] <SWPadnos> Applications -> Accessories -> Terminal
[16:55:28] <SWPadnos> stepconf
[16:55:31] <Dallur> alex_joni: yup added a reset to updown
[16:56:57] <RonA> Thanks - now i really gotta go
[17:03:12] <SWPadnos> see ya
[17:11:01] <toastatwork> wow
[17:11:09] <toastatwork> third round of layoffs and i'm still here
[18:16:57] <fenn> toastatwork: maybe you should ask for a raise :)
[18:29:50] <toastatwork> i asked earlier this year!
[18:29:52] <toastatwork> well
[18:29:54] <toastatwork> late last year
[18:30:25] <toastatwork> if sales comes back (and the company doesn't go under) i'll probably get one
[18:31:05] <toastatwork> also if we don't have another three rounds of layoffs wherein i get laid off
[18:31:13] <toastatwork> high risk, this whole venture
[18:31:19] <toastatwork> can't say i approve
[18:31:52] <skunkworks_> Good luck - I think we all need it.
[18:32:33] <archivist> the big question...jump before the ship sinks...or wait for the pay off when redundancy comes
[18:32:37] <shrdlu-> hm, does G64 persist through the job if I place it at the start of the file
[18:32:39] <shrdlu-> ?
[18:33:28] <toastatwork> i'm going to wait it out and see what happens, because i like working here - if not, I'm going to call around the diamond turning companies and see if they're still looking
[18:33:49] <toastatwork> they have a bear of a time trying to get people to run DT lathes, so i've got a vain hope there's something out there
[18:36:59] <skunkworks_> shrdlu-: yes. unless you have a g61 in there
[18:37:22] <shrdlu-> ok. I'll forget about that helping me then :)
[18:38:16] <shrdlu-> I kinda understand why I can't do my raster thing now. I guess I need to do some kind of timed firing of the laser, instead of telling the axis to stop at a certain point then fire it
[18:38:35] <shrdlu-> is there any way to do timing dependent commands?
[18:39:00] <skunkworks_> I thought the whole point of the z axis was timed firing?
[18:39:02] <shrdlu-> like, 'after you've fired x pulses, then do this..'
[18:39:57] <shrdlu-> it doesn't make any difference whether I use the x or spindle, the axis still thinks it has to come to a halt when it reaches the end of the move
[18:40:02] <shrdlu-> I think
[18:40:06] <shrdlu-> x/y
[18:40:27] <shrdlu-> or the start of the move, both I guess
[18:40:54] <shrdlu-> erm sorry 'z or spindle'
[18:44:11] <shrdlu-> every move has a stop does it not?
[18:45:31] <shrdlu-> oooh
[18:46:28] <shrdlu-> I just took out all the spindle commands and it's scanning perfectly
[18:46:44] <shrdlu-> or at least better
[18:47:11] <skunkworks_> heh - so you had spindle commands with z?
[18:48:07] <shrdlu-> I've been swapping between both
[18:48:27] <shrdlu-> neither work well
[18:48:57] <shrdlu-> if I could get it to fire in time to how it's moving now, with only x and y command, it would be easily good enough
[18:53:58] <skunkworks_> is it on-off - or does it have to be proportional?
[18:54:17] <shrdlu-> on off is fine for the moment
[19:09:04] <pjm_> btw could someone just confirm about 3phase motor direction, to reverse direction u swap 2 of the phases?
[19:09:12] <SWPadnos> yes
[19:10:37] <pjm_> ok thanks ;-)
[19:27:19] <shrdlu-> hm some guy sent me an extra $10 today because he thought I was undercharging
[19:28:22] <archivist> whats his address, we need customers like that
[19:28:33] <shrdlu-> hah
[19:29:42] <shrdlu-> he hasn't got his stencil yet, he might be less inclined to do that after he sees the mess i've made of it with this new setup ;)
[19:31:12] <shrdlu-> although it does seem to be scanning well now. I had to drop the top speed to get the acceleration high enough for it not to 'jump'
[19:31:18] <shrdlu-> and get rid of all the rapid moves
[19:31:49] <shrdlu-> now back to the problem of synching this up :/
[19:31:58] <archivist> set the maximum rate correctly then put rapids back
[19:32:31] <shrdlu-> ah, yeah, good point
[19:50:24] <shrdlu-> guess I'll have to have different setups for rastering and vector cutting
[20:14:28] <UncleGemc> anyone make their own stepper motor driver board from spare parts, such as a printer?
[20:15:13] <archivist> UncleGemc, some do but thats never as good
[20:16:25] <UncleGemc> archivist: I don't expect it to be, but it would be a great learning process
[20:17:09] <UncleGemc> I have multiple computers and various other electronics around, I might as well try to build one from scrap then when I get the real goods things should go much smoother.
[20:17:28] <archivist> you can go from the unipolar stepper drivers to switching bipolar
[20:17:47] <shrdlu-> I dunno man, you can get a 4 axis 1.5amp controller for like $50 these days
[20:18:00] <shrdlu-> plugs straight in
[20:18:16] <shrdlu-> driver, sorry
[20:19:12] <UncleGemc> It seems I could buy things, but I think I could go red-green show all the way and make it work :)
[20:20:16] <shrdlu-> ah, I see all my hobby stuff as potential income, so faster is always better for me
[20:20:45] <fenn> sounds like its making real income now
[20:21:16] <shrdlu-> sure, this one has been upgraded to a small business :)
[20:22:01] <shrdlu-> http://www.smtstencil.co.uk/
[20:22:19] <shrdlu-> that's what I'm doing, now with EMC. (assuming I can get it working right)
[20:22:55] <fenn> i wonder if one could do that with black plastic film and a CD-R laser diode
[20:23:26] <fenn> what sort of laser are you using?
[20:23:39] <shrdlu-> 40W CO2
[20:24:00] <fenn> how long does it take to cut a typical board?
[20:24:06] <fenn> like the one in the photo
[20:25:04] <shrdlu-> well, up till now I've been using the plotter card and the crappy software that comes with it. To get a high quality cut it was taking about 30-60 mins on average
[20:25:10] <shrdlu-> because I have to raster them
[20:25:23] <shrdlu-> cutting them was never possible with old setup
[20:25:36] <fenn> huh raster seems like the wrong way to do it for sure
[20:25:36] <shrdlu-> I'm hoping with more control via EMC I can speed things up
[20:25:53] <fenn> people send you gerber files right?
[20:25:57] <shrdlu-> yeah
[20:26:19] <fenn> you could try gcam.js.cx
[20:27:05] <shrdlu-> that looks nice, but how would it help?
[20:27:32] <fenn> plot around the parts you want to cut out instead of raster scanning
[20:27:43] <fenn> i think it'd be many times faster
[20:28:02] <shrdlu-> it's faster, sure, but the quality is nowhere near as good
[20:28:44] <shrdlu-> if I can get PWM control of the laser and work out the routines properly, hopefully it'll be better and faster
[20:29:14] <shrdlu-> but if I just put a gerber straight in there, at the moment, snot good at all
[20:29:25] <shrdlu-> and was completely useless on the old setup
[20:29:36] <fenn> you need to use some kind of cam software to offset the toolpath (to account for beam width)
[20:30:11] <shrdlu-> yah I think that'd help
[20:30:44] <geo01005> I'm looking for some suggestions for how to get an analog input into EMC.
[20:31:20] <shrdlu-> that's cool that GCAM it loads gerbers and outputs gcodes though. I hadn't found a tool to do that yet
[20:31:21] <shrdlu-> thanks a lot
[20:31:36] <shrdlu-> I was having to go into windows
[20:32:27] <jepler> geo01005: in terms of ADCs that are compatible with emc, or in terms of how to use the analog value in gcode once you've got it from the ADC?
[20:33:03] <geo01005> I'm looking for hardware to aquire the analog signal.
[20:33:28] <SWPadnos> I can sell you an expensive board that plugs into a Mesa FPGA card
[20:33:39] <fenn> heh
[20:33:48] <SWPadnos> 6 inputs, 8 outputs, scans at about 175 KHz max, all 16 bit resolution
[20:34:11] <fenn> geo01005: try hacking a usb joystick and using hal_input
[20:34:12] <geo01005> I'm running a Mesa 7i43, and it would be great if the driver could be modified to to just read a simple SPI or i2c ADC.
[20:34:22] <SWPadnos> simultaneous sampling and simultaneous output update (but in and out aren't necessarily synchronous)
[20:34:42] <SWPadnos> there is a card coming from Mesa (7i64?) that will have some analog I/O I think
[20:34:54] <jepler> geo01005: yes, there's a new card from mesa, but it's not supported in emc2 yet.
[20:34:59] <SWPadnos> and there's an SPI block for the FPGA, but I don't know that it's supported by the driver yeu
[20:35:01] <SWPadnos> yet
[20:35:06] <fenn> SPI or i2c would be great
[20:35:33] <fenn> wasnt there a multi-channel SPI driver for the parport? or did i imagine that?
[20:35:37] <jepler> for cards supported today, the only one I know of is the "motenc-lite" 7541 with 8 analog inputs
[20:35:38] <geo01005> What I really want to read is some thermocouples for the Rep-Rap project.
[20:35:43] <archivist> geo01005, or write a driver to call any other hardware
[20:36:09] <fenn> geo01005: does it have to be a thermocouple? you could use a thermistor in a 555 circuit
[20:36:12] <SWPadnos> yeah, a driver for a measurement computing board would probably be pretty easy
[20:36:23] <SWPadnos> if they still have something like the CIO/DAS08
[20:36:36] <archivist> geo01005, see http://emcar.sourceforge.net/ they did an input driver
[20:38:45] <SWPadnos> shrdlu-, I don't think I ever saw what version of EMC2 you're using - which is it?
[20:38:51] <jepler> if speed is not important, you could also just bit-bang SPI / i2c on any compatible I/O pin
[20:39:12] <fenn> archivist: they did? "Presently, the system requires stopping at each measurement point while the measurement data are triggered and recorded. Perhaps, with some help from the gurus of EMC2 and the HAL, the measurements can be made "on-the-fly" in the future?"
[20:39:22] <fenn> i hardly call that a driver
[20:40:02] <archivist> fenn true done with an external prog/grab/whatever
[20:40:23] <archivist> and define stop
[20:40:53] <fenn> zero relative velocity
[20:40:55] <geo01005> Well so one option is to use a Arduino to get the signal and then trasmit over the serial line/USB, but no real time, and no guarantees.
[20:41:00] <archivist> they have more issues with the analyzers settling as well
[20:42:19] <fenn> geo01005: what about the 555 circuit or some other voltage to frequency idea?
[20:42:43] <SWPadnos> frequency measurement in software is unlikely to be very accurate
[20:42:53] <SWPadnos> unless you use very low frequencies
[20:42:54] <archivist> and there is delay
[20:43:09] <fenn> bah
[20:43:24] <archivist> not realtime :)
[20:43:33] <fenn> PWM?
[20:43:33] <SWPadnos> it may be ok though, since the encoder/counter module has a velocity estimate
[20:43:37] <SWPadnos> it's realtime
[20:43:55] <SWPadnos> just high latency due to the low frequency :)
[20:44:04] <archivist> realtime can and is overused as a term
[20:44:18] <SWPadnos> yes indeed
[20:44:21] <geo01005> Would it be possible to augment the Hostmot2 driver to read SPI devices?
[20:44:52] <geo01005> I suppose a new bitfile would have to be made as well.
[20:45:15] <geo01005> That would open up lots of applications for the Mesa devices.
[20:45:44] <SWPadnos> it's planned, once the hardware is released (as far as I know)
[20:46:03] <jepler> yes, it would require an updated firmware and hal driver though
[20:46:06] <SWPadnos> the driver should be maintained such that it supports all the "standard" hardware that Mesa makes
[20:46:26] <fenn> 7i43 isn't "standard"?
[20:46:27] <SWPadnos> (of course they won't necessarily support whatever FPGA stuff you make yourself)
[20:46:34] <SWPadnos> 7i43 is, the 7i64 isn't out yet
[20:46:43] <SWPadnos> unless PCW corrects me on that
[20:47:03] <PCW> Just happen to be hacking VHDL for 7I65 now
[20:47:06] <geo01005> what would be the cost of the 7i64?
[20:47:09] <SWPadnos> ah, 7i65
[20:48:01] <PCW> Its expensive because of 8x +-10v 16 bit DACs and isolation $279 qty1
[20:48:32] <SWPadnos> excellent! mine's only twice the price
[20:48:35] <SWPadnos> or maybe 3x :)
[20:48:42] <SWPadnos> yeah, closer to 3x
[20:49:21] <fenn> and hand-soldered (well, part of it) so it has "craft" value
[20:49:37] <archivist> cruft value ?
[20:49:58] <SWPadnos> no, the only piece I put on myself was the fence for the shield can
[20:50:10] <fenn> i thought you had to flip some capacitors around
[20:50:31] <SWPadnos> I did on the test run, but not on the later production run
[20:50:47] <SWPadnos> (they still had them screwed up, but they fixed it by hand ;) )
[20:51:52] <geo01005> So I'm not sure I understood that, so is SPI, or I2C communication going to be suppored by the hostmot2 driver? By the way somebody asked what version I'm using, I run the trunk.
[20:52:07] <PCW> But theres nothing stopping anyone from using the raw SPI interface
[20:52:09] <PCW> With HM@ you might be able to do it with just the raw-read and raw-write HAL commands
[20:52:10] <PCW> (not that it would be pretty or fun)
[20:52:12] <PCW> Either SSPI or BSPI
[20:52:14] <PCW> Just need to add support to the driver
[20:52:15] <PCW> If anyone wants to mess with it, I can make a bitfile
[20:52:39] <PCW> (HM2)
[20:53:10] <SWPadnos> PCW, I don't think there's a way to tell the driver to enable the SPI functions
[20:53:39] <PCW> raw-write ATM
[20:53:47] <SWPadnos> I also think there's no way to read values automatically in HAL
[20:54:11] <SWPadnos> it can probably be done with halcmd, so it could be possible with a custom M-code
[20:54:20] <PCW> raw-read
[20:55:13] <jepler> looking at it another way, it's a perfect opportunity to contribute -- get an SPI-capable firmwrae from PCW, and write the PC-side bits that suit your use
[20:56:16] <SWPadnos> indeed. that's a good way to look at it :)
[20:57:34] <geo01005> I wish I had better programing skills, Most all the programming I have done is with numerical analysis in engineering. I feel a little underqualified to do such a thing.
[20:58:05] <SWPadnos> PCW, are the 7i65 available for purchase yet?
[20:58:11] <PCW> Yes
[20:58:18] <SWPadnos> cool
[20:58:27] <PCW> 7I64 as well
[20:58:32] <SWPadnos> geo01005, you can send me one and I'll write the code :)
[20:58:43] <PCW> OK!
[20:58:52] <SWPadnos> oh, you can too ;)
[20:59:07] <PCW> (OH rats)
[20:59:11] <SWPadnos> heh
[20:59:12] <cradek> haha
[20:59:55] <SWPadnos> actually, a set of 7i64/7i65/7i66 would be a good thing for development
[21:00:10] <SWPadnos> there's no ID on the cards though, right?
[21:00:17] <SWPadnos> (ie, not way to detect what's there)
[21:00:21] <SWPadnos> s/not/no/
[21:02:03] <Dallur> It's only 2 hours until the time will be 1234567890
[21:02:29] <SWPadnos> hmm. actually, I think I asked this before, but what's the practical limit on SPI clock speed/cable length for those daughterboards?
[21:02:53] <PCW> I looked at the driver Didn't look terribly difficult (If I Knew C)
[21:02:54] <PCW> (the 7I65 has an EEPROM) but the 7I64 and 7I66 do not (they are 4 wire SPI devices)
[21:02:56] <PCW> They could possibly return a SPI cookie
[21:02:57] <PCW> but the shift register lengths are different so you would have to probe for each
[21:03:06] <skunkworks_> * skunkworks_ thinks PCW has a computer scanning the irc logs for any mention of mesa hardware. (then a large alarm and light go off in his shop)
[21:03:21] <jepler> skunkworks_: either that, or he has a spy on the channel that we just haven't ferreted out yet
[21:03:26] <PCW> (you found me out)
[21:04:16] <SWPadnos> sure, actually stick a SPI serial number chip (assuming those exist) at the end of the line, and clock in 16 bits or whatever at init time
[21:05:07] <SWPadnos> oh hmm. you don't really want different numbers for every board now, do you
[21:05:39] <skunkworks_> ;)
[21:05:55] <geo01005> So do the 7i64/7i65/7i66 boards communicate via SPI?
[21:06:16] <SWPadnos> yes, they communicate back to the FPGA card (or other port I suppose) with SPI
[21:06:54] <PCW> SWP output only devices can be 16 MHz or so (thats what I'm testing the 7I65 DACs with)
[21:06:56] <PCW> Read data is slower because of isolator skew (7I65) and cable delays
[21:06:57] <PCW> (I will eventually add the receive skew adjust to the SPI interface)
[21:07:13] <SWPadnos> ok, cable lengths in the 6' range or so?
[21:07:35] <PCW> Yes 6' is ok (All terminated)
[21:07:46] <SWPadnos> right
[21:07:53] <SWPadnos> and maybe longer with slower clocks
[21:07:56] <geo01005> So is it trivial to read any SPI device from a Mesa FPGA card? (After a driver is writen)
[21:08:06] <SWPadnos> is the clock rate programmable?
[21:08:15] <SWPadnos> geo01005, well, I wouldn't call it trivial
[21:08:39] <SWPadnos> the hostmot2 driver has several layers so that it can support all the different FPGA cards
[21:08:40] <PCW> geo1005: Yes except the 7I65 encoder reading is not SPI
[21:08:52] <SWPadnos> including PCI and parallel port connected ones
[21:09:11] <SWPadnos> and it also supports a number of functions (like step generator, PWM generator, encoder ...)
[21:09:43] <SWPadnos> SPI is another function that can be in the FPGA, but that's just a generic interface to "something"
[21:10:00] <PCW> yes, clock rate is programmable (in SSPI) and selectable via channel descriptor (in BSPI)
[21:10:07] <fenn> Dallur: it only happens once :)
[21:10:23] <SWPadnos> to make that "something" useful, the driver needs to know how much data to send/receive, and what the data means (is that 16 bits a single analog input or 16 digital inputs?)
[21:10:43] <SWPadnos> so there are a couple of layers that need to be dealt with for proper SPI support
[21:11:05] <SWPadnos> however, if you want a quick and dirty implementation for a specific device, that should be pretty easy
[21:11:10] <Dallur> fenn: yup, now that we have 64 bit time :P
[21:12:00] <fenn> for another half a terayear at least
[21:12:28] <Dallur> fenn: yeah, but you know terayears, they fly by
[21:13:38] <PCW> Right, per device clock speed, number of bits, echo data or no echo, CS delays, CPOL, CPHA,
[21:13:41] <PCW> and as SWP says send and recv data interpretation
[21:14:43] <SWPadnos> PCW, how does that work for multiple devices chained together?
[21:15:21] <SWPadnos> like sending 128 bits to 8 analog outputs
[21:16:43] <PCW> Just had to add some gnarly bits to BSPI to access EEPROM
[21:16:44] <PCW> unlike ADCs and DACs, EEPRMS have variable length SPI ops
[21:16:46] <PCW> SWP what I added will help this also (The "dont-clear-Frame at end of cycle" bit
[21:16:58] <PCW> )
[21:17:06] <SWPadnos> ok
[21:17:34] <SWPadnos> no buffer, but the driver can write several words and reset the "don't EOF" flag on the last one
[21:17:43] <geo01005> So lets say for starters, that I want to use a MAX6675 thermocouple chip. I don't know for sure thats what I want to use, but what would the steps be to get it work?
[21:18:02] <PCW> Theres a 16 by 36 xmit FIFO
[21:18:08] <SWPadnos> oh, excellent
[21:18:12] <SWPadnos> any read FIFO?
[21:18:25] <PCW> Yes 16 x32
[21:18:32] <SWPadnos> luvvly
[21:19:13] <PCW> bbiab lunch!
[21:19:21] <geo01005> It has an SPI interface, spits 16 bits at you when you pull the chip select down.
[21:19:33] <SWPadnos> geo01005, assuming that it needs no setup, you'd need to make the hostmot2 driver enable an SPI block, then configure that block to transfer the right number of bits
[21:20:22] <SWPadnos> then add a function to read the SPI data - whatever is needed to transfer it from the MAX6675 in the first place, then to read it from the FPGA
[21:21:07] <SWPadnos> I don't know the specifics of the SPI block (as you can see from my questions to PCW), so I can't really give you any detail
[21:22:04] <alex_joni> jepler: the STG also has some ADCs on it (on some versions)
[21:22:14] <SWPadnos> and vigilant, I think
[21:22:19] <alex_joni> I doubt anyone still finds one of those thoug
[21:22:21] <alex_joni> h
[21:23:02] <alex_joni> shrdlu-: I think in the next version of emc2 you can use short segments and laser on/off triggered using M62/M63
[21:23:23] <SWPadnos> that was my thinking when asking what version he's using now
[21:23:25] <shrdlu-> oh really?
[21:23:33] <shrdlu-> what does short segments mean?
[21:23:42] <alex_joni> but it depends how short segments you need
[21:23:58] <alex_joni> you can activate/deactivate a digital IO at the beginning of a move
[21:24:05] <shrdlu-> ahh
[21:24:10] <alex_joni> that doesn't break blending or cause slowdowns
[21:24:19] <SWPadnos> I don't think the segment length is a problem, except when trying to use Z as the laser control :)
[21:24:26] <shrdlu-> I take it that wont be released in the next 15 hours ;)
[21:24:29] <alex_joni> SWPadnos: it can be for raster
[21:24:37] <alex_joni> shrdlu-: april 15th or so
[21:24:43] <SWPadnos> no, but you can compile it yourself in the next 15 minutes or so
[21:24:54] <SWPadnos> (depending on the speed of your internet connection)
[21:25:03] <shrdlu-> really?
[21:25:05] <alex_joni> SWPadnos: if he's doing raster, going on a line and having lots of dots to laser
[21:25:10] <shrdlu-> * shrdlu- hot foots it
[21:25:23] <alex_joni> then segments size/frequency might be a limiting factor
[21:25:27] <SWPadnos> sure
[21:25:58] <fenn> alex_joni: did you do analog i/o too?
[21:26:00] <SWPadnos> are the analog outputs also working synchronized now?
[21:26:02] <SWPadnos> heh
[21:26:27] <cradek> to my surprise, there are no analog outputs
[21:26:42] <SWPadnos> oh, isn't that interesting
[21:26:49] <cradek> yours too?
[21:26:52] <SWPadnos> yes
[21:26:59] <alex_joni> fenn: nope, not yet
[21:27:05] <alex_joni> not sure how to specify it in g-code
[21:27:13] <geo01005> Is there any documentation for Hostmot2 and spi?
[21:27:14] <SWPadnos> P / Q?
[21:27:16] <alex_joni> actually I have an idea, but that would be a new feature.. wouldn't it
[21:27:24] <SWPadnos> geo01005, man hostmot2
[21:27:38] <SWPadnos> also noline at http://www.linuxcnc.org - click the documentation link
[21:27:42] <SWPadnos> online
[21:27:46] <alex_joni> SWPadnos: I'd go with Exx for the analog output number, and some other letter for value
[21:27:55] <alex_joni> probably only for M62 and M64
[21:28:19] <SWPadnos> other M codes (like custom, for instance) use P and Q
[21:28:24] <SWPadnos> wouldn't that work?
[21:28:48] <SWPadnos> M6xx P1.5 Q6 (set output 6 to 1.5)
[21:29:20] <alex_joni> I already used M61 Pxx for digital inputs, and Exx for analog inputs
[21:29:26] <SWPadnos> ah, ok
[21:29:43] <seb_kuzminsky> geo01005: hostmot2 currently doesnt do spi
[21:29:56] <SWPadnos> oh, so it's the same M code for both, and you use E vs P to decide whether to read analog or digital
[21:30:12] <alex_joni> indeed
[21:30:20] <SWPadnos> hmmm. bummer :)
[21:30:23] <alex_joni> why?
[21:30:48] <SWPadnos> well, this seems like something that should be more along the lines of L - select which "bank" you're operating on
[21:30:57] <SWPadnos> M61L1P2
[21:31:07] <SWPadnos> read digital bank1 pin 2
[21:31:12] <SWPadnos> M61L2P2
[21:31:21] <SWPadnos> read analog bank pin 2
[21:31:33] <alex_joni> that could have worked too
[21:31:43] <alex_joni> but I guess it's done differently now ;)
[21:32:08] <SWPadnos> then you have symmetry for the output codes also. use M65(?) L1P2Q1
[21:32:17] <SWPadnos> digital pin 2 set to 1
[21:32:21] <SWPadnos> yeah :)
[21:32:40] <alex_joni> yeah, only one output m-code would be more "philosophically" correct
[21:32:46] <SWPadnos> that's in 2.2 also, so we can't really change it willy nilly, can we?
[21:32:59] <alex_joni> M62 is there since 2.0.x I think
[21:33:38] <SWPadnos> actually, separate codes are OK, if it's legal to specify both digital and analog at the same time
[21:34:10] <SWPadnos> (M61 xxx M610 yyy - enable laser and set PWM value ;) )
[21:37:05] <shrdlu-> so I can get those features via "cvs -z5 -d:ext:anon@cvs.linuxcnc.org:/cvs co emc2"
[21:37:11] <shrdlu-> sorry, I is a dev noob
[21:37:33] <SWPadnos> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2
[21:37:38] <alex_joni> now this seems like a bad idea to me: http://www.youtube.com/watch?v=9BnLbv6QYcA&feature=channel
[21:37:54] <SWPadnos> specifically this section: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#On_Ubuntu_6_06_or_8_04_from_source
[21:38:17] <shrdlu-> thanks
[21:38:23] <SWPadnos> sure
[21:38:28] <SWPadnos> read carefully, then act ;)
[21:38:53] <seb_kuzminsky> alex_joni: ha ha
[21:39:28] <shrdlu-> flash sound just randomly stopped working for me today
[21:39:38] <SWPadnos> "everything is just a few hundred clicks away" :)
[21:39:43] <alex_joni> seb_kuzminsky: I tried to word it as polite as possible
[21:39:58] <seb_kuzminsky> it's the onion, you dont have to be polite ;-)
[21:40:21] <alex_joni> I meant in here :P
[21:40:57] <SWPadnos> heh. I like the bit about the Sudoku killer at the end :)
[21:41:11] <SWPadnos> "will kill either 1, 4, or 9 victims next"
[21:41:46] <SWPadnos> I liked this one: http://www.theonion.com/content/video/in_the_know_should_the_government
[21:41:57] <SWPadnos> though I'm sure I didn't need flash 10 for it before (bastards)
[21:42:54] <alex_joni> yeah, just installing it now
[21:45:03] <shrdlu-> should I specify a particular branch?
[21:45:07] <SWPadnos> no
[21:45:10] <shrdlu-> k
[21:45:15] <SWPadnos> if you specify nothing you get the latest
[21:45:18] <SWPadnos> warts and all ;)
[22:00:33] <shrdlu-> I was strongly considering buying mach3 the other night
[22:00:37] <shrdlu-> then I went back and used it
[22:00:41] <shrdlu-> and changed my mind
[22:01:11] <SWPadnos> heh
[22:01:19] <SWPadnos> what was wrong with it?
[22:02:01] <shrdlu-> it's kind of ugly. stepconf is way better than anything that comes included for setup
[22:02:14] <shrdlu-> I just couldn't be bothered to calibrate it, and use that interface
[22:02:58] <shrdlu-> plus I have a daydream about building a control panel for inkscape and integrating it with emc
[22:02:59] <alex_joni> SWPadnos: don't get me started on that ;)
[22:03:05] <shrdlu-> for laser cutting
[22:03:17] <SWPadnos> cool
[22:03:32] <SWPadnos> have you seen Xara LX?
[22:03:47] <SWPadnos> it's the closest thing I've seen to Corel Draw (other than Corel Draw, of course)
[22:04:23] <shrdlu-> wow
[22:04:26] <shrdlu-> looks good
[22:04:46] <SWPadnos> it ain't Corel, but it is pretty darned good
[22:05:05] <shrdlu-> I'd say inkscape was probably at least as good as illustrator
[22:05:26] <archivist> SWPadnos, they are related iirc corel owned it at sometime
[22:05:30] <SWPadnos> yep
[22:05:42] <SWPadnos> Corel around version 3 or so was Xara
[22:05:44] <shrdlu-> I don't like gimp much
[22:05:48] <SWPadnos> different tools
[22:06:04] <SWPadnos> bitmap editor vs. vector editor, completely different ralms
[22:06:08] <SWPadnos> realms
[22:06:39] <alex_joni> SWPadnos: do you know any tool/viewer/gismo that can open corel files?
[22:06:53] <SWPadnos> nope, only Corel AFAIK
[22:06:58] <alex_joni> except corel draw ;)
[22:07:09] <SWPadnos> the latest we have is version 9 though, I haven't been in that world for a while
[22:07:58] <shrdlu-> ah, I always thought corel was 50/50 with paint and vector
[22:08:04] <shrdlu-> I've never used it though
[22:08:24] <SWPadnos> the Corel suite has Corel Draw, which is vector, and Corel PhotoPaint, which is bitmap
[22:08:36] <shrdlu-> ah
[22:08:44] <alex_joni> I got cdr 11 I think
[22:08:45] <shrdlu-> I've always used adobe
[22:08:50] <alex_joni> the last one they sold separately
[22:08:52] <SWPadnos> you can use bitmaps as elements in Draw, and do some manipulation, and you can do some line drawing and text and stuff in PhotoPaint
[22:09:03] <alex_joni> now you can only get the suite, for a overpriced bill
[22:09:09] <SWPadnos> they have another interesting tool that's meant to be used for technical drawings
[22:09:16] <SWPadnos> like annotating 3D models and stuff
[22:09:28] <alex_joni> * alex_joni likes e-Drawings
[22:09:46] <alex_joni> you can do "some" annotations with that
[22:09:53] <SWPadnos> * SWPadnos doesn't particularly like SolidWorks (the company)
[22:10:30] <shrdlu-> "And although this image consists of more than 1200 vector objects (most of them involving graduated transparency) this image renders in under 1/10th second on a normal PC"
[22:10:31] <shrdlu-> nice
[22:10:59] <SWPadnos> there's an Ubuntu package if you enable the universe repos
[22:12:01] <alex_joni> SWPadnos: for what?
[22:12:06] <SWPadnos> XaraLX
[22:12:12] <alex_joni> ah
[22:12:40] <shrdlu-> haha, the win version is commerical
[22:20:57] <skunkworks_> ummm http://linux.softpedia.com/get/Science/Enhanced-Machine-Controller-based-Antenna-Range-37079.shtml
[22:22:42] <archivist> skunkworks, some add pusher giving http://emcar.sourceforge.net/
[22:23:53] <archivist> crafty getting themselves above sourceforge
[22:48:05] <shrdlu-> Hmm, how do I specify which pin M62/M63 uses?
[22:53:09] <jepler> shrdlu-: by using 'net' commands in your .hal file
[22:53:42] <jepler> something like 'net dout-00 motion.digital-out-00 => parport.0.pin-01-out' but you would have to use the correct pin names instead of the incorrect ones from my memory.
[23:01:38] <shrdlu-> dout, great, thankyou
[23:08:13] <shrdlu-> "Can't run a program when not homed"?
[23:08:16] <shrdlu-> thats new
[23:08:32] <SWPadnos> there's an ini option to defeat that, if you want to use it
[23:08:42] <SWPadnos> I don't remember the name
[23:08:42] <shrdlu-> hokay
[23:09:16] <shrdlu-> I need to get this CRT away from my machine
[23:09:26] <shrdlu-> it makes the screen flutter badly
[23:09:31] <shrdlu-> I feel motion sick
[23:14:30] <shrdlu-> hm, I can't see anything. Is this supposed to be added to my config ini?
[23:15:06] <SWPadnos> yes
[23:15:25] <SWPadnos> ini files often don't contain all possible options, they're too confusing if they do
[23:19:39] <jepler> [TRAJ]NO_FORCE_HOMING -- http://linuxcnc.org/docview/devel/html/config_ini_config.html#sub:[TRAJ]-section
[23:20:14] <shrdlu-> heh, thanks, I couldn't find it anywhere
[23:30:33] <jepler> >>> time.time()
[23:30:33] <jepler> 1234567826.6769791
[23:30:38] <jepler> it's just about here -- second 1234567890
[23:31:30] <jepler> yay 1234567890
[23:31:35] <seb_kuzminsky> yay
[23:32:05] <jepler> now, I am not sure about how leap seconds relat to this
[23:32:17] <seb_kuzminsky> they dont, i think
[23:33:40] <jepler> I think that if UNIX counted TAI seconds instead of UTC seconds, you would have had second 1234567890 34 seconds later
[23:34:55] <seb_kuzminsky> i thought time_t *was* TAI
[23:35:00] <jepler> I thought it was UTC
[23:36:44] <shrdlu-> How come I get that Homed error in the dev version but not the release?
[23:37:08] <shrdlu-> Disabling my soft limits isn't ideal
[23:38:01] <alex_joni> SWPadnos: disabling soft limits?
[23:38:23] <alex_joni> emc2-2.3.x will only run programs or MDI commands once it's homed
[23:38:25] <SWPadnos> I said nothing about doing that
[23:38:31] <jepler> shrdlu-: the behavior "require home before mdi / program" isn't in 2.2.x at all, it's new in 2.3 (and in 2.3 can optionally be disabled, to get the 2.2.x behavior instead)
[23:38:32] <SWPadnos> only homing, not limits
[23:38:38] <alex_joni> the old behaviour was that you can always do that
[23:38:53] <alex_joni> even if the machine is in the "who knows where I am" state
[23:39:09] <shrdlu-> but now I can just jog right past the soft limits
[23:39:30] <shrdlu-> "Warning: Using this will allow the machine to run past soft limits while in operation and is not generally desirable to allow this. "
[23:40:57] <jepler> shrdlu-: if you don't home the machine, then soft limits aren't enforced (They can't be, beacuse emc doesn't actually know where it is)
[23:41:04] <jepler> shrdlu-: once you home, soft limits are enforced
[23:41:27] <shrdlu-> I did home it, at least, I told it to home
[23:41:39] <shrdlu-> and it still brought up the error
[23:41:51] <shrdlu-> I don;t have any switches set up
[23:42:03] <alex_joni> what error?
[23:42:05] <jepler> shrdlu-: "if you do this" means "if you turn on [TRAJ]NO_FORCE_HOMING and don't home, you don't get soft limits"
[23:42:30] <shrdlu-> "Can't issue MDI command when not homed"
[23:42:45] <shrdlu-> ooh. maybe it's the y axis
[23:42:48] <shrdlu-> err
[23:42:49] <shrdlu-> z
[23:42:55] <jepler> you have to home all axes defined in your inifile
[23:43:03] <alex_joni> good night all
[23:43:10] <jepler> see you alex_joni
[23:43:16] <shrdlu-> right, ok, that makes sense
[23:43:17] <shrdlu-> thanks
[23:43:32] <shrdlu-> yeah, I don;t actually have one of those
[23:43:35] <shrdlu-> night alex
[23:43:37] <alex_joni> shrdlu-: you can define and use home-all
[23:43:53] <alex_joni> but that's probably for another time ;)
[23:45:41] <jepler> ugh, unix handling of leap seconds makes me queasy. http://en.wikipedia.org/wiki/Unix_time
[23:46:36] <seb_kuzminsky> yeah that's some icky stuff
[23:46:46] <seb_kuzminsky> it totally tains the euphoria i felt at 1234567890
[23:49:44] <jepler> hah: "In some newer operating systems, time_t has been widened to 64 bits. In the negative direction, this goes back more than twenty times the age of the universe, and so suffices. In the positive direction, whether the approximately 293 billion representable years is truly sufficient depends on the ultimate fate of the universe, but it is certainly adequate for most practical purposes."
[23:50:05] <seb_kuzminsky> i dont know, it depends on those leap seconds
[23:50:10] <BigJohnT__> BigJohnT__ is now known as BigJohnT
[23:50:21] <seb_kuzminsky> Big John!
[23:50:36] <cradek> the funniest part is that "0" is still 1970
[23:51:10] <alex_joni> seb_kuzminsky: maybe you'll like 1500450271 then
[23:51:30] <seb_kuzminsky> is that when you're finally going to bed? ;-)
[23:51:43] <alex_joni> might be so
[23:52:12] <alex_joni> it's prime, so I might just
[23:52:14] <jepler> that's a prime time; is it the next prime time?
[23:52:21] <seb_kuzminsky> see y'all later
[23:52:25] <jepler> (I actually was wondering that just a bit ago...)
[23:52:25] <alex_joni> not sure if it's the next prime time
[23:53:15] <jepler> well, somebody's developed a website which they promise will "instantly" tell you whether a number is prime or not! http://www.math.com/students/calculators/source/prime-number.htm
[23:54:08] <alex_joni> hah
[23:54:14] <jepler> there's also "List of prime numbers - Wikipedia, the free encyclopedia"
[23:54:14] <alex_joni> 1234567891 was a prime
[23:54:19] <jepler> alex_joni: oh well heck
[23:54:21] <jepler> I missed it
[23:54:28] <alex_joni> 1234567907 next
[23:54:39] <alex_joni> missed that one too
[23:55:39] <alex_joni> 123457067 is still up
[23:55:59] <alex_joni> I mean 1234567067
[23:57:18] <jepler> (hm, here's a good one: a proposal to abolish the leap second and instead force our descendants to have a leap hour .. why not save it up and have a whole leap year somewhere down the road?)