#emc | Logs for 2010-06-19

[00:49:04] <andypugh> Woohoo! At last! I have my Arduino/IRAMS Resolver-feedback AC Servo thing working properly!
[01:02:54] <andypugh> And at 300V DC bus voltage it flies! I can't imagine how fast it will move my axes, probably far faster than is even slightly sensible.
[01:03:40] <andypugh> 3000rpm * 5mm pitch = 15m/min....
[01:04:00] <andypugh> Travel is 250mm....
[01:04:17] <andypugh> Better get those limit switches fitted.
[01:50:56] <morfic> when i opened my case to install new mobo, cpu, ram, it was like opening a time capsule, pc been in storage for a while, nostalgia
[01:51:51] <morfic> hope the "error loading operating system" does not mean i forgot about a failed drive somewhere :>
[02:29:38] <madsci44> is it me or is axis incorrectly plotting something; when i program a G92 to set an offset, and reload the drawing translates accordingly, - so far so good..
[02:31:00] <madsci44> but when i program G92.2 to deactivate the offset (but retain it in the storage variables) the drawing stays according to the last offset - while the plot of the actual moves is translated back relative to the machine origin
[02:32:47] <cradek> madsci44: hit reload
[02:32:53] <madsci44> right
[02:33:13] <madsci44> still shows it according to the offset
[02:33:35] <cradek> (if you use touch off instead of g92 you won't have this problem)
[02:34:28] <cradek> hm, I misread the firs time what you typed
[02:34:35] <cradek> first
[02:34:57] <cradek> you are saying that even if you do reload after g92.2 it plots according to the saved vars
[02:35:08] <madsci44> yes
[02:35:38] <cradek> I think that is a fairly fundamental limitation of how axis works (it copies the var file)
[02:36:10] <madsci44> but if i run the program - it runs in the proper location (ignoring saved variables) but still shows the drawing plot in the gl window at the offset position - the movement plots are in the correct location
[02:36:11] <madsci44> ah i see
[02:36:41] <cradek> g92.2 is kind of an oddball - nothing else writes vars but then ignores them
[02:39:33] <madsci44> i thought it would be handy for manually indexing to a corner - and it functions correctly so its not a big deal - was just confusing till I realised what was going on
[02:40:15] <cradek> thanks for the report but I'm not sure what to do about it
[02:40:27] <cradek> it might not hurt if you'd file a bug report on the sourceforge bug tracker so it's not forgotten
[02:41:00] <madsci44> ok sure - do i need to know much to use that?
[02:41:13] <cradek> no it's just a web thingy.
[02:41:26] <madsci44> (never filed one before) ok cool should be easy nuff
[02:42:31] <madsci44> im setting up a machine for a friend and he wanted that ability - to hit a button and call it zero
[02:43:06] <madsci44> so i added the buttons to axis "zeroxy, unzeroxy, re-zeroxy"
[02:43:12] <madsci44> in the manual tab
[02:43:27] <madsci44> (he is not one who likes to type g-codes)
[02:43:34] <cradek> but that's what touch-off does (and it works better)
[02:43:54] <madsci44> its on a plasma gantry
[02:44:15] <madsci44> unless i am not understanding touchoff
[02:45:16] <madsci44> is that based on probing? (i have to read up on that anyway)
[02:45:29] <alex|lappy> Hmm, anythoughts on hal configuration so I can get EMC to estop when a pin goes high
[02:45:33] <alex|lappy> err
[02:45:34] <alex|lappy> goes low
[02:53:53] <madsci44> maybe just AND it with the input loop?
[02:54:11] <madsci44> (invert it first of course)
[03:04:14] <madsci44> alex|lappy: you already have an external estop input dont you?
[03:04:32] <madsci44> like are you trying to add another?
[03:07:51] <alex|lappy> well
[03:07:54] <alex|lappy> i have gecko320s
[03:08:06] <alex|lappy> they have a fault output that goes low on fault
[03:08:14] <alex|lappy> and I want that to trigger a system wide estop
[03:08:40] <cradek> you must do that in wire, not software
[03:09:09] <alex|lappy> cradek, what?
[03:10:36] <alex|lappy> as in, use a 5v relay or something?
[03:12:46] <cradek> relays are ok, usually estop is a (NC) chain
[03:13:01] <cradek> I have to run - goodnight
[03:13:07] <madsci44> gnite
[03:20:32] <madsci44> is there something i have to have set up to enable the touchoff button
[03:20:35] <madsci44> mines greyed out
[03:35:39] <madsci44> that wouldnt work for this guy anyway he just wants to hit a button and have that be zero
[03:38:07] <alex|lappy> madsci44, what about a g92 command?
[03:38:46] <alex|lappy> that shifts machine wcs zero to the current position
[03:38:52] <alex|lappy> it is... not very elegant
[03:38:54] <alex|lappy> but it works
[03:39:17] <alex|lappy> er, at least it does on the old anlam 1400 I use it on all the time :-p
[03:39:19] <madsci44> which the touch off?
[03:39:34] <madsci44> yeah thats what im using is g92
[03:39:47] <madsci44> that works perfect except axis doesnt display the plot correctly
[03:39:51] <madsci44> but it will do the job
[03:40:45] <madsci44> it does when you invoke it, and when you temporarily disable it, like g92.2 it doesnt translate the visual back to origin
[03:40:59] <madsci44> but still he can just "hit a button" this way so that will do
[03:42:23] <alex|lappy> I mean, can he not just punch in a couple of very easy to obtain numbers and use a g54 like the rest of the world?
[03:44:56] <alex|lappy> hmm, so it looks like I can tie all the gecko drives fault outputs together, then it at least faults all the drives at the same time
[03:45:21] <alex|lappy> that would just leave the spindle running, which while not ideal, I could hit the physical estop button after that
[03:45:47] <morfic> you might actually want that
[03:46:20] <morfic> sometimes harder to deal with a burried endmill than with one spinning free
[03:46:39] <alex|lappy> and then I can just have the estop flip a relay that cuts AC power to the servos, so that when I hit the estop button it will def. cut power to the drivers
[03:46:54] <alex|lappy> or at least that will work until I can build a real current dump circuit
[03:47:10] <alex|lappy> and put a relay in that that will control the estop on my BOB
[03:47:52] <madsci44> he doesnt want to enter numbers :) hey its his machine what works for him...
[03:48:42] <alex|lappy> madsci44, heh, tell him to go buy a haas then
[03:48:55] <alex|lappy> they have a really nice touch off button on the front
[03:48:58] <alex|lappy> :D
[03:49:07] <madsci44> tell taht to any hobbyist and you will get the same result :)
[03:59:06] <madsci44> i think the whole point of having software based controllers - besides economy - is flexibility - adapt it to what you want, have it work the way you want. Especially EMC2 - being open source - choice of guis, etc...
[04:04:24] <renesis> mmmmm haaaaas
[04:15:33] <madsci44> that was odd
[04:20:57] <alex|lappy> actually I am honestly really really underwhelmed by the haas machines my school just got
[04:21:33] <madsci44> how so
[04:21:58] <alex|lappy> we got a TL-1 and a VF
[04:22:08] <alex|lappy> madsci44 the software on the control is terrible
[04:22:15] <alex|lappy> a couple of times a day it stops reading off usb keys
[04:22:22] <alex|lappy> and you have to reboot
[04:22:32] <alex|lappy> sometimes the lathe freaks out with g50s
[04:22:39] <alex|lappy> not letting go of offsets after you delete them
[04:22:49] <alex|lappy> and sometimes there are programs that will not run no matter how hard you try
[04:22:57] <alex|lappy> then restart the machine and bam, works on the first time
[04:23:15] <alex|lappy> It would not be so bad except for the fact that they make me feel like I am going crazy because they are so inconsistant
[04:23:29] <madsci44> wow
[04:23:35] <alex|lappy> oh oh
[04:23:39] <alex|lappy> they did not come with network card
[04:23:50] <alex|lappy> and they want $2500 per machine
[04:23:54] <alex|lappy> w.t.f.
[04:24:57] <alex|lappy> also, we got a 4th axis donated
[04:25:10] <alex|lappy> they want a couple grand to enable the outputs for it
[04:32:36] <madsci44> i hate that sort of stuff
[04:33:05] <madsci44> proprietary crap - paying $2k for something that should be $200
[04:33:31] <alex|lappy> madsci44, yeah, and if we were making money off it that would be one thing, but it is a Community College
[04:33:47] <madsci44> but probably thats their strategy - when they buy the machie they are in for a penny - if they need anything else they are in for a pound (or a tonne)
[04:33:50] <alex|lappy> there is no money for that stuff, and why in the world would they not just enable ALL of it?
[04:33:55] <alex|lappy> it makes no sense at all
[04:33:58] <Jymmm> Then call and ask them to donate it, and give them a tax receipt
[04:34:06] <alex|lappy> Jymmm, oh trust me
[04:34:08] <alex|lappy> we have tried
[04:34:10] <alex|lappy> again and again
[04:34:41] <alex|lappy> they won't do it
[04:35:09] <alex|lappy> I just don't get why you would not give free features (ones that just need to be turned on) to a .edu
[04:35:27] <alex|lappy> everyone they train is going to be used to these nice new haas machines, and will want to buy those features later
[04:35:36] <alex|lappy> except not, because why would you ever want to do rigid tapping?
[04:36:26] <alex|lappy> * alex|lappy is bitter
[04:37:28] <madsci44> im amazed a name like that would be so glitchy
[04:38:07] <Jymmm> Well, force them to.
[04:40:24] <alex|lappy> madsci44, I wish I was just going crazy, but it is not just me :/
[04:41:12] <elmo40> any bloggers? http://code.google.com/p/googlecl/
[04:41:25] <madsci44> seems like anything is a risk these days - like companies dont actually try/use their own stuff before they put it in prouction
[04:42:04] <madsci44> ths kybrd is a god exmple - heeh
[04:42:06] <elmo40> like winVista? :P
[04:42:25] <madsci44> yeah like that
[04:46:51] <elmo40> alex|lappy: the college is making money off of it
[04:46:58] <elmo40> buy training students!
[04:47:14] <elmo40> do who cares what the price is. they make it back in tuition
[04:47:26] <elmo40> plus, it is a tax right off.
[04:47:38] <elmo40> don't for once think a college has no money.
[04:48:46] <alex|lappy> elmo40, huh? my community college is not making any money for anyone
[04:49:53] <alex|lappy> without grant money and donations we would not have any money for new equipment at all
[05:02:25] <madsci44> ok i cant find a link to the sourceforge bug tracker
[05:02:44] <madsci44> no wait
[05:03:48] <madsci44> no thats not it
[05:20:50] <madsci44> somehow i was hoping my first contribution to emc would be something more positive than a bug report.. heh
[06:14:05] <alex|lappy> so, my estop plan is thus
[06:14:07] <alex|lappy> http://www.geckodrive.com/upload/ElectromechanicalEStop.pdf
[06:14:09] <alex|lappy> I am going to make that
[06:17:36] <alex|lappy> and put a NO relay that is activated @120 volt in series with the estop, and then have it switch a +5 into my BoB
[06:18:13] <alex|lappy> so that when the AC is cut EMC gets Estopped.
[06:34:45] <madsci44> what about when an estop is issued from the emc side?
[06:46:56] <alex|lappy> madsci44, oh, another relay in series with the estop switch attached to one of the relays on my BoB
[06:47:25] <alex|lappy> so when the computer estops the circuit charging the relay coil still gets broken
[06:59:22] <madsci44> kewl
[07:00:05] <alex|lappy> hmm, now I am worried about the amp rating for the Estop switch
[07:00:37] <madsci44> only has to be enough to drive relay no?
[07:02:35] <alex|lappy> yes
[07:02:56] <alex|lappy> the relay I am looking at has a 5.2kohm coil
[07:03:16] <alex|lappy> which means there is only .02amps going through it
[07:04:02] <alex|lappy> * alex|lappy orders parts on digikey
[07:09:30] <madsci44> what kinda machine is this setup going to be for?
[07:15:27] <alex|lappy> madsci44, a knee mill
[07:16:18] <alex|lappy> madsci44, my build thread I started and have not updated in a week :-p
[07:16:18] <alex|lappy> http://cnczone.com/forums/showthread.php?p=788127#post788127
[07:17:11] <madsci44> my knee mill is next
[07:19:52] <alex|lappy> madsci44, I have been really working on it for about a month
[07:19:57] <alex|lappy> well, more then that
[07:20:05] <alex|lappy> pretty heavy reading for the last 6 months
[07:20:20] <alex|lappy> but focused on actually figuring out what I wanted to buy/make for the last month or two
[07:20:33] <alex|lappy> and then have only really started purchasing equipment in the last two weeks
[07:21:31] <alex|lappy> it has been fun, I have all of my servos working with EMC, once i get this Digikey order I will be able to have a real Estop circuit and current dump for my motors. Tomorrow i will get the analog out I got working to controll my VFD
[07:21:41] <alex|lappy> just need to find a set of ball screws :-p
[07:23:08] <Valen> alex|lappy: linearmotionbearings2008 on ebay
[07:25:10] <alex|lappy> well, I might be getting some from a teacher at my school, he happens to have a set lying around that might be perfect
[07:26:06] <madsci44> ballscrews are always murder
[07:26:29] <madsci44> mine has some really nice ground acme's in it - ill probaby just run those for a while to get started
[07:29:55] <alex|lappy> Valen, those are all rolled :/
[07:30:11] <Valen> really cheap though ;->
[07:30:21] <Valen> spend what you save on linear scales ;->
[07:31:41] <alex|lappy> Valen, heh, I have already laid down too much dough to skimp on things in general
[07:31:55] <alex|lappy> I keep thinking things like "damn, I really don't want to spend $200 on a sealed enclosure"
[07:32:05] <Valen> i would think the linear scales method would be better in the end
[07:32:11] <alex|lappy> but I also don't want a thousand bucks in electronics to die
[07:32:26] <Valen> because to get a ballscrew as accurate as the scale would be *really* expensive
[07:33:22] <madsci44> my thinking is the same - there are some decent rolled available now, just let the scale do the accuracy part
[07:33:51] <alex|lappy> can you get linear scales without the DRO attached?
[07:33:53] <Valen> you do still want encoders on the motors though
[07:33:56] <Valen> yeah
[07:34:01] <alex|lappy> just with like a qudarature encoder output?
[07:34:03] <alex|lappy> where at?
[07:34:05] <Valen> we got .001mm scales for ~$1000 all up
[07:34:13] <Valen> local machine tool shop
[07:34:22] <Valen> sorry machenery shop
[07:34:25] <Valen> hare and forbes
[07:34:36] <Valen> what you have to look at though is backlash
[07:35:03] <Valen> if theres any in the system it will have some issues with low speeds
[07:35:50] <madsci44> Nook (and i suppose others) have sort of a precision grade of rolled screw now
[07:36:06] <madsci44> can get anti backlash nuts with those
[07:36:22] <Valen> things like stiction and the like can also look like backlash
[07:36:27] <Valen> thats our problem at the moment
[07:36:49] <Valen> wiki.linuxcnc.org/cgi-bin/emcinfo.pl?JakeAndRussells is our mill, its just using the scales to close the servo loop
[07:36:53] <madsci44> end bearings, flex in the plates,
[07:38:03] <madsci44> stretch in drive belts
[07:38:17] <alex|lappy> Valen, ah, see, scales for my mill would be a little more I think
[07:38:44] <Valen> alex|lappy: why is that?
[07:40:33] <alex|lappy> Valen, it is larger
[07:40:40] <Valen> photo?
[07:40:46] <alex|lappy> 36" of travel in X and 19" in Y
[07:41:14] <Valen> thats not actually that much bigger
[07:41:30] <alex|lappy> Valen, http://www.flickr.com/photos/33383121@N04/4263848415/sizes/l/
[07:41:33] <Valen> ours is 750mm x 250mm
[07:41:48] <Valen> so same ballpark
[07:42:41] <madsci44> very decent travels
[07:43:39] <alex|lappy> madsci44, she was well loved whe I got her, but I have a feeling that after I finish this summer I will have a machine to be proud of
[07:43:49] <alex|lappy> especially after i get the spindle bearings redone
[07:44:42] <madsci44> no doubt
[07:45:32] <alex|lappy> if only i knew somone old enough to teach me how to scrape the ways
[07:45:37] <Valen> we are looking to put a spindle on ours
[07:45:48] <alex|lappy> I actually think I am going to send mine in
[07:45:58] <alex|lappy> there is a company nearly local to here that does it super cheap
[07:45:59] <madsci44> there are some very cool things available for spindles now
[07:46:33] <madsci44> yeah here theres a place like that too - they will do the whole 9 yards, spindle, quill, balance everything, etc
[07:46:36] <alex|lappy> $475 for a complete rebuild, and they will re grind your taper while it is seated in it's bearings
[07:46:44] <Valen> thats cheap
[07:46:57] <alex|lappy> http://www.cm-spindle.com/services.html
[07:47:08] <Valen> problem with ours is 1600RPM max speed
[07:47:54] <fragalot> Brapp brapp
[07:48:02] <alex|lappy> I think mine will do 2500 at the fastest
[07:48:59] <madsci44> mines supposed to do 4000 but i dont trust it past 2000
[07:49:22] <Valen> so we are looking at those chinese $500 spindle + inverter combos
[07:49:42] <Valen> thing is I don't think they will have any ooomph at lower speeds
[07:49:57] <alex|lappy> Valen, that is almost half the cost of my entire mill!
[07:50:02] <alex|lappy> D:
[07:50:18] <alex|lappy> get something used, old and american, your life will be happy again
[07:50:29] <alex|lappy> of course, there will be swearing, and the being covered in grease
[07:50:32] <alex|lappy> and the stripped screws
[07:50:35] <alex|lappy> and the sharp edges
[07:50:40] <alex|lappy> the scratches and dings...
[07:50:48] <Valen> alex|lappy: and then ship it to australia and you could have baught a new CnC for the same price
[07:50:54] <alex|lappy> maybe you did make the right choice :D
[07:53:30] <madsci44> my dad has an old adcock & shipley
[07:53:58] <Valen> so yeah, any suggestions on a spindle for metalwork rather than wood?
[07:54:09] <madsci44> its smaller table and travels old as hell, but its SO solid compared to mine
[07:55:18] <alex|lappy> madsci44, what have you got?
[07:55:37] <madsci44> mines a taiwan import
[07:55:49] <madsci44> 10x50 VS
[07:55:49] <fragalot> mine's dead >.>
[07:56:26] <madsci44> dead?
[07:56:34] <alex|lappy> !8 just wait until class starts on tuesday, or call and pester my teacher for ballscrews tomorrow?
[07:56:38] <fragalot> the controller decided to short out on me
[07:56:42] <alex|lappy> oh, no magic 8ball in this channel.
[08:03:16] <madsci44> heres a recent photo of mine alex|lappy: http://img816.imageshack.us/img816/518/1002234.jpg
[08:03:38] <Valen> far too clean
[08:03:55] <alex|lappy> madsci44, shiny
[08:04:08] <alex|lappy> mine is going to get a fresh coat for suuuure
[08:04:12] <madsci44> hah
[08:04:12] <alex|lappy> I am not quite sure what color yet
[08:04:44] <alex|lappy> for some reason this caught my eye
[08:04:45] <alex|lappy> http://i531.photobucket.com/albums/dd358/gonekayatgmailcom/monarch07.jpg
[08:04:48] <madsci44> it looks nice once a week
[08:05:01] <alex|lappy> I would kill for a 10EE
[08:05:05] <alex|lappy> they are so pretty ;_;
[08:06:25] <fragalot> alex|lappy: for some reason' .. probably because it's bright orange :P
[08:07:01] <alex|lappy> fragalot, it is cool
[08:07:07] <alex|lappy> like bow ties, and math
[08:08:16] <storagecy> this isnt a channel for EMC storage? ie, clariion and such?
[08:08:36] <alex|lappy> noes
[08:08:42] <alex|lappy> stay and build CNC machines with us though
[08:08:47] <alex|lappy> its more fun, I swear.
[08:08:51] <fragalot> :D
[08:08:56] <fragalot> what's EMC storage?
[08:09:10] <storagecy> fragalot: http://www.emc.com/products/category/storage.htm
[08:09:23] <alex|lappy> fragalot, I dunnknow, but how could it possible be as amusing as mashing up metal using electricity and other bits of metal
[08:09:37] <storagecy> lol
[08:10:03] <storagecy> thanks anyways guy
[08:10:06] <fragalot> alex|lappy: assuming your controller doesn't magically short out like mine
[08:10:13] <fragalot> * fragalot glares at it from a distance
[08:10:19] <madsci44> carriage on that looks massive
[08:10:20] <alex|lappy> fragalot, chips?
[08:10:33] <alex|lappy> madsci44, on the 10EE?
[08:10:43] <alex|lappy> that art deco look is so amazing
[08:10:47] <alex|lappy> * alex|lappy starts looking at ebay
[08:10:49] <madsci44> yeah
[08:10:58] <madsci44> its half the entire bed!
[08:11:26] <alex|lappy> hmmm, they are not that expensive actually
[08:11:44] <fragalot> alex|lappy: It's at my friends' place to ship it back for warranty, nfc which chips were on there, bt it had 3 4DIP switches, left 2 controlling the microstepping (up to 1/8), and the right 2 selecting the mode 'mix' 'fast' 'slow' etc
[08:11:51] <alex|lappy> http://cgi.ebay.com/MONARCH-10EE-TOOL-ROOM-LATHE-/290322903213?cmd=ViewItem&pt=BI_Lathes&hash=item439897f0ad#ht_720wt_907
[08:12:00] <fragalot> it said V2.0 090907 printed on the board, and that's about all I can give as a description
[08:12:00] <madsci44> what happened to your controller?
[08:12:07] <fragalot> madsci44: it randomly shorted out.
[08:12:12] <alex|lappy> $2k USD, looks like the threading gearbox is dead but no one here gives a fuck about that
[08:12:16] <madsci44> man not good
[08:12:18] <fragalot> everytime I switch on the PSU it just goes *Poof*
[08:12:48] <madsci44> same machine you were running last weekend?
[08:12:52] <alex|lappy> fragalot, oh, sorry, I was actually asking if it got hit with some chips :o
[08:12:57] <fragalot> http://www.cnc-sable.nl/images/driverview.jpg
[08:13:16] <fragalot> alex|lappy: ah, no. and even if it did it's only wood dust
[08:13:31] <fragalot> (literally dust, no chips.. only done some engraving sofar)
[08:13:40] <alex|lappy> :/
[08:13:51] <alex|lappy> huh, where are all of you people located?
[08:13:52] <Jymmm> fragalot: Did you ALWAYS make sure you had a GOOD connection to the motors? even during testing
[08:14:00] <madsci44> << ontario
[08:14:00] <fragalot> Jymmm: yes
[08:14:10] <alex|lappy> <- Eugene Oregon USA
[08:14:15] <Valen> sydney australia
[08:14:18] <Jymmm> fragalot: No chance at all that a wire could have come loose?
[08:14:51] <fragalot> Jymmm: measured wires before connecting, and after i noticed something was wrong.
[08:14:56] <fragalot> checked out in both cases
[08:15:30] <Jymmm> fragalot: Not measured, I mean could a wire have come loose at ANY TIME, even briefly?
[08:16:08] <fragalot> only wire that could have come loose would be one of the limit switches that was giving me a hard time
[08:16:18] <Jymmm> ok
[08:16:39] <alex|lappy> fragalot, are you in netherlands?
[08:16:44] <fragalot> alex|lappy: belgium
[08:18:40] <Valen> http://cgi.ebay.com.au/Fanuc-spindle-motor-A06B-1066-B923-rebuilt-/120581354757?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item1c13357505#ht_1086wt_1139 would be a nice addition to our motor lol
[08:18:48] <Valen> addition to our machine rather
[08:20:20] <alex|lappy> Valen, you should do a big DC servo
[08:20:26] <alex|lappy> get yourself some rigid tapping in the process
[08:20:44] <Valen> would need to be pretty dern big lol
[08:20:52] <Valen> we have some 400W ones i'm tempted to use
[08:21:08] <madsci44> i bought a beautiful yaskawa AC servo motor to use as spindle drive but came with the shaft locked
[08:21:20] <alex|lappy> madsci44, bad bearings?
[08:21:25] <madsci44> second time burned for a servo motor from ebay
[08:21:34] <madsci44> nah i thik magnet or misaligned shaft
[08:21:41] <fragalot> :/
[08:21:47] <alex|lappy> have you taken it apart?
[08:21:50] <madsci44> no
[08:21:58] <alex|lappy> sounds like fun
[08:21:58] <Valen> madsci44: you could more than likley fix a magnet
[08:22:36] <madsci44> shaft turns about maybe 5 deg, feels like its hitting metal in both directions
[08:22:42] <madsci44> i never tried disassembling one
[08:22:49] <Valen> thats probably fixable then
[08:23:04] <Valen> if the shaft was bent it'd just get harder not go clunk
[08:23:13] <madsci44> i presume they are hard to put together/take apart but theres probably a trick
[08:23:20] <madsci44> ?
[08:23:47] <alex|lappy> nah, probably take front and back cover off and just slide it out
[08:23:53] <alex|lappy> slide the rotor out that is
[08:24:03] <madsci44> wounnt the magnets make it hard to do nicely?
[08:24:08] <madsci44> wouldnt
[08:24:08] <Valen> if it was something bent it wont generally go clunk
[08:24:25] <Valen> it'll be hard to pull out
[08:24:31] <Valen> and scary to put in
[08:24:46] <Valen> but just be a bit carefull and you should be ok
[08:24:47] <madsci44> yeah - i contemplated jigging it on the lathe somehow
[08:24:54] <Valen> how big is it?
[08:25:09] <madsci44> mount face is about 10 inches
[08:25:12] <madsci44> square
[08:25:15] <madsci44> maybe 8
[08:25:16] <madsci44> in there
[08:25:25] <Valen> what wattage?
[08:25:35] <madsci44> 4kw
[08:25:39] <alex|lappy> I mean, the magnet in the rotor is not *that* big
[08:25:53] <madsci44> if i remember correctly its been sitting for a couple years since i got it
[08:25:53] <Valen> at 4Kw in a servo it's probably getting pretty large
[08:25:54] <alex|lappy> the real stron ones are the coils, which are obviously not energized
[08:26:02] <alex|lappy> strong
[08:26:29] <madsci44> its an AC servo
[08:26:43] <Valen> if its AC then there will be no magnets in it
[08:26:52] <Valen> perhaps a bit of residual field
[08:27:23] <madsci44> hmm
[08:27:52] <madsci44> i thought ac servos had permanent magnets, driven like a brushless DC servo with a sine wave instead of trapezoidal
[08:28:03] <madsci44> and via encoder instead of hall sensors
[08:28:41] <Valen> AC is generally an induction motor
[08:29:02] <Valen> there are brushless DC motors and they have magnets
[08:29:32] <fragalot> the 'brushless' bit sortof indicates the precense of magnets :P
[08:30:06] <madsci44> heh ..or field coils :O
[08:30:35] <fragalot> detail
[08:30:44] <Valen> field coils would be uncommon in a brushless motor
[08:31:01] <Valen> (mainly because you would need brushes to get the current to them)
[08:32:02] <madsci44> shaft on this thing is around 1.25 inch dia
[08:32:03] <fragalot> could do it via induction? :P
[08:32:47] <madsci44> well can an induction motor hold full torque and its shaft stationary?
[08:32:47] <Valen> fragalot: yeah, then it'd be called an AC motor ;-P
[08:33:00] <fragalot> Valen: ^_^
[08:33:10] <Valen> generally not "full torque" but they are definatly above zero with a good driver
[08:34:52] <alex|lappy> I have a VFD that does .5hz (about 15rpm) with large amounnts of torque
[08:34:58] <alex|lappy> mmmmm
[08:35:47] <Valen> http://cgi.ebay.com.au/Dover-Air-Bearing-Spindle-Motor-Drive-ATC-BDC-0610-/180521741998?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item2a07ef2eae#ht_5719wt_913
[08:36:37] <madsci44> ac servo i think is a different beast
[08:37:18] <alex|lappy> Valen, no .2" DOC for you!
[08:37:36] <Valen> we are lucky to get 1mm at the moment lol
[08:37:55] <alex|lappy> I... think I would kill myself
[08:38:15] <Valen> also 1600RPM max speed
[08:38:37] <alex|lappy> as much as the controllers of the haas at school are shit
[08:38:40] <alex|lappy> the hardware is very nice
[08:39:10] <alex|lappy> 5k RPM with a half inch carbide doing HSM toolpaths is a blast
[08:39:28] <Valen> http://cgi.ebay.com.au/1996-JOHNFORD-CNC-MILL-MITSUBISHI-SPINDLE-MOTOR-200V-/150455608814?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item2307da65ee#ht_9549wt_1171 more your speed alex|lappy?
[08:40:49] <alex|lappy> Valen, wait
[08:40:55] <alex|lappy> oh
[08:40:57] <alex|lappy> spindle, not motor
[08:41:40] <alex|lappy> Valen is that... 200V DC?
[08:41:51] <alex|lappy> @96 amps?
[08:42:07] <madsci44> http://www.yaskawa.com/site/Products.nsf/products/Rotary%20Servomotors~SGMSHSigmaII.html
[08:42:30] <madsci44> similar to photo on that page - i think same series
[08:42:47] <Valen> its a 7.5Kw motor
[08:42:57] <Valen> so 9.6 i'm guessing
[08:43:09] <alex|lappy> thats 10hp 0.o
[08:43:43] <alex|lappy> I thought my little .6Kw servos were pretty baller
[08:55:31] <fragalot> * fragalot casts a stare at his .5W servos
[08:55:33] <fragalot> :(
[08:56:54] <Valen> we are using 400w scooter motors ;-P
[08:57:34] <madsci44> hey they did the job frag! nothing wrong with that :)
[08:57:51] <madsci44> and easier to carry
[08:58:29] <fragalot> madsci44: :) used 'm in my robotic walker
[08:58:44] <fragalot> later found out if I let it carry it's battery pack it ground to a halt.
[08:59:02] <alex|lappy> ooooh
[08:59:16] <alex|lappy> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=310205926480&rvr_id=&crlp=1_263602_263622&UA=%3F*S%3F&GUID=113376291250a0437190df50ffcc6352&itemid=310205926480&ff4=263602_263622#ht_1682wt_1133
[08:59:24] <alex|lappy> that would look perfect on my X axis
[08:59:52] <alex|lappy> yup, that would look nice indeed
[09:02:35] <Valen> urgh all the non crap spindles are 10Kw + and $1000
[09:03:59] <madsci44> fragalot: that walker sounds like a fun project
[09:05:26] <fragalot> madsci44: it was :)
[09:05:41] <madsci44> alex:lappy: maybe in the y? that one only has 22" travel no?
[09:05:51] <WalterN> I need to get around to building a CNC mill... :/
[09:06:31] <WalterN> for engraving
[09:06:43] <WalterN> was thinking about basing it on this... http://www.artcotools.com/compete-nsk-e3000-series-spindle-system-60000-rpm-pr-16758.html
[09:06:53] <WalterN> unless somebody has better suggestions
[09:06:55] <alex|lappy> err
[09:06:57] <alex|lappy> madsci44, yeah
[09:07:20] <Valen> spendy
[09:07:24] <alex|lappy> the problem is most of what I am finding on the ebay is either too small, or needs to get turned down
[09:07:27] <Valen> WalterN: seen the chinese spindles?
[09:07:33] <alex|lappy> and the idea of cutting ballscrew thread off
[09:07:37] <alex|lappy> just makes me wanna puke a little
[09:07:41] <Valen> ~$500 or so for 2.2kw spindle + vfd
[09:07:57] <WalterN> Valen, this is to make these... http://www.rockymountainwatch.com/
[09:08:14] <WalterN> Valen, for the engraving part anyway
[09:08:20] <Valen> pretty teensy
[09:09:03] <WalterN> it would be nice if it was portable, so he could take the machine with him to shows, and just engrave what they need there at the show
[09:09:39] <Valen> might be worth looking at those chinese ones, if you drop the $5k spindle you would be crying so much lol
[09:09:59] <alex|lappy> god, I would hate to have someone looking over my shoulder while I worked with endmills and engravers near their nice watches
[09:10:11] <WalterN> Valen, where I used to work, I set one of those systems up on a lathe for a C-axes
[09:10:19] <Valen> what do the watches sell for out of curiosity?
[09:10:23] <WalterN> Valen, very nice, and really quiet
[09:10:25] <alex|lappy> WalterN, I bet that worked mighty fine
[09:10:42] <WalterN> Valen, the cheapest ones are $600ish each
[09:10:50] <Valen> the cheap ones only go up to 25K or so
[09:12:08] <WalterN> hmm
[09:12:13] <WalterN> now that I'm thinking about it
[09:12:21] <WalterN> I dont think portable is really feasable
[09:12:24] <WalterN> heh
[09:13:16] <Valen> why not?
[09:13:24] <WalterN> enh
[09:13:45] <fragalot> I've got a Sable 2015 .. it's portable
[09:14:06] <Valen> if its just the engraving you could make a machine to do it
[09:14:09] <fragalot> the controller is useless as it died after a week though >.>
[09:14:13] <madsci44> I've turned quite a few ballscrews its not that big a deal
[09:15:30] <madsci44> usually like a 4150 and about a C50-55 hardness
[09:16:34] <alex|lappy> no no
[09:16:38] <alex|lappy> its not the process that I don't like
[09:16:45] <alex|lappy> it is the idea of cutting off precision ground threads
[09:16:49] <alex|lappy> they are just so nice and pretty
[09:16:53] <madsci44> hehe
[09:16:56] <alex|lappy> and you paid so god damn much
[09:17:11] <alex|lappy> nah, to turn them I will just throw them in the mori seki at school
[09:17:14] <alex|lappy> it will make quick work
[09:17:18] <madsci44> yeah - but you would have paid even more with the journal you wanted as well
[09:20:21] <fragalot> lol
[09:23:31] <madsci44> this is my gantry setup http://img710.imageshack.us/img710/7481/lt44.jpg
[09:25:53] <fragalot> did you make that yourself/
[09:26:17] <madsci44> yup
[09:26:21] <fragalot> how accurate is it?
[09:26:31] <alex|lappy> madsci44, is that a RoGR?
[09:27:04] <madsci44> pretty accurate depends on what its doing
[09:27:53] <madsci44> positional res is 0.001 repeatability is maybe around 0.003-0.005
[09:28:07] <madsci44> if its moving fast or pushing a load of course deflection can add to that
[09:28:22] <madsci44> for plasma it doesnt matter much since thats a sloppy process anyway
[09:28:38] <madsci44> but i made it so i could swap work heads and use it for routing as well
[09:28:46] <madsci44> whats a rogr?
[09:29:09] <alex|lappy> madsci44, a put it together yourself type bot that has a simliar construction
[09:29:45] <madsci44> ah not familliar with it - i designed this from scratch , based on another design i did from scratch that was similar about 10 years ago
[09:30:44] <madsci44> the one pictured was my first EMC based version :)
[09:32:14] <WalterN> madsci44, what did you use to power the servos with?
[09:34:51] <madsci44> those are steppers
[09:34:59] <madsci44> i designed my own closed loop drivers
[09:35:03] <WalterN> hmm
[09:35:16] <madsci44> they position microsteps like a servo
[09:36:30] <madsci44> i can get 600 ipm out of them but not with emc yet
[09:37:33] <madsci44> each motor is profiled to the drive for torque at each position in the cycle
[09:37:56] <alex|lappy> oh man, the grand nagus is bangin' quark's mom
[09:37:58] <alex|lappy> pwned
[09:38:58] <WalterN> I'm going to bed
[09:39:09] <WalterN> more questions to come
[09:39:15] <madsci44> cool
[09:39:15] <WalterN> after sleeps :)
[09:39:20] <madsci44> have a good one
[09:39:53] <madsci44> http://www.kelinginc.net/CNCSpindleandController.html
[09:40:05] <madsci44> might be a source if you havent checked them already
[09:43:42] <alex|lappy> hmmmm
[09:43:52] <alex|lappy> what is the OD of that?
[09:44:01] <madsci44> spindle?
[09:44:03] <alex|lappy> yeah
[09:44:04] <madsci44> theres a few there
[09:44:51] <alex|lappy> it would be neat to mount one of those to my mill's quill...
[09:45:57] <alex|lappy> oooh, the small one is only 2
[09:45:59] <alex|lappy> 2"
[09:46:02] <alex|lappy> OD
[09:46:19] <madsci44> yeah i was contemplating that for the router/engraving head
[09:46:24] <alex|lappy> 5
[09:46:27] <alex|lappy> 5" long
[09:46:43] <alex|lappy> make a R8 taper holder for it that also screws into some holes drilled and tapped into my quill
[09:46:59] <alex|lappy> it would be sweet
[09:48:34] <madsci44> 2.56 dia - you could almost make an adapter sleeve hehe
[09:48:52] <madsci44> they have the bt30 versions below
[09:50:47] <alex|lappy> hmmm, what I really want to do is make a 4th axis
[09:51:20] <madsci44> like rotary table?
[09:51:56] <madsci44> or like horizontal axis
[09:52:25] <alex|lappy> http://www.youtube.com/watch?v=glYDe94hxb0&feature=player_embedded
[09:53:04] <alex|lappy> I realllly like his setup
[09:53:10] <alex|lappy> it seems sweet and simple to me
[09:53:59] <madsci44> yeah thats cool - looks like lever cams a brake lever against the pulley tho??
[09:54:36] <alex|lappy> nah, he mounted a motor cycle disk break
[09:54:44] <alex|lappy> so it clamps a disk
[09:54:58] <madsci44> ah yeah that would be my choice now i have to look again hah
[09:55:36] <madsci44> oh duh yeah i didnt even see it
[09:56:23] <madsci44> i was thiking once of making like a mini lathe setup that bolts to the mill table using it as a bed
[09:56:57] <madsci44> ill not have time for that until my next life im sure
[09:59:09] <alex|lappy> I have a 10x22 lathe
[09:59:19] <alex|lappy> it could probably fit wholesale on my mill table
[09:59:28] <alex|lappy> dunnknow that it could support like 1200lbs though
[10:01:09] <madsci44> i was thinking like to use supported shafting on a jig that goes on table surface as way guides, and bushings overtop
[10:01:50] <madsci44> simple tailstock and a headstock with a decent bore
[10:02:28] <madsci44> without all the transmission stuff i the headstock, and without the gearbox and apron and all that
[10:02:49] <madsci44> and using the table as the bed structure i think it would be low enough in weight
[10:03:48] <madsci44> im learning now tho its better to have seperate tools
[10:03:51] <alex|lappy> hmmm
[10:03:56] <madsci44> because setup always takes so much time
[10:04:05] <alex|lappy> I wonder if you could find a lathe with totally junked ways, but a good headstock and bearings
[10:04:11] <madsci44> but for that idea it would be almost worth dedicating to a mill as a machining center concept
[10:04:27] <alex|lappy> just ditch the ways entirely, mount the lathe tooling to the side of the quill
[10:04:32] <alex|lappy> er
[10:04:34] <alex|lappy> side of the head
[10:04:46] <alex|lappy> make a little gang tooling setup
[10:05:55] <madsci44> maybe - lathes have short stout supports for tooling so i dont know if you could get that from a mill head where theres this longer structure looping around to the part
[10:06:10] <madsci44> like maybe would chatter?
[10:06:31] <madsci44> so i would still make a little carriage for the turning jobs
[10:06:55] <madsci44> or more like saddle
[10:07:57] <alex|lappy> I mean, you could just machine a large ass block of steel with 2 or 3 1/2" lathe tool pockets in it
[10:08:10] <madsci44> hmm yeah actually
[10:08:21] <madsci44> you have x and y there already so theres a point there
[10:08:31] <alex|lappy> bolt it to the head so that they hang just a bit below the quill when it is fully retracted
[10:08:44] <alex|lappy> I mean, you won't be taking any huge cuts
[10:08:52] <madsci44> or mount a tool block to the column - like where you would have a horizontal hhead in a universal mill
[10:09:26] <madsci44> is your column dovetail ways or box?
[10:09:36] <alex|lappy> even on a larger knee mill I bet you would be lucky to take .1 on a pass
[10:09:40] <alex|lappy> dovetail
[10:09:46] <alex|lappy> but probably 2x as wide as a bridgeports
[10:10:13] <madsci44> well if you put the tool block there i bet that would help
[10:10:33] <madsci44> clamp on the upper section or something
[10:10:56] <madsci44> mine has box ways but could do same idea
[10:11:04] <alex|lappy> hahhaa, I could just install a quickchange toolpost on the column
[10:11:23] <madsci44> then still have the mill head available for those ops
[10:11:53] <alex|lappy> oh shit
[10:11:56] <alex|lappy> it is like 3am
[10:12:20] <madsci44> yeah 6 here
[10:12:24] <madsci44> ugh!
[10:12:42] <alex|lappy> madsci44, http://www.youtube.com/watch?v=p0kIWpylrqk&feature=related
[10:14:35] <madsci44> theres the idea hah
[10:15:01] <madsci44> perfect spindle concept - but see? look how close the gang tooling comes to the bearing block
[10:15:07] <alex|lappy> meh
[10:15:19] <alex|lappy> pretty near is not touching
[10:15:54] <alex|lappy> I mean, I spent a lot of time within .05 of the chuck on lathes
[10:15:56] <madsci44> thought it doesnt ahve to go past center either so maybe not so important
[10:16:14] <madsci44> but his is more like a bed mill than a knee mill
[10:16:26] <alex|lappy> yeah, that is the problem
[10:16:41] <alex|lappy> its why I kinda like the idea of mounting the toolpost on the column
[10:16:58] <alex|lappy> if you had the lathe tools on the same plane as the table
[10:17:11] <alex|lappy> you could set lathe tool height with the knee
[10:17:20] <alex|lappy> and Z axis on the mill using the quill
[10:18:19] <madsci44> yeah
[10:18:27] <madsci44> fersure more rigid I think
[10:18:41] <madsci44> plus in that case if your turning your table is back up close to the column
[10:19:30] <alex|lappy> well, i have a bit more throw on Y then a normal bridgeport does, which would allow me to push the ram out a little bit and be able to have a fair bit of room to manouver
[10:20:06] <madsci44> yeah that helps
[10:20:48] <madsci44> mine has 16" y travel, 34" x (more like 30 practially tho)
[10:23:14] <StonedSlacker> G'mornin' mah kneegrows! Say, can anyone tell me what MAX_ACCELERATION does?
[10:23:38] <alex|lappy> uh, sets your max acceleration rate in units/s^2
[10:23:45] <alex|lappy> so if it is 1
[10:23:50] <alex|lappy> then after 1 second your speed is 1
[10:24:01] <alex|lappy> after 2 seconds your speed is 2
[10:24:02] <alex|lappy> etc etc
[10:24:18] <StonedSlacker> ah, okay, duh. That is simple lol
[10:24:22] <StonedSlacker> how about STEPGEN_MAXACCEL
[10:24:29] <madsci44> what is it at 0.5 seconds
[10:24:39] <StonedSlacker> .5 :D
[10:24:45] <madsci44> hehe
[10:25:18] <madsci44> stepgen_maxaccel is same idea except sometimes the software needs to erm think ahead
[10:25:56] <madsci44> so you make that maybe somewhere between 5 to 15% higher than maxaccel
[10:26:04] <StonedSlacker> I'll just keep it on 1.0, screw it
[10:26:14] <StonedSlacker> ahh, okay
[10:26:28] <madsci44> if you get following error it can be because there is not enough margin between those two values i think
[10:27:05] <celeron55> stepgen_maxaccel is the absolute maximum and max_acceleration is the maximum the software tries to obey i guess
[10:27:17] <madsci44> .25
[10:28:27] <StonedSlacker> that makes sense
[10:28:45] <madsci44> im not exactly sure its something like that
[10:28:48] <StonedSlacker> Man, etch-a-sketches have wicked backlash lol
[10:29:07] <madsci44> lol
[10:29:47] <madsci44> at least you dont need coolant
[10:29:49] <StonedSlacker> I might shelve it today and start something else. A baby mill or lathe
[10:30:12] <StonedSlacker> madsci44:You havent seen me etch :^D
[10:30:33] <madsci44> hah - ah yea the carbide age
[10:30:55] <StonedSlacker> Seriously, I get fuckin chatter lmao
[10:31:04] <alex|lappy> StonedSlacker, you cant take care of that in the ini?
[10:31:29] <StonedSlacker> I'm working on it that, alex
[10:32:01] <alex|lappy> hmm, i think the first step would be to figure the turn ratio
[10:32:09] <StonedSlacker> I have to get my scale set right first
[10:32:12] <alex|lappy> huh, wait
[10:32:25] <alex|lappy> is it the same over the entire travel?
[10:32:36] <StonedSlacker> Should be
[10:32:47] <alex|lappy> not necessarily
[10:32:57] <StonedSlacker> If the line builds up then no
[10:33:17] <StonedSlacker> I havent taken it apart, so I dunno
[10:33:41] <StonedSlacker> Maybe after I get my medicine and get settled in this morning I will take it apart and see
[10:33:48] <alex|lappy> no, it is a pully and wire system
[10:33:53] <alex|lappy> http://entertainment.howstuffworks.com/question317.htm
[10:34:22] <alex|lappy> so it does have the same ratio over the entire travel
[10:35:58] <StonedSlacker> hmm, Imma take this shit aprt :D
[10:36:27] <alex|lappy> i need to get my little engines running....
[10:36:32] <StonedSlacker> nah, I'll go buy a new one for that. I gotta have one for drawin'
[10:37:11] <alex|lappy> http://www.flickr.com/photos/33383121@N04/4670150830/sizes/l/
[10:37:14] <alex|lappy> they are soooo close
[10:37:57] <madsci44> what are those?
[10:38:06] <alex|lappy> my project last term in school
[10:38:09] <StonedSlacker> I cant wait to get moved back into somewhere that I can use my lathe and build shit again. Alex, I so jealous I might have to hate on you for a while.
[10:38:16] <madsci44> look like little rc engines
[10:38:20] <alex|lappy> me and a couple of kids made a couple of little engines
[10:38:21] <alex|lappy> yeah
[10:38:27] <alex|lappy> 1.8cc diesel
[10:38:31] <madsci44> too cool
[10:38:42] <alex|lappy> none of them work yet, I have one more super simple part to make
[10:38:49] <alex|lappy> just have not gotten around to it
[10:38:58] <StonedSlacker> That's the same thing in my wife's hundyai
[10:39:31] <madsci44> hehe
[10:41:21] <alex|lappy> why did I start watching another DS9 episode
[10:42:26] <madsci44> you are curious if the FCA and grand naggis will get investigated by the obama administration?
[10:43:46] <madsci44> i need to ream c95400
[10:44:01] <madsci44> have you ever worked with that stuff alex? /
[10:45:45] <alex|lappy> madsci44, not personally, but I know a couple people who have
[10:46:41] <madsci44> its supposed to have a a rating of 60% - i have to ream a 1" dia hole 6 " deep on a lathe
[10:47:13] <alex|lappy> I was watching somoene drill and bore some, and it looked like a horrific process
[10:47:17] <alex|lappy> the end product was really nice
[10:47:21] <madsci44> yeah
[10:47:27] <alex|lappy> but it is just so fucking hard
[10:47:31] <alex|lappy> well
[10:47:37] <alex|lappy> not hard, but... I don't even know
[10:47:40] <alex|lappy> almost plastic?
[10:48:00] <alex|lappy> it yeilds a bunch and then flakes
[10:48:03] <alex|lappy> it is strange
[10:48:10] <madsci44> i read its a bit like stainless in some ways, alloy in others, and idonno
[10:48:14] <alex|lappy> yeah
[10:48:21] <alex|lappy> I would take large cuts to avoid work hardening
[10:48:32] <alex|lappy> well, as large as you have the HP and rigidity to do
[10:48:35] <madsci44> wondering how much to leave for reaming - the work hardening is what scares me the most
[10:48:57] <alex|lappy> but I would also drill really close to the finsih ID
[10:49:15] <madsci44> yeah - like maybe 0.005 on the dia
[10:49:18] <alex|lappy> I mean, 63/64ths
[10:49:44] <alex|lappy> hmm, 1" ID?
[10:49:47] <madsci44> oh thats like .015 under
[10:49:49] <madsci44> yeah
[10:49:50] <alex|lappy> might be better bore it
[10:49:53] <madsci44> yeah
[10:49:57] <alex|lappy> you will have more info on what is happening
[10:49:59] <madsci44> boring shouldnt be too bad
[10:50:18] <madsci44> its hollowbar so i only have to come up from 3/4
[10:50:37] <alex|lappy> how big of a lathe
[10:50:48] <madsci44> 1440 china
[10:51:20] <alex|lappy> so a 3hp motor?
[10:51:25] <madsci44> yeah
[10:51:31] <alex|lappy> yeah, go for a fat pass
[10:51:39] <alex|lappy> .075 if you can take it
[10:51:41] <madsci44> bearings are good
[10:51:46] <alex|lappy> depends on how nice of a boring bar you have
[10:51:53] <madsci44> yeah thats the main thing
[10:52:37] <madsci44> reamer is a straight flute chucking reamer
[10:52:46] <madsci44> problem is i dont have a chuck to fit the shank on it hah
[10:52:48] <alex|lappy> you making a bushing?
[10:52:55] <madsci44> yeah
[10:53:21] <madsci44> i want to ream because i need to do alot of them and they need to be close
[10:53:34] <alex|lappy> are you machining shafts to fit?
[10:53:43] <madsci44> shafts are predone
[10:53:50] <madsci44> 0.9990
[10:53:57] <alex|lappy> oooh
[10:54:15] <madsci44> i have a 0.9995 reamer and a 1.0000
[10:54:39] <alex|lappy> got a inch of stock to test on?
[10:54:51] <madsci44> yes
[10:55:05] <madsci44> thats a good idea
[10:55:16] <madsci44> instead of going full depth right off the bat
[10:55:32] <alex|lappy> I mean, you could just try drilling to 63/64ths and jam the reamer in there
[10:55:38] <madsci44> yeah
[10:56:25] <madsci44> ill still bore it tho i hate drilling that big
[10:58:18] <alex|lappy> mad, with a 3/4 thru hole already there you might be able to push the drill really fast
[10:58:28] <alex|lappy> hence avoid some of that hardening
[10:58:40] <madsci44> oh! i see what you are getting at yeah thats a thought
[11:00:46] <madsci44> lots of coolant
[11:01:06] <alex|lappy> flood that mofo
[11:01:40] <alex|lappy> oh wow, if you could get coolant running through the headstock bore and into the back of the part
[11:01:44] <alex|lappy> so awesome
[11:01:59] <madsci44> hahahaha
[11:02:11] <madsci44> i have to hook the fire hose up for that
[11:02:49] <madsci44> actually you know now that i think about it
[11:02:55] <madsci44> i have this little jet head
[11:02:59] <madsci44> if i aim it just right
[11:03:03] <madsci44> thats brilliant!
[11:03:55] <alex|lappy> if you actually do that, I demand a picture
[11:04:29] <madsci44> i will be the guy in the life jacket
[11:08:56] <madsci44> yep thats exactly what ill do ill flood it from the back
[11:11:26] <alex|lappy> madsci44, http://www.copperinfo.co.uk/alloys/bronze/downloads/pub-83-al-bronze-alloys-for-industry.pdf
[11:15:31] <alex|lappy> anyway, good night all!
[11:15:41] <alex|lappy> madsci44, i really do want picture. that is going to be crazy messy.
[11:18:06] <madsci44> i will see what I can do
[11:18:18] <madsci44> thanks for the suggstions and have a good night!
[11:23:27] <micges> madsci44: hi
[11:23:37] <madsci44> hi
[11:26:26] <micges> I'm trying to reporduce bug you've submitted
[11:26:48] <madsci44> oh ok
[11:26:57] <micges> on what emc version you've noticed it?
[11:27:10] <madsci44> i was just going to say i forgot to write that - 2.4.1
[11:27:27] <madsci44> downloaded last weekend
[11:27:38] <micges> from source?
[11:28:13] <madsci44> I tried both, now im not totally sure which one i ended up building because I switched a couple times due to another issue
[11:28:13] <micges> or from packages?
[11:28:22] <madsci44> i did not use packages - just the tarball
[11:28:28] <madsci44> run in place
[11:28:34] <madsci44> is it not working the same on yours?
[11:28:51] <micges> I see, what is another issue?
[11:29:17] <micges> it works the same
[11:29:46] <madsci44> i built for RTAI 3.8 on kernel 2.6.34 - and during the compile there was an issue in any driver that called the a pci read function to test if a device was present
[11:31:12] <madsci44> hal_m5i20.c: In function \u2018init_module\u2019:
[11:31:26] <madsci44> hal_m5i20.c:288: error: implicit declaration of function \u2018pci_find_device\u2019
[11:31:35] <madsci44> hal_m5i20.c:288: warning: assignment makes pointer from integer without a cast
[11:32:09] <micges> ah I remember, youve told me that
[11:32:13] <madsci44> it happened in several other drivers as well - i simply replaced the function call with a null since I do not need those drivers
[11:32:23] <_TwisT_> hi all!
[11:32:27] <micges> hi
[11:32:28] <madsci44> i figure it was probably due to a compiler version or something
[11:32:41] <madsci44> so i did not post it as a bug
[11:32:57] <madsci44> hi
[11:34:01] <madsci44> if you want me to find any other info - logs or whatever i am more than happy to try
[11:36:29] <micges> cool, thanks, we will back to compile problem, now I'll try to locate G92 problem
[11:36:50] <_TwisT_> I'm wondering, how the standard step/dir motor driver should behave, when there is no "step" signals on it's input?
[11:37:20] <_TwisT_> Should it pull down all mosfets or just leave them open accurding to last phase?
[11:38:51] <micges> when you pull down mosfets there is no way to say if motor didn't move
[11:39:29] <micges> our driver pull down current to 1/4 when motor stops for few sec
[11:40:00] <madsci44> the ratio of currents between the phases should stay constant
[11:43:27] <_TwisT_> micges: I have an leadscrew transmission, so (I think) there is no need to lock the spindle
[11:45:33] <micges> _TwisT_: so you have stepgen driving spindle? with encoder?
[11:47:55] <micges> on steppers there is issue when you decrease current below some level it can 'skip' to full step phase
[11:50:27] <celeron55> why couldn't one just keep the full current flowing
[11:50:44] <celeron55> stepper systems should be designed so that it can be done anyway
[11:51:56] <micges> celeron55: yes I know
[11:52:50] <micges> we should ask _TwisT_ why is he asking
[11:53:50] <_TwisT_> I develop my own step/dir controller :)
[11:54:13] <_TwisT_> and there is no encoders or any other type of feedback yet
[11:54:16] <celeron55> just use full current always
[11:55:23] <micges> current decreasing is only to decrease of heating of motors
[11:55:26] <celeron55> or maybe 1/2 when idle, if torque isn't needed at all when stopped
[11:57:46] <celeron55> anyway, never decrease it near 0
[11:58:07] <_TwisT_> micges: thanks for pointing me out the problem with full step phase
[11:58:11] <celeron55> you can't assume a stepper to stay where it is when there is no current
[11:58:29] <_TwisT_> celeron55: ok, thanks for advice
[11:59:01] <micges> _TwisT_: welcome
[12:23:37] <StonedSlacker> My motors are current starved
[12:24:12] <StonedSlacker> ATX power supply is low on balls
[12:25:03] <cradek> madsci44: thanks for filing that report
[14:14:37] <Valen> if you dont load the 5V rail an atx supply wont put out its 12V well at all
[14:15:07] <Valen> and at higher currents they expect the load to be balanced at a certain % between 5v and 12v
[14:15:28] <Valen> we use a car blinker bulb as a load on 5V to get the +12 somewhere near right
[14:15:33] <celeron55> atx power supplies have only a single switcher regulator with many secondary coils
[14:15:52] <celeron55> and if a single output voltage raises too high, it shuts down
[14:16:33] <celeron55> they control the switcher based on something like an average output voltage of all the outputs
[14:17:04] <celeron55> (if somebody needed an explanation)
[14:28:19] <thiel> Hi from germany
[14:28:44] <frallzor> heil from Germany?
[14:28:50] <frallzor> jk =)
[14:28:53] <frallzor> I love germans
[14:29:04] <frallzor> or the girls atleast :P
[14:29:48] <thiel> trying to run axis in lucid linux
[14:31:02] <micges> thiel: and what is result?
[14:31:32] <thiel> i even do not find the page where are the instructions
[14:34:51] <micges> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Ubuntu10.04Notes
[14:38:11] <thiel> micges: this is for full install i only need it to view the toolpath so there wars a no reeltime install of axis anywhere for 9.10
[14:41:22] <micges> thiel: you can download source, install packages listed on bottom of that page and compile emc simulator (with no rt part) to preview paths
[14:41:29] <thiel> oh i see there are packetches pant for this month to 10.04
[14:47:57] <thiel> got it
[15:16:41] <thiel> micges: YES it runs !
[15:17:06] <thiel> axis 2.5.0
[15:17:58] <micges> cool
[15:18:28] <thiel> i workt thru this http://wiki.linuxcnc.org/emcinfo.pl?Installing_EMC2#Building_emc2_simulator
[15:18:54] <thiel> so by
[15:30:55] <jthornton> hmmm I got the busybox prompt while trying to load 10.04 on the new Zotac motherboard
[15:32:33] <jthornton> 8.04.4 seems to load ok
[15:33:34] <jthornton> well no I got the busybox prompt on that one too
[16:21:43] <boot|sleep> boot|sleep is now known as bootnecklad
[16:35:42] <cncmachineguy> can anyone tell me which doc to look in for info on custom m codes? how to define them and such?
[16:36:00] <Endeavour> NO!
[16:36:11] <Endeavour> ;)
[16:36:27] <cncmachineguy> well fine then
[16:36:28] <Endeavour> (No idea, just harassing.)
[16:36:29] <cncmachineguy> lol
[16:38:18] <cncmachineguy> i feel teased, everything i read keeps telling me m101-m1xx is for user defined, but cant seem to find how to define them
[16:40:16] <SWPadnos_> Cncmachcncmachineguy, I don't know which manual, but they're just executables in the nc_files dir named M100, M101 ...
[16:40:37] <SWPadnos_> Err - phone paste
[16:41:55] <cncmachineguy> executables for who? hal? emc? ubuntu?
[16:42:59] <SWPadnos_> Anything executable
[16:43:18] <cncmachineguy> ok, thanks
[16:44:10] <cncmachineguy> will emc wait for it to execute before continuing?
[16:44:21] <SWPadnos_> So it could be a bash script that uses halcmd, maybe
[16:44:24] <SWPadnos_> Yes
[16:46:31] <cncmachineguy> cool, so i prolly need to figure out halcmd's to do what i need to do.
[16:47:38] <cncmachineguy> this is so much fun! i love all this control.
[16:48:18] <SWPadnos_> Most likely halcmd sets <signalname> <value>
[16:48:33] <SWPadnos_> Yeah, it's pretty cool
[16:49:15] <cncmachineguy> then next hal cmd wait <signalname2> true, then done?
[16:55:45] <cncmachineguy> thanks sw
[17:03:03] <jthornton> cncmachineguy it is in the User Manual
[17:03:04] <SWPadnos_> There's no wait in halcmd - when it exits, the job was done
[17:03:27] <SWPadnos_> Gotta run, about to start driving
[17:04:45] <jthornton> cncmachineguy http://www.linuxcnc.org/docview/html//gcode_main.html#sec:M100-to-M199:
[17:26:23] <davidf> Hi
[17:27:43] <davidf> Does anyone here own a Proxxon hand-held rotary tool?
[17:28:05] <KimK> Is that like a Rotozip?
[17:28:16] <davidf> Hey Kimk yes
[17:28:28] <davidf> or Dremel
[17:28:37] <davidf> But better quality
[17:28:47] <KimK> Don't own a Rotozip, but I've seen them on TV, lol
[17:29:09] <KimK> Had a Dremel once, now gone.
[17:29:22] <davidf> I need to know the diameter of the threaded shaft the mounts the collet nut.
[17:30:23] <KimK> Making your own spindle?
[17:30:42] <davidf> I'm making a tool that will hold a Proxxon collet nut and collets. I have the collet nut and collets, but I don't have a way to deduce the exact size of the male.
[17:30:58] <davidf> Sort of. A pin vise is more like it.
[17:31:01] <frallzor> why use that?
[17:31:08] <frallzor> and not something more standard?
[17:32:13] <davidf> Making a holder for mandrels used in glass blowing. Has to handle pretty high heat. Proxxon collets are steel.
[17:34:56] <Endeavour> KimK: Thinking about getting this - http://www.newegg.com/Product/Product.aspx?Item=N82E16813121399
[17:40:39] <frallzor> intel boards rock
[17:43:18] <morficmobile> nothing wrong with a tasty amd board
[17:43:51] <morficmobile> but those atoms seem to get pimped a lot in emc world :)
[17:44:32] <frallzor> yeah I got my first atom ever to use with emc
[17:44:39] <frallzor> im real happy with it
[18:25:24] <KimK> Endeavour: Yes, a lot of people have mentioned atoms, maybe I'll get to see one at the workshop.
[18:36:14] <MattyMatt> it seems a trifle bizarre to use an atom (low power & quiet) on a machine tool
[18:36:48] <fragalot> MattyMatt: most professional CNC machines still use 200Mhz cpus
[18:37:01] <fragalot> a 1.6Ghz atom is going to do just fine.
[18:37:04] <MattyMatt> it sounds like a job for a Pentium D, to me
[18:37:27] <MattyMatt> something that too hot & noisy for a desktop
[18:37:40] <fragalot> why would you waste power?
[18:38:05] <fragalot> atoms are effecient, and can be cooled passively, so you can have a closed case, so no chips can get in.
[18:38:26] <MattyMatt> to save capital. I suppose atoms are about as cheap as another mini-itx solution tho
[18:38:41] <fragalot> MattyMatt: you save with an atom in the long run
[18:38:49] <fragalot> and they're dirt cheap anyway
[18:38:54] <Roguish> atoms are great, add an SSD !!!!
[18:40:24] <Roguish> under $200 for atom board + SSD + ps + 1g mem
[18:40:37] <fragalot> s/SSD/SD card/
[18:40:55] <Jymmm> under $100 actually
[18:40:55] <fragalot> proper SSDs are too expensive, and SD cards provide enough speed/storage for this purpose.
[18:41:14] <bootnecklad> bah... who needs 200hmz CPUs, 1.6GHz atoms, SSDs or w/e
[18:41:27] <bootnecklad> why dont we all develop systems using good 'ol TTL
[18:41:35] <Roguish> 16g SSD $90 at Fry's
[18:41:41] <fragalot> bootnecklad: I actually have one of those laying around here
[18:41:47] <MattyMatt> those cheap ssd for fast booting are OK, about $80 or so for 16Gb or so
[18:41:50] <fragalot> bootnecklad: but it doesn't have a parport. :P
[18:42:00] <fragalot> MattyMatt: thats expensive imo.
[18:42:06] <bootnecklad> fragalot thats awesome
[18:42:26] <fragalot> bootnecklad: :)
[18:42:35] <fragalot> bootnecklad: it boots via tapes :P
[18:42:50] <MattyMatt> if you have to buy a hdd instead, $80 isn't so bad, IMO :)
[18:43:01] <MattyMatt> those cheap ones have crap write speeds tho
[18:43:08] <bootnecklad> thats just... crazly awesome...
[18:43:21] <fragalot> MattyMatt: if you're going for 16gb, why not an sd card for less than half the price?
[18:43:33] <fragalot> the speed 'l be the same as a cheap ssd
[18:43:40] <Roguish> if it's just a machine controller, speed shouldn't matter so much.
[18:43:41] <fragalot> hell, it *is* an ssd :P
[18:43:43] <bootnecklad> I am thinking of making a simple plotter, maybe using some old stepper motors from amstrad PCW printers and other simple parts. just need a TTL system that uses G code
[18:43:54] <bootnecklad> time to learn G-code!
[18:43:59] <MattyMatt> fragalot, I went even further, all my backups or on usb sticks now :)
[18:44:25] <MattyMatt> it was cheaper than kitting all my machines out with SD reader
[18:44:44] <fragalot> hell usb sticks work just fine too :P
[18:45:11] <fragalot> so you just need a spare 4gb usb stick, mobo, cpu, ram, psu, done.
[18:45:26] <MattyMatt> I preume an atom mobo could boot from usb stick without problems
[18:45:29] <fragalot> yeah
[18:45:49] <fragalot> done it on my Eee lots of times
[18:46:03] <MattyMatt> bootnecklad: got a whole pcw system?
[18:47:06] <bootnecklad> I had a few but I only need the one
[18:47:11] <bootnecklad> pcw9512s
[18:47:24] <bootnecklad> I also have a +, so I have one with a 3" and one with a 3.5" drive
[18:48:44] <bootnecklad> have a few z80s from the boards, quite a few 256k*1 DRAM chips, never used DRAM before
[19:08:14] <bootnecklad> [19:41] <fragalot> bootnecklad: I actually have one of those laying around here
[19:08:20] <bootnecklad> some pictures would be excellent!
[19:08:26] <bootnecklad> video would be even better
[19:09:01] <fragalot> bootnecklad: it's a Sinclair QL
[19:09:13] <fragalot> it's an old personal computer, and is entirely TTL based.
[19:09:24] <pjm> cool fragalot ! I have a QL also, they were wicked
[19:09:30] <fragalot> pjm: ^__^
[19:09:52] <pjm> i think i have almost 1 of each sinclair micro
[19:10:00] <pjm> they rocked bigtime
[19:10:28] <fragalot> i love how you needed t run the tapes in RAID1 because they were so reliable
[19:10:48] <bootnecklad> Based on a Motorola 68008 processor clocked at 7.5 MHz
[19:11:15] <pjm> i still have a bunch of the 3" disks for the 9512 also
[19:11:21] <pjm> as they were used in the speccy +3
[19:11:37] <bootnecklad> I
[19:11:55] <bootnecklad> I need to replace the belt on my amstrad, then get cp/m and locoscript running
[19:18:07] <bootnecklad> btw pjm, have I linked you to infomation on my cpu? You can find videos: http://www.youtube.com/user/bootnecklad They may get boring becuase I start to ramble... something oh look a butterfly
[19:18:22] <bootnecklad> and stuff
[19:18:24] <bootnecklad> http://marc.cleave.me.uk/cpu/index.htm
[19:18:30] <pjm> lol cool i will look
[19:18:31] <bootnecklad> please excuse the crude website
[19:18:41] <pjm> i like old micros from the 80's
[19:18:59] <pjm> my missus reckons i still live in that decade! how dare she have an opinion!!
[19:19:21] <bootnecklad> hmm, I'd like to be in the '60s as I am now
[19:19:39] <bootnecklad> I could make computers that people would actually use!
[19:20:36] <pjm> heh
[19:20:57] <pjm> we have a old core-store card in a picture frame in the house, my missus hates it with a passion, but i reckon its cool
[19:21:38] <bootnecklad> magnetic core memory? in a frame?
[19:22:16] <pjm> well ferrite core store, its about 4 X 4 panels of the stuff
[19:22:36] <bootnecklad> that makes you a proper nerd
[19:22:39] <pjm> give me 5 mins and i'll take it out and photo it
[19:22:44] <pjm> it is cool!
[19:24:07] <bootnecklad> i should buy a single card or so, implement it into my CPU and then watch some vintage-collector-nerds rage at me for using a priceless object
[19:25:41] <pjm> lol my missus just said "are you finally taking down that crap"
[19:25:43] <pjm> what a cheek
[19:26:14] <bootnecklad> it hurt a little inside, how could she say such a thing!
[19:26:39] <bootnecklad> oh wow 8k!
[19:26:41] <bootnecklad> http://cgi.ebay.com/Data-General-NOVA-800-8K-CORE-Memory-PCB-1971-GOOD-RARE-/130400596970
[19:27:13] <fragalot> * fragalot still smells like 2stroke fuel after showering twice
[19:27:14] <fragalot> D:
[19:27:21] <bootnecklad> :D
[19:28:11] <pjm> bootnecklad http://pjm.dyndns.org/mem.jpg
[19:28:21] <pjm> and http://pjm.dyndns.org/mem1.jpg
[19:28:38] <pjm> the card has an inspected OK sticker, 26/3/71
[19:28:42] <pjm> it existed before me!
[19:29:03] <fragalot> haha
[19:29:20] <fragalot> I used to have a 3D game for my sinclair!
[19:29:24] <fragalot> BJ in 3D land iirc
[19:30:01] <pjm> brb need to re-hang picture, taking ear defenders this time
[19:30:16] <fragalot> o.O
[19:30:38] <bootnecklad> ear defenders?!??!
[19:31:08] <bootnecklad> jesus christ
[19:32:45] <bootnecklad> thats a lot of ferrite cores
[19:33:00] <pjm> back, funnily enough nothing heard from the missus
[19:33:06] <pjm> i saw her mouth moving though
[19:33:10] <fragalot> pattex 'no nails' seems to be utter crap
[19:33:33] <bootnecklad> bahaha you live dangerously pjm
[19:33:57] <bootnecklad> pjm: you didnt really take ear defenders?
[19:34:01] <pjm> lol i do, i have returned to my workshop
[19:34:04] <pjm> yep
[19:34:08] <fragalot> I glued 2 bits of rough wood together this morning roughly 100x100cm, just pulled it apart with my hands :/
[19:37:45] <bootnecklad> so pjm by now you should be fluent in titan machine code, go into a dark room and make me an operating system
[19:39:04] <pjm> lol
[19:39:14] <pjm> thats an interesting cpu design
[19:39:44] <bootnecklad> thanks!
[19:41:36] <fragalot> hah
[19:41:44] <fragalot> just looked up the spec for that pattex glue
[19:41:49] <fragalot> "100kg/cm2"
[19:42:08] <fragalot> so supposedly I just pulled 1 kiloton apart with my hands ^___^ *strong*
[19:42:13] <bootnecklad> awesome fragalot!
[19:45:07] <bootnecklad> pjm I have 256 available general purpose single byte I/IO ports
[19:45:14] <bootnecklad> what I/O should I have?
[19:46:43] <pjm> are u gonna build the uarts etc out of 74xx logic, or use 8255/s etc?
[19:47:20] <bootnecklad> I havnt thought of UARTs yet
[19:47:25] <bootnecklad> dont even know how they work >.<
[19:47:33] <bootnecklad> serial communication yet
[19:47:36] <bootnecklad> all I know though lol
[19:47:57] <bootnecklad> i need a parralel to serial shift register converter thing
[19:48:00] <bootnecklad> or something
[19:48:09] <pjm> i guess it would be worth making some 8 bit ISA compatible interface
[19:48:13] <pjm> so you can use old PC cards
[19:48:59] <bootnecklad> that would be epic
[19:57:44] <archivist> * archivist just returened from the vintage computer festival at bletchley
[19:57:54] <pjm> cool! i want to go there
[19:58:00] <pjm> to see the crypto machines
[19:58:13] <bootnecklad> arrrgh
[19:58:19] <archivist> on tomorrow too
[19:58:20] <bootnecklad> * bootnecklad envys archivist
[19:59:39] <bootnecklad> do you think they would let me take home some of collousousous
[19:59:39] <Endeavour> Anyone here have any experience with the D510 Atom Motherboard and EMC2?
[20:02:15] <frallzor> the one you showed earlier?
[20:08:21] <Endeavour> frallzor: Asking me?
[20:08:27] <frallzor> yup
[20:08:31] <Endeavour> http://www.newegg.com/Product/Product.aspx?Item=N82E16813121399
[20:08:33] <Endeavour> That one
[20:08:39] <frallzor> was thinking about that, no parallell port?
[20:08:45] <alex|lappy> hmm, i have been working in my shop at home 10-12 hours a day since the 8th
[20:08:46] <Endeavour> Has a header for a port.
[20:08:49] <alex|lappy> I shold probably take a day or two off
[20:09:01] <skunkworks> http://www.youtube.com/watch?v=4nuRea6615s
[20:09:29] <skunkworks> Did I mention how awesome emc2 is?
[20:10:10] <alex|lappy> what the hell is the taper on that machine?
[20:10:13] <alex|lappy> WHAT THE HELL IS THAT MACHINE?
[20:10:18] <alex|lappy> err
[20:10:20] <alex|lappy> without caps
[20:10:24] <frallzor> its a monster
[20:10:39] <skunkworks> of the hundreds of cycles of testing - no missed tool
[20:12:28] <skunkworks> 32767 tool possiblitys ;)
[20:12:52] <Endeavour> Know anyone who has used it, frallzor ?
[20:13:21] <frallzor> I cant see why it wouldnt work, I use similar from intel too :)
[20:13:42] <elmo40> but no par port.
[20:13:58] <skunkworks> what we have in there now is tool 13317, 18513, 10, 15363, 7178 and 5
[20:14:04] <skunkworks> ;)
[20:17:16] <Endeavour> elmo40: It has a header for the parallel port.
[20:21:13] <skunkworks> http://electronicsam.com/images/KandT/oldkandt.JPG
[20:21:42] <alex|lappy> holy shit, i wish machines still looked like that
[20:21:54] <alex|lappy> skunkworks, that is great
[20:22:36] <alex|lappy> do yo guys have any suggestions on cheap but good enclosures?
[20:22:59] <alex|lappy> I don't want to spend $250 on a sheet metal box, but I also don't want like $1k in electronics to let out their blue smoke
[20:58:54] <Jymmm> SWPadnos: You rang (last night)?
[21:15:19] <alex|lappy> lulz
[21:15:25] <Jymmm> SWPLinux: : You rang (last night)?
[21:15:38] <alex|lappy> cheif Obrian is using a "micro lathe" to make doll house furniture
[21:15:49] <alex|lappy> or at least he says he is, they don't show it ;_;
[21:17:51] <SWPLinux> yeah, kinda
[21:18:35] <Jymmm> SWPLinux: Broker? Pimp? Mule? Controlled Substance?
[21:18:52] <SWPLinux> module name
[21:19:30] <Jymmm> SWPLinux: Module Named 'Tranquility' to Honor Apollo 11
[21:19:37] <Jymmm> SWPLinux: Anything else?
[21:19:44] <SWPLinux> gonna write a "safety relay" component (like charge pump, but one that monitors multiple heartbeats, etc), but of course I don't want to call a software module "safety xxx"
[21:20:07] <Jymmm> SWPLinux: Give me a moment to think about it
[21:21:47] <Jymmm> SWPLinux: diagnostic, monitor, environment
[21:23:10] <SWPLinux> ok, take your time
[21:24:05] <Jymmm> SWPLinux: diagnostic, monitor, environment, risk, management, asylum, safeguard
[21:24:34] <Jymmm> SWPLinux: I personally like 'monitor module' as it can be generic in nature.
[21:25:18] <Jymmm> SWPLinux: Kinda reminds me of SNMP too.
[21:25:50] <Jymmm> SWPLinux: "diag module"
[21:26:02] <Jymmm> SWPLinux: "diagnostic module"
[21:26:13] <SWPLinux> one word :)
[21:26:26] <Jymmm> SWPLinux: CYA
[21:26:37] <Jymmm> SWPLinux: CYA <--- That's one word
[21:26:38] <SWPLinux> heh
[21:26:47] <SWPLinux> MEGAMonitor
[21:26:53] <Jymmm> SWPLinux: CYA == Cover your Ass
[21:27:09] <Jymmm> not See'Ya
[21:27:13] <andypugh> Diagnostic?
[21:27:25] <andypugh> Triagnostic?
[21:27:31] <andypugh> Diatheist?
[21:28:01] <Jymmm> SWPLinux: "The MM (Monitor Module) will allow you to...."
[21:29:23] <Jymmm> SWPLinux: DAMN Module! "The Diagnostic And Monitor Node module would allow you to...."
[21:29:58] <alex|lappy> sooo I have a quick safety question
[21:30:13] <Jymmm> no such thing
[21:30:23] <alex|lappy> http://www.geckodrive.com/upload/ElectromechanicalEStop.pdf
[21:30:31] <alex|lappy> I am planning on making an estop like this
[21:30:48] <Jymmm> alex|lappy: No you're not. it has a defect in the design
[21:30:52] <alex|lappy> oh?
[21:30:56] <alex|lappy> please do tell
[21:31:00] <Jymmm> alex|lappy: No
[21:31:03] <alex|lappy> ;_;
[21:31:07] <alex|lappy> just gonna let me fry?
[21:31:09] <alex|lappy> or crush my hand
[21:31:11] <Jymmm> Yep
[21:31:18] <alex|lappy> so sad!
[21:31:23] <Jymmm> Eh
[21:31:32] <alex|lappy> anyway, my estop is only rated for lik 10A
[21:32:05] <Jymmm> No, not anyway. It really does have a MAJOR defect in the design that will fail in a very bad way.
[21:32:08] <andypugh> Dropping out the main contactor really ought to work.
[21:32:20] <alex|lappy> my rudimentary electronics knowledge says that the Estop is only going to have the current that the DPDT relay across it
[21:32:40] <alex|lappy> Jymmm, then really, would you like to explain why?
[21:32:48] <Jymmm> Just use a REAL safety relay and dont try to be a cheap bastard.
[21:32:49] <alex|lappy> i believe you, but i kinda want a working estop :D
[21:33:33] <Jymmm> alex|lappy: If you want to know, setup a bench test and find the flaw.
[21:33:47] <Jymmm> I found it in two minutes.
[21:33:53] <SWPLinux> I think you're overstating the problem
[21:33:55] <andypugh> A _REAL_ safety relay is more than my whole machine cost.
[21:33:56] <alex|lappy> wtf?
[21:34:06] <alex|lappy> Jymmm, uh, why would you not share the information?
[21:34:07] <Jymmm> andypugh: So?
[21:34:09] <alex|lappy> kind of a dick move
[21:34:31] <andypugh> I can grow new limbs
[21:34:39] <alex|lappy> I mean, if it was an issue of a document I should read, that is fine
[21:34:52] <alex|lappy> but honestly, I don't have the money to buy the stuff to test that circuit, and then not use it
[21:35:25] <SWPLinux> alex|lappy: think about what happens if you hold the start button in, regardless of whether the stop button is held
[21:36:06] <alex|lappy> ah, shorts out thru the relay
[21:36:15] <SWPLinux> the problem can be solved with a diode between the high sides of the two transformers
[21:36:45] <SWPLinux> so the button can't drive the main transformer, but can drive the holding coil (but only through the estop button)
[21:36:51] <andypugh> Talking of safety, bear in mind that in this video there is 300V DC onto the motor terminals, but I am actually too cheap to buy connectors.... http://www.youtube.com/watch?v=oyeJfNg3NfQ
[21:37:50] <andypugh> But I am using a 99p eBay isolation transformer.
[21:39:36] <alex|lappy> SWPLinux, a diode between the relay coil and the main coil, allowing current to only go main coil--> relay coil
[21:39:51] <alex|lappy> * alex|lappy is just trying to make sure he really understands
[21:43:29] <SWPLinux> yes
[21:44:39] <alex|lappy> SWPLinux, I also have no idea how to choose the correct component
[21:45:44] <alex|lappy> i have not had time to take a digital electronics class ;_;
[21:46:05] <SWPLinux> oh hmm, those are AC, so diodes aren't so useful
[21:48:23] <andypugh> Gosh Darn it chaps, I put my life on the line making that video at those voltages, and it has zero viewrs!
[21:48:45] <alex|lappy> andypugh, shows as 8 for me
[21:49:30] <Roguish> andypuch: how is the bldc comp cominig along
[21:49:44] <alex|lappy> I have been thinking that when I test my motors at full voltage I should probably ratchet strap them to the bench first
[21:50:03] <alex|lappy> not real tight, but enough for them to not jump off my desk if they accelerate real fast by accident
[21:50:05] <andypugh> Two are released to the devs, one is still in development.
[21:51:43] <Roguish> dumb question? is this something that could be done in the fpga?
[21:52:05] <Roguish> the commutation I mean.
[21:53:15] <andypugh> Yes, but why bother. Worst case you would need a different PIN file for every Hall sensor pattern
[21:53:46] <Roguish> ok
[21:54:04] <Roguish> just thinking about unloading the cpu.
[21:54:15] <andypugh> Anyway, the Hall sensors were a complex and inferior solution, the sinusoidal driver is simpler, smoother and better.
[21:54:40] <andypugh> But doing trignometry on an FPGA is asking for trouble
[21:57:00] <Roguish> ok another dumb question. sinusoidal does not need hal?
[21:57:27] <andypugh> Hal yes, Hall no. :-)
[21:57:51] <Roguish> sorry, hall sensor.
[21:58:03] <andypugh> Saves 3 IO pins at the expense of a motor wiggle at startup.
[22:00:31] <Roguish> ok, guess it depends on how much 'wiggle' is acceptable.
[22:02:52] <andypugh> 180 motor _electrical_ degrees. on my 4-pole servos that is 90 motor degrees and 45 ballscrew degrees, so 5/8 mm movement.
[22:06:27] <alex|lappy> SWPadnos, what if I just put a low rated fast blow fuse in series with the momnetary NO switch?
[22:06:40] <alex|lappy> I mean, the switch I have is rated for 15A
[22:07:14] <alex|lappy> so can I just put a 15A fast blow fuse in there? It should in theory never get tripped unless the transformer is under heavy load when the switch gets pushed in
[22:07:32] <alex|lappy> err
[22:07:49] <alex|lappy> yeah
[22:09:57] <andypugh> What's the machine? You are probably beeing too cautious if it is an Etch-a-Sketch and under-cautious if it is a 200Tonne 100m/min monster :-)
[22:10:08] <alex|lappy> andypugh, a knee mill
[22:10:12] <alex|lappy> bit bigger then a bridgeport
[22:10:52] <alex|lappy> .6KW motors
[22:10:54] <alex|lappy> http://www.flickr.com/photos/33383121@N04/4263848415/
[22:11:07] <alex|lappy> so it could be very dangerous if it gets out of hand
[22:11:08] <andypugh> Yeah, that could hurt. Is it only you, or are other people likely to get tangled up in it?
[22:11:37] <alex|lappy> just me, but I don't want to have to wory about it hurting me or someone else
[22:11:39] <alex|lappy> it is in my house
[22:11:46] <alex|lappy> I don't want my house to burn down
[22:11:58] <alex|lappy> especially because a switch got some coolant in it and got stuck close
[22:12:01] <andypugh> An E-stop won't help with that
[22:12:39] <andypugh> Well, it might, but I think you would have the time to walk to the wall outlet.
[22:12:40] <alex|lappy> andypugh, a defective estop shunting a lot of current through a switch not rated to handle it could
[22:13:12] <andypugh> One thing I would say... Are those safety handles?
[22:13:24] <alex|lappy> ?
[22:13:57] <alex|lappy> that is pre conversion, the hand wheels on it are going to get altered so the handles flip in and won't nut you by accident
[22:13:59] <alex|lappy> if that is what you are asking
[22:14:01] <andypugh> The BP at work has these neat handles that only engage when you twist the hand-grip (it works pretty intuitively).
[22:14:26] <andypugh> Yes, handwheels in the nuts was my worry.
[22:15:10] <alex|lappy> heh, yeah, it is something I thought of
[22:15:24] <andypugh> I say intuitively, I had to take one apart the first time I used the machine before I figured it out.
[22:16:26] <alex|lappy> but am I right in thinking a fuse there would keep the button from blowing up?
[22:17:13] <andypugh> You need to hold the handle tight enough that it twists in the wheel as you turn it, which operates a cam, which works a plunger in the spoke which engages a clutch in the hub. A bit over-complex but does mean the handles don't spin on rapids.
[22:17:35] <andypugh> Sorry, I have no opinion on your e-stop chain.
[22:18:16] <alex|lappy> the supermax we have at school has springloaded folding handles, you just pull them out against the sprint pressure and flip them in twards the center
[22:18:22] <alex|lappy> cheap, easy, simple
[22:18:33] <alex|lappy> i mean, there are still rotating disks out there, but uh, dont wear a tie
[22:19:19] <andypugh> The Bridgeport one are automatic and failsafe. And probably cost more than my EMC2 machine
[22:22:00] <andypugh> I don't need it, do I? http://cgi.ebay.co.uk/Colchester-Student-6-Gap-Bed-Lathe_W0QQitemZ370394376642
[22:23:15] <alex|lappy> 106 GBP?
[22:23:20] <alex|lappy> how can it be that inexpensive?
[22:23:46] <alex|lappy> andypugh, did you see the bright orange monarch 10EE i linked yesterday?
[22:23:57] <andypugh> I anticipate a flurry of bidding, but they go for £300 or so anyway. Lovely machines though.
[22:24:11] <SWPLinux> Jymmm: cradek suggested "EKG" :)
[22:24:34] <alex|lappy> SWPadnos, did you seem my fuse idea?
[22:24:52] <andypugh> This is rather nice too: http://cgi.ebay.co.uk/Elliot-mini-jig-borer-milling-machine-/190406911522
[22:25:05] <alex|lappy> andypugh, http://i531.photobucket.com/albums/dd358/gonekayatgmailcom/monarch07.jpg
[22:25:20] <alex|lappy> I have been thinking of doing my mill in the same color
[22:25:32] <cradek> MY EYES
[22:25:47] <andypugh> It's a very functional finish.
[22:25:54] <alex|lappy> andypugh, that is a very clean little machine
[22:26:03] <alex|lappy> cradek, :d
[22:26:05] <alex|lappy> D
[22:26:07] <alex|lappy> that is
[22:26:30] <alex|lappy> imagine what a machine 15-20% bigger then a bridgeport would look like with that ^_^
[22:31:47] <andypugh> That little Elliot is a lovely machine, I think I will let it go to someone who won't spoil it. (like I would plan to). Moe details at http://www.lathes.co.uk/elliott%20mini%20jig/index.html
[22:32:53] <alex|lappy> andypugh, that is an impressive belt set up for such a small machine
[22:33:30] <andypugh> It's a very serious little machine. I don't think the size was an attempt at cheapness.
[22:34:31] <alex|lappy> no, it seems to be built for exaclty one thing, and to do it damn well
[22:35:03] <andypugh> viz that Monarch 10EE: The ATK Company, manufacturer of the Space Shuttle solid propellant boosters (RSRM) was still, in 2009, still using a Monarch Lathe manufactured on December 7th, 1941, for machining flares and dissecting igniter cases.
[22:35:09] <Jymmm> SWPLinux: I'd suggest MCM
[22:36:10] <alex|lappy> andypugh, there are a bunch of lathes at the shop at school, probably 25 including a couple of 15hp American Turnmasters, a 20HP Clausing and a bunch of 2 or 3 year old Harrisons
[22:36:18] <Jymmm> SWPLinux: The Miss Cloe Module =)
[22:36:20] <alex|lappy> I use a 1940s LeBlond
[22:36:26] <alex|lappy> because god damn, it is nice
[22:36:54] <alex|lappy> holds a tenth no problem, a really amazing c5 collet system
[22:37:04] <andypugh> They didn't get any better at making lathes after about 1940. They just got cheaper.
[22:37:19] <alex|lappy> can easily take off .2" of aluminium in a pass even with my shitty indexible carbide tools
[22:37:27] <alex|lappy> I should take a picture of it next time I am in the shop
[22:37:50] <alex|lappy> the only thing about it is the max spindle speed is a bit low
[22:37:56] <alex|lappy> like 1800 RPM
[22:38:51] <andypugh> Though my 1938 Colchester is actually rather rubbish.
[22:38:52] <andypugh> Not desperately stiff for a a 6' between centres lathe, and a 6 tpi leadscrew on the cross slide with a 1.25" micrometer thimble is just bizarre.
[22:39:38] <alex|lappy> 6tpi? on an acme thread?
[22:39:43] <alex|lappy> that does seem a little strange
[22:41:24] <andypugh> So, 20 graduations on the thimble, each is 16 thou on diameter.. The only way to use the machine is with a DTI on the tool slide.
[22:43:26] <alex|lappy> why don't you cut a larger diameter thimble?
[22:43:54] <andypugh> It would foul the cross-slide. You can only actually see it on small work
[22:44:23] <andypugh> And it is in my dad's workshop, 200 miles away, so I am not that bothered :-)
[22:44:57] <frallzor> hmm how to stop wood from deforming after milling?
[22:45:05] <andypugh> PEG?
[22:45:47] <frallzor> talking to me? =)
[22:45:55] <andypugh> I was
[22:45:56] <alex|lappy> frallzor, kiln dry it
[22:46:03] <frallzor> whats PEG then? =)
[22:46:10] <frallzor> kil n dry+
[22:46:11] <frallzor> ?
[22:47:44] <alex|lappy> frallzor, a lot of the warping is caused by moisture
[22:48:20] <alex|lappy> honestly, just call a art supply store, see what they suggest
[22:48:24] <alex|lappy> for carving and the like
[22:48:34] <andypugh> Poly Ethylene Glycol
[22:48:38] <frallzor> one would expect that kitchen counters are dry when you buy them
[22:49:08] <frallzor> what does that do then andypugh?
[22:49:14] <andypugh> http://owic.oregonstate.edu/pubs/peg.pdf
[22:51:03] <andypugh> alex|lappy: Lathes.co.uk goes into raptures about that Monarch 10EE lathe, seems to be about the very best lathe you could ever get. http://www.lathes.co.uk/monarch/page2.html
[22:51:51] <alex|lappy> andypugh, they are not that expensive either
[22:52:02] <alex|lappy> there is one on ebay right now for $1500 that has a junked gearbox
[22:52:08] <alex|lappy> it would be the perfect CNC conversion
[22:52:49] <andypugh> I class that as expensive. My lathe was £750 brand new. And worth almost half that.
[22:53:24] <alex|lappy> ah, 2k
[22:53:27] <alex|lappy> http://cgi.ebay.com/MONARCH-10EE-TOOL-ROOM-LATHE-/290322903213?cmd=ViewItem&pt=BI_Lathes&hash=item439897f0ad#ht_720wt_907
[23:00:14] <alex|lappy> andypugh, I just really can't get over the art deco look
[23:01:14] <andypugh> They are 40 years after Art-deco, it must be art-stiffness
[23:05:04] <jthornton> * jthornton needs to get the cd drive and hard drive off of the box of cracker jacks and into the case :/
[23:06:57] <andypugh> Cases are generally over-rated, but not in at atmosphere with significant metal content ;-)
[23:12:04] <skunkworks_> http://www.cnczone.com/forums/showthread.php?p=790006#post790006
[23:13:59] <andypugh> He obviously didn't try the IRC link
[23:15:32] <alex|lappy> "there aren't any drivers for the mobo for linux"
[23:15:36] <alex|lappy> I don't think he gets it.
[23:16:07] <alex|lappy> and honestly, I hopped on this channel like 4 days ago, and I don't think I have ever had another linux software IRC channel be as helpful as you guys have been
[23:16:35] <skunkworks_> it is pretty great.
[23:16:53] <celeron55> "This is based on EMC2's latency test when compared to Mach3's equivalent"
[23:17:15] <celeron55> sounds like a fair comparison...
[23:18:16] <celeron55> well, depends on the quality of the drivers i guess
[23:18:36] <celeron55> (of the hardware that is in use)
[23:19:28] <alex|lappy> well, i think the first thing is, why is it EMC's fault that ubuntu does not support his mobo properly (even though I doubt that)
[23:28:29] <foxtrot> hi
[23:28:40] <andypugh> Hi.
[23:28:46] <foxtrot> So ive built a 3axis x,y,z machine with dremel and 3motors
[23:28:54] <foxtrot> ive got 3x pololu stepper drivers
[23:29:04] <andypugh> I should warn you that according to a CNC zone thread we are horribly unhelpful here :-)
[23:29:33] <foxtrot> im trying to figure out a wiring diagram for hooking them directly to the parallel port for emc
[23:29:37] <foxtrot> heh
[23:29:43] <foxtrot> well cnczone never helped me
[23:29:51] <andypugh> Err "direct" is a bad plan
[23:30:07] <andypugh> P-Port can handle 5mA...
[23:30:24] <foxtrot> i thought i just needed to give x,y,z step/direction control
[23:30:29] <andypugh> What's the spec of the motors for current/voltage?
[23:30:43] <foxtrot> 2A iirc
[23:30:50] <andypugh> Sorry, I missed the "driver" part.
[23:30:52] <foxtrot> ive got an ATX PS to run all that
[23:31:19] <foxtrot> yeah the Pololu A4983 Allegro based bipolar stepper motors
[23:31:29] <foxtrot> controller
[23:31:36] <andypugh> You can pretty much wire it any way you choose,
[23:31:47] <foxtrot> what do i haveto wire
[23:31:50] <andypugh> But what is the connector on the driver?
[23:32:18] <foxtrot> um it has 16 solder terminals each one is labeled
[23:32:47] <foxtrot> http://www.pololu.com/catalog/product/1201
[23:32:51] <foxtrot> its that one
[23:33:11] <andypugh> Yes, I was already there.
[23:33:16] <foxtrot> 3 of thoser, hooked to three of these http://www.alltronics.com/cgi-bin/item/28M035/search/Lin-Engineering-4218L-01-10-bipolar-stepper-motor
[23:34:40] <andypugh> it should be a simple Stepgen configuration.
[23:35:10] <foxtrot> stepgen?
[23:35:24] <andypugh> Sorry, I meant stepdonf
[23:35:38] <andypugh> No I didn't. Stepconf. That's the word
[23:36:27] <jthornton> oh my http://www.cnczone.com/forums/showthread.php?p=790107&posted=1#post790107
[23:36:58] <foxtrot> heh, still dont know what you mean, stepconf?
[23:37:30] <jthornton> the StepConfig Wizzard on the CNC menu
[23:37:32] <andypugh> Work out which P-Port pins are inputs and which are outputs, (either from wikipedia, or by starting stepconf and seeing what it offers) then decide which pins you _want_ to use as enable, step and direction for each driver, and give those settings to stepconf.
[23:37:37] <jthornton> oh only one z
[23:38:23] <foxtrot> ok
[23:38:31] <foxtrot> i dont have any linuxcnc stuff installed
[23:38:39] <foxtrot> should i just download the livecd to test this?
[23:38:46] <andypugh> Yes.
[23:38:55] <jthornton> yes
[23:39:10] <foxtrot> im running netbsd, so im trying to avoid any compatibility issues while im testing this beast for the first time
[23:39:10] <jthornton> is there an echo in here andypugh?
[23:39:29] <andypugh> If you have the drivers and a P-Port plug you are pretty much ready to install and go,
[23:40:20] <foxtrot> i have both those things
[23:40:24] <andypugh> jthornton: PLCguy really knows how to pour oil onto troubled forest fires.
[23:40:38] <jthornton> it seems so
[23:40:44] <foxtrot> but i dont have a zero position encoder for x,y,z tho
[23:40:48] <foxtrot> how important is that
[23:41:05] <andypugh> You mean a home switch?
[23:41:13] <foxtrot> yea
[23:41:19] <jthornton> foxtrot, steppers?
[23:41:33] <foxtrot> my machine is simply mechanics + 3 drivers + 3 motors + spindle
[23:41:46] <andypugh> You can manage without, add one later. There is certainly no point bothering until the motors are moving a machine.
[23:42:15] <jthornton> you can move to a match mark and call that home without home switches
[23:43:59] <foxtrot> 'match mark'?
[23:44:52] <jthornton> a pencil mark or something to kinda get you close so your soft limits work properly
[23:45:44] <foxtrot> whats the definition of limits
[23:45:45] <jthornton> say you move everything to the center of the travel and click on the home button then EMC thinks that is 0,0,0
[23:45:52] <jthornton> you define them
[23:46:21] <jthornton> based on the physical properties of your machine
[23:46:27] <foxtrot> ah, cool
[23:46:31] <andypugh> That too is handled by the wizard, you say what the min and max travel of the axis is, and where the home position is.
[23:46:53] <jthornton> you set them up so and EMC won't allow you to jog into the hard stop
[23:47:24] <jthornton> * jthornton wanders off to fire up the barbie
[23:47:31] <Endeavour> Hello everyone.
[23:47:38] <Endeavour> I'm getting latencies around 80000ns
[23:48:21] <Endeavour> Checked the system info and all four cores of the i7 are operational
[23:48:24] <Endeavour> Memory is fine.
[23:48:43] <Endeavour> Usage of all resources seems fairly low.
[23:49:12] <andypugh> That's shocking latency for about the fastest CPU on the planet
[23:49:32] <andypugh> Which kernel?
[23:50:20] <andypugh> Ah, you actually already answered the question, to an extent.
[23:50:44] <andypugh> You are running an SMP kernel, patched for RTAI?
[23:54:20] <morfic> is it all the "let's power down the cores we don't need" magic at work?
[23:54:55] <andypugh> I was waiting for him to show signs of life before posting the isolocpus link
[23:55:38] <morfic> i need to finish building stuff i need so i can build emc deps and rtai on my new desktop, kind of curious how much havoc all the power savings options i turned on in bios cause, to get an idea what to buy for work, just not warming up to those atoms
[23:55:54] <andypugh> morfic: Want me to make you a motor driver for your lathe? http://www.youtube.com/watch?v=oyeJfNg3NfQ
[23:56:51] <andypugh> I don't see why not, cheap, good latency, 20x the CPU power of the best machine you could buy when the lathe was made....
[23:57:39] <morfic> andypugh: that's the thought that remains, more cpu power than any of our machines
[23:58:27] <morfic> andypugh: running plain old AC motors and stopping them at any angle with the arduino? or "just" running servos?
[23:58:31] <andypugh> And EMC is _really_ not very demanding
[23:58:58] <andypugh> That video is a brushless, Resolver, Servo.
[23:59:23] <morfic> yeah, that's why i was asking, to be our hero, it would have to make some 40HP motor magic ;)
[23:59:57] <andypugh> And I have not done the induction motor test yet, that would just be HAL config and a willingness to try something that common sense says is foolish.