#emc | Logs for 2009-04-10

[00:13:44] <tomp> BigJohnt 'sending pulses to encoder' ..'A or B' are yousending data to the place where encoder feedback goes? thats tricky trickery. if you 'simulate' motion by faking signals. you'd need to keep track of quadrature to send the correct A B Z.
[00:28:17] <wildrice> Been reading and trying to understand the HAL API. But I do not understand what the difference between a HAL PIN and a HAL parameter. Normaly, I would think of a parameter as like a function parameter in C or other such language. But with the HAL API, I thought that that was what the HAL PINS were. So what is the difference?
[00:34:30] <tomp> the difference is smaller as time goes on...meaning many parameters have become pins. originally it had the 'flavor' of a write only parameter. but it is evolving. short answer... not much difference, you need to evaluate per case.
[00:45:23] <fenn> parameters are more like settings than input stream
[00:45:46] <fenn> like a knob versus an input jack
[00:46:06] <fenn> oh there are output parameters too, i guess that's like a display
[00:46:15] <wildrice> There is no net associated with them? What hooks them with the outside world?
[00:46:45] <fenn> noting, you setp and forgetp :P
[00:47:52] <wildrice> The python code I am trying to understand does this:
[00:47:54] <wildrice> for port in range(8):
[00:47:54] <wildrice> c.newpin ("digital-out-%02d" % pinmap[port], hal.HAL_BIT, hal.HAL_IN)
[00:47:54] <wildrice> c.newparam("digital-out-%02d-invert" % pinmap[port], hal.HAL_BIT, hal.HAL_RW)
[00:48:20] <wildrice> so In the HAL file, I do a SETP to set that value?
[01:32:43] <dareposte> hi all
[01:54:27] <dareposte> ahhh stove top parkerizing... very fun
[02:01:39] <Skullworks1> http://www.liveleak.com/view?i=8d3_1239175602
[02:04:16] <dareposte> wow that noise is really really shrill
[02:04:42] <dareposte> cool machine though
[02:04:54] <dareposte> i bet the kinematics were a b***
[02:05:07] <Skullworks1> he uses one of those tunnel prop RC airplane motoer for the router.
[02:05:55] <dareposte> i wonder if the camera just picks it up funny, i don't remember rc motors sounding so whiney
[02:05:56] <Skullworks1> it walks away at the end...
[02:43:48] <dareposte> http://www.youtube.com/watch?v=IfAdNu6nV88
[02:43:55] <dareposte> there's my lovely video
[02:44:06] <dareposte> not nearly as cool as a hexapod machining robot
[02:57:07] <dareposte> cradek: I slowed it down from yesterday and took a bigger cut, and the surface finish was much better. thanks for the advice
[03:05:55] <Skullworks1> dareposte: You have a speed control on your 9x20 or just off/on?
[03:16:25] <dareposte> it has speed control
[03:16:54] <dareposte> constant surface speed
[03:19:07] <dareposte> Skullworks1: why do u ask, seem too slow?
[03:19:48] <fenn> dareposte: what's the solution made of that you're using?
[03:20:07] <dareposte> phosphoric acid
[03:20:10] <dareposte> zinc something or other
[03:20:19] <dareposte> a bit of rusty iron
[03:20:35] <fenn> i like to use paste wax for rust prevention (furniture polish, comes in a can)
[03:20:49] <fenn> when you buff it off, nice and shiny, you can't even tell it's wax
[03:21:12] <dareposte> really
[03:21:26] <dareposte> sounds cool, whats the can look like
[03:21:31] <fenn> bright yellow
[03:21:39] <dareposte> i think if you put cosmoline on the parkerized parts they get to be pretty shiny
[03:21:44] <dareposte> but i don't have any cosmoline
[03:21:48] <fenn> http://images.rockler.com/rockler/images/57547-01-200.jpg
[03:22:06] <dareposte> oh yeah that's a great idea
[03:22:32] <dareposte> i think someone told me to put that stuff on black-oxide coated parts, and it would make them look really nice too
[03:22:41] <fenn> one can will last forever
[03:22:51] <fenn> i use it on all my tools
[03:23:04] <dareposte> that's a good idea, i'm going to get some this weekend
[03:23:13] <fenn> good for machine parts because it doesn't add thickness the way paint does
[03:23:27] <dareposte> it probably doesn't wear well does it?
[03:23:46] <fenn> it rubs off after a while, and according to cradek it won't keep rust from forming underneath a vise with flood coolant
[03:23:57] <dareposte> i'm going to try some black oxide coating this weekend, but i have to do that outside because it apparently releases huge clouds of toxic ammonia gas
[03:24:25] <cradek> I tried it a couple times, but liberally-applied WD40 worked much better for rust prevention
[03:24:41] <fenn> well, wd does stand for "water displacement" after all
[03:24:53] <dareposte> yeah but it doesn't last in my experience
[03:25:06] <dareposte> unless you respray it frequently
[03:25:07] <cradek> nope, you'd have to do it whenever mounting the vise
[03:25:37] <cradek> I use it if I intend to leave the vise on for more than a couple days. normally I just move it after I'm done
[03:26:00] <dareposte> what coolant do you use
[03:26:11] <dareposte> i haven't had mine rust anything yet, but i'm just misting
[03:26:17] <cradek> 'kutzwell' or some silly name like that
[03:26:51] <dareposte> i saw one named "rustlick" or something like that
[03:27:06] <dareposte> sounds like it should not make rust form under the vise
[03:27:16] <cradek> hm the internet doesn't know about "kutzwell" so that must not be the name
[03:27:44] <cradek> dareposte: they all are supposedly rust preventive - also, they all cause rust
[03:28:09] <cradek> coolant is 95% water, not much you can do about it
[03:28:35] <dareposte> i think i am using Kool Mist #77
[03:28:53] <cradek> I am tempted to try mist, now that I have a big air supply
[03:29:02] <dareposte> it doesn't actually take that much air
[03:29:15] <cradek> does it fog up the shop?
[03:29:25] <dareposte> i run mine off a 1hp 2.9 cfm compressor, and it keeps up just fine
[03:29:33] <dareposte> my lathe is actually in my living room
[03:29:44] <dareposte> but no it doesn't fog up the house
[03:29:44] <cradek> oh wow, that isn't much air, I'm surprised
[03:29:52] <cradek> huh
[03:29:57] <dareposte> if you turn it on full blast it will fog some
[03:30:11] <dareposte> but just tweak it down to keep the tool cool without blowing coolant all over the place, and it works great
[03:30:50] <fenn> i hear ethanol is good for aluminum
[03:31:12] <fenn> (and trichloroethylene is good for steel, but you can't have any)
[03:31:55] <dareposte> ethanol is too valuable to waste as coolant
[03:31:56] <cradek> and an atmosphere of pure hydrogen gas
[03:32:16] <cradek> (who comes up with this stuff?)
[03:33:11] <Skullworks1> The Darwin awards nominating committee?
[03:33:45] <cradek> yeah, sounds like
[03:34:06] <fenn> * fenn mutters something about armies of robots from outer space
[03:34:09] <cradek> goodnight
[03:34:37] <Skullworks1> Rustlick WS5050 is good - but the mixed coolant will grow anarobic bacteria if you don't keep a bubbler in it long term
[03:34:56] <dareposte> can't you add some bleach or something?
[03:35:46] <Skullworks1> nope - different coolants use different bases
[03:36:11] <dareposte> pj
[03:36:12] <dareposte> oh
[03:36:52] <Skullworks1> some use sulfur chloride - add more Cl and the mix is wrong
[03:37:26] <Skullworks1> but Ozone will keep the bacteria from growing
[03:37:47] <dareposte> how do you know what base it has, do they publish that?
[03:38:01] <Skullworks1> MSDS
[03:39:14] <dareposte> * dareposte looks for msds for kool mist 77
[03:40:01] <dareposte> "GREEN LIQUID/MILD ODOR"
[03:40:14] <Skullworks1> actually if you have good air supply there is a near dry lube that works great.
[03:40:52] <dareposte> i have a crappy air supply
[03:41:18] <dareposte> it fits my crappy lathe
[03:41:21] <Skullworks1> They use it on the huge carbide chop saws down in Guaymas
[03:41:34] <dareposte> is that the air chiller thing?
[03:42:09] <Skullworks1> chips come out light fluffy and dry - virtually no oil residue
[03:42:34] <Skullworks1> no not a Vorsh tube
[03:43:03] <Skullworks1> micro drop lube in a small fine air stream
[03:43:11] <dareposte> sounds pretty cool
[03:43:20] <dareposte> this MSDS doesn't say anything except to not add nitriles
[03:43:56] <Skullworks1> hmm - Ill see if I can find the sight - I think they have so good online demo films
[03:44:36] <Skullworks1> see - add the wrong thing and the coolant goes south
[03:45:24] <Skullworks1> bwt I used up a 5gal of coolmist 77 and decided it was not worth the trouble
[03:45:39] <Skullworks1> 5gal concentrate
[03:45:41] <dareposte> it's my first coolant ever
[03:45:45] <dareposte> whats bad about it?
[03:46:24] <Skullworks1> no real EP properties for cutting steels
[03:46:30] <dareposte> so far it's worked okay for me, i just got a 1gal concentrate but have used up about 2 gals of mix now
[03:46:56] <Skullworks1> and it leaves a sticky film that builds up everwhere in the shop
[03:47:17] <dareposte> were you flooding or misting
[03:47:46] <Skullworks1> misting and flood at first - then only mist
[03:48:04] <dareposte> what do you use now?
[03:48:47] <Skullworks1> mainly the Rustlick WS5050 - it works for all metals
[03:48:49] <dareposte> i was trying to pick one out, but they all seemed pretty much the same in the catalog, and the 77 claimed to be useful for misting so that's what i wound up with
[03:49:31] <dareposte> cool i will make a note for when i use this up
[03:49:42] <dareposte> pricy stuff though isn't it
[03:49:57] <dareposte> i need to find some coolant for my band saw now too
[03:50:01] <Skullworks1> yeah some other coolants will cause you to evac if you mist them - they don't settle out of the air fast enough
[03:50:52] <dareposte> what is the practical difference when you use rustlick vs another... finish? tool wear?
[03:51:12] <Skullworks1> well we used it because we run stainless one day, aluminum the next and it works for all
[03:51:42] <Skullworks1> tool life is much better
[03:52:45] <fenn> i hear mist coolant will kill you..
[03:52:57] <fenn> bwahahaha
[03:53:10] <Skullworks1> wear is minimal on coated carbide until the coating breaks down - then the tool will wear quickly
[03:53:26] <fenn> but seriously, i'd rather breathe ethanol than some random industrial chemicals
[03:53:38] <dareposte> i would also rather breathe ethanol :)
[03:53:49] <dareposte> lungs are very good at absorbing it into the bloodstream
[03:54:07] <fenn> do not operate heavy machinery after operating heavy machinery!
[03:54:11] <Skullworks1> for AL kerosene is great!
[03:54:58] <dareposte> heh
[03:55:04] <Skullworks1> I also soak all my stones in it
[03:55:07] <dareposte> isn't wd-40 mostly kerosene
[03:56:09] <Skullworks1> it is somewhat similar - but much different
[03:56:51] <Skullworks1> and price per gallon is crazy different
[03:58:12] <dareposte> a machinist at work was telling me cinnamon tap magic works for aluminum
[03:58:19] <dareposte> i didn't know what he was talking about though
[03:58:34] <dareposte> i gave up turning aluminum as soon as i found out i could do steel
[03:58:58] <Skullworks1> It has a strong cinimon type oder but its a thick clear fluid
[03:59:50] <Skullworks1> I saw your vid - use the right type of insert and you will love Al
[04:00:08] <dareposte> which type would that be?
[04:00:21] <dareposte> i've tried a few different ones and nothing works really well yet
[04:00:26] <fenn> i think tap magic is just biodiesel
[04:00:28] <dareposte> ground HSS is what i settled on for al
[04:00:50] <Skullworks1> those TPG type holders are the bottom of the ladder just above hand ground HSS
[04:01:04] <SWPadnos> the old Tap Magic Aluminum was stinky and carcinogenic
[04:01:27] <SWPadnos> the new Tap Magic Aluminum is nice cinnamon-smelling stuff, and it makes aluminum turn into mirrors when you machine with it
[04:01:39] <SWPadnos> s/with//
[04:01:57] <Skullworks1> 1-1-1 tri-cloroethane or some such
[04:02:09] <Skullworks1> nasty stuff
[04:02:16] <dareposte> Skullworks1: I have a CCMT insert holder too, how do those stack up
[04:02:30] <Skullworks1> Very good!
[04:02:53] <SWPadnos> cobalt seems to work nicely on Aluminum
[04:03:14] <Skullworks1> A Ross Arno insert in there will seem like your lathe has 2x the available HP
[04:03:50] <dareposte> so 2hp instead of 1 lol
[04:03:57] <dareposte> that sounds nice
[04:04:36] <dareposte> i haven't gotten the ccmt inserts out yet because i like the tpg's pretty well so far
[04:05:24] <dareposte> huge improvement just over having ground hss bits with crappy geometry
[04:05:29] <Skullworks1> well the high polish inserts will work wonders in AL
[04:05:41] <dareposte> are those the ones with big positive rake?
[04:06:01] <Skullworks1> yeah - let me shoot a pic
[04:06:13] <dareposte> cool
[04:06:27] <dareposte> look like a little crown?
[04:06:32] <dareposte> i think i know the ones you mean
[04:07:22] <dareposte> i only have tpg-222 inserts that i've used, and i think the radius is too large for the small cut depths i can take
[04:08:15] <dareposte> it's a 0.8mm radius on the tip instead of the 221's 0.4mm, and i think that it might make things harder
[04:08:41] <dareposte> somehow the finish gets better when i speed up feed beyond what seems reasonable
[04:09:15] <dareposte> 0.4mm / rev gives a very decent finish, but 0.1 or 0.2mm/rev is lacking
[04:10:06] <dareposte> i am fantasizing that if i got some 0.4mm radius inserts then i could take 0.2mm/rev and have good finish, but i'm not sure what impact the tool radius actually has on the feed rate and finish
[04:10:06] <fenn> your lathe makes a sort of rattling sound
[04:10:19] <fenn> i wonder if your gibs are tight and bedded properly
[04:10:31] <fenn> or maybe the spindle bearings are not preloaded
[04:10:34] <dareposte> fenn: they are tight, but the gib itself doesn't fit
[04:10:52] <fenn> i've seen where people rub them on a sharpening stone to get it flat
[04:11:32] <dareposte> fenn: where do you hear the rattling sound?
[04:11:38] <fenn> on the video you just linked to
[04:11:44] <dareposte> yeah but which part, the whole thing?
[04:11:50] <fenn> especially when doing the parting off cut
[04:12:48] <dareposte> oh that
[04:12:58] <dareposte> yeah thats a dull parting tool
[04:13:41] <dareposte> i think the spindle bearings are loosening up again too
[04:13:52] <dareposte> good call i'll check them tomorrow now that you mentioned it
[04:14:14] <dareposte> i crashed the heck out of my cross slide today too, so that might have osmething to do with it
[04:14:30] <dareposte> MDI mode... typed "g0 z15", but it came out as "g0 z1"
[04:14:40] <fenn> z'eaux
[04:14:47] <dareposte> which happened to be about 10mm behind the chuck jaws
[04:15:11] <fenn> there ought to be some way to keep it from doing that
[04:15:32] <fenn> CRASH_LATHE = 0
[04:16:04] <dareposte> that would be nice
[04:16:20] <dareposte> lucky for me i have steppers on there
[04:16:40] <dareposte> and when it hit the chuck was spinning fast enough that the tool didn't get under the jaws
[04:16:48] <dareposte> just bounced off and machined a bit of the jaws off
[04:17:20] <dareposte> machined... i meant gouged
[04:23:34] <dareposte> it will be big trouble when i get a full sized turning center, those probably cost a lot more to wreck
[04:23:58] <Skullworks1> MDI and G00 is evil
[04:25:07] <dareposte> "real" cnc's have jog wheels right
[04:25:27] <dareposte> or some way besides mdi to get it to move where you want it to somewhat accurately?
[04:26:22] <Skullworks1> http://imagebin.ca/view/8yYr881G.html
[04:26:41] <Skullworks1> but you have a kyb
[04:28:38] <dareposte> yeah
[04:28:48] <dareposte> it doesn't jog so great though
[04:29:35] <dareposte> you can get there but its somewhat annoying
[04:36:02] <fenn> jog wheel is what you want..
[04:36:27] <fenn> i sort of wonder what people use these joysticks for
[04:38:47] <dareposte> jog wheel sounds good
[04:39:02] <dareposte> if i could hook up a ps2 controler it would be fantastic
[04:42:34] <fenn> well, you can
[04:43:10] <fenn> http://www.dealextreme.com/details.dx/sku.18022
[04:43:30] <fenn> not 100% guaranteed to work, but i bet linux will recognize it as a usb HID device
[04:43:58] <fenn> john thornton got a logitech usb gamepad to work, looks exactly like a ps2 controller
[04:44:12] <dareposte> cool
[04:44:20] <dareposte> and it jogs all around?
[04:44:46] <dareposte> ideally there would be a normal jog, and a button to hold to either rapid jog or slow jog
[04:46:04] <fenn> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Simple_Remote_Pendant
[04:49:42] <dareposte> that's pretty neat
[04:50:08] <dareposte> i'm lost on the hal stuff though, i need to sit down and dig into it some day soon
[04:52:06] <Skullworks1> half the battle is won just by knowing its allready be done.
[04:52:21] <dareposte> yeah
[05:00:35] <dareposte> i'm about to scrap my current lathe and get a bigger one though
[05:00:42] <dareposte> no sense in putting lipstick on a pig
[05:03:14] <dareposte> i do wonder if an old screw machine might be a good carcass for cnc conversion
[05:04:33] <dareposte> one of those old cam driven things
[05:15:42] <Skullworks1> prolly not
[05:16:31] <dareposte> i've never seen one close enough to have a good look at the internals
[05:17:16] <Skullworks1> unless you need real big - you could get a Hardinge CHNC cheap and the conversions have been proven
[05:17:33] <Skullworks1> 8 tool turret
[05:17:42] <dareposte> yeah
[05:17:46] <dareposte> i looked at those
[05:18:02] <dareposte> i'm kind of lusting over a used mori-seki at the moment
[05:18:47] <Skullworks1> dream big - thats good
[05:18:49] <dareposte> but it will pass i'm sure
[05:19:44] <dareposte> there's a machinery reseller near me that had a pretty nice one, late 90's or so for about $7k
[05:20:51] <dareposte> hard to beat that, its always the problem of how to get it in my garage that kills my plans
[05:22:11] <dareposte> that combined with my frequent crashing of my stepper lathe
[05:22:27] <dareposte> and knowing what a 2kw servo would do to a 15hp spindle motor
[05:41:14] <Skullworks1> my Uncles Mori was scary when we first got it - now it seems typical
[05:42:57] <Skullworks1> a mill - 1280 IPM rapids, 10K 30hp spindle, 9.2hp servos, .9 sec tool change
[05:43:10] <Skullworks1> you get used to it
[06:48:40] <UncleG> Anyone know how to achieve a reflective finish on aluminum?
[06:48:55] <archivist> polish
[06:49:18] <UncleG> I would like to do it on the machine with minimal time involved
[06:50:09] <archivist> turned finish?
[06:50:21] <UncleG> if possible
[06:50:40] <UncleG> Could a specialized carbide cut such a finish?
[06:51:26] <archivist> there are some ground extra sharp carbides inserts available
[06:51:33] <UncleG> http://www.displays2go.com/product.asp?ID=9682
[06:51:51] <UncleG> thats the best finish required
[06:52:09] <archivist> heh no where near mirror
[06:52:59] <UncleG> well
[06:53:14] <UncleG> the particular parts I have you can see yourself in them rather well but not mirror
[06:53:20] <archivist> http://www.collection.archivist.info/archive/DJCPD/PD/2006_09_07_pendulum/ for mirror
[06:53:38] <UncleG> I would prefer them to come off of the machine finished
[06:54:04] <UncleG> beautiful
[06:54:11] <UncleG> absolutely perfect
[06:54:23] <UncleG> material type and tooling?
[06:54:32] <archivist> cast iron
[06:54:38] <UncleG> lol
[06:54:47] <UncleG> looks like a tin coated carbid
[06:54:52] <UncleG> carbide
[06:54:59] <archivist> carbide tool, and wet and dry paper to finish
[06:55:25] <SkinnYPuP> I like wenol for polishing myself
[06:55:56] <UncleG> not so much as a line in the finish
[06:56:02] <UncleG> must be a nice lathe
[06:57:02] <archivist> Schaublin circa 1956
[06:58:25] <UncleG> is it yours?
[06:58:51] <archivist> currently the bosses :( but Im working on it :)
[06:59:13] <archivist> very worn in places
[06:59:43] <UncleG> :)
[07:00:06] <UncleG> can one use a spray weld and resurface the ways if needed?
[07:00:20] <SkinnYPuP> Know that feelin, first 18" away from the chuck on my colchester is like oiled glass... lil further down it tightens up
[07:00:53] <archivist> must be 10-15 thou of wear
[07:00:59] <UncleG> yuck:(
[07:01:01] <SkinnYPuP> ooooo
[07:01:30] <archivist> broken screw cutting
[07:01:41] <pjm__> good morning
[07:01:42] <UncleG> I have had armature shafts spray welded and reground, it seems one could do the same for a way.
[07:01:45] <UncleG> gm
[07:02:57] <archivist> on this one just grinding straight could make if useful
[07:17:41] <pjm__> archivist i found my 'noise' problem!
[07:17:58] <pjm__> now and then the machine would stop wiht a joint2 limit error
[07:18:42] <alex_joni> faulty switch line?
[07:18:47] <pjm__> turns out that since i changed the limit switch output to be always +5 when its not on a limit, the vibration causes the contacts in the switch to vibrate, thuse signalling that a limit had been reached!
[07:19:25] <pjm__> so this morning i'm gonna replace the switch with one of these omron inductive proximity sensors
[07:25:13] <archivist> how good are the omron switches
[07:30:17] <alex_joni> inductive almost always beats regular switches
[07:30:36] <alex_joni> trip point might not be that exact, but for limits they are best
[07:30:58] <alex_joni> (except when you start milling magnetic stuff that sticks to the sensor :D)
[07:31:00] <archivist> trip point is what I was looking for
[07:31:25] <alex_joni> archivist: the apps where I used them didn't need that exact tripping point
[07:31:29] <alex_joni> couple mm was ok
[07:32:01] <JymmmEMC> Gawd I love my netbook =)
[07:32:16] <JymmmEMC> I'd marry it if it had 4G =)
[07:32:29] <alex_joni> JymmmEMC: what brand/type?
[07:32:37] <archivist> alex_joni, one day I want homing to microns
[07:32:52] <JymmmEMC> Lenovo IdeaPad S10
[07:32:55] <alex_joni> then you gotta start collecting them :D
[07:33:09] <alex_joni> the microns I mean :p
[07:33:52] <archivist> I can do about 4 microns on one axis here but not sure of the stability
[07:34:15] <pjm__> i like the idea of homing on a switch + encoder pulse
[07:35:05] <JymmmEMC> alex_joni: Mine's BLK http://blog.laptopmag.com/up-close-with-lenovos-ideapad-s10-one-sweet-10-incher
[07:35:05] <archivist> yup that could be very good
[07:35:25] <JymmmEMC> alex_joni: It does 1920x1400 external video
[07:35:32] <alex_joni> cool
[07:35:34] <alex_joni> battery life?
[07:35:42] <archivist> 2 mins
[07:35:50] <JymmmEMC> alex_joni: Unlike the Acer, it has an Express34 slot
[07:36:12] <JymmmEMC> alex_joni: about 2.5 hrs on the 3cell battery, around 7 with the 6 cell (I've heard)
[07:36:29] <alex_joni> there is one thing that would prevent me from buying it
[07:36:35] <alex_joni> (from a quick glance at it)
[07:36:43] <JymmmEMC> ?
[07:36:43] <alex_joni> the leftmost "Fn" key
[07:37:09] <JymmmEMC> Nah, it's the right shift that's a lil funky, but you get used to it.
[07:37:15] <alex_joni> I hate these designs where they put the Fn before Ctrl
[07:37:18] <alex_joni> I use Ctrl a lot..
[07:37:37] <JymmmEMC> Also the trackpad has a very cool scroll feature in where you make lil swirles with your finger
[07:37:46] <JymmmEMC> I use ctrl just fine on it
[07:38:27] <JymmmEMC> Believe me, I have big hands and am a heavy kybd user, it's well layed out for what it is
[07:38:33] <alex_joni> JymmmEMC: I'm sure you get used to it, but if you switch frequently from a regular keyb to it, it gets annoying
[07:38:45] <JymmmEMC> No not at all, very easy
[07:38:52] <alex_joni> I have a 12" which I love, that suffers from the same "symptom"
[07:39:08] <alex_joni> I get on just fine with it, but switching to another laptop is painful
[07:39:08] <JymmmEMC> Believe me, I'm very picky when it comes to this stuff.
[07:39:30] <JymmmEMC> I like it's kybd better thant he MBP
[07:41:54] <alex_joni> I have an Amilo Pro v3205 from FSC
[07:43:07] <JymmmEMC> I like the size of the IdeaPad
[07:43:12] <alex_joni> it's a bit bigger I guess (1.8kg)
[07:43:18] <alex_joni> but with a dvd-rw
[07:43:26] <alex_joni> and a Core Duo inside
[07:43:32] <alex_joni> about 4-5h autonomy
[07:44:00] <JymmmEMC> I have 4gb usb sticks for storage, and if I need optical drive there is the usb powered one
[07:45:55] <tom1> archivist: My-com switch, 1um repeatability, usedd em on agies, they really are 1um ( was 1/2 um steppers ) http://www.baumerelectric.com/1216+M560cdebd35c.html?&L=1 and baumer is easy to get
[07:47:27] <archivist> tom1, price ?
[07:57:03] <tom1> dunno, check uk, supplier , top of lin, top quality, ruby tipp to prevent wear error, i bet 60-120$ US, but off the shelf perfection. ( or used agie dealer.. look in side uv head of wire ems )
[07:57:44] <tom1> edms
[07:58:57] <tom1> they were for oen loop systes, so no scale involved. home repeated good enuf for carbide trrim passes between power downs
[07:59:20] <tom1> open loop systems (fskn kbd)
[07:59:41] <archivist> I need to get the slides a lot better as well
[08:00:07] <archivist> perhaps on version 2
[08:17:29] <archivist> oo retrofit fodder but too heavy and abroad ebay 110372003948
[08:33:32] <JymmmEMC> Man this tobacco has such a sweet smell to it
[09:09:36] <Valen> "alex_joni: I have a 12" which I love"
[09:09:45] <Valen> can i quote that our of context around the place?
[09:42:26] <tomp> dont inhale (B Clinton)
[10:11:35] <Valen> http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=390040865682QQssPageNameZMERC_VI_RCRX_Pr4_PcY_BIN_Stores_IT&refitem=110372272408&itemcount=4&refwidgetloc=active_view_item&usedrule1=CrossSell_LogicX&refwidgettype=cross_promot_widget&_trksid=p284.m184&_trkparms=algo%3DCRX%26its%3DS%252BI%252BSS%26itu%3DISS%252BUCI%252BSI%26otn%3D4
[10:38:34] <pjmcnc> great! proximity sensor for Z limit + home seems to have fixed the problem totall!
[10:38:40] <pjmcnc> totally
[11:18:30] <archivist> wee getting brave...cutting metal with an opensource database project building and testing in the background
[11:18:45] <archivist> http://askmonty.org/buildbot/waterfall
[13:08:50] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[16:09:09] <issy> hi all
[16:12:05] <issy> haw to deal with the zero return switches when i have 2 motors on single azis?
[16:15:25] <cradek> this comes up frequently and was discussed on emc-users in the last few days - did you look there?
[16:15:41] <issy> no
[16:16:29] <issy> the problem is that according to the descriptions , i have to connect all switches together
[16:16:53] <issy> and this is a problem when comes to the gaintry axis
[16:17:19] <cradek> http://news.gmane.org/gmane.linux.distributions.emc.user
[16:17:27] <cradek> go to page 2 and you will see the latest discussion
[16:19:23] <SWPadnos> but be careful, I don't think the logic is correct in Len Shelton's recent post
[16:19:41] <issy> it is not
[16:19:53] <SWPadnos> I've got a response in an editor window, and boy does it get complex to do it his way :)
[16:20:28] <SWPadnos> unless you use the -in and -in-not pins for the physical inputs (instead of inverters)
[16:20:34] <issy> the real way is to use 2 sensors and when go out of conntact to cleat at the first Z impuls of the encoder
[16:21:01] <SWPadnos> well he's not using encoders, so that's a different scenario
[16:21:28] <issy> yes , i am using brushless servos with encoders
[17:17:27] <Valen> dang, i wonder what servo serup issy is using
[17:20:30] <Master15> hi... so so alex_joni ist auch da ;-)
[17:23:16] <Master15> did somebody has a cnc-step machine?
[17:27:44] <alex_joni> Master15: 'tuerlich, 'tuerlich
[17:27:53] <alex_joni> Master15: btw, you can try asking at cncecke.de
[17:28:02] <alex_joni> there's an emc2 forum there, and lots of german users
[17:28:08] <alex_joni> maybe there are people with a cnc-step
[17:28:49] <Master15> I have already search in the forum but i can't found an cnc-step user
[18:29:46] <archivist> pjm__, your counting the spindle on the parport? whats your max count rate
[18:49:40] <alex_joni> heh http://failblog.org/2009/04/10/headline-fail-4/
[18:53:27] <SkinnYPuP> lol
[18:54:13] <pjm__> archivist the base period gives me about 35KHz
[18:54:30] <pjm__> in practice it stops reading speed accuratly just about 1500rpm
[18:55:59] <archivist> showing fohn stevensone here
[18:58:36] <skunkworks_> what is your encoder again?
[19:04:07] <fenn> * fenn glares at fenn__
[19:21:12] <pjm__> skunkworks , my encoder is a 360ppr quadrature type
[19:21:32] <pjm__> this is why I need to get a mesa card, so it can run at full chat
[19:59:57] <Optic> moo
[20:18:00] <jmkasunich> finaly found my round tuit
[20:18:08] <jmkasunich> * jmkasunich admires nice shiny 2-4-6 blocks
[20:19:12] <cradek> cool
[20:19:43] <jmkasunich> graphic example of power laws in action
[20:20:05] <jmkasunich> if you are used to 1-2-3 blocks, you think "twice the size, no biggie"
[20:20:19] <jmkasunich> then you pick them up, and realise 2x linear = 8x volume/weight
[20:20:34] <cradek> yep they're quite surprisingly large and heavy
[20:20:55] <cradek> also why they're $60 instead of $5
[20:21:04] <jmkasunich> $45 ;-)
[20:21:17] <cradek> that's about the lowest I've seen - are they square? :-)
[20:21:24] <jmkasunich> haven't checked
[20:22:54] <jmkasunich> seem to be
[20:23:20] <jmkasunich> don't have any particularly precise squareness checking technology here
[20:24:48] <cradek> I screw a parallel to a 123 block so it can pivot, set the 123 on the surface plate, place it against the test part, reverse it, look for light
[20:25:16] <cradek> so the error is double - should be able to see a very small deviation
[20:25:32] <cradek> (sorry if this is obvious)
[20:26:52] <jmkasunich> I realised right after I typed that - I do have tech
[20:27:12] <jmkasunich> both blocks sitting on the 4x2" end on the surface place, 6x2" faces touching
[20:27:14] <jmkasunich> feeler gages
[20:27:30] <jmkasunich> seems to be off by 0.001 or so over the full 6"
[20:28:33] <cradek> that sounds plenty good enough to build a tower with
[20:31:12] <jmkasunich> most of the error seems to be on one block
[20:31:27] <jmkasunich> turning that block 180 changes it, turning the other doesn't
[20:32:07] <jmkasunich> someday when I'm bored I'll test them more carefully
[20:32:19] <jmkasunich> there are many permutations
[20:32:36] <jmkasunich> 12 edges per block, any number of which can deviate from 90
[21:22:28] <jmkasunich> ick! (see pic at bottom right) http://www.moldmakingtechnology.com/articles/030702.html
[21:23:03] <cradek> who hasn't seen that once or twice?
[21:23:17] <jmkasunich> me
[21:23:25] <jmkasunich> I don't tend to run those speeds
[21:23:25] <cradek> reallllly
[21:23:40] <cradek> I've sure seen it in plastic a lot...
[21:24:00] <cradek> I guess in Al I just usually break the cutters off if I screw up
[21:24:21] <toastydeath> hahaha i've done that to many cutters, bottom right picture
[21:24:53] <jmkasunich> I had a very nice high helix 3-flute 1/2" endmill, it snuck out of the collet increasing depth of cut till it broke
[21:25:08] <jmkasunich> never clogged tho (that was on the Van Norman, very modest RPM)
[21:25:26] <skunkworks_> our big supprise was trying to cut steel with a HSS mill thinking it was carbide.
[21:25:33] <jmkasunich> melt
[21:25:43] <skunkworks_> glowing red.
[21:25:56] <jmkasunich> so - I need some help here
[21:26:10] <skunkworks_> estop - then about a minute later it went pop as it must have broke away from the part.
[21:26:15] <jmkasunich> working on a spindle design that will be used for both PCB milling and light regular milling
[21:26:26] <cradek> I've glowed a 3/4 end mill on steel - way too fast - sparks flying out in a circle like fireworks
[21:26:46] <jmkasunich> wtf were you thinking?
[21:27:07] <cradek> ...
[21:27:07] <toastydeath> i saw a kid run a 3/4" endmill backwards in steel at 3000 rpm
[21:27:12] <toastydeath> at school
[21:27:24] <toastydeath> looked like a grinder
[21:27:25] <cradek> haha, I've seen drills run backward - they do sort of work a bit, for a while
[21:27:26] <jmkasunich> ever heard of friction stir welding ;-)
[21:27:33] <alex_joni> * alex_joni did
[21:27:37] <toastydeath> they do work, which is the scary part
[21:27:48] <toastydeath> but you'd think the sparks and molten steel flying would dissuade a person from continuting
[21:27:51] <toastydeath> *continuing
[21:28:05] <toastydeath> this kid just kept at it for 4" of cut until the thing disintegrated completely
[21:28:07] <jmkasunich> common sense is less common than you'd think
[21:28:09] <alex_joni> oh.. cool.. sparks.. lets crank FO higher
[21:28:13] <SWPadnos> but sparks are COOL!
[21:28:37] <toastydeath> I SUPPOSE SO
[21:28:55] <alex_joni> although to be precise, sparks are actually HOT!
[21:29:00] <jmkasunich> anyway - what is a good speed for end milling AL with "plain old carbide endmills" ?
[21:29:13] <toastydeath> anything you can get
[21:29:27] <toastydeath> you can do 2000-3000 sfm without much difficulty with plain, uncoated carbide
[21:29:40] <toastydeath> hm, but that's with coolant
[21:29:48] <cradek> I think the right speed is "a little faster than that"
[21:29:49] <toastydeath> so 1500 is what i'd use without
[21:29:51] <jmkasunich> this would be dry, maybe air blast to clear chips
[21:30:16] <toastydeath> without flood, stick under 2000 sfm
[21:30:33] <toastydeath> just make sure you can support the feed rate required, or back the rpm down until you can
[21:30:35] <jmkasunich> holy cripes - for a 1/8" cutter, 1500 SFPM is 45K RPM
[21:30:47] <toastydeath> yeah
[21:31:04] <jmkasunich> ok, that actually makes things simpler for me
[21:31:13] <jmkasunich> I have a 1/2HP, 18000 RPM motor
[21:31:17] <toastydeath> also, make sure it's not four flute
[21:31:29] <jmkasunich> designed a belt drive to take it up to about 36K for the tiny PCB mills
[21:31:29] <toastydeath> 1/2 might be an issue
[21:31:55] <jmkasunich> I was thinking I needed another belt step for slower speeds when doing real milling (1/8" cutter in Aluminum)
[21:32:02] <toastydeath> 72 inches per minute on a 1/8 cutter
[21:32:08] <toastydeath> .002/tooth, 2 flute
[21:32:28] <jmkasunich> I was thinking about 10K rpm --> 327 SFPM
[21:32:58] <jmkasunich> but it sounds like I can get away with one pulley ratio for both kinds of milling
[21:32:58] <toastydeath> that'd make the feed more managable for sure
[21:33:00] <cradek> for hss with coolant that would be good...
[21:33:13] <toastydeath> ^^ true
[21:33:34] <cradek> (man, wish I had 10k)
[21:33:43] <cradek> hell, wish I had 5k
[21:33:55] <jmkasunich> I have a VFD for the motor, but power is proportional to speed - its 1/2HP at 18K on the motor, 1/4HP at 9K, etc
[21:34:14] <toastydeath> ah
[21:34:46] <jmkasunich> maybe I should compromise - choose pulleys that give me 30K or so
[21:34:58] <toastydeath> you might want to gear it to run at the 1/2 hp
[21:35:03] <jmkasunich> right
[21:35:27] <jmkasunich> the big issue is - 1 ratio or 2
[21:35:34] <fenn> 2
[21:35:42] <toastydeath> 2
[21:35:58] <jmkasunich> there are tradeoffs (as always)
[21:36:02] <toastydeath> one to one and then a 18000 to 9k stepdown
[21:36:22] <fenn> step up and step down
[21:36:38] <fenn> one to one is silly, you can just use the built in collet
[21:36:57] <jmkasunich> the main tradeoff is: the pulleys are between the bearings and the tool, so making a two step pulley increases tool stickout
[21:37:02] <jmkasunich> fenn: what built in collet?
[21:37:09] <fenn> the bosch thingy has a collet doesnt it?
[21:37:10] <toastydeath> ooh
[21:37:11] <jmkasunich> this is an AC induction motor
[21:37:20] <fenn> ok nevermind
[21:37:35] <jmkasunich> the bosch thingy has an integral step up gear - the actual spindle goes 50K
[21:37:37] <fenn> must be some other 1/2 HP 18000 rpm motor
[21:37:42] <fenn> oh wow
[21:37:55] <jmkasunich> but I'm not gonna use the spindle part - too much runout, not enough stiffness
[21:38:03] <jmkasunich> its made for handheld use - like a die grinder
[21:38:10] <jmkasunich> I _am_ gonna use the motor
[21:38:20] <fenn> why are the pulleys between the bearings and tool? that sounds like bad design
[21:38:36] <jmkasunich> its the lesser of two or three evils
[21:38:54] <jmkasunich> drive on the back means the spindle has to be much longer
[21:39:04] <jmkasunich> the drawbar has to be much longer
[21:39:09] <fenn> but it's longer on the side that doesn't matter
[21:39:13] <jmkasunich> the drawbar hole in the spindle has to be much longer
[21:39:53] <jmkasunich> my plan right now is to have a quill, made out of 1.5" turned/ground/polished and hard-chromed 1045 bar
[21:40:00] <jmkasunich> the quill will be about 7" long
[21:40:12] <jmkasunich> the spindle is only about 2" long, in the front of the quill
[21:40:21] <fenn> and the pulley is on the end of the quill?
[21:40:24] <jmkasunich> the Z screw is centered in the quill, from the top
[21:40:34] <jmkasunich> the pulley is just in front of the front spindle bearing
[21:40:37] <jmkasunich> the quill doesn't spin
[21:40:59] <jmkasunich> quill = the part of a drill press that goes up and down, and has a rack cut in it for the feed handle
[21:41:03] <fenn> so it's really just a round z axis
[21:41:15] <jmkasunich> yes, with the spindle and Z screw both on the centerline
[21:41:39] <jmkasunich> when all the way up, the bottom of the Z screw will hit the top of the drawbar and compress the bellvilles - tool release
[21:41:51] <fenn> clever
[21:41:59] <jmkasunich> a 3/16" wide belt is enough for the power I have
[21:42:06] <fenn> don't crash it at 36k rpm :)
[21:42:15] <jmkasunich> so a one-speed drivetrain is maybe 1/4" of tool stickout
[21:42:21] <fenn> you might want to make some kind of centrifugal lock-out
[21:42:26] <jmkasunich> but doubling that isn't so much fun
[21:42:33] <jmkasunich> classicladder
[21:43:14] <jmkasunich> soft limit will be _before_ the screw hits the drawbar, I'll add a small offset to bring it up the last 0.050 and release
[21:43:22] <jmkasunich> the offset, etc, will be in hal
[21:43:31] <fenn> what ID on the bearings?
[21:43:34] <jmkasunich> and be interlocked with the spindle drive
[21:43:35] <jmkasunich> 10mm
[21:43:55] <fenn> 1 diameter stick-out isnt too bad
[21:44:38] <fenn> timing belt drive?
[21:44:51] <jmkasunich> either MXL timing belt, or a flat belt
[21:45:08] <fenn> you know you can get 1/8" MXL belts
[21:45:11] <jmkasunich> yep
[21:45:23] <jmkasunich> but 1/8" is marginal for 400W at that speed
[21:45:51] <jmkasunich> I had a nice discussion with a belt design engineer this morning
[21:46:11] <jmkasunich> normally the kind of person who isn't allowed to waste time talking to a guy who will buy one belt
[21:46:19] <fenn> heh
[21:46:38] <jmkasunich> but it was slow today, and as he said "I'm twiddling my thumbs, so sure, I'll help you figure stuff out"
[21:47:03] <jmkasunich> I'm still torn between flat belt and MXL timing
[21:47:14] <fenn> what are the benefits of flat belt?
[21:47:46] <jmkasunich> less noise, and I can make the pulley myself, including final finishing when its on the spindle - so it will be perfectly concentric
[21:48:13] <jmkasunich> a timing belt pulley is a buy item - the $ isn't much, but I'll have to bore it, and it will be hard to get the new bore perfectly concentric
[21:48:24] <jmkasunich> any runout will cause vibration at 30K RPM
[21:48:25] <fenn> i hate that
[21:48:36] <fenn> why can't they just sell pulleys with a variety of bore sizes
[21:49:13] <jmkasunich> because there is a _very_ wide variety of needed bores
[21:49:22] <jmkasunich> I want "press onto 10mm shaft"
[21:49:35] <archivist> taper lock hubs :)
[21:49:53] <jmkasunich> archivist: 19mm diameter pulley ;-)
[21:50:08] <archivist> can be done
[21:50:19] <fenn> anything can be done if you're stubborn enough
[21:50:31] <jmkasunich> anything can be done
[21:51:04] <jmkasunich> but I can't find a place to buy a 19mm diameter MXL pulley with either a 10mm press fit bore _or_ a taper lock bore
[21:51:17] <fenn> sudo make me a timing pulley
[21:51:17] <jmkasunich> so either way, I wind up boring it myself
[21:51:48] <fenn> you're turning the spindle so it doesn't have to be 10mm does it?
[21:52:10] <jmkasunich> I'm planning to use 10mm drill rod for the part the bearings fit on
[21:52:15] <jmkasunich> more accurate than anything I turn
[21:52:30] <jmkasunich> press a pulley on the front, which also serves as the front bearing shoulder
[21:52:37] <fenn> right
[21:52:59] <jmkasunich> thread the back M10x1 for a nut which pushes on the bellville bearing preload springs
[21:53:33] <jmkasunich> two abec7 bearings - 10mm ID, 26mm OD, 8mm thick
[21:53:35] <fenn> the bearings are spring loaded?
[21:53:55] <jmkasunich> outer races spaced apart by 2 or 3 26mm OD x 8mm thick spacers
[21:53:58] <jmkasunich> yes
[21:54:23] <jmkasunich> deep groove bearings turn into angular contact when you apply axial preload
[21:54:47] <fenn> yep i do that with skate bearings :P
[21:57:26] <jmkasunich> http://jmkasunich.com/pics/eighth-inch-spindle-1.pdf
[21:58:06] <jmkasunich> the "spacers" are the outer races of abec1 bearings - $1.49 each ;-)
[21:58:22] <jmkasunich> the red things at the top are the preload bellvilles
[21:58:54] <jmkasunich> as-drawn, there is a 3/8" wide flat belt
[21:59:02] <jmkasunich> I now know I can use about half that width
[21:59:45] <jmkasunich> assuming I don't try to cram a two-step pulley in there
[22:04:47] <alex_joni> hi seb
[22:05:42] <seb_kuzminsky> hiya
[22:06:19] <alex_joni> what's new?
[22:06:28] <seb_kuzminsky> another day, another bug ;-)
[22:06:43] <seb_kuzminsky> i'm hoping to take a look at the hm2 stepgen overshooting problem tonight or this weekend
[22:07:03] <seb_kuzminsky> it works fine with the hm2-stepper config, but not with the one that you sent me i guess
[22:07:10] <seb_kuzminsky> more testcases is good ;-)
[22:08:14] <fenn> jmkasunich: is that a standard taper?
[22:09:15] <fenn> probably mt0
[22:11:56] <jmkasunich> not standard
[22:12:06] <jmkasunich> morse is about 0.6" per foot
[22:12:25] <jmkasunich> this is 2" per foot - so I won't have to hit it with a hammer to release
[22:12:39] <jmkasunich> standard ISO taper (CAT30, CAT40, etc) is 3.5" per foot
[22:13:28] <fenn> tailstocks work the same way you're describing, what's the difference?
[22:13:48] <fenn> (not like it's particularly easy to find mt0 collets anyway)
[22:14:35] <jmkasunich> the difference is that what feels like just a little force to your hand on a 4" diameter tailstock wheel is actually quite a bit of torque for a small stepper or servo
[22:39:16] <jmkasunich> decision made - 42 tooth MXL pulley driving 28 tooth MXL pulley, 1.5:1 step up, 27000 RPM max at the spindle
[22:39:20] <jmkasunich> only one step
[22:39:40] <jmkasunich> pulley is 0.39" wide, about what I have on the drawing
[22:48:34] <BigJohnT> seb lives under a tree according to the map :)
[22:49:36] <seb_kuzminsky> heh
[22:49:43] <seb_kuzminsky> i didnt type in my street address, just the city
[22:49:48] <seb_kuzminsky> i guess boulder is under a tree
[22:50:06] <alex_joni> heh
[22:50:16] <alex_joni> city only is just fine
[22:50:35] <BigJohnT> when you zoom in it points to a rock under a tree on someones driveway :)
[22:51:00] <alex_joni> it's a boulder, not a rock
[22:51:16] <BigJohnT> heh seb_kuzminsky I found a cheap way to connect to a 50 pin cable
[22:52:17] <Skullworks> Boulder where you stand an equal chance of being killed by a lighting strike or pulverized by small boulder sized hail...
[22:52:34] <seb_kuzminsky> or run over by a hippie in a Range Rover
[22:53:28] <Skullworks> higher odds on the Ranger Rover - they do sell like crazy here
[23:03:34] <BigJohnT> I tried the V-F board with the software encoder but the results were not good
[23:04:58] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/50pin01.jpg
[23:05:10] <BigJohnT> http://i47.photobucket.com/albums/f163/johnplctech/50pin02.jpg
[23:05:14] <BigJohnT> $5
[23:21:28] <eric_unterhausen> what's a v-f board, and do the bad results have anything to do with the fact that you didn't solder your connector?
[23:22:26] <BigJohnT> LOL I have not used the connector yet... I had it wired into a BOB and then into my parallel port
[23:22:41] <BigJohnT> It is soldered now :)
[23:22:51] <BigJohnT> voltage to freq
[23:23:55] <BigJohnT> the velocity output from the software encoder is too lumpy so I'm hooking it up to the 5i20 next
[23:53:34] <PCW> BigJohn: you would need a longer averaging time for the Software counted velocity
[23:57:14] <BigJohnT> PCW: I kinda figured that but I'm going to test it with the 5i20 as I got my headers in also the pulse past 3 or 4 volts input is too fast for the computer I had it hooked to
[23:57:52] <BigJohnT> * BigJohnT goes back to cooking fresh fish :)
[23:57:56] <PCW> In divide by 128?
[23:59:13] <PCW> fresh fish !