#emc | Logs for 2009-12-17

Back
[00:00:06] <toastydeath> or some ridiculously tiny number
[00:00:32] <toastydeath> it's just providing pressure on top of the paint to get it down the tube, you never actually lose pressure other than what volume you gain as the paint runs out
[00:00:46] <andypugh> I know that the paint isn't normally pumped, but there is no reason it couldn't be
[00:00:52] <toastydeath> you'd blow the seal
[00:00:56] <toastydeath> on the airgun
[00:01:22] <toastydeath> an airgun can't handle more than 30 or so psi coming down the paint tube
[00:01:24] <andypugh> There is a seal?
[00:01:27] <toastydeath> yes
[00:01:39] <toastydeath> the needle seal
[00:01:49] <ds3> are you telling an airgun works on roughly the same principle as a mustard bottle?
[00:01:50] <andypugh> Throw away the needle
[00:01:56] <toastydeath> you can't throw away the needle
[00:02:02] <andypugh> I just did
[00:02:05] <toastydeath> that's what laminates the flow to get razor thin lines
[00:02:17] <ds3> i.e. airpressure from a compress instead of air pressure from squeezing a bottle?
[00:02:34] <toastydeath> ds3, no, airguns normally work on siphon
[00:02:47] <L84Supper> http://www.youtube.com/watch?v=C4fFh0FLyBA
[00:02:56] <toastydeath> but you can feed most siphon guns with a light pressure similar to squeezing a mustard bottle
[00:02:59] <ds3> ah
[00:03:15] <ds3> then what about finding a way to modulate the air pressure instead?
[00:03:25] <andypugh> Here is my idea. You run the airbrush effectively starved of paint.
[00:03:32] <ds3> with all the pneumatics out there, this should be a pretty stock thing
[00:03:32] <toastydeath> that was one of the options, the question is which one is better
[00:03:51] <toastydeath> pressure variation to maintain the paint ratios, or restrictions
[00:03:53] <ds3> gotcha
[00:04:04] <toastydeath> andypugh, artist airbrushes don't work that way
[00:04:07] <toastydeath> they are dual action
[00:04:17] <toastydeath> one of them admits air to the body of the gun, with zero paint
[00:04:20] <andypugh> You then let is have carefully measured amounts of paint from tiny pumps. Not enough to make any pressure, in fact less that it might normally draw.
[00:04:35] <toastydeath> so the brush is now spraying air around the needle
[00:04:47] <andypugh> Yes..
[00:05:09] <toastydeath> when you pull back, the needle slides back and opens a small ring on the circumference of the nozzle
[00:05:20] <toastydeath> air AND paint go through this small opening and mix with the larger air mass
[00:05:38] <toastydeath> the further you pull the needle back, the more paint comes out
[00:05:57] <andypugh> Yes, so you only let it have the paint you want it to have, from a series of tiny pumps
[00:06:17] <toastydeath> you can do that, but then you lose the laminar quality of the paint flow
[00:06:26] <andypugh> Why?
[00:06:27] <toastydeath> that would work brilliantly in a larger brush for coverage
[00:06:43] <L84Supper> another digital airbrush http://www.youtube.com/watch?v=WoH0eOwCi8M&feature=related
[00:06:52] <toastydeath> because the needle surface drag slows the air containing the paint down
[00:06:54] <andypugh> You need to give it the amount it would normally pull, but that is just NC
[00:06:56] <toastydeath> holding it close to the needle
[00:07:03] <toastydeath> as the air in other layers rushes around it
[00:07:18] <toastydeath> so if you place the needle essentially on the paper and fire it, you get a razor sharp line
[00:07:37] <toastydeath> L84Supper, those are awesome links, ty
[00:07:43] <Valen> thats cool
[00:07:49] <andypugh> what normally controls the amount of paint in the flow?
[00:08:04] <toastydeath> the needle
[00:08:25] <andypugh> It comes down a tube, you just control the amount of each paint allowed into that tube
[00:08:57] <toastydeath> i'm not saying it doesn't work, what i'm saying is that removing the needle removes other features and behavior of the airgun that are critical to using it artistically
[00:08:58] <andypugh> You then need to control the needle position and the pump speeds, but that is just maths
[00:08:59] <L84Supper> we do it using articulating robots
[00:09:32] <tomp> tomp is now known as tom3p
[00:09:46] <Valen> I'm suprised its not done with a rotary axis
[00:10:12] <L84Supper> sure, for cans, bottle and tubes
[00:10:27] <Valen> or helmets say
[00:10:31] <toastydeath> andypugh, what you're saying will absolutely work
[00:10:42] <toastydeath> i am not willing to give up the characteristics of a needle
[00:11:12] <toastydeath> i don't want to spend that much work on that many pumps and trying to figure out exactly how much flow to have
[00:11:13] <andypugh> I amnow saying keep the needle, control the paint flow, and use maths..
[00:11:21] <toastydeath> how do i do that manually?
[00:11:31] <andypugh> You don't.
[00:11:50] <toastydeath> right, so it's an option down the road if I do this and decide it's worth making electronic
[00:11:51] <andypugh> Gear pumps are easy to make. Tiny steppers are very cheap.
[00:12:24] <toastydeath> but right now i'm looking at the 10 dollars per needle valve or 10 dollars per regulator, and it looks like something a person could cobble together in an afternoon
[00:13:02] <andypugh> You could run the tiny steppers direct from a P-Port or (more likely, given the number you want) from a Mesa card with only a ULN2003 as a current buffer.
[00:13:50] <toastydeath> at the moment, a manual configuration is superior if one is to be had, because I don't want to have to write the software to control for paint variables that such a system would introduce
[00:14:23] <Valen> if you ran a liquid sprayer it wouldn't be so bad
[00:14:50] <toastydeath> Valen, liquid sprayer?
[00:14:59] <Valen> no air
[00:15:07] <Valen> just squirt liquid paint directly
[00:15:09] <L84Supper> hvlp
[00:15:33] <andypugh> I was assuming that you would be using EMC and just setting stepgen velocities (which would map direct to paint volumes)
[00:15:37] <L84Supper> airless sprayers
[00:15:44] <toastydeath> andypugh, if it goes electronic than yeah
[00:16:00] <MattyMatt> fuel injectors would be your airless sprayers
[00:16:13] <Valen> nah they are on/off
[00:16:18] <toastydeath> but you have to agree there's a huge difference between "screw a regulator onto an airline" and setting up emc
[00:16:44] <Valen> I reckon the gear pump with a braided line to the head
[00:16:46] <MattyMatt> piezo = either pulsed drops or PWM regulation. either would work
[00:17:05] <toastydeath> and even still, this is far astray of the problem as stated
[00:17:15] <MattyMatt> a hi-freq combo of the 2 is what I'd expect
[00:17:17] <andypugh> Yes, but injectors go on-off very rapidly. Minimum opening is <1 uS
[00:17:38] <toastydeath> how do you passively control paint ratios given arbitrary and unknown flow rates
[00:18:06] <MattyMatt> camera watching the output?
[00:18:12] <toastydeath> dear lord.
[00:18:20] <MattyMatt> PID loop via pantone charts
[00:18:22] <toastydeath> passive, passive control
[00:18:45] <Valen> run it like stepper, open loop
[00:18:54] <MattyMatt> * MattyMatt does passive aggresive control
[00:19:03] <toastydeath> i can't run steppers without electricity
[00:19:05] <MattyMatt> yes, it's a NICE colour
[00:19:15] <toastydeath> nor emc
[00:19:21] <andypugh> You can't passively control, so don't try.
[00:19:34] <toastydeath> why not?
[00:19:43] <andypugh> It's an oxymoron
[00:19:51] <toastydeath> okay, i'm not using electronics
[00:19:56] <toastydeath> if it can't be done manually, it won't be done
[00:20:23] <toastydeath> simple specification.
[00:20:46] <Valen> what are you trying to do?
[00:20:48] <andypugh> This is the EMC channel. If it's not electronic, it's off-limits
[00:20:51] <toastydeath> if it can be done manually, and it turns out to be a worthwhile concept, then it's worth the time and money to do it right electronically.
[00:21:05] <toastydeath> not until then
[00:21:15] <toastydeath> Valen, control paint mixing to an airbrush.
[00:21:34] <Valen> your going to mix the paints then spray?
[00:21:38] <toastydeath> yes
[00:21:53] <Valen> won't work if your trying to spray an image
[00:22:08] <toastydeath> for which reason
[00:22:09] <Valen> the colours will lag
[00:22:24] <toastydeath> what system are you imaging where the colors lag?
[00:22:35] <Valen> the tube with the mixed paints in it before the atomiser
[00:22:35] <MattyMatt> raster scanning
[00:22:38] <toastydeath> the colors are forced under pressure into the feed tube
[00:23:16] <toastydeath> so as soon as the needle opens to admit paint, the 20 psi sitting down in the manifold forces the paint to refill
[00:23:57] <MattyMatt> that plotter made from a spraycan on 2 wires to top corners was cool
[00:25:49] <Valen> I would use a seperate paint system per colour
[00:25:49] <toastydeath> indeed it was
[00:26:49] <toastydeath> seperate paint is what i'm trying to get away from
[00:26:58] <Valen> why?
[00:27:07] <MattyMatt> airbrushes are cheap, once you've got a compressor
[00:27:30] <toastydeath> because I'm not interested in graffiti type artwork, I'd like to be able to actually mix and refine colors
[00:27:45] <lilalinux> lilalinux is now known as lilalinux_away
[00:27:45] <Valen> you should be able to mix them on the job
[00:27:54] <Valen> its how ink/bubblejet printers work
[00:28:02] <Valen> I'd be tempted to just use one of those
[00:28:18] <toastydeath> okay, let me back up again - i admit there are other solutions to problems similar to mine
[00:28:21] <MattyMatt> yeah especially water based. that'll mix in the air on the way to the job
[00:28:36] <toastydeath> the problem, and the only problem i have, is not with mixing, the brush itself, or anything else
[00:28:51] <toastydeath> it's how to meter multiple fluids in ratios across a small band of flow rates
[00:29:36] <toastydeath> experimenting is probably the only way to find out
[00:29:37] <L84Supper> collect the multiple fluids and the pump that to the nozzle
[00:29:47] <L84Supper> sp the/then
[00:30:01] <MattyMatt> variable apertures, with constant total cross section
[00:30:23] <MattyMatt> so gate/butterfly/needle valves all on steppers
[00:30:42] <toastydeath> you feel the restriction method is probably superior to varying the pressure itself, then?
[00:30:58] <MattyMatt> yes, keep the overpressure constant
[00:31:14] <toastydeath> because that's really the internal dilemma i'm having, pressure as variant or aperture as variant
[00:31:18] <MattyMatt> unless it's a small pump near the cup
[00:32:02] <toastydeath> nah, it's air over the fluid providing the pressure, only the individual paint lines go up to the cup
[00:32:04] <L84Supper> all the pressures would need to stay in the same proportion to each other, how will you do this manually?
[00:32:18] <Valen> If you did do the PWM with various paints it'd be closer to how modern printers work
[00:32:31] <Valen> and they do work
[00:32:44] <toastydeath> L84Supper, you're saying i need to have the pressure proportional to the aperture?
[00:33:00] <Valen> you will if you want a constant spray pattern
[00:33:18] <toastydeath> if i was using needle valves, all the paints would be on the same shared pressure manifold
[00:34:09] <toastydeath> so every paint line would have the same 20 psi sitting behind it
[00:35:15] <MattyMatt> that pressure would be 10 psi half way up the pipe
[00:35:17] <Valen> I'm liking MattyMatts idea of the fuel injectors
[00:35:29] <MattyMatt> that was andypugh's
[00:35:32] <toastydeath> my concern is that, say when first mixing the paints, when I go from .001 gpm to .025 gpm or something
[00:35:37] <L84Supper> this is usually when I ask for a block diagram
[00:35:49] <toastydeath> that the ratio will change
[00:36:14] <Valen> this is why I'm liking the injector idea, put a constant pressure behind it, you will get a set spray pattern
[00:36:20] <Valen> one injector per color
[00:36:25] <toastydeath> Valen, the paint just oozes out of the end of the tube
[00:36:27] <toastydeath> it does not spray
[00:36:31] <MattyMatt> yeah, meter each pigment individually
[00:36:33] <toastydeath> this is a feed tube for an airbrush
[00:36:47] <toastydeath> so each color has it's own tube, and it's own needle valve
[00:36:57] <Valen> I'm saying an airbrush is the wrong application
[00:36:58] <toastydeath> all those tubes go up to the brush and combine, and then feed the paint cup
[00:37:02] <Valen> wrong tood for the job
[00:37:06] <Valen> tool
[00:37:30] <toastydeath> ...you're telling me that an airbrush is the wrong tool to produce airbrush art?
[00:37:36] <dan1mal_garage> ha
[00:37:42] <toastydeath> i could use crayons, i suppose
[00:37:50] <dan1mal_garage> yea, buy a wagner power painter
[00:37:52] <Valen> its the wrong tool to produce airbrush art from a computer
[00:37:59] <toastydeath> i'm not producing art from the computer
[00:38:01] <toastydeath> i'm holding it in my hand
[00:38:02] <L84Supper> high pressure paint behind the injector since it has a low duty cycle
[00:38:13] <toastydeath> if the manual version of the color mixer works, THEN i will make it emc-based
[00:38:18] <toastydeath> not the airbrush itself
[00:38:26] <MattyMatt> do you have enough machine shop to make a rotary selector/manifold?
[00:39:09] <toastydeath> i was trying to steer away from the rotary selector method
[00:39:18] <toastydeath> they already exist and don't work for mixing more than two paints
[00:39:42] <MattyMatt> you could cascade them
[00:39:43] <toastydeath> you can get red->yellow, or yellow->blue, but unless it's a 3 color selector, you can't loop back around to blue->yellow
[00:40:23] <toastydeath> that is true
[00:40:55] <toastydeath> L84Supper, i have a question for you
[00:41:03] <L84Supper> http://americas.kyocera.com/kicc/automotive/piezo.html Piezoelectric-Stack
[00:41:05] <toastydeath> you mentioned air proportioning
[00:41:20] <toastydeath> i was hoping the air didn't actually matter since there's no computer controlling this yet
[00:41:39] <toastydeath> that, if i had a shared 20 psi over each color, i could just open and close a valve until I got the right mix
[00:41:46] <MattyMatt> "faders" for each pigment would be most versatile, I suppose
[00:41:58] <toastydeath> say, set yellow half open, and then start opening blue until I got the purple i wanted
[00:42:09] <MattyMatt> try it and see. no doubt it'll work at some settings
[00:42:11] <toastydeath> and then opened white until it was the correct lightness
[00:42:13] <dan1mal_garage> can you airbrush a naked lady on the hood of my car
[00:42:26] <toastydeath> not yet, I'm getting there slowly.
[00:42:44] <toastydeath> i am confident the method i just described works for a constant flow
[00:43:05] <andypugh> That Kyocera thing: Note that that is a piloted-valve. The needle is opened by the fuel pressure all the Piezo does is release the pressure on the top of the needle.
[00:43:06] <toastydeath> i just wasn't sure if the ratios of the paint flowing through those valves was going to change after I start spraying more or less paint
[00:43:33] <toastydeath> i.e. with the yellow mostly open and the blue mostly closed, as I demand more paint, the hue would shift more toward yellow
[00:43:43] <toastydeath> or if it would remain mostly linear
[00:43:43] <andypugh> And RoHS is going to make PZT piezos illegal some time next year.
[00:43:58] <MattyMatt> * MattyMatt does AoHS
[00:44:09] <toastydeath> aohs?
[00:44:11] <MattyMatt> Addition of :)
[00:44:42] <MattyMatt> I've got enough 60/40 to last until 2040
[00:44:58] <toastydeath> i still don't know what the hs stands for
[00:45:07] <MattyMatt> hazardous substances
[00:45:08] <toastydeath> =(
[00:45:13] <toastydeath> oh
[00:45:16] <toastydeath> ha
[00:46:19] <andypugh> PZT is lead zirconium titanate. It makes no sense to worry about stable lead compounds, but they have decided that lead is bad regardless, so out it goes.
[00:46:37] <dan1mal_garage> what about solder?
[00:46:49] <MattyMatt> that's gone man
[00:46:55] <dan1mal_garage> nooo!
[00:47:10] <MattyMatt> it's all zinc & tin now, with 3% gold if you're rich and want it usable
[00:47:19] <dan1mal_garage> how am i going to fix my cell phone charger everytime my dog eats it
[00:49:43] <MattyMatt> http://www.goldmine-elec-products.com/prodinfo.asp?number=G16133
[00:49:48] <MattyMatt> >:)
[00:50:05] <toastydeath> hahaha
[00:50:14] <toastydeath> you rebel
[00:50:38] <dan1mal_garage> haha
[00:50:42] <dan1mal_garage> sold!
[00:50:47] <dan1mal_garage> added to cart
[00:51:02] <dan1mal_garage> how it works for cnc's!
[00:51:06] <dan1mal_garage> hope*
[00:51:15] <MattyMatt> have you seen the 49c pcb?
[00:51:26] <MattyMatt> that looks like a typo
[00:51:27] <dan1mal_garage> no
[00:51:44] <MattyMatt> http://www.goldmine-elec-products.com/prodinfo.asp?number=G2626
[00:51:55] <MattyMatt> bargain of the century, that one
[00:53:04] <dan1mal_garage> cheap
[00:53:23] <MattyMatt> lotsa bargains there. 99c solenoids etc
[00:53:27] <L84Supper> that price is about right if it's surplus, it's not G10 or FR4
[00:54:11] <dan1mal_garage> i'm not much of an electronics guy
[00:54:29] <andypugh> I don't bother with anything but FR4 nowadays, it's hard to break, I like that
[00:54:42] <dan1mal_garage> my mill retrofit is about as technical as i've gotten
[00:57:20] <MattyMatt> aw we're not even on the rip off list. they'll ship to S Korea or Aus for $50 but not here at all
[00:57:55] <MattyMatt> unless we count as a US Protectorate :)
[00:59:04] <MattyMatt> * MattyMatt sends order from Gitmo
[01:00:09] <eric_unterhausen> I'm pretty sure that surplus circuit board is offcuts from a panel where they just wasted the space
[01:01:05] <MattyMatt> of course. nothing wrong with that
[01:01:21] <MattyMatt> 8" x 4" could be an unpopular size
[01:01:46] <MattyMatt> or they got a batch in that's tarnished or badly squared
[01:02:04] <MattyMatt> would you get 2x PC104 on 8x4?
[01:03:08] <MattyMatt> ah well, I'm a way away from milling pcbs yet
[01:03:40] <MattyMatt> my Z is way too heavy for the ballscrew slide to hold it by itself. I need another X slider
[01:06:48] <MattyMatt> probably a pair, and put the ballscrew in the middle. this unit has seen service and the slide has a bit of lash when there's enough leverage
[01:07:43] <MattyMatt> about 5mm worth in Y with a 300mm Z
[01:10:31] <MattyMatt> I'm using this new kind of slider I found in B&Q again. "pocket door slider" it's called. 550mm long with a 100mm slide with a constant 20 balls contact
[01:11:15] <MattyMatt> 10gbp per pair, but one would be enough for a very light axis
[01:14:33] <dan1mal_garage> for a machine??
[01:17:50] <MattyMatt> a non-contact machine :) a reprap or a laser or a plotter etc
[01:18:01] <dan1mal_garage> ugh now that i added a coolant relay seperate from the spindle, i need to add an m8 and m9 to every tool and every program i have
[01:18:15] <dan1mal_garage> cool
[01:19:19] <MattyMatt> why you'd use a 5q slider to position a 5kq laser is beyond me, but it'd work OK
[01:19:31] <andypugh> Probably still a bit light. Try a car scrapyard for electric seat rails. The have preload so they don't rattle
[01:19:33] <MattyMatt> maybe you stole the laser :)
[01:20:24] <MattyMatt> yeah I need a scrappy that'll let me scramble
[01:22:47] <MattyMatt> I'm taking way too long for a bodge job
[01:22:51] <andypugh> The backlash isn't even that huge, and the force is high (if the speed is low). One could probably even use them as-sourced
[01:24:56] <MattyMatt> maybe even as-intended. my car needs those :)
[01:25:19] <MattyMatt> heated leather seats too
[01:25:40] <MattyMatt> has anyone put the massager in emc?
[01:26:16] <jimbo8288_> I would like to add states to some of the buttons inside of the TOUCHY interface. On one of the buttons inside of the GTK I have a handler of on_estop_clicked. I have used grep to look at everything inside of the touchy folder and can find no referance to this label. Is this label referenced inside of HAL?
[01:27:41] <MattyMatt> is there a glade file?
[01:28:15] <MattyMatt> grep would find that tho
[01:28:53] <andypugh> Does halui pick them up?
[01:30:27] <jimbo8288_> I will look for halui. Thank
[01:36:23] <dan1mal_garage> i wonder if gas tank sealer would seal my coolant tank and keep it from rusting
[01:36:51] <dan1mal_garage> i use water based coolant, and the coolant tank is in the base of the machine, which is cast iron
[01:36:56] <toastydeath> leave it open
[01:37:05] <dan1mal_garage> and my coolant is now a lovely shade of rust
[01:37:08] <toastydeath> bacteria will eat the coolant over time and it will start to outgass
[01:37:22] <toastydeath> are you sure that's not waylube
[01:37:31] <dan1mal_garage> yes
[01:37:54] <dan1mal_garage> i have to fill the tank often enough so there really isnt any bacteria smell
[01:38:00] <toastydeath> coolant goes in pale white/blue/green, and after a day goes brown because of the emulsifiers pulling tramp oil off the surface
[01:38:20] <dan1mal_garage> na it's rust
[01:38:58] <dan1mal_garage> there really isnt any way lube
[01:39:06] <dan1mal_garage> i have a one shot oiler on it
[01:39:32] <dan1mal_garage> so it's not pumping in tons of it like a bed mill or something
[01:39:58] <dan1mal_garage> it's just the fact that it's a bare cast iron coolant tank
[01:41:13] <dan1mal_garage> i'd paint it but water based coolanty isnt exactly easy on paint
[01:41:28] <Jymmm> pweder coat it
[01:41:34] <dan1mal_garage> or easy on anything for that matter
[01:42:01] <dan1mal_garage> hmmm yea i could plate it
[01:42:43] <MattyMatt> explosive SS plating
[01:43:17] <dan1mal_garage> explosive diahrea plating?
[01:43:48] <andypugh> Polythene bag?
[01:44:03] <MattyMatt> dunno if that works with thin sheet, but it's an amazing way of getting 1 cm of SS on 10cm of plain steel
[01:44:27] <dan1mal_garage> the problem is that the cast iron base of the machine is the coolant tank
[01:44:35] <dan1mal_garage> i cant get in there easily at all
[01:44:37] <toastydeath> i have one machine that you can't see the actual coolant because there's so much waylube sitting on top of the water
[01:44:37] <toastydeath> it looks like a tank of motor oil
[01:44:37] <toastydeath> until you stick something in it and the water splashes out
[01:44:46] <andypugh> Empty it, put in a big bin-bag, refill. Job done
[01:45:30] <dan1mal_garage> the only holes are 2 for little screen filters, and a 6" hole for the coolant pump
[01:45:44] <dan1mal_garage> i dont think it's even 6", probably more like 4"
[01:46:16] <MattyMatt> that's enough to lay the charges one-handed :)
[01:46:21] <dan1mal_garage> no way to really bag it
[01:47:01] <dan1mal_garage> i dont want to blow up my machine matty lol
[01:47:23] <MattyMatt> it's a very controlled explosion :)
[01:48:20] <andypugh> You could, theoretically, electroplate. But I doubt you would get a good result.
[01:48:37] <Jymmm> powder coat it
[01:48:39] <dan1mal_garage> it doesnt need to look pretty
[01:48:49] <dan1mal_garage> cant get in there Jymmm
[01:49:02] <Jymmm> try harder
[01:49:08] <dan1mal_garage> there's several chambers and it's impossible to get to them
[01:49:13] <ds3> use better powder ;)
[01:49:23] <andypugh> Assuming it was cast, I expect you could split the machine and get in.
[01:49:49] <dan1mal_garage> haha hey i'm supposed to be the only smarty ass around here
[01:49:55] <dan1mal_garage> smart*
[01:50:07] <andypugh> But better might be to just ignore that reservoir, and go to an external tank and pump
[01:50:39] <dan1mal_garage> andypugh, i thought about that but it would be extremely difficult to catch the coolant
[01:51:44] <andypugh> Ah, yes. I was thinking something like this: http://www.axminster.co.uk/product.asp?pf_id=21084&name=coolant&user_search=1&sfile=1&jump=0 but you have a point
[01:52:21] <andypugh> Perhaps you need a rust-inhibiting coolant formulation (I have no idea if they exist)
[01:52:39] <MattyMatt> central heating antifreeze?
[01:53:00] <dan1mal_garage> yea, i'll have to do something
[01:54:33] <dan1mal_garage> man, i put this handy program resume button on my machine, and i keep forgetting to use it
[01:54:49] <toastydeath> program resume?
[01:55:01] <toastydeath> i just step through programs until I'm back where i need to be
[01:55:09] <toastydeath> hop over stuff
[01:55:25] <dan1mal_garage> pregram resume after an M1
[01:55:31] <dan1mal_garage> program*
[01:55:31] <andypugh> http://www.americanmachinist.com/304/Issue/Article/False/78564/Issue
[01:55:46] <andypugh> Perhaps switch to straight oil?
[01:55:57] <dan1mal_garage> yea, it'd be a pain to deal with
[01:56:14] <dan1mal_garage> i run mass production, it'd take too long to clean off the parts
[01:57:06] <andypugh> Interesting article thought, it suggests that your water matters
[01:58:06] <dan1mal_garage> yea, technically i should be using dionized water
[01:58:21] <dan1mal_garage> deionized*
[01:59:19] <dan1mal_garage> i've been using this stuff for over 12 years now, works ok with tap water, but not the best
[01:59:48] <andypugh> depending on your volumes, it is cheap enough to buy at one end, and cheap enough to make at the other. What's the betting you are in the middle?
[02:00:26] <andypugh> Anyway, sleep time
[02:01:23] <dan1mal_garage> most of mine goes on the floor
[02:03:04] <MattyMatt> has your water stayed the same? sand in the water is common in Holland
[02:03:40] <MattyMatt> new treatment or just silting up could leave a brown suspension
[02:03:58] <dan1mal_garage> it's definately rust
[02:04:14] <dan1mal_garage> from the tank
[02:04:37] <dan1mal_garage> if you saw it when i emptied it, you'd agree 100%
[02:05:23] <MattyMatt> hat's what I said to the doctors when I diagnosed myself with syphillis
[02:05:39] <dan1mal_garage> lol
[02:05:57] <dan1mal_garage> hanging around with some unsavory women?
[02:06:07] <MattyMatt> only some
[02:06:22] <MattyMatt> I felt deprived at the time
[02:06:55] <dan1mal_garage> or you just wanted the doctor to think you get laid?
[02:07:54] <MattyMatt> microscope + med textbook + itch
[02:08:47] <cradek> + swab + don't want to think about it
[02:09:13] <dan1mal_garage> haha
[02:09:19] <MattyMatt> I had to drop one of the sciences in school, so that was biology out
[02:09:40] <MattyMatt> I wasn't giving up physics or chem :)
[02:10:06] <dan1mal_garage> i'm unedumicated
[02:10:18] <dan1mal_garage> only trade schools for machining
[02:10:22] <MattyMatt> I'm worse. self educated
[02:10:50] <dan1mal_garage> machining is all i've done and all i know
[02:11:05] <MattyMatt> good plan
[02:11:17] <dan1mal_garage> hopefully it'll last another 40 years till i can retire
[02:12:18] <MattyMatt> it'll last as long as any job. AI will do it all eventually
[02:12:38] <MattyMatt> Management class think they'll be last with jobs. Ha!
[02:14:49] <MattyMatt> lawyers will be first
[02:15:01] <dan1mal_garage> hope so
[02:15:41] <dan1mal_garage> cant wait to have time to work on my lathe
[02:16:25] <MattyMatt> yet when you relax and claim welfare, you get called lazy, but it's the way we're all going
[02:18:42] <dan1mal_garage> my plan is to make my fortune young and buy up some properties and retire with a sweet work shop
[02:18:54] <MattyMatt> yeah that was mine
[02:19:01] <dan1mal_garage> already got the shop, just need to make the fortune :)
[02:19:35] <MattyMatt> I made the fortune, but couldn't keep it because I never had my own shop
[02:20:40] <MattyMatt> so, your shop IS your fortune
[02:20:55] <dan1mal_garage> i told myself i'd have my own shop as soon as i bought a house
[02:21:13] <dan1mal_garage> well i hope not, my shop isnt really all that great
[02:21:14] <MattyMatt> same thing :) house = shop + bunk
[02:21:48] <dan1mal_garage> it's sweet for a hobby shop, but not super for production
[02:22:28] <MattyMatt> as long as its yours, so your machines are never homeless
[02:22:53] <MattyMatt> a man can live on the street for a while, a machine can't
[02:23:16] <dan1mal_garage> well there's the problem, if i'm homeless, so are my machines
[02:23:52] <MattyMatt> got a mortgage?
[02:24:10] <dan1mal_garage> yea
[02:24:27] <dan1mal_garage> i've only had this house for a year and a half
[02:26:29] <dan1mal_garage> and houses in southern california are not cheap unfortunately
[02:26:47] <MattyMatt> k, so you need a fortune next :)
[02:27:09] <dan1mal_garage> i need a fortune just to pay the mortgage lol
[03:32:15] <Dave911_> Dave911_ is now known as Dave911
[05:52:44] <Dave911_> Dave911_ is now known as Dave911
[05:56:25] <KimK> KimK is now known as KimK_afk
[06:21:55] <lilalinux_away> lilalinux_away is now known as lilalinux
[07:00:41] <lilalinux> lilalinux is now known as lilalinux_away
[07:12:19] <tom3p> hektor the grafitti robot ( 2 axis plus spray can ) has a variant 'rita' that is a whiteboard plotter with pen changer http://video.google.com/videoplay?docid=-283528724322180681#
[07:34:10] <lilalinux_away> lilalinux_away is now known as lilalinux
[09:07:48] <lilalinux> lilalinux is now known as lilalinux_away
[10:55:08] <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/"
[11:00:09] <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/"
[11:30:43] <frallzor> eyey
[14:43:56] <christel> [Global Notice] Hi all, we are (surprise, surprise) still experiencing DDoS, after a quiet period it just started up again. Apologies for the inconvenience.
[16:17:50] <KimK_afk> KimK_afk is now known as KimK
[16:26:43] <Jymmm> 10lbs of mercury $100
[16:27:04] <SWPadnos> disposal of same, $100,000,000
[16:27:20] <Jymmm> Or gold mining I guess
[16:27:52] <Jymmm> I just can't think of a good use for it
[16:28:07] <SWPadnos> a really big tilt switch
[16:28:18] <archivist> we had a customer who wanted a mercury pendulum. so it was easy to get rid of some from here
[16:28:24] <jmkasunich> a really big thermometer
[16:30:10] <Jymmm> I dont even know how much that is.... 8 ounces?
[16:30:30] <SWPadnos> a thermometer probably has less than a gram
[16:30:34] <skunkworks> you could go into production of cfl light bulbs
[16:30:39] <SWPadnos> a tilt switch maybe a couple of grams
[16:30:46] <SWPadnos> 10 pounds is a shitload
[16:31:05] <Jymmm> it's FS on craigslist
[16:33:56] <SWPadnos> hmmm. I may have to light the pilot in the downstairs heater
[16:34:06] <SWPadnos> it's 4 degrees outside at the moment
[16:35:41] <Jymmm> * Jymmm hands SWPadnos 2 gal of gasoline and a flare
[16:36:04] <SWPadnos> I'm thinking the gas heater is a better idea
[16:36:57] <Jymmm> SWPadnos: Got Snow?
[16:37:04] <SWPadnos> yeah, a little bit
[16:37:10] <SWPadnos> maybe an inch or two
[16:37:13] <lilalinux> lilalinux is now known as lilalinux_away
[16:37:34] <lilalinux_away> lilalinux_away is now known as lilalinux
[16:37:35] <Jymmm> SWPadnos: Ok, make snow angels, pour in the 2 gal of gas and ignite
[16:38:09] <SWPadnos> "making snow angels" is the opposite of geting warm
[16:38:41] <eric_unterhausen> my daughter was doing an art project that involved fire
[16:38:46] <Jymmm> SWPadnos: No worries, in the end you'll get REALLY warm.... GOT MARSHMALLOWS?
[16:39:07] <eric_unterhausen> I probably should have given a better safety lecture first
[16:39:09] <Jymmm> eric_unterhausen: eating?
[16:39:15] <SWPadnos> oh. I missed the part about staying in the snow angel while it's on fire
[16:39:56] <frallzor> http://www.pici.se/p/glKvzfiCF/ my new family members :)
[16:40:04] <Jymmm> SWPadnos: Chestnuts roasting on an open fire......
[16:40:25] <SWPadnos> chest, nuts roasting on an open fire
[16:40:28] <SWPadnos> (I fixed that for you)
[16:41:05] <Jymmm> lol
[16:42:33] <eric_unterhausen> that reminds me, I learned a new self defense move -- "monkey steals the peach"
[16:43:19] <SWPadnos> I sense some nut-grabbing in that
[16:43:26] <Jymmm> eric_unterhausen: Yeah? Me too.... "hollow point make big hole"
[16:43:43] <SWPadnos> must be tai chi
[16:44:07] <eric_unterhausen> ninja
[16:44:32] <Jymmm> Ninjas suck, Pirates rule!
[16:44:33] <eric_unterhausen> I think there are more web sites about that than lolcats
[16:44:44] <Jymmm> and I have proof!
[16:45:37] <MattyMatt> jymmm, golf clubs
[16:46:18] <MattyMatt> floating glass
[16:46:26] <MattyMatt> making percussion caps
[16:46:45] <MattyMatt> and dageurrotypes
[16:46:51] <Jymmm> Just google "Ninja Master"
[16:47:09] <MattyMatt> also, injecting into oranges in the supermarket that dissed you
[16:47:13] <eric_unterhausen> I was trying to disassemble a part yesterday, tried a sledgehammer and a arbor press
[16:47:21] <eric_unterhausen> turned out it was threaded
[16:48:56] <Jymmm> eric_unterhausen: That's silly, you just needed one of these... http://sfbay.craigslist.org/sby/grd/1514098250.html
[16:58:20] <frallzor> did you see my beautiful pic MattyMatt? =
[16:58:22] <frallzor> =)
[16:59:06] <archivist_attic> I didnt
[16:59:15] <archivist_attic> linky?
[16:59:17] <frallzor> http://www.pici.se/p/glKvzfiCF/
[16:59:22] <frallzor> voilá
[17:00:04] <MattyMatt> I'm not jealous. it's too heavy for mine :) What's the box? the driver?
[17:00:18] <frallzor> yup
[17:00:49] <frallzor> cnc people dont get jelous, its allways fun to see hardware :P
[17:01:10] <MattyMatt> yeah but we're all male here >>:)
[17:01:38] <MattyMatt> there is inevitably an element of "mine is bigger than yours"
[17:01:47] <frallzor> haha :P really?
[17:01:50] <frallzor> never noticed
[17:01:59] <MattyMatt> well at least until nanotech takes off
[17:02:20] <MattyMatt> any girls here?
[17:02:52] <MattyMatt> there are girls with repraps
[17:05:02] <frallzor> my plan was to make one of those before the mechmate took of
[17:05:05] <frallzor> *off
[17:05:26] <MattyMatt> I'll try an extruder on mine
[17:05:37] <MattyMatt> they are cheap enough to make
[17:06:52] <MattyMatt> you could put a cement extruder on the mechmate and print garden furniture
[17:08:04] <skunkworks> I don't think cement would dry fast enough....
[17:08:05] <frallzor> haha
[17:08:23] <frallzor> well now im sure I can mill metall atleast =)
[17:22:01] <MattyMatt> so you make your next mechmate yourself? or is it still easier to get the steel plasma/lasered?
[17:22:51] <frallzor> next? :P
[17:22:53] <frallzor> never!
[17:23:22] <frallzor> now when you mention it, I could probably mill the part with some luck :P
[17:23:26] <frallzor> if it can do steel
[17:23:35] <eric_unterhausen> selling those things is a tough business to be in
[17:23:41] <MattyMatt> wait until that carpenter wants 24.5/7 work doing ;)
[17:24:32] <frallzor> well alot of people involved doing me favours or getting me dutyfree stuff :P
[17:24:52] <issy> hi all
[17:24:54] <frallzor> all want to see it done so they might tell a friend and then they tell a friend etc etc :P
[17:25:11] <MattyMatt> that's how you'll end up making them :)
[17:25:18] <eric_unterhausen> I have a part where I could make a decent amount of money if I could only find a $5 collet chuck
[17:25:20] <MattyMatt> even if you don't want to
[17:25:23] <frallzor> rather make stuff for ppl :P
[17:25:27] <frallzor> Im a master cadder too :P
[17:25:34] <frallzor> and studying design
[17:25:48] <MattyMatt> yeah custom kitchen cabinets
[17:25:57] <MattyMatt> is where the money's at
[17:25:58] <frallzor> ikea furnitue :P
[17:26:04] <frallzor> flat flat flat!
[17:26:09] <frallzor> *furniture
[17:26:10] <MattyMatt> yeah too flat
[17:26:24] <MattyMatt> kitchenswithcurves.com
[17:26:53] <MattyMatt> dammit, it's a porn site already :)
[17:27:11] <MattyMatt> fat mommas cooking in the nuddy
[17:27:16] <eric_unterhausen> nice
[17:27:41] <MattyMatt> sorry, overactive imagination :)
[17:28:06] <MattyMatt> they were right about me
[17:29:14] <eric_unterhausen> I was looking at a woodworker's shop pics, he was making a bombe chest
[17:29:16] <frallzor> http://camcraftsa.com/cnc-routing-textures.htm this is my dream :P
[17:29:23] <frallzor> I wanna learn how to do this
[17:29:41] <eric_unterhausen> then he decided he didn't like it and threw away $1000 U.S. in wood
[17:29:43] <frallzor> tried making the pattern and get paths but no go yet
[17:29:54] <bill20r3> neat!
[18:17:17] <MarkusBec_away> MarkusBec_away is now known as MarkusBec
[18:26:38] <ds3> a
[18:26:49] <eric_unterhausen> quick brown fox
[18:37:01] <archivist_attic> jumped
[18:38:39] <MattyXM> over my sphinx of quartz
[18:46:32] <danimal_garage> is that code?
[18:46:41] <danimal_garage> like blue moon rises at sunset?
[18:47:03] <danimal_garage> my washing machine is having a fit
[18:47:21] <danimal_garage> i think it's going to die
[19:12:55] <L84Supper> MattyMatt: custom cabinets "is where the money's at", interesting, in the US it's "monkey see, monkey do" you can't be too custom
[19:20:59] <ries> frallzor: seems like they cut on a Mechmate ;)
[19:21:05] <ries> frallzor: how are you doing?
[19:22:52] <frallzor> what what who
[19:23:27] <ries> frallzor: you are building a mechmate, right?
[19:23:34] <frallzor> yes
[19:23:44] <frallzor> but who cut on a mechmate? =)
[19:24:13] <ries> aAhhh sorry, the people on http://camcraftsa.com/
[19:24:16] <frallzor> its the mechmatemakers son that runs that shop if refering to that site :P
[19:24:18] <ries> I wonder who it is...
[19:25:06] <frallzor> geralds son that is =)
[19:25:19] <ries> Ahh yeaa... I did just read it :)
[19:25:37] <ries> Did you continue your rail grinding?
[19:25:45] <frallzor> nah not yet
[19:25:49] <frallzor> but got new material
[19:25:59] <frallzor> and spirit level bulbs to use =)
[19:27:01] <ries> good good...
[19:27:04] <L84Supper> http://www.techno-isel.com/CNC_Routers/index.htm sells sub-assemblies and unfortunately, windows only complete units
[19:27:16] <ries> I failed to get my transformer in the US :( so I will let some company make one here....
[19:27:21] <frallzor> next week the skeleton should be coming home too
[19:27:22] <ries> I hope to do some work this saterday or next week
[19:27:27] <frallzor> if I can clean the garage
[19:28:06] <ries> I also need to find a computer still and didn't 'dare' to buy one off the shelf in the US... first I will play with my laptop, then I will see how that goes
[19:28:24] <frallzor> probably will work nicely
[19:28:31] <frallzor> i used an old one for my first mill
[19:29:10] <ries> with the laptop I had some problems latency tests, but I didn't see any sign of mal function on my kitchen table test.... We will see
[19:29:38] <frallzor> now Ive got my vfd and spindle too :P
[19:32:11] <ries> cooool
[19:32:34] <ries> I want to buy one also, but first I need to get the system to work and see how that will owrk out
[19:32:43] <ries> When I am going to use it a lot, I will buy that spindle
[19:32:58] <frallzor> got a nice deal on a used really nice vfd so I thoughy I might aswell get a spindle
[19:33:02] <ries> a week to floriday did set me back 3.5K :s
[19:33:03] <frallzor> Äthought
[19:33:06] <ries> so I need to take it easy
[19:33:11] <frallzor> nice =)
[19:33:24] <ries> it was really nice!
[19:35:53] <ries> brb
[19:38:37] <skunkworks> Skiing tonight!
[19:54:31] <tarzan> skunkworks, falling painful?
[19:58:19] <skunkworks> yes!
[19:59:08] <skunkworks> new skis so - probably
[20:16:42] <SWPadnos> too funny. we just got a package that says "Perishable. Freeze Immediately!"
[20:17:04] <SWPadnos> it's 7 degrees outside
[20:28:22] <KimK> Hi all. John is considering adding encoders to his stepper-driven Bridgeport II, but wants to understand the benefits first. I said that encoders can continuously "DRO" the current position with no lost steps, and detect and fault on excessive following error, but cannot close the positioning loop as with servos. Is that correct?
[20:28:39] <KimK> I want to be sure I'm giving John the right information.
[20:29:09] <cradek> yes, you could use encoders to tell when position is lost, and in that case, stop with a following error
[20:30:27] <danimal_garage> i need to do that
[20:30:27] <cradek> I wonder if he could get better motors for a similar cost though... to fix the problem instead of just sensing it
[20:30:41] <danimal_garage> can you still use an index line for homing as well?
[20:30:59] <danimal_garage> assuming the encoders had an index line
[20:31:24] <cradek> I don't think you can do that currently, with steppers
[20:32:01] <cradek> well I'm changing my answer to "I don't know" :-)
[20:32:05] <danimal_garage> hmm
[20:32:25] <cradek> I think kim and john are using the ppmc cardset, which might allow homing steppers to index - I'm really not sure
[20:32:57] <danimal_garage> i know you can still configure the ini to look for an index line if you have steppers, but i dont know if it would work
[20:33:10] <danimal_garage> i have a 7i33
[20:33:17] <KimK> Well, that's something to think about, but I would give back to John the comment that he gave me earlier: "It is what it is." I'd rather see him get it to the "real nice" stage and then sell/trade it for something better. But he's kind of attached to it, so we'll see.
[20:33:38] <KimK> He's kind of swimming in big steppers right now.
[20:33:42] <cradek> danimal_garage: 7i33 is not a stepper device
[20:33:46] <danimal_garage> same here KimK
[20:33:50] <danimal_garage> i know
[20:33:55] <danimal_garage> i ment for the encoders
[20:34:18] <danimal_garage> i have one for my stepper machine for when i put on a spindle encoder
[20:34:23] <danimal_garage> and jog wheel
[20:34:42] <danimal_garage> plus i wanted the analog output
[20:36:17] <KimK> Yes, John is going to try adding the CUI capacitive encoders. I had some objections at first, but after looking over the data sheet, I relented. Does anyone have an actual EMC2 usage report on them yet?
[20:36:23] <danimal_garage> hey the x connection on the 7i33 is for the index, correct?
[20:39:01] <danimal_garage> what if i tested it without an encoder by putting some sort of momentary switch on the index input of the 7i33 and try homeing it with the index search turned on?
[20:39:25] <cradek> yes you could sure try it
[20:39:30] <KimK> John is currently using the factory high-quality roller lever arm limit switches (I forgot brand) for home switches, and getting surprising position repeatability from them. Easily better than .005", if you're lucky, maybe .002".
[20:39:30] <danimal_garage> if it stops then we know it'll work
[20:39:33] <cradek> are you using mesa stepgen or software stepgen?
[20:39:46] <danimal_garage> mesa
[20:39:56] <danimal_garage> i have a 5i20
[20:40:17] <cradek> I'm fairly sure it won't work right
[20:40:35] <danimal_garage> wont cost anything to try
[20:40:40] <danimal_garage> other than some time
[20:40:42] <cradek> nope you are right
[20:40:46] <cradek> please let us know
[20:40:50] <danimal_garage> will do
[20:41:20] <danimal_garage> i'm sure you're right as you obviously know loads more than me about this stuff, but it would be neat if it worked
[20:42:21] <danimal_garage> i forgot to set the index search line in the ini to NO and it kept homing until i stopped it, so i know at least do that
[20:42:23] <cradek> don't bet on me being right about anything that has a lot to do with steppers :-)
[20:42:48] <danimal_garage> hey you were right about my stepping issue
[20:43:11] <cradek> I don't remember it...?
[20:43:19] <danimal_garage> resonance
[20:43:42] <cradek> ah
[20:43:50] <danimal_garage> i increased my microstepping and it smoothed out and my loss of steps went away
[20:46:53] <KimK> cradek: Thanks for your advice on adding the encoders. BTW, I was reading back (to yesterday?) where you posted about having to make a fuel line decoupler(?) because the cheap plastic one they sell at the auto parts store didn't work. Neat! Maybe that's a product? ("...now try ours, no cheap plastic, and it really, really works, guaranteed!")
[20:47:39] <cradek> the competition is the $9 set - even if it doesn't work, people are going to buy that instead :-)
[20:47:54] <danimal_garage> hey this might be kinda hack, but cant you have it feet to one home/limit switch then latch onto a different switch? what if you used an index line on an encoder as the second home switch to get a super acurate home position?
[20:48:17] <cradek> I'm thinking of wiring mine to the top of the gas tank for the next sucker to find (which will probably be me of course)
[20:48:47] <danimal_garage> feed*
[20:48:57] <cradek> danimal_garage: you definitely could do that. you'd have to move slow enough that you don't miss the index pulse. it is very short.
[20:49:05] <danimal_garage> yea
[20:49:16] <cradek> danimal_garage: you could do it entirely in HAL actually, since the "home state" for each axis is a hal pin
[20:49:27] <danimal_garage> right
[20:49:29] <danimal_garage> cool
[20:49:37] <KimK> I'm not so sure. For the home mechanic, OK. But those Snap-On trucks (and now clones) keep coming around, selling high price high quality tools to the pros, so I'm pretty sure they keep selling stuff.
[20:50:11] <cradek> you are probably right
[20:50:31] <cradek> I bet there are real/good ones around, but my auto parts store just doesn't have them
[20:51:32] <cradek> another thing I could do is put on my website the ID/OD that works, and anyone with a lathe can make their own
[20:51:47] <KimK> danimal_garage: Before John said he wanted to try real encoders, I was wondering about making an optical flag to attach to the ballscrew, and then detecting that optically. Sort of a poor man's index pulse.
[20:52:15] <danimal_garage> not a bad idea
[20:52:38] <cradek> KimK: the HNC has 'coarse home' (prox on the sliding bit) and 'fine home' (prox 1ppr on the ballscrew)
[20:52:53] <cradek> also index, of course
[20:53:29] <danimal_garage> i'm thinking about my mill which has steppers
[20:53:31] <cradek> but if you set it up so both coarse and fine trigger simultaneously in only one place, you're good
[20:53:35] <KimK> cradek: Tell John and I the ID/OD instead, and we'll make them and buy you a nice dinner if we sell any, LOL
[20:53:50] <cradek> heh
[20:55:08] <danimal_garage> i actually just got my home switches hooked up and working yesterday
[20:55:11] <KimK> The HNC prox I can understand, but how do they get 1ppr fine home?
[20:55:17] <danimal_garage> my mill didnt have any at all
[20:55:56] <KimK> Oops I confused 1ppr and 1ppm, sorry
[20:56:25] <KimK> I thought that was pretty darn fine, LOL
[20:57:45] <KimK> "Yes, I'd line to buy a 250,000 line encoder... what do you mean, you don't make them?"
[20:58:25] <danimal_garage> ha
[21:05:24] <KimK> I was hoping that an EMC2 user might have some experiences to report on the CUI capacitive encoders. They have some "features". Strong magnetic fields (back of a stepper) may confuse the 1-of-8 (45 deg) hall effect detector (I think the capacitive count sequence repeats every 45 degrees, and the hall effect picks one as origin).
[21:07:05] <KimK> The quadrature count may oscillate/hunt at certain settings, but it never errors. But if you're using just the A or just the B output for RPM, say, you'll get "excessive counts" (not really, *must* use quadrature)
[21:10:17] <skunkworks> are those the ones that mariss swears by now?
[21:10:32] <skunkworks> digikey for like $20
[21:12:51] <KimK> I don't know about mariss, but yes, they're on Digikey for about $29 in singles, single-ended outputs. And they have a "line driver cable" for about $9(?) with a shrink-tube-lump about 3" down from the encoder. 5 wires to the encoder connector, 8 wires (loose) out.
[21:15:51] <KimK> They are pretty cheap for an encoder, even compared to US Digital (which I like). I guess we'll see if they're "too cheap"? What did mariss have to say about them? John and I would like to know. I'll go get a couple of relevant links for everyone, brb.
[21:17:11] <KimK> I wanted to test them for linearity before installing them, but John's dividing head only goes to 360 (1 degree), so I guess I won't.
[21:17:54] <KimK> They're supposed to be within 0.5% or so (could it have been 0.25%)?
[21:24:55] <KimK> The 102 and 103 are the same, except the connectors come off along the shaft (102), or at 90 degrees to the shaft (103).
[21:25:11] <KimK> The encoders: http://www.cui.com/adtemplate.asp?invky=333508&catky=112362&subcatky1=994182&subcatky2=997829&subcatky3=
[21:26:17] <KimK> And the driver cables: http://www.cui.com/adtemplate.asp?invky=878825&catky=112362&subcatky1=994182&subcatky2=997829&subcatky3= ***** http://www.cui.com/adtemplate.asp?invky=970739&catky=112362&subcatky1=994182&subcatky2=997829&subcatky3=
[21:29:09] <KimK> And Digikey: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1307-ND ***** http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1308-ND
[21:33:47] <KimK> skunkworks: What does mariss say about them, if these are the ones?
[21:35:47] <skunkworks> http://www.cnczone.com/forums/showpost.php?p=574704&postcount=18
[21:37:28] <SWPadnos> he likes them a lot
[21:37:38] <SWPadnos> thinks they're the best thing since sliced bread
[21:38:40] <skunkworks> well - when your drives have a +/-128 count error window... ;)
[21:39:13] <cradek> if they had a 20mm hole in the center, I'd be on it
[21:39:56] <KimK> Sorry to hear he had a bad experience with US Digital. I always order the "D" version (built-in differential drivers) and have never had any trouble with them.
[21:40:30] <KimK> I've never used US Digital's cable-driver.
[21:40:50] <cradek> I thought to get differential the only option was a thing that plugs in
[21:40:51] <KimK> cradek: Yes, sorry, looks like 8mm is the biggest.
[21:40:55] <SWPadnos> they apparently don't output a high or low enough voltage all the time, a problem which is exacerbated by long cables
[21:42:16] <skunkworks> I thought someone on the linuxcnc message board used the stepgen in velocity mode and used scales as feedback. (and pid)
[21:42:37] <skunkworks> sudo servo - just don't push it. ;)
[21:42:42] <cradek> I wouldn't be surprised if someone had done that
[21:45:00] <andypugh> Any sign of that graphical Hal-builder yet? I have 36 bits of IO to set up tomorrow night....
[21:45:18] <SWPadnos> no
[21:45:45] <skunkworks> andypugh: http://emergent.unpy.net/01200498465
[21:45:48] <andypugh> Ah well.
[21:45:48] <skunkworks> :)
[21:46:09] <skunkworks> link is dead. darn
[21:46:14] <SWPadnos> skunkworks, don't forget crapaHALic, or the eagle schematic library tomp put together
[21:46:50] <SWPadnos> (neither of which can be used to actually make HAL setups, especially if you want to include hardware drivers)
[21:47:33] <andypugh> I have hardware drivers.
[21:47:54] <SWPadnos> that's the problem :)
[21:48:12] <skunkworks> wow - this little atom board running raid with mdadm (8 esata drive bay) works suprisingly well. I wonder what will happen when I get 40 people hooked into it.
[21:48:15] <andypugh> I wasn't expecting it to have magically appeared without me noticing, and I am perfectly well aware what a non-trivial task it is.
[21:48:20] <KimK> On the US Digital models I like, the E6 (and the E5 if no room for the E6), about the "D" version it says: "For differential versions: the internal differential line driver (26C31) can source and sink 20mA at TTL levels. The recommended receiver is industry standard 26C32." http://www.usdigital.com/products/encoders/incremental/rotary/kit/e6/ and http://www.usdigital.com/products/encoders/incremental/rotary/kit/e5/
[21:52:01] <cradek> KimK: interesting, thanks
[22:01:32] <skunkworks> this is the guy
[22:01:32] <skunkworks> http://www.youtube.com/watch?v=M8ijWbg_seI
[22:03:34] <skunkworks> http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/id,286/catid,10/limit,6/limitstart,0/lang,english/
[22:03:53] <cradek> huh 'modified hostmot2 stepgen to enable home on index.'
[22:04:06] <cradek> too bad he didn't contribute that (that I saw?)
[22:06:22] <skunkworks> I thought he got it from seb or peter
[22:07:01] <cradek> hmm
[22:08:35] <skunkworks> cool
[22:08:36] <skunkworks> http://www.youtube.com/watch?v=OegjQz5mzW0
[22:09:37] <andypugh> Serious looking drives
[22:13:23] <skunkworks> http://www.youtube.com/watch?v=93xNiZ6FJUE
[22:13:53] <skunkworks> I don't know about having to wrap my part in tin foil though.. ;)
[22:15:21] <ds3> is a candle a reasonable item to use to debug (get the hang of it) single point threading with EMC on a lathe?
[22:15:56] <archivist_emc> I used air then steel :)
[22:16:06] <skunkworks> air, wood, steel. ;)
[22:16:19] <archivist_emc> pfft who needs wood
[22:16:22] <skunkworks> heh
[22:16:31] <skunkworks> thats what she said?
[22:16:46] <toastydeath> oho
[22:17:16] <archivist_emc> softer materials may break out the chuck
[22:18:55] <ds3> yes but threading tools are $$$$
[22:19:00] <archivist_emc> is he doing cnc turkey with that probing
[22:19:21] <ds3> and threading mistakes crash too easily
[22:19:43] <archivist_emc> ds3, nah grind up a bit of high speed steel as the tool
[22:20:23] <archivist_emc> watch it cut air till you are confident
[22:20:27] <ds3> archivist_emc: doesn't that mean a whole new setup each time I regrind or you got some way to grind and not have to reset w/CNC?
[22:21:03] <archivist_emc> ds3, yes use tipped tools after you know your machine
[22:21:27] <ds3> Good point
[22:22:58] <archivist_emc> wax and wood wont be easy to measure to correct your tool table
[22:23:28] <ds3> what makes wax hard to measure?
[22:24:37] <archivist_emc> soft
[22:25:15] <archivist_emc> unless you have the required touch or optical methods
[22:26:01] <ds3> oh
[22:26:26] <ds3> wax SEEMS like the ideal thing to try threading on... it is self lubricating but I have never tried to accurately measure a candle
[22:26:35] <archivist_emc> and being soft means hard to hold in the chuck
[22:26:39] <andypugh> I have seen adverts for machinable wax for exactly the proposed purpose
[22:27:18] <archivist_emc> schools use machinable wax, but accuracy is perhaps less important
[22:27:18] <ds3> that's not an issue... I hate chucks enough that they are a last resort thing; collets whenever possible
[22:29:05] <ds3> heh... school is where I learned how easy it is to crash a threading tool
[22:29:58] <archivist_emc> parting tools seem more fragile to me
[22:31:10] <ds3> parting tool fragility seems to depend a lot on the material... I have 75% change of snapping one on alum but almost 0% on steel (mild and 12L14)
[22:32:34] <andypugh> So, Aluminium gets expensive very quickly then
[22:33:10] <ds3> sort of, I just hacksaw it off and face it instead of parting
[22:33:31] <ds3> no real advantage to parting for me anyways... no stock puller and no automatic chuck/collet
[22:33:42] <archivist_emc> ally needs lube when parting
[22:34:13] <ds3> I drench it cutting oil... saturated acid brush on the stock near the top of the parting tool
[22:34:40] <archivist_emc> it also work hardens and sticks to the tool
[22:35:04] <archivist_emc> pecking to clear helps
[22:35:09] <ds3> that's what the instructor mentioned the first time I broke one
[22:35:46] <ds3> he also suggested I go in a little bit, come out, move it about 30thou to one side, go in again, come and repeat 30thou to the tother side
[22:37:36] <archivist_emc> yup that helps a lot
[22:38:13] <archivist_emc> the swarf jams the slot, making it wider allows clearance
[22:39:01] <ds3> somehow I manage to snap an insert on a Haas even after programming that in
[22:39:22] <ds3> the isntructor's comments there was something about the coolant running low
[22:41:12] <archivist_emc> a big problem is tool spring if you dont have a neutral parting tool so it rubs one side
[22:41:47] <archivist_emc> or dont set it square
[22:42:50] <ds3> is that a possible problem on a dedicated insert holder made for the Haas toolchanger?
[22:44:40] <cradek> I don't think I've ever broken a parting tool (except I crashed one into the end of the work recently)
[22:44:57] <cradek> while parting I mean
[22:45:10] <cradek> it's sure true that parts always seem to veer to one side a bit
[22:46:19] <archivist_emc> and the pip left is a pain, so you need a biased tip
[22:46:35] <cradek> I have never actually made that work
[22:46:49] <archivist_emc> upside down on a capstan we never broke one iirc
[22:46:58] <cradek> I just always assume a parted part is unfinished
[22:47:45] <archivist_emc> get the tip on center and with the right bias and parts are right first time
[22:47:59] <archivist_emc> save ages
[22:48:03] <ds3> there is also the skills need to control which side the little stub wind up and it is usually on the wrong side :(
[22:51:00] <archivist_emc> if correct you overrun and have no pips either side
[23:14:19] <MarkusBec> MarkusBec is now known as MarkusBec_away