#emc | Logs for 2007-08-18

[00:01:14] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[00:06:29] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[00:16:44] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[00:30:59] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[00:37:35] <toast_> http://cgi.ebay.com/24-x-150-MONARCH-DYNA-SHIFT-ENGINE-LATHE-24901_W0QQitemZ190122672596QQihZ009QQcategoryZ97230QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[00:37:44] <toast_> http://cgi.ebay.com/20-x-60-cc-IKEGAI-Engine-Lathe-Model-A20_W0QQitemZ110124549785QQihZ001QQcategoryZ97230QQssPageNameZWDVWQQrdZ1QQcmdZViewItem#ebayphotohosting
[00:37:49] <toast_> i would love either of those lathes
[00:38:02] <toast_> especially that monarch
[00:38:06] <toast_> dear lord.
[00:46:22] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[01:03:48] <toast_> toast_ is now known as toastydeath
[01:28:55] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[01:34:08] <jmkasunich> well, I'm pretty happy with my $5 surplus UPS
[01:34:29] <jmkasunich> we had a 30 minute power failure today, and it kept this system up
[01:34:52] <jmkasunich> I only expected 15 mins, and even that was iffy considering the unknown condition and age of the battery
[01:34:52] <toastydeath> nice
[01:45:26] <toastydeath> does anyone listen to like, thrash metal
[01:45:43] <toastydeath> i'm trying to find new bands i like since i've been listening to 80's industrial for the last two years
[01:45:51] <toastydeath> and apparently missed three hundred or so new bands
[01:46:10] <SWPadnos> darn. I seem to have missed all of them too
[01:46:24] <SWPadnos> and the ones from the '80s
[01:47:47] <jmkasunich> classic rock man!
[01:48:04] <toastydeath> put the zz top down
[01:48:11] <toastydeath> and step away from the cd player
[01:48:18] <jmkasunich> :-P
[01:48:21] <SWPadnos> heh -you'd like the triple-play concert at our fair this year: Joan Jett, Foghat, and Blue Oyster Cult
[01:48:34] <jmkasunich> ~
[01:54:40] <toastydeath> i have an affinity for music that makes one's ears bleed
[01:55:36] <SWPadnos> I have an affinity for non-blood-soaked ears
[01:55:44] <toastydeath> http://www.youtube.com/watch?v=VrvtTakbx00
[02:15:00] <jepler> darn, jlmjvm left. pluto is documented in the, er, documentation: http://linuxcnc.org/docs/devel/html/hal/drivers/#sec:Pluto-P:-generalities http://linuxcnc.org/docs/devel/html/hal/drivers/#sec:Pluto-step:-Hardware-step
[02:15:32] <toastydeath> that thing with the words
[02:46:40] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[02:50:58] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:08:37] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:10:54] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:21:24] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:31:33] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:50:52] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[03:57:01] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[04:01:07] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[04:45:02] <fenn> cradek: fixed the web poll. sorry bout that
[04:57:50] <fenn> toast: lost foam is really a much better process for casting one-off machine parts
[04:58:12] <toast> well i'm coming at it from a recyclable medium
[04:58:27] <fenn> some of the wax gets burned
[04:58:27] <toast> plus i've seen the surface finishes a person can get by using wax + ceramic casing
[04:58:40] <fenn> is it accurate though?
[04:58:43] <toast> oh yeah
[04:58:46] <toast> very accurate
[04:58:59] <fenn> those reprap things dont look like they make nice surface finish :)
[04:59:12] <toast> yeah, but it's easier to fix the surface on wax
[04:59:14] <toast> than on metal
[04:59:15] <fenn> i was thinking you could squirt out wax and then finish machine it
[04:59:23] <toast> exactly
[04:59:31] <toast> and make a larger version, to do ways
[04:59:34] <fenn> then you could cast other materials in wax molds too
[04:59:49] <fenn> like silicone or epoxy
[05:00:14] <toast> would those handle the pressure?
[05:00:25] <fenn> you dont need much pressure
[05:00:39] <toast> rockin
[05:00:52] <toast> what do you want to make out of silicone?
[05:01:08] <fenn> i think plastic injection molds use higher pressure so they can run stuff at lower temperature to get more parts faster
[05:01:14] <fenn> sex toys
[05:01:15] <toast> i was just thinking that, if i started a small business, that would almost eliminate machining for a lot of stuff
[05:01:33] <toast> oh, i've seen those molded
[05:01:38] <toast> they DO do it just like you said
[05:01:40] <fenn> what else do you make out of silicone? sheesh :)
[05:01:45] <toast> poured into an open mold
[05:02:06] <toast> plastic injection machines use high pressure because that helps melt the pellets
[05:02:18] <toast> you can use different extruder screws for different plastics
[05:02:23] <toast> like interchangable tooling
[05:02:25] <toast> on a mill
[05:02:40] <fenn> er. they dont use heating elements?
[05:02:42] <toast> from what i understand the compression does most of the melt
[05:02:51] <toast> and the heating elements just provide a little secondary assistance for the high speed units
[05:02:58] <toast> but the older ones just used screw pressure alone
[05:03:40] <toast> heating elements work great for a small flow
[05:03:56] <toast> but some of the bigger ones shoot 80-400 oz of plastic in 4-7 seconds
[05:04:03] <toast> every 4-7 seconds
[05:04:10] <toast> the heaters don't work fast enough
[05:04:17] <toast> at least, that's the theory i've had explained
[05:06:01] <fenn> that makes sense
[05:06:21] <fenn> smooshing granules generates heat inside them, whereas heating the outside means you have to wait for the heat to dissipate through the granule
[05:06:47] <toast> the question i have about that rapid prototyping
[05:06:51] <toast> is how big a wax form can a person make
[05:07:12] <fenn> well.. they make wax models of people
[05:07:18] <toast> like, could a shop cast a bridgeport with a larger machine
[05:07:24] <toast> lost wax
[05:07:34] <fenn> hmm
[05:07:51] <fenn> lost wax sculptures are usually cast in plates not more than 2 feet by 2 feet or so
[05:07:51] <toast> because being able to cast in place a lot of components for machines
[05:07:55] <toast> hmm
[05:08:00] <fenn> then they weld together and grind off the welds
[05:08:19] <fenn> it might have to do with the shrinkage difference between the metal and ceramic
[05:08:36] <fenn> solidifying metal is not very strong and can tear as it cools down
[05:08:45] <toast> yeah
[05:08:59] <toast> that would be something to investigate
[05:09:02] <fenn> when you're sand casting, the sand sorta crushes to give the metal room
[05:09:19] <toast> and even with sand casting, you could make your cores out of plastic
[05:09:25] <toast> er not cores
[05:09:27] <toast> but pattern
[05:09:34] <toast> so that would still speed the process up
[05:09:39] <toast> and i guess you could make a core box out of plastic
[05:09:50] <fenn> you could make it out of wax too :\
[05:10:03] <toast> i just kind of imagine my shop having a couple of those machines, allowing me to work on a lot more stuff at once
[05:10:23] <fenn> they are very slow right now
[05:10:27] <fenn> but that might just be the academic attitude
[05:10:49] <toast> they're trying to keep it inexpensive, right?
[05:11:07] <toast> even if it's slow, they're cheap
[05:11:10] <fenn> reprap is, fab@home seems to do whatever's easiest for them
[05:11:22] <fenn> see reprap is planning on taking over the world
[05:11:23] <toast> build a bunch of them, tune them as much as a person can
[05:11:31] <toast> and just use them as a fab farm for cores and stuff
[05:11:37] <toast> even if it's slow, you've got twenty of them doing stuff
[05:11:54] <toast> because they're cheap
[05:11:58] <fenn> sure, but i'd rather have one machine that's twenty times faster
[05:12:02] <toast> so would i
[05:12:05] <toast> no doubt, so would i
[05:12:17] <toast> but i am completely ignorant of what the limitations are of that technology
[05:12:34] <fenn> so are the people working on it
[05:12:39] <toast> hahaha
[05:12:58] <fenn> they seem to be limiting themselves to a small idea for some reason
[05:13:18] <toast> i get the sense they're very proliterat
[05:13:24] <toast> in design philosophy
[05:13:32] <fenn> yes that's all and good
[05:13:55] <toast> ha
[05:13:58] <fenn> but there's a lot more that could be done if they expanded the idea to include more than just one process
[05:14:05] <toast> like what
[05:14:08] <fenn> like, say you also had an oven, and a robot arm
[05:14:16] <fenn> or build the oven into the fabbing platform
[05:14:41] <fenn> then you could do ceramics, and melt wax, cure epoxy, etc
[05:14:51] <toast> ah, like sort of a cell manufacturing thing
[05:14:59] <fenn> yeah that's the whole point isnt it
[05:15:14] <toast> haha, for someone who wanted to do it as a part of their business, yes
[05:15:31] <toast> for someone who wants to overthrow capitalism, probably not
[05:15:31] <fenn> no i mean just to have in your basement
[05:15:39] <fenn> it has to be useful to be .. useful
[05:15:51] <toast> ..PERHAPS
[05:16:05] <fenn> you liberal arts types.. always waffling on everything :P
[05:16:37] <toast> but i would say that the integration of that into another process is easily done by hand
[05:16:45] <toast> especially at the current speed
[05:16:56] <fenn> yeah but i want my automatic robot flying jetpack
[05:16:57] <toast> like, you can't even feed one homemade oven with that thing
[05:17:02] <toast> at the current speed
[05:17:28] <fenn> i wonder if that's just because they're using slow-ass steppers with 1/4-20 threaded rod
[05:17:43] <toast> probably
[05:17:56] <toast> i'd bet money that if one of your crazy emc guys got a hold of one of those things
[05:17:58] <fenn> and silly one size fits all cam algorithms
[05:18:02] <toast> it would be stripped, have new hardware, and run at 120 ipm
[05:18:34] <toast> it would cost twice or three times as much, but it would be several orders of magnetude more productive
[05:18:43] <fenn> i dont see why it would have to cost a lot
[05:18:51] <toast> compared to what they're shooting for, i mean
[05:19:01] <toast> i think it would still be very reasonable
[05:19:01] <fenn> thats what i mean, they're thinking too small
[05:19:23] <toast> i agree
[05:19:26] <fenn> gyarr
[05:19:27] <toast> how do they do the solid
[05:19:37] <fenn> i should go build my damn motor drives instead of talking on irc
[05:19:44] <toast> like, is there 50000 lines of g-code
[05:19:44] <toast> or some equivilent
[05:19:48] <toast> that gets spit out
[05:19:49] <fenn> its.. java
[05:19:53] <fenn> * fenn shudders
[05:20:02] <toast> it just figures out the solid, and builds it from the ground up?
[05:20:09] <fenn> um, i think so
[05:20:13] <toast> this is all very mysterious.
[05:20:37] <fenn> it divides a model into layers, then zigzag's the interior and traces around the outline for each layer
[05:20:40] <toast> haha i just started to wonder, what if you had a hanging part
[05:20:49] <toast> would it try to make the hanging part in midair
[05:20:57] <toast> and just drool plastic on the floor
[05:20:59] <fenn> yes
[05:21:05] <toast> hahahahahaha
[05:21:06] <toast> that sucks
[05:21:12] <toast> well i mean. It's interesting.
[05:21:18] <fenn> someone was experimenting with making temporary supports out of cake frosting
[05:21:35] <toast> or maybe use solidworks to make a negative
[05:21:42] <toast> have it fab the negative
[05:21:50] <toast> let it cool, coat it with something like aluminum foil
[05:21:55] <toast> and then let it spit the feature out that way
[05:22:01] <toast> oh, but collisions.
[05:22:06] <toast> nevermind.
[05:22:41] <fenn> you could do that in stages
[05:23:03] <fenn> like with the slabs of wax idea
[05:23:04] <toast> or what if you had the machine build the supports as it went
[05:23:10] <toast> and just snap them off
[05:23:16] <toast> sand down.
[05:23:38] <fenn> that's what STL machines do these days
[05:23:51] <fenn> i think its kinda silly
[05:24:10] <toast> but that begs the question, would you need a 5 axis machine
[05:24:18] <fenn> no
[05:24:19] <toast> to build features from a "central body" algorithm
[05:24:48] <fenn> a what?
[05:24:54] <toast> like, say you had a cup with a handle
[05:25:12] <toast> and the algorithm build it up from the ground, but then when it hit the lower part of the cup with the handle
[05:25:18] <toast> it would go back to build the handle at that point
[05:25:39] <fenn> why do it that way?
[05:25:49] <toast> because in the case of like, lost wax casting
[05:25:52] <toast> for complex parts
[05:25:59] <toast> that has a ton of that kind of feature
[05:26:12] <toast> and it would be precarious to try and remove supports
[05:26:32] <fenn> well, you could use different kinds of supports, like water soluble or with different melting points
[05:26:37] <toast> because right now the mass production wax casting is done in injection molds, which don't have many restrictions
[05:26:52] <toast> and a 5 axis head would give the prototyper that same flexibility on one-off parts
[05:26:56] <fenn> uh actually injection molds have the same kinds of problems
[05:27:04] <fenn> undercuts
[05:27:10] <toast> slides
[05:27:20] <toast> there are like, ten part injection molds
[05:27:31] <fenn> hm
[05:27:33] <toast> *music*
[05:27:52] <fenn> bwoot oot at er ut
[05:28:10] <toast> in general they try to make injection molds only two parts plus the stripper pins
[05:28:38] <toast> but on more expensive items, like on lost wax, it sometimes makes sense to build a hella expensive mold to get around extra machining
[05:28:54] <fenn> i always thought they just glued the wax pieces together by hand
[05:29:03] <toast> ?
[05:29:17] <fenn> divide the desired pattern into parts you can make with 2 part molds
[05:29:33] <toast> they do that too
[05:29:40] <toast> depends on how automated they want to make it
[05:29:47] <toast> some lost wax stuff is lights out
[05:29:55] <toast> just like legos
[05:30:03] <fenn> mmm.. lights out foundry
[05:30:07] <fenn> i'd like to see that
[05:30:10] <toast> ha ha good pun
[05:30:15] <toast> =)
[05:30:23] <fenn> its a pun?
[05:30:28] <toast> "lights out... i
[05:30:30] <toast> 'd like to see that"
[05:30:39] <fenn> well, the metal would be glowing yellow
[05:30:46] <toast> perhaps
[05:30:57] <fenn> unless its like, zinc or some crap
[05:31:15] <fenn> nobody does lost wax with zinc do they?
[05:31:19] <toast> i don't believe so
[05:31:25] <toast> lost wax is usually ferrous
[05:31:28] <toast> in high production
[05:31:34] <toast> because it's close in quality to die casting
[05:32:03] <toast> not that you couldn't do zinc if you had some incentive
[05:32:10] <fenn> i usually think inconel turbine blades
[05:32:22] <toast> i thought those were single crystal
[05:32:26] <toast> machined
[05:32:34] <fenn> hm
[05:32:41] <toast> but i don't know for sure
[05:32:43] <toast> that's just what i thought
[05:32:46] <fenn> i know they make hobby airplane engines by lost wax
[05:32:48] <toast> ohhh
[05:33:02] <toast> i thought you meant like, concorde Olympus supersonic jet engines
[05:33:03] <fenn> big aerospace companies arent exactly broadcasting their manufacturing methods all over youtube
[05:33:15] <toast> yeah
[05:33:24] <toast> i could see some turbines being lost wax
[05:33:43] <toast> but high stress stuff is almost universally single crystal ingots
[05:33:56] <fenn> i think for single crystal blades they cast the blade in the right shape, and then there's a sort of dovetail that holds it in place on the turbine wheel
[05:34:11] <toast> can you cast single crystal?
[05:34:36] <toast> i guess if you had the right control over the cooling you might
[05:34:43] <toast> cool it from one side or something very slow
[05:35:13] <fenn> looks like it
[05:35:15] <fenn> http://www.msm.cam.ac.uk/phase-trans/2003/Superalloys/superalloys.html
[05:35:21] <fenn> about halfway down they show blades with sprues attached
[05:35:29] <toast> rockin'
[05:35:36] <toast> that is good to know
[05:35:49] <fenn> The image on the left shows a wax assembly for making the mould used in the casting of single-crystal nickel based superalloy turbine blades.
[05:35:50] <toast> dear internet: webpages only
[05:35:54] <toast> oops
[05:35:55] <toast> wrong window
[05:36:08] <fenn> heh i concur
[05:36:19] <fenn> http and ssh
[05:36:26] <fenn> no more no less
[05:36:28] <toast> i second your motion
[05:36:48] <toast> so that brings the question
[05:37:00] <toast> could, if one had the knowledge, use a reprap to get the wax form
[05:37:10] <toast> and then do a monocrystalline casting
[05:37:11] <toast> for a home turbine blade
[05:37:21] <toast> someone needs to come up with an open source oven
[05:37:31] <fenn> why that's.. that's imPOSsible!!
[05:37:43] <fenn> terrorists!!
[05:37:47] <toast> possible is the opposite of unicorns
[05:38:19] <toast> yeah i'd imagine someone in the government would get hella pissed if open source stuff for monocrystalline casting got all over the internet
[05:38:24] <fenn> oh, you have to cast turbine blades because they need intricate cooling channels routing through the inside
[05:38:32] <toast> oh
[05:38:32] <toast> good point
[05:38:43] <fenn> toast: ITAR is a motherfucker
[05:38:44] <toast> cooling? we don't need no stinking cooling
[05:38:49] <toast> itar?
[05:38:58] <toast> googlin' itar
[05:39:08] <toast> oh
[05:39:24] <toast> how bout a disclamer
[05:39:32] <toast> "nobody from outside the us view this webpage"
[05:39:35] <toast> "Ty"
[05:39:42] <fenn> basically it makes it illegal for an american to distribute information that might possibly be used to make a weapon
[05:39:55] <fenn> of course its totally ridiculous but that didnt stop anyone
[05:39:59] <toast> doesn't that mean uh
[05:40:06] <toast> anyone who ever discussed how to operate a lathe
[05:40:09] <toast> is a felon
[05:40:22] <toast> certainly line boring or deep drilling
[05:40:35] <fenn> there's a specific list of itar restrictions
[05:40:46] <toast> item number three on ITAR:
[05:40:52] <toast> "anything cool, ever"
[05:40:58] <fenn> yeap pretty much
[05:41:39] <toast> i wonder if the riprap could use a multiple-nozzle head
[05:41:44] <toast> so you could lay down a ton of stuff at once
[05:41:53] <toast> or maybe a riprap lathe
[05:41:58] <toast> vertical lathe
[05:42:12] <fenn> how would that work?
[05:42:22] <toast> which idea
[05:42:21] <fenn> just spins the table instead of moving in a circle?
[05:42:24] <toast> yep
[05:42:27] <toast> two axis
[05:42:29] <fenn> why bother
[05:42:44] <toast> i dunno, really
[05:42:52] <toast> just to see what it was better at
[05:42:57] <toast> you might get better concentricity
[05:43:09] <toast> just experementation, really
[05:43:24] <fenn> surface finish: 2000000 ra; concentricity 5 micron
[05:43:33] <toast> hahahahahahaha
[05:43:53] <toast> what if you used a Bullard VTL
[05:43:59] <toast> 56" vertial lathe
[05:44:07] <toast> put a rapid prototype head on the left arm
[05:44:15] <toast> and a smoothing tool on the right
[05:44:23] <toast> that would be so horrible.
[05:44:26] <toast> pretend i never said that
[05:44:46] <fenn> have you seen the contour crafting head?
[05:44:51] <toast> no sir?
[05:45:45] <fenn> http://www.isi.edu/craft/CC/Welcome_files/resources/media/CCmachine.wmv
[05:45:48] <fenn> that's the little one
[05:46:31] <fenn> ah dammit they hid all the other files
[05:46:32] <toast> downloadin'
[05:46:53] <fenn> well basically it has a squirter and a paddle that goes along and smooths one side
[05:46:58] <toast> oh
[05:47:09] <fenn> the idea is to squirt out concrete to make houses
[05:47:16] <toast> yeah but the bullard takes standard insert tooling and has a 56" chuck
[05:47:21] <fenn> i think it has more applications though, with reprap in mind
[05:47:50] <toast> oh hey that is neat
[05:48:14] <fenn> it stalls after 194kb for me
[05:48:20] <toast> i think i got the whole thing?
[05:48:58] <toast> i could totally see using girders
[05:49:01] <toast> to build a house with that
[05:49:14] <fenn> i think the best way is to set up 3 towers and hang a platform from them with cables
[05:49:33] <toast> would that do concrete
[05:49:36] <fenn> sure
[05:49:37] <toast> accurately?
[05:49:42] <fenn> why not?
[05:49:45] <toast> i guess it doesn't have to move THAT fast the first time
[05:49:57] <fenn> why wouldnt it be fast?
[05:50:06] <fenn> do you need more than 1g accel?
[05:50:13] <toast> well i was thinking more oscillation
[05:50:17] <toast> in the lines
[05:50:21] <toast> from moving at all
[05:50:27] <toast> with a head big enough to build a house
[05:50:28] <fenn> well, you'd want a jerk limited trajectory planner
[05:50:41] <toast> or active damping
[05:50:48] <toast> or both
[05:50:55] <fenn> there's not a lot of vibration involved in squirting out concree
[05:51:10] <toast> there is in transferring motion
[05:51:17] <toast> since the cords are stretchy
[05:51:28] <toast> lag between when the servo fires and when the thing moves
[05:51:36] <fenn> hm
[05:51:59] <fenn> with enough weight on the platform you could preload all the elasticity out of the lines
[05:52:04] <toast> i don't think that would be a problem on smaller units, but with enough cord to build a house
[05:52:09] <fenn> or most of it
[05:52:15] <toast> hmm
[05:52:19] <toast> how much weight would that be though
[05:52:26] <toast> that would be interesting if it didn't take much
[05:52:27] <fenn> and an octagonal frame for the towers
[05:52:42] <fenn> well it depends how wide the cables are?
[05:52:47] <toast> true
[05:52:57] <fenn> uh.. stress.. strain
[05:53:04] <fenn> i should really read about that stuff
[05:53:06] <toast> hahaha
[05:53:16] <toast> i'm taking statics this semester
[05:53:29] <toast> SO I'LL GET BACK TO YOU ON IT
[05:53:35] <fenn> OKAY
[05:54:05] <toast> that's gonna be so wicked if someone builds a rapid prototyper able to lay down wax on a production basis
[05:54:17] <fenn> i just think its ridiculous this guy has a picture of this huge cast iron gantry thing that's squirting out igloos on the moon
[05:54:28] <fenn> like, how are you gonna get this huge thing on the moon?
[05:54:33] <toast> link?
[05:54:39] <toast> build it there, i guess
[05:54:49] <toast> haven't you played Master of Orion 2
[05:54:49] <fenn> http://www.isi.edu/CRAFT/CC/space1.html
[05:54:53] <toast> send a colony ship
[05:55:37] <toast> yeah that's a little ridiculous
[05:56:00] <fenn> its a cute picture though
[05:56:11] <fenn> i wish i knew how to draw like that
[05:57:55] <fenn> of course the whole idea is stupid; you'd be baked by radiation if you had such thin walls
[05:58:15] <fenn> and the lava tubes on the moon are like 2 miles in diameter so why bother building above ground
[05:58:28] <toast> those are very good poitns
[05:58:28] <toast> *points
[05:58:36] <toast> and evidence of why i do not live on the moon
[05:58:58] <fenn> no it's all reagan's fault
[05:59:04] <fenn> damn you ronald reagan!
[05:59:06] <toast> well i am saying i don't know anything about living there
[05:59:10] <toast> clearly, i would be here
[05:59:09] <toast> on earth
[05:59:12] <toast> talking to you, on the moon
[06:01:21] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[06:01:28] <fenn> with a with a 2.6 second time delay
[06:02:51] <toast> that's not bad
[06:02:57] <toast> i get 3.6-4 seconds sometimes.
[06:03:04] <toast> round trip
[06:04:26] <fenn> well, 2.6 is the minimum
[06:04:58] <toast> yeah
[06:11:50] <fenn> toast: check out my hovercraft airplane-car http://fenn.dyndns.org/sketches/tadpole-plane-0.2-web.jpg
[06:12:47] <toast> haha what
[06:13:16] <toast> are you planning on building that?
[06:13:18] <fenn> the big circle around the body is the hovercraft skirt
[06:13:26] <fenn> the square box is batteries or supercapacitors
[06:13:38] <fenn> and the funny thing next to the wheel is a gas turbine engine
[06:13:45] <toast> hahahahah
[06:13:46] <toast> natch
[06:13:51] <fenn> wings bolt on the sides for long distance flights
[06:14:05] <toast> hahahaha.
[06:14:21] <fenn> i think the air ducts might be too small
[06:15:16] <toast> i have no idea man, i know very little about the CFM requirements of a small turbine engine
[06:15:20] <toast> and aerodynamics
[06:16:03] <fenn> well its more a problem of getting enough thrust at low speeds
[06:16:44] <fenn> air is only at 14.7psi :(
[06:17:07] <toast> but the aerofoils work both of the bernoulli effect and newton's second law
[06:17:28] <toast> so it's not purely a vacuum effect
[06:17:33] <fenn> at zero forward velocity the only way to get air into the engine is vacuum
[06:17:34] <toast> use ROCKETS
[06:17:39] <toast> oh
[06:17:42] <fenn> yeah i thought about rockets
[06:17:53] <fenn> but its just not practical for day to day stuff
[06:17:58] <toast> or legal
[06:18:03] <fenn> well, neither is a car-plane
[06:18:06] <fenn> legal i mean
[06:18:08] <toast> yeah
[06:18:10] <toast> not yet anyway
[06:18:11] <fenn> fuck it
[06:18:15] <toast> soon
[06:18:20] <fenn> i'll take me and my car plane to mexico
[06:18:23] <toast> hahahaha
[06:19:53] <fenn> i just posted that to a discussion about open source automobiles and nobody responded
[06:20:18] <toast> don't open source automobiles always look like they were designed by failed art students
[06:20:22] <toast> that's what i've noticed
[06:20:33] <fenn> you must be thinking of electric cars
[06:20:50] <toast> no i saw an open source car design
[06:20:52] <toast> and it looked retarded
[06:21:36] <toast> i think i've seen more than one
[06:23:47] <fenn> googling "open source automobile" doesnt turn up much
[06:23:55] <toast> in the news
[06:24:25] <fenn> oh i spelt it wrong
[06:24:57] <fenn> cool http://en.wikipedia.org/wiki/Open_source_hardware
[06:30:23] <toast> gg wikipedia
[06:32:08] <fenn> welp oscar looks alright, the others definitely do suck for sure
[06:32:28] <toast> lol openrisc
[06:32:40] <fenn> but then maybe they're just imitating the car companies
[06:33:14] <fenn> heh i didnt realize OScar is a name
[06:33:22] <fenn> fools!
[06:37:13] <toast> yeah i'd totally read that "oscar" as well
[06:43:36] <fenn> are there any other thin-kerf ways to cut sheet metal besides laser and waterjet?
[06:43:42] <fenn> like <1mm
[06:44:03] <toast> i wouldn't call either one of those thin kerf
[06:44:05] <fenn> and need a turning radius of about 2mm
[06:44:05] <toast> but wire EDM
[06:44:14] <toast> bandsaw
[06:44:22] <fenn> how thin can bandsaws go?
[06:44:32] <fenn> i mean can you cut out arbitrary shapes with a bandsaw?
[06:44:32] <toast> i'm not entirely sure, but less than 1mm
[06:44:40] <toast> by hand, yes
[06:44:48] <fenn> well prefereably cnc
[06:44:52] <toast> you're fucked.
[06:44:57] <toast> bandsaws won't do cnc
[06:45:01] <fenn> why not?
[06:45:06] <toast> the blades always bind up
[06:45:06] <toast> and snap
[06:45:17] <fenn> doesnt wire edm break all the time too?
[06:45:27] <toast> not as much as a CNC bandsaw would
[06:45:38] <toast> wire edm also does more shapes
[06:45:42] <toast> but it's sloooow
[06:45:50] <fenn> slow is no prob
[06:46:13] <fenn> i'm hoping for a process that's do-able by a reasonably competent DIY'er
[06:46:27] <toast> i hear wire edm is fairly hard to implement
[06:46:32] <fenn> waterjet is damn near impossible to make
[06:46:40] <fenn> and dangerous
[06:46:51] <toast> yeah
[06:46:55] <fenn> do you think building a laser would be easier or harder than wire edm?
[06:47:34] <fenn> i'm thinking harder because of the lenses
[06:47:43] <fenn> and the electronics is more sophisticated
[06:48:00] <fenn> EDM's are full of crap i dont understand though
[06:48:30] <toast> nervous control, dielectric fluid
[06:48:34] <toast> yeah
[06:48:39] <toast> i don't understand it either, don't feel bad
[06:48:40] <fenn> i hope you can use kerosene instead of DI water
[06:49:37] <fenn> or propylene glycol maybe
[06:50:06] <fenn> does alcohol dissolve salt?
[06:51:49] <fenn> oh maybe diesel oil would work better
[06:51:57] <fenn> er, not work better, but, not highly flammable
[06:57:02] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:01:21] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:05:37] <fenn> it's always funny when people are trying to make an absurdity of something and end up making a good point
[07:05:52] <toast> ?
[07:06:11] <fenn> the open source guns part http://members.aol.com/erichuf/Linux4.html
[07:09:12] <toast> toast is now known as toastydeath
[07:16:53] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:18:30] <fenn> omfg
[07:18:55] <fenn> "By comparison, computer programming does not require much knowledge. Computer languages consist of only a few dozen words, and there are only a few rules to remember. Computer programming is so easy to pick up that many people learn it on their own. A person who does not have an ability to sit still in a classroom can easily become a computer programmer because he does not need to sit in a class. A person who cannot read a book or use a
[07:31:05] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:33:44] <ds2> that is obvious well believed... look at all the outsourcing ;)
[07:46:37] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:50:48] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[07:57:00] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[08:07:12] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[08:11:23] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[08:16:34] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[08:47:00] <lerneaen_hydra> well someone appears to be having network issues
[08:57:18] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[09:01:14] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[09:14:38] <alex_joni> fenn: omg, that guy really is deranged
[09:21:27] <lerneaen_hydra> fenn, that must be parody/humor
[09:21:30] <lerneaen_hydra> at least I really hope it is
[09:21:46] <fenn> its not humor
[09:23:56] <Vq^> fenn: source?
[09:24:11] <lerneaen_hydra> http://www.framebox.de/creations/3d/salad/
[09:24:42] <alex_joni> http://members.aol.com/erichuf/Linux4.html
[09:25:37] <Vq^> heh, i saw that one yesterday
[09:26:51] <fenn> it was pretty painful to read it, i dont suggest that anyone do so as there isnt much of value
[09:27:16] <fenn> i only read it because i thought i might gain some new insight about how we could improve our image
[09:27:41] <alex_joni> fenn: with people like that there's nothing you could/should do
[09:28:12] <fenn> i dont even think he's a "troll" because there is just so much material
[09:29:55] <alex_joni> otoh one should keep in mind he wrote it in 2001
[09:30:18] <alex_joni> back then linux desktop wasn't as good as today
[09:30:26] <Vq^> "Defective people often end up in jobs in the government, where standards for behavior are lower than they are for businesses, and some of them end up as computer programmers (and network managers), where mental disorders are tolerated."
[09:31:06] <alex_joni> I sold a machine a year or so ago
[09:31:33] <alex_joni> it was for checking compression on a flexible link (between engine and exhaust on big diesel motors)
[09:32:04] <alex_joni> I did it with HAL and some GUI I wrote, gave them the sources with the machine aswell
[09:32:08] <alex_joni> it was quite profitable
[09:32:17] <fenn> gratz
[09:32:26] <fenn> i think i remember that
[09:32:38] <alex_joni> it also saved me some money than a proprietary version
[09:32:51] <alex_joni> I just used parport to count an encoder, etc
[09:33:14] <alex_joni> the 'alternative' would have been some 'smart' PLC with encoder input
[09:37:05] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[09:47:24] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[09:53:04] <lerneaen_hydra> alex_joni, any idea of why most people really want to have a plc with smart nodes etc etc instead of a fast central machine with stupid endpoints? (ie plc+profibus vs emc-parport like systems=
[09:53:49] <alex_joni> lerneaen_hydra: I guess some people like stuff, they pay lots of money for, better
[09:54:13] <alex_joni> like.. it must be better if it's more expensive
[09:54:35] <alex_joni> (that goes for linux vs. win, or mach vs. emc, etc)
[09:54:40] <lerneaen_hydra> right, is there any real benefit to "smart" systems?
[09:54:55] <alex_joni> lerneaen_hydra: depends on the goal
[09:55:06] <lerneaen_hydra> it would seem to me that there's just more things that can go wrong
[09:55:14] <lerneaen_hydra> (for smaller systems that is=
[09:55:15] <lerneaen_hydra> *)
[09:55:17] <alex_joni> if you want an industrial grade machine, then getting an industrial PC is also quite expensive
[09:55:24] <alex_joni> and the PLC might be a cheaper system
[09:55:50] <lerneaen_hydra> right, still, couldn't you use the PLC to drive stupid endpoints?
[09:56:07] <alex_joni> sure
[09:57:19] <lerneaen_hydra> so it's mainly industrial inbreeding that makes people want smart nodes with "neat" protocolls between them?
[09:59:31] <alex_joni> probably so
[09:59:40] <alex_joni> and the advantage of a digital connection
[09:59:52] <alex_joni> vs. an analog speed prescription for example
[10:00:09] <alex_joni> which really makes a difference at > 10m
[10:00:22] <alex_joni> fenn: still around?
[10:00:22] <lerneaen_hydra> oh, I was thinking along the lines of say passing an encoder signal over to a controller (PLC/PC)
[10:00:43] <lerneaen_hydra> wrt either quadrature vs. canbus/I2C etc
[10:00:51] <alex_joni> I rearely have seen encoders connected to a bus
[10:00:57] <alex_joni> but servos.. all the time
[10:01:47] <lerneaen_hydra> hmm that's true
[10:02:07] <alex_joni> that's different for absolute encoders or glass scales
[10:02:35] <alex_joni> the ones I worked with are rs4xx, newer ones might be *bus
[10:09:58] <fenn> alex_joni: ya
[10:11:10] <fenn> mach doesnt cost a lot so there isnt that illusion of "better"ness
[10:11:48] <fenn> lerneaen_hydra: often there's a cost advantage to modular construction
[10:11:55] <fenn> and troubleshooting is easier
[10:12:21] <fenn> it's unfair to compare PLC's with commodity PC hardware.. try comparing with a huge complex industrial control center
[10:12:31] <fenn> there's some acronym but i forget
[10:12:40] <lerneaen_hydra> right, for larger systems I can certainly beleive it's more effective
[10:13:14] <fenn> and if you're already familiar with <favorite PLC> then you save a lot of development effort
[10:13:25] <fenn> by using it for small things
[10:13:33] <lerneaen_hydra> but people I know want canbus/modular systems controlling digital I/O with something like 5 endpoints
[10:13:41] <alex_joni> fenn: was meaning to ask you something
[10:13:43] <lerneaen_hydra> true
[10:13:45] <fenn> i mean, alex could have used a $1.50 microcontroller which could have done the job better
[10:14:04] <alex_joni> fenn: probably.. but this was cheaper as it was the customers PC
[10:14:07] <alex_joni> :P
[10:14:12] <alex_joni> well.. 'cheaper'
[10:14:20] <fenn> guess i should read the whole backlog before running my mouth
[10:14:27] <alex_joni> anyways, different topic
[10:14:38] <fenn> * fenn clears pipelines
[10:14:42] <alex_joni> I'm trying to implement reading some inputs from g-code
[10:14:58] <alex_joni> and there's some problems I'm facing
[10:15:01] <fenn> like, probe xyz and then draw a circle around it?
[10:15:10] <alex_joni> no, like wait for an input
[10:15:25] <alex_joni> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?IOCapabilities
[10:15:40] <alex_joni> or read an input into a variable
[10:15:56] <fenn> there should be a hal pin like with probing
[10:16:06] <alex_joni> the hal pins is the easy part
[10:16:22] <alex_joni> and getting the stuff from motion to canon is also quite easy for me
[10:16:34] <alex_joni> the real problem is the interpreter reading ahead
[10:16:44] <fenn> why is that a problem?
[10:17:09] <alex_joni> lets say one codes a read input into var #5678 and the next line will be a o-word if-branch based on that value
[10:17:18] <fenn> ah
[10:17:40] <alex_joni> the interpreter reads the input when it's interpreting, not when actually running
[10:17:55] <fenn> does the interpreter ever wait?
[10:18:01] <alex_joni> so the value of the input might be bogus, if it's something affected by one of the earlier g-codes
[10:18:05] <alex_joni> fenn: not so far :/
[10:18:26] <alex_joni> except the queue is full, then it stops interpreting
[10:19:23] <alex_joni> I've been giving this quite some thought, and the only possible solution right now would be to cause interp stop interpreting until that line has been executed
[10:19:36] <fenn> yes
[10:20:00] <fenn> otherwise you'd have to clear everything in the queue which is messy
[10:20:03] <alex_joni> the alternative is implement o-words in CANON too
[10:20:07] <fenn> um..
[10:20:07] <alex_joni> fenn: yes
[10:20:15] <fenn> how would that work?
[10:20:36] <alex_joni> well.. it would just interpret in the interp, then pass all possibilities to the executor
[10:20:43] <alex_joni> which would do the branching at the appropiate time
[10:20:52] <fenn> wouldnt i be just a conditional (if) command "tag"?
[10:21:06] <alex_joni> sure.. but with both branches passed to canon
[10:21:18] <alex_joni> then canon uses only one of them
[10:21:28] <fenn> ok, that's not really o words
[10:21:47] <alex_joni> well.. it's an 'if"
[10:22:05] <alex_joni> it's just like with any compiler
[10:22:18] <fenn> really its a branch statement
[10:22:24] <alex_joni> yes
[10:22:30] <fenn> analogy here being that CANON is like assembly
[10:22:38] <alex_joni> exactly
[10:23:04] <fenn> ok so "branch if #5678 = 1"
[10:23:20] <alex_joni> yeah, but #5678 is only in interp
[10:23:26] <alex_joni> canon has no idea about that
[10:23:28] <fenn> ew
[10:23:29] <alex_joni> :/
[10:23:48] <fenn> i thought the reason for stupid registers was so that simple code like realtime stuff could handle it
[10:24:02] <alex_joni> stupid registers?
[10:24:16] <fenn> #5678 is like a pointer, or a poor imitation of one
[10:24:36] <fenn> an int pointer
[10:24:58] <fenn> i havent looked at any of that code to see what it really does, that's just my impression of it
[10:25:19] <alex_joni> well.. not very far from that
[10:26:33] <alex_joni> so.. to summarize
[10:26:49] <alex_joni> we have an interpreter which can only serialize stuff apriori
[10:27:05] <alex_joni> and an executor (canon + task) which is quite simple/dumb
[10:27:28] <fenn> and it doesnt allocate new memory
[10:27:41] <alex_joni> right, and it discards already executed stuff
[10:27:46] <alex_joni> so no looping as it is now
[10:28:25] <fenn> is there any good reason not to allocate new memory?
[10:28:36] <alex_joni> don't think so
[10:28:44] <fenn> not to get sidetracked or anything but if you have 50 different execution paths..
[10:29:00] <fenn> hmm
[10:29:03] <fenn> actually nevermind
[10:29:19] <fenn> there is a good reason not to allocate new memory, and that's so we dont have to change stuff
[10:29:23] <alex_joni> you still have them serial in a linked list
[10:29:38] <alex_joni> just the possibility to jump from one position to another
[10:30:13] <alex_joni> the problem is deciding when to start throwing things away
[10:30:23] <fenn> how do you jump from a to b? is there a pointer in one element of the list?
[10:30:34] <alex_joni> but I guess o-words provides some help for that
[10:30:53] <alex_joni> fenn: probably you have a canon call : jump to item 54378
[10:31:05] <fenn> er, but if its a linked list you arent sure where that is really
[10:31:06] <alex_joni> which can be a call, or a loop
[10:31:23] <alex_joni> fenn: then it must be changed to a numbered list
[10:31:39] <fenn> ok
[10:31:43] <alex_joni> or you have a diff. table with pointers to interesting branchpoints
[10:32:09] <alex_joni> because you know location 2345 can be a place to jump to from later on, when you encounter the o100 in the g-code
[10:32:09] <fenn> where's the list structure? i'm looking at rs274_pre.cc right now
[10:32:29] <alex_joni> I know it's interp_list in emccanon.cc
[10:32:42] <alex_joni> interpl.hh
[10:33:15] <fenn> gah i wish find had more standard syntax
[10:33:30] <fenn> oh n/m i can use grep the way i want
[10:33:39] <alex_joni> it's in src/emc/nml_intf
[10:34:08] <alex_joni> class LinkedList * linked_list_ptr;
[10:36:16] <fenn> long long?
[10:36:20] <fenn> wonder what that's for
[10:37:42] <fenn> ok so add a command number to NML_INTERP_LIST_NODE
[10:37:42] <alex_joni> beats me
[10:38:12] <alex_joni> and add a couple of methods to set the current pointer
[10:38:35] <alex_joni> the thing that troubles me though is that get() seems to discards old nodes
[10:38:45] <fenn> int set_line_number(int line);
[10:38:45] <fenn> int get_line_number();
[10:39:13] <alex_joni> get() needs a rewrite
[10:39:34] <alex_joni> parse linked list, search for line_number == node.line_number
[10:39:52] <fenn> you mean get_line_number()?
[10:39:57] <alex_joni> although this might be a wrong approach
[10:40:07] <alex_joni> let's think about loops for a while
[10:40:16] <alex_joni> do .. while #5367 < 10
[10:40:59] <fenn> uh
[10:41:08] <fenn> i think its better to stick to the assembly analogy
[10:41:20] <alex_joni> you have registers in assembly
[10:41:30] <alex_joni> that's how you do loops and conditions
[10:41:41] <fenn> and have the interp output a series of commands that checks 53467 and if some condition is not met it goto's somewhere
[10:42:44] <fenn> you wouldn't have a 'while' command it would either fall through the branch point or goto somewhere
[10:42:56] <alex_joni> yes.
[10:43:03] <fenn> and to do a loop you would have a goto start of the loop after the branch
[10:43:03] <alex_joni> so to make it work you need:
[10:43:06] <alex_joni> goto
[10:43:12] <alex_joni> check value
[10:43:15] <alex_joni> change value
[10:43:26] <alex_joni> if value then goto
[10:43:39] <fenn> what's check value? get the number from HAL or something?
[10:43:54] <alex_joni> I think I meant the last one
[10:43:58] <alex_joni> if value then goto
[10:44:03] <fenn> ok
[10:44:15] <fenn> no i/o in canon then? :P
[10:44:17] <alex_joni> sure
[10:44:27] <alex_joni> change value with something from io
[10:44:42] <alex_joni> this is quite a big change
[10:45:15] <fenn> our little turing machine is all grown up *sniff*
[10:45:34] <alex_joni> wonder when jepler wakes up to give us his 2 cents ;)
[10:45:41] <alex_joni> fenn: haha :P
[10:46:01] <fenn> hey what harm could another ad-hoc programming language be
[10:46:17] <alex_joni> there are already a couple hundred
[10:46:22] <fenn> i mean in emc
[10:46:22] <alex_joni> so another one is no harm :))
[10:46:38] <alex_joni> didn't get that..
[10:46:51] <fenn> well, i'm not a big fan of o-words if you havent noticed
[10:46:59] <fenn> and neither is jepler
[10:47:19] <alex_joni> I'm not a big fan of g-code ;)
[10:47:24] <fenn> so i was extrapolating his response based on that, and then making light of it for purposes of fine sarcasm
[10:47:36] <fenn> it might have been hard to follow
[10:47:46] <alex_joni> a bit
[10:48:16] <fenn> hmmm
[10:48:37] <alex_joni> anyways.. I think this change might wait for after 2.2
[10:48:46] <fenn> ya
[10:48:57] <fenn> you could branch it if you wanted to start right away
[10:49:23] <fenn> btw this bugs me:
[10:49:25] <fenn> "using a gcode subset is better than using canon, because for instance using canon means you have to account for tool length, coordinate system offsets, and (worst!) cutter radius compensation yourself"
[10:50:21] <fenn> but you cant use gcode if you want to call some macro halfway through a part program
[10:50:39] <fenn> so that means you'd have to write a new set of routines that do tool length, coords etc
[10:50:57] <fenn> which isnt such a bad thing in itself but then you'd have to modify the interpreter..
[10:51:14] <fenn> um, so i think maybe there should be another layer between interp and canon
[10:52:01] <fenn> actually, split interp into "gcode specific stuff" and "generic high level machine utilities"
[10:52:10] <alex_joni> hmm
[10:52:30] <fenn> then writing a new language parser should be a lot easier
[10:52:34] <fenn> for LOGO or whatever
[10:52:58] <fenn> actually step-nc is the only real contender i think
[10:56:05] <fenn> oh how could i forget my python shell
[10:56:17] <alex_joni> right
[10:56:43] <fenn> what's the opposite of "canonical" as used here?
[10:57:33] <alex_joni> what do you mean?
[10:57:57] <fenn> hehe
[10:58:02] <fenn> canonical definition: accepted
[10:58:12] <fenn> Antonyms: unacceptable, unauthorized,
[10:58:23] <fenn> that's the wrong definition
[10:58:32] <fenn> they use canonical as "simplest possible thing that could work"
[10:59:52] <fenn> i had no idea there were so many definitions for canonical
[11:01:13] <alex_joni> heh
[11:01:55] <alex_joni> lerman: can you please read back on this discussion with fenn, and let us know what you think?
[11:03:18] <fenn> hey it looks like ron ginger is thinking the same thing
[11:03:23] <fenn> is that what prompted this?
[11:04:38] <alex_joni> yes and no
[11:04:43] <alex_joni> I started it a bit earlier
[11:05:19] <fenn> must be the global mind control satellites then
[11:05:32] <alex_joni> haha
[11:07:16] <alex_joni> fenn: why aren't you in #emc-devel ?
[11:07:22] <fenn> i'm not cool enuf
[11:07:26] <alex_joni> ha :P
[11:07:36] <alex_joni> this would have been better in there
[11:32:34] <fenn> can we kick a-l-p-h-a?
[11:34:47] <alex_joni> :)
[11:34:50] <alex_joni> be nice
[11:35:23] <Vq^> alex_joni: give him access to a more stable machine for the irc-client?
[11:42:54] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[12:19:10] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[12:21:07] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[12:27:15] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[12:31:45] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[12:50:55] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[13:09:44] <jlmjvm> good morning all
[13:13:43] <Vq^> hello mr Jvm
[13:17:57] <alex_joni> yo
[13:19:14] <jlmjvm> when is 2.2 supposed to be out,and will it have pluto_step
[13:19:44] <alex_joni> pluto_step is not 100% finished
[13:19:55] <jlmjvm> darn
[13:20:00] <alex_joni> and 2.2 is probably at least a month away
[13:20:17] <alex_joni> I think jeff said he has some version working, but it needs more testing (pluto_step)
[13:20:55] <jlmjvm> im very interested in it
[13:21:09] <jlmjvm> just found out last nite
[13:21:09] <alex_joni> jepler is around on and off.. just ask him
[13:21:52] <jlmjvm> cool
[13:22:45] <jlmjvm> i like the price of the pluto
[13:26:27] <jlmjvm> jepler: r u here
[13:29:15] <jepler> jlmjvm: yeah, for the moment
[13:29:53] <jepler> jlmjvm: did you find the documentation for pluto-step yet? http://linuxcnc.org/docs/devel/html/hal/drivers/#sec:Pluto-P:-generalities http://linuxcnc.org/docs/devel/html/hal/drivers/#sec:Pluto-step:-Hardware-step
[13:30:11] <jlmjvm> you read my mind
[13:31:31] <jepler> pluto-step is very new and probably buggy. the feature set is not guaranteed to remain the same as now (4x 300kHz stepgen + misc digital I/O).
[13:32:26] <jlmjvm> do u use a pluto
[13:32:42] <jepler> I don't have a mill
[13:33:08] <fenn> there is a piece of plastic with your name on it
[13:34:15] <jlmjvm> im interested,like the price,and have a mill
[13:34:40] <lerneaen_hydra> jepler, what are you driving off the pluto?
[13:34:44] <lerneaen_hydra> loose steppers?
[13:34:52] <jepler> lerneaen_hydra: an oscilloscope
[13:34:57] <lerneaen_hydra> oh, ok
[13:35:08] <fenn> that's pretty funny actually
[13:35:13] <lerneaen_hydra> didn't you borrow cradek's servo lathe?
[13:35:25] <lerneaen_hydra> though that won't be useful with pluto
[13:35:33] <jepler> lerneaen_hydra: cradek's lathe is run using the older, more tested pluto-servo configuration
[13:35:40] <lerneaen_hydra> oh, right
[13:35:52] <jepler> for a time I had a borrowed sherline lathe with steppers, but I hadn't worked on pluto-step yet
[13:37:58] <jlmjvm> are you the person developing pluto_step
[13:38:03] <jepler> and I will soon have a step+direction engraving machine so at that point it'll be important to me that pluto-step works properly
[13:38:22] <jepler> jlmjvm: knjn.com develops/sells the pluto-p board (no connection to me)
[13:38:35] <jepler> jlmjvm: but I'm the developer of the pluto_step and pluto_servo firmwares for it
[13:39:38] <lerneaen_hydra> jepler, something akin to skunkworks' engraver?
[13:40:32] <jepler> lerneaen_hydra: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=190141490742
[13:40:57] <lerneaen_hydra> oh, neat
[13:41:00] <lerneaen_hydra> not that expensive
[13:41:17] <jepler> on paper it sounds like what I want
[13:41:23] <jepler> we'll see how reality measures up
[13:41:32] <lerneaen_hydra> heh, exactly
[13:42:00] <lerneaen_hydra> I'm a bit sceptical to using plastic as the frame material
[13:42:09] <lerneaen_hydra> modulus of elasticity and all
[13:42:20] <fenn> should be fine for pcb's
[13:42:32] <jepler> yeah my main goal is to do pcbs
[13:42:41] <lerneaen_hydra> oh, right
[13:43:07] <jepler> repeatability, backlash, and axis alignment are all going to be important though
[13:43:20] <fenn> axis alignment?
[13:43:34] <jepler> I don't think that's quite the word I'm looking for
[13:43:43] <fenn> unless you're doing double sided pcb's i dont see why squareness would be an issue
[13:44:01] <jepler> what I mean is, if there's even a very small rise in the table over the width of a pcb it will make for bad trace isolation milling
[13:44:02] <fenn> and even then, if you flip it over the right way it shouldnt matter
[13:44:11] <fenn> ah, flatness
[13:44:38] <fenn> you can skim off the table i presume
[13:46:38] <jepler> if it actually goes 70ipm and I use 8x microstepping, I'll be just a bit out of emc's software step generation range (37.3kHz step rate)
[13:47:19] <fenn> i noticed steppers didnt work so well if you make them go very fast
[13:48:12] <jepler> yeah, torque drops off as velocity increases -- but then all that is moving is a few ounces of plastic
[13:50:52] <jlmjvm> how do you attach your signals to a pluto,what else do you need
[13:51:59] <jepler> jlmjvm: the "ACEX1K datasheet" I allude to in the documentation is http://www.altera.com/literature/ds/acex.pdf
[13:52:06] <jepler> jlmjvm: that document gives information about drive strength of outputs, etc
[13:53:54] <jepler> jlmjvm: For pluto_servo, chris and I found that it could reliably accept feedback from 5V encoders and control the 5V logic of an L298 with its outputs
[13:54:16] <jepler> the highest voltages involved are around 40V and there is no optoisolation in that system
[13:54:41] <jepler> see page 46 of acex.pdf for information about output voltages & currents
[13:55:06] <jepler> use the entries for Vccio = 3.0v
[13:55:06] <jlmjvm> just found out last week my breakout board isnt optoisolated
[13:55:22] <lerneaen_hydra> jlmjvm, the hard way?
[13:55:56] <jlmjvm> no,just by ereading specs
[13:56:01] <lerneaen_hydra> oh, right
[13:56:07] <jlmjvm> reading,lol
[13:56:31] <jlmjvm> i guess that is the hard way according to my spelling
[13:58:05] <jlmjvm> jepler: what i meant was what do you need to actually attach signals to the pluto
[13:58:23] <fenn> its just terminal strips i think
[13:58:41] <fenn> nope
[13:59:00] <fenn> apparently you have to solder your own connectors on
[13:59:09] <jlmjvm> ribbon cable to a breakout?
[13:59:23] <fenn> yeah
[13:59:24] <jepler> jlmjvm: the board has a series of 2-row .100" spacing headers
[13:59:33] <jepler> one of them is presoldered with a connector, the rest are bare
[13:59:48] <jepler> I haven't come up with any better ideas than ribbon cable to breakout
[14:00:31] <jlmjvm> so that will work,ribbon to breakout?
[14:00:32] <jepler> actually, my plan is to make a circuit board that adapts the pluto_step "main connector" pinout to the xylotex pinout, but that's pretty specific to my needs
[14:01:26] <fenn> jepler: you can get IDC DB-25 connectors
[14:01:31] <fenn> so you'd just have a ribbon cable
[14:01:35] <fenn> and do the pinout in software
[14:02:11] <jepler> fenn: unfortunately, no
[14:02:32] <fenn> http://www.futurlec.com/Connectors/DSUBIDCF25.shtml
[14:03:00] <skunkworks> fenn: did you see the pictures?
[14:03:02] <jepler> fenn: the pluto_step main connector has GND, VCC and dedicated-input pins in places that are incompatible with the layout of parport signals on a 26-pin connector
[14:03:09] <fenn> skunkworks: yeah pretty neat
[14:03:31] <skunkworks> those where made from plans
[14:03:53] <fenn> jepler: sure you arent just looking at it upside down? :)
[14:07:31] <jepler> oh probably -- I overlook the simple solution to problems everyday
[14:08:03] <fenn> or you could pull the ribbon cable apart and swap the wires around, then clamp it together with the idc connector
[14:08:15] <fenn> but by that point you might as well make a pcb
[14:49:00] <lerman> alex_joni: fenn: Are you still here?
[14:51:07] <lerman> (or not)
[14:51:51] <lerman> I read alex_joni's request that I read back your discussion.
[14:52:16] <lerman> My conclusion is that there is a straightforward way to accomplish what you would like.
[14:53:00] <lerman> But the margin is too small to include it here. :-)
[14:53:17] <lerman> I'll "see" you later.
[15:30:40] <gene> hey guys, good morning, whats left of it
[15:31:54] <gene> hello, anybody here?
[15:33:12] <DanielFalck> hi gene
[15:36:53] <gene> Hi Daniel, qquestion about sim_inch.ini for test purposes
[15:37:41] <DanielFalck> what's your question? (I might not be able to help, but someone surely can)
[15:38:05] <gene> I need to (probably copy/paste) for ABC so I can test Ian Wrights code which uses an A axis too
[15:38:24] <gene> what 'type' should I change those tto?
[15:39:15] <gene> This is the thread I've been trying to help on on the mailing list
[15:39:26] <DanielFalck> as in 'rotary or linear' ?
[15:40:09] <gene> I'd assume rotary, anything else to change for the simulation?
[15:40:36] <DanielFalck> I don't think you have to change anything else
[15:40:57] <gene> Ok, BBL after I've tried that. Thanks
[15:40:58] <DanielFalck> I'll look at a simulated 'Axis' here
[15:41:30] <gene> I'll check back on this screen occasionally
[15:45:38] <gene> That seems to have worked, axis is showing 6 axis's now
[15:45:51] <gene> Now to load Ian's code...
[15:47:30] <gene> I'd made one change in the o100 block, and now the load progress bar goes to 100% instead of 40, but its still hangs.
[15:48:14] <gene> with, to quote Ian, the budsy doofer :)
[15:48:24] <gene> busy
[15:48:49] <DanielFalck> I can't seem to get 'B' axis to show movement on my screen
[15:48:57] <DanielFalck> 'A' works fine though
[15:49:39] <gene> my shop keyboard is troo high and still gets full of swarf... Lemme kill it I see if I can get motion on the restart
[15:52:12] <gene> Neither can I. Never having run the sim, I was surprised to hear the keyboard beeper simulate the stepper noises!
[15:52:30] <gene> No motion on A, B , or C here.
[15:54:20] <DanielFalck> I'm going back into my config files to try and remember what I did to get where I am. I did a *.ini file and an *.hal file
[15:55:32] <gene> Here is the first piece I pasted:
[15:55:39] <gene> #+ fourth axis
[15:55:42] <gene> [AXIS_3]
[15:55:46] <gene> TYPE = ROTARY
[15:55:50] <gene> HOME = 0.0
[15:55:53] <gene> MAX_VELOCITY = 1.2
[15:55:58] <gene> MAX_ACCELERATION = 20.0
[15:56:01] <gene> STEPGEN_MAXVEL = 1.4
[15:56:06] <gene> STEPGEN_MAXACCEL = 21.0
[15:56:09] <gene> BACKLASH = 0.000
[15:56:15] <gene> INPUT_SCALE = 4000
[15:56:17] <gene> OUTPUT_SCALE = 1.000
[15:56:21] <gene> MIN_LIMIT = -2.0
[15:56:25] <gene> MAX_LIMIT = 4.0
[15:56:30] <gene> FERROR = 0.050
[15:56:33] <gene> MIN_FERROR = 0.010
[15:56:37] <gene> HOME_OFFSET = 0.0
[15:56:41] <gene> HOME_SEARCH_VEL = 0.0
[15:56:45] <gene> HOME_LATCH_VEL = 0.0
[15:56:45] <gene> HOME_USE_INDEX = NO
[15:56:45] <gene> HOME_IGNORE_LIMITS = NO
[15:56:47] <gene> That S/B A.
[15:56:55] <gene> major problem, empty coffee cup, brb
[15:57:08] <DanielFalck> in the 'TYPE=' section try 'ANGULAR'
[15:58:20] <gene> ok, will do
[15:58:40] <cradek> 1.2 degrees/second is pretty darn slow
[15:58:51] <cradek> you're better off starting with a working 4 or 6 axis config
[15:59:10] <cradek> the numbers for angulars are very different, because they're in degrees
[15:59:31] <DanielFalck> how about 90 then?
[15:59:49] <cradek> http://cvs.linuxcnc.org/cvs/emc2/configs/sim/axis_9axis.ini?rev=1.3
[15:59:57] <cradek> if you're running trunk there is a 9 axis sim config
[16:00:02] <DanielFalck> ok thanks
[16:00:12] <cradek> if you're adding axes, don't forget to do the hal part too (loopback the positions etc)
[16:00:51] <DanielFalck> I think that's were I screwed up. I did a hal file but am just fumbling through it
[16:00:49] <gene> 2.1.7 Chris
[16:01:16] <gene> aha, where's that?
[16:01:42] <DanielFalck> in the 'configs' directory
[16:02:08] <gene> I can get rid of the 5th & 6th I guess, Ian's code doesn't need them.
[16:02:28] <gene> brb, see what I can do to hal
[16:05:48] <gene> Ah, I just noted AXES was still set to 3, fixed that and the COORDINATES list, but I don't see a sim_hal?
[16:06:53] <gene> Humm, still no A motion, and BC are still there????
[16:10:20] <gene> Ok fixed that, forgot to save. suggestions for sim_pinout.hal to add another axis?
[16:11:00] <gene> amy old unused pin since its a sim anyway?
[16:11:07] <gene> amy/any
[16:14:13] <gene> Can I use speaker.0.pin.08 and 09.outs for Astep and Adir?
[16:17:20] <DanielFalck> I need to shut down here. my cpu fan is dead.
[16:17:38] <gene> Not getting anywhere, those pins are not found
[16:20:34] <gene> Where is he hal manpage Chris? I've tried pins up to 12 now, but emc fails to init.
[16:27:48] <gene> Chris, my printed manual, where hal starts on page 43, never mentions these speaker.0.pin-n-out's at all.
[16:29:37] <gene> How do I 'hook' it up for simulated A motion?
[16:29:46] <jepler> gene: so use the tools available to you to find out what they are
[16:30:10] <jepler> $ halrun
[16:30:10] <jepler> halcmd: loadrt hal_speaker
[16:30:10] <jepler> halcmd: show pin
[16:30:10] <jepler> Component Pins:
[16:30:10] <jepler> Owner Type Dir Value Name
[16:30:12] <jepler> 03 bit IN FALSE speaker.0.pin-00-out
[16:30:15] <jepler> ...
[16:30:45] <jepler> you'll see that there are pins numbered 00 through 07
[16:30:57] <jepler> bbl
[16:43:02] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[16:47:11] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[16:49:37] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[16:49:40] <gene> Humm, which are already in use according to the ssim_hal file
[16:50:29] <gene> I rearranged pin 1 for Adir, and pin 0 for Astep, now it can't find Astep??
[16:56:55] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[17:02:27] <jepler> troubleshoot. be methodical. remember that you can insert "show" commands right in your .hal files to see the state of things before stuff goes wrong.
[17:11:37] <gene> well, I'm just trying to see if I can help the watchmaker from that thread last week.
[17:13:08] <gene> The simulation doesn't seem to be mentioned in my printout and it did seem to be a good way to exersize his code.
[17:15:11] <gene> So trying to add a 4th axs to the sim was the idea. It appears that the 4th axis is moving, but at a truly glacial pace, with lots of following errors, about .008 degrees so far, but making the sim beep the speaker, or show this motion isn't possible I guess.
[17:25:05] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[17:31:30] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[17:39:01] <jepler> stepper-xyza would be another starting point to consider
[17:46:42] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[18:11:01] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[18:35:17] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[18:41:26] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[19:06:42] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[19:11:06] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[19:14:58] <lerneaen_hydra> a-l-p-h-a sure seems to be indesicive today
[19:22:27] <toast> lol
[19:22:41] <toast> my internet knows how his internet feels
[19:31:12] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[19:37:29] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[19:49:51] <Skullworks-PGA1> http://www.yahoo.com/s/652565 Girl gets her i-phone bill from AT&T.
[19:51:08] <anonimasu> heh
[19:51:13] <Ziegler> you all see that company that cloned the i-phone?
[19:51:18] <Ziegler> made it bette
[19:51:20] <Ziegler> better
[19:51:52] <Ziegler> http://www.popsci.com/popsci/technology/e7e48a137b144110vgnvcm1000004eecbccdrcrd.html
[19:52:15] <Ziegler> the miniOne
[19:52:22] <Skullworks-PGA1> after working for Apple, I will never trust any device they make, or sell.
[19:53:24] <Ziegler> Then that Chinese company is right up your alley!
[19:54:48] <Skullworks-PGA1> * Skullworks-PGA1 dropped is hardline dialtone in 2000 - and dropped cell service in 2005
[19:55:20] <Skullworks-PGA1> I am pleasantly disconnected phone wise - except my skype #
[19:56:51] <Ziegler> I have no cell
[19:57:00] <Ziegler> but I still have land line
[19:57:43] <Skullworks-PGA1> I used to have 2 hardlines - 1 was a 800#
[19:58:29] <Skullworks-PGA1> then some scam artist started selling forged prepaid phone cards with my 800# on it
[19:58:45] <Ziegler> hehe
[19:58:50] <Ziegler> bet that was intersting
[19:59:09] <Skullworks-PGA1> phone was ringing off the hook with people complaining there calls were not going thru
[19:59:24] <Skullworks-PGA1> so I dropped that
[19:59:28] <Ziegler> I would have hooked up a pay service
[20:00:17] <Skullworks-PGA1> later I figured I was paying 3x more in taxes than for actual use - so I dropped that and used only the cell
[20:00:28] <Ziegler> "dial 123 to begin speaking to an operator at $2 a min"
[20:02:41] <Skullworks-PGA1> I use skype now which usually works great - allthough it HATES hold music
[20:03:44] <Skullworks-PGA1> music is continuous - no time for the buffers to catch up on the on-the-fly compression.
[20:05:56] <Skullworks-PGA1> Back when I had the 800# I had my own biz - my phone needs are much simpler now.
[20:16:52] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[20:26:45] <anonimasu> *yawns*
[20:27:02] <toast> aww yeah
[20:27:05] <toast> hot anonimasu yawning action
[20:29:31] <anonimasu> yeah
[20:29:45] <anonimasu> what's up with you?
[20:34:32] <toast> chillin
[20:36:57] <anonimasu> ok
[20:37:01] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[20:37:02] <anonimasu> made anything cool recently?
[20:37:37] <toast> no, actually =(
[20:37:43] <toast> i finished the 18" thrust bearings
[20:37:48] <toast> and now i'm back to pressing "go" on a cnc
[20:37:56] <toast> until something else neat comes into the shop
[20:38:03] <anonimasu> hehe ok
[20:38:04] <toast> then i'll go do that, and it will be better again
[20:38:13] <toast> on vacation now though
[20:38:14] <anonimasu> :/
[20:38:28] <anonimasu> I have 80 parts comming up soon for a run
[20:38:34] <toast> what's the cycle time
[20:38:48] <anonimasu> I dont know maybe 15 minutes
[20:38:59] <toast> yeah, i'm okay with 15 minute runs
[20:39:06] <toast> this 1.5 minute crap
[20:39:08] <toast> is what bothers me a lot
[20:39:14] <anonimasu> though there's lots of toolchanges..
[20:39:15] <toast> freakin' set it up on a the bar fed lathe, damnit
[20:39:17] <anonimasu> like 5 per run..
[20:39:27] <toast> oh, you don't have a toolchanger
[20:39:28] <anonimasu> manual..
[20:39:29] <anonimasu> no
[20:39:30] <anonimasu> not yet
[20:39:35] <anonimasu> but I'll run 10 parts at a time..
[20:39:38] <anonimasu> or 20
[20:39:56] <toast> yeah
[20:40:17] <anonimasu> last time I made 40 we made one and one..
[20:40:18] <anonimasu> it was a pain
[20:40:56] <toast> yeah, that would suck
[20:42:15] <anonimasu> :)
[20:42:51] <anonimasu> I need to design a toolchange/machine parts for one
[20:43:21] <toast> after reading some i kind of subscribe to the "bring the machine head to the toolchanger" philosophy
[20:43:33] <anonimasu> my head's fixed
[20:43:34] <anonimasu> moving tablew
[20:43:35] <anonimasu> table
[20:43:38] <anonimasu> err knee..
[20:43:39] <toast> yeah, most are
[20:44:18] <anonimasu> im thinking a drum and a changer arm
[20:44:25] <toast> yeah
[20:44:35] <toast> well the other option is the carousel type changer
[20:44:42] <toast> or umbrella, whatever you want to call it
[20:44:42] <anonimasu> mounted on the _other_ side of the machine(away from the cabinet with amsp..)
[20:44:42] <anonimasu> yeah
[20:44:50] <anonimasu> I want a vertical drum..
[20:44:54] <toast> a lot faster, yeah
[20:44:54] <anonimasu> like haas has them..
[20:45:16] <toast> yar
[20:45:37] <anonimasu> I could do with a 4 tool rack.. actually
[20:45:45] <anonimasu> and a linear axis..
[20:46:06] <anonimasu> it'd be enough
[20:46:15] <anonimasu> :)
[20:47:33] <toast> lol
[20:47:35] <toast> yeah
[20:47:44] <anonimasu> only problem is that I have a threaded power drawbar..
[20:48:05] <anonimasu> and lock | | things in the spindle..(requires rotating tools)
[20:48:44] <anonimasu> or spindle to line up :)
[20:49:20] <Skullworks-PGA1> remove the drive pin
[20:49:24] <anonimasu> I have no idea how commerical machines do it..
[20:49:28] <anonimasu> is that really a good ida?
[20:49:30] <anonimasu> idea?
[20:49:59] <Skullworks-PGA1> yes - its one of the first things we do with any R8 machine
[20:50:13] <anonimasu> Skullworks-PGA1: really?
[20:50:18] <anonimasu> I thought you had to have them
[20:50:20] <Skullworks-PGA1> yes
[20:50:44] <anonimasu> "yes?!"
[20:51:06] <Skullworks-PGA1> it is to make it easier when changing tools that the collet can't spin as draw bar is turned
[20:51:22] <Skullworks-PGA1> has to drive function
[20:51:31] <anonimasu> guide it..
[20:51:49] <anonimasu> dosent they provide more clamping force/keeping the tool from rotating?
[20:51:48] <Skullworks-PGA1> has NO drive function - that is all done by the grip of the taper
[20:51:51] <anonimasu> err holder..
[20:51:55] <anonimasu> oh!
[20:51:56] <anonimasu> sweet
[20:52:09] <Skullworks-PGA1> think about it
[20:52:19] <anonimasu> if that's the case toolchanging is easy..
[20:52:29] <anonimasu> well, a sane challenge
[20:52:38] <Skullworks-PGA1> a .100" dia pin sticking out .05" - there is no load capacity there
[20:53:43] <Skullworks-PGA1> but what WILL be a pain if it happens is that if drawbar is not tight - and you are using a fixed tool holder like an EM holder
[20:54:42] <Skullworks-PGA1> is that the pin will form a radial groove 90 drees from the alignment grrove
[20:55:19] <Skullworks-PGA1> and now the tool holder will be stuck in the spindle until you can line up the groove.
[20:55:35] <anonimasu> yeah ofcourse :)
[20:55:46] <anonimasu> EM? holder+
[20:56:14] <anonimasu> http://www.bftburzoni.com/immagini/prodotti/8_punte_inserti/schede/69871_PR_b.jpg
[20:56:21] <anonimasu> they dont have a pin they have some square things
[20:57:03] <anonimasu> :)
[20:57:07] <anonimasu> Skullworks-PGA1: this helped me lots ;)
[20:57:15] <Skullworks-PGA1> OK - thats a CAT 40? not an R8
[20:57:20] <anonimasu> iso40
[20:57:24] <anonimasu> yeah
[20:57:35] <anonimasu> http://thumbs.kapaza.nl/thumb/2022472.jpg
[20:58:17] <anonimasu> does that make a difference?
[20:58:22] <anonimasu> are they still just guides?
[20:58:37] <Skullworks-PGA1> well still - proper torque transfer from spindle to tool is from the taper grip
[20:59:25] <Skullworks-PGA1> my 10Hp 12,000rpm Okuma VMC holds the drive lugs on the spindle face with 4mm scres
[20:59:36] <Skullworks-PGA1> screws
[20:59:40] <anonimasu> yeah
[21:00:20] <Skullworks-PGA1> they hold the tool during toolchange to preserve alignment.
[21:00:47] <anonimasu> how does the toolchanger handle it?
[21:00:51] <anonimasu> alignment?
[21:01:03] <anonimasu> does it always rotate the spindle to a know position then take the tool out?
[21:01:23] <Skullworks-PGA1> - goto go now - have friends BBQing and they are waiting on me.
[21:01:42] <anonimasu> ok
[21:02:17] <Skullworks-PGA1> yes M6 does a internal M19 spindle orient before the toolchanger movement
[21:02:32] <anonimasu> ah ok
[21:02:34] <Skullworks-PGA1> bbl
[21:03:00] <anonimasu> that's a pain :)
[21:17:19] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[21:29:05] <anonimasu> tomp: drop me a message when you wake up :)
[21:29:27] <alex_joni> hi guys
[21:29:38] <alex_joni> lerman: I am now around
[21:29:50] <JymmmEMC> alex_joni: stop eating so much
[21:29:59] <alex_joni> who says I was eating?
[21:30:08] <JymmmEMC> alex_joni: you said you were around
[21:30:16] <alex_joni> funny :/
[21:30:24] <JymmmEMC> alex_joni: not really ;)
[21:30:40] <alex_joni> I noticed
[21:30:46] <anonimasu> hm, when doing rigid tapping you dont nescessarily need a servo spindle do you?
[21:30:49] <JymmmEMC> couldn't tell =)
[21:31:55] <SWPadnos> doesn't have to be servo, but does need to be controllable by EMC (fwd/rev/on/off) and have good speed feedback
[21:32:12] <anonimasu> * anonimasu nods
[21:35:34] <lerneaen_hydra> SWPadnos; good speed feedback?
[21:35:51] <SWPadnos> yes, good speed feedback. index also required, I believe
[21:35:54] <lerneaen_hydra> ie good resolution wrt spindle position?
[21:35:56] <lerneaen_hydra> oh, right
[21:40:53] <anonimasu> hm..
[21:47:36] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[21:52:40] <anonimasu> * anonimasu yawns
[21:56:53] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[22:00:40] <alex_joni> * alex_joni recompiles emc
[22:00:49] <JymmmEMC> * JymmmEMC recompiles alex_joni
[22:00:57] <JymmmEMC> KERNEL PANIC
[22:02:11] <CIA-24> 03alex_joni 07TRUNK * 10emc2/src/emc/kinematics/tp.c: spelling fix in error
[22:03:24] <alex_joni> core dump
[22:03:36] <JymmmEMC> lol
[22:04:28] <JymmmEMC> cradek: Something you might appreciate http://www.leapsecond.com/pages/atomic-nixie/
[22:07:06] <a-l-p-h-a2> a-l-p-h-a2 is now known as a-l-p-h-a
[22:32:12] <jmkasunich__> jmkasunich__ is now known as jmkasunich
[22:54:16] <LawrenceG> JymmmEMC: haro.... come in Tokyo
[22:54:43] <JymmmEMC> LawrenceG:
[22:54:50] <JymmmEMC> AK
[22:55:27] <LawrenceG> hey... what are the oem 650 and 750 drives worth.... I just started following them on ebay?
[22:56:14] <JymmmEMC> LawrenceG: I've seen 650's for roughly $85, rarely seend 750's
[22:56:27] <JymmmEMC> what are you looking at currently?
[22:57:02] <LawrenceG> 160144766199
[22:57:23] <JymmmEMC> $40 and he's selling three of them http://cgi.ebay.com/Parker-Compumotor-OEM750-Series-Drive-Automation_W0QQitemZ330155855119QQihZ014QQcategoryZ71394QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[22:58:09] <LawrenceG> those too.... have you compared 650 and 750 on your motors?
[22:58:17] <JymmmEMC> but being sold AS-IS, not DOA guarntee
[22:58:36] <JymmmEMC> The 650 dont have mid band compensation, not I havne't yet
[22:58:53] <JymmmEMC> caompared that is, but skunk said he never had an issue
[22:59:21] <LawrenceG> I have a few spare steppers here that might get used for something if I had drives.... looking for deals
[23:00:07] <JymmmEMC> with heatsink http://cgi.ebay.com/Compumotor-Parker-OEM650-Microstepping-Drive-w-Heatsink_W0QQitemZ160144766199QQihZ006QQcategoryZ71394QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[23:00:17] <LawrenceG> does it look like the 650/750 could be fixed if they blew up?
[23:00:54] <JymmmEMC> except for the SMT stuff, yeah.
[23:01:08] <JymmmEMC> the 750 still mfg'ed so you coudl return them to parker for repair.
[23:01:34] <LawrenceG> cool... If I buy drives I would like a set of 3...
[23:01:58] <JymmmEMC> I understnad. The750 has dip switches, the 650 has jumpers
[23:02:55] <JymmmEMC> speak of the devil
[23:03:10] <JymmmEMC> skunkworks: LawrenceG has parker Q's
[23:03:48] <skunkworks> uh oh
[23:03:59] <LawrenceG> not many!... just wondering if anyone had compared 650 series with 750 series
[23:04:15] <skunkworks> I gave him the link for the manual - what more does he want.
[23:04:17] <skunkworks> ;)
[23:04:26] <skunkworks> I have only used the 650's
[23:05:15] <JymmmEMC> LawrenceG: You're asking the comparision of performance between the two?
[23:05:24] <JymmmEMC> or features?
[23:05:58] <LawrenceG> got manual... thanks... watching some on ebay... would also like some oem 675..... perceived benefit of mid band resonance feature
[23:06:46] <JymmmEMC> LawrenceG: mid-band compensation is what geckos have, xylotex doesn't.
[23:07:10] <JymmmEMC> On that alone, is whu I wanted the 750's.
[23:07:14] <JymmmEMC> why
[23:07:40] <LawrenceG> understand.. would like to know if its worth the extra cost.... my current stepper drives dont have it and they run ok
[23:08:36] <JymmmEMC> skunkworks said he's never had an issue with the 650's, but when I looked around and found the 750's had the mid band, I went for it.
[23:08:47] <alex_joni> good night all
[23:08:53] <JymmmEMC> nite alex
[23:08:56] <LawrenceG> sweet dreams
[23:09:12] <alex_joni> I'll be dreaming about NML & interp :/
[23:10:27] <toast> does anyone have any large manual lathes or mills
[23:10:31] <toast> that they particularly like
[23:20:01] <anonimasu> hm
[23:20:32] <LawrenceG> JymmmEMC: we just finished an exciting lightning and flash flood type storm here....
[23:21:05] <LawrenceG> JymmmEMC: now its sunny again.... just very damp
[23:23:39] <skunkworks> been raining here all day. Steve miller band concert was canceled
[23:23:47] <skunkworks> (outdoor)
[23:30:07] <jmkasunich> * jmkasunich hates infernal combustion engines
[23:30:18] <jmkasunich> (damn lawnmower won't start)
[23:30:45] <cradek> you win then
[23:30:51] <jmkasunich> for today anyway
[23:30:56] <cradek> neighbor kid + $20?
[23:30:58] <jmkasunich> sooner or later the grass will win
[23:31:11] <cradek> or is it $40 by now
[23:31:24] <jmkasunich> if they provide the mower, probably 40
[23:31:33] <skunkworks> heh - fuel and spark.. fuel and spark.
[23:31:35] <jmkasunich> kids don't do that anymore - its all landscapers now
[23:31:56] <cradek> skunkworks: fuel, spark, compression
[23:31:58] <jmkasunich> got spark (pulled plug, set on crankcase, cranked, sparky sparky)
[23:31:57] <cradek> 3 things
[23:32:16] <jmkasunich> got fuel, at least in the carb (the inside of the air inlet tube gets wet after cranking
[23:32:32] <skunkworks> if you don't have compression - you have bigger problesm
[23:32:32] <jmkasunich> got compression I think, but how much? dunno
[23:32:34] <cradek> spark plug wet or dry?
[23:32:38] <jmkasunich> plug is dry
[23:32:38] <skunkworks> problems
[23:33:42] <robin_sz> evening
[23:33:48] <jmkasunich> evening
[23:33:48] <skunkworks> a can of starter fluid is your friend.. Or if your adventurouse - a little gas down the carb.
[23:34:08] <jmkasunich> there's plenty of gas on the inlet side of the carb
[23:34:27] <robin_sz> then the lug should be wet
[23:34:33] <robin_sz> plug
[23:34:39] <jmkasunich> you'd think so
[23:34:57] <robin_sz> 2 smoke or 4 smoke?
[23:35:06] <jmkasunich> 4
[23:35:20] <robin_sz> hmmm
[23:36:10] <jmkasunich> ran fine when last used (June - we've had a dry spell)
[23:37:41] <robin_sz> ahh, mower?
[23:37:41] <cradek> this is a basic push type mower?
[23:37:44] <jmkasunich> yep
[23:38:01] <jmkasunich> one cylinder, horizontal draft carb
[23:38:08] <robin_sz> probably needs fresh gas
[23:38:30] <jmkasunich> robin_sz: can't rule that out of course, but I'd be astonished
[23:38:46] <jmkasunich> this gas is pretty fresh
[23:38:47] <cradek> does it have the carb that has a fuel pump made of gasket material plus a spring with a cap on it? those gaskets can tear and make it so it can't draw gas up anymore
[23:39:05] <robin_sz> the stuff in the floatbowl will be low aromatics by now
[23:39:17] <jmkasunich> no fuel pump - gravity feed - carb is below the tank
[23:39:33] <cradek> huh
[23:39:41] <robin_sz> normal for mowers
[23:39:49] <jmkasunich> robin_sz: I've had far less trouble starting it for the first time in spring after a full winter
[23:39:49] <robin_sz> diaphragm carb
[23:39:55] <jmkasunich> this has only been a couple months
[23:39:59] <robin_sz> of hot wx?
[23:40:02] <jmkasunich> warm months tho
[23:40:14] <robin_sz> evaporation of the aromatics ...
[23:40:17] <jmkasunich> yeah
[23:40:18] <robin_sz> ??
[23:40:26] <robin_sz> drain the bowl at least
[23:40:31] <robin_sz> maybe?
[23:40:42] <jmkasunich> easier said than done on this one (but I should go look again)
[23:40:44] <cradek> I've never had that problem
[23:40:57] <robin_sz> wait, does it say "briggs and stratton" on the engine, or "tecumseh"?
[23:41:01] <jmkasunich> old one (20+ years old) had a spring button on the bottom of the bowl, this one doesn't have that
[23:41:23] <jmkasunich> crapsman (so its one or the other, but they disguise it)
[23:41:43] <robin_sz> right, because briggs and stratton are not actually 4 stroke gas engines
[23:41:54] <cradek> eh?
[23:42:02] <robin_sz> no, they look like gas engines
[23:42:09] <robin_sz> but really, they are clockwork
[23:42:22] <robin_sz> you pull and pull and pull ... that just winds it up
[23:42:37] <robin_sz> then it runs for 2 minutes before you wind it up again.
[23:43:00] <robin_sz> hint: get a honda ;)
[23:43:12] <jmkasunich> bah
[23:43:21] <jmkasunich> this one has worked fine for about 10 years
[23:43:29] <robin_sz> some do, to be fair
[23:43:37] <jmkasunich> the one prior to that (also crapsman) worked for close to 20 (we got it from my wife's parents)
[23:56:12] <jmkasunich> http://www3.sears.com/Paul/Leadimage.asp?productTypeID=1503500&brandID=0247&modelDesc=ENGINE&modelNumber=143434332&documentID=00064147&documentTypeID=PLDM&documentClassCode=PM&titleType=REPLACEMENT%20PARTS&titleID=00001&
[23:59:04] <cradek> probably a carb problem, but I'd check compression to be sure before working too hard on it
[23:59:31] <jmkasunich> any easy way to check compression?
[23:59:55] <cradek> you can borrow my gauge