#linuxcnc | Logs for 2015-04-08

Back
[02:21:06] <Deejay> moin
[02:21:15] <Jymmm> hi Deejay
[02:21:28] <Deejay> hi Jymmm :)
[03:15:38] <miss0r> I am experiencing some backlash issues with my home converted mill. Some of the backlash came from too weak bearings locking the ball-rod in place. I have mended that. Now it seems that I have something like 0.15mm backlash in the ballnut itself. My question is as follows: is it possible that I am simply asking too much of the ballnut? as in: am I trying to moove too much with this stuff?
[03:15:38] <miss0r> are you guys familiar with this?
[03:16:34] <miss0r> or did I perhaps buy it too cheap
[03:22:03] <archivist> cheap screws are cheap, look at the spec
[03:22:34] <archivist> c6,c7 or whatever
[03:26:32] <miss0r> so, if nothing is mentioned about any C-value where I bought, it means it is probally shit :) Can you reccomend somewhere good to get a sfu1605 with ballscrew that are trustworthy?
[03:29:10] <archivist> I remember reading the catalogue I got from a show http://www.misumi-europe.com/en/bestseller/screws, but there are plenty of makers
[03:30:59] <archivist> where I got the cx specs data from, cant find my copy today though
[03:31:31] <miss0r> alright. thanks. I will see what I can find. because what I've got is basically worthless
[03:40:16] <archivist> I just program my machine to cut one way without climb milling so backlash is not much of a problem for me
[03:41:31] <miss0r> hmmm.. But I would never be able to make a precise pocket with the backlash
[03:42:36] <archivist> sure you can, add the backlash compensation value to the config og the machine
[03:43:06] <miss0r> so, you are telling me, I can just program my way out of this?
[03:43:20] <archivist> yes
[03:43:28] * miss0r is reading up on that
[03:44:27] <archivist> you can even map pitch error of the screw and calibrate it out
[03:44:44] <miss0r> woah... one step at a time :)
[03:44:54] <archivist> hehe
[03:45:15] <archivist> you want the big step too?
[03:45:19] <miss0r> I wish I knew how it can do this?
[03:45:23] <miss0r> sure
[03:46:22] <archivist> backlash is on http://www.linuxcnc.org/docs/html/config/ini_config.html
[03:47:06] <miss0r> Thanks
[03:47:07] <archivist> BACKLASH = 0.0000 - Backlash in machine units. Backlash compensation value can be used to make up for small deficiencies in the hardware used to drive an axis. If backlash is added to an axis and you are using steppers the STEPGEN_MAXACCEL must be increased to 1.5 to 2 times the MAX_ACCELERATION for the axis.
[03:47:41] <miss0r> why must the max_acceleration be increased?
[03:48:25] <archivist> I think a fault in the stepgen acceleration code, that noone has bothered fixing yet
[03:49:08] <miss0r> i'm pretty sure I don't understand how it can possibly make this work properly. Do you know of a link explaining how it compensates?
[03:51:17] <archivist> just put that .15 in there and retest
[03:51:41] <archivist> magic happens
[03:51:49] <miss0r> I'm not sure I like what I see. no matter how much I compensate, the cutter will still be able to move the table
[03:52:46] <archivist> yes that can happen hence I only conventional mill so it always pushes the screw the right way
[03:53:25] <miss0r> hmm... I will go test. Thanks
[07:03:53] <skunkworks> http://www.cnczone.com/forums/tormach-pathpilot-/264676-well-pci-pp-back-stock-ordered-mine.html
[07:14:12] <jthornton> yea! the cave computer is back on line
[07:14:44] <skunkworks> what was it?
[07:15:53] <Tom_itx> seems the Tormachers are happy with their new stuff
[07:15:58] <jthornton> I had to edit the etc/network/interfaces
[07:16:20] <Tom_itx> i do that so seldom i gotta check my notes every time
[07:16:22] <jthornton> add a couple of lines so the pci network card worked
[07:16:43] <jthornton> the onboard network died and I added a PCI card
[07:19:37] <Loetmichel2> how do you even kill a Twisted pair nic?
[07:19:45] <skunkworks> it happens
[07:19:55] <Loetmichel2> they are supposed to withstand 2kV on each pin in respect to ground
[07:21:12] <jthornton> we had a power blink and it died
[07:21:33] <Loetmichel2> meh, thats chinese quality then
[07:25:37] <_methods> jthornton: what torch did you put on your plasma?
[07:25:45] <_methods> hypertherm 80?
[07:26:28] <_methods> or powermax 45?
[07:27:33] <jthornton> hypertherm 1250
[07:27:42] <_methods> ahh
[07:28:12] <jthornton> they don't make that model anymore
[07:28:22] <_methods> yeah looks like that is the powermax 65 now
[07:28:32] <_methods> you buy new?
[07:29:10] <jthornton> aye
[07:29:23] <_methods> you happy with it?
[07:29:26] <jthornton> then added the machine torch
[07:29:32] <jthornton> yes
[07:29:35] <_methods> kk
[07:29:45] <_methods> you running on 220?
[07:29:49] <_methods> single phase for it
[07:30:47] <_methods> what's the max pierce for yours?
[07:30:51] <jthornton> yes
[07:30:55] <_methods> the powermax 65 says 5/8"
[07:31:04] <jthornton> let me look
[07:32:15] <jthornton> aww that directory is not shared on the plasma... I can look in a bit when I go out to the shop
[07:32:23] <_methods> k
[07:32:25] <_methods> no rush
[07:32:45] <_methods> my oxyfuel torch mixing assembly melted on me last night
[07:32:54] <_methods> and i was like you know i should just get my plasma now
[07:34:16] <_methods> i can use it for roughing out all my parts for my plasma table build
[07:36:52] <renesis> the torch handle itself?
[07:36:55] <renesis> scary shit
[07:40:51] <Loetmichel2> i once ran a fire INSIDE the oxygen hose to the bottle
[07:41:11] <Loetmichel2> beat it and closed the bottle/detached the hose...
[07:41:16] <Loetmichel2> was close tho
[07:41:36] <Loetmichel2> that could have been a bad idea if i would have been a bit slower
[07:49:07] <Loetmichel2> since then i look for missing backfire suprossors on the hoses before starting an oqy/acetylene torch ;-)
[08:07:56] <JT-Shop> _methods, the 1250 will pierce 5/8" and side start up to 1"... the thickest I've pierced is 3/8"
[08:08:14] <JT-Shop> and I usually cut 11ga or thinner with the FC nozzle
[08:19:59] <renesis> first time i did anything with oxy fuel was 1" from the side, 1/8" cut across 12x12 plate
[08:20:25] <renesis> when i got maybe 3/4 in, i look at the start and its opened up like 2"
[08:21:15] <renesis> i was doing the machining program at the time, so going through 1" like that with a 1/8" cut with virtually all the metal vaporized was mind blowing
[08:21:55] <renesis> pretty sure it took way under a minute, too
[08:33:31] <_methods> yeah the acetylene mixing chamber melted
[08:33:54] <_methods> i guess it back fired in there, i did have a couple pop offs from getting too close
[08:37:53] <_methods> thanks JT-Shop those sound about the same specs as the hypertherm powermax 65
[08:37:58] <_methods> i guess that's what i'll pick up
[08:50:31] <_methods> piercing 5/8" should be good for about 95% of what i plan on doing
[08:53:37] <_methods> JT-Shop: is it an air hog?
[08:53:49] <JT-Shop> not as bad as my CHNC lol
[08:53:54] <_methods> haha
[08:54:07] <_methods> looks like you can run it on nitrogen
[08:54:31] <JT-Shop> I have an air dryer inline when I run the plasma
[08:54:50] <_methods> damn i'll need to upgrade my compressor and get a dryer
[08:54:55] <_methods> maybe i'll wait on this
[08:55:17] <JT-Shop> I'd go the Powermax85 if I was doing 5/8" all the time
[08:55:32] <_methods> i'm sure i'll be in the 1/8"-3/8" range
[08:55:47] <_methods> if you had it to do over you'd get a more powerful pak?
[08:55:51] <JT-Shop> 65 will do fine for that with occasional 5/8"
[08:56:05] <JT-Shop> no, I do mostly 11ga and thinner
[08:56:08] <_methods> k
[08:56:20] <JT-Shop> some 1/4 and 3/8
[08:56:30] <_methods> how is the cut quality on that thinner stuff?
[08:57:33] <_methods> guy down teh street from me works at kaliburn
[08:57:55] <_methods> i wonder if i should talk to him about getting one of theirs refurb or something
[08:59:53] <_methods> guess i need to get my priorities straight first
[09:00:00] <_methods> dry air would be the first step
[09:02:41] <JT-Shop> yea, dry air is important for consumable life
[09:02:55] <_methods> yeah, not nearly as exciting as plasma
[09:03:05] <_methods> sux when you have to spend money on boring stuff
[09:03:28] <_methods> like buyin tires for your car
[09:03:54] <JT-Shop> and both my cars need tires
[09:04:03] <_methods> haha
[09:04:19] <_methods> dry air or sexy plasma machine with sparks and flames
[09:04:24] * JT-Shop heads to town to mail packages
[09:05:11] <JT-Shop> or moist air and shorter consumable life... having said that I cut for a year with shop air and no dryer
[09:05:32] <_methods> i have a crappy little harbor freight compressor
[09:06:09] <_methods> little 2gallon
[09:06:23] <_methods> i can barely fill a tire with it
[09:06:28] <Jymmm> 2gal would drive me nuts
[09:06:35] <JT-Shop> it could work for shorter cuts, if the pressure gets too low the plasma cuts out
[09:06:40] <_methods> yeah it's horrible but i don't have a lot of air tools
[09:06:47] <JT-Shop> mine uses 85psig when cutting I think
[09:06:49] <_methods> so thankfully i don't have to fire it up often
[09:07:01] <Jymmm> I just use nitrogen tank instead
[09:07:03] <_methods> yeah said like 6.5scfm at 85psi
[09:07:10] <JT-Shop> no, 75 psig
[09:07:16] <_methods> k
[09:07:30] <JT-Shop> gotta run
[09:07:42] <_methods> later thx
[09:08:43] <ssi> morn
[09:09:05] <_methods> hola
[09:09:33] <_methods> what torch you use on your plasma ssi
[09:09:54] <ssi> I have a powermax 45, and I used the stock hand torch on it for ages, but I have the t65m machine torch now
[09:09:59] <ssi> only downside there is it uses different consumables
[09:10:08] <_methods> yeah i'm lookin at the powermax 65
[09:10:20] <ssi> actually it might not be the t65m? I forget the model
[09:10:22] <ssi> it's the SHORT one
[09:10:23] <_methods> but looks like i need to get my air right first
[09:10:24] <ssi> without a rack
[09:10:33] <_methods> yeah that mini torch
[09:10:35] <ssi> yeah
[09:10:44] <ssi> 180?
[09:10:46] <ssi> something like that
[09:10:49] <ssi> http://www.bakersgas.com/HYP059481.html?gclid=CK31zdrs5sQCFdcXgQodSxwASA
[09:11:06] <_methods> yeah
[09:11:45] <_methods> i guess i might just get a big harbor freight compressor
[09:12:53] <_methods> http://www.harborfreight.com/air-tools/air-compressors/5-hp-60-gallon-165-psi-two-stage-air-compressor-93274.html
[09:14:24] <ssi> do you need 165psi?
[09:14:34] <ssi> cause I can show you the value in compressors if you're ok with single stage
[09:15:30] <ssi> http://www.northerntool.com/shop/tools/product_211720_211720
[09:15:38] <ssi> I've bought two of them, and they're great compressors for the money
[09:22:26] <_methods> k
[09:22:42] <_methods> yeah i'm not picky
[09:22:44] <_methods> i just need air
[09:22:49] <ssi> yea
[09:22:58] <ssi> that IR is awesome... 18.1cfm
[09:23:02] <ssi> it's got a decent 5hp motor on it
[09:23:05] <ssi> the HF one I'd be suspect
[09:23:13] <_methods> um yeah it's HF
[09:23:14] <_methods> lol
[09:23:16] <ssi> :)
[09:23:41] <_methods> that plas says it need 6.5 scfm at 85
[09:24:31] <_methods> i think i'd rather have IR than HF
[09:25:33] <ssi> I know I would
[09:25:39] <ssi> and it's more money but not a lot more
[09:25:59] <_methods> wow lot of bad reviews on northern tool though
[09:27:34] <_methods> wonder if i should just get teh HF and plan on it blowin up and put the money in a better aftermarket compressor and motor
[09:27:47] <_methods> when it dies
[09:28:21] <ssi> I lost a motor in one of mine, but it was like seven years in
[09:29:03] <_methods> i'm always watching auctions to try and get one
[09:29:15] <_methods> but haven't seen one at the right price yet
[09:29:21] <ssi> also the HF compressors that I've seen are extremely noisy
[09:29:23] <ssi> but the IR is quiet
[09:29:56] <_methods> the shop next to us just gave us a 25hp gardner denver rotary screw
[09:30:04] <_methods> i'd love to have that
[09:30:10] <_methods> but that's so overkill lol
[09:33:34] <ssi> yuuuup
[09:35:56] <_methods> heh 2hp 20gal dental air compressor for sale on craigslist for $1000
[09:36:11] <_methods> probably won't do enough cfm for what i'll need though
[09:36:28] <_methods> http://charleston.craigslist.org/tls/4939425020.html
[09:37:45] <ssi> that's pretty cool
[09:37:53] <ssi> that'd be a great laser compressor
[09:37:56] <_methods> yeah clean dry air for sure
[10:17:29] <_methods> oh nice got a line on a speedaire 60gal for around $400
[10:17:45] <ssi> very nice
[10:18:36] <_methods> http://imgur.com/7dG1NPE
[10:19:06] <dirty_d> whoa https://www.youtube.com/watch?t=18&v=Jt9hEvjmHjI
[10:19:11] <JT-Shop> that's about the same as my Ingersol Rand
[10:19:32] <_methods> yeah uv cure
[10:20:02] <_methods> we'll see what the price ends up going to
[10:20:10] <_methods> but i'll try and get it for under $500
[10:20:51] <ssi> I wonder how you get that stuff unstuck
[10:20:56] <_methods> uv
[10:21:00] <_methods> uv cure and uncure
[10:21:03] <ssi> weird
[10:21:04] <dirty_d> how?
[10:21:15] <_methods> the uv is used for a certain time to cure
[10:21:51] <_methods> http://www.mmsonline.com/articles/light-activated-glue-holds-and-releases-workpieces-in-a-flash
[10:21:58] <_methods> they can explain better than me lol
[10:22:10] <_methods> that article if from 2004 lol
[10:22:24] <ssi> I need a lot more money
[10:22:25] <ssi> D:
[10:22:32] <dirty_d> i bet in the future they will have some kind of nano gel that responds to some kind of signal to harden and unharden
[10:22:57] <malcom2073> "even ice"
[10:23:02] <malcom2073> That's scary
[10:23:38] <dirty_d> in these videos, it just looks like they break it
[10:23:41] <dirty_d> with a bolt
[10:24:31] <_methods> not sure when i used it you had to hit it with a second shot of uv
[10:24:37] <_methods> then you could tap it free
[10:24:40] <dirty_d> neat https://www.youtube.com/watch?v=dNdua_KpL9Y
[10:24:48] <dirty_d> they have a tool for cleaning off the grippers
[10:31:31] <dirty_d> http://www.reddit.com/r/Machinists/comments/31ttmn/any_of_you_guys_try_this_product/
[10:31:38] <dirty_d> superglue, never though of that
[10:32:09] <dirty_d> i guess if you take light cuts, theres no raeson it wouldnt work
[10:43:13] <JT-Shop> dang 1.001" reamer was too long
[10:48:43] <Rab> Wonder if that Blue Photon stuff is related to dental composite filling material.
[10:50:35] <Rab> If it's UV release, I wonder if it's contra-indicated for certain processes like welding or plasma cutting. ;)
[10:51:25] <dirty_d> i dont get it, i dont see anythign with any springyness here https://www.youtube.com/watch?v=YY05iBfHszs
[10:53:08] <Rab> Fixed preload, you would have to change to a thicker shim if the system wears.
[10:53:51] <dirty_d> ahh
[10:54:40] <archivist> every metal is elastic under strain
[10:57:05] <dirty_d> i made a cut last night on my machien that was supposed to bring a part down to 1.5" height, came out to 1.494"
[10:57:16] <dirty_d> not sure what the hell happened, thats more than my backlash
[10:57:51] <dirty_d> i clamped it in a vice with a thin parallel under it, tightened the vise, tapped the part down to make sure the parallel was tight under it
[10:58:06] <dirty_d> touched the tool off the parallel
[10:58:25] <dirty_d> moved the tool to 1.5" for the finishing cut of about 0.01"
[10:58:41] <dirty_d> cant see how it came out that far off
[10:59:36] <_methods> JT-Shop: you reamed over?
[10:59:42] <dirty_d> its good enough for what i was making, but my OCD is bothering me
[11:01:37] <archivist> 6 thou under is a lot, check the vice does not lift when tightened
[11:02:04] <_methods> make sure you end mill didn't pull out of tool holder too
[11:02:20] <archivist> seen that too :)
[11:02:32] <_methods> yep
[11:02:49] <_methods> and is it 1.494 all over
[11:02:49] <JT-Shop> I reamed 1.001" and it is a perfect fit whew
[11:02:57] <_methods> heh nice
[11:03:02] <_methods> is it an over reamer?
[11:03:16] <archivist> or a blunt reamer
[11:05:43] <dirty_d> archivist, i touched off after i tighted it though
[11:05:57] <JT-Shop> over reamer?
[11:06:06] <dirty_d> _methods, yea
[11:06:08] <_methods> yeah you can buy reamers over and under
[11:06:32] <JT-Shop> yea it is an over reamer, the bushings are a bit over and the hole spec was 1.001"
[11:07:02] <_methods> yeah that will put you to the high side
[11:07:03] <JT-Shop> I still have to ream the bushing after assembly as the pins are a half a thou over
[11:07:41] <_methods> you can play with the feed rate some usually to make the hole bigger or smaller
[11:07:59] <_methods> if you want to make it smaller speed up the feed a bit
[11:08:27] <archivist> or lap for the tenths
[11:08:48] <_methods> yeah or burnish it
[11:10:31] <_methods> http://phillipecantin.blogspot.com/2015/03/teeny-tiny-build.html
[11:10:35] <_methods> heh that's pretty cool
[11:12:13] <archivist> but we had those displays in the dark ages of the 1970s
[11:13:00] <_methods> yeah lol
[11:13:03] <archivist> MK14 kit had them, bugger to play moonlander on it
[11:13:21] <zeeshan|2> im never ordering something from china again
[11:13:22] <zeeshan|2> lol
[11:13:24] <zeeshan|2> it takes forever!!
[11:13:57] <_methods> yeah one of the hidden reqmnts of sourcing from china is a large reserve of patience
[11:14:19] <_methods> and always order 2
[11:16:18] <CaptHindsight> _methods: todays funny http://www.3ders.org/articles/20150408-never-lose-your-drill-bits-again-with-this-useful-3d-printed-tool-wristband.html
[11:16:56] <syyl_ws> uh yeah
[11:17:01] <syyl_ws> that looks as usefull as
[11:17:02] <syyl_ws> hm
[11:17:06] <syyl_ws> an smartwatch?
[11:17:10] <archivist> impossible to get out quickly
[11:17:37] <CaptHindsight> http://www.3ders.org/articles/20150408-ir3-industrial-revolution-iii-3d-printer-pick-place-assembly-bot-launches-on-kickstarter.html is that thing already warping or is it just the angle of the pic?
[11:17:52] <_methods> buwhahahaha
[11:18:04] <syyl_ws> "industrial revolution"
[11:18:19] <syyl_ws> and the guys with 20 SiPlace laugh their ass off
[11:20:24] <archivist> no strength in that design to lift anything weighing more than a feather, nor press fit the parts
[11:21:23] <archivist> not fit for purpose, must try harder
[11:21:47] <CaptHindsight> if was designed and built by the girl in the pic then it's impressive
[11:21:57] <malcom2073> Makes me want to stick grippers on my printer, it can push down with enough force to press-fit things heh
[11:22:40] <malcom2073> I'd be excited about it for having nuts and the like dropped in, except I discovered melt-in plastic nutcerts and haven't looked back since
[11:22:41] <archivist> I think thaey are designing for a piss fit rather than press fit
[11:22:46] <_methods> https://www.kickstarter.com/projects/817819607/the-industrial-revolution-iii-become-a-global-desi
[11:22:49] <_methods> hahaha
[11:22:57] <_methods> they got $4.47
[11:23:07] <malcom2073> People need to stop buzzwording stuff
[11:23:27] <CaptHindsight> the price is right
[11:23:37] <CaptHindsight> <$5 in junk parts
[11:24:08] <archivist> I need some ground rod, must back it for 1p
[11:24:26] <CaptHindsight> how little can one donate?
[11:24:28] <malcom2073> "WE EXTEND NO WARRANTIES REGARDING YOUR FITNESS OR THAT OF OUR PRODUCTS FOR ANY PURPOSE. "
[11:24:32] <malcom2073> Well, at least they're honest
[11:24:36] <_methods> wow
[11:24:39] <CaptHindsight> do they also accept good advice?
[11:24:50] <_methods> there is a whole "team" behind making this
[11:25:10] <malcom2073> No wonder they need 75k
[11:25:11] <CaptHindsight> yes, Mrs Smiths 4th grade science class
[11:25:21] <_methods> the pic looks that way
[11:25:30] <_methods> scroll to the end of the kickstarter for much lulz
[11:26:11] <CaptHindsight> http://createspacelondon.org/
[11:27:01] <malcom2073> There's a makerspace starting up in a town not far from me
[11:30:05] <archivist> seems like none of these new inventors ever did any structural design
[11:31:00] <CaptHindsight> it's made of metal so it must be strong.... just look at bridges, tanks and hammers
[11:31:13] <Rab> $225 billet aluminum brick looks pretty structural to me.
[11:31:30] <Rab> Especially after its li-poly pack craps out.
[11:32:48] <Rab> (Yesterday's design lol:) https://www.kickstarter.com/projects/1363754907/brick-lamp-reveal-the-light
[11:33:03] <archivist> I like looking at bridges, I see design and thought, not when I look at repcrap though
[11:34:04] <CaptHindsight> Rab: now that is magic
[11:34:39] <archivist> it contains snake oil
[11:35:53] <CaptHindsight> https://www.adafruit.com/products/1946 turns any device with a uart into a quad band phone
[11:37:45] <CaptHindsight> https://www.nabitablet.com/nabi-big-tab going to put one in here and use as my new smartphone
[11:38:10] <roycroft> i'm assuming that 3d printed wrist band screwgun bit holder is an april fools' joke posted a week late
[11:38:28] <malcom2073> CaptHindsight: Please say you'll put a microphone on the bottom, speakeron the top, and hold it to your face?
[11:38:39] <CaptHindsight> malcom2073: thats the plan
[11:38:42] <malcom2073> Sweet
[11:39:00] * archivist sets pagespeed looking at that crappy adafruit cake site
[11:39:02] <JT-Shop> one more pair to go and it's lunch time
[11:39:05] <CaptHindsight> kickstarter here we come
[11:42:21] <PetefromTn_> Hey linuxCNC junkies!
[11:44:40] <JT-Shop> http://ibin.co/1xdot9nhWbd0
[11:46:05] <archivist> production!
[11:47:45] <Rab> CaptHindsight, trying to figure out if that big tablet is actually battery-operated.
[11:47:54] <CaptHindsight> Rab: it is
[11:48:00] <CaptHindsight> saw them at Walmart
[11:48:29] <CaptHindsight> https://www.nabitablet.com/nabi-big-tab/specs
[11:48:36] <Rab> 1650 mAh, Rechargeable Lithium-ion Battery 30 minute backup for portability*
[11:48:39] <Rab> haha
[11:48:45] <CaptHindsight> tiny
[11:49:25] <_methods> ermahgerd pete lives
[11:51:00] <PetefromTn_> so far anyways LOL
[11:51:09] <JT-Shop> hey Pete
[11:51:21] <PetefromTn_> Hey JT HOWSITGOIN?
[11:51:23] * JT-Shop notices it is half past lunch... be back later
[11:51:24] <_methods> where you been hidin
[11:51:33] <JT-Shop> busy as a three legged cat
[11:51:38] <ssi> peet
[11:51:46] <PetefromTn_> well I went on vacation for eleven days to Florida
[11:52:04] <JT-Shop> one more op on those blocks...
[11:52:07] <JT-Shop> nice
[11:52:07] <PetefromTn_> and when I got back tore my kitchen apart and been working on refinishing my cabinets
[11:52:19] <PetefromTn_> glad you are busy JT
[11:52:22] <PetefromTn_> nice
[11:52:27] <JT-Shop> when your done you can drop by here lol
[11:52:40] <ssi> PetefromTn_: I fired my contractor last week
[11:52:42] <JT-Shop> I'll be glad when I'm caught up for a change
[11:52:45] <ssi> and I'm just gonna do it my damn self
[11:52:47] <ssi> hahahahaha
[11:52:51] <PetefromTn_> beleive it or not I have several projects for the CNC shop I am working on right now!!!
[11:53:08] <PetefromTn_> oh yeah what kinda contractor?
[11:53:15] <ssi> general
[11:53:15] <ssi> heh
[11:53:25] <PetefromTn_> are you rebuilding your home?
[11:53:28] <ssi> the only thing they've done is R&R the roof
[11:53:29] <ssi> yeah I have to
[11:53:42] <PetefromTn_> Damn that kinda sucks huh
[11:53:47] <ssi> yeah
[11:53:59] <ssi> I'm gonna get my buddy's brother who is a builder in michigan to come down and do the bulk of the work
[11:54:03] <PetefromTn_> maybe you can get it fixed up nice and get it sold...
[11:54:09] <ssi> insurance is issuing me the ACV check, which is like $78k
[11:54:24] <ssi> and then when it's done I get another $40k in replacement cost
[11:54:33] <PetefromTn_> wow nice you could probably almost build a new house for that much
[11:54:41] <ssi> yeah seriously
[11:54:47] <ssi> which is why I didn't want to pay a stupid contractor to waste it
[11:55:19] <PetefromTn_> I have several projects I am working on for the race shop guys right now
[11:55:32] <ssi> excellent
[11:55:40] <PetefromTn_> and I got a text from a local company that I have been frequenting lately that has kind been blowing me off
[11:55:53] <ssi> hey you want to look at a print and quote me?
[11:55:55] <PetefromTn_> they want me to come by today and look at a production piece they want done..
[11:56:07] <PetefromTn_> sure...
[11:56:15] <PetefromTn_> my CNC lathe is STill not done tho..
[11:56:22] <ssi> yeah hurry up with that
[11:56:23] <ssi> haha
[11:56:31] <ssi> if you can do this entirely on the mill, I'd have you do it immediately
[11:56:38] <ssi> needs to be 304 ss
[11:56:49] <ssi> and those tapped holes don't have to be thru all, they can be thru to the bore, either way
[11:57:14] <ssi> I need 500 pcs
[11:57:49] <PetefromTn_> are you in a real hurry?
[11:57:55] <ssi> yea kinda
[11:58:41] <PetefromTn_> what is the time frame because I have committed to do some CNC work for the race shop that will probably take the next week or two to finish.
[11:59:02] <ssi> I dunno, they said ASAP
[11:59:38] <ssi> I was planning on trying to get the hnc running and doing it with some manual millwork
[11:59:44] <PetefromTn_> but I probably could do that on the mill making them from stainless barstock if the tolerances are not too tight
[11:59:53] <ssi> the bore tolerances are the only important ones
[12:00:24] <ssi> let me know what you'd want to do the lot, and if I don't get my stuff set up by the time you're ready to roll then you can do them
[12:00:34] <PetefromTn_> I got a question for you guys...
[12:00:34] <jthornton> what kind of part ssi?
[12:00:37] <PetefromTn_> OK
[12:00:42] <ssi> http://www.prototechnical.com/~imcmahon/Adapter.PDF
[12:00:47] <ssi> oh yeah you have the chnc
[12:01:08] <PetefromTn_> I have a customer that is having me machine some parts for him that are kinda odd shaped
[12:01:27] <PetefromTn_> and after they are machined they are sending them out to be high speed balanced.
[12:01:38] <ssi> neat
[12:01:43] <Rab> <INSERT COMPANY NAME HERE>
[12:01:52] <ssi> that's what I named my company
[12:02:07] <jthornton> busy little part
[12:02:12] <PetefromTn_> I was wondering what would be involved in getting some kinda balancing apparatus setup here?
[12:02:15] <ssi> yeah it's a pain in the ass
[12:02:41] <ssi> I can do the turning work on the hnc if I get it running again, but the setscrew holes confound me
[12:03:05] <ssi> best thing I've come up with is to make a softjaw to hold them in the manual mill with two locations, one with a dowel hole
[12:03:15] <ssi> and do the driling and tapping there, with a tapping head
[12:03:22] <jthornton> are they tapped from each side or through?
[12:03:22] <ssi> it'll be like a six operation part that way :(
[12:03:29] <ssi> can be through or just to the bore
[12:03:31] <ssi> doesn't have to be through
[12:03:32] <PetefromTn_> yeah I would do them in soft jaws as well in the VNC
[12:03:36] <ssi> I'm not sure why I drew it that way :)
[12:03:55] <jthornton> does it hold something in the center bore?
[12:03:58] <ssi> yeah
[12:04:02] <PetefromTn_> two jaws vertical and horizontal
[12:04:03] <ssi> the bores are toleranced, they need to be reamed
[12:04:07] <ssi> the rest .01" is fine
[12:04:14] <jthornton> it would hold better if the set screws were 90 degrees
[12:04:23] <ssi> well there's just one setscrew
[12:04:26] <ssi> again, not sure why I drew it thru
[12:04:41] <jthornton> oh, ok that makes a difference
[12:04:43] <ssi> yeah.
[12:05:00] <ssi> also, because of the drill/tap operation, I figure the reaming op would need to happen after the tapping op
[12:05:08] <PetefromTn_> exactly
[12:05:34] <PetefromTn_> anyone ever built any sort of balancing machine?
[12:06:13] <_methods> not i
[12:06:24] <ssi> my plan was: op 1: face/chamfer barstock, drill .095" or so to .900", pull and part; op 2 (or 2-5 if on a manual machine), drill and tap setscrews on a mill; op 3 in a collet with stop, face and chamfer back side, drill .120", ream .125 to .375, ream .100 thru
[12:06:29] <ssi> or something like that
[12:06:30] <archivist> nope but there are some opensource turbine balancing machine designs
[12:06:51] <ssi> that's how I'm going to do it if I can get the hnc going in the near future, but if someone's ready to rock to make that part I'll happily sub it
[12:06:58] <archivist> I have some old code for balancing machine
[12:07:53] <_methods> do you have a toleranced print of the part?
[12:08:24] <PetefromTn_> archivist code?
[12:08:39] <_methods> and what is your deadline?
[12:08:55] <jthornton> I would think a thread forming tap for that job
[12:08:56] <archivist> PetefromTn_, http://www.collection.archivist.info/searchv13.php?searchstr=AN162
[12:09:15] <ssi> jthornton: yeah that's the conclusion i came to also
[12:09:19] <ssi> but I need to acquire a tapping head
[12:09:20] <_methods> the qty is 500 right?
[12:09:22] <ssi> since I have no cnc mill :(
[12:09:24] <ssi> _methods: correct
[12:09:35] <ssi> no toleranced print, I could make one
[12:09:43] <_methods> yeah i'll need that to quote it
[12:10:10] <archivist> PetefromTn_, just needs some accelerometers and a data grabbing system (and the rotation of course)
[12:10:23] <ssi> archivist: that wouldn't be too hard
[12:10:28] <_methods> what was the material?
[12:10:30] <_methods> 5052?
[12:10:34] <_methods> or 6061?
[12:10:40] <ssi> 304ss
[12:10:46] <_methods> ah crap that's right
[12:10:54] <archivist> piezo buzzer can make very sensitive accelerometers
[12:10:56] <_methods> gotta make it difficult eh
[12:10:58] <_methods> lol
[12:11:01] <ssi> yeah :(
[12:11:21] <PetefromTn_> I don't see anything there man what am I missing?
[12:12:06] <PetefromTn_> never had much luck with form taps in stainless...
[12:13:33] <jthornton> it take more than luck to tap stainless lol
[12:13:39] <Rab> Lots of lube + precise hole dia is critical.
[12:14:14] <ssi> well so, they didn't specify the material
[12:14:18] <PetefromTn_> I have tapped thousands of stainless holes I am well aware...just never had much luck with form taps in it.
[12:14:20] <ssi> they want "some kind of steel that won't corrode unfinished"
[12:14:24] <ssi> suggestions? :D
[12:14:25] <archivist> PetefromTn_, here is one I spotted at an exhibition http://www.collection.archivist.info/searchv13.php?searchstr=balancing+machine
[12:16:43] <archivist> PetefromTn_, sensing device is made with http://www.ebay.co.uk/itm/10Pcs-12mm-Thickness-1-0mm-Piezo-Disc-for-Buzzer-Pressure-Sensor-Speaker-DIY-/171120915838
[12:16:53] <archivist> or similar
[12:17:19] <dirty_d> how do those work as pressure sensors?
[12:17:58] <archivist> use them to block a port, measure the voltage they produce
[12:18:08] <PetefromTn_> honestly this part is pretty large and odd shaped. Probably weights about five to six pounds.
[12:18:28] <PetefromTn_> must be balanced to 12k RPM
[12:18:38] <archivist> use a smaller more sturdy accelerometer
[12:18:39] <ssi> what the heck is it
[12:19:03] <PetefromTn_> I'd rather not say ;)
[12:19:06] <ssi> hahah
[12:19:42] <PetefromTn_> but they currently send them out for balancing and they pay handsomely for it. They said if I could do it they would let me do everything on this part.
[12:19:53] <ssi> it'd be pretty neat to design a balancing machine that could weigh the part, spin it up, measure the accel vector, and then output data that could be post-processed into G-code whcih your mill would use to precisely mill away material to balance it
[12:19:53] <PetefromTn_> would mean some good consistent work in here
[12:20:09] <archivist> but the rc jet turbine makers are running up to over 60k rpm
[12:20:15] <ssi> and not exactly trivial, but well within the limits of diyp
[12:20:21] <ssi> -p
[12:20:45] <PetefromTn_> I know I had a friend that made a simple weight balancing machine for motorcycle wheels and we used it to balance the wheels and tires on racebikes
[12:20:45] <jthornton> actually it looks like a spiral point tap is the best for 304
[12:21:01] <archivist> does that mean I have to scan that doc?
[12:21:20] <jthornton> you don't want to break too many at $45 a pop
[12:21:30] <ssi> eesh
[12:21:34] <archivist> is in basic so should be easy to translate to any other language
[12:21:39] <ssi> what taps are $45/ea?
[12:21:45] <ssi> I was lookxing at form taps in 6-32 and they're like $13-18
[12:21:47] <archivist> over priced ones
[12:21:57] <jthornton> the OSG ones for tapping 304
[12:22:07] <PetefromTn_> buy a GOOD tap if you intend to do that job yourself
[12:22:09] <jthornton> hmm they are all H3
[12:22:30] <PetefromTn_> OSG are excellent I have several here
[12:22:53] <jthornton> it's all I use
[12:23:57] <PetefromTn_> I have some morse and even some fastenal that are decent for larger stuff. Smaller taps in tougher materials I always pay for better quality ones.
[12:24:32] <jthornton> bbl
[12:24:43] <ssi> exotap va-3 is the one the osg thing recommended me
[12:24:47] <ssi> and it says $14.17 each
[12:24:49] <ssi> I can handle that :D
[12:25:00] <ssi> ooh they have a georgia warehouse
[12:40:17] <_methods> +1 for OSG
[12:43:25] <_methods> is there a mesa card that can control both steppers and servos?
[12:43:40] <malcom2073> Daughterboard?
[12:43:43] <malcom2073> Or you mean at once
[12:43:54] <_methods> sure daughterboard
[12:44:14] <ssi> not that I'm aware of, but you could run two daughterboards on one io card and accomplish that
[12:44:44] <_methods> yeah that's what i mean with one io card can i control a combination of steppers and servos
[12:44:50] <ssi> yep
[12:44:51] <CaptHindsight> close your eyes, click your heels twice and say "is there a mesa card that can control both steppers and servos? pcw_home
[12:44:57] <_methods> hahah
[12:45:13] <ssi> you could easily run, say, a 7i76 and 7i77 on one 5i25
[12:45:17] <ssi> but I dunno if there's a firmware for it
[12:45:40] <malcom2073> The 7i77 has tons of extra I/O, can you run step/direction out of some of those?
[12:46:39] <CaptHindsight> there are 7i76 with 7i77 configs for 5i25
[12:47:33] <_methods> i wasn't sure if you could run the servo stuff and stepgen stuff all at the same time
[12:47:41] <ssi> sure
[12:48:12] <_methods> so just get a 5i25/6i25 and 7i76 and 7i77
[12:48:16] <ssi> hell, you can have as many mesa cards as you have PCI slots, and each of them have two or three daughtercards, and you can run all that horseshit at once
[12:48:22] <ssi> but you'd run out of AXES in linuxcnc ;)
[12:56:11] <CaptHindsight> _methods: working on a screw machine?
[12:58:39] <_methods> yeah kinda
[12:58:44] <_methods> want a real 4th axis
[12:58:52] <_methods> turn my minimill into a live tooling set up
[12:58:54] <_methods> http://littlemachineshop.com/products/product_view.php?ProductID=2299
[12:59:03] <_methods> i think i'm going to use that for my 4th axis
[12:59:06] <CaptHindsight> I asked the devs about expanding beyond 9 axes and they said it was possible
[12:59:14] <_methods> i only need 4
[12:59:24] <_methods> i just want to use a servo for the 4th
[12:59:39] <ssi> I should set my g0602 up and refit it with some simple live tooling
[12:59:46] <ssi> that'd be an ideal machine for something like this little part I need to make
[13:00:00] <_methods> yeah
[13:00:10] <_methods> would work well for that part
[13:00:13] <ssi> yep
[13:00:31] <ssi> even just a qctp toolholder with a small spindle in it
[13:00:36] <ssi> and spindle orient
[13:00:44] <ssi> (main spindle orient tht is)
[13:01:13] <XXCoder> ssi: once saw someone use small drill mounted on spidle to do 90 degree drills
[13:01:15] <XXCoder> small one
[13:01:18] <ssi> yeah
[13:01:19] <CaptHindsight> I've only made it to 8 axes so far
[13:01:44] <ssi> I wonder how that works with the JA branches
[13:01:44] <_methods> there's a guy that's done all this to his little x2
[13:01:53] <ssi> is it still a 9 axis limit, but more joints, or is it a 9 joint limit?
[13:02:05] <_methods> https://www.youtube.com/watch?v=b2-Kdud7eiA
[13:02:53] <CaptHindsight> I believe 9 axes but more joints
[13:07:10] <ssi> that's a pretty nice setup
[13:07:25] <XXCoder> yeah, almost had impact though
[13:07:25] <ssi> god a decent spindle like that on my VMC would be everything I'd need
[13:07:32] <XXCoder> drill was mm away from top
[13:10:12] <_methods> he's got balls droppin the end mill down in the tslots lol
[13:10:27] <_methods> thats one way to get some extra room
[13:11:21] <_methods> yeah i would imagine you could really do some work with something like that on a real vmc
[13:11:28] <_methods> with a tool changer
[13:11:32] <ssi> yep!
[13:11:38] <_methods> you could put the turning tools in spindle
[13:11:39] <ssi> could make some custom tool holders to hold lathe tools
[13:11:40] <ssi> yeha
[13:11:44] <_methods> and spindle orient
[13:11:51] <ssi> could even change rake angle that way
[13:11:54] <_methods> yep
[13:12:08] <_methods> ghetto live tooling lathe hehe
[13:12:12] <XXCoder> yeah
[13:12:32] <XXCoder> on cnc with toolchanger can add lathe tools so its live swappable
[13:17:08] <CaptHindsight> ghetto live tooling lathe or next kickstarter?
[13:17:22] <XXCoder> CaptHindsight: kickstarter kickstarter kickstarter!!!!!
[13:17:22] <_methods> oh damn
[13:17:30] <_methods> this is our chance
[13:17:34] <XXCoder> CaptHindsight: new collets
[13:17:35] <_methods> nah nm this is actually useful
[13:17:41] <XXCoder> lather holder collets
[13:17:46] <XXCoder> *lathe tooling
[13:17:48] <ssi> ooh you know what
[13:17:51] <ssi> square ER collets
[13:17:55] <ssi> would hold lathe tools
[13:17:55] <ssi> bam
[13:17:57] <ssi> live tool'd
[13:18:02] <_methods> yeah
[13:18:05] <XXCoder> it exists already?
[13:18:27] <ssi> shit yeah
[13:18:35] <CaptHindsight> _methods: could be used to make wristbands to hold end mills
[13:18:40] <_methods> hahaha
[13:18:45] <_methods> or pancake printers
[13:19:15] <XXCoder> well. first invent it, then invent really crappy use for it
[13:19:23] <XXCoder> let smart people see actual use for it
[13:19:40] <ssi> actually I'm having a hard time fining square ER collets
[13:19:44] <ssi> square 5C are common tho
[13:19:48] <_methods> yeah
[13:19:54] <_methods> i was going to say 5c
[13:20:03] <CaptHindsight> http://reprap.org/mediawiki/images/f/f2/Printrun_fr.png and be sure to use this GUI
[13:20:03] <ssi> but I have ER tooling for the VMC
[13:20:06] <ssi> cat40 ER chucks
[13:20:07] <_methods> they make cat40 5c adapter i'm sure
[13:20:12] <ssi> yeah probably
[13:20:13] <XXCoder> cool. I recently finally setup and use 4th axis. .. it was interesting.
[13:20:23] <archivist> endmills with er shape shanks so length is preset
[13:20:53] <XXCoder> heyyyy turn it around. invent stock holder for spidle
[13:21:02] <_methods> maybe not on the adapter lol
[13:21:06] <XXCoder> turn cnc mill into lathe like hmm ssi menioned?
[13:21:14] <CaptHindsight> http://www.3ders.org/articles/20150406-us-navy-researcher-3d-prints-adjustable-groove-joint-pliers.html for plastic fasteners
[13:21:24] <XXCoder> also, table mount lathe tools
[13:22:09] <CaptHindsight> "postdoctoral research associate"
[13:22:30] <ssi> lol postdocs
[13:22:57] <CaptHindsight> I guess tey don't let you have real pliers until after postdoc
[13:23:07] <ssi> you know what would have done that job perfectly adequately?
[13:23:13] <ssi> SMOOTH JAW PLIERS
[13:23:18] <ssi> like knipexb
[13:23:19] <ssi> -b
[13:23:29] <XXCoder> well research can be useful
[13:23:33] <ssi> http://www.amazon.com/Knipex-8603250-10-Inch-Pliers-Wrench/dp/B000X4OG94/ref=sr_1_1?s=hi&ie=UTF8&qid=1428516112&sr=1-1&keywords=pliers+knipex
[13:23:36] <ssi> problem solved
[13:23:38] <XXCoder> on space you can't dock and get more tools.
[13:24:06] <XXCoder> if something unexpected happened and you need certain tools, its good to know how to print it so it stands up to task,.
[13:24:33] <XXCoder> If I recall, there was first 3d print recently
[13:24:43] <ssi> and what do you do if you're in space and some dickhead used up all the plastic filament on moustaches and fake sunglasses
[13:24:43] <PetefromTn_> Wonderful some fucking asshat just phoned a bomb threat into my kids school. They evacuated and did not find anything.
[13:24:52] <_methods> buwhahahahhaa
[13:24:59] <XXCoder> not very funny
[13:25:09] <XXCoder> hope they trace and arrest that person
[13:25:20] <_methods> i was laughin at the mustache comment
[13:25:24] <XXCoder> ssi: yeah space that guy
[13:25:27] <PetefromTn_> I hope they kick his ever lovin' ass into the ground....
[13:25:28] <_methods> not bomb threat you nipplehead
[13:25:36] <ssi> lol
[13:25:48] <XXCoder> _methods: didnt see funny one yeah sorry
[13:25:56] <_methods> :)
[13:26:04] <PetefromTn_> nipplehead...teehee
[13:26:15] <_methods> haha it's my go to
[13:27:32] <JT-Shop> ssi, would 7075 be strong enough?
[13:28:12] <ssi> JT-Shop: honestly probably, but they specifically asked for steel
[13:28:30] <XXCoder> oh yeahj
[13:28:30] <ssi> it's a shaft adapter for a vibrapencil
[13:28:35] <XXCoder> did you guys ever mill tungesin
[13:28:44] <ssi> and aluminum would probably deform the bore
[13:28:46] <XXCoder> ssi: "pencil" right ;)
[13:32:00] <_methods> ssi: what's the tolerance on teh 2 through holes i'll just write it in so i can quote it
[13:32:21] <XXCoder> .0000001
[13:32:29] <_methods> and are the tapped holes thru?
[13:32:51] <_methods> i seem to remember you saying they weren't supposed to go through allt he way
[13:32:58] <XXCoder> actually if I recall, Lego uses that tolence. 0.0000001
[13:33:11] <archivist> patent pending! http://www.genswiss.com/er_taper_integrated_tiloc_tooling.htm
[13:33:31] <ssi> _methods: they can be thru or to bore... I think I drew it thru cause I figured it'd be easier to do
[13:34:03] <_methods> well less tapping is probably better with 304
[13:34:08] <ssi> the holes ought to be reamed to size, they didn't give me an actual tolerance but the .125 bore is slip fit on a shaft, the .100 bore isn't as critical but probably +/-.002 is a good target
[13:34:18] <_methods> k that works
[13:34:36] <ssi> everything else is cosmetic
[13:34:41] <ssi> OD can be barstock dimension
[13:35:04] <_methods> so is stock scale fine?
[13:35:09] <_methods> or does it need to be turned?
[13:35:12] <ssi> thread class 2B is fine, hole location +/-.01" is fine
[13:35:24] <ssi> um I think mill scale is fine
[13:35:31] <_methods> k
[13:35:36] <ssi> 304 is pretty bright in scale isn't it?
[13:36:34] <_methods> bar stock usually isn't
[13:36:58] <_methods> sheet's usually not too bad
[13:37:35] <_methods> but bar stock is usually a dull gray
[13:37:41] <ssi> that's probably fine
[13:37:45] <_methods> i might have some out there i can take a pic of
[13:37:47] <_methods> one sec
[13:37:57] <ssi> although I imagine that OD finish turning wouldn't add a ton of machine time and doesn't add an OP
[13:41:02] <_methods> i found some 303 it looks pretty good
[13:41:06] <_methods> 304 looks about the same
[13:41:46] <XXCoder> whats you guys making
[13:43:11] <_methods> http://imgur.com/akWYQ3O
[13:44:32] <_methods> http://i.imgur.com/oz2stty.jpg
[13:44:43] <_methods> sorry horrible pic trying to hold it and camera
[13:45:32] <ssi> yea that's plenty fine
[13:45:37] <_methods> k
[13:45:40] <ssi> it's bright compared to, say, A36 :)
[13:48:48] <_methods> alright i got the boss man doin a quote now
[13:48:54] <ssi> awesome thx
[13:48:57] <_methods> np
[13:49:11] <ssi> when you have it I'll call them and make sure they didn't find someone else, and if it's inline I'll get approval and have you jam it up
[13:51:56] <_methods> cool
[13:52:23] <_methods> unless someone else in here wants/needs teh work
[13:52:25] <_methods> i'll stay out
[13:54:00] <_methods> i'd give pete or jt first dibs
[13:54:36] <_methods> PetefromTn_ or JT-Shop were you going to get with ssi on his parts?
[13:55:06] <ssi> pete's back a couple weeks and doesn't have a lathe at the moment
[13:55:23] <ssi> he's making big kid parts right now :)
[13:55:27] <_methods> hehe
[13:57:56] <_methods> http://i.imgur.com/ZheHF4a.jpg
[13:58:10] <ssi> HAHAHA
[13:58:24] <_methods> haha
[13:58:28] <_methods> i have no idea if that's true
[13:58:32] <_methods> but it sure was funny
[13:58:33] <PetefromTn_> honestly I would like to do the parts but I JUST got some work to do actually several jobs for one customer and I am heading over to another customer that wants me to look at something right now. If any of these go thru I will be working for a little while on them.
[13:58:50] <_methods> right on
[13:59:03] <_methods> well i don't want to take any work away from anyone
[13:59:12] <PetefromTn_> if he decides that there is not a big rush I am working on a quote for him.
[13:59:14] <_methods> i'd rather see you guys get it then my shop get it
[13:59:38] <PetefromTn_> would be easier if my CNC lathe was working DOH!
[13:59:44] <_methods> heheh
[14:01:14] <PetefromTn_> yeah I wish I could laugh about it but it is a sore spot right now. Been approached with several jobs that would be good for it.
[14:01:21] <ssi> I feel the same way man
[14:01:24] <ssi> if I had my vmc going
[14:01:27] <ssi> or even just my hnc
[14:02:04] <FinboySlick> You guys seen this yet? https://www.youtube.com/watch?v=LVODJm05plw
[14:02:28] <_methods> heh
[14:02:32] <_methods> i've never nuked a bearing
[14:02:46] <PetefromTn_> I thought it was a bad thing to nuke metal parts?
[14:02:57] <FinboySlick> I guess dipping it in boiling water would work too.
[14:03:02] <XXCoder> PetefromTn_: pointy stuff would make plasma
[14:03:25] <XXCoder> even so, it's not very good idea
[14:03:39] <PetefromTn_> Gotta go meet a NEW potential customer right now wish me luck guys!
[14:03:45] <XXCoder> good luck man
[14:03:50] <FinboySlick> Break a leg, PetefromTn_.
[14:03:56] <PetefromTn_> ;)
[14:03:58] <_methods> go get it
[14:04:09] <PetefromTn_> BBL
[14:04:34] <mknawabi> anyone here good with isilon ACLs? :D
[14:04:48] <XXCoder> also 15 sec isnt enough for any serious issue
[14:05:06] <_methods> i don't even know what that is
[14:05:46] <_methods> mknawabi: not emc
[14:05:46] <XXCoder> any of you guys ever nuked a cdrom or dvd? you should do it once in least. only 1 or 2 seconds!
[14:05:53] <mknawabi> o shit
[14:05:54] <mknawabi> sorry
[14:05:57] <_methods> mknawabi: it's linuxcnc lol
[14:18:34] <JT-Shop> well them bad boys are done and now some of the pins are a bit tight and need to be polished
[14:57:36] <cnc1> hallo to all
[14:58:29] <_methods> hello
[14:58:35] <XXCoder> hey
[14:58:51] <cnc1> i have question
[14:59:12] <_methods> fire away
[14:59:20] <cnc1> MDI_COMMAND = O<probez> call
[14:59:47] <_methods> calling a sub in mdi?
[15:00:58] <cnc1> the make error unable to open file
[15:01:13] <_methods> are you doing this in axis gui
[15:01:17] <_methods> under the mdi tab
[15:01:23] <cnc1> sorry i'm slowly
[15:02:05] <cnc1> yes axis
[15:03:39] <_methods> i honestly don't know what you can or can't do in axis mdi
[15:03:57] <cnc1> with a button
[15:04:59] <_methods> is there like a more advanced axis gui doc than this
[15:05:02] <_methods> http://linuxcnc.org/docs/html/gui/axis.html
[15:06:00] <JT-Shop> cnc1, the file has to be in the path specified in the ini file and a few more things
[15:06:51] <cnc1> http://imagebin.ca/v/1xenx74cV9FU
[15:07:28] <_methods> ah integrator manual has more adv axis stuff i guess
[15:09:14] <JT-Shop> in the MDI tab type o<probez> call and see if it runs
[15:09:30] <JT-Shop> o<probez> call
[15:09:42] <cnc1> ok
[15:10:44] <cnc1> no
[15:10:54] <cnc1> same error
[15:11:14] <cnc1> unable to open file
[15:11:19] <JT-Shop> what directory is the file in?
[15:12:00] <cnc1> 0<probez> sub
[15:12:01] <cnc1> G38.2 Z-35 F60
[15:12:03] <cnc1> G92 Z-31.97
[15:12:04] <cnc1> G0 Z10 F400
[15:12:06] <cnc1> 0<probez> endsub
[15:13:35] <cnc1> i make this tuturial
[15:13:37] <cnc1> http://rlirc.blogspot.de/2014/03/verwendung-des-stepcraft.html
[15:14:52] <cnc1> the last section
[15:15:16] <JT-Shop> where did you put the file?
[15:16:05] <cnc1> in same folder were is ini and hal
[15:17:41] <skunkworks_> the probe.ngc needs to be in the nc_files directory
[15:18:03] <cnc1> aha
[15:18:06] <JT-Shop> The file must be in the directory pointed to by PROGRAM_PREFIX or SUBROUTINE_PATH in the ini file.
[15:18:22] <JT-Shop> and you must restart Axis so it knows it is there
[15:18:47] <skunkworks_> ^better answer
[15:26:11] <Cromaglious> here I was thinking it was a head gasket, and it was the radiator.. well just spent $92 on a radiator..
[15:26:43] <cnc1> thank to all
[15:26:50] <cnc1> thanks to all
[15:27:06] <cnc1> it's work
[16:46:23] <Deejay> gn8
[17:13:14] <furrywolf> I need entertainment for a plane trip. I was going to download mythbusters, but looks like they haven't made any new episodes in the last few months since the last plane trip. Suggestions? No stupidity.
[17:13:44] <furrywolf> non-fiction. (so no "reality tv" either)
[17:14:34] <zeeshan|2> mythbusters is so shit now
[17:14:43] <zeeshan|2> they're doing all this movie myth busting nonsense
[17:32:57] <furrywolf> it's far more interesting than most of what's on TV. Of course, that doesn't take much.
[17:35:29] <Tom_itx> look out the window
[17:35:50] <furrywolf> at cross-country cruising altitude at night, there's not much out the window.
[17:35:59] <Lowridah> secialized set of circumstances busters
[17:36:02] <Lowridah> specialized
[17:36:05] <Tom_itx> you might see some aliens
[17:40:22] <furrywolf> I might, but it's very unlikely.
[17:40:50] <adb> gn
[17:41:40] <furrywolf> so, suggestions? Something at least marginally intelligent, that's easy to download.
[17:43:10] * furrywolf doesn't usually watch tv, so doesn't know what's currently available
[17:44:11] <LeelooMinai> I watch TV series only when taking breaks to eat during the day.
[17:44:36] <LeelooMinai> But that's, I don't know, I guess "fiction" in your book
[17:46:01] * furrywolf isn't sure how watching tv during food breaks implies fiction
[17:47:35] <furrywolf> hrmm, they made a new season of survivorman? might be worth getting.
[17:47:37] <LeelooMinai> I wrote TV series...
[17:47:55] <LeelooMinai> But movie-like TV series, not shows like Mythbusters
[17:48:07] <furrywolf> ... does "TV series" mean something different there than it does in us english?
[17:48:40] <LeelooMinai> Not sure - is there a special term for the "movie-like" ones?
[17:49:12] <furrywolf> made-for-tv movie
[17:49:14] <furrywolf> ?
[17:49:46] <LeelooMinai> But that includes just... movies. I mean series that have lot's of parts:)
[17:50:16] <furrywolf> here "tv series" just means anything that airs on tv and has more than one episode, including wildlife documentaries, the simpsons, modern marvels, jersey shore,... :P
[17:50:25] <furrywolf> that is, it tells you nothing about what the show is.
[17:50:47] <LeelooMinai> Yes, right, so I am not sure how to describe the ones that are movie-like.
[17:52:26] <Lowridah> dramatic nonfiction?
[17:52:39] <furrywolf> oh goody, my favorite easy-as-fuck-to-download-from streaming site has some of the last survivorman series.
[17:53:02] <LeelooMinai> Lowridah: But movie-like includes fiction...
[17:53:23] <Tom_itx> download crockadile man
[17:53:31] <furrywolf> it's a streaming site that wants you to watch their crap, but it's plain html5, downloads at 10mbit, and doesn't seem to have any detection whatsoever that you're wgeting ten things at once. :)
[17:53:34] <Lowridah> movies are like fruit, they can be fiction or an orange, but not all oranges or fiction are fruits and movies?
[17:53:44] <Lowridah> either way just download BBC's a history of art in 3 colors
[17:53:53] * furrywolf thinks Lowridah might be a bit fruity
[17:55:28] <Lowridah> could be could be
[17:56:25] <LeelooMinai> furrywolf: Won't downloading video take 2 months on your connection? :)
[17:57:44] <furrywolf> I'll be spending a day at someone else's house before the plane trip (leaving my car there and getting a ride), can use broadband there.
[17:58:28] <furrywolf> downloading them to my webhosting now
[17:58:31] <LeelooMinai> I see, temporary return to civilisation
[17:58:45] <furrywolf> hrmm, their server is slacking, only 185K/sec now.
[18:01:02] <andypugh> Lowridah: That was rather a good series, certainly.
[18:01:30] * furrywolf doesn't much like art
[18:02:05] <andypugh> Then you can read “Pattern Recognition” for more abour International Klien Blue.
[18:03:07] <andypugh> furrywolf: It is more about colours than it is about art.
[18:03:31] <Lowridah> color and material scarcity
[18:06:32] <Lowridah> sorry, colours, it is a BBC show after all
[18:07:11] <furrywolf> hrmm, they seem to have bumped your speeds down to 185K/sec and limited you to 4 files at once now. oh well.
[18:07:47] <furrywolf> that's still >700K/sec at least. :)
[18:07:49] <andypugh> I obsessively watch anything with Lucy Worsley too.
[18:09:22] <furrywolf> nevermind, it's just that video is broken. another works fine. downloading five at once now.
[18:09:33] <furrywolf> never heard of (her?)
[18:13:15] <andypugh> Cute, clever, sexy lisp. https://www.youtube.com/watch?v=0QTRYPcaGUo
[18:14:21] <furrywolf> I have written plenty of lisp. I never found it cute, clever, or sexy. :P
[18:14:26] <furrywolf> oh, you don't mean the language. :P
[18:17:02] <furrywolf> yay, getting around 1.3MB/second. too bad I can't get them from my webserver to here at that rate. heh.
[18:17:41] <furrywolf> I downloaded something the other day from youtube to my webserver at 34MB/sec. good to know google doesn't ratelimit.
[18:18:31] <furrywolf> I then downloaded it from there to here, which took much longer. I then found out it utterly fucking sucked, and it was a waste of bandwidth twice.
[18:20:00] <andypugh> www.speedtest.net says I get 6Mb up and 106Mb down.
[18:22:44] <furrywolf> I get around 20-100K/sec down, 5-20K/sec up
[18:22:49] <furrywolf> when it's working
[18:23:17] <andypugh> Do you really mean K? And are you taking bits or bytes?
[18:24:10] <furrywolf> big-K. kilobytes.
[18:24:45] <furrywolf> when it's not working, it can range from barely usable to unusable...
[18:26:42] <_methods> 6-32 tap drill
[18:26:46] <_methods> oops
[18:27:09] <furrywolf> 7/64
[18:27:17] <_methods> correct
[18:27:18] <_methods> hehe
[18:28:46] <furrywolf> andypugh: if it were kbits, it'd be even worse. :)
[18:28:47] * _methods needs to get a tap drill chart to hang up
[18:29:21] <andypugh> It’s easy. You subtract the pitch from the diameter.
[18:29:43] <andypugh> M6 x 1 = 5mm. M25 x 2 = 23mm…
[18:30:01] <andypugh> M4 x 0.8 = 3.2mm
[18:30:29] <andypugh> As the angles are the same that should work in your quaint units too.
[18:30:56] <furrywolf> It's even easier than that. you look at a chart. :)
[18:31:11] <furrywolf> the chart will tell you the nice fraction, rather than you having to figure it out. :P
[18:31:56] <_methods> yeah well i'm havin to tap for stupid mobo standoffs that use 6-32
[18:32:05] <_methods> metric is so much mo bettah
[18:35:10] <andypugh> furrywolf: You take a digital caliper to drills in the box smaller than the screw :-)
[18:35:28] <furrywolf> lol
[18:36:00] <furrywolf> I've picked drills by just holding them up to the tap and seeing what looks about like the right one...
[18:36:46] <andypugh> So, 6 - 32 would be 5.96875 tapping size…. Wait, when things get small you use random code digits rather than a size? You are on your own then.
[18:38:23] <cpresser> andypugh: for most metric threads the subtraction method for the core-drill-diameter works. but not for all. DIN recommends a table
[18:38:33] <furrywolf> lol
[18:38:36] <cpresser> there is a table on the back of my caliper, thats handy
[18:39:15] <furrywolf> I was working on a car project that had M9 and M11 bolts. yay hard to find.
[18:39:27] <andypugh> furrywolf: Bugatti?
[18:39:35] <furrywolf> historic-ish subaru
[18:40:32] <andypugh> The baffle in a Puch maxi moped silencer is held in with an M7 nut. I have no idea why.
[18:41:36] <furrywolf> also had m10 in a very fine pitch, which is also quite hard to find. had to order them from a specialty fastener place (no, not fastenal - they couldn't get them, as always).... twice. because the first time the factory put normal fine pitch bolts in the box even though the sticker had the correct extra-fine pitch...
[18:42:31] <beebee> hi, does LinuxCNC work with r323?
[18:43:07] <furrywolf> subaru flywheel bolts are the same size and pitch as toyota driveshaft bolts, and neither is easy to find. :)
[18:43:22] <andypugh> beebee: What is r323?
[18:43:23] <_methods> rs232?
[18:43:42] <beebee> andypugh, _methods : yes rs323 sorry
[18:44:02] <beebee> 232*
[18:44:07] <furrywolf> that would explain why my googling for r323 cnc was finding exactly nothing...
[18:44:14] <andypugh> A bit. What do you actually need to do?
[18:45:54] <beebee> oops, rs323, i'm sure this time. i need LinuxCNC to speak to a rs323 controller
[18:46:35] <cpresser> beebee: we will need details on that controller. specs, datasheet, ...
[18:47:16] <beebee> conquer design and engineering
[18:47:48] <beebee> m140 i think
[18:48:02] <andypugh> What data is transferred by RS232 in this case?
[18:48:41] <furrywolf> yes another product google says doesn't exist...
[18:49:58] <beebee> furrywolf: sorry found it http://www.eaziform.co.uk/index.php?products,m401
[18:50:40] <furrywolf> It helps if you know what you have before asking about it. :)
[18:50:50] <andypugh> This is nice :-) http://www.ebay.co.uk/itm/Max-Muller-Geidermeister-CNC-Machine-Lathe-Metal-Turning-/151636899731?&_trksid=p2056016.l4276
[18:51:04] <beebee> yeah i suppose it does
[18:51:18] <beebee> it's for my uncle so i'm not 100%
[18:51:45] <andypugh> OK. LinuxCNC can’t talk to that because LinuxCNC does _exactly_ the same job as that.
[18:51:48] <beebee> so can LinuxCNC work with that controller I posted above@?
[18:51:53] <beebee> oh
[18:51:53] <furrywolf> It is very unlikely to work with linuxcnc without some coding.
[18:52:12] <beebee> how can a piece of software do the same job as a piece of hardware?
[18:52:12] <andypugh> You can _replace_ that with LinuxCNC
[18:53:13] <andypugh> Well, You also need some hardware to be controlled by LinuxCNC. Anything from parallel port (yes, we know) to an FPGA IO card
[18:53:17] <beebee> andypugh: that doesn't make sense.. did that link work? it's a piece of hardware
[18:53:42] <furrywolf> I'm sure linuxcnc could be made to work with it, but I'm not aware of a driver for it.
[18:53:43] <andypugh> It is a piece of hardware with an embedded CNC controller in firmware. LinuxCNC is a CNC controller.
[18:54:20] <beebee> so is that hardware made for windows?
[18:54:37] <furrywolf> is it a full controller, or just a pulse generator?
[18:54:58] <andypugh> I am not sure. I suspect that the M401 expects to be sent G-code as text
[18:54:59] <beebee> pulse generator
[18:55:44] <beebee> it came with software called EaziCNC
[18:55:47] <furrywolf> yeah, if it has a controller built-in, then you get to hack the firmware too... :P
[18:56:34] <andypugh> No, it is more than a pulse generator as it does motion control. (Bezier curves implies multi-axis coordinated motion)
[18:56:55] <furrywolf> is this a card you have, or one you were thinking of getting?
[18:57:05] <beebee> card I have
[18:57:20] <beebee> it works with EaziCNC on windows but it stutters a lot
[18:57:25] <andypugh> So, the M401 does pretty much the same thing as LinuxCNC + Parallel Port (probably better, as the Parallel port is a very sub-optimal solution)
[18:58:58] <beebee> would you recommend a USB controller to work with EaziCNC?
[18:59:19] <beebee> or maybe even better, with LinuxCNC
[18:59:39] <furrywolf> Mesa boards are generally suggested for linuxcnc
[18:59:41] <andypugh> If you want something that would be an upgrade from the M401 then you would be looking at installing Linux, realtime kernel, and probably an FPGA card such as the Mesa 5i25 - 7i76 combo.
[18:59:56] <furrywolf> there are no realtime usb solutions that I know of, just attempts that may not have worked...
[19:00:56] <andypugh> beebee: You would be looking at $200 to replace that M401 with a llinuxCNC-supported equivalent. http://store.mesanet.com/index.php?route=product/product&path=83_84&product_id=215
[19:01:51] <andypugh> But it seems to me that the M401 ought to work without stuttering, and the first port of call should be the manufacturer to ask why their product doesn’t work.
[19:02:55] <beebee> andypugh: so linux + Mesa 5i25 - 7i76 would be realtime?
[19:03:05] <andypugh> Yes.
[19:03:32] <beebee> m401's motion seems stuttery
[19:03:49] <beebee> the cuts are fine, it's just not smooth
[19:03:50] <andypugh> With one caveat: You need a PC motherboard that can run well with the realtime kernel.
[19:04:04] <Tom_itx> but they're only about $60
[19:04:57] <furrywolf> you don't need a good motherboard to use a mesa board. only parallel port is really picky...
[19:05:06] <andypugh> As a first step download the LinuxCNC LiveCD, boot from the CD and run a latency test. You can do all this without making any changes to the Windows installation.
[19:05:19] <SpeedEvil> new
[19:05:20] <furrywolf> you can run a mesa board with >1000us jitter. :)
[19:05:34] <SpeedEvil> (but it probably makes sense to get a new mb)
[19:05:36] <beebee> the computer can run ubuntu. i've also read the ethernet or usb controllers can fun smoother
[19:05:44] <beebee> *be
[19:06:11] <beebee> i think windows is interrupting the smoothness of the steps
[19:06:27] <furrywolf> if the card operates like their website claims, the smoothness is entirely in hardware
[19:06:33] <andypugh> The USB controllers (as supprted by Mach3) are very much like the M401. They are a complete motion controller in a box being sent high-level commands from the Windows software.
[19:08:35] <andypugh> beebee: Windows shouldn’t be able to interfere with M401 steps. It is more likely that the windows software is starving the motion queue so that the motion stops ocassionally. This would fit with the parts coming out well. If the step-pulses were interrupted between moves then the motors would tend to stall and not recover. So it sounds more like the M401 is stopping and waiting between moves.
[19:09:12] <beebee> andypugh: that's exactly it
[19:10:20] <furrywolf> or their controller has a piss-poor look ahead algorithm...
[19:10:49] <furrywolf> brain-damaged path planning stops between every command
[19:10:53] <andypugh> So, in that case you could try to see what the problem is with the windows system. Or you could just accept that things sound worse that they are and carry on as you are and make parts :-)
[19:11:14] <beebee> the machine it's controlling is a precsix404
[19:11:48] <andypugh> LinuxCNC and 5i25 - 7i76 will work better. And you can hack it to do anything you like. But there will be a learning curve and down-time.
[19:12:18] <beebee> are there tests i run to find out where the issue is and come back with more info?
[19:12:21] <andypugh> So if you are making money from the current system it isn’t a trivial decision to make.
[19:12:22] <furrywolf> or an ethernet mesa card if you like small cables
[19:12:45] <furrywolf> beebee: you can send g-code directly to your controller with any terminal program, bypassing their included software entirely.
[19:13:24] <furrywolf> my suspicion would be poor firmware on their part, or poor g-code on yours (such as generated g-code with ten billion tiny moves, that the serial port can't send fast enough).
[19:14:00] <andypugh> If you ask on the Mach3 forums they will tell you to buy a Smoothstepper and Geckos and that you will be back up and running in a week. I will tell you that you should bet on a month to get up to speed again, either with Mach3 or LinuxCNC.
[19:14:33] <beebee> i generate smooth vectors with very little nodes created in aspire
[19:14:44] <andypugh> furrywolf: Yes, tiny moves that the machine can complete before the serial port can send them seems very likely.
[19:14:50] <furrywolf> how big is your generated g-code file?
[19:15:58] <beebee> furrywolf: 6600 lines for a silhouette of a cat
[19:16:01] <andypugh> beebee: Can you change the serial data rate? It might be instructive to halve it to see if things get worse as a diagnostic step (doubling it to see if it gets better is probably less diagnostic)
[19:16:02] <beebee> i can post the gcodfe here
[19:16:12] <furrywolf> and how long does it take to cut it?
[19:16:40] <beebee> furrywolf: 12minutes
[19:17:00] <furrywolf> ok, so g-code sending shouldn't be the limit, then... that's only a few lines a second.
[19:17:33] <furrywolf> do they have commands to switch between exact stop and any kind of trajectory planning?
[19:18:01] <beebee> i'm not sure. i can post the file here.
[19:18:28] <furrywolf> it'll be in their programming/configuration manual, which I'm too lazy to download again. :)
[19:19:27] <furrywolf> ah-ha! apparantly splines are handled by the software, the firmware doesn't natively support them. you said you were generating splines?
[19:20:46] <furrywolf> do you have your accelerations set correctly?
[19:21:03] <beebee> what are splines?
[19:21:11] <furrywolf> nevermind, you said vectors
[19:21:20] <furrywolf> andy mentioned splines earlier
[19:21:25] * furrywolf is doing too many things at once
[19:21:55] <beebee> yes I mentioned vectors
[19:22:17] <beebee> could it be that i haven't configured aspire correctly?
[19:22:52] <andypugh> I think what we are saying is that despite being LinuxCNC enthusiasts, we think that your existing system ought to work, and that switching to our pet project is likely to be more effort for you. My feeling is that you wouldn’t get quite such a balanced view from the Mach3 crowd. This is based on advising someone doing a retrofit on his commercial machine that it would take at least 2 months before he was making money a
[19:22:52] <andypugh> with LinuxCNC (keeping existing drives and motors). And the Mach3 forums saying that swapping all his servos for steppers, Geckos, Smoothstepper and Mach3 would see him back making money in 2 weeks. Last time I looked the poor chap was 6 months in and still not making parts :-(
[19:22:56] <furrywolf> I know absolutely nothing about aspire, including what it is.
[19:23:31] <furrywolf> ... replace servos with steppers? I think I'll sell my car and get a horse too.
[19:24:46] <andypugh> beebee: If you are looking at Conqueror products, are you in the UK?
[19:24:58] <beebee> andypugh: Ireland
[19:25:10] <andypugh> Our bit or the other bit :-)
[19:25:22] <furrywolf> looking at their programming guide, the card isn't that bright... even circular interpolation is handed off to the software. it does talk g-code, not proprietary, so their software must be splitting every line needed interpolation into many simply g0s... which could be a lot of them.
[19:25:44] <beebee> andypugh: ROI
[19:26:34] <beebee> the other bit :P
[19:28:16] <andypugh> I have been there. A few times actually. Mainly the underground bits.
[19:29:14] <furrywolf> G5P1 and G5P0 might cause weird stuttering issues, but it sounds like their software should be setting them automatically
[19:29:50] <andypugh> The only bit of Ulster I have visited is Derry/Londonderry/Doire which is a place with such complicated politics that they can’t even decide what to call it.
[19:29:55] <furrywolf> reading their programming guide, brain-damaged firmware seems increasingly likely. :)
[19:30:51] <andypugh> Anyway, yes, it seems likely that your nice curves are becoming short segments and starving the queue.
[19:30:52] <furrywolf> if you do G5P1, it'll queue multiple moves... but it ignores acceleration, and will break (lose steps) if you send it any kind of sudden direction change. G5P0 will respect the acceleration settings... but comes to a complete stop between every command? wtf?
[19:31:48] <andypugh> If that is true then I recant my suggestion of persisting with the M401 and insist that you convert to LinuxCNC immediately :-)
[19:33:03] <beebee> i've been looking a few options. smoothstepper and cslabs ethernet controller. also i don't use mach3. mach3 won't work with the m401.
[19:33:32] <beebee> didn't you say LinuxCNC won't talk to the m401?
[19:33:54] <furrywolf> correct. linuxcnc will NOT talk to the m401.
[19:34:15] <andypugh> Yes. If you switch to either Mach3 or LinuxCNC then you will have to ditch the M401.
[19:34:36] <beebee> so should I look at the mesa combo and LinuxCNC?
[19:34:46] <Tom_itx> it's a good combo for sure
[19:34:49] <andypugh> But from what furrywolf has said that is where the problem is
[19:35:07] <andypugh> You could initially look at LinuxCNC and the parport.
[19:35:19] <andypugh> Do you have a spare PC?
[19:35:44] <furrywolf> the manuals mention newer and older versions of cards, and contradict themselves repeatedly.
[19:36:01] <beebee> andypugh: yes I do have a spare PC
[19:36:06] <andypugh> You could set up a LinuxCNC parport system to drive the parts downstream of the M401 without disabling the machine.
[19:36:33] <furrywolf> http://www.eaziform.co.uk/files/ProgrammingGuide.pdf you can see what you make of the G5 command. :)
[19:37:04] <andypugh> You may find maximum speed limited by the parport, but you would at least get some confidence with the software for very little capital outlay.
[19:37:57] <furrywolf> how the hell do they have a 32mhz processor and they can't do circular interpolation on-card?
[19:38:08] <SpeedEvil> furrywolf: Circles are hard.
[19:38:42] <furrywolf> not that hard. :P
[19:39:38] <furrywolf> "Full 4-axis motion support including cubic and Bezier splines" is one of their features... but if you read the manual, splines are done in software... and circular interpolation... and imperial units... and etc. heh.
[19:39:41] <andypugh> beebee: Euro-source for Mesa is: http://eusurplus.com/index.php?route=product/category&path=63 or www.mesanet.eu (the former is someone active on the forums)
[19:39:53] <Tom_itx> you might be setting up the configs twice though
[19:40:59] <andypugh> In _theory_ you could hack the right layer of LinuxCNC to send motion control commands to the M401. But I am not sure what that would achieve in the end.
[19:41:53] <furrywolf> it'd achieve an ugly hack? :)
[19:42:24] <beebee> 7I76-5I25 + LinuxCNC would work on any computer?
[19:42:41] <Tom_itx> any computer with decent latency
[19:42:46] <cpresser> most computers
[19:43:22] <andypugh> beebee: No. You need a PCI port for the 5i25. You can use a 6i25 in a PCIE port: http://eusurplus.com/index.php?route=product/search&search=7i76&sub_category=true&description=true
[19:43:28] <Tom_itx> you can burn the iso and run a latency test
[19:43:41] <andypugh> (the cards are otherwise identical)
[19:44:12] <furrywolf> again, you don't need good latency for a mesa card.
[19:44:23] <andypugh> You need good-ish latency.
[19:44:33] <Tom_itx> you still need to check it
[19:44:44] <furrywolf> pcw says they'll work fine even if you have to do a 2ms thread. :)
[19:44:48] <beebee> my computer has free PCI slots
[19:44:48] <andypugh> The 6mS I got with a RaspPi wasn’t usable :-)
[19:45:31] <furrywolf> lol
[19:45:32] <beebee> can I run a latency test on windows XP?
[19:45:39] <Tom_itx> no
[19:45:50] <Tom_itx> you can run from the live cd though
[19:46:02] <andypugh> Start cheap. Download the DVD, try a parallel port config driving existing hardware. if it looks good then go for the Mesa kit.
[19:46:14] <Tom_itx> oh yeah... it's a DVD now
[19:46:19] <beebee> I see, I'll run the test. what is an example of a good score?
[19:47:13] <andypugh> Anything under 20,000nS is good for software stepping. Less than 100,000nS is good for the Mesa card, but you might be able to get away with more.
[19:47:15] <beebee> i'm not using parport, just rs323
[19:47:34] <andypugh> Do you have a parport on the spare PC?
[19:47:39] <Tom_itx> rs232 is a no go
[19:47:40] <furrywolf> less than 2,000,000ns is fine according to pcw
[19:47:56] <furrywolf> rs232 is pretty slow...
[19:47:57] <beebee> andypugh: yes I do
[19:48:22] <Tom_itx> furrywolf you need the IO pins
[19:49:50] <andypugh> So, first boot that machine from the DVD and do the latency test. And have a play around with LinuxCNC and the simulator configs. If that all looks OK then go for the Linux install and try a parallel port config to drive your existing hardware, then if that looks good, get the Mesa or Pico or General Mecahtronics or…. stuff.
[19:50:53] <andypugh> Unless you just want to go for it in one go and €200 isn’t a big deal, of course :-)
[19:53:19] <andypugh> I never like to encourage anyone to rip into a working machine.
[19:53:22] * furrywolf still can't afford a 7i76e
[19:54:03] <beebee> i'm gonna need to look under the hood and see exactly what's going on. i'll come back with better information sometime. thank you very much everyone for your help :~)
[19:54:58] <andypugh> That’s why we hang out here.
[19:55:44] <Tom_L> andypugh, what have you torn into lately?
[19:55:57] <andypugh> My thumb?
[19:56:03] <Tom_L> hah
[19:56:07] <beebee> :)
[19:56:49] <andypugh> I performed an autotrephination last week. http://www.instructables.com/id/How-To-Drain-Your-Subungual-Hematoma-blood-pocke/
[19:57:37] <Tom_L> hot needle or drill bit?
[19:57:51] <andypugh> Strange how drilling a hole in yourself seems like a really bad idea until it starts to feel like a really _good_ idea.
[19:58:09] <andypugh> 0.8mm drill bit, twiddled by hand.
[19:58:17] <Tom_L> until it grabs the last little bur and digs in a bit too far
[19:58:42] <andypugh> Hence the hand-twiddling.
[20:00:10] <Tom_L> i notice he has the bourbon close by
[20:00:36] <Tom_L> a little for the wound and alot for the gullet
[20:03:28] <andypugh> I injure myself in minor ways all the time, and nearly everything stops hurting after 20 minutes or so. The thumb was a whole new experience, it got worse and worse for 24 hours until I drilled the hole.
[20:04:59] <andypugh> I wonder how I can find someone locally with a planer? I want to re-machine a lathe bed.
[20:05:37] <andypugh> Not a common machine to find nowadays, but exactly what I need.
[20:07:35] <SpeedEvil> I have a planar thicknesser :)
[20:08:04] <andypugh> Not _quite_ the machine I have in mind
[20:09:23] <andypugh> Maybe this one? http://www.lathes.co.uk/milnesplaner/page3.html
[20:10:38] <furrywolf> one of my favorite tv moments was some construction show, where someone was framing a house... and slipped with the nail gun. put a framing nail clean through his thumb. grabbed the vise grips from his toolbelt, yanked the nail back out, kept on working...
[20:10:39] <furrywolf> bbl
[20:26:34] <_methods> yeah good luck finding a shop with a planer anymore
[20:27:02] <_methods> i think there is one shop in my area that has one
[20:27:08] <_methods> at the ship yards
[20:30:21] <andypugh> Stuart S has a _huge_ sectional one in storage
[20:34:18] <malcom2073> Oh man, that's what I need
[20:34:28] <_methods> what size part are you wanting to plane?
[20:35:13] <malcom2073> Here I am asking about surface grinding huge parts, that seems like a much easier option
[20:35:33] <_methods> good luck finding one
[20:35:40] <malcom2073> Yeah heh
[20:36:15] <malcom2073> Man that looks perfect though :P
[20:36:24] <malcom2073> Like a giant shaper
[20:36:41] <_methods> shaper and planer are analogous
[20:37:17] <malcom2073> Not seen many shapers this size though: https://www.youtube.com/watch?v=w4kPDhrhlgk&NR=1
[20:37:50] <_methods> well usually a large shaper is called a planer
[20:38:02] <malcom2073> Makes sense
[20:38:06] <_methods> but they are same thing
[20:38:59] <andypugh> No, they are very different
[20:39:03] <_methods> heh when a part comes loose on one of those bad boys
[20:39:16] <malcom2073> Doesn't a shaper usually have a moving tool?
[20:39:26] <andypugh> A shaper has a moving tool, a planer has a stationary tool and moves the workpiece
[20:39:52] <_methods> ok sounds good to me
[20:41:10] <andypugh> I posted a picture earlier of a much bigger one.
[20:42:02] <andypugh> And I think that the one that SWS has is 100’ if he puts all the bed sections together. Though that is a travelling-bridge plano-miller really.
[20:42:14] <_methods> i don't even see them at auctions often anymore
[20:42:19] <_methods> they are scarce
[20:43:40] <_methods> i keep gettin out bid on keyseaters too
[20:56:53] <furrywolf> I have a planer... of course, it's the small wood type, not the giant anything type.. :P
[20:57:26] <furrywolf> mine does wood 15" wide by 6" thick.
[21:03:45] <andypugh> That’s technically a horizontal miller :-)
[21:04:05] <andypugh> I have machined wood on a shaper :-)
[21:04:37] <furrywolf> well, it's not THAT small... >400lbs
[21:05:05] <furrywolf> not including the dust vacuum system
[21:05:16] <furrywolf> of course, I don't have a shop, so it lives under a tarp in the driveway.
[21:05:48] <SpeedEvil> I'm pondering getting a lot of 3*2 timber, and making me up some shop.
[21:05:59] <andypugh> I made a Backgammon set by machining accurate lozenges in wood glueing together, machining square, then slicing the block up to make very thick marquetry.
[21:06:05] <furrywolf> 3x2?!
[21:06:26] <SpeedEvil> furrywolf: trusses
[21:08:06] <furrywolf> why would you use an odd size? or is that common there?
[21:08:57] <SpeedEvil> It's stiff enough, and significantly cheaper than 4*2
[21:09:53] <furrywolf> it's impossible to find here. I tried finding a piece once, tried every lumber yard in the area... finally found one that had it, but only in cedar. inside the walls here, there is ONE cedar stud. lol
[21:10:03] <SpeedEvil> :)
[21:10:16] <furrywolf> I was enlarging the water heater box and needed smaller wood to make more room
[21:10:40] <SpeedEvil> http://www.mkmbs.co.uk/Menu/timber-%7Band%7D-sheet-materials/List/type:carcassing~rough-sawn_contentmode:list#/contentmode:thumb
[21:11:08] <SpeedEvil> - 30% higher or so than I'm paying when I login - but the best I've found :/
[21:11:22] <SpeedEvil> It's at least usable for worst-case budgeting
[21:12:40] <furrywolf> 2x4 is cheap here
[21:12:45] <furrywolf> it's what everything is built with
[21:13:39] <SpeedEvil> yeah - 2*4 is overkill for most of the structure
[21:14:05] <SpeedEvil> 60cm centres, single-story structure, 2.4m or so high - it's really not needed
[21:15:14] <SpeedEvil> Even doing the numbers for 80MPH winds
[21:16:31] <furrywolf> 2x4s are probably required by law here... I've never checked the codes (always used them for everything), but there's all sorts of requirements for earthquake-proofing...
[21:16:42] <Tom_itx> 70mph will knock down hwy signs
[21:16:54] <furrywolf> Tom_itx: no, it won't.
[21:17:13] <Tom_itx> i can guarantee it will. i have first hand experience
[21:17:14] <furrywolf> not unless you have really shitty, dangerous signs. :P
[21:17:21] <Tom_itx> I beam
[21:17:30] <Tom_itx> bent over like a twig
[21:17:39] <furrywolf> ok... we get gusty winds over that here, and they survive. also, a close-passing truck will give a wind blast pretty close to that...
[21:17:50] <Tom_itx> we get sustained high wind here
[21:18:15] <SpeedEvil> furrywolf: this is technically a shed
[21:18:30] <SpeedEvil> No foundations
[21:18:48] <SpeedEvil> The roof will have a lot of weight around the edges
[21:20:28] <furrywolf> I've seen videos of hurricanes etc, and it takes a lot before the signs go...
[21:20:57] <furrywolf> then again... i-beams? who the heck uses ibeams? caltrans uses PT 4x6 for small signs. :)
[21:43:08] <zeeshan|2> hi andypugh
[21:44:13] <andypugh> hi
[21:44:17] <zeeshan|2> :)
[21:44:22] <andypugh> not here much longer
[21:44:27] <zeeshan|2> damn it!
[21:44:36] <zeeshan|2> damn the time zone difference
[21:45:12] <andypugh> In fact, I have to go now.
[21:45:20] <andypugh> Bye :-)
[21:45:20] <zeeshan|2> cu
[22:18:32] <furrywolf> now that I agree with.
[22:25:10] <zeeshan|2> ive about had it with this polymer shit
[22:25:46] <furrywolf> it's ok. graduating is overrated.
[22:25:57] <zeeshan|2> dude
[22:25:59] <zeeshan|2> i did the same experiment
[22:26:04] <zeeshan|2> and im getting a completely differennt result
[22:26:17] <zeeshan|2> i understand viscoelastically from a theoretical point of view
[22:26:26] <zeeshan|2> but i wasn't applying it into my experiments
[22:27:03] <furrywolf> time to update your model, then. :)
[22:28:21] <furrywolf> what result did you get this time?
[22:28:33] <zeeshan|2> do you know about viscoelasticity
[22:29:41] <furrywolf> it's what makes my tempurpedic mattress so nice and comfy.
[22:29:51] <zeeshan|2> well you know how regular steel
[22:29:56] <zeeshan|2> has a modulus of elasticity yea?
[22:30:10] <zeeshan|2> you put a force on it, and it deforms an x amount, and if you remove that force, it returns
[22:30:22] <zeeshan|2> almost instantaneously
[22:30:59] <zeeshan|2> with polymers, and other viscoelastic materials, the time duration of the force will determine it's overall deformation
[22:31:09] <zeeshan|2> so if you put 100lb for 5 min you might stretch it twice
[22:31:22] <zeeshan|2> you put it it for 1 hour it might stretch three times
[22:31:33] <zeeshan|2> they are always moving
[22:31:42] <furrywolf> yes
[22:32:07] <zeeshan|2> adding heat is a solution to speed up the time it takes to get to its final deformation
[22:32:12] <furrywolf> and your polymer isn't behaving like an ideal shape memory polymer, and is instead exhibiting viscoelastic effects too?
[22:32:14] <zeeshan|2> but then you activate another mode ! creep!
[22:32:25] <zeeshan|2> no my smp is being a smp
[22:32:31] <zeeshan|2> but the polymer im comparing it to is being a bastard
[22:32:44] <zeeshan|2> it's not supposed to be recovering back to its original shape and it is!
[22:32:51] <furrywolf> lol
[22:33:06] <furrywolf> compare it to, say, wet clay instead? :)
[22:33:14] <zeeshan|2> haha
[22:35:07] <furrywolf> "To improve model reference accuracy, a commercially available kaolinite-based material was used as a surrogate for a polymer exhibiting minimal viscoelastic effects under the experimental parameters"
[22:35:23] <zeeshan|2> dude this is the thing.
[22:35:39] <zeeshan|2> either someone gave me the wrong film
[22:35:44] <zeeshan|2> or something fucked up is going on
[22:36:02] <zeeshan|2> all i know is im fed up
[22:36:04] <zeeshan|2> :/
[22:36:21] * furrywolf has no idea what this film is, or enough of an idea of what happened to answer that
[22:37:18] <furrywolf> I do, however, know that I've read enough crap to produce sadly-realistic overstuffed fake-educated-sounding writing...
[22:37:27] <pcw_home> I once worked for a company that made a asphalt tester (resilient modulus apparatus )
[22:37:28] <pcw_home> it squeezed samples and measured the expansion at right angles to the applied force
[22:37:59] <zeeshan|2> lol furrywolf
[22:38:41] <zeeshan|2> why is resilience important for asphalt?
[22:38:53] <pcw_home> lifetime i think
[22:39:28] <zeeshan|2> pcw_home: i might have pretty much design a micro indentor apparatus soon
[22:39:37] <zeeshan|2> it's going to be crazy challenging for me
[22:39:45] <zeeshan|2> i have access to a very precise load cell
[22:39:54] <zeeshan|2> im thinking of using servo control and a glass scale
[22:40:03] <furrywolf> precise load cells are a lot of work. long springs are easy.
[22:40:05] <zeeshan|2> along with ball screw
[22:40:27] <zeeshan|2> and then temperature controll will be used using a combination of peltiers and a heating element
[22:40:47] <furrywolf> asphalt never really hardens, and slowly squishes flatter and wider with time... some batches moreso than others.
[22:41:01] <zeeshan|2> and this whole thing will be enclosed in an insulated box with the top face of the box being a quartz glass window
[22:41:05] <furrywolf> isn't a heating element a bit redundant when you have peltiers?
[22:41:05] <pcw_home> probably precise timing also for viscoelastic materials
[22:41:07] <zeeshan|2> and then ill nitrogen purge the chamber
[22:41:29] <zeeshan|2> furrywolf: the problem is with peltiers you still need to release the heat from the hot junction when youre trying to cool
[22:41:47] <zeeshan|2> so if i have a heat pipe or some sort of other mechanism to remove heat
[22:41:50] <pcw_home> yeah you just run Peltiers from a hbridge
[22:42:00] <zeeshan|2> it'll just be designed for removing heat
[22:42:04] <pcw_home> (or add heat when reversed)
[22:42:09] <furrywolf> zeeshan|2: that effect works to your advantage
[22:42:24] <zeeshan|2> but the thing is my peltiers will be wedges within an aluminum block
[22:42:26] <zeeshan|2> 2 half plates
[22:42:35] <zeeshan|2> (atleast thats how im imagining it)
[22:42:55] <zeeshan|2> like do you guys really think i could reverse the polarity fast enough
[22:42:57] <zeeshan|2> to get precise temp control?
[22:43:00] <zeeshan|2> within a c?
[22:43:02] <zeeshan|2> 1 c
[22:43:09] <pcw_home> yep
[22:43:11] <furrywolf> during cooling, the hot side gets hot. when you want to heat, the heat is stored on the former hot side, and you get an initial period of extra-efficient heating. when you go to cool, the new hot side is already nice and cold, so you get an initial period of much more efficient cooling.
[22:43:48] <furrywolf> 1C is coarse. you could get much more precise control.
[22:43:54] <zeeshan|2> hmm
[22:43:57] <zeeshan|2> i will look up designs that do this
[22:44:40] <zeeshan|2> so basically my control parameters are: load, displacement, temperature, time
[22:45:08] <zeeshan|2> i might drop this whole idea if it's too complex to do over 4 months
[22:45:10] <pcw_home> PCR machines use Peltiers for this
[22:45:15] <furrywolf> for an application requiring fast thermal cycling, put a lot of thermal storage (i.e. a block of copper) on the reject side. when cooling, you store heat here, that you can reuse for heating, which "stores cold", that you can reuse for cooling...
[22:46:04] <zeeshan|2> http://i.imgur.com/FMw425E.png
[22:46:18] <zeeshan|2> all the faces that you see there with the exception of the top
[22:46:21] <zeeshan|2> will be insulated though
[22:46:31] <zeeshan|2> so i guess the copper will really be hot
[22:46:46] <zeeshan|2> i think ill need to have heat pipes coming out
[22:47:04] <furrywolf> if you're testing a very thin film, the ambient air will have quite an effect, not just the material it's on.
[22:47:19] <zeeshan|2> thats why ill have a fan
[22:47:24] <zeeshan|2> circulating on the temp control side
[22:47:32] <zeeshan|2> thats pretty much the standard thing these guys do
[22:47:44] <furrywolf> just checking. :P
[22:47:53] <zeeshan|2> concept :P
[22:48:06] <zeeshan|2> is the voltage draw of a peltier constant?
[22:48:12] <zeeshan|2> of temperature dependant
[22:48:17] * zeeshan|2 doesnt know anything :)
[22:48:19] <furrywolf> is "voltage draw" a meaningful term?
[22:48:29] <zeeshan|2> well what im trying to understand is
[22:48:39] <zeeshan|2> if one side of the chip is 10C and another is -10C
[22:48:43] <furrywolf> peltiers work at any voltage. there is no minimum Vf.
[22:48:45] <zeeshan|2> and then in case 2, its 20C and -20C
[22:48:57] <zeeshan|2> what is the electrical consequence on the peltier
[22:49:09] <furrywolf> however, when there's a temperature differential, they'll fight you by generating voltage of their own.
[22:49:29] <pcw_home> they are basically high voltage Thermocouples
[22:49:52] <pcw_home> (externally anyway)
[22:49:56] <furrywolf> if you're not doing simple off/hot/cold switching, you'll probably want to drive it off a constant-current supply.
[22:50:12] <furrywolf> hrmm, or maybe constant-voltage... I don't remember. lol
[22:50:29] <zeeshan|2> whats the lowest temp i can hit on the cold side?
[22:50:31] <zeeshan|2> -20?
[22:50:34] <zeeshan|2> -50000000
[22:50:39] <zeeshan|2> absolute 0?
[22:50:39] <zeeshan|2> :D
[22:51:15] <pcw_home> They stack them for really low temps
[22:51:28] <zeeshan|2> id like to hit -20c
[22:51:31] <furrywolf> no, the efficiency drops as the differential increases, to the point where all you get is more heat, no more cooling.
[22:51:47] <zeeshan|2> okay so that means that the hot side definitely needs heat pipes
[22:51:49] <furrywolf> -20C shouldn't be too hard with good cooling of the hot side and a well-insulated cold side.
[22:51:51] <pcw_home> Yeah how big a sample do you have?
[22:51:53] <zeeshan|2> going to a heat exchanger
[22:52:03] <zeeshan|2> they will be 3"x3"
[22:52:15] <furrywolf> heat pipes are hard. water is hoses is easy.
[22:52:18] <zeeshan|2> in a chamber with nitrogen with a space of like
[22:52:32] <zeeshan|2> 1" tall by 3"x3"
[22:52:35] <furrywolf> especially since you can buy a commercial cpu water cooler complete with peltier mount and save some fab time
[22:52:44] <zeeshan|2> furrywolf: can't use liquids
[22:52:53] <zeeshan|2> if i could, i woulda just used liquid nitrogen
[22:52:57] <zeeshan|2> and a heating element
[22:53:06] <zeeshan|2> wait, why am i not doing that?
[22:53:08] <zeeshan|2> :-)
[22:53:17] <furrywolf> my test with a peltier showed the cold side, with a ~2x4" aluminum plate on it, would grow about 3/8" of frost.
[22:53:29] <zeeshan|2> nice!
[22:53:34] <furrywolf> from a $6 ebay 100W peltier
[22:54:00] <zeeshan|2> our enviromental chambers used pid to control injection of liquid n2 and heating of an element
[22:54:04] <zeeshan|2> to get desired temp
[22:54:10] <zeeshan|2> they can easily hold within .2C
[22:54:26] <zeeshan|2> *use
[22:54:44] <furrywolf> yeah, know what? precision control of liquid n2 is a lot harder than precision control of a peltier. :P
[22:55:02] <pcw_home> We used to have a environmental chamber that used CO2
[22:55:04] <zeeshan|2> for me it might be easier
[22:55:10] <zeeshan|2> cause i know how heating elements work
[22:55:14] <zeeshan|2> and its a matter of switching it on and off
[22:55:26] <zeeshan|2> and on the cold side, its a matter of switching the liquid n2 solenoid on and off
[22:55:46] <zeeshan|2> pcw_home: this takes c02 but then you're limited to like -30C
[22:55:52] <pcw_home> same with Peltier except you can heat and cool
[22:55:53] <zeeshan|2> with liquid n2 i think you can hit -80C
[22:56:09] <zeeshan|2> yea ill loook it up
[22:56:16] <furrywolf> if the problem is space, can a large fansink simply be mounted 6" away from the sample box with a heavy copper bar connecting it to the peltier?
[22:56:17] <zeeshan|2> maybe i can buy a controller for it
[22:56:40] <zeeshan|2> furrywolf: thats what i meant by heat pipes
[22:57:01] <furrywolf> 2" wide by 1/4" thick nickel-plated copper bar is widely available, new and scrap. it's used for bus bars in big electrical things.
[22:57:08] <zeeshan|2> i guess with heat pipes you'd have a medium that transfers heat
[22:57:29] <zeeshan|2> lemme show you the pic of the area im constrained to
[22:57:41] <zeeshan|2> http://i.imgur.com/DhfLYxA.jpg
[22:57:43] <pcw_home> we used it for -40 (dry ice is -78C)
[22:58:12] <furrywolf> oh. a heat pipe is a specific device, using evaporation of a suitably choosen refrigerant at hot points that condenses at cold points, with wicking in a layer of some other material to move it back to the hot side... not a copper bar. :P
[22:58:21] <zeeshan|2> furrywolf: youre right
[22:58:36] <zeeshan|2> copper bar would be easier to implement
[22:58:37] <furrywolf> hence why I said heat pipes are hard.
[22:58:41] <furrywolf> copper bars are easy. :P
[22:58:53] <furrywolf> oh, you have tons of room.
[22:59:01] <zeeshan|2> just limited to the Z height
[22:59:11] <zeeshan|2> i tried using quartz glass to see how much it distorts the images
[22:59:12] <zeeshan|2> it doesnt at all
[22:59:15] <furrywolf> mill a pocket in the bar for the peltier
[22:59:18] <pcw_home> what maximum temperature do you need (Peltiers are pretty easy to melt)
[22:59:19] <zeeshan|2> it just changes the index of refraction a bit
[22:59:24] <zeeshan|2> pcw_home: 80C
[22:59:38] <furrywolf> measuring through glass might tweak the z measurements slightly...
[22:59:40] <zeeshan|2> 100C would be nice
[22:59:48] <zeeshan|2> furrywolf: i tested that today
[23:00:24] <zeeshan|2> from my memory i recall an insignficant difference
[23:00:24] <furrywolf> as long as you look perfectly straight, you're probably ok...
[23:00:26] <zeeshan|2> like 5 micron
[23:00:46] <zeeshan|2> will the peltier melt at 100c?
[23:01:00] <furrywolf> for cooling the hot side, cpu cooler fansinks are dirt cheap and perfectly sized.
[23:01:22] <zeeshan|2> furrywolf: i was thinking of brazing an array of them
[23:01:22] <zeeshan|2> haha
[23:01:32] <furrywolf> ... you don't need that much cooling.
[23:01:34] <zeeshan|2> if i can demonstrate this stupid polymer bullshit
[23:01:42] <zeeshan|2> they'll approve the funding for this device
[23:01:50] <zeeshan|2> furrywolf: make it big or go home
[23:01:53] <furrywolf> you probably will have a 30W continuous heat load tops...
[23:02:47] <furrywolf> you might not even need fans.
[23:03:05] <zeeshan|2> better to over design it
[23:03:09] <zeeshan|2> and just turn off a fan
[23:03:13] <zeeshan|2> vs have to add crap later on
[23:03:18] <pcw_home> https://tetech.com/peltier-thermoelectric-cooler-modules/high-temperature/
[23:03:25] <furrywolf> no, it's better to have it done within 4 months.
[23:03:33] <pcw_home> (200C models)
[23:03:34] <zeeshan|2> 200C!
[23:03:36] <zeeshan|2> nice!
[23:04:11] <furrywolf> heh, I was about to paste a page from the same site. https://tetech.com/faqs/
[23:04:38] <zeeshan|2> there you go
[23:04:43] <zeeshan|2> looking at the datasheet
[23:04:51] <zeeshan|2> voltage has a direct effect o nthe heat removal
[23:05:05] <zeeshan|2> and waste heat
[23:05:29] <zeeshan|2> https://tetech.com/wp-content/uploads/2013/11/VT-199-1.4-0.8.pdf
[23:05:47] <zeeshan|2> and since thiis company is using matlab to plot those
[23:05:48] <furrywolf> all sorts of useless FAQs, including suggested PWM frequencies for temperature controllers. :P
[23:05:50] <zeeshan|2> they might gain me as a customer!
[23:05:59] <furrywolf> (which I think you asked once)
[23:06:50] <zeeshan|2> hm
[23:06:53] <zeeshan|2> they even make a temp controller.
[23:07:18] <zeeshan|2> rs232 communication too
[23:07:19] <zeeshan|2> nice!
[23:08:09] <furrywolf> looks like they can sell you every part you need, in exchange for you supplying as many zeroes as they need.
[23:08:20] <zeeshan|2> well
[23:08:23] <zeeshan|2> the temp controller looks like $189
[23:08:35] <zeeshan|2> can control 0-50V, upto 20A
[23:08:47] <zeeshan|2> my only q is can i connect it to more than 1 peltier
[23:08:50] <zeeshan|2> i don't see why not.
[23:08:51] <furrywolf> what, you're not getting the one with the blue lcd? :P
[23:09:03] <furrywolf> you do remember I said shiny is a very important part of a project, right? :P
[23:09:07] <zeeshan|2> lol
[23:09:18] <zeeshan|2> they havent given me a number
[23:09:29] <zeeshan|2> but i think most of the money will go towards the load cell controller
[23:09:31] <furrywolf> yes, you can connect it to any number of peltiers that draw less than 20A total at the voltage you're using.
[23:10:12] <furrywolf> you might use two stacked for your application
[23:10:20] <zeeshan|2> why stacked
[23:10:58] <furrywolf> larger temperature differential
[23:11:17] <furrywolf> parallel = more heat transfer, but same limits. series = less heat transfer, but larger limits.
[23:11:22] <archivist> I hate the conduction of heat back through an unpowered peltier, have to maintain control, not on off control
[23:11:41] <zeeshan|2> so according to the spec sheet of what i posted
[23:11:47] <zeeshan|2> 80C is the max temp difference
[23:12:30] <furrywolf> you don't want to design for the maximum. it's at zero heat movement. since your part will be leaking, you'll have heat movement, and won't see it.
[23:13:00] <furrywolf> you might get away with one, but the cost of the peltiers is pretty small compared to the cost of everything else needed.
[23:13:45] <furrywolf> also, the performance diagrams in that pdf only show 65C maximum with zero heat moved...
[23:14:57] <furrywolf> running them up to the max just results in a lot of heat you have to get rid of...
[23:15:31] <zeeshan|2> yea
[23:15:32] <zeeshan|2> the cop drops
[23:15:35] <furrywolf> if you figure a relastic differential of 40C, and you want -20C, you'd have to have a hot side of only 20C... not happening. so you need two stacked.
[23:16:00] <zeeshan|2> i wonder if that controller
[23:16:07] <zeeshan|2> can reverse the polarity
[23:16:10] <zeeshan|2> and heat the stage too
[23:16:11] <furrywolf> extra capacity also results is faster cooling, rather than having to wait a half hour while it cools the last couple degrees...
[23:16:20] <furrywolf> probably not
[23:16:26] <furrywolf> s/is/in
[23:17:41] <furrywolf> "The TC-48-20 is an easy to use yet powerful cool-only or heat-only PWM temperature controller."
[23:17:43] <zeeshan|2> yep
[23:17:48] <zeeshan|2> you need a "bipolar" module
[23:17:50] <zeeshan|2> to do what im doing
[23:17:53] <zeeshan|2> so the more expensive one :/
[23:18:10] <furrywolf> The TC-720 is a bipolar, heat and cool, thermoelectric temperature controller with USB communications. only $617! :P
[23:18:23] <zeeshan|2> you really like the 617 version
[23:18:24] <zeeshan|2> lol
[23:18:37] <furrywolf> you said you're building a CNC thing of some form to do the indents?
[23:18:39] <zeeshan|2> he Model TC-36-25 RS232 is a bi-polar (heat and cool) proportional-integral-derivative temperature controller that can modulate power input from 12 V up to 36 V, or from 0 V to 36 V with a second power supply, at currents of up to 25 A.
[23:18:44] <zeeshan|2> yea
[23:18:52] <furrywolf> linuxcnc should have no problem functioning as a temperature controller. :P
[23:19:05] <zeeshan|2> how bad ass would that be? :)
[23:19:12] <zeeshan|2> using linuxcnc's pid to control temp
[23:19:18] <zeeshan|2> and strain and displacement
[23:19:18] <zeeshan|2> haha
[23:19:49] <furrywolf> use any servo driver connected to the peltiers, and feed the thermister into one of the analog inputs on a mesa board... tell it that's the position encoder... tune your PID... :P
[23:19:59] <zeeshan|2> haha
[23:20:11] <zeeshan|2> since it uses a h-bridge
[23:20:13] <zeeshan|2> that'll work :)
[23:20:21] <furrywolf> yep :P
[23:20:29] <zeeshan|2> i'd need an amplifier between the thermocouple input
[23:20:30] <zeeshan|2> though
[23:20:53] <pcw_home> typically you need a inductor for a straight HBridge
[23:20:54] <furrywolf> yes, and you'd want it calibrated. an off-the-shelf thermometer unit that spoke something you could read might be easier.
[23:21:13] <zeeshan|2> we have a couple of omron thermoucples with calibration datasheets
[23:21:25] <zeeshan|2> but i can calibrate a tehrmcouple no problem
[23:21:28] <furrywolf> pcw_home: eh? h-bridges love resistive loads. no inductive currents they have to shunt back to the supply.
[23:21:35] <zeeshan|2> cause we have a master temp thing i can compare with
[23:21:36] <renesis> AD596/AD597 ftmfw
[23:22:04] <furrywolf> in that case, skip hardware calibration, and just have a map in your software somewhere.
[23:22:06] <zeeshan|2> don't put ideas in my head furry
[23:22:13] <zeeshan|2> usiung linuxcnc for doing this
[23:22:13] <zeeshan|2> haha
[23:22:26] <pcw_home> Umm no you need the inductor (the HBridge is functioning as a step down converter)
[23:22:27] <zeeshan|2> it might be easier and quicker to use seperate pids
[23:22:39] <zeeshan|2> and just have a master pythoin program to set the set points
[23:22:48] <zeeshan|2> depending on the test
[23:23:08] <archivist> lovely peltier graph of the td drop with power transfer http://www.heatsink-guide.com/peltier.htm
[23:23:14] <furrywolf> pcw_home: not if you have a device that doesn't mind being driven at full rail-rail voltage.
[23:23:16] <pcw_home> you don want the Peltier voltage to be > than required
[23:23:21] <furrywolf> I'm thinking a small chinese one running off 24V, not a big mains one
[23:23:57] <pcw_home> any more than required and you will just heat both sides
[23:25:09] <pcw_home> (more than Peltier voltage is just lost in Ohmic heating)
[23:26:40] <furrywolf> and, in any case, tossing a random inductor in series with the peltier is not a major project. :)
[23:26:42] <zeeshan|2> ^ too much electrical speak for me
[23:27:34] <furrywolf> zeeshan|2: inductance will smooth the square-wave pwm into something more like a constant current with ripple
[23:27:54] <pcw_home> sleepy time for old folks... 'nite all
[23:28:00] <zeeshan|2> ah okay
[23:28:03] <furrywolf> which results in a happier peltier
[23:28:06] <furrywolf> almost bed time for me too
[23:28:36] <zeeshan|2> same
[23:28:40] <zeeshan|2> it'll be an early morning tomorrow
[23:28:43] <zeeshan|2> industry open house
[23:28:47] <zeeshan|2> we have to show off our research
[23:28:56] <zeeshan|2> more like im gonna vent about WTF at polymers
[23:30:57] <furrywolf> there's nothing wrong with polymers, only your modeling of their behavior.
[23:31:02] <zeeshan|2> fu
[23:31:11] <zeeshan|2> im not modeling anything
[23:31:13] <zeeshan|2> im experimenting!
[23:32:40] <furrywolf> "Experiment results indicate the existing model of polymer behavior may not sufficiently account for additional variables"
[23:33:59] <zeeshan|2> youre right about that
[23:34:12] <zeeshan|2> you use the maxwel model for stress relxation
[23:34:12] <zeeshan|2> etc
[23:34:16] <zeeshan|2> but im not even at that point
[23:36:57] <furrywolf> bbl, wolfy bedtime
[23:37:47] <furrywolf> now you need to get a bunch more peltiers and some plutonium pellets, so it powers and heats itself.
[23:38:15] <furrywolf> polymer tester and RTG! :)
[23:39:05] <zeeshan|2> lol