#homecmos IRC log for Monday, 2011-08-01

azonenbergGot busy and didnt have time to do any work today... hoping to do a test of slow-etching copper (for better repeatability) tomorrow01:38
bart416So alive on this beautifull afternoon azonenberg ?11:43
azonenberg_workLol, yep11:48
azonenberg_workBut it's 0800 for me11:48
azonenberg_workjust got in to the office :p11:48
bart416I'm halfway through my ARM Sitara technical documents!11:49
azonenberg_workYay lol11:49
bart416Also, I was thinking11:50
bart416Can you evaporate copper on paper?11:51
azonenberg_workI dont think paper is UHV compatible12:11
azonenberg_workIt'd probably outgas12:11
azonenberg_workalso, the metal vapor is quite hot12:12
azonenberg_workthe surface has to be able to handle what's quite literally boiling metal condensing on it12:12
bart416yeah but it's not much12:12
azonenberg_workIt won't burn, due to lack of air, but might char12:12
azonenberg_workand in the process contaminate your chamber12:12
bart416that's what I'm worrying about12:12
bart416If it would only char it'd be perfect12:12
azonenberg_workAll of the pyrolysis products would make a huge mess though12:13
bart416yeah, again that's my worry12:14
bart416But it solves the hard to get rid off polymer12:14
bart416Though using glass as substrate would also be nice12:14
bart416Going to see if I can use the materials lab at college12:16
bart416If I do I could also use the xray equipment to test the result12:16
azonenberg_workWhat exacly is your goal?13:14
bart416to make a multi layer pcb13:15
azonenberg_workon paper?13:15
bart416Was just considering paper for the simple reason that it's easy to chemically remove13:16
azonenberg_workWhat about your ILD?13:16
azonenberg_workYou could use blank FR4 as the substrate if its vacuum compatible (dont think it is though)13:16
bart416What the hell do you mean by ILD13:17
bart416I've been trying to figure that out for days13:17
bart416We're not big on acronyms in non-English speaking countries13:17
azonenberg_workinter-layer dielectric13:18
bart416I looked up the properties further13:18
bart416It's perfect for this13:18
bart416Can do thick layers13:18
bart416good mechanical and chemical properties13:19
azonenberg_workSo you want to do su-8 on paper?13:19
azonenberg_workWhy not silicon13:19
azonenberg_workget a cheap solar wafer, $50 per 25 pack on ebay13:19
azonenberg_workput down one layer of su-8 over that13:19
azonenberg_workAnd do you have what it takes to develop/coat it?13:19
azonenberg_workAnd do you have a source for it?13:19
bart416I just need a substrate that I can dispose of after I'm done13:19
azonenberg_workI've never heard of an amateur attempting it13:19
bart416SU-8 can be done by standard contact lithography equipment13:20
azonenberg_workWhat i mean is, can you get hold of it and the developer?13:20
bart416Through college, sure :|13:21
bart416I've placed weirder orders for chemicals than that :S13:21
bart416Would be amazed if I couldn't get it13:22
azonenberg_workI see13:22
bart416Ever ordered a large canister of liquid oxygen and a bag of KNO3 on the same day?13:22
azonenberg_workUm... nope13:22
bart416I did :P13:22
azonenberg_workMy school is a little skittish about letting students get access to stuff13:22
azonenberg_workmost i was ever offered was some NaOH crystals13:22
bart416As long as the cost isn't too high they don't care13:23
bart416If it's cheap they won't even charge you13:23
azonenberg_workyou can do an lot in a lab but they dont like you sneaking it out13:23
bart416Like if you need a few cheap transistors or some cheap chemicals (like developer for more common photoresist for pcb etching) they'll just give it to you the next day most of the time13:23
bart416And if you ask for the cost "nevermind"13:23
azonenberg_worklol nice13:24
bart416Though larger things like a scope or power supply are harder to get through the paperstack (you get college discount eh :P )13:27
bart416But it happens13:27
azonenberg_worklol nice13:27
bart416Most expensive thing somebody got through was an 8000 euro R&S scope13:27
bart416somebody tried an even more expensive scope once but got busted by taxes :(13:28
azonenberg_workWhat school is this? :P13:28
azonenberg_workAnd the scope i plan to buy later this week is http://www.rigolna.com/products/digital-oscilloscopes/ds1000d/ds1102d/13:29
azonenberg_workaround 900 USD after educational discount13:29
bart416I'm looking at picking up the 50 MHz rigol13:29
bart416and unlocking it13:29
azonenberg_workthe MSO or just the analog capture?13:30
azonenberg_workthis one has 16 digital inputs13:30
bart416I just think it's not worth it13:30
bart416If I were to get a mixed signal I'd pay 200 USD more and get the agilent13:30
azonenberg_worki see13:30
bart416Just a matter of reliability13:31
bart416But actually going to wait until rigol responds to my inquiry13:32
bart416and until after the flea market :P13:32
azonenberg_worki see13:32
bart416It's just if you can get an agilent for $200 more, on $1000 it's not going to matter13:33
bart416especially with the current exchange rates13:33
bart416+ don't you do anything except messing with electronics?13:34
Action: bart416 pokes azonenberg and wonders where he gets the $1000 USD budget13:40
azonenberg_workNope, i dont really :P14:18
azonenberg_workWell, that and coding14:18
azonenberg_workbut that doesnt need hardware14:18
azonenberg_workAnd re budget, i finally got paid for the summer :P14:18
azonenberg_workand i need an oscilloscope so i figured, why not14:19
bart416Then again, I probably earned more than you during the summer, just sucks that I have to blow it all on things cause my parents turned greedy >_>14:19
bart416Also SU-8 has adhession problems to glass14:19
azonenberg_worklol i see14:19
azonenberg_workCan you use a thin adhesion layer of something else?14:20
azonenberg_workI mean, HMDS might work for starters14:20
bart416Yeah they suggest it in the brochure14:20
bart416Also 1mm of SU-8 will survive 40 kV :P14:21
bart416I have a feeling it'll do for PCB design :')14:22
bart416Density is a bit higher than water14:23
bart416And doesn't start to degrade until about 350°C14:23
azonenberg_workDid i add you to the googlecode page yet btw?14:24
azonenberg_work*user list14:25
azonenberg_workif not, PM me your google code email and i'll set you up with access for wiki etc so you can document your work when you actually start running experiments14:25
bart416I'll first see if it's financially viable14:28
bart416Nobody publishes the price of SU-814:28
bart416So I'm starting to have fears14:28
bart416They suggest AP30014:38
bart416Have you tried that azonenberg ?14:38
bart416The main problem would actually be the top and bottom surface14:44
bart416Though I guess using normal resist for this wouldn't be much of a problem14:45
bart416cross linked SU-8 is fairly resistant to acetone anyway14:45
bart416also azonenberg_work http://www.homalite.com/cr39.htm14:45
azonenberg_workCross linked SU-8 is pretty much indestructible from what i hear :P14:50
azonenberg_workIt practically takes piranha to remove14:50
bart416they report etch rates of 1µm/minute14:52
azonenberg_work1um or nm?14:53
azonenberg_workThats pretty high actually14:53
bart416Using chemicals specifically manufactured to etch SU814:53
bart416But better rates are achieved using O2 + F radicals plasma14:54
bart416azonenberg, you know the price of SU-8?15:00
bart416$600 / liter o_O15:00
bart416This would make for very expensive PCBs15:07
bart416Not worth considering15:07
bart416Need something cheaper15:13
azonenberg_workYeah, i figured it would be something like that15:21
azonenberg_workAnd do you plan to use O2 + F plasma in your living room? :P15:21
bart416lol, no15:27
bart416But I don't need to etch crosslinked SU-8 anyway15:27
bart416Even if I were to use it15:27
bart416What I won't cause $600 / liter would give a bad turn over rate15:29
bart416mhhh, other idea15:30
bart416coat plastic with Cu15:30
bart416Should be a lot cheaper15:33
bart416+ I can etch both acrylics and copper easily15:34
bart416And without damaging the other15:34
bart416And pressing them together is easy as well15:34
bart416oh btw, if anybody needs anything from springer, sciencedirect or web of knowledge15:36
bart416now is the time15:36
bart416damn the new search system they employ sucks o_O15:43
azonenberg_workplastic coated with Cu was my suggestion too16:24
azonenberg_worksu-8 is not cheap :P16:24
bart416what would you suggest16:26
bart416electroless plating?16:26
bart416Also, you suggested glass :P16:27
bart416Not acrylics16:27
azonenberg_workI had suggested FR4 originally17:32
azonenberg_workre plating, not sure17:32
azonenberg_workevaporation probably wont work for it17:33
bart416FR4 phenolic seems like something that'd be horrible to etch though17:41
bart416That's why I'd use normal acrylic17:45
bart416etch holes17:46
bart416apply photoresist17:46
bart416remove unexposed17:46
bart416add electroless plating solution17:47
bart416apply palladium ions to make plastic conductive17:47
bart416remove resist17:48
bart416Meh, I need to go and talk to a chem professor anyway sooner or later17:49
bart416Or I could ask my ex-girlfriend :|17:59
bart416But I so do not want to do that17:59
azonenberg_workYou dont etch FR418:03
azonenberg_workYou use the FR4 as the *substrate*18:03
bart416Yeah but then I have only a single layer18:03
azonenberg_workThen use something else (details TBD) as the ILD18:03
bart416that I can mount actual devices on18:03
bart416Removing FR4 would be horribly complicated18:03
azonenberg_worklike i said, you dont remove it18:04
azonenberg_workyou'd drill those holes by hand18:04
azonenberg_workor CNC even18:04
azonenberg_workthey'd be huge, the equivalent of a TSV18:04
bart416Sub millimeter holes are hard, even for CNC machines...18:04
bart416especially if you're looking at 500+ of them18:04
azonenberg_workyou dont do 500+18:04
bart416Actually considering my current design for this damned arm that's about what I need18:05
azonenberg_workwhat are you making?18:05
bart416Easy enough for a professional plant18:05
azonenberg_workoh, did i mention i soldered a 5-ball CSP BGA last night?18:05
bart416Ever tried a 500 pin BGA on a double layer pcb?18:05
azonenberg_work520 x 570 micron grid with one extra ball in the middle, all 200 micron diameter18:05
azonenberg_workand nope18:05
azonenberg_workthis was my first attempt at a bga18:06
azonenberg_workIt appears to have been successful so i'll probably try larger in the future18:06
azonenberg_worki have some 11x11 0.8mm XBGA18:06
azonenberg_workwould be cool to get them to work18:06
bart416I'm starting to become as insane as one of my analog circuit design professors18:08
bart416I like things that don't work and figuring out how to get them to work18:08
azonenberg_workAnd i like building stuff18:16
azonenberg_worklike, all-out building stuff :P18:16
azonenberg_workfrom the kernel to the CPU RTL to the wafer fab18:16
bart416I'd love to do that but we don't have the facilities goes too much money goes to stupid business and humanities18:59
bart416yes business and humanities get all the scholarships for studying abroad and research grants >_>19:00
bart416Electronics gets less for research than business gets for rewards to give to people for their surveys...19:00
azonenberg_workAnd RPI just built a new athletics facility and performing arts center19:02
azonenberg_workthe old ones were fine19:02
bart416Our college decided to make a new artificial grass soccer field19:03
bart416You know, the type you find in professional sports stadiums19:03
bart416Then again, two years ago they did spend money on something useful19:03
bart416A new library19:03
bart416Believe me, it was necessary19:04
bart416But yeah, electronics department had to beg to buy a new network analyzer19:04
bart416well, it sort of costs the same as a new small building but yah :P19:05
bart416But yeah, 50 GHz network analyzers don't come cheap :P19:08
bart416So they bought it together with the university19:09
bart416Though technically starting from within a few weeks we're part of the university and not the college anymore19:09
bart416They moved engineering to university cause they didn't think it was right that a college does masters degrees >_>19:09
berndjbart416, dj delorie makes his own multilayer boards, afaik he uses some sort of hacked laminator to get the layers together20:49
bart416any website?20:49
berndjwww.delorie.com, but i don't think i've seen much doc on his process - rather pics of the results20:50
berndjare you on the homebrew_pcbs yahoo group? seems like lots of clever people hang out there20:51
--- Tue Aug 2 201100:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!