#qi-hardware IRC log for Saturday, 2014-06-28

DocScrutinizer05heat dissipation is the key limiting factor00:49
whitequarkdos1: flash on SGS2 could actually be turned on for at least half an hour without any permanent damage06:27
whitequarkand maybe longer06:27
whitequarkand it would actually (temporarily) burn in your retina from 1.5m, so it's pretty bright06:27
astrhow many degrees would you say these leds are? 45° ?08:38
whitequarkum08:50
whitequarkit would get really hot08:50
whitequarkmore like 60?08:50
astrwhitequark, sorry not temperature, umm... angle of the light coming out of the led10:16
astris that 60°C or °F10:17
astri guess °C10:18
whitequarkyes10:24
whitequarkno idea about the angle10:24
DocScrutinizer05whitequark: in HH?13:05
whitequarkDocScrutinizer05: hm? what in hh?13:07
DocScrutinizer05astr: beam opening angle is obviously meant to have a uniform brightness of what the camera shoots. So for the usual F of a mobile cam more like 60°13:08
DocScrutinizer05whitequark: wasn't that hardware hacking congress this weekend?13:09
whitequarkDocScrutinizer05: EHSM, yes. if you meant to ask whether I'm in Hamburg, then I'm there for 26-30 June13:09
whitequarkright now at EHSM, in fact.13:09
DocScrutinizer05mhm13:09
DocScrutinizer05greetings from Doc to whomever it may concern! :-)13:10
whitequarkI only really know the people who are also at #qi-hw...13:10
DocScrutinizer05you might be mildly interested in http://neo900.org/stuff/eaglefiles/20140517-Version2-wip/GTA04b7_upper.pdf   http://neo900.org/stuff/eaglefiles/20140517-Version2-wip/GTA04b7_lower.pdf13:11
whitequarkfull neo900 schematics?13:12
DocScrutinizer05not exactly full13:12
DocScrutinizer05protoV213:12
DocScrutinizer05comprehensive peripherals test version13:12
DocScrutinizer05"Accelerometer" to "Zener of backlight booster"13:13
DocScrutinizer05no SoC yet13:13
whitequarkthanks... I'll take a look a bit later13:13
Action: DocScrutinizer05 watches DTM races in TV, took a 20 min to realize the Live TV is from only a 3km distance @.@13:21
Action: DocScrutinizer05 listens out the kitchen window13:21
DocScrutinizer05qualifying only13:24
DocScrutinizer05race is tomorrow: http://www.ardmediathek.de/tv/Sportschau-live/Livestream-DTM-am-Norisring/Das-Erste/Video?documentId=21820958&bcastId=72323013:24
DocScrutinizer05http://www.ardmediathek.de/tv/Sportschau-live/Livestream-das-DTM-Qualifying-am-Noris/Das-Erste/Video?documentId=21820936&bcastId=72323013:25
astrDocScrutinizer05, thanks14:22
astroh and thx whitequark14:23
wpwrakDocScrutinizer05: you should watch .cl vs. br. amazing match. both teams very strong and playing as if the loser would face a firing squad.16:53
DocScrutinizer05watched parts of it16:55
DocScrutinizer05for all I know the loser could end in hell. NFC about that event actually16:55
DocScrutinizer05not even know if they fight for going to next round16:56
DocScrutinizer05anyway time for a bit of shopping before shops close16:57
wpwrakthis is the 1/8 finals. so the loser goes home, the winner moves to the 1/4.17:01
wpwrakdamn. they're going into an extension. how will i do my shopping, if .uy vs. .co is just two hours away from now ?17:55
wpwrak(loser going to hell) well, if the loser is brazil, that's pretty much the outlook. that world cup also has a significant political dimension in brazil. a lot of brazilians are not happy at all with the dubious infrastructure work that had been done.18:00
wpwrak(dubious) e.g., gigantic stadiums that will have no use once the world cup is over.18:01
wpwrakand after the extension still 1:1, now the penalties ... impressive18:40
whitequarkDocScrutinizer05: so remember your laser printer idea?19:08
whitequarkEHSM features just such a machine19:08
whitequarkerr19:08
whitequarkdirect UV exposure with laser printer19:08
DocScrutinizer05sure, i always remember my ideas19:09
DocScrutinizer05and I'm used to find them done by somebody else19:09
whitequarkit doesn't use a lens at all (I think they correct in software)19:09
whitequarkalso a single line of exposure is 80 rotations of the mirror19:09
whitequarkso you get about 150µm of resolution due to slight variations in speed19:10
DocScrutinizer05:-)19:10
DocScrutinizer05how do they create the UV laser? blueray?19:10
whitequarkseems like a typical consumer laser19:11
DocScrutinizer05o.O19:11
whitequarkwell19:11
whitequarkhttp://pcbwriter.github.io/19:11
whitequarkhere's it.19:11
DocScrutinizer05never heard of blue consumer lasers19:11
DocScrutinizer05well, UV consmer lasers19:11
whitequarkoh19:11
whitequarkactually I didn't either19:12
whitequarksadly I missed the talk, I should ask the guy tomorrow19:12
DocScrutinizer05prolly slaughtering a blueray drive (or get one of those LEDs) could do the trick19:12
whitequarkoh. no need. "Thus, the IR laser diode was replaced by a (much stronger) UV laser diode (~1.5W, 445nm) from eBay."19:12
DocScrutinizer05mhm19:13
whitequarkhttp://pcbwriter.github.io/about.html19:13
whitequarkdefinitely thinking about building one, because I'm kinda tired of fiddling with photonegatives19:14
DocScrutinizer05nice :)19:15
DocScrutinizer05you'll love it19:16
whitequarkalso there is a LOT of people interested in reproducible pcb photolitography19:17
whitequarkactually two of them (100% of people I've talked to about this) also want homebrew multilayer19:17
whitequarkand one already has working copper plating method using graphite-based ink19:17
whitequarkI didn't find it earlier because it was in french...19:17
DocScrutinizer05another typ of the Doc: to check out the exposure and all, you can use a sheet of phosphorescent plastic19:17
whitequarkDocScrutinizer05: there's actually no need for that. the photoresist very visibly changes color when exposed19:18
DocScrutinizer05yeah, but it costs real money19:18
DocScrutinizer05while for a quick go/nogo check you rather want to use sth cheaper19:18
whitequarkeh, it's cheap, perhaps even cheaper than phosphorescent plastic + delivery19:19
whitequarkquite sure it IS cheaper19:19
whitequarkfor like ten cm² of board, it's maybe a dozen cents19:19
DocScrutinizer05no it's not since you discard it after use19:19
whitequarkI could use cat piss, it's phosphorescent too19:20
DocScrutinizer05no, it's fluorescent19:20
whitequarkactually cat piss is THE reason consumer UV lamps are sold19:20
DocScrutinizer05like any piss19:20
whitequarkoh19:20
whitequarkright, fluo/phosphorescent, I always mix it up19:20
whitequarkI now see what you mean19:20
whitequarkmmmm, still need to buy it19:20
DocScrutinizer05get a emergency exit plastic sign19:21
DocScrutinizer05backside should be fine for this usecase19:21
whitequarknever seen phosphorescent ones19:21
DocScrutinizer05since you never searched for them19:22
DocScrutinizer05fluorescent plastic comes in all forms and colors19:22
DocScrutinizer05for pretty cheap19:22
DocScrutinizer05err phosphorescent19:22
whitequark;D19:22
whitequarkok I'll look19:22
whitequarkalso these people were quite happy with my extensive documentation, and I'll probably be collaborating on documenting a complete modern homebrew PCB process with one of them later19:23
whitequarkso I guess the work that went into lab.wq.org just paid off19:23
DocScrutinizer05http://www.amazon.de/Art-Manufacture-Design-Nachtleuchtfarbe-250-ml-gelbgrün/dp/B0030IE94M19:25
whitequark20 EUR, I could buy several square meters of resist19:26
whitequarkOH19:26
whitequarkwait, I just realized what you're actually suggesting19:26
whitequarkyou're suggesting this to debug optics19:26
whitequarkI'm dumb.19:26
whitequarkyes, indeed I would want something like that19:27
whitequarkthanks19:27
DocScrutinizer05yw19:27
whitequarkI've seen so many laser cut parts at the ehsm19:28
whitequarkI now want a laser cutter pretty badly19:28
whitequarkyou see, there's a lot of cases laser cut19:29
whitequarkin fact I have seen a case with rounded (d=40mm) corners made out of five pieces of acrylic and no glue at all19:29
DocScrutinizer05of course for exact tests a proper 10MP camera with infinite exposure time (can get simulated in software by adding 0.5s snapshots) and a plain sheet of paper works too19:29
whitequarkone top plate, one bottom plate, a plate with a lot of thin slits that is bent around, and two more plates to latch top+bottom19:30
whitequarkI could never make stuff like that with a mill, the cutters just don't work like that19:30
whitequarkand even very simple cases with tabbed edges won't work, because I have to manually file down every corner and it just sucks19:30
DocScrutinizer05wait till you see the presentation of real professional grade 3D printing, e.g. with a bath of UV curing cyanometacrylate19:32
whitequarkkickstarter is full of SLS 3D printers19:33
DocScrutinizer05or metal sintering 3D printer19:33
whitequarkin a year or two they will be very cheap19:33
whitequarkand in fact a presenter at EHSM was going to build a metal sintering 3D printer using his electron beam welder19:33
whitequarkhe even tried it with zinc oxide, but it just shat all over the vacuum chamber and into the pump19:34
DocScrutinizer05lol19:34
whitequarka great guy from Warsaw. invited me into his lab19:35
whitequarkthat reminds me, I need to book a hotel...19:35
DocScrutinizer05recently a company doing professional metal 3D for aeronautic etc (!!!) wen IPO (stock exchange)19:35
whitequarkheard19:36
whitequarkI think19:36
DocScrutinizer05their printers are a tad expensive19:36
whitequarkthere's a gaping hole where all the DIY metal sintering printers should be19:37
DocScrutinizer05yup19:37
whitequarkI just wish vacuum equipment was cheaper19:37
DocScrutinizer05I wonder if sputtering could get done19:37
whitequarkyou need something heavy to sputter with19:38
whitequarkand that's probably really expensive19:38
whitequarkI mean, sputtering is apparently done using gold ions or smth like this19:38
whitequarkargon, too, but it's not as efficient19:38
DocScrutinizer05or (attention, *weird*) vaporizing19:38
whitequarkand really expensive19:38
DocScrutinizer05maybe the metal atoms could even get transported exactly into place by an electron beam?19:39
whitequarkso you'd make the workpiece the cathode?19:40
DocScrutinizer05prolly only feasible for *really* tiny objects19:40
DocScrutinizer05yup19:40
whitequarkI think the problem is that if you want it fast, you'd need high beam current19:40
whitequarkand that melts / destroys workpiece19:40
DocScrutinizer05yup19:40
whitequarkmaybe if you machine, idk, wolfram19:40
whitequarktungsten?19:40
DocScrutinizer05well, the trick is to make the workpiece get hot enough to have nice integration of the metal atoms into a crystal, but not melt the whole thing down19:41
whitequarkuh, I don't think you can grow a crystal like that19:41
whitequarkit would be incredibly stressed19:41
DocScrutinizer05metal is a crytal maze, no?19:41
whitequarkyou'd have a very, very, very imperfect crystal19:42
DocScrutinizer05unless it's molten19:42
DocScrutinizer05yes, sure19:42
whitequarkyou could anneal it afterwards, I guess19:42
whitequarkdamn. so many cool ideas to work on19:42
DocScrutinizer05I'm more worried about effects like xray secondary radiation when operating with such types of electron beams ;-)19:44
whitequarkthat's at 10kv and higher19:44
DocScrutinizer05even 3019:44
DocScrutinizer05how do you create an electron beam of several Ampere strength?19:45
whitequarka big power supply?19:45
whitequarkI mean, there's not exactly a limit. you can pump hundreds of amperes into a xenon continuous discharge lamp19:45
DocScrutinizer05aaah, right, it might have a impedance like a gas discharge lamp19:45
whitequarksame principle here19:45
whitequarkbut you need very precise pressure control19:46
whitequarktoo high, and it sparks. too low, and beam dies19:46
DocScrutinizer05I was thinking absolute vacuum19:46
whitequarkno beams in absolute vacuum19:46
whitequarkand no such thing too19:46
DocScrutinizer05uhß19:46
DocScrutinizer05?19:46
whitequarkwell19:47
DocScrutinizer05what's that Braunsche Roehre then, CRT?19:47
whitequarkany vacuum you can achieve on earth is not even close to simply outer space19:47
whitequarkhm19:47
whitequarkbrb, laptop discharged19:47
whitequarkback19:54
whitequarkDocScrutinizer05: http://www.sparkbangbuzz.com/crt/crt6.htm19:59
whitequarkthis guy explicitly says that if the pressure is too low, the CRT won't work19:59
DocScrutinizer05sorry, for all I ever heard that's mere bs20:03
whitequarkhm20:03
whitequarkmaybe20:03
whitequarkoh, I realized the mistake20:04
whitequarkyou would only need some ions if you want cold cathode20:04
whitequarkfor hot cathode you can indeed have any kind of vacuum20:04
DocScrutinizer05for CRT in former times you even had ion traps to keep any residual gas atom in the CRT from damaging the fluorescent paint20:05
DocScrutinizer05nowadays vacuum is good enough so they don't need it anymore20:05
DocScrutinizer05unlike the electrons the ions don't get diverted that much by the magnetic field, due to higher mass, and thus created a dark spot in center of CRT screen20:06
whitequarkwhy would ions be accelerated towards the screen?20:07
DocScrutinizer05err, dunny, maybe because they are also electrically charged? Or because the electrons simply "kick" then?20:11
DocScrutinizer05see electromigration20:11
DocScrutinizer05dunno*20:12
whitequarkI mean, the ions should be accelerated into the opposite direction20:13
whitequarkelectrons kicking them... hmmm, maybe that20:13
DocScrutinizer05https://de.wikipedia.org/wiki/Kathodenstrahlröhre#Strahlerzeugung  >>Im nebenstehenden Bild ist das Strahlsystem einer veralteten Fernsehbildröhre mit Ionenfalle zu sehen. <<20:16
DocScrutinizer05https://de.wikipedia.org/wiki/Ionenfleck20:17
whitequarkI see20:19
DocScrutinizer05I was wrong regarding why it's not needed anymore though. The fix been a 0.2µm alu coating on the phosphor20:20
DocScrutinizer05electron tubes have a barium pill that gets ignited after production, to absorb all remaining gas atoms/molecules20:22
DocScrutinizer05https://de.wikipedia.org/wiki/Getter20:23
whitequarkyeah, know about that20:28
wpwrakand now, for some conspiracy theory that would surprise even The Onion: http://worldtruth.tv/rothschild-inherits-a-semiconductor-patent-for-freescale-semiconductors/22:18
wpwraknote how they beautifully mix three at best vaguely related products / processes ;-)22:19
--- Sun Jun 29 201400:00

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