#qi-hardware IRC log for Tuesday, 2015-01-20

Luke-Jranyone here work with the TP-Link routers?00:13
Luke-Jrxiangfu around maybe? :00:13
whitequarkLuke-Jr: me00:22
Luke-Jrwhitequark: I'm having trouble soldering to the GPIOs.. they're so tiny I can barely see them, and the solder won't stick :/00:22
whitequarkoh, hardware wise00:23
Luke-Jrany tricks?00:23
whitequarkcan you post a photo00:23
Luke-Jrhttp://dangerousprototypes.com/forum/download/file.php?id=8244&sid=c40c3a027a423661e495159435318f14&mode=view is what I am trying to do00:23
Luke-Jrdoes a photo of mine specifically matter? it looks the same minus the wires00:23
Luke-Jroh, and the resistors are still there00:24
Luke-Jrnot sure if my use case needs them removed yet00:24
whitequarkno i just want to see the general layout00:24
Luke-Jr(nor how to remove them if it does)00:24
whitequarkoh i see00:24
whitequarkdid you tin the wire?00:25
Luke-Jryes, I think so00:25
Luke-Jrmelted solder and spread it over the wire00:25
whitequarkok00:25
Luke-Jr(at this point I'm questioning everything I thought I knew about soldering)00:25
whitequarkso the issue is probably the router using pb-free solder00:25
whitequarkput a *very small* blob of your pb/sn solder on the resistor terminal and make sure it melts well and mixes with whatever was on board00:26
whitequarkthe iron should be set to 300, maybe even 330°C if you work with pb-free00:26
Luke-Jrit just sticks to my soldering iron00:27
whitequarkthen it is not hot enough00:28
whitequarkthe tip isn't, at least00:29
wpwraktrivial question first: are you using flux ? :)00:29
whitequarkoh ^00:29
Luke-JrI don't know what flux is?00:29
whitequarkwpwrak: you win okay00:29
wpwraksometimes it's just too easy ;-)00:29
Luke-Jrthe solder is labelled "Standard Rosin-Core Solder"00:30
wpwrakLuke-Jr: flux is what makes the difference between such a solder job being your worst nightmare and it actually working without a fuss :)00:30
wpwrakfor SMT and such, you need to add extra flux. what little is in your solder is not enough.00:31
Luke-Jrdoes this count as SMT? :o00:31
whitequarkrosin-core solder would probably work, if you put the tip at the resistor terminal and then touch the heated terminal with the solder00:32
wpwrakflux can come in many forms. more common ones are syringes (gel), felt pens (liquid), etc.00:32
whitequarkyes, of course this counts as SMT, it's an SMD resistor00:32
whitequarkwpwrak: rosin-core solder pretty much always worked for me for SMT jobs00:34
Luke-Jrno terminal visible :/00:34
whitequarkLuke-Jr: huh? the terminals are the metal caps on the ends of the resistor00:34
Action: Luke-Jr wonders if it's normal to have a hard time keeping hand steady00:34
wpwrakLuke-Jr: the felt pens are the most convenient. if you find a pen that dispenses flux through a pipe (a bit like a syringe), they're also nice to use. syringes tend to be a bit messy because it's hard to get the right quantity and you may need to spread it with a toothpick or such.00:34
wpwrakLuke-Jr: only when sober ;-)00:35
Luke-Jrwhitequark: yes, but it's so small I can't really see it - just a reflection of a tiny speck of light00:35
Luke-Jreh, soberness is bad for this? XD00:35
whitequarkLuke-Jr: get a loupe then00:35
wpwrakwhitequark: (rosin-core) i wouldn't even know when i tried the last time without extra flux :) 00:36
whitequarkwpwrak: i usually don't bother. i always use flux for reworking, but not really when initially soldering00:37
DocScrutinizer05Luke-Jr: hey! long time no see01:06
Luke-JrDocScrutinizer05: hi :03:08
Luke-Jrtrying to pick up some hw basics :p03:08
DocScrutinizer05Luke-Jr: yeah, a beer help a lot with soldering SMD, and you definitely want to use flux. The difference is like drilling a hole in a brick wall with your finger vs drilling that hole with http://www.professional-power-tool-guide.com/wp-content/uploads/2008/08/hilti-demolition-tools-300x209.jpg03:20
DocScrutinizer05literally03:21
whitequarkDocScrutinizer05: wait, do you literally mean that being sober makes it harder?03:21
DocScrutinizer05indeed03:22
DocScrutinizer05one beer helps a relax03:22
DocScrutinizer05to*03:22
whitequarkhuh03:23
DocScrutinizer05at least that's my mileage03:23
DocScrutinizer05a lil bit of muscle relaxant, plus more patient03:24
DocScrutinizer05of course the size/number of beer(s) needs adjustment to the individual disposition03:25
DocScrutinizer05and some stoic/phlegmatic characters might not see any use in beer at all03:26
DocScrutinizer05admittedly you don't need or benefit from beer when you do that stuff every day03:28
DocScrutinizer05and maybe that's actually a specifically bavarian method03:31
Luke-Jryay it works http://imgur.com/lh0EP9a03:36
Luke-JrDocScrutinizer05: hm, I've never drunk alcohol before. I guess this is as good a reason as any03:37
Luke-Jrunless there's something else that works03:37
xiangfuwpwrak: we receive all the components except this KL26 and the PCB.05:36
xiangfueach component I bought 50 of them.05:37
wpwrakwow ;-)05:39
wpwrakis the KL26 a problem ? or does it just take a little longer ?05:40
qi-bot[commit] Werner Almesberger: sfc/slicer.py: when flipping, also flip Z positions used for selecting levels (master) http://qi-hw.com/p/cae-tools/4cc43e106:21
xiangfuwpwrak: it is just slow. I will get it at 27 Jan.06:54
wpwrakah yes, that's pretty slow for this kind of chip. well, nothing we can do about it.06:57
whitequarkniiiice, bundestag has video recordings of meetings available online12:50
DocScrutinizer05german bundestag?14:24
whitequarkyes14:24
DocScrutinizer05wow14:24
whitequarkhttp://www.bundestag.de/mediathek/?isLinkCallPlenar=1&action=search&ids=4439410&instance=m187&categorie=Plenarsitzung&destination=search&mask=search14:24
DocScrutinizer0580. Sitzung vom 16.01.2015  typical pathetic number of joining dudes14:35
whitequarkyou seem to have a particularly unflattering view of germany14:36
DocScrutinizer05ssort of, mostly. yes14:41
DocScrutinizer05GroKo, coalition between left and right mainstream, opposition is negligible. Always a poor government, not giving a f* about what the people thinks and wants14:44
whitequarkisn't the approval ratings in ~80s?14:46
DocScrutinizer05approval ratings?14:46
whitequarkhttp://uk.reuters.com/article/2014/08/07/uk-germany-poll-satisfaction-idUKKBN0G71T32014080714:47
DocScrutinizer05aaah satisfaction, well the Germans are satisfied with politics that don't annoy. Which means the less news the better14:48
DocScrutinizer05Merkel's secret sauce14:48
DocScrutinizer05wpwrak: air asia - same maneuver as that other airbus: steep climb, then sudden drop (prolly after stall). I bet on similar combination of idiocy in autopilot / stall warning programming and instrumentation misguiding20:58
DocScrutinizer05also funny: google goes global sat internet21:04
DocScrutinizer05bright future for mobile phones everywhere21:05
DocScrutinizer05though I doubt google will allow VoIP (whatever flavor) except their very own service, and most likely not for free either21:06
DocScrutinizer05I just wonder how to make some bucks from that emerging market for 3 billion dirt cheap embedded mobile sat transceivers21:08
wpwrakDocScrutinizer05: (QZ-8501) siehe http://avherald.com/h?article=47f6abc7&opt=021:15
DocScrutinizer05:-)21:15
wpwrak"the aircraft was climbing at a vertical rate of up to 6000 fpm exceeding the aircraft's capabilities"21:15
wpwrakthat sounds more like a rather violent wind system than an AF447-type incident21:15
DocScrutinizer05well, when the stall warning stops yelling when aircraft speed < 40 knots, what do you expect the pilots to do? Particularly when airspeed iinstrumentation flawed by e.g. ice21:16
wpwrakalso, the pilot was extremely experienced. former fighter pilot, etc. the sort of AF447 mess seems extremely unlikely there.21:17
DocScrutinizer05hmm, sound reasoning21:17
DocScrutinizer05but why the crahs after that extreme climb then?21:17
DocScrutinizer05AF447 the pilot wasn't even in cockpit afaik, he joined later (pissing? coffee? sex?) and failed to grok the situation21:19
wpwrakthe plane may have ended up at a place where it couldn't fly on its own. or maybe it suffered structural damage as a consequence of that ride.21:19
DocScrutinizer05structural damage from winds? not heard of during last ~50 years, right?21:20
wpwrakwell, the PIC was out of the cockpit. but also note that AF447 was a long flight. QZ-8501 was comparably short. so they didn't need to rotate the cockpit crew.21:20
DocScrutinizer05(except during landing)21:20
DocScrutinizer05hope they soon will decode the voice recorder21:22
wpwrak(structural damage in flight) doesn't happen often, but it does happen. also, there have been numerous incidents in the preceding weeks just in the general area with severe turbulences. so one can assume that the conditions were worse than normal for the season21:25
DocScrutinizer05however >>the aircraft was climbing at a vertical rate of up to 6000 fpm exceeding the aircraft's capabilities, its airspeed decreased and it appears the aircraft entered a stall<< sounds exactly like AF44721:25
DocScrutinizer05extreme conditions been what canceled out the autopilot on AF44721:26
DocScrutinizer05afaik21:27
wpwrak(conditions) here's a nice structural failure example. happened at low altitude, but still (look at the pictures): http://aviation-safety.net/database/record.php?id=19930331-021:28
wpwrakAF447 was fully flyable at the time it stalled. QZ-8501 may not have had that luck.21:29
DocScrutinizer05W*T*F?!21:29
wpwrakand yes, pilot overwhelmed when the AP disconnects is a surprisingly common cause of death21:30
wpwrakbut the again, this was a pilot with real flying experience. not just "take off, turn on AP, take a nap, autoland, go home"21:30
DocScrutinizer05hehe21:31
wpwrakof course, even experienced people can make mistakes. but in this case, it doesn't look like a basic aviation failure like in AF44721:32
DocScrutinizer05as scary as all AF447 series of events was, my first reaction been "I prolly had shut down all instrumentation and flew the bird from my ass sensors only"21:33
DocScrutinizer05(virtually shut off, read 'ignored')21:33
DocScrutinizer05I mean, you literally _hear_ a difference between low speed stall and high speed stall21:34
wpwrak(turn off instruments) that's surprisingly difficult ;-) http://avherald.com/h?article=47d74074&opt=021:35
DocScrutinizer05I don't believe in a 150dBm stall warning + extremely high speed display when my ears tell me the thing almost standing still21:36
DocScrutinizer05(difficult) thus virtually21:36
wpwraknot sure how reliable your senses are. also considering weather and darkness. i.e., you don;t really know what you're flying into.21:36
wpwrakwell, in the case of AF447. QZ8501 on the other hand had daylight ...21:37
wpwrakoh, and if you want fun, look at the video here (best enjoyed on youtube and in full-screen): http://avherald.com/h?comment=47f968b7&opt=021:39
DocScrutinizer05yeah, I know my senses, so that's prolly why I say so. A frequent pattern when I run into weird instrumentation readouts of any kind. I'd prolly trigger the sparks of my car engine manually when I suspect the engine control is doing nuts21:39
wpwrakthe #1 rule in aviation if you can't see where you're going is to only trust your instruments, not your senses21:40
DocScrutinizer05hmm, damaged landing gear? Doesn't look unusual or spectacular21:40
wpwrakguess there were many dead pilots who trusted their senses as much as you trust yours ;-)21:40
wpwrakit's not the blown tires, it's the taxiing ...21:41
DocScrutinizer05ooh, ignoring all rules?21:41
wpwrakthese dead pilots may be the reason why these rules are there ...21:42
DocScrutinizer05"damg thing doesn't woll, give a tad more throttle!"21:42
wpwrakyeah :)21:42
DocScrutinizer05omg21:42
DocScrutinizer05"dang thing doesn't roll, give a tad more throttle!"21:42
DocScrutinizer05I doubt the pilot even noticed21:42
wpwrak"let's just taxi with take-off thrust" :)21:43
wpwrakjust read the comments ;-)21:43
wpwrakor look at this example of how little of an issue pilots do notice: http://avherald.com/h?article=47f95b98&opt=021:44
DocScrutinizer05(trust senses) look, the difference is: I only overrule the instruments when they act extreme (stall warning etc). When my senses act extreme while instrumentation is pretty normal, I'd never even poncer which to trust more21:45
DocScrutinizer05ponder*21:45
wpwrakhehe :)21:45
wpwrakbut yes, LH-1829 was a very nice example of what to do when the plane is acting weird. don't bother debugging the computer, just fly.21:46
DocScrutinizer05I'm not going to fly head down thinking all horizons went mad21:47
DocScrutinizer05exactly21:47
DocScrutinizer05(not knowing LH-1829 case, just ack'ing the general approach)21:48
DocScrutinizer05http://avherald.com/h?article=47d74074&opt=0 ta "search"21:48
wpwrakLH-1829 was the 4000 ft dive, link above21:49
wpwrakyup :)21:49
DocScrutinizer05FL270 fett * 10?21:50
DocScrutinizer05feet*21:50
DocScrutinizer05or * 100?21:50
wpwrak*10021:50
wpwraki.e., cruise altitude21:51
wpwrak... where nothing normally happens :)21:51
wpwrakhere's another nice accident. very unfortunately without pictures. http://avherald.com/h?article=46fadfa1&opt=021:52
wpwrakNB: in the comments someone explains that this aircraft has a wingspan of 34 m21:54
DocScrutinizer05>> actually airbus should Have a simple computer protection override switch<< hehe!21:55
wpwraki guess it's a two-edged sword: airbus are pretty aggressive with automation, and i think this does make the planes safer. on the other hand, it also means that pilots with insufficient training can fly them for years without any sign of trouble.21:59
DocScrutinizer05also the meaning of "insufficient training" suddenly changes when the computer can outsmart the pilot22:00
DocScrutinizer05when APA(?) can force node down and pilot can't overrule that -  e... :-((22:01
DocScrutinizer05nose*22:01
DocScrutinizer05I shouldn't type today22:01
wpwrakit's the ADRU that implements alpha protection (alpha being the angle of attack. i.e., the thing that was waaaay off in AF447)22:02
DocScrutinizer05AOA22:03
DocScrutinizer05yep22:03
DocScrutinizer05no idea about all that, it's my friend and landlord who got a license, not me ;-)22:04
wpwrakAF447: by the way, you may find this interesting: http://www.bea.aero/fr/enquetes/vol.af.447/videos/animation.deroulement.du.vol.mp422:04
wpwrakmentioned here: http://avherald.com/h?article=41a81ef1/0096&opt=022:04
DocScrutinizer05I think that's what he told me about - weird messy actions on horn (sidestick), right?22:05
wpwraki kinda wonder if the computers in the airbus also do a longer-term assessment of the effect of their actions. e.g., would LH-1829 eventually have stopped the descent on its own ?22:05
wpwrakyup. and look at the angle of attack.22:06
DocScrutinizer05sorry, mimetype odd, doesn't play in any video player on my machine22:06
wpwrakthat AF447 simulation reminded me a bit of how i "fly" in a flight simulator when i try it for the first time and haven't read the instructions22:07
DocScrutinizer05(would have stopped) aiui nope22:07
wpwrakdownload, then view ?22:07
DocScrutinizer05the suppositious angle of attack?22:09
DocScrutinizer05(download) meh!22:09
DocScrutinizer05"open with" vlc22:10
DocScrutinizer05bwoahh scary22:24
DocScrutinizer05what's the A321 going to "do" when lft and right sidestick are diametrically opposite steering position?22:25
DocScrutinizer05seems the pilots trained on a spaceshuttle ;-)22:27
DocScrutinizer05though, only regarding angle of attack and sink rate/mode, not regarding taking into account the fact that such a critter reacts on steering like a wet sponge22:28
DocScrutinizer05my friend told me there's just two things to care about, after making sure fuel=100% and liftoff: altitude and speed. lots of both22:32
DocScrutinizer05I really don't understand what the pilots tried to do. Why didn't they push throttle to max and nose down?22:33
DocScrutinizer05I mean, even in their coffe cups they must have seen that the plane is nose-up or at least not too much nose-down22:34
DocScrutinizer05and the airspeed meter was between 60 knots and zero most of the time, right?22:35
DocScrutinizer05did they just check altimeter and conclude they need to pull up to stop decline?22:36
DocScrutinizer05don't pilots train to get a plane out of stall?22:37
DocScrutinizer05they must have been in that situation in simulator before, no?22:38
wpwrak(two things to care about) that's the aviation part. you may also want to pay attention to navigation from time to time. especially if you live in a mountainous area :)22:46
wpwrakthere seems to be a specific "unreliable airspeed" procedure that would have saved them. why they didn't will remain their secret forever ...22:47
wpwrak(opposed sidestick) that's when you get the "Dual input" warning22:49
DocScrutinizer05(spaceshuttle) even for that the "landing" procedure would be incorrect - way too low airspeed22:56
DocScrutinizer05(dual inp) I guess captain's sidestick rules then22:57
DocScrutinizer05hmm, maybe not22:57
DocScrutinizer05I bet there's some other weird rule that aims to cope with erratic input from epileptic attacks, defect sidestick, et el22:58
--- Wed Jan 21 201500:00

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