#qi-hardware IRC log for Monday, 2012-06-11

qi-botThe build has FAILED: http://fidelio.qi-hardware.com/~xiangfu/building/Nanonote/Ben/openwrt-xburst.full_system-20120610-0412 00:04
qi-bot[commit] Adam Wang: scj368r1.fpd: added 3.5mm audio stereo phone jack (master) http://qi-hw.com/p/kicad-libs/12ffd3002:29
qi-bot[commit] Adam Wang: Merge branch 'master' of projects.qi-hardware.com:kicad-libs (master) http://qi-hw.com/p/kicad-libs/354d60f02:29
qi-bot[commit] Adam Wang: renamed scj368r1.fpd -> phonejack.fpd:  phonejack-${size}-${N} (master) http://qi-hw.com/p/kicad-libs/dc51a1b04:01
wpwrakcladamw: did you see ? i make a script that generates connector symbols, for CONN_1 to CONN_40 and CONN_1X2 to CONN_40X210:02
wpwrakthe script is components/gencon.pl and the library it generates is gencon.lib10:03
cladamwwhat's the syntax ?10:05
wpwrakit's Perl :)10:05
wpwrakunfortuanately, we don't have something like fped for schematics symbols ...10:06
cladamw*.pl is a language of Perl, right ?10:06
wpwrakyes10:06
cladamwso how to generate ?10:06
cladamwi've never used Perl before. :-)10:07
wpwraki also made a variant of pads.fpd, called pads-array.fpd, which puts this sort of contacts in an array. for now, i use it with the "2mm" pattern at 50 mil pitch, for soldering ribbon cables. (like you can to with UBB)10:07
wpwrak(perl) just  "make gencon.lib"10:07
wpwrakor you could run things manually, ./gencon.pl >gencon.lib10:08
wpwrakmaybe i should just have "make" build any generated files (in components/). right now, it's equivalent to "make sch", but typing "make sch" isn't all that much more work ...10:10
cladamw(pads.fpd) yeah ... these 'round' pad are good variants10:11
cladamw(phonejack.fpd) f.g. I used similar rounded pads but includes holes.10:11
wpwrakthe only problem we have now is that kicad sometimes ends up with endless lists of items :)10:12
cladamwin initial, I didn't know how to edit a round pad. :O  Then I know how to edit a round pad with PTH after i saw your this (pads.fpd) examples. :-)10:13
qi-bot[commit] Werner Almesberger: components/Makefile: "make" now generates instead of running "make sch" (master) http://qi-hw.com/p/kicad-libs/49fa1ab10:13
wpwrakheh :)10:13
cladamwletmme git pull and try. :-)10:14
wpwrakbtw, all the .fpd files should be listed in modules/Makefile, so that one can convert them to kicad .mod with a simple "make"10:14
cladamwgood, seems I ONLY generated *.ps to check dimensions. :-)10:15
cladamwif 'Write KiCad' cmd can be auto 'made' by make which will be smartly and useful to me at least. :-)10:17
wpwrak"make" is a source of power and magic ;-)10:18
cladamwgood, now i see that   gencon.lib 10:19
qi-bot[commit] Werner Almesberger: componens/adxl32x.lib: add pin for pad; put into components.pro (master) http://qi-hw.com/p/kicad-libs/6ecb85110:20
wpwraklet me update the catalog ...10:21
wpwrakwolfspraul: btw, generating the catalog could be a task for your magic server scripts10:22
wpwrakah, and we don't translate NC. let's fix that ...10:26
wpwrakcladamw: there's the ambiguity whether "NC" means "no internal connection" (so you can connect something if you want, e.g., to help with routing), or whether it means "do not connect" (so the pin has a function but you're not supposed to use it)10:29
wpwrakif we called the latter "DNC" instead of just "NC", do you think people would recognize the acronym ? or would it be too uncommon ?10:29
cladamw(DNC vs NC) severely speaking ... mmm good point now. we now need to redefine both. :-)10:32
cladamwpeople would hard to recognize. 10:33
cladamwdo you want to say DO NOT CONNECT ?10:34
wpwrakin kicad, you can already express this condition: "no internal connection" would be pin type "Unspecified". "do not connect" would be pin type "NC"10:34
cladamwa NC in sever datasheet will always describe very well as "no internal connection"10:35
wpwrakmy question is now for the catalog. should i follow kicad's example and translate "NC" to "NC", or should i show "DNC" instead ?10:35
wpwrak(NC in data sheets) yes, some say "no internal connection", some say "do not connect", and others just say "NC". the ones i trust most are those that have "do not connect" and "no internal connection" in the same data sheets (for different pins, of course). that means that they actually thought about it :)10:36
cladamwi prefer NC. :-)10:37
cladamw(in sch) we can define some texts there. :-)10:37
wpwrakalright. NC it is then10:37
DocScrutinizer05https://community.rapid7.com/community/metasploit/blog/2012/06/11/cve-2012-2122-a-tragically-comedic-security-flaw-in-mysql !!!!!10:59
wpwrakDocScrutinizer: subtle :) makes me wonder how/why they implemented such an odd check11:17
DocScrutinizer05a hash table of size 256?? dafaq11:18
DocScrutinizer05well, actually man 3 memcmp11:22
DocScrutinizer05WTF would anybody cast the return code to int8?11:24
DocScrutinizer05why would anybody care at all about the actual value, if !=011:24
DocScrutinizer05my CentOS box seems safe11:26
DocScrutinizer05at least the crack oneliner didn't succeed11:27
qi-bot[commit] Werner Almesberger: b2/test/subesc: more tests; also for values; add explanation of escaping (master) http://qi-hw.com/p/eda-tools/7f24ef011:55
qi-bot[commit] Werner Almesberger: genex/expand-pintype: add translation for "new" pin type NC (master) http://qi-hw.com/p/eda-tools/3949ead11:55
qi-bot[commit] Werner Almesberger: genex/expand-pintype: reduce character width (master) http://qi-hw.com/p/eda-tools/4cb56eb11:55
qi-bot[commit] Adam Wang: rj45.fpd: RJ45-8-LED added 8 pins ethernet connectors with two leds version from Molex. (master) http://qi-hw.com/p/kicad-libs/178fbc312:22
qi-bot[commit] Adam Wang: Merge branch 'master' of projects.qi-hardware.com:kicad-libs (master) http://qi-hw.com/p/kicad-libs/b9a392712:22
wpwrakinteresting ... i have a few batteries (AA primary cells) that are depleted to the point of having a negative voltage. i didn't even know they could do that :)13:20
viricyou didn't know? when you finish the batteries, you place them upside-down and loop ad infinitum13:27
viricit's a big pendulum effect13:28
viricthe fastest you unload the batteries, the more inverse charge.13:28
DocScrutinizer05lol13:36
DocScrutinizer05wpwrak: actually this indicates you mixed batteries in series with differing capacity. Rechargeables will get killed by that13:37
DocScrutinizer05actually the effect is a vicious circle: the battery with least capacity in your bat-pack gets damaged further the most and the first13:39
whitequarkpositive feedback!13:42
DocScrutinizer05that's btw a valid reasoning why you indeed *should* charge even LiIon *packs* for 12h before first use. And not discharge further than to 70% remaining during the first maybe 5 cycles13:42
DocScrutinizer05for LiIon single cells, all that doesn't apply13:43
wpwrakDocScrutinizer: hmm, i think i replaced them all at the same time. most of them are < 100 mV, one a bit more two read negative. they've been in wiimotes, so a pair in each. the wiimote has pretty large idle consumption.14:17
wpwrakwolfspraul: how's the PCB-making going ? have you made the first acid test yet ?14:40
wolfspraulhe. pushing it out :-)14:41
wpwrakdon't fear :)14:41
wolfspraulI 'almost' started a few times the last days, but I'm in a good run on the fpgatools so that's first right now14:41
wpwrakheh, okay. too much multitasking is bad, too :)14:43
Action: kristianpaul multitask all day at work :-|18:40
--- Tue Jun 12 201200:00

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