#qi-hardware IRC log for Saturday, 2012-04-21

viricmth: ok, thank you!07:25
qi-bot[commit] Adam Wang: fixed LP38511TJ-ADJ's GND to power input (master) http://qi-hw.com/p/kicad-libs/f36ddbf12:46
qi-bot[commit] Adam Wang: changed JS28F256J3F105's BYTE as Input (master) http://qi-hw.com/p/kicad-libs/261b47013:01
qi-bot[commit] Adam Wang: changed MIC2550AYTS's VTRM to Power input (master) http://qi-hw.com/p/kicad-libs/2c9b9b413:09
qi-bot[commit] Adam Wang: changed testpin to Passive (master) http://qi-hw.com/p/kicad-libs/966affe13:14
qi-bot[commit] Adam Wang: changed MT46V32M16P's VREF to input. (master) http://qi-hw.com/p/kicad-libs/9cd6b1113:22
LunaVoraxHello everyone!14:24
wolfspraulhi :-)14:24
wolfspraulI'm almost calling it a day though...14:25
rjeffriesreading backlog. someone asked if I pimp qi-hardware in Raspberry Pi channel. A: NO. I was not promoting Raspberry Pi here, merely mentioning the release of board schematics (a Good Thing, IMO). 17:22
rjeffriesNot sure I understand the latent hostimity to a $35 computer that will run Linux. While not perfectly open, they have released datasheet on the SoC and now we see schematics. As best I can tell, they are not forcing people to purchase the board.17:24
rjeffriesWhy not let 1,000 flowers bloom? Raspberry Pi may well fail, if so, then it failed. It is not a threat to Ben Nanonote, rather it has very different target market. 17:25
rjeffriesBen NN has been an interesting (ongoing experiment). I am not sure Raspberry Pi can not achive a similar level of open-ness as Ben NN. Both have published SoC datasheets. 17:27
rjeffriesThe life forces of qi hardware have already shifted to Milkymist, which has promise, although it will never be a $35 system. Will Milkylist someday be converted to an ASIC? It could happen. 17:29
DocScrutinizerwhat damn latent hostility?18:17
DocScrutinizerwpwrak: (precharge) hah, well you want some proper current limiters in a number of places18:22
DocScrutinizerwpwrak: and I don't think you could place "the capacitor to the right side of switch" here18:32
DocScrutinizeranyway we had sth similar with Vsys on GTA02, no? ;-D18:33
DocScrutinizer(current limiter) a LDO with /ENA input is probably the better alternative to a switch with current limiter18:38
DocScrutinizerthis way you got a decent decoupling from system VDD by design18:39
DocScrutinizerand a LDO probably is even cheaper than a switch with current limiter18:39
DocScrutinizerof course this requires some power rail providing Uout + Udrop_LDO18:40
DocScrutinizerrunning a commerical device without any on-board power stabilization is ... a bit unusual. As are good stabilized 5V wallwart PSU18:42
wpwrakDocScrutinizer: in GTA02, we solved the Vsys problem just by placing a cap on the other side of the switch :)20:07
wpwrakan LDO may be a bit inefficient. but yes, it's a possiblity, too.20:09
wpwrakjust the absence of inrush current limiting is generally a bad idea, as we keep seeing :)20:10
DocScrutinizerunless you have no separate power domains and a proper power-up sequence20:15
DocScrutinizerone of the main reasons for power-up sequences is to avoid such effects20:16
DocScrutinizerbasically you can take a boader approach than "limiting inrush": always consider both ends of the rope20:18
wpwrakDocScrutinizer: power-up sequences don't help in this case - it's all hot-plugged20:57
wpwrakor at least can be hot-plugged. same worst case :)20:58
--- Sun Apr 22 201200:00

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