01:44 <qi-bot> [commit] Werner Almesberger: sod.fpd (SOD-123): add SOD-123 (~4.5 x 1.7 mm diode) footprint (master) http://qi-hw.com/p/kicad-libs/34bf59d

03:21 <xiangfu> wpwrak: I received two Y-cables. one of them is works just fine.

03:21 <xiangfu> the other one have a small chip on it....

03:34 <wpwrak> great ! could you please check how the "nice" cable is connected internally ?

03:35 <wpwrak> i.e., if it's like http://downloads.qi-hardware.com/people/werner/anelok/tmp/y-cable.pdf

03:35 <wpwrak> or if they had some new ideas :)

03:35 <wpwrak> chip sounds interesting. what does it do ? ;-)

03:38 <xiangfu> we (me and archang) will upload some pics.

03:38 <xiangfu> it is excatly like http://downloads.qi-hardware.com/people/werner/anelok/tmp/y-cable.pdf.

03:38 <xiangfu> archang: can you please check what is that chip for?

03:39 <wpwrak> ah, you brought reinforcements. nice ! welcome, archang ! :)

03:39 <archang> oh. hello

03:40 <wpwrak> (exactly) including ID connected to GND on the Micro B (or A) side ?

03:40 <xiangfu> the working calbe is 9.9RMB.

03:40 <xiangfu> the one have chip is 22RMB. :)

03:40 <archang> both are Micro B with the ID connected to GND

03:41 <wpwrak> (ID) great, thanks ! it's nice when the industry anticipates my wishes ;-)

03:42 <xiangfu> the picutre on taobao is not same with the one they shipped to me.:)

03:42 <wpwrak> in the 9.9 RMB cable, how long are the branches ?

03:42 <xiangfu> wpwrak: I will send you two or three of them.

03:42 <xiangfu> both like 15CM.

03:44 <wpwrak> (picture) i guess they may just sell whatever they can find. but it's good that lower price means that it's more likely to work

03:44 <xiangfu> should we upload to anelok.com/people ?

03:44 <wpwrak> (15 cm) hmm, inconvenient if you don't use a laptop. but it's a good start nevertheless.

03:45 <wpwrak> yes, please

03:45 <wpwrak> and maybe post to the list as well

03:45 <xiangfu> wpwrak: ok. I will email the ssh key to you for anelok.com.

03:45 <xiangfu> wpwrak: list: yes.

03:46 <wpwrak> oh, sorry. didn't pay attention ... no, please upload to qi-hw

03:47 <xiangfu> wpwrak: ok. then I will create a wiki page.

03:47 <wpwrak> anelok.com is just my private server, trying to look important :)

03:47 <xiangfu> sure. just asking :)

03:48 <wpwrak> or just put it on http://downloads.qi-hardware.com/people/xiangfu/

03:51 <xiangfu> uploading...

03:52 <xiangfu> wpwrak: here is the taobao page: http://item.taobao.com/item.htm?spm=a1z09.2.9.9.R4rHd4&id=37895461380&_u=69mahte5c59

03:53 <xiangfu> wpwrak: there is another one. it is just like hte y-cable: http://item.taobao.com/item.htm?spm=2013.1.w4018-8662707918.8.xvGcUo&scm=1007.168.0.i37895461380&id=36212526756&ad_id=&pvid=6a28154b-3e7b-4eb6-9e37-e0c8904d4637&jlogid=p061152517607e

03:54 <wpwrak> ah, but that one has 2 x micro. my y-cable.pdf has only one micro. the full-size USB A male goes into PC or laptop

03:55 <wpwrak> yes, the ¥ cable looks like what i have in mind

03:57 <xiangfu> Yes.

03:57 <xiangfu> ￥cable. :)

03:57 <wpwrak> though the description seems to suggest that it may not be able to deliver power to the micro B side

03:57 <wpwrak> err, ¥ 12 ;-))

03:58 <xiangfu> archang: have tested on his phone.

03:58 <wpwrak> so "¥" means "money" ? i didn't know it was used for RMB, too, only noticed it for Yen.

03:58 <xiangfu> ¥ both RMB JPY

04:01 <wpwrak> heh, just like "$" is also used for all sorts of "pesos". in some countries, they add letters (which doesn't necessarily help. e.g., in uruguay, they put a "U" for uruguay and an "S" because it's plural. so you get "U$S", which you'll of course read as "US$" (which would be a lot more expensive). very confusing.)

04:08 <archang> Cellphone test result on 9.9 RMB cable: OTG works fine, tested on usb keyboard/mouse and usb disk; the power line only supports the otg slave device, no charging to my phone.

04:20 <wpwrak> hmm, so it's not quite what we want

04:23 <archang> we need to charge both the slave and host?

04:23 <wpwrak> it's nicer for anelok if it doesn't have to use its battery

04:24 <wpwrak> not that there would be any charging (i hope :)

04:27 <wpwrak> by the way, i'm making good progress with fixing the board errata. almost done with the principal changes. then i'll have to update and check the BOM.

04:30 <wpwrak> the coming version of the board will be almost identical to the prototype i have now, except that it's 1 mm wider (so the FPC of the OLED will suffer less), the rfkill switch is will really be just rfkill (and not system-kill), the crystal should be a bit more reliable (i had a layout bug there), and there will be a better diode on VUSB

04:30 <wpwrak> (plus a bunch of minor improvements)

07:43 <archang> wpwrak: an interesting thing about the 9.9 cable: cannot charge my nexus5 but ok on a nexus4...

07:54 <archang> and can power up a raspberry pi...

07:54 <archang> so it should meet the need

09:23 <wpwrak> archang: interesting. i think what makes the behaviour of such cables difficult to understand is that there are two mechanisms at play: one is the physical connection of VBUS and the other is the charger detection the software on the phone/tablet/etc. does

09:24 <wpwrak> the device only charges if there is a physical connection and the software decides to use it. so the same cable can work with some devices and not with other. worse, this may change after a sw update.

09:25 <wpwrak> and who knows, maybe there's some random element in this, too ;-)

09:26 <wpwrak> in the case of anelok it's simple: anelok will just assume the port can do 500 mA and it will tell this to the device. anelok itself needs only very little power, so as long as the port can do 500 mA, it'll be fine.

09:27 <wpwrak> looking at the pictures ... oh wow, the second cable is pretty complex !

21:38 <DocScrutinizer05> no chargiing not automatically means the power isn't connected to the micro plug. Actually I can't think of a Y-cable that doesn't connect VBUS to each plug

21:39 <DocScrutinizer05> ooh, [2015-01-06 Tue 08:43:38] <archang> wpwrak: an interesting thing

23:14 <wpwrak> DocScrutinizer05: it would kinda make sense not to connect power to the "data host" if there's a concern about overloading the "power host". but it's hard to say (without physical examination) whether the cable manufacturers do something like this.

23:22 <wpwrak> the development of such cables may be an evolutionary process: someone makes a nonstandard cable, sees what happens. then the next one makes a slightly different one, see if that causes less trouble in the field, and so on. once a design seems to work really well (at least with a product that's hot at the time), everyone clones that design. then things change (new device becomes fashionable, new droid version, etc.), etc.