**** BEGIN LOGGING AT Sun Jun 15 02:59:58 2014
Jun 15 05:13:26 <Psi> cool, that short i had on my n900 USB VCC to USB D- was the connector itself
Jun 15 05:13:30 <Psi> just unsoldered it and the short is gone
Jun 15 05:13:42 <Psi> so now the question is, has this short damaged the USB driver or not
Jun 15 05:16:12 <Psi> hopefully D- is 5v tolerant i/o
Jun 15 05:20:14 <ecc3g> i've seen some USB power outlets that have D+ shorted to D- which is in turn shorted to +5 ...
Jun 15 05:22:03 <Psi> cool that would point to them being 5v tolerant
Jun 15 05:22:57 <ecc3g> or that power outlet manufacturer wants to blow up devices :D
Jun 15 05:26:43 <Psi> we will see soon enough
Jun 15 05:26:52 <Psi> wiring up a usb cable to USB D- D+ GND VCC pads
Jun 15 07:50:08 <Psi> ok, im calling it, this N900 is deceased
Jun 15 07:54:15 <Psi> S/N 356938031287163    2010 - 2014
Jun 15 07:55:06 <Psi> RIP
Jun 15 08:08:27 <Humpelstilzchen> what have you done? :(
Jun 15 08:08:43 * joga lits candle
Jun 15 08:42:58 <Psi> Humpelstilzchen: it just stopped detecting USB cable for charging or data
Jun 15 08:43:57 <Psi> so i installed another kernel (incase the power kernel i had installed a few weeks back was causing usb issues)  except i must have installed the wrong one because it wouldnt boot at all after that
Jun 15 08:44:21 <Psi> so now it wont boot and pc cant detect USB to reflash it
Jun 15 08:45:08 <Psi> i removed usb socket (it had a short between D- and VCC) and soldered directly to the test pads instead. But PC still cant recognise the usb device
Jun 15 08:45:20 <Psi> i think the short to VCC damaged the usb chip
Jun 15 08:45:31 <bencoh> and ... you dont have uboot on it ?
Jun 15 08:45:44 <Psi> nope, no boot menu
Jun 15 08:46:18 <Psi> i used to have one, but last month it stopped booting and i had to do a full hardware reset so lots all that stuff
Jun 15 08:46:33 <Psi> i hadnt got around to putting it back on
Jun 15 08:46:46 <Psi> otherwise yeah, could boot from usd
Jun 15 08:47:36 <Psi> if i had a BGA rework station i could probably unsolder the flash chip and reflash the kernel externally but meh, too much effort
Jun 15 08:47:55 <Psi> plan to buy a LG g3 when it comes out next month
Jun 15 08:48:13 <hurrian> Psi: you'd need one hell of a rework station, the OneNAND is on top of the OMAP, and the pads are /tiny/
Jun 15 08:49:33 <Psi> true, the kernel is on the oneflash isnt it, not the 32gb flash
Jun 15 08:49:48 <hurrian> i've got a OMAP3530 on my desk, and trying to focus on any features hurts my head, I need a loupe.
Jun 15 08:50:33 <Psi> could also probably fix it if i knew the jtag headers and specs
Jun 15 08:50:45 <hurrian> you'll probably just want to chuck the N900 in a drawer and wait until you're motivated to put on a new USB port :P
Jun 15 08:51:05 <bencoh> or replace it with a neo900 board
Jun 15 08:51:07 <Psi> na, ive already soldered to the USB test pads and tried it, its definitly dead
Jun 15 08:52:05 <Psi> i also found one loose smt cap in the FM TX section and one ESD chip broken while i was looking at the usb socket
Jun 15 08:52:39 <Psi> its possible one of those loose bits caused the problem by shorting something
Jun 15 08:53:28 <hurrian> if you've got the massive balls and 1337 h4xx0r skills to do so, rip out the OneNAND, put back the PoP DRAM, and try to flash x-loader using omapflash to the eMMC :D
Jun 15 08:55:38 <hurrian> come to think of it, we probably should've asked OSSO that some time ago - enable OMAP ROM booting from USB!
Jun 15 09:08:53 <Psi> how does omapflash work?
Jun 15 09:09:05 <Psi> nevermind
Jun 15 09:12:09 <Psi> is there a rough idea what the neo900 upgrade board will cost?
Jun 15 09:13:19 <bencoh> around 500 afaik
Jun 15 09:13:41 <bencoh> maybe more
Jun 15 09:15:42 <Psi> 1ghz isnt really fast enough
Jun 15 09:16:25 <Psi> a n900 overcloked to 1ghz is still quite slow at some things
Jun 15 10:56:44 <DocScrutinizer05> hurrian: attach your N900 to a linux PC, check syslog of PC. You'll find "ROM boot" enumerating
Jun 15 10:57:03 <DocScrutinizer05> hurrian: that's how coldflash works
Jun 15 10:57:22 <DocScrutinizer05> Psi: 700
Jun 15 10:58:13 <DocScrutinizer05> and sorry for that, but that's actaully what it costs to build such thing
Jun 15 10:58:45 <DocScrutinizer05> we're not buying the chips from TI in 100k qty
Jun 15 11:00:35 <DocScrutinizer05> NB this is still only an educated estimation, we don't know yet what it will cost in the end, since it's not finalized yet
Jun 15 11:12:13 <Psi> yeah, i wasnt bitching at the price
Jun 15 11:12:45 <Psi> i expect about that
Jun 15 11:12:56 <Psi> due to low qty
Jun 15 11:26:39 <bencoh> hmm, what would one use to fetch/parse json in a maemo/hildon app ?
Jun 15 11:27:18 <bencoh> is there some simple api or is it back to roots with libjson and curl/gsoup ?
Jun 15 11:39:17 <DocScrutinizer05> Psi: you probably could flash via the test pads under battery, using phenix software and a jtag rig
Jun 15 11:40:02 <DocScrutinizer05> Psi: anyway the OMAP should enum on USB during power-on 
Jun 15 11:40:21 <DocScrutinizer05> no NAND or kernel involved in that
Jun 15 11:43:02 <Psi> yeah, but i suspect my usb port driver is damaged
Jun 15 11:43:30 <DocScrutinizer05> that would be 1704/1707 PHY then
Jun 15 11:43:46 <Psi> yeah
Jun 15 11:44:33 <DocScrutinizer05> D4280
Jun 15 11:44:42 <DocScrutinizer05> in schematics
Jun 15 11:44:43 <Psi>  D- got shorted to VCC on connector (must have been something metalic jammed under connector)
Jun 15 11:44:58 <DocScrutinizer05> *should* not kill the 1707
Jun 15 11:45:12 <Psi> i unsoldered the usb connector and the short went away but the usb still doesnt work
Jun 15 11:45:38 <DocScrutinizer05> are you sure the USB connector itself isn't damaged?
Jun 15 11:46:11 <Psi> probably
Jun 15 11:47:03 <Psi> but i used the usb test pads with the connector unsoldered (short problem fixed) and my PC still cannot recognise the usb device 
Jun 15 11:47:51 <Psi> any idea what component is, i suspect ESD protection http://psi.abcom.co.nz/n900pcb.jpg
Jun 15 11:48:55 <Psi> i found it to be smashed in half when looking at usb socket. Its right near where the stylis tip ends up inside the case which is a bit suspicious. I suspect it has nothing to do with usb prob
Jun 15 11:51:35 <DocScrutinizer05> well, apart from my SIM card I see there: V6180 FMTX Ant ESD protector
Jun 15 11:51:56 <Psi> hehe yeah, grabbed pic from goggle image search hehe
Jun 15 11:53:31 <DocScrutinizer05> FMTX largely unrealted to USB ;-)
Jun 15 11:53:37 <Psi> yeah
Jun 15 11:54:25 <Psi> whats interesting is that windows detects that the usb device is malfuntioned as soon as i plug it in, before even the n900 starts up in usb mode  (pressing u key)
Jun 15 11:54:37 <DocScrutinizer05> but that component being damaged indicates some severe abuse of device
Jun 15 11:55:10 <Psi> i wouldnt call being dropped 3 times in its life abuse
Jun 15 11:55:15 <DocScrutinizer05> please check with linux, watch syslog!
Jun 15 11:56:01 <Psi> http://pastebin.com/949NjWzL
Jun 15 11:56:27 <Psi> however, that's when it was shorted, ive not tried it in linux since fixing the short
Jun 15 11:56:28 <DocScrutinizer05> I know there are diagnostic tools for USB on windows, but I don't know s*t about windows
Jun 15 11:58:20 <DocScrutinizer05> USB is not tolerant to bad wiring
Jun 15 11:58:54 <DocScrutinizer05> even some regular USB cables smetimes might introduce data errors so ENUM fails
Jun 15 11:59:05 <DocScrutinizer05> when poor cable quality
Jun 15 12:00:15 <hurrian> mmm, negotiates speed at Low Speed/1.0, cables might be the culprit after all
Jun 15 12:00:35 <DocScrutinizer05> D+ and D- form a differential 50Ohm(?) line, which means the two wires must run parallel in always same distance and no sharp bends or other incontinuities
Jun 15 12:00:52 <DocScrutinizer05> hurrian: right
Jun 15 12:01:36 <Psi> this is with short fixed http://pastebin.com/SzXkMyPc
Jun 15 12:02:09 <kerio> DocScrutinizer05: are you saying that my 0.20€ chinese usb cable is not certified? D:
Jun 15 12:02:57 <dos1> (differential line) ah, that explains a lot :)
Jun 15 12:03:40 <dos1> I have some cheap Y-cable with mini and micro plugs at its two ends, and I wondered why it's working randomly
Jun 15 12:04:53 <dos1> it's great for charging anyway ;)
Jun 15 12:07:10 <Psi> na i soldered a 20cm usb cable to the pads, i doubt its a cable issue
Jun 15 12:11:08 <Psi> http://psi.abcom.co.nz/n900cable.jpg
Jun 15 12:14:40 <DocScrutinizer05> that looks like a massive incontinuity in D+/- lines
Jun 15 12:14:54 <DocScrutinizer05> might work, but not guaranteed to do
Jun 15 12:15:13 <Psi> you think i should twist them right up until the pads?
Jun 15 12:15:42 <DocScrutinizer05> keep them parallel and straight (without sharp bends)
Jun 15 12:16:14 <DocScrutinizer05> and don't approach the steel too much, next to the pads
Jun 15 12:17:23 <DocScrutinizer05> btw soldering to the battery contact blades is begging for later trouble. The tin coating never makes proper contact to the battery, since it oxidizes
Jun 15 12:17:31 <DocScrutinizer05> ask SpeedEvil
Jun 15 12:19:19 <DocScrutinizer05> but probably you really have a problem in your N900 hardware. Just the question is, what exactly
Jun 15 12:20:17 <Agge> Doc did I tell you that I got 3 china nokia chargers that have the same serial numbre
Jun 15 12:21:27 <DocScrutinizer05> lol, nice
Jun 15 12:22:26 <DocScrutinizer05> though it's questionable if chargers actually have a SERIAL-number
Jun 15 12:22:46 <Agge> original nokia ones do
Jun 15 12:23:12 <Agge> opened one and an original nokia one and it was a masive difference in the build quality
Jun 15 12:23:16 <DocScrutinizer05> seems applying an individula unique number to every single charger is overkill and too expensive even for genuine original nokia chargers
Jun 15 12:23:37 <DocScrutinizer05> ok, opening is a good proof
Jun 15 12:23:39 <DocScrutinizer05> :-)
Jun 15 12:24:03 <Agge> the original one that I opened also had a serial numbre
Jun 15 12:24:26 <DocScrutinizer05> do you have two originals that have different serial?
Jun 15 12:24:26 <Psi> yeah i knew soldering batt things was bad, but i was careful to only put solder on the top, not on the surface that touches the battery
Jun 15 12:24:48 <DocScrutinizer05> Psi: smart! :-)
Jun 15 12:25:16 <Agge> Psi: or further down then the battery touch?
Jun 15 12:25:37 <Agge> not from the same production line. have acces to one old round and own one that was dilivered with my old N900
Jun 15 12:25:47 <Agge> microusb one
Jun 15 12:27:09 <Psi> http://psi.abcom.co.nz/n900cable2.jpg
Jun 15 12:27:12 <Psi> 2nd attempt
Jun 15 12:27:13 <DocScrutinizer05> carefully removing tin by desolder braid, and then with some weak acid (I'm not a good at chemicals, maybe HCl) should also fix solder problems on blades, by removing solder but not really ruining the gold coating
Jun 15 12:28:43 <DocScrutinizer05> Psi: that looks much better, yes
Jun 15 12:29:37 <Psi> exactly the same responce in logs
Jun 15 12:30:42 <DocScrutinizer05> possibly parts of the protective gear on D+/- got fried
Jun 15 12:31:20 <Psi> i had a look on the scope when it had the shorted problem
Jun 15 12:31:54 <Agge> Psi: are you using a original nokia battery or an aftermarket one?
Jun 15 12:32:00 <Psi> origonal
Jun 15 12:32:06 <Agge> hade some problems with aftermarket ones
Jun 15 12:32:09 <Psi> well.. a lab psu atm, so no batt at all
Jun 15 12:33:55 <Psi> as soon as i connect vbat power i get the white nokia logo screen, then it goes back for a few sec, vibrates and starts again with the white screen
Jun 15 12:34:10 <Psi> back=black
Jun 15 12:34:29 <Psi> pressing U gets the USB logo
Jun 15 12:34:41 <Agge> exactly what hapened with one phone when I tryed to use a aftermarket battery
Jun 15 12:34:43 <Psi> but thats about all ive seen it do
Jun 15 12:35:06 <Psi> i wonder if it doesnt like the missing temp sensor wire being disconnected
Jun 15 12:35:07 <Agge> exactly what do the third battery conector pin do?
Jun 15 12:35:56 <Agge> migh be some risistance or somthing that needs to be ablied for it to like the battery?
Jun 15 12:36:07 <DocScrutinizer05> Psi: definitely a problem
Jun 15 12:36:12 <DocScrutinizer05> (missing BSI)
Jun 15 12:36:36 <DocScrutinizer05> 80kR to ... GND(?)
Jun 15 12:36:38 <Psi> i dont think thats the cause of the usb issue though
Jun 15 12:36:53 <DocScrutinizer05> NOLO already checking BSI
Jun 15 12:37:30 <DocScrutinizer05> there's even a special "test battery" that switches device into debug mode via special resistor value on BSI
Jun 15 12:37:30 <Agge> Psi: what about the lab suply and usb conection to a computer?
Jun 15 12:37:49 <DocScrutinizer05> JTAG jig also has a special BSI resistor
Jun 15 12:38:05 <DocScrutinizer05> APE *and* modem +both* check BSI value
Jun 15 12:38:19 <Psi> Agge: ?
Jun 15 12:39:16 <Agge> Psi: did you get any difference to the syslog output after you begun using the lab power suply insted of a battery?
Jun 15 12:39:35 <DocScrutinizer05> Psi: attach 70..80kR BSI - GND!
Jun 15 12:40:53 <DocScrutinizer05> it's probably unrelated to your USB trouble, but for sure introduces even more random weird effects when BSI is missing
Jun 15 12:41:08 <Psi> Agge: ah, not sure
Jun 15 12:41:14 <Psi> its a good question though
Jun 15 12:41:20 <DocScrutinizer05> please test with linux PC and powered down N900
Jun 15 12:41:29 <Psi> i have put 100R onto BSI (same as offical bat)
Jun 15 12:41:40 <DocScrutinizer05> you should see a proper Omap ROM Boot ENUM
Jun 15 12:41:50 <DocScrutinizer05> WUT???
Jun 15 12:41:57 <Psi> sorry 100k
Jun 15 12:42:09 <Agge> Psi: I would also test if I wold be able to get the rescue os to boot
Jun 15 12:42:12 <Psi> i measued the offical batt and got 100k
Jun 15 12:42:15 <DocScrutinizer05> well, original pat has 75..80
Jun 15 12:42:19 <DocScrutinizer05> but irrelevant
Jun 15 12:42:27 <DocScrutinizer05> 100k works too
Jun 15 12:42:44 <Psi> Agge: rescue os?
Jun 15 12:43:43 <Agge> an image that boot loads into ram and boot the phone from it. I found it a good way to see if stuff wants to work att all
Jun 15 12:43:51 <Agge> ~rescueos
Jun 15 12:44:07 <Agge> I am probibly spelling it wrong :(
Jun 15 12:45:39 <DocScrutinizer05> ok, you're right. 99.8k on one of my batt
Jun 15 12:46:01 <DocScrutinizer05> infobot dead
Jun 15 12:46:05 <DocScrutinizer05> mompls
Jun 15 12:46:38 <DocScrutinizer05> should be back soon
Jun 15 12:46:45 <Agge> wont do anything if the flacher cant see the phone. but when you tested most logical things it is time to try the strange ideas :)
Jun 15 12:47:00 <DocScrutinizer05> :-)
Jun 15 12:47:46 <Psi> slightly different behavour with 100R
Jun 15 12:47:47 <DocScrutinizer05> Agge: doesn't make sense when USB doesn't even enumerate
Jun 15 12:47:56 <Psi> 100k damnit
Jun 15 12:47:57 <Agge> yeah sorry
Jun 15 12:48:05 <infobot> DocScrutinizer: infobot joined!
Jun 15 12:48:10 <DocScrutinizer05> ~wb
Jun 15 12:48:11 <infobot> thx
Jun 15 12:48:23 <Psi> logs on pc are the same, but with the 100K i can get to the non backlit nokia logo screen
Jun 15 12:48:31 <Psi> before it would always start on the white one
Jun 15 12:48:52 <Psi> i cant seem to get the phone to actally turn off
Jun 15 12:49:12 <Psi> its either in usb mode or doing infinite reboots
Jun 15 12:50:25 <Agge> Psi: shut of the PSU for at least one minute?
Jun 15 12:50:44 <Psi> the phone always starts automatically when you insert a battery
Jun 15 12:50:48 <DocScrutinizer05> 2014-06-15T14:49:59.996024+02:00 localhost kernel: [2037151.880303] usb 2-1.5.3: new high-speed USB device number 11 using ehci-pci
Jun 15 12:50:50 <DocScrutinizer05> 2014-06-15T14:50:00.082030+02:00 localhost kernel: [2037151.966356] usb 2-1.5.3: unable to get BOS descriptor
Jun 15 12:50:51 <DocScrutinizer05> 2014-06-15T14:50:00.083027+02:00 localhost kernel: [2037151.967541] usb 2-1.5.3: New USB device found, idVendor=0421, idProduct=0106
Jun 15 12:50:53 <DocScrutinizer05> 2014-06-15T14:50:00.083030+02:00 localhost kernel: [2037151.967545] usb 2-1.5.3: New USB device strings: Mfr=34, Product=38, SerialNumber=0
Jun 15 12:50:54 <DocScrutinizer05> 2014-06-15T14:50:00.083030+02:00 localhost kernel: [2037151.967548] usb 2-1.5.3: Product: Nokia USB ROM
Jun 15 12:50:56 <DocScrutinizer05> 2014-06-15T14:50:00.083031+02:00 localhost kernel: [2037151.967551] usb 2-1.5.3: Manufacturer: Nokia
Jun 15 12:51:00 <Psi> yeah, i know what it supposed to do
Jun 15 12:51:44 <DocScrutinizer05> that however it is supposed to do from ROM Bootloader, *unconditionally* (unless hw is defect)
Jun 15 12:52:14 <Agge> Psi: yeah but I have found that even if you remove the battery it wont start corectly unless you leve it unconected for a minute or so
Jun 15 12:52:26 <Psi> i should trace the d- and d+ back to the usb driver chip
Jun 15 12:52:27 <DocScrutinizer05> while this is NOLO already:
Jun 15 12:52:29 <DocScrutinizer05> 2014-06-15T14:50:00.586041+02:00 localhost kernel: [2037152.469920] usb 2-1.5.3: USB disconnect, device number 11
Jun 15 12:52:31 <DocScrutinizer05> 2014-06-15T14:50:01.276026+02:00 localhost kernel: [2037153.159029] usb 2-1.5.3: new high-speed USB device number 12 using ehci-pci
Jun 15 12:52:32 <Psi> make sure they actually get there
Jun 15 12:52:32 <DocScrutinizer05> 2014-06-15T14:50:01.439030+02:00 localhost kernel: [2037153.321663] usb 2-1.5.3: New USB device found, idVendor=0421, idProduct=0105
Jun 15 12:52:34 <DocScrutinizer05> 2014-06-15T14:50:01.439046+02:00 localhost kernel: [2037153.321676] usb 2-1.5.3: New USB device strings: Mfr=1, Product=2, SerialNumber=5
Jun 15 12:52:35 <DocScrutinizer05> 2014-06-15T14:50:01.439048+02:00 localhost kernel: [2037153.321677] usb 2-1.5.3: Product: Nokia N900 (Update mode)
Jun 15 12:52:37 <DocScrutinizer05> 2014-06-15T14:50:01.439049+02:00 localhost kernel: [2037153.321679] usb 2-1.5.3: Manufacturer: Nokia
Jun 15 12:52:38 <DocScrutinizer05> 2014-06-15T14:50:01.439049+02:00 localhost kernel: [2037153.321680] usb 2-1.5.3: SerialNumber: 4D554D323931313733
Jun 15 12:53:14 <Psi> Agge: ive started it many times after working on it unpluged for 10+ minutes and it always starts when the battery is connected
Jun 15 12:53:27 <Agge> Psi: ok
Jun 15 12:54:43 <DocScrutinizer05> Psi: you should check if ROM bootloader enumerates on USB
Jun 15 12:54:54 <Psi> how do i check that
Jun 15 12:55:08 <Psi> is it a different hold key or something
Jun 15 12:55:24 <DocScrutinizer05> connect powered-down device to linux machine, then provide power via USB
Jun 15 12:56:03 <DocScrutinizer05> ROM boot will not check for any keys pressed
Jun 15 12:56:27 <DocScrutinizer05> ROM BL only checks USB and enums, every single time the CPU boots
Jun 15 12:56:53 <DocScrutinizer05> this is how clodflash works
Jun 15 12:57:01 <DocScrutinizer05> ~coldflash
Jun 15 12:57:02 <infobot> extra, extra, read all about it, coldflash is http://mg.pov.lt/maemo-irclog/%23maemo.2010-10-31.log.html#t2010-10-31T23:34:01 (jacekowski's instructions), or http://wiki.maemo.org/Updating_the_firmware/Cold_Flashing
Jun 15 12:57:33 <Psi> same thing in logs, and phone just keeps rebooting
Jun 15 12:57:57 <DocScrutinizer05> it's NOT (I *repeat: NOT) recommended to do cold flashing ever, unless you have very strong ndications that NOLO is defective
Jun 15 12:58:24 <DocScrutinizer05> your logs are incomplete
Jun 15 12:58:41 <DocScrutinizer05> they only show end of a failing enum
Jun 15 12:59:00 <DocScrutinizer05> to see what's going on you need first few lines, not last
Jun 15 12:59:45 <DocScrutinizer05> it doesn't matter if device reboots or whatever
Jun 15 13:00:33 <DocScrutinizer05> as soon as CPU gets powered up, it executes ROM BL and (one of) first things ROM BL does is ENUM on USB, to check if a flasher asks for coldflash
Jun 15 13:01:25 <DocScrutinizer05> ROM BL is the initial bootloader, stored to ROM in CPU and immutable
Jun 15 13:01:43 <Psi> this is all i get in kern.log http://pastebin.com/WSxx2zDX
Jun 15 13:01:52 <Psi> maybe i need to enable better loging
Jun 15 13:02:06 <DocScrutinizer05> ROM BL does USB ENUM before it initializes any LCD or anything else
Jun 15 13:03:25 <Psi> im familar with the message MSwindows is giving when i plug it into that, its the message you get when there is serious USB communication problems
Jun 15 13:03:26 <DocScrutinizer05> if that's the very first line that shows up after you attach your powered-down N900, then you got a hw problem (or driver problem on PC side, but that's pretty unlikely)
Jun 15 13:03:49 <Psi> yeah
Jun 15 13:04:09 <Psi> i think there must be a broken data line to the driver chip. Or the chip is fried
Jun 15 13:04:43 <DocScrutinizer05> particularly it shows that *some* of USB data still works, or otherwise the PC wouldn't even notice the USB device attached
Jun 15 13:05:21 <DocScrutinizer05> broken data line is possible but unlikely, particularly when you attached cable the way you did
Jun 15 13:05:49 <DocScrutinizer05> a short of one data line to something else is highly likely
Jun 15 13:06:25 <DocScrutinizer05> fried PHY is also possible but only likely when some terribly broken charger got attached
Jun 15 13:06:45 <Psi> well it can be VCC or GND as i have checked ohms with DMM and its in the meg range between D- and D+ to GND or VCC
Jun 15 13:06:55 <DocScrutinizer05> 1707 PHY supposed to survive 5V on data line, and for sure supposed to survive data line grounded
Jun 15 13:07:43 <DocScrutinizer05> please check voltage on open-end D- D+
Jun 15 13:07:47 <DocScrutinizer05> vs GND
Jun 15 13:08:01 <Psi> open-end ?
Jun 15 13:08:10 <DocScrutinizer05> I don't recall the exact values from top of my head, but for sure not 0V
Jun 15 13:08:18 <DocScrutinizer05> not connected to PC
Jun 15 13:08:26 <DocScrutinizer05> only device powered up
Jun 15 13:08:30 <honestly> USB uses differential signaling
Jun 15 13:08:36 <DocScrutinizer05> I know
Jun 15 13:08:55 <DocScrutinizer05> naming of D- D+ suggests that already
Jun 15 13:09:01 <honestly> so idle connection should have both lines pulled up somewhat
Jun 15 13:09:10 <DocScrutinizer05> yup
Jun 15 13:09:12 <honestly> at 2V or something like that
Jun 15 13:09:16 <DocScrutinizer05> :nod:
Jun 15 13:09:20 <honestly> (:
Jun 15 13:10:08 <Psi> both D- and D+ goto about 0.3V on DMM for about 0.5second right at power on when it vibrates
Jun 15 13:10:12 <Psi> then goto zero volts
Jun 15 13:10:55 <Psi> however DMM maybe lying, should really  use scope
Jun 15 13:11:51 <DocScrutinizer05> actually I think *idle* USB device PHY has both pulled up weak (range of  100k), while active device (Vbus +5V) PHY is supposed to pull up one by 1.5k and pull down other by 15k or somesuch, depending on low speed or full speed it's either D+ or D- that's pulled up. I can't recall details from top of my head
Jun 15 13:12:33 <DocScrutinizer05> Psi: that sounds conclusive
Jun 15 13:13:06 <DocScrutinizer05> well, except for the 0.3V which probably are somewhat higher, but DMM is too slow
Jun 15 13:13:47 <Psi> yeah, scope says its actually 0.6v
Jun 15 13:13:48 <DocScrutinizer05> Psi: I take it you have a lab PSU attached to battery?
Jun 15 13:13:53 <Psi> yep
Jun 15 13:14:01 <Psi> at 4.0V
Jun 15 13:15:00 <DocScrutinizer05> a battery voltage of <3.0V should cause CPU not booting and instead bq24150 charger chip asking 1707 PHY about D+/- short and depending on that it lits up the steady yellow indicator LED
Jun 15 13:15:19 <DocScrutinizer05> make Vbatt 2.3V please
Jun 15 13:15:33 <DocScrutinizer05> then apply 5V Vusb
Jun 15 13:16:00 <DocScrutinizer05> do this one time with open D+/- and one time with D+/- short
Jun 15 13:17:47 <DocScrutinizer05> sorry I have no discharged battery here, so can't reproduce the procedure. So you have to trust in my poor and clouded memory, or check the schematics and bq24150 datasheet yourself
Jun 15 13:19:00 <DocScrutinizer05> I *think* it's even sufficient to connect Vbatt+ and Vbatt-GND with a e.g. 100R, and not apply any voltage at all
Jun 15 13:19:44 <Psi> yep, led comes on with D- and D+ as they are.  If i short them and repeat i get no led
Jun 15 13:19:58 <DocScrutinizer05> ooh
Jun 15 13:20:07 <DocScrutinizer05> that's kinda weird
Jun 15 13:20:19 <DocScrutinizer05> should be exactly other way round
Jun 15 13:20:33 <DocScrutinizer05> LED is amber?
Jun 15 13:20:38 <Psi> yellow
Jun 15 13:20:47 <DocScrutinizer05> or is it red, or green, or flashing?
Jun 15 13:20:56 <Psi> not red or green
Jun 15 13:21:00 <DocScrutinizer05> steady yellow, yep that's correct
Jun 15 13:21:01 <Psi> not flashing, just solid
Jun 15 13:21:15 <DocScrutinizer05> weird
Jun 15 13:21:21 <Psi> it wouldnt be able to draw the full 1A charge from usb PC 5v anyway
Jun 15 13:21:39 <DocScrutinizer05> np, that's not what it tries to do
Jun 15 13:22:29 <DocScrutinizer05> it however would try to charge up battery from charger with as much as it can pull from USB
Jun 15 13:22:44 <DocScrutinizer05> can you see on lab PSU how much it pulls
Jun 15 13:22:45 <DocScrutinizer05> ?
Jun 15 13:23:18 <DocScrutinizer05> or see voltage on battery pins going up, or sth?
Jun 15 13:23:28 <Psi> yep, i can see that, it pulls the 2.3v up to 3.25V
Jun 15 13:23:43 <Psi> can see it on lab psu meter
Jun 15 13:24:20 <DocScrutinizer05> It might be that it charges "battery" with 100mA when D+/- ope circuit, and tries with 500ma on D+/- short, thus causing Vbus supply cutting out and no yellow LED
Jun 15 13:24:41 <DocScrutinizer05> 3.25 is correct for charging, yes
Jun 15 13:24:57 <DocScrutinizer05> it's emergency charging end voltage
Jun 15 13:25:11 <Psi> ah, makes sense
Jun 15 13:25:46 <Psi> do you know if the usb driver IC is BGA or not
Jun 15 13:25:52 <DocScrutinizer05> when end voltage reached, bq24150 quits emergency charging and allows CPU to boot up
Jun 15 13:26:04 <DocScrutinizer05> it's BGA, yes
Jun 15 13:27:08 <DocScrutinizer05> http://wiki.maemo.org/N900_Hardware_USB#USB_Phy
Jun 15 13:27:33 <Psi> is there a manual for the n900 which matches the part numbers in the schmatic diagram to physical locations on the pcb?
Jun 15 13:28:59 <DocScrutinizer05> the schematics itself has that, at end (page 15 orso)
Jun 15 13:29:10 <Psi> really, i must have missed it
Jun 15 13:29:11 * Psi looks
Jun 15 13:29:58 <DocScrutinizer05> sorry, page 16
Jun 15 13:30:09 <Psi> yeah, im looking at it now
Jun 15 13:30:18 <Psi> for some reason i thought that was missing
Jun 15 13:30:30 <DocScrutinizer05> http://wiki.maemo.org/N900_Hardware_Schematic#Page_16_Component_Finder_.28Top.29
Jun 15 13:31:55 <DocScrutinizer05> anyway your tests suggest that 1707a PHY can detect a D+/- short correctly. That tells some story about what must be OK and working on your device
Jun 15 13:32:36 <Psi> yeah
Jun 15 13:32:57 <DocScrutinizer05> particularly it does suggest that none of D+ D- traces to phy is broken
Jun 15 13:33:08 <Psi> now that i know how to find components on the board and trace the circuit i will have another play with the meter looking for faults
Jun 15 13:33:37 <Psi> oh, there is a missing cap on the FM tx that i found loose in the phone
Jun 15 13:33:42 <Psi> C6184
Jun 15 13:33:58 <Psi> its part of the sleep clock input to the FM tx chip
Jun 15 13:34:06 <DocScrutinizer05> ouch
Jun 15 13:34:19 <Psi> the cap was still there, inside the metal can
Jun 15 13:34:21 <Psi> just loose
Jun 15 13:34:28 <DocScrutinizer05> OUCH!
Jun 15 13:35:22 <Psi> yeah, im beginning to wonder if one of the two broken components inside the phone may have touched something and shorted it out
Jun 15 13:35:31 * DocScrutinizer05 wonders how to find out which cap got loose, without opening can and while doing that, at least remove the loose cap
Jun 15 13:36:23 <Psi> yeah, i have removed both the broken ESD chip and the borken cap
Jun 15 13:36:25 <DocScrutinizer05> yeah, a loose component is always a high risk to short other stuff randomly
Jun 15 13:36:38 <Psi> cap looks ok, its still intact but esd chip i only found one half of it
Jun 15 13:37:01 <DocScrutinizer05> wtf? how can such a tiny chip break into parts
Jun 15 13:37:24 <Psi> have a look where that chip is, its right at the point of the stylis
Jun 15 13:37:41 <DocScrutinizer05> hmm, not really
Jun 15 13:37:52 <DocScrutinizer05> some millimeter space between the two
Jun 15 13:38:09 <Psi> i dunno, just seems an odd coincidence
Jun 15 13:38:26 <Psi> i do push the stylis in with some force
Jun 15 13:38:41 <DocScrutinizer05> sure, when you bend the whole device a 10mm... ;-P
Jun 15 13:38:47 <Psi> heh
Jun 15 13:39:18 <Psi> ok, i need to sleep, its 1:40 here and i have to get up for work at 6
Jun 15 13:39:26 <DocScrutinizer05> or insert something that's not exactly a correctly shaped stylus
Jun 15 13:39:27 <Psi> thanks a lot for your help
Jun 15 13:39:41 <DocScrutinizer05> yw, good luck!
Jun 15 13:39:47 <Psi> na, ive never put anything into the stylus hole thats not a stylus
Jun 15 13:40:13 <Psi> maybe the chip came loose and got hit by the vibrator
Jun 15 13:40:23 <DocScrutinizer05> possible
Jun 15 13:40:23 <Psi> that might crack it
Jun 15 13:40:29 <DocScrutinizer05> yup
Jun 15 13:40:55 <Psi> ok, see ya.  Thanks
Jun 15 13:40:59 <DocScrutinizer05> cya
Jun 15 18:54:57 <emma> hello
Jun 15 18:55:06 <emma> May I ask what is the latest good news about Maemo?
Jun 15 19:56:40 <DocScrutinizer05> maybe the upcomming coding competition?
Jun 15 20:00:37 <DocScrutinizer05> maybe the rumor that Neo900 UG ponders to donate (an option for) a device or alternatively a sustantial money price for a yet to get specified port of fremantle?
Jun 15 20:01:15 <DocScrutinizer05> suBstantial*
Jun 15 20:01:30 <DocScrutinizer05> and prize of course
Jun 15 20:03:03 <honestly> I still can't parse that properly
Jun 15 20:03:11 <honestly> who is donating what for what?
Jun 15 20:03:20 <honestly> do you mean offering a bounty?
Jun 15 20:06:51 <emma> why is there a bounty?
Jun 15 20:09:27 <DocScrutinizer05> errr, a competition is about prizes, winners and all that, no?
Jun 15 20:10:19 <DocScrutinizer05> honestly, is that so obscure? http://talk.maemo.org/showthread.php?t=93334
Jun 15 20:10:43 <honestly> oh, I see.
Jun 15 21:09:44 <Gh0sty> hello I was wondering ...
Jun 15 21:10:07 <Gh0sty> it's possible to run the n900 as mifi device (sharing 3G over wifi) 
Jun 15 21:11:05 <Gh0sty> but from what I read it's the same as most other linuxes ... it only supports WEP encryption? 
Jun 15 21:12:14 <honestly> as far as I know, yes.
Jun 15 21:12:38 <honestly> you can also share 3G via usb.
Jun 15 21:12:51 <Gh0sty> just for my interest but why do android phones support wpa2 accesspoint mode
Jun 15 21:13:06 <Gh0sty> but thinks like an n900 or for example my laptop don't ? :/ 
Jun 15 21:13:20 <Gh0sty> something to do with the driver or is that even a feature that needs hardware support? 
Jun 15 21:20:15 <DocScrutinizer05> who says N900 doesn't support WPA?
Jun 15 21:20:52 <DocScrutinizer05> not saysing that it does. Just wondering why you think it doesn't
Jun 15 21:21:18 <honestly> there are no mobile hotspot apps offering it (:
Jun 15 21:21:52 <DocScrutinizer05> it supports WLA(2) for normal (infrastructure) mode device
Jun 15 21:22:04 <DocScrutinizer05> WPA
Jun 15 21:23:24 <DocScrutinizer05> it's quite possible that the chip doesn't support WPA AP mode, actaully I think the chip doesn't support genuine AP mode at all, it's an emulated AP mode. So you would have to emulate WPA as well, which aiui is what WPA-supplicant is made fore
Jun 15 21:23:27 <DocScrutinizer05> -e
Jun 15 21:25:37 <DocScrutinizer05> iirc it emulates infrastructure AP by exploiting AdHoc mode which by design can't use WPA. At least that's my take on it
Jun 15 21:27:35 <DocScrutinizer05> so when the chip doesn't support WPA but WEP for AdHoc, it's what the emulated AP mode will support too, as long as you don't emulate WPA AP in software, which should be perfectly possible
Jun 15 21:36:14 <ShadowJK> it's ad-hoc mode, not ap mode
Jun 15 21:36:22 <ShadowJK> recent androids don't supprot ad-hoc btw
Jun 15 21:43:38 <DocScrutinizer05> that's what I said. AdHoc mode simmulating AP mode
Jun 15 21:44:07 <Gh0sty> ah yes
Jun 15 21:44:15 <Gh0sty> indeed that was one of the problems
Jun 15 21:44:29 <Gh0sty> that it's doing ad-hoc so no android device can connect :/
Jun 15 21:44:30 <DocScrutinizer05> and AdHoc only can do WEP, by principle of operation I guess
Jun 15 21:44:37 <Gh0sty> think it's the same with my laptop
Jun 15 21:44:50 <Gh0sty> for infrastructure mode the hardware needs to be better? :/ 
Jun 15 21:45:07 <DocScrutinizer05> not better, just different
Jun 15 21:45:13 <Gh0sty> well yes :p
Jun 15 21:45:22 <DocScrutinizer05> actually probably only the firmware needs to be different
Jun 15 21:45:43 <DocScrutinizer05> the firmware we got doesn't support AP
Jun 15 21:45:59 <Gh0sty> ah the firmware is not open from that part of the phone? :/
Jun 15 21:46:01 <DocScrutinizer05> but it supports injection :-)
Jun 15 21:46:27 <DocScrutinizer05> the firmware of the chip, not the linux system of the phone
Jun 15 21:47:48 <DocScrutinizer05> aiui injection is a sort of "raw" mode that allows you to send whatever you like. So it *can* do proper AP mode when you write (or find) a FOSS software that does the encoding and protocol stack
Jun 15 21:48:09 <honestly> probably not too hard, just a lot of work.
Jun 15 21:48:39 <DocScrutinizer05> cave: this will keep the APE CPU busy and eat your battery like candy
Jun 15 21:48:52 <Gh0sty> so in other words
Jun 15 21:48:59 <honestly> om nom electrons.
Jun 15 21:49:02 <Gh0sty> I will just buy one of those huawei toys :p
Jun 15 21:49:33 <Gh0sty> if it was easy enough the n900 could still have a purpose but now ... :)
Jun 15 21:49:45 <Gh0sty> not sure, still sits in the drawer :/
Jun 15 21:50:03 <honestly> send it to me for when mine croaks >_>
Jun 15 21:50:26 <Gh0sty> well the led controller of mine is acting funky
Jun 15 21:50:37 <Gh0sty> one of the reasons why i bought myself an nexus 5
Jun 15 21:50:46 <Gh0sty> not too happy with that one either but ... 
Jun 15 21:50:46 <DocScrutinizer05> o.O
Jun 15 21:50:50 <DocScrutinizer05> funny?
Jun 15 21:50:51 <javispedro> hm, softapd to the request?
Jun 15 21:51:04 <javispedro> *to the rescue :)
Jun 15 21:51:18 <DocScrutinizer05> javispedro: \o/
Jun 15 21:51:24 <DocScrutinizer05> hi!!!
Jun 15 21:51:59 <javispedro> hi!
Jun 15 21:59:30 <DocScrutinizer05> javispedro: how about coding competition?
Jun 15 21:59:56 <DocScrutinizer05> or about joining the FPTF team?
Jun 15 22:01:25 <DocScrutinizer05> I heard rumors that some weirdo wants to promise a prize of 1000EUR for a fremantle based on open source only and capable to forward cmtspeech audio to an USB attached audiocard instead of modem
Jun 15 22:01:31 <DocScrutinizer05> ;-)
Jun 15 22:02:14 <javispedro> ah well :)
Jun 15 22:02:32 <javispedro> if I had the time a systemd-free distro is on the top of my list for now :)
Jun 15 22:02:35 <DocScrutinizer05> further conditions allegedly are: binary compatibility to maemo-extras packages, semi-decent power management
Jun 15 22:03:58 <DocScrutinizer05> aiui http://www.omappedia.com/wiki/Maemo_Getting_Started is already halfway tehre, maybe even further
Jun 15 22:04:46 <DocScrutinizer05> I think your understanding of how audio/mixer on N900 works would help a lot on achieving the cmtspeech goal
Jun 15 22:07:21 <DocScrutinizer05> (sidenote: it's allowed to import closed blobs from an existing fremantle installation, so the developer has such a system at hand. This wouldn't disqualify the solution)
Jun 15 22:07:54 <DocScrutinizer05> above rule doesn't apply for kernel
Jun 15 22:08:14 <javispedro> as of lately I have more time and I'm more on this channel, so feel free to ask
Jun 15 22:08:28 <DocScrutinizer05> well, obviously kernel isn't a closed blob on fremantle, so nevermind
Jun 15 22:09:19 <DocScrutinizer05> teams are allowed, the price will get shared on discretion of those who decide, and might even get upgraded
Jun 15 22:09:32 <DocScrutinizer05> prize :-S
Jun 15 22:11:36 <DocScrutinizer05> It occurred to me, a possible approach for audio (last resort) would be to treat the USB audio card as a SIP telephony account and just convert RTP to plain S16LE datastream to feed to the card
Jun 15 22:12:14 <honestly> ugh, this is getting ridiculous
Jun 15 22:12:18 <DocScrutinizer05> but of course a genuine modem ("phone") account for the voicecalls would be preferred
Jun 15 22:12:38 <DocScrutinizer05> honestly: honestly?
Jun 15 22:12:44 <honestly> honestly
Jun 15 22:12:52 <honestly> my n900 just won't charge properly anymore
Jun 15 22:12:53 <DocScrutinizer05> elaborate!
Jun 15 22:12:55 <nox-> doesnt that recorder app already record calls too?
Jun 15 22:13:04 <DocScrutinizer05> nox-: yep
Jun 15 22:13:16 <DocScrutinizer05> nox-: and javispedro understands how it does
Jun 15 22:13:25 <nox-> ah
Jun 15 22:13:25 <DocScrutinizer05> I don't
Jun 15 22:13:30 <honestly> there must be a loose contact somewhere
Jun 15 22:13:51 <DocScrutinizer05> honestly: check springs in battery
Jun 15 22:14:09 <honestly> huh.
Jun 15 22:14:12 <honestly> okay.
Jun 15 22:14:16 <DocScrutinizer05> you can gently bend them to "close"
Jun 15 22:14:18 <honestly> that's a new one
Jun 15 22:14:48 <DocScrutinizer05> using a needle or somesuch tool
Jun 15 22:15:03 <honestly> seems like the battery is sitting in there very solidly
Jun 15 22:15:14 <DocScrutinizer05> sure, it does
Jun 15 22:15:26 <DocScrutinizer05> not the springs in phone holding the battery
Jun 15 22:15:39 <honestly> oh, I see
Jun 15 22:15:44 <DocScrutinizer05> the tiny golden critters in the battery's slots
Jun 15 22:15:56 <honestly> right
Jun 15 22:16:22 <DocScrutinizer05> abusing them makes them gap open
Jun 15 22:16:50 <DocScrutinizer05> e.g by charging battery in external charger with thicker contact blades
Jun 15 22:17:42 <DocScrutinizer05> take care not to short anything when you stirr in there with a needle
Jun 15 22:17:42 <honestly> I haven't done any abusing like that
Jun 15 22:17:48 <honestly> yeah.
Jun 15 22:18:14 <DocScrutinizer05> visual control is OK. Lemme find a photo of a proper contact spring
Jun 15 22:18:30 <honestly> oh, the phone had turned off due to low battery.
Jun 15 22:18:39 <honestly> is it possible that it stayed in low-current emergency charging mode?
Jun 15 22:24:04 <DocScrutinizer05> http://maemo.cloud-7.de/share-service/20120504_002.jpg
Jun 15 22:24:12 <DocScrutinizer05> dang, the first one (almost)
Jun 15 22:24:23 <DocScrutinizer05> that took a while, bottom up
Jun 15 22:25:06 <honestly> yeah, my battery contacts look like that.
Jun 15 22:25:36 <honestly> let me see if I can't take a better picture :P
Jun 15 22:27:09 <DocScrutinizer05> o/ afk
**** ENDING LOGGING AT Mon Jun 16 02:59:59 2014