I'm planning on trying to figure out what the pads below the battery actually are (got Open Logic Sniffer in the mail a while back), at least for those that output something.

Current problem is getting probe wires on those pads without doing anything destructive (like soldering, besides it's cramped in there), current plan is to mill pieces of polycarbonate to fit the slots and drill wire sized holes, CNC mill would be nice but I don't have one...).

I'm hoping there is I2C or SPI on the pads (which would enable us to make plans for a "bolt-on" accessory, we could even design a "shields" [ala Arduino] system for N900 )

I'll update the wiki when I get around to doing this (which might take a while since I'm at aKademy mon-wed and on a training camp for a week starting on friday...)

Made a sorta-adapter, unfortunately as later comments show I had a brainfart and thus got nothing out, need to try again later (but not today)

Edit: Did some schematic reading, as result updated the debug ports documentation.

Just a quick update: My adapter is crap (poor connection and no connection at all on some pads) and my lab supply is puny so I can't power the device without battery. So I need to either find very small springs (that could be used to make good connections on a number of pads from a new adapter or some other nondestructive way of getting wires on the pads, or I need to get a more powerfull supply (working on it but it'll take a while)

OTOH: at http://hacklab.fi/ I have access to a 48ch (actually 96ch but we have only one bag of probes...) logic analyzer so once I figure out a good way to get the probes to the pads it's possible to look at all of them at the same time...

Bit the bullet and got a new 2x5A (can be used in parallel for 10A) lab supply (even got a rather good "special price - only for you" -discount since I'm a regular at the electronics store , still it cost over 200EUR )

But I have a milling job to do for a fellow hacklabber that needs to be done by tomorrow (we need it at ASM2010, which is where I'll be spending a lot of the weekend too...) so this will still take a while.

Status update: figured out the need for 47k resistor between BSI and GND pins and now can boot my device with a bench supply, with a multimeter and a small handheld scope located some promising pads for further study and now need to figure a good way to get the logic analyzer probes to stay on them.

IIRC, BL-5J has a 100K resistor.