hello > Usually it's best to heat up the PCB while pulling the chip away from > it. You will need about 260°C to make the solder melt, but the chip > should not be heated above that temperature for more than 10 seconds. So > you need something that heats up the PCB really quickly. One usually > uses some kind of hot air blowers for that, but if you want to take the > risk, you can try to do it with some kind of gas burner. >
>>> resoldering : its possible to solder each ball to a wire >>> >>> > Well, if you have appropriate soldering equipment, it's possible to do that. > i have acess to all that is needed for such operations hot air station, smd solder iron with magnification, etc... > Connect A0-A18 and DQ0-DQ15 to the AVR, CE#, OE# and Vss to GND, WE# and > Vdd to 1,8V. Depending of the kind of outputs the AVR has, you may also > need to use resistors for the Axx connections, or you may need to > connect Axx to 1,8V through some resistors. A datasheet of the AVR would > clarify this. > exactly then a very simple loop program who reads out the content and sends it to the uart (serial line), a pc logs then the result. > The data we will get out of it will probably be some kind of ARM > executable code, which can probably be disassembled by IDA Pro etc. > In case they should have taken precautions against that, the content of > that chip may also be encrypted with the decryption code and key being > in a small internal ROM in the ARM, supposed it has one. Then we would > need to "sniff" the contents of the RAM at runtime, or somehow record > the data transmitted by the ARM to the RAM at boot time. > it is very hard to read out the ram, but remains possible. a fpga is needed to simulate or to acquire the ram. if someone sends me an old nano 2G, i can try to read out the flash. christoph _______________________________________________ Linux4nano-dev mailing list [email protected] https://mail.gna.org/listinfo/linux4nano-dev http://www.linux4nano.org
