Recently, my Thinkpad T530 suddenly broke and only flashed the LEDs at power on and then did nothing. I stripped all unnecessary hardware but the behavior did not change. Also, pressing certain key combinations during power on which shall clear remaining charges or BIOS settings did not help.

As a last measure before giving up, I wanted to check if replacing the BIOS/UEFI with coreboot would give any hint on what is going on as the original firmware did not even show an error message or play a beep code. While coreboot did not give me any information on what was wrong, the laptop just started to boot again after I flashed coreboot and setup the grub bootloader to boot in BIOS mode instead of UEFI. Eventually, I guess either the BIOS settings or the BIOS code itself got corrupted in the memory and flashing coreboot and resetting the BIOS config resolved the issue.

While the instructions to flash coreboot are mostly similar to other Thinkpads from the same time, one specialty of this particular laptop is the placement of the two EEPROM chips that contain, among others, the BIOS/UEFI code. The smaller 4 MB large chip is below the chassis and the large 8 MB chip is close to the chassis. So, the small one cannot be flashed with one of those common clips and an external programmer and the large one was too close to chassis (or my clip was to large). Luckily, as stated in the coreboot documents, most pins of the two chips are connected. Only the chip select (CS) pins can be controlled individually. Hence, in my case a programmer can be connected with the large chip using individual wires and only the CS pin must be connected directly to the chip that shall be flashed. As I wanted to avoid soldering in such a small space, i planned to use test clips to connect the programmer and chips. However, the test clips were too large to connect them all to one chip with such a small pitch. Hence, I distributed the test clips among both chips as can be seen in the following image:

t530 eeprom.jpg

To make things worse, only connecting the CS pin of the to-be-flashed chip with the programmer was not enough. The CS pin of the other chip had to be connected with Vdd so it will not interfere with the other chip's communication. Luckily, the pins of the two chips are also connected to the unequipped pads labeled with "CN100". As you can also see in the image above, the Vdd and CS pin of the 8 MB chip were next to each other on the left of the side towards the display and I was able to simply connect them using a small screwdriver. The two pads for the 4 MB chip are just on the other side facing away from the display and can be connected in the same way if needed.

While this setup is not ideal - I had to start the dump and flash process several times until it successfully finished - I was able to avoid soldering on small space and to disassemble the whole laptop.