fzabkar wrote:
I would measure the voltages on the donor and determine which are the Vcc and Vccq supplies for the NANDs, then check whether these are shorted on the patient.
I see a shorted bypass capacitor at the corner of one of the NANDs, but I don't know it's function.
Here is information I found to help me with this, please make any corrections if you feel it is incorrect.
Quote:
The correct voltage levels for VCC and VCCQ on the Samsung EVO 860 SSD are:
- VCC: 5V +/- 5%
- VCCQ: 1.8V or 3.3V +/- 5%
If you do not have access to datasheets or online resources to help you trace the VCC and VCCQ lines, you will need to manually trace the lines on the board. Here are some steps you can follow:
1. Use a multimeter to measure the resistance between VCC and VCCQ and ground. The resistance should be very high or infinite. If the resistance is low or zero, there may be a short on the VCC or VCCQ lines.
2. Use a magnifying glass or microscope to inspect the board for any visible signs of damage or short circuits. Look for any components that may have been damaged or have visible signs of a short circuit, such as burn marks or discoloration.
3. Follow the traces on the board to locate any components that are connected to the VCC and VCCQ lines. Look for any components that may have been damaged or have visible signs of a short circuit.
4. Use a multimeter to measure the resistance between the VCC and VCCQ lines and the components that are connected to them. If the resistance is low or zero, the component may be shorted and may need to be replaced.
5. If there are no visible signs of damage or short circuits, use a continuity tester or multimeter to trace the VCC and VCCQ lines on the board. Start from the PMIC and follow the traces to locate any components that are connected to these lines.
Voltage injection with thermal camera onto these lines resulted in the ARM chip heating up directly in the center with no other additional heat spotted (0.8V used). Removing the shorted component you mentioned near NAND was unsuccessful, but it is on the same line as PMIC (unsure if vcc or vccq).
Lardman wrote:
fzabkar wrote:
To me, this SSD is not recoverable.
The op has multiple donors I'd pull the nand and transplant (if I remember correctly they're bga -316) . The level and location of shorts would also be consistent with technician damage under the bgas and there's always the hope any electrical problems hit the sdram or controller, rather than the nand.
It would be really helpful if schematics or even boardviews fell of a truck in China.
I believe that I have the skills to transplant this chip but have no access to software or tools which could be used to recover the data. I cannot do firmware adaptation as I cannot currently access the firmware of this chip (unless any of you know how) so transplanting to a donor may be useless as I do not have access to specialized software. I am willing to fork over money to purchase software if this seems like a viable option (as I've been quoted $1250 for a transplant recovery, and deepspar for example is under this price although I'm not sure if it capable of such recovery). Maybe a shop with pc3k could assist me after a successful transplant, but if anyone has any alternatives or ideas for recovery than I am all ears. If this recovery goes successful than I am willing to shave off the top layer of this pcb to show the boardview for all future repairs.
I am not giving up on this repair.