GIGABYTE SSD 1TB PS3111 CONTROLLER

[speakerbox]

Hello,
I have this drive that showed 20MB
Soldered a jumper wire for safe mode.
Uploaded recommended loader and created translator. I have access to data.
I need a very big file (aprox 90GB) and opened a DE task to try to copy the list of ranges.
However I get slow reading in most parts (120KBps) and 1.5MBps when it goes fast.
I get bunches of 32 black blocks from time to time ("page marker checking error" (invalid data)). This is what slows a lot the process.
Could this be the loader used?

Also have bad MFT entries in the NTFS partition
I noticed when trying to save the file that it would only copy 1GB stating wrong size

Could memory chips be degraded so fast? Drive is from Jan 2021

[northwind]

Could memory chips be degraded so fast? Drive is from Jan 2021

That's one and a half year. I'd say it exceeded the usual life span
OK, exaggerating, but we've seen chips degrade in a blink of an eye, so...

I think the translator you've built is probably fine.
It's the chips that are in bad condition.
Try heating/freezing, if you're VERY lucky, it might help you.

However, I have to say that getting 90GB intact from a drive in this condition... I'd say chances are close to zero.

[arvika]

Degradation of chips. Try reread many times.

[speakerbox]

Could memory chips be degraded so fast? Drive is from Jan 2021

That's one and a half year. I'd say it exceeded the usual life span
OK, exaggerating, but we've seen chips degrade in a blink of an eye, so...

I think the translator you've built is probably fine.
It's the chips that are in bad condition.
Try heating/freezing, if you're VERY lucky, it might help you.

However, I have to say that getting 90GB intact from a drive in this condition... I'd say chances are close to zero.

Degradation of chips. Try reread many times.

Thank you both for your input!! I will let you know the outcome

[speakerbox]

Well, situation didnt change either using freeze spray or heating the chips up to 160 degrees celsius
As this is a *.vsddsqlzip file, we can assume it is deeply f*cked

[fzabkar]

One more thing you could try would be to adjust the NAND Vcc supply in-circuit. Run it through the range from 2.7V to 3.6V.

[speakerbox]

One more thing you could try would be to adjust the NAND Vcc supply in-circuit. Run it through the range from 2.7V to 3.6V.

where is this feature? I cannot find it

[fzabkar]

One more thing you could try would be to adjust the NAND Vcc supply in-circuit. Run it through the range from 2.7V to 3.6V.

where is this feature? I cannot find it

That's because it was a suggestion I put forward at the HDD Oracle. The idea is that you intercept the PCB's 3.3V supply and inject your own variable supply.

[speakerbox]

One more thing you could try would be to adjust the NAND Vcc supply in-circuit. Run it through the range from 2.7V to 3.6V.

where is this feature? I cannot find it

That's because it was a suggestion I put forward at the HDD Oracle. The idea is that you intercept the PCB's 3.3V supply and inject your own variable supply.

I assume I would have to cut the trace for that and then inject V

[speakerbox]

Update so far:
I retrieved the 97GB file but has 0.24% corrupted data, so this one sucked. As it was a compressed file database it couldnt be loaded.
Tried with freeze spray, heating etc... no luck.
Garbage chips.

Anyway there where a few negligences here...
- Server did not have the UPS turned on, and the lights went off
- Corrupted firmware in SSD
- They make the BACKUP of the system from that SSD to THAT SELF SSD in another folder. How stupid is that?
- They make 100GB backups every few days and the memory chips where degraded. Only a year and a half from new.

[fzabkar]

One more thing you could try would be to adjust the NAND Vcc supply in-circuit. Run it through the range from 2.7V to 3.6V.

where is this feature? I cannot find it

That's because it was a suggestion I put forward at the HDD Oracle. The idea is that you intercept the PCB's 3.3V supply and inject your own variable supply.

I assume I would have to cut the trace for that and then inject V

Remove the inductor and inject the voltage into the load side of the inductor.

[fzabkar]

Would it be possible to upload a resource dump for this SSD? It would be interesting to see what the defect lists and SMART look like.