HDD GURU FORUMS
http://forum.hddguru.com/

Sandisk 960 Ultra II SSD
http://forum.hddguru.com/viewtopic.php?f=10&t=35472
Page 1 of 1

Author:  ocooool [ June 15th, 2017, 19:31 ]
Post subject:  Sandisk 960 Ultra II SSD

Dear
I have the following case, a Sandisk 960 Ultra II SSD which is damaged, is not detected by Bios
I have the intention to buy others and exchange the meori, someone has experience in that ??
PC3000 recommend me for recovery but the high cost leaves me out of possibilities.

Ignore the damaged component

Sorry my english is of Translator.

Image

Author:  fzabkar [ June 16th, 2017, 0:05 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

ocooool wrote:
Ignore the damaged component

Why?

Author:  ocooool [ June 16th, 2017, 8:33 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

Because it is not the exact photo of the disk, if it is the same model and same components, but the original or where the data died after a suspension already detected nothing.
I want to buy another identical model and just replace the memories, my question is if someone has performed that procedure and if to effective result.

Author:  HaQue [ June 16th, 2017, 10:25 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

What is the marvell chip number?

Swapping won't work.

at the bare minimum, if it were possible, you would probably need to match exactly:

firmware version / revision
controller version / revision
nand chip part number
board revision

- hope that whatever bad blocks / bad columns are handled correctly
-hope no tables, config data etc are stored on the controller
-hope SSD does not instantly write anything to chips for whatever reason if things dont look right to its initialisation.

If I wanted data, the only way I would try this is to dump each nand chip first during the process while they are off anyway.

Marvell controllers are not dumb algorithmic controllers like Phison / SMI / Alcor. They are more like a microprocessor. There are alot of buzzwords used when they talk about how the controller stores data economically and safely, which translates to "hard to recover"

Author:  ocooool [ June 16th, 2017, 11:39 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

It is perfectly clear to me that a change of brute force will not work.

And extracting the raw memories via a NAND reader, I figure I'll have to rebuild a sort of virtual RAID from the SSD. ?

In general my questions may be somewhat awkward recently research on the subject seems incredibly interesting

Can you recommend information to better understand the subject?


EJ: http://rusolut.com/


HaQue wrote:
What is the marvell chip number?

Swapping won't work.

at the bare minimum, if it were possible, you would probably need to match exactly:

firmware version / revision
controller version / revision
nand chip part number
board revision

- hope that whatever bad blocks / bad columns are handled correctly
-hope no tables, config data etc are stored on the controller
-hope SSD does not instantly write anything to chips for whatever reason if things dont look right to its initialisation.

If I wanted data, the only way I would try this is to dump each nand chip first during the process while they are off anyway.

Marvell controllers are not dumb algorithmic controllers like Phison / SMI / Alcor. They are more like a microprocessor. There are alot of buzzwords used when they talk about how the controller stores data economically and safely, which translates to "hard to recover"

Author:  fzabkar [ June 16th, 2017, 15:28 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

ocooool wrote:
Because it is not the exact photo of the disk, if it is the same model and same components, but the original or where the data died after a suspension already detected nothing.
I want to buy another identical model and just replace the memories, my question is if someone has performed that procedure and if to effective result.

Unfortunately the translation is poor, but IIUC you have done no basic testing, yet you now plan to reduce the SSD to a pile of bits.

If you know how to use a multimeter, and if you can upload detailed photos of both sides of your PCB, I could mark the voltage test points for you.

Author:  pcimage [ June 16th, 2017, 15:33 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

fzabkar wrote:
ocooool wrote:
Because it is not the exact photo of the disk, if it is the same model and same components, but the original or where the data died after a suspension already detected nothing.
I want to buy another identical model and just replace the memories, my question is if someone has performed that procedure and if to effective result.

Unfortunately the translation is poor, but IIUC you have done no basic testing, yet you now plan to reduce the SSD to a pile of bits.

If you know how to use a multimeter, and if you can upload detailed photos of both sides of your PCB, I could mark the voltage test points for you.


Agree, don't take the chips off.

Author:  HaQue [ June 16th, 2017, 20:06 ]
Post subject:  Re: Sandisk 960 Ultra II SSD

fzabkar wrote:
ocooool wrote:
Because it is not the exact photo of the disk, if it is the same model and same components, but the original or where the data died after a suspension already detected nothing.
I want to buy another identical model and just replace the memories, my question is if someone has performed that procedure and if to effective result.

Unfortunately the translation is poor, but IIUC you have done no basic testing, yet you now plan to reduce the SSD to a pile of bits.

If you know how to use a multimeter, and if you can upload detailed photos of both sides of your PCB, I could mark the voltage test points for you.


agree would try this way first - but OP didnt answer my first question and if recent other threads are anything to go by, it is going to be another dozen posts before anything substantial comes out. I was answering based on the OP's question.

Page 1 of 1 All times are UTC - 5 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/