usb key not recognized

[opethfan]

hi
i must recover files from usb key not recognized at all.
i show you some pictures.

thank you

Attachments

[HaQue]

A data recovery pro in your area should be able to read the nand memory by using another device with a regular tsop48 nand chip as a reference and discovering the signals at the controller. Willing to take it on if you want to send it to AU at a forum discount, or maybe someone closer wants to offer?

[data-medics]

I'm not doubting your expertise HaQue, I don't even do flash recoveries myself and I know you're good at them, but how can you even tell it's a TSOP48 under that blob of black goo?

[jermy]

what else can it be ?

[HaQue]

I know it isn't a tsop... the raw nand underneath still must have the same signals connecting to controller. The controller has standard pins for each model, look on one device where D0-D7 go to controller and ALE,CE etc, and it will be the same on this device. we use the controller to reference the signals, not the chip in this case.

As Jeremy is suggesting, underneath the "clothes" they are all the same

[arvika]

It is not TSOP48, it is just blob You can connect to nand from controller side as HaQue suggest.

[jeremyb]

It's silicon motion so finding the datasheet/pin-out for the controller to locate the signals should be easy.. be gentle soldering, those chinese PCB's are shit.

Willing to take it on if you want to send it to AU at a forum discount

Do you have so much time on your hands that you're willing to offer a "forum discount" now? If so I have a few monolithic's I'd like to send your way that need pin-outs, I'm MORE than happy to get a forum discount

Caution: Some may have buried traces

under that blob of black goo?

Has anyone tried removing that goo? It's not ceramic like with monolithics, it's some type of glue, it gets tacky when I heat it up.

[HaQue]

I think there are a few different types as I had one looking similar to this and hitting it with SMD Rework hot air did nothing except smell bad. I think it would be too easy to damage the raw NAND and tiny wire bonds anyway. laser and probe may be an option but time/cost/reward ratio sucks....

[opethfan]

i think that the glue is epoxy resin...
what do you think about reading the nand chip from the controller side?
is there a tutorial to do that?

[HaQue]

a tutorial? Muahhh!!! haha!!!! haha!!.. oh wait, your serious...

seriously though..

do you have a way to read a regular NAND chip? and do you have data recovery tools, Rusolut, SC, PC3K or SD?

if yes: find a datasheet or pinout diagram of the controller, wire it up. recover.

if not, find some one to do it for you. I should say "attempt" as these can be REALLY crappy quality, affecting recovery.

[jeremyb]

is there a tutorial to do that?

I'd start here:
http://www.alldatasheet.com/datasheet-p ... 08R0B.html

It gives you the basic signals you'll need to know, from there either bitbang or use an FPGA

[fzabkar]

What is the part number of the controller?

[opethfan]

sm3257en
s1k137.1

[HaQue]

Probably full number is SM3257EN Q AA which means maybe a WL chip, and bad cols, in software, makes it even harder for DIY.

[fzabkar]

Here is an application circuit:

[HaQue]

1. you need a reader.
2. you need to identify signals at the controller
3. you should remove controller, solder thin wires from controller pads to reader
4. if you can ID the NAND, read it.
5. If chip is WL, you need to deal with bad columns
6. remove whitening and mix and whatever else needs to be done to fix and create image
7. recover files using DR software

IMHO, steps 1 4,5,6 are not going to be a cakewalk. If you don't have flash experience then you probably won't even get the chip read properly.

this case should be outsourced.

[jermy]

If you don't have flash experience then you probably won't even get the chip read properly.

this case should be outsourced.

+1