USB drive - no power

[maxwell1883]

Hi!

I want to get into data recovery and I have a faulty USB drive here that isn´t recognized by any computer.

It´s a no name drive with i guess 2 GB.
Controller: IT112E48A
NAND: H61AG080I

When the drive is plugged in, nothing happens (also no LED light on pcb). However, the controller gets hot and reaches a temperature of about 38 °C.
I didn´t find any unbond legs on the controller or nand. Also the other components look fine.
I have checked the caps for shorts.

On C3 i am measuring 3.3 V and on C2 1.75 V.
On the NAND i am measuring 3.3 V.
I didn´t find any schematics for the controller so I don´t know, which signal belongs to which pin.

My next steps would be:
- Measure the oscillation of the quartz with an oscilloscope
- Resolder all components
- Remove the NAND and see if something changes

Can someone give me a hint where to look next or does somebody know such behavior?
Can somebody share a schematic of the NAND and/or controller?
Is there some generic schematic for these types of USB drives available? I guess they are all quite the same except the used controller and NAND.

Thanks!

Attachments

NAND

[fzabkar]

Most likely the firmware has panicked.

Your NAND flash, H61AG08OI, has an ID of 98D5943276.

https://rusolut.com/software-update-23-october-2015/

This is associated with Toshiba 2GB MLC flash.

https://nand.gq/#/searchId?id=98D5943276

[HaQue]

Th Flash is likely refurbished Toshiba. Likely only solution is to read nand and use data recovery software to put the image back together. Building a homebrew reader might be tricky, and I would suggest contacting a data recovery lab, it should be straightforward as long as data is on nand and no other issues. IMHO, without DR software, and reader, you will have a year of learning to knock up any kind of home solution, if at all. Unless you strike it lucky and find a fixable issue with PCB, but as voltages dont look to crazy, unlikely

[maxwell1883]

Thanks for the info!

What options do I have to repair the firmware? Are there any options without spending thousands of dollars for the start?

Edit: What options do I have to read the nand? Is for example the RT809H Reader an option?

[HaQue]

I doubt that programmer would support it. The MLC flash is a bit different to the few NAND chips that are not SPI. You need either Flash-extractor, Rusolut or ACE Labs.

The controller is probably shot, so doubt you could write firmware. There are some MPtools here but you want to be VERY careful with anything from this site. Make sure you bring your Malware A-Game too!

https://www.usbdev.ru/articles/a_ite/ut165mfpt-repair/

[Arch Stanton]

I doubt that programmer would support it. The MLC flash is a bit different to the few NAND chips that are not SPI. You need either Flash-extractor, Rusolut or ACE Labs.

The controller is probably shot, so doubt you could write firmware. There are some MPtools here but you want to be VERY careful with anything from this site. Make sure you bring your Malware A-Game too!

https://www.usbdev.ru/articles/a_ite/ut165mfpt-repair/

But using MPtool means give up on data, right?

[Arch Stanton]

Most likely the firmware has panicked.

But then you'd expect some sings of life, in Disk Management for example 'no media' or some size in MB range?

ChipGenius may help tell if controller is detected or controller + NAND. If controller + NAND detected + No media / MB size then most likely NAND itself is the issue and chip-off would be only way.

[maxwell1883]

I doubt that programmer would support it. The MLC flash is a bit different to the few NAND chips that are not SPI. You need either Flash-extractor, Rusolut or ACE Labs.

The controller is probably shot, so doubt you could write firmware. There are some MPtools here but you want to be VERY careful with anything from this site. Make sure you bring your Malware A-Game too!

https://www.usbdev.ru/articles/a_ite/ut165mfpt-repair/

Thanks, but i fear that MPTools only works if I get a connection to the drive. Currently it is sort if dead, i dont get any signs of recognition on the computer. It is as if nothing is connected to the usb connector.

I tried to get Flash extractor, but the support has ended and the needed hardware is sold out. And as far as i have seen, ACE Labs is similar expensive as Rusolut right?

For ChipGenius i also need some sort of recognized drive, right?

[HaQue]

you have it pretty right, yes. I would suggest finding someone close to read the chip for you. If your country just had 2 more letters in it, I would do it for you. could probably recover the data as well. A data lab near you should be able to do it, and cost for this type *shouldnt* be all that expensive, way cheaper than getting tools, and much quicker.

[Arch Stanton]

i dont get any signs of recognition on the computer.

Then I assume controller is dead or no connection to it. If you scrape these traces do you get continuity between x's? Can not really judge how this looks from the image. Or just use corresponding pins on controller

[maxwell1883]

i dont get any signs of recognition on the computer.

Then I assume controller is dead or no connection to it. If you scrape these traces do you get continuity between x's? Can not really judge how this looks from the image. Or just use corresponding pins on controller

continuity.jpg

Yes, I have continuity between the USB connector contacts and the corresponding controller pins.
I have ordered some noname drives that look the same, with the same size. Let´s see if I have luck, otherwise my only chance is to get the NAND read out by someone with the proper tools.

[maxwell1883]

Hi!

I have an update.
I ordered the RT809H programmer and it looks like I was able to read the NAND.
Log:

Code:

011:  The currently selected:NAND_AUTO
012:  You have selected the GENEREAL MODE, it will ignore the ID verification and using compatible with most manufacturer's General algorithm.
013:  Download device programming algorithm......
014:  Algo update OK
015:  >------------------------------------OK---------------------------------------<
016:  Chip pins contact is detected OK.
017:  TotalPageNum: 0x100,PageNumInBlock: 64,PageSize: 2112
018:  TOSHIBA, ID: 0x98D5943276540A00, 3.3V, CE:1, 8bit, Col: 2, Row: 3, MLC.
019:  PageSize:4096B, OobSize:64B, BlockSize:512KB, ChipSize:2080MB.
020:  I:\B\Vermesser\NAND_AUTO_5211\NAND_AUTO_5211
021:  Start reading chip......
022:  Buffer data checksum: 16bits_0xF0DE £¬32bits_0xDA61F0DE :
023:  Read successful£¬Elapsed time£º352.4 seconds¡£
024:  Auto verifying...
025:  92 bytes verification are inconsistent.
026:  exceed 1 bit/512B Check for inconsistent pages£º1 pages
027:  exceed 4 bit/512B Check for inconsistent pages£º0 pages
028:  exceed 24 bit/512B Check for inconsistent pages£º0 pages
029:  Verification successful£¬Elapsed time£º352.4 seconds¡£
030:  Elapsed time: 704.7 seconds£¬average speed of 6189800 bytes/sec.
031:  >------------------------------------OK---------------------------------------<

How can I proceed now? Which software can be used to get valid data from the bin file?

[Arch Stanton]

https://adreca.net/NAND-Flash-Data-Reco ... okbook.pdf

You're at page 37 or thereabouts.

[maxwell1883]

Thanks for this treasure. One question, is flash extractor also capable of XORing, page/block management etc.?
It looks like the software is still available for a couple of hundret $.

[HaQue]

Thanks for this treasure. One question, is flash extractor also capable of XORing, page/block management etc.?
It looks like the software is still available for a couple of hundred $.

If the scheme used by manufacturers has been seen before, or it is fairly common, then yes it might support it. Flash recovery uses words like: possibly, might, should, could and maybe. and not : will, or definitely.

You do need a license for the software. but once you have a decent dump of the chip, it is easy to share to someone with this to recover for you.

[maxwell1883]

Thanks, was able to recover it with PC3000 Flash