Page 1 of 1

SM2246 trouble

Posted: 16.03.18, 16:53
by northwind
Hi guys

Got this Sandisk SSD Plus with SM2246 controller inside.
It stays in BSY.

I follow procedure and I get into utility, init the drive and then start to create translator.
Capture1.JPG
Capture1.JPG (5.28 KiB) Viewed 7684 times


During the procedure, I get errors after a while,
Capture2.JPG
Capture2.JPG (5.45 KiB) Viewed 7684 times


and it will finish
Capture3.JPG
Capture3.JPG (175.1 KiB) Viewed 7684 times


However, when I create task in DE, it will read zeroes :(
Capture6.JPG
Capture6.JPG (7.03 KiB) Viewed 7684 times

Capture4.JPG
Capture4.JPG (348.07 KiB) Viewed 7684 times


This goes on until the end of the drive.

Any insight please?

Re: SM2246 trouble

Posted: 17.03.18, 14:29
by northwind
An update.

I did the procedure several times, always with same errors during the translator creation.
Once, DE behaved differently and this time it wasn't reading zeroes, but actual data, however the errors are more than 60% (which obviously will lead to 0 good files).

Capture9.JPG
Capture9.JPG (8.66 KiB) Viewed 7674 times


Is there any hope?

Re: SM2246 trouble

Posted: 19.03.18, 08:38
by Roman_TS
Hello!

Unfortunately SM2246 are not working with TLC chips so they can't use ReadRetry function for self-error correction. That's why in some cases might be the situation as yours - a huge number of bad sectors in some areas.

Anyway, after reading the image, try to create a submap with incorrect sectors (yellow, red, black) and reread them with several retries and different reading settings in DE.

Also, during rereading you can try to freeze the NAND chips - temperature influence is also very important for that.

Re: SM2246 trouble

Posted: 20.03.18, 11:31
by northwind
Hi Roman, I hope you're fine!

I tried your suggestions, temperature control did not provide any different results.

When I create submap with problem sectors, it looks like it's reading a few sectors more each time. However, the result is still very bad, not more than 25-30% of sectors are being read (in some areas, close to 1-2%!).
I guess I'm gonna read what I can and then analyze the image and see if I can get anything out of this.

Thank you for the insight!