SATAFIRM S11 or How to bring an SSD to life on a Phison PS3111

Hello. I used method 2 on the cx400 fw sbfm61w5 disk, I uploaded sbfm61.5, now the disk is missing in my computer and in management everything was uploaded without errors, visible only through the flasher as ps3111 serm07.0

bioly2500 wrote:

Hello. I used method 2 on the cx400 fw sbfm61w5 disk, I uploaded sbfm61.5, now the disk is missing in my computer and in management everything was uploaded without errors, visible only through the flasher as ps3111 serm07.0

I had a similar experience, I uploaded the same thing a second time and it worked.

Unfortunately, after the second upload, the same thing happened to me

I think it`s time to admit that this disk is unsalvageable. If you do everything in accordance with the above and the effect is different than that of other successful users, maybe it`s a waste of time?

As I clearly noted, the success of such a "repair" is not likely to last forever. If a drive has failed once, it has a high chance of failing again. Of course, these 2-3% of drives could have a problem due to, for example, an error in the firmware - after upgrading to the latest (compatible) they should work, but 97-98% fail due to NAND degradation.

bioly2500 wrote:

Unfortunately, after the second upload, the same thing happened to me

You have one of the NAND cells damaged. Repair is absolutely unprofitable. In my case, all methods did not help, but after full recertification of the NAND, it rose for a moment, showing damage to one of the NAND chips.


Finally, my broken Cx400 with a new name was restored, but not with full capacity, but only 480 GB, because RDT showed that CE12 is FF, which means that one of the NANDs is more degraded than the other, or something is not in contact with the PCB. One day I may be tempted to reball the nand bones and we will check again how this dead thing will behave.
RDT has prepared a new list of defects and it does not indicate that the nands are degraded, each cell has a maximum of 1%, it is difficult to say how much it will work, I suspect that trying to write to files will destroy the nands in the blink of an eye. Regards.

My disk, strangely enough, has been working for several dozen hours and not even 200 starts. It crashed during the installation of Win10 - a blue prompt after restarting the computer (after everything was copied) the partition on my computer was visible but it could not be formatted. The warranty was not recognized because I supposedly sent them back the disk that was not the one I sold them - Of course, there is no invoice that supposedly shows the SN - which I didn`t even check.

bioly2500 wrote:

My disk, strangely enough, has been working for several dozen hours and not even 200 starts.

Mine was similar, not all 500 hours and a few starts, in total it was 4.5 months. As you can see in my example, it was the Phison controller that died, not the NAND, the NANDs are in perfect condition

bioly2500 wrote:

It crashed during the installation of Win10 - blue prompt after restarting the computer (after everything was copied)

Mine said enough, when I restarted the system it didn`t recover.

bioly2500 wrote:

The warranty was not recognized because I supposedly sent them back the disk that was not the one I sold them - Of course, there is no invoice that supposedly shows the SN - which I didn`t even check.

Mine was also theoretically under warranty, but I know they would also turn up their noses and not honor the warranty.

That`s what I thought, finally, while copying less than 20 GB of test data, it crashed and returned to ROM mode, as if it had turned off the power. Don`t waste your time on these SSDs.
Out of boredom, I reballed the nands, changing the order, and it turns out that it is not the nands that are causing the disks to disappear, but the Phison controller itself.
In my case, channel 1 is damaged because after resoldering the NAND, the problem does not transfer to the second theoretically functional channel, signaling a Cell problem. everything is working, but the problem remains on the same channel, i.e. 1, this time the problem is in Cell 8, previously there was a problem in Cell 12.
Reballing the controller itself didn`t help, the problem remains. So replacing PS3111 alone should restore the drive to full functionality. Most write that NAND is degraded, but they forget that the main controller can also be damaged.
In the first photo, there is visible damage in channel 1, cell 8 of the NAND memory, which after transferring to channel 0 is functional, while after transferring NAND, cell 12 shows damage, but again, transferring to channel 0 shows all targets visible with the same ID.
The second photo shows us NAND defects. As you can see, on cell 15 we have less than 1% damage to NAND blocks, so tell me where do you find NAND degradation?
btw, I insisted on it and the post is intended to show and perform an in-depth analysis whether NAND is actually the reason for the disappearance of these disks, or the Phison controller. As you can see in my example, it was the controller that died, not the NAND.

Recently I found in a drawer 4 broken GOODRAM CX300 disks which were waiting for better times 2x 120GB with F/W: SBFM21.1 and 2x 240GB with F/W: SBFM71.0 and SBFM71.1. All of them with SATAFIRM S11 error. As they were broken anyway I decided to experiment on them. The first way did not work on any of them. The second way works! I only made a mistake the first time because I used s11-flasher2-micron instead of s11-flasher2-toshiba.... I had imagined they were on toshiba cubes. As you can guess, I've messed up the drive completely. After taking it apart and making sure there is a micron logo on the cube, I shorted (painted with a pencil) 2 gold pads (there were no holes in my cube like a few posts earlier) and the drive was detected. This time it was possible to upload the correct FW. It was even possible to upgrade the software version from goodram to version 71.2 (the latest) but the drive name is "SATA SSD" instead of GOODRAM no but it works. Other disks with the correct fw version uploaded have the numbers and name preserved. The goodram software recognises them and updates the fw to the latest version 21.2 for 120GB and 71.2 for 240GB respectively. If it wasn't for my mistake at the beginning it would have been a complete success.
I haven't tested the drives thoroughly yet, I just test uploaded a few iso images (~50GB) to see if it lives. There was no error, but the write speed waved a lot and there were drops even to a few MB/s. I don't know if these drives were like this or if they are simply running out of juice, but they are fine for testing/service.

Hey I have this problem as well. My firmware says that it is SBFK61W1 and that firmware cannot be found in the firmware dump provided. I have provided a picture. I anyone can help me with this it would be much appreciated because I do not know how many more times I can watch a Russian man try to explain to me how to do this.

Hello
I have a Goodram cx400 gen2 1TB drive that got a shot when powered most likely from some kind of adapter. What should be the power supply from the converters? There are three, one dfn8 will give 3.3V for sure, there are two more (unfortunately I could not find anything about them) small sot23-5 marked 578 from resistor measurements I get 1.2V and 1.15V . It would seem odd that the nand TKBHG93AYV gets 3.3V and 1.2V and the Phison PS3111-S11-13 gets 3.3V and 1.15V.
Has anyone measured the power supply in the drive and remember how much it should be?

StynkS wrote:

What should the inverter supplies be?

.
The ones you have specified are rather good.
Can you put the drive into ROM mode by shorting the service pins?
Here, using the example of the SM2258H controller (Adata SU800) you have described how to check for possible damage in a SATA SSD.

Well that's what it comes out to from measuring the components but I was wondering if sometimes it should be one 1.8V on the NAND instead of two 1.2V ones.
There is no full documentation anywhere for TLC in civilian releases. Normally the older NANDs used to have a second voltage of 1.8V if the supply was 3.3V.
Do you have any papers for the PS3111-s11 controller itself, where the power supply is specified?
Normally hands are falling down, if you do not write off the measurements, someone will not tell you because he has on the table, no company is willing to put the information. Apart from the fact that sometimes no one admits to the circuit .

ps. The drive inverters were faulty so I gave the voltages from calculations after measuring the resistors on the compression. Hence my question if anyone has measured them in real life.

Greetings

According to "made in China: G5728BTO1U (code:578B) SOT23-5

In the gen.2 model on the ctrl Maxio MAS0902A needed: 5V, 1.2V and 3.3V
This is a typical budget model on Phison PS3111-S11-13 (1C/ 2-channel). 32 MB buffer is sufficient for basic operations.
Described with tests Link .
* or Link

Hello

ATTENTION!!!
I have completed the data recovery from a Goodram cx400 gen2 1TB drive. It had damaged all inverters after overvoltage. After plugging in external power supplies the drive started the data was copied .
Goodram cx400 Gen2 1TB drive on PS3111 -S11-13 memory TKBHG93AYV power supply 3.3V and 2x1.2V I set 1.16V.

regards
StynkS

Bravo you ! Congratulations. Thanks to this, we now know that it can be done.... The knowledge of tensions remains for posterity.

Hello everyone!

Would any of you who have managed to upload firmware and successfully get the drive back up and running be willing to repair my GoodRam CX400 gen 2 for a fee? I don't care about the data, and I could theoretically buy a new drive, but I'd rather try to fix it first - it's greener, and at the end of the day, a 500 GB SSD is still a 500 GB SSD .

The drive was bought 3 years ago (hence the warranty no longer applies) and has been lying unused, so I don't expect NAND wear. I recently plugged it into the PC, checked it was working fine, then did a complete wipe and GPT partition cleanup during which a power failure occurred. After booting up, the famous SATAFIRM S11 showed up. I think it's just the translator gone awry, hence a software fix should help .

I've tried to bring it back to life myself using both methods, but have been unsuccessful - programs from colleagues across the eastern border don't recognise the drive for me, and Firmware Updater communicates that I should connect it with AHCI. The snag is that I connect the drive directly to SATA just with AHCI and use the standard Windows driver, and despite this and many other attempts (different SATA sockets, different power supplies, different versions of programs) nothing has started. I expect this is due to my rather exotic industrial motherboard, so I gave up.

@gearsosprey As long as you read, give us this info still -> https://www.elektroda.pl/rtvforum/topic4067195.html#21174055

I have just scored a "Satafirm S11" on a Gigabyte 256GB SSD. The drive is under warranty, unfortunately I haven't made a copy for a couple of months and some data irrevocably lost ( I think?) .
I'm going to advertise it, is it 100% possible to deactivate it ( data ), format/zero/other?, if there is even a hint of possibility I wouldn't want anyone to know my innermost secrets, hehehe.
https://www.elektroda.pl/rtvforum/topic4084538.html

The data is most likely to be recoverable, but I doubt that anyone on the guarantor side will do so. Worse if it falls into the hands of some Ugandan recycling company who may be tempted to recover more than the raw materials. You could try to repair the drive w/ the procedure described earlier - repairing it will destroy the data.

MM2X wrote:

I have just passed the "Satafirm S11" on a Gigabyte 256GB SSD. The drive is under warranty, unfortunately I haven't made a copy for a couple of months and some data has been irretrievably lost ( I think?) .
I'm going to advertise it, is it 100% possible to deactivate it ( data ), format/zero/other?, if there is even a hint of possibility I wouldn't want anyone to know my innermost secrets, hehehe.
https://www.elektroda.pl/rtvforum/topic4084538.html

.

Yes, of course! The possibility of data recovery hits you as soon as you carry out the procedure to REPAIR it I know, sounds cursory, but that's exactly what it is. You will get a more or less functional media without data .

a_jablon wrote:

The possibility of recovering the data will hit you as soon as you carry out the procedure to REPAIR it I know, sounds cursory,

.
I want to do just that, unfortunately I've been trying to get my computer back on its feet for the last two days after this disk meltdown.
https://www.elektroda.pl/rtvforum/topic4084538.html#21287830

Could this damaged disk cause problems with Bios crashing, up until now I had no problem with this in various situations there was always a second bone ready ( Dual_Bios) and now both crashed.

SSD failure is not related to the BIOS. These are two separate topics.

There was a symptom of "SATAFIRM S11" ? If yes, the problem is indeed with the SSD.... If NO, I would not prejudge the problem with the SSD. A USB-SATA adapter and a drive plugged into another computer in this way might brighten things up a bit. CrystalDiskInfo could show e.g.: amount of reallocation, in general: SMART parameters.

I looked into the linked topic: there you have DualBios on the motherboard and the protection mechanism detected content corruption (content inconsistent with the checksum) so it took a rescue action. I don't want to malign, but it could be a problem with the FLASH cube itself, on which the BIOS content is stored. Contrary to appearances: it's not at all as "immortal" as many think.

a_jablon wrote:

Was the symptom "SATAFIRM S11" ?

.
Yes, it was on the basis of this description that I came across your topic here.

a_jablon wrote:

Contrary to appearances: it's not at all as "immortal" as many think.

So, when looking for a new board in the near future don't suggest "Dual-Bios", in the last few years I have had a message at startup that the computer is using this option.
I managed to get it working by installing the components on another disk and so far it is running . Now the "repair" procedure, maybe I'll figure it out .

Is your board alive after performing the above-mentioned operation? If yes, don't combine, use it. If the bios base cube was hit, you would have bios recovery in a loop, i.e.: every reboot, possibly a failure message, as the checksum is checked after writing.

Dual Bios is a nice thing, sometimes it comes in handy. I wonder if it's a primary criteria for choosing a motherboard, and it comes out to me rather not. It's nice to have it of course. Personally, I like Gigabyte's boards - trouble-free looking back at my and customers' PCs. Sticking with this manufacturer is not a bad option.... Which doesn't mean forbidding a look at menus from other manufacturers .

a_jablon wrote:

After the aforementioned surgery was performed, is your disc alive?

.
Yes, the patient came back to life.

a_jablon wrote:

If so, don't combine, use it.

My ability to combine ( whatever that doesn't mean, hehehe ) is limited anyway as if, what .

a_jablon wrote:

If the bios base cube was hit, you would have a bios recovery in a loop, ie: every reboot, possibly a failure message,

.
I thought that one is already dead and from the other one the Bios is not read every time.

a_jablon wrote:

I personally like Gigabyte boards -

.
Somehow I've always had Gigabyte and Intel, my requirements are medium/low - no gaming etc.

a_jablon wrote:

Sticking to this manufacturer is not a bad option.... Which does not mean forbidding a look at menus from other manufacturers

.
Also I think there is strength in diversity but sometimes it loses out to habit.

Thank you mate Works on software "SBFKB1H5_13112019.BIN"

I have this problem. First, reheat SSD NAND. If not, show all ID. Reball NAND.
Look at this page. I fixed it and re-MP program.
https://flashboot.ru/flash_recovery/2023/01/26/ps3111-ssd-fix-hardware-problem.html

I managed to put on my feet a Gigabyte 256GB ssd (Satafirm-S11) , admittedly I had a different idea / disk under warranty and I wanted 100% data erasure and here as "for spite" works / .

.
Thanks fest .

>>21441640 It won't last too long, until the first nand fill, then it will disappear again.

Marvell_88i8945 wrote:

It won't last long, until the first nand fills up, then it disappears again.

.
And that's the point .