Drive: SanDisk X110 256GB, SD6SB1M-256G-1022I, 2.5" SATA
Controller: Marvell 88SS9175, MLC NAND
PCB: 54-50-05079-02 (underside label 54-90-20128-256G)
Firmware on label: X231600

History:

I connected the drive to a Promise FastTrak 376 RAID card, and it stopped working. It reported about 80% health before that.
On a different SSD that was on the same card, HxD shows a “Promise Technology, Inc.” block 63 sectors before the end of the disk, so the card probably wrote something similar to the X110.
The drive is no longer detected. The BIOS (ASRock B75) freezes while waiting for it. It’s also not detected on a second PC or through a USB-SATA adapter.
I never formatted, initialized, or flashed it.

Console findings:

UART at 115200, 3.3 V, on the pads along the top edge. I read it with an ESP32, and the output is scrambled in the usual way.
With no SATA link, it repeats a “no SATA PHY” message about every 0.64 s.
With the SATA link connected, it prints one burst and stops. I decoded it by matching it against the HardMaster SD6SP1M log: a [FATAL FAILURE] dump, RTOS ErrNo 0x10A00082, RTOS Task ID [0x4 = STAR_4], module STAR_ErrorHandler.c. The values are approximate. (These are according to claude AI)

Questions:

Has anyone seen this fault on an X110 or Ultra Plus after a RAID card wrote to the drive?
Does anyone know the safe-mode pads for PCB 54-50-05079-02?
Does PC-3000 SSD (VanGogh) handle the 88SS9175 well enough to recover the data?

There are two pads at the corner of the SATA connector. I don't know their function, and I would be reticent to short them. Instead, it might be worth grounding the Dout pin of the temperature sensor. This might provoke a firmware panic.

ADT7302, Analog Devices, +/-2degC Accurate, Micropower Digital Temperature Sensor, marking T02:
http://www.analog.com/media/en/technical-documentation/data-sheets/ADT7302.pdf

_________________
A backup a day keeps DR away.

The Promise metadata explanation fits what Zeelo99 described: if the FastTrak wrote its own configuration block over the start of the drive, the Marvell controller can fail to find a valid translation table and stall before it ever brings up a SATA PHY. That would explain the repeating message with no link, and the single burst once a host is attached. On the safe-mode pads, I would probe the board with a scope before shorting anything, since a wrong pad on this PCB can leave the controller latched. For PC-3000, the 88SS9175 support is worth confirming with the vendor rather than trusting forum claims.