If the MCU survived the surge, then I suspect that retrieving its embedded ROM contents may be as simple as adding two external supplies (+2.5V and +1.2V) and asserting the POR (power-on reset) signal to the MCU.

_________________
A backup a day keeps DR away.

Assuming that it would work and assuming you wouldn't jtag to the MCU then you would still need for either the TX/RX X-Modem component of the circuit to be functional or for the SATA components to be functional in order to issue VSC to the MCU and extract the ROM in PSV/Kernel mode.

_________________
1Q9xrDTzTddUXeJAFRn37aqh1Yr6buDCdw - (Bitcoin Donations)
paypal.me/Spildit - (PayPal Donations)
The HDD Oracle - Platform for OPEN research on Data Recovery.

All those components are part of the MCU. JTAG wouldn't be required.

_________________
A backup a day keeps DR away.

Interesting ! It looks like doable in particular if there is still some points on the PCB with those voltages. One could wire those points to the necessary pins on the MCU unless is BGA without traces to solder to.

_________________
1Q9xrDTzTddUXeJAFRn37aqh1Yr6buDCdw - (Bitcoin Donations)
paypal.me/Spildit - (PayPal Donations)
The HDD Oracle - Platform for OPEN research on Data Recovery.

The voltage test points would be easy to find. I haven't checked my PCB, but I think the attached image is correct.

NPOR is probably one of the 8 signals that run between the SMOOTH IC and the MCU. It is probably high when the supply rails are OK and low when they are out of spec. That's a matter for experimentation, though. I would use a 1K resistor to test each of these signals, either as a pullup to Vio or pulldown to ground

_________________
A backup a day keeps DR away.

Even assuming that it would work the chances are high for a pre-amp damage.
I think the OP would want to start with a HOT SWAP to confirm head/pre-amp functionability first.
If the drive can still read with native heads then geting the original ROM or a ROM that works might be an alternative to a samrt hot swap.

_________________
1Q9xrDTzTddUXeJAFRn37aqh1Yr6buDCdw - (Bitcoin Donations)
paypal.me/Spildit - (PayPal Donations)
The HDD Oracle - Platform for OPEN research on Data Recovery.

The OP hasn't provided a photo or the results of any TVS diode tests, so it's not possible to say whether the surge hit the +5V supply or the +12V supply. A 12V surge is survivable.

_________________
A backup a day keeps DR away.

Interesting.
Let's wait for the OP to reply then.

_________________
1Q9xrDTzTddUXeJAFRn37aqh1Yr6buDCdw - (Bitcoin Donations)
paypal.me/Spildit - (PayPal Donations)
The HDD Oracle - Platform for OPEN research on Data Recovery.

The first (and unique at the moment) test we did,
has been to completely remove the SMOOTH ic.
Since it made no difference, we abandoned the idea to go further with diagnoses of the board.
But all above is really interesting and requires some tests and measures.

The SMOOTH IC generates the supply voltages. Without it, you get 0V for both Vcore and Vio.

_________________
A backup a day keeps DR away.

Uh-oh
Never dealt with those type of researches, thank you Fzabkar you are always an authority in matter of knowledge.

That faked our test so, since we thought other components could been burned, while SMOOTH ic could be the unique one so.

Instead we thought (but for a while only) to feed power to the Marvell IC alone, but without schematics...

now you provided important hints.

Thank you

Would you attempt to replace the SMOOTH ic?
We have the tools and the experience to do that but again without schematics, the worry is to burn it because of other possibly damaged components.

I would measure the resistance (on the 200 ohms range) between ground, +5V and each of Vio and Vcore. This will tell you whether the MCU or Q2 are shorted. Also check the zero-ohm resistors in that area, eg R46.

If there are no shorts, then I would be willing to try a SMOOTH replacement.

_________________
A backup a day keeps DR away.