Gentlemen, just wanted to update you on the progress so far. I've managed to borrow an oscilloscope and noticed that the HDD does not pull its Tx line to 0V when communicating with the world. Instead it uses ~1.9-2.0V for "1" and ~0.6-0.8V for "0". Using PL-2303HXA datasheet I was able to tweak its Serial I/O voltage so that the built-in Schmitt trigger interprets these levels correctly. It turns out that
3.8-4.0V applied to the VDD_325 pin works a treat. Because noise on the HDD's Tx line increases a lot with voltage, I sticked to 3.8V, which boils down to Schmitt thresholds of around 1.25V and 1.9V (negative-going and positive-going, respectively). (Sharing these so that future generations may tweak their interfaces accordingly.)
I can also confirm that the
correct interface values for this drive are: 38400 speed, 8 data bits, no parity, one stop bit, no flow control.
The second good news is that - judging by the prompt - the
drive is indeed an F3 generation.
Off to trying some commands then