Firmware for WD1200JB-00REA0 (Starling)

[dsockt]

Specifically, WD1200JB-00REA0-2060-701292-001

I suspect the SA is no longer readable as the drive spins up and "clicks". Has been progressively failing to detect in BIOS over the past few resets. Was reading extremely slow when it was last detected a month ago, with just a few bad sectors, but data was being read OK. Now appears "blank" in BIOS. ROM reads fine.

Using scraped-together tools (non-PC3K/MRT/etc.) calling WD VSC ATA commands gained access, so it would be great to send over firmware while in safe mode to see if the unit can read again.

Yes, the drive is old and replacements are cheap. It's the contents of the drive I'm after. Non-critical but like-to-have items, plus it's a learning experience for me.

Any suggestions and related comments appreciated!

ds

[Zero Alpha]

Have you looked at the firmware section to see if its already available?

[dsockt]

I have;

There's a similar package, for a WD1600JB-00REA0; of course, different parameters (LBA, etc). Lots of other families appear to be available, but not Starlings.

Nonetheless, the drive has begun to read again, so ddrescue continues.

There appear to be two issues with this drive; the encoder/decoder chip (?) gets extremely hot as felt from the underside, and/or the heads are going bad. Perhaps one is the cause of the other.

I'm not equipped to perform a head or a ROM chip swap at this time, so we'll let it ride.

ds

[fzabkar]

If the MCU is getting hot, measure the Vcore and Vio voltages.

[dsockt]

I'll check tomorrow, but looking at photos and descriptions, it's not the MCU that's getting hot, it's the VCM. I read that 100C is acceptable for this chip??

Sure feels like I can pop a kernel of corn on it.

ds

[fzabkar]

If you remove the PCB and power it up on its own, does it still heat up?

[dsockt]

Sorry for the delayed reply;

Tested PCB alone, no temperature issues. Same in safe mode. Seems to draw nominal wattage while (attempting) to seek/read, gets very hot while motor/heads are active.

Also, was asked to check voltages to some chips; I am unable to find official schematics and don't want to be guessing that pin1 is Vcc on each.

Power was lost to the drive recently and it's back to clicking looking for the SA.

Curious, having the firmware to 1600jb-00rea0 available, how damaging would it be to attempt to send that firmware in safe mode, unmodified?

Unmodified since I wouldn't know what bytes to modify in the 1600's modules to make it appear as a 1200.

Thank you in advance.

[SWM]

The disk firmware has nothing to do with the described problems.
If the data is extracted, you can also clean the contacts on the board (to the hermetic block) from oxides.
A school eraser and then alcohol.

[fzabkar]

When you find "official schematics", please tell the world. As for "guessing that pin1 is Vcc on each", I have no idea what you are talking about. I'm asking for the Vcc and Vio supply voltages (Q8 and R050). C6 and D4 are the -5V preamp supply.

[dsockt]

My apologies, I thought you were referring to the microchips’ voltages.

I’ve attached an image with values requested and some additional I found on another forum post of a similar drive troubleshooting

This is drive not attached and with ground on the molex cable. Hope this is useful.

Also, I scrubbed with a pencil and isopropyl on the contact pads and pins to the drive, same result.

ds

Attachments

[fzabkar]

-5V and +3.3V are OK. I think there was a similar PCB in another thread where the researcher measured +1.37 as the Vcore. In short, I think these voltages may be OK.

https://forum.hddguru.com/viewtopic.php?p=298644#p298644

[dsockt]

That's a good read, thank you.

I checked a few other resistors mentioned in that thread, mine seem okay.

I'm thinking these heads are trash. That's why I asked for firmware because of the three times (over the course of the last year) that it did detect on the PC, I was able to read _something_, albeit with tons of random read errors. My thought is that it's read erroring the service area on power up, and that I might be able to bootstrap it's detection and continue to rescue the data.

Thanks for the help, all.

ds