Seagate Barracuda es.2 pcb st3500320ns -- dead

[13hm13]

I've had this 500GB dead Barracuda drive in storage for about 9 years. For fun, it'd be cool to have it work again.

I think the "SMOOTH" motor-control IC may be shorted to ground???

Inside the black square (lower right), I have probed (with DMM in resistance mode) some passive components.

So, for example, across that black ST c744 diode, or surrounding ceramic caps, I get about 0.7 ohms across them. I have removed the c744 diode and probed the open PCB pads (where the diode was soldered) -- same 0.7 ohm reading!
I do not see any blow holes or other heat damage on that IC or any other component.
But something in or around that black square has apparently shorted.

I have not swapped that PCB not much less gone any further in troubleshooting. I don't have many proper tools (hot air rework stat., etc) to really proceed.

Anyone think of a quick solution?

For the image below, ignore the red circle and its caption. This PCB image is not mine (but the PCB is same model as mine).

Attachments

[fzabkar]

Did you test the 5V and 12V TVS diodes near the SATA power connector?

See http://www.users.on.net/~fzabkar/HDD/TVS_diode_FAQ.html

BTW, "744" appears to be a YWW (Year/Week) date code.

[13hm13]

I'll check 'em out.
BTW: What's the most-common failure mode for these HDDs (IC failure -- if so, which one), diode short/open, capacitor failure, motor failure, voice coil.

For this Barracuda, the motor does not spin up; and the USB signal is totally absent. The Cooler Master external case is fine; it's definitely this drive that's faulty.

[fzabkar]

Most of the PCB failures I've seen in the storage forums have been due to external events, primarily overvoltages. These are nearly always contained by the TVS diodes and their series connected protection devices.

I have compiled some information here:

PCB protection devices:
http://www.hddoracle.com/viewtopic.php?f=100&t=1615

The following datasheet should help you to understand the function of SMOOTH chips.

L7250, SMOOTH, spindle motor + VCM controller, ST Microelectronics:
http://www.datasheetcatalog.org/datashe ... Xyuswx.pdf

[13hm13]

It's none of the TVS diodes nor ceramic caps. I've removed most of those as a process of elimination and the short to ground remains.
SEVERAL pins on the SMOOTH chip are near 0 ohms to gnd. I think my original assumption was correct.

[fzabkar]

There would probably be 3 Rsense pins for the spindle motor current. One parallel array of 1R00 resistors provides current sensing for the spindle, the other for the voice coil. The former is grounded at one end. These pins would account for some of the low resistance readings.

I don't understand the function of the subject diode, but I'm wondering if it forms part of a Vboost circuit. Can you identify any other markings on the diode? Since you are measuring 0.7 ohm across it, then the short must be on the IC side of the 1R00 resistor, which does then point to the IC as the culprit.

Are the Vio and Vcore regulators working? These voltages would be present at the junction of each 3R3 coil and the associated filter capacitors.

Just to satisfy my curiosity, would you be able to tell me where the two opamps (adjacent to the shock sensors) get their Vcc supply (pin #8)? Is it provided by either of the 3R3 coils (Vio) or the +5V or +12V SATA supply?

[13hm13]

I'm giving up on this one for while. I see eBay has spare boards around $15, so I may spring for one of those sometime.