PCB fault: 201221500B REV B -Seagate EXOS X20 ST20000NM007D-3DJ103

Hi. I have a damaged PCB on this model, it will not power on. The problem came about after SATA power GND pin 6 came into contact with something (I think) there is some charring between SATA power GND pin 4 & 5. I don't see any other surface damage on the PCB. I have removed GND pin 6 as it was damaged.

I suspect the fault is on a 5V TVS diode or 5V e-fuse. But I need help from more knowledgeable people.

I have measured a whole lot of things with the help of AI, but I have started confusing myself at this point I think. I have not tested with power on, only continuity & resistance. I am a bit cautious to power it up, as the power pin plastic backplane has broken off, the backplane still exists and can be carefully placed in place so a SATA power connector can be inserted, but I would prefer to test as much as possible without power first!

-

The conclusion I've come to, but am unsure of: Because the protection switch inside J9 (5V e-fuse) is stuck permanently open, and the rest of the circuit is 100% healthy, bridging J9's input pins (1–3) directly to its output pins (9–11) bypasses the broken internal switch and will safely restore 5V power to the drive.

It would be nice to confirm this. I will follow testing directions.

There is also the component 'K03' near the 5V lines, which I don't know if is relevant, since 'J9' was the 5V e-fuse.

Pictures here:
https://imgur.com/a/AoJw0oy

"K03" is a MOSFET which functions as a flywheel diode in a buck converter.

UT6K3, Rohm, N-ch + N-ch middle power MOSFET, 30V, 5.5A, 2W, marking K03, DFN2020-8D:
https://fscdn.rohm.com/en/products/databook/datasheet/discrete/transistor/mosfet/ut6k3tcr-e.pdf

I would measure the resistance between ground and the Vin and Vout pins of "J9". If there are no shorts, I would power up the board on its own and measure the voltages at Vin and Vout.

_________________
A backup a day keeps DR away.

Hi, thanks for the reply! Before I try to power it on, which might be a bit messy with the state the sata pins are in, here is the info I've gathered sprinkled by AI guidance:

1. Resistance & Diode Measurements (Unpowered):

• J9 Input (Vin) to Ground: Open / Infinite Resistance (OL / "1")
  -> Confirms no short circuit on the incoming SATA 5V power entry.
  
• J9 Output (Vout) to Ground: ~1.6 MΩ Resistance / 0.500V Diode Mode drop
  -> Confirms no downstream short to ground. Main controller, RAM, and regulators are intact.
  
• J9 Vin to Vout (Across Chip): Open / Infinite Resistance (OL / "1")
  -> Confirms the internal switch of the J9 e-fuse is permanently blown open.
  
• 12V Rail (ADTP): Open / Infinite Resistance (OL / "1") to Ground
  -> Confirms the 12V rail survived the event completely unharmed.
  
• Downstream Ceramic Capacitors (MLCCs): Open (OL) across pads / 0.500V Diode Drop
  -> No shorted capacitors on the 5V line.
  
• Power Inductor / Choke: Direct continuity (0 Ω / Beep)
  -> Continuous, healthy trace path through the power filter.

There are zero shorts to ground anywhere on the 5V or 12V rails.

The 5V e-fuse (J9) absorbed the power surge and failed open, protecting the downstream logic.

Because the downstream circuit is completely healthy, bridging Vin (Pins 1–3) directly to Vout (Pins 9–11) on J9 should safely restore 5V power to the drive.

-

It is my understanding (which could be wrong) that the e-fuse is "open" until powered, which is why I don't know if I can go by this information without doing the powered on tests.

btw is the component I have annotated a TVS diode, or do TVS diodes always have that bar line on one side & letters on top of them?

Because testing continuity on that component I get beeps, i.e. shorted

Your "5V TVS Diode" is an inductor. It is just a coil of wire wrapped around a ferrite core. It normally tests shorted.

_________________
A backup a day keeps DR away.

Thanks for info. I added some more info, but that post had to be approved above. Power on test will have to later on.

There are no shorts. You won't know if the e-fuse is open until you power it up.

If you want to limit the current, you could use a USB-SATA adaptor.

_________________
A backup a day keeps DR away.

a fuse should be testable by resistance across the component. it should show a short but if it does no it is likely failed. AFAIK this hard disk has a 5 year warranty so send it back on RMA is my suggestion

As for USB adapter, I have looked for one that has a plug on it to supply 12V like the enclosures I use as the cable I have is only USB powered. The cable is laptop oriented where 2.5" drives are standard.

Recent docks that hold a disk vertical has some potential.The enclosures I use all have the component side down making them useless for anything other than testing and copying.

_________________
USB 3.5" enclosures are $19 USD postage paid from China, I own several

I have already talked to Seagate, unfortunately since the plastic backing of the power pins has broken of and 1 pin is fully removed, it seems it will not qualify for warranty RMA.

@Amra, please ignore "Hardcore Games". This member is either a troll or a bot. In any case, he/she/it knows nothing about data recovery or electronic fuses. It's all just noise.

_________________
A backup a day keeps DR away.

Fuses are used in some older cheap laptops but more recent machines use MOSFETs to handle power. Depends on the vintage. Poking around with the black lead and ground you get a better idea when shorts exist outside the usual ground connections.

Recently an Asus G751J came into my shop and it is 4th gen but it powers up so maybe it has some potential. It has a GTX 980 for discrete graphics so it's a heavy machine.

_________________
USB 3.5" enclosures are $19 USD postage paid from China, I own several

Bot, even trolls are not that stupid.

I am considering a pcb swap service, so that I get a fully functioning sata connector again with the plastic backing intact, I know that aspect would bother me even if I manage to fix the short with soldering or whatever. I'd like to keep using this 20TB drive into the future, as it has become 200-300% more expensive since it was initially purchased.

The data on the drive is not really important, I just want to keep using the drive.

Do forum members have some recommendation of a service where I send in my pcb, chips get transferred to a fresh one and I get both back? I have seen hdd-parts.com mentioned, however I can't find my board on their site. Actually I can't find many newer boards on that site, so I'm not sure if they keep it updated.

They do have a repair service.

https://www.hdd-parts.com/13080601.html

_________________
A backup a day keeps DR away.