Corrupted Seagate ROM

[Sling Shot]

It's a strange case where the customer isn't telling me everything about the drive, despite being asked, but thats nothing new, I guess.

Anyway, despite all the screws missing from the drive lid (apparently they removed themselves), it looks like they haven't been inside the drive itself.

However, the PCB has had a diode removed from its powerline side (again, it removed itself).

Anyway, the drive will spin up, but it knocks like crazy. 

Cannot read ROM in boot code mode so programmer read only (dump attached)

Loaded ROM into 3K utility and get a warning:

ROM image contains crytical damaged objects
[CAP, RAP, SAP]

The drives serial matches in the limited info from ROM dump, so it might still be the native PCB.

The question is, can this ROM be fixed, or is it game over ?

Dump text and pics attached: 

Thanks in advance.

[pbzcbf...@gmail.com]

Without actually examining the ROM, if the drive spins up, then the ROM is probably OK. The message about damaged adaptives usually means that only the first half of the ROM was dumped.

[pbzcbf...@gmail.com]

Actually, it's a bad dump. You need to dump the ROM at least twice and compare the dumps.

[Sling Shot]

Thanks Franc, will try again and come back to you.

[pbzcbf...@gmail.com]

I would measure the Vcc voltage on the PCB (pin #8) to make sure your programmer is set correctly.

[Sling Shot]

Did a few re-reads of the ROM, and the first version was definitely a bad dump.

Versions 3 and 4 are identical, so I settled on version 4.

Version 4 ROM and dump text attached.

Vcc on pin 8 was 4.94v, but my programmer doesn't allow for voltage control, so not sure how I can do anything with that unfortunately.

[pbzcbf...@gmail.com]

What!!!

Please get rid of your programmer before you destroy your customers' ROMs. Today's ROMs are 1.8V devices. 5V ROMs disappeared decades ago.

[Sling Shot]

Can't say we've had any issue with this programmer to date, but sure, what do you recommend using instead? 

Also, this 5V case is from 2010, so it should be okay with this one. 

Do you have any ideas on the contents of the ROM info? Can it be fixed? 

[t...@desertdatarecovery.com]

Might be some confusion here. Franc you asked him to measure pin 8 on the PCB and it came back as 5v. Nothing to do with him programmer (or maybe I missed something).

 

Tim Homer - Lead Engineer

Desert Data Recovery

+1 (602) 686 2622

t...@desertdatarecovery.com

www.desertdatarecovery.com

--
Data Recovery Certification Group / for issue with google group please email sc...@myharddrivedied.com
---
You received this message because you are subscribed to the Google Groups "DataRecoveryCertification" group.
To unsubscribe from this group and stop receiving emails from it, send an email to datarecoverycertif...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/datarecoverycertification/47bac36d-bfc3-48f0-bc49-2b3541533c4bn%40googlegroups.com.

[img]

I think he measured the programmer's voltage because on that PCB there is a Winbond 25X40BLSP9 which is a 2,5-3,3V flash so 4,94V definitely an overvoltage for that.

https://www.alldatasheet.com/datasheet-pdf/pdf/443818/WINBOND/W25X40BL.html

4,94V is a typical USB rail voltage which can be originated from the USB powered CH341A without voltage mod. CH341A needs vmod for 3,3V:

https://wiki.chucknemeth.com/usb-devices/ch341a/3v-ch341a-mod

Or you can buy another version from Aliexpress (v1.7 with built-in voltage selector switch):

https://www.aliexpress.com/item/1005009264729721.html

To view this discussion visit https://groups.google.com/d/msgid/datarecoverycertification/00f001dc0b1d%24529a8e10%24f7cfaa30%24%40desertdatarecovery.com.

[Sling Shot]

Just for clarity.

The PCB is powered via a SATA power cable.

Using a multimeter, I have measured the voltage going to each leg of the ROM chip.

Legs 1,2,3 all received 2.5V

Leg 4 receives 0v

Legs 5,6,7 all received 2.5V

Leg 8 receives 0v

The 4.94v was measured from pin 8 of the SATA power on the PCB.

My programmer is a 3.3v unit that hasn't caused me any issues yet, but it doesn't have any switches to drop it to 2.5 or 1.8.

I repeated the process using a donor PCB (that hadn't been tampered with), and I got the same results as above.

The customer has since miraculously remembered plugging the drive into a laptop's PSU, so we can assume the drive was electrocuted.

It also went to the local computer shop before it came to us.

The patient ROM looks like is missing information.

The question is, is this fixable? 

[img]

Thank you for the clarification.

'The customer has since miraculously remembered plugging the drive into a laptop's PSU, so we can assume the drive was electrocuted.' -> It makes sense then.

'Leg 8 receives 0v' -> It is interesting because pin 8 is the positive power supply (VCC) input of the flash chip so if there's 0V then it means the chip does not get power supply or at least something is shorted on that power rail.

Can you measure 3,3V anywhere on that board?

'My programmer is a 3.3v unit' -> Okay, that will be fine for this type of flash.

'I repeated the process using a donor PCB (that hadn't been tampered with), and I got the same results as above.'

So you replaced the flash chip from the patient board to a donor PCB and then the patinet drive with the donor PCB acted with the same symptoms? Or you just measured the untouched donor board for reference?

To view this discussion visit https://groups.google.com/d/msgid/datarecoverycertification/bfae9976-c437-4b3a-9e70-ffcc4e7f1b1an%40googlegroups.com.

[img]

And IDK from the top of my head which is the read channel on that board but you should check that - and the preamp - for (partial) short or other damages due to the PCB is 'electrocuted'.

[Sling Shot]

I'll jump back on it tomorrow and see what I can measure.

As for the donor board, I didn't swap the ROM chip over; I just used an untouched board as a reference.

My initial feeling was a nuked pre-amp, but the entire interaction with the customer—missing screws, tampered PCB, weird ROM dumps, and half-truths extracted through painstaking repeated questioning—left me somewhat suspicious of the whole debacle.

[Sling Shot]

Theres only one place I can measure 3.3v (marked on below photo).

Doing a continuity test on the read channels and preamp isnt finding a short either.

Just for clarity on the ROM legs, leg 1 is by the dot and leg 5 is opposite the dot. 

[img]

'Just for clarity on the ROM legs, leg 1 is by the dot and leg 5 is opposite the dot.'

As I expected too.

I didn't found the exact same board but the one with the same Winbond flash: on the flash's pin 8 there must be some voltage, probably 2,5V (not 3,3V so my initial assumption was wrong). If there's absolutely no voltage: what resistance to the ground can you measure on that pin (without powering on the board)? If no low resistance (dead short or partial short) measured (and also on the desoldered chip) then the fastest way is to move the flash ROM to an identical donor board and test the patient drive with that. And also in PC3K for reading / writing the ROM.

If you're in the mood of measuring, you can check up some things on the damaged patient board:

If the TVS diode is shorted on the 12V power rail than the the Smooth motor controller can be damaged, which also provides a few sub power rails to the components.

'Doing a continuity test on the read channels and preamp isnt finding a short either.'

That is good news.

On the PCB (in idle state, without the HSA attached) you can measure 1,3V (LSI MCU power supply), 2.5V but also there is 3,3V and 5V. Are they there? Some of the voltages are appears only when the drive is reading or writing, so you can find only these in idle state.

Can you measure very low resistance to the GND, on both sides of the bigger capacitors anywhere?

To view this discussion visit https://groups.google.com/d/msgid/datarecoverycertification/489e53ab-3782-4586-b4db-501898f320e4n%40googlegroups.com.