SEAGATE ST3500320AS

[Piper]

Hi,

I have a kind of similar situation.

It is about Seagate ST3500320AS disk with firmware SD04 not seen by the bios or windows all the time until it was not recognized at all.
The disk was spinning fine, no funny clicks or something. I presumed the PCB is dead. I got new PCB. I tried it without changing the bios and disk is recognized again only parameters are not correct.
Then with use of the ch341 programmator I swapped bios from original PCB to donor, everything went well. Now I have two PCBs that don't even spin up the disk. Then I've read somewhere that this ch342 programmator does not use correct voltage for signal, so I thought I've burned PCBs.
Then I got a new (second donor) PCB, check it with the disk without changing the bios and the disk spins fine. This time I've swapped the bios with soldering and again afterward the disk does not spin.
Now I am about to get another PCB.
Could somebody point me in the right direction, what I am doing wrong here?

I did save the data from this disk at the beginning of failing of the disk, but the problem is that the disk where data was saved to is dead - making scratching noise and I think it would be very hard to get anything from it. I have more chance to get data from original disk.

Thanks,
AJ

[Piper]

Here are the Rom dumps.

Original PCB dump
Donor 2 dump
SD04 from online source dump

On original PCB all I did was dumping rom with flash tool CH341 and now is dead.
On donor1 I flashed rom with original rom and now is dead.
I solder rom from donor1 to donor2 and donor2 is dead.
I solder back original rom from donor2 to donor2 and the donor2 is still dead.

Is it possible that something is wrong with my PSU? Other drives are working fine.

Attachments

Seagate_Donor2_PCB_dump1.7z

(411.93 KiB) Downloaded 155 times

Seagate_SD04_from_online.7z

(414.42 KiB) Downloaded 181 times

Seagate_Original_PCB_dump1.7z

(414.65 KiB) Downloaded 156 times

[pepe]

change offset 000001 like this:

Code:

Offset      0  1  2  3  4  5  6  7   8  9 10 11 12 13 14 15
              vv
00000000   1A 23 00 00 40 00 00 00  40 82 00 00 5C 31 04 00    #  @   @‚  \1 
00000016   02 38 00 00 04 38 16 06  06 48 18 06 03 58 9C 06    8   8   H   Xś
00000032   0B 58 CC 07 05 58 DD 07  0E 70 FD 07 00 00 00 08    XĚ  XÝ  pý     
00000048   00 00 00 00 00 00 00 00  2E A0 31 02 40 82 00 00           . 1 @‚ 

[Piper]

Thanks for help.
OK, I can do that but I am not there yet.
The problem here is that I am doing something fatally wrong here, and I don't know what.
I wanted to revert donor2 PCB with soldering back on the original donor2 rom, but the PCB is still not working.
Looks that I am burning these PCBs somehow.

[fzabkar]

@Piper, your patient ROM had a strange checksum error, but your donor ROMs are OK. Can you upload photos of your no-spin PCBs?

BTW, this tool may be of interest to you:

https://www.hddoracle.com/viewtopic.php?p=18531#p18531

[Piper]

Ha, ha, all of PCBs are now no-spin.
Thanks for help.

I already have this useful tool.

Attachments

PCB_donor2.7z

(1.11 MiB) Downloaded 166 times

PCB_original.7z

(1.08 MiB) Downloaded 155 times

[Piper]

I did notice that on patient PCB a small resistor is missing left from the bios chip.
All I can think of is that maybe I did break away this resistor with clamp for flashing bios.

[fzabkar]

Measure the resistance between ground (any screw hole) and each of V1, V2, V3. If there are no shorts, measure the voltages at those same points. However, before powering up the PCB, measure the SATA +5V and +12V outputs from your PSU.

As for the missing resistor, it appears to be the pullup resistor for one of the other pins, possibly /HOLD, /WP or /CS. If the /HOLD pin is floating at a low level, then this would prevent the ROM from being read. I think something like 4K7 should suffice.

Attachments

reg3.jpg (119.03 KiB) Viewed 53824 times

reg2.jpg (102.05 KiB) Viewed 53824 times

reg1.jpg (63.94 KiB) Viewed 53824 times

[Piper]

PSU 5,13V and 12,08V


PATIENT
V1 2.59V 0.863kOhm
V2 1.126V 71.8 Ohm
V3 or Vneg 0V 11.43 kOhm


DONOR
V1 2.61V 0.856kOhm
V2 1.14V 49.8 Ohm
V3 or Vneg 0V 9.28 kOhm

[fzabkar]

Vneg should be -5V when the drive is spinning, so that result is inconclusive. The other voltages (Vcore and Vio) are OK. I don't know what else you could do with a multimeter.

You could measure the Vcc pin of the ROM ...

[Piper]

OK, thanks.

Do you have any experience with flashing Rom on this old Seagate drives? Is this even possible, I've read somewhere on this forum that in these drives family you can only read Rom, but you can not write to the Rom.
Before working on this drive I did play around with newer Seagate drive ST31000524AS and everything was smooth sailing. I did not manage to burn the PCB .
Maybe I will get another PCB (they are cheap).

[pepe]

@Piper, your patient ROM had a strange checksum error, but your donor ROMs are OK.

it wasn't a CS error, byte at offset 1 was read bad, but the rest of the dump appeared to be fine. The above correction did the job.

the missing part is a bypass capacitor, likely not causing any issue if missing.

What do you get from the terminal when powering it up?

[fzabkar]

@Piper, your patient ROM had a strange checksum error, but your donor ROMs are OK.

it wasn't a CS error, byte at offset 1 was read bad,..

Yes, I realise that. I didn't mean to imply that the checksum was at fault.

[Piper]

What do you get from the terminal when powering it up?

Unfortunately, I do not have serial cable for terminal connection yet.
How do I know the PCBs are blown?
If I connect working PCB (only PCB without drive) I get some kind of drive visible in win devices.
After my magic PCBs manipulation, I get nothing in win devices when PCB is connected.
So, my conclusion is a have a supernatural power to make the hard drives PCBs brain-dead.

[fzabkar]

Recheck your soldering. Use leaded solder rather than lead-free, if you can find it.

[Piper]

Little update

I've managed to communicate with the drive with my cable.

1. Original PCB; as you figured it out the rom is corrupt, so drive does not spin

2. Donor PCB; I have managed to flash rom with corrected version, drive spins. If I connect it to sata, drive is not visible in BIOS nor windows. Windows works very slow.
On terminal when drive starts, there is no information (no copyright, no nothing like on working drive). After a while with ctrl Z I get F3 T> prompt

With ctrl l I get following info

With this donor pcb and corrected rom I think that the drive is behaving like before my magic interventions.

Can we make a conclusion now that the heads are guilty of this strange drive behavior?

[pepe]

it would be a lot easier to get my hands on the drive and recover it than continuing this thread and scratching my head after each post...

[SWM]

After powering up the ctrl-Z and ctrl-X.
Connecting the SATA of a faulty drive to Windows is a bad idea.
p.s. And yes, Pepe is right. I didn't see the previous post while I was typing it...

[fzabkar]

Try the "F" commands in the following ZOC script (select level T):

https://www.hddoracle.com/viewtopic.php?p=10886#p10886

Hopefully they will stabilise the drive enough so that you can clone it.

WARNING: The commands are Case Sensitive.

See this thread:

https://forum.hddguru.com/viewtopic.php?f=1&t=38411

Code:

F"READ_SPARING_ENABLED",0,22
F"WRITE_SPARING_ENABLED",0,22
F"OFFLINE_SPARING_ENABLED",0,22
F"DAR_ENABLED",0,22
F"DISABLE_IDLE_ACTIVITY",1,22
F"BGMS_DISABLE_DATA_REFRESH",1,22
F"ABORT_PREFETCH",1,22
F"READ_LOOKAHEAD_DISABLED_ON_POWER_UP",1,22
F"READ_CACHING_DISABLED_ON_POWER_UP",1,22

[Piper]

it would be a lot easier to get my hands on the drive and recover it than continuing this thread and scratching my head after each post...

I am very sorry that you have to scratch your head because of me
Are you able to check the drive through TeamViewer or you need hands on the drive?

After powering up the ctrl-Z and ctrl-X

Try the "F" commands in the following ZOC script (select level T):

OK, I will check this tomorrow, for now it's kind of rocket science to me.