Hello,

recently my 1 TB Samsung HD103SJ disk died, it's not recognized any more by the BIOS and also not spinning or making any clicking noises.

TVS diodes (and 0-Ohm "resistors") are okay, DC-DC converters on the PCB are running (2.5 V, 1.8 V and -5 V).

When connecting the disk via serial connection the terminal shows only a view lines and then stops:
-------------------------------------------------------------------------
*PA VID=0000 PN=0004 Rev=0002- 785x Found
*PA VID=0000 PN=0004 Rev=0002- 785x Found
ChipRev = 63B0
U
S_0
[DEBUG HALT @MDL_RSRC_BufferAlloc():10034AB5 L0 1003A471]

RV En Sensor Circuit
En Shock Sensor Circuit
-------------------------------------------------------------------------

Also bought the exact PCB (BF41-00359A, Rev 01), swapped the rom but still same issue, disk is not doing anything (still have to check the serial output for this combination).

Any idea what is wrong and if there is a chance to fix it?


Thanks for your help,

Urmel

Did you try the pcb without move the ROM...if not, try..it can be bad ROM.

_________________
Rescue IT Datarecovery service Sweden
Rescue IT Dataräddning Göteborg AB
http://www.rescue-it.se

First reply 5 minutes after my original post.... that's impressive !

So the output for the "Replacement-Board" with its original rom looks like this:
---------------------------------------------
DDR size detected = 32MB
*PA VID=0000 PN=0004 Rev=0002- 785x Found
*PA VID=0000 PN=0004 Rev=0002- 785x Found
ChipRev = 63B0
U
S_0SHOS-(1AJ437fP.d26)
ENG>
----------------------------------------------


I've also access to the same Samsung hdd with a later manufacturing date, when this PCB is mounted
to the drive under investigation the output is:
----------------------------------------------
*PA VID=0000 PN=0004 Rev=0002- 785x Found
*PA VID=0000 PN=0004 Rev=0002- 785x Found
ChipRev = 63B0
U
S_0
RV En Sensor Circuit
En Shock Sensor Circuit
SO_1
SpinStartUp: mcSpinRPM = 0
RPM at Handoff: 578
Temp : 21 degC
SpinOk
mS1 00000003
SK C:117236 H:0
Loaded FIT ( 0: 0: 1)
CalibTable Loaded. Rev:0x14
Selective MARC NX Loaded
ResoTable Loaded. Rev:0x01
Ovly loaded to 0x0001CD00
Ovly loaded to 0x1005B800
----------------------------------------------


Does this help?


Thanks,

Urmel

As he says and also did you swap the right chip and do a good solder job?

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

Yep, swapped the 8-pin Winbond 25x chip, all this was done with a professional soldering station under a microscope, so I'm quite sure the soldering was done correctly. Chip orientation on the PCB is also correct.

The first serial dump in my second post shows the replacement-board with the rom that came with it, so this is the best answer I can give to mr_spokk question right now.

Oh, by the way, when the disk shows the "eng>" prompt, is it supposed to react to serial commands like "HE" ?


Take care,

Urmel

Bad ROM then...

I am quite novice, but I think that you should try the PCB under investigation with the donor HDA to see if there is something wrong /shorted inside the patient HDA. There is a risk the mini-pcb inside the donor HDA might get damaged.

Looks like bad (corrupted) Rom, if the drive spinns up ok with patient Pcb and Rom and stays quiet with patient Pcb and original Rom.

_________________
Rescue IT Datarecovery service Sweden
Rescue IT Dataräddning Göteborg AB
http://www.rescue-it.se

ISTM that if you would like to confirm a problem with the flash memory, then you could see how the board behaves when you lift one of the chip's important pins, or you could short one of its pins to ground (but you would need to be sure which pin was safe).

_________________
A backup a day keeps DR away.

Hello,

thanks for all the suggestions.

Tried the PCB from the working (but newer HD103SJ) on my "defective" drive, after connecting it to the SATA port of the motherboard the drive was detected by BIOS, started spinning and made some head seeking noises, but looks like track alignment (?) was wrong as seeking noise kept going forever. My conclusion is that spindle motor and head electronics are good, do you agree?

Tomorrow I'll swap the Rom from the defective drive with the Rom of the replacement board (same Rev-Level) and see how both combinations will behave, Ill keep you posted.
Tried that already (disk was not spinning up), but did not look at the serial dumps, maybe there is some valuable information there.

Another question, if the Rom is really bad/defective, is there any chance to fix that?


Regards,

Urmel

Hello all,

finally had the time to swap the Roms again, meaning that Rom from the original PCP is now on replacement PCB and vice versa.
Bad news is that the "[DEBUG HALT @MDL_RSRC_BufferAlloc():10034AB5 L0 1003A471]" error (my original problem) is "moving" with the Rom, so its occurrence is independent of the rest of the PCB hardware, so I guess this confirms that the Rom is bad.

When I connected the HDD with the original PCB and the Rom from the replacement PCB to the SATA port of the mother board the disks starts spinning, but there are "strange" noises from heads, like they are hitting a hard stop, the serial output looks like this:
------------------------------------------------------
DDR size detected = 32MB
*PA VID=0000 PN=0004 Rev=0002- 785x Found
*PA VID=0000 PN=0004 Rev=0002- 785x Found
ChipRev = 63B0
U
S_0SHOS-(1AJ437fP.d26)
ENG>@
[RESET:1 00000000:00000288:00070000]
[LINK:0123]
@
[RESET:1 00000000:000002C5:00070000]
[LINK:0123]
[HOST_ER:2004 A1.0000 00000000 00000656 3]
@
[RESET:2 00000000:00000715:00070027]
[HOST_ER:2004 B0.00D8 00000000 00000746 3]
BT C:2064, H:0

S_0SO_1
VOP!
PK_4 C: 0 H:0
SpinStartUp: mcSpinRPM = 0
RPM at Handoff: 575
Temp : 22 degC
SpinOk
SR:0
BT C:2064, H:0
SR:1
BT C:2064, H:0
SR:2
BT C:2064, H:0
SR:3
VOP!
PK_1 C: 0 H:0
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
AP:2 C: 0->156930 H:0->8 AP Cnt: 0
VOP!
PK_3 C:156753 H:8
UF6 GE
AP:2 C: 0->121862 H:0->9 AP Cnt: 0
VOP!
PK_3 C:123510 H:9
UF6 GE
AP:2 C: 0->122005 H:0->0 AP Cnt: 0
VOP!
PK_3 C:122151 H:0
UF6 GE
AP:2 C: 0->121856 H:0->1 AP Cnt: 0
VOP!
PK_3 C:123220 H:1
UF6 GE
AP:2 C: 0->121502 H:0->2 AP Cnt: 0
VOP!
PK_3 C:121648 H:2
UF6 GE
AP:2 C: 0->121932 H:0->3 AP Cnt: 0
VOP!
PK_3 C:122076 H:3
UF6 GE
AP:2 C: 0->123114 H:0->4 AP Cnt: 0
VOP!
PK_3 C:123289 H:4
UF6 GE
AP:2 C: 0->121368 H:0->5 AP Cnt: 0
VOP!
PK_3 C:123153 H:5
BT C:2064, H:0
BT C:2064, H:0

S_0SR:0
BT C:2064, H:0
SR:1
BT C:2064, H:0
SR:2
BT C:2064, H:0
SR:3
VOP!
PK_1 C:123153 H:0
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
UF6 GE
AP:2 C: 0->156588 H:0->8 AP Cnt: 0
VOP!
PK_3 C:156416 H:8
UF6 GE
AP:2 C: 0->122958 H:0->9 AP Cnt: 0
VOP!
PK_3 C:123101 H:9
UF6 GE
AP:2 C: 0->122880 H:0->0 AP Cnt: 0
VOP!
PK_3 C:124275 H:0
UF6 GE
AP:2 C: 0->122247 H:0->1 AP Cnt: 0
VOP!
------------------------------------------------------

Any tips on what the dump tries to tell me and if there is a way to get the combination of replacement Rom with original PCB to work ?


Thanks for your help,

Urmel

In short, No. The DIY has reach it's limit and you need to hand it over to someone that has the experience and proper tools...sorry to say.

_________________
Rescue IT Datarecovery service Sweden
Rescue IT Dataräddning Göteborg AB
http://www.rescue-it.se

Do you have any facility to read the ROM contents?

_________________
A backup a day keeps DR away.

Hello,

@mr_spokk:

This is of course an option, still have not decided how much money I would be willing to spend on a professional data recovery service. Based on the information posted, would it be a quick job for a professional?


@fzabkar:

Just had a look at the datasheet of the Winbond 25x40, it's an SPI based chip, so yes, I could hook it to an Arduino or something and read out the data. Is there some information on what data is stored in the rom, meaning which data range contains what kind of (unique ?) information about the hdd?


Regards,

Urmel

Hello,

can't edit my last post (as stated in the board FAQ), so had do a new posting ..... sorry.

Please be aware that I'm a bloody beginner to hdd trouble shooting, but it's a very interesting topic, so I'm quite eager to expand my horizon.


Besides removing the rom and read it out in an external electronics environment there should also be a way to read/write it via serial connection, at least this is what some posts on the internet claim. The "HE" output shows lots of commands, but it very hard to find any good information on the commands and syntax on the internet.
Anyone knows which command (and syntax) is for reading / writing the rom?

The posting in viewtopic.php?f=1&t=26366 shows a list of commands

The LB command causes the "[DEBUG HALT @MDL_RSRC_BufferAlloc():10034A41 L0 10039689]" error, on a good HD103SJ disk it shows several lines of information like serial number etc.

All other "L" commands (with my bad disk) gives some output, so my hope is that not all of the rom is bad. If helpful, I can create a dump of all outputs "L" commands and upload it.

Any suggestions on how to proceed?



Regard,

Urmel

Hello all,

spent the last days working on a python script to do the serial communication with the hdd,
finally I'm able to read memory data and store it to a binary file.

Searched the internet for a command list and found this:

D addr [num] Display Data memory
DP addr [num] Display Program memory
DB addr [num] Display Buffer memory
DW addr [Num] Word Display Puffer memory
DF? see manual
M addr Modify Data memory
MD addr <data> Modify Data memory
MB addr <data> Modify Buffer memory
MW addr <data> Word Modify Buffer memory
MP addr <data> Modify Program Memory

The DF-command (Display flash memory ?) is not working for the disk, all other Dx commands give some
reasonable looking output. Can anyone tell me which D-command is the right one for reading the rom on a HD103SJ?


Thanks,

Urmel

Hello,

in the meantime I was able to read the roms from the defective PCB and from the replacement PCB (see attachments).

As mentioned in the thread, defective.bin causes "[DEBUG HALT @MDL_RSRC_BufferAlloc():10034AB5 L0 1003A471]" while "china.bin" starts the drive (its also detected in bios) but results in head seeking noise without any access to data.


If anybody has a tip on how to proceed I would be very thankful.


Regards,

Urmel

Hello all,

besides the defective hdd I have another (but younger) HD103SJ, today I read the rom and tried it on my defective hdd, the rom of the "good" disk is attached below.

The serial output of the "good" rom on the "defective" disk give the following output:

*PA VID=0000 PN=0004 Rev=0002- 785x Found
*PA VID=0000 PN=0004 Rev=0002- 785x Found
ChipRev = 63B0
U
S_0
RV En Sensor Circuit
En Shock Sensor Circuit
SO_1
SpinStartUp: mcSpinRPM = 0
RPM at Handoff: 575
Temp : 16 degC
SpinOk
mS1 00000003
SK C:112415 H:0
Loaded FIT ( 0: 0: 1)
CalibTable Loaded. Rev:0x14
Selective MARC NX Loaded
ResoTable Loaded. Rev:0x01
Ovly loaded to 0x0001CD00
Ovly loaded to 0x1005B800

=> Result is not surprising as good hdd has a different manufacturing date:

Defective hdd:
"HD103SJ/D", Rev A, FW 1AJ1001, Date 2011.02
PCB: BF41-00359A 00

Good hdd: "HD103SJ", Rev. A, Date 2011.06 (no info about firmware on label)
PCB: BF41-00353A 00


Maybe the information of all three roms in summary is helping the error analysis ?


Regards,

Urmel

You are wasting your time (after 1 month...). Either you find a pcb that work (assuming there is nothing else) or you have to make one work.
The real problem is if someone is willing to help you the way you want.
P.S. when I see damaged or diy-beaten pcbs or when the customer say "IT HAS BEEN ALREADY TOUCHED/ATTEMPTED TO REPAIR" the pcb goes to the recycle box straight away, I assume I don't have a pcb then go to the second option abovementioned. Up today it always worked. So, instead of speculating and wasting time, if you really want to discover something useful, concentrate on solution(s) to do that. If I succeeded, it is possible.

Hi Stefan,
You can compare the contence of the ROM you read, and the ROM from the donor Pcb.
But as said earlier, DIY is over. If you care for your information don't play with it anymore.


Regards/ Bosse

_________________
Rescue IT Datarecovery service Sweden
Rescue IT Dataräddning Göteborg AB
http://www.rescue-it.se