Hi everyone,

New user here. I'm a newbie at recovery. I have some basic knowledge in electronics as I studied electrical engineering for one year before switching to mechanical engineering.

So the two drives below come from my previous machine in which all components died when the PSU itself blew a cap for an unknown reason . They've been sitting there for 3 years, I knew someday I would do something about them. Today I was about to send the PCBs to have their firmware transferred on replacement boards, but I figured I might research more. Then I found a few threads that got me worried about the preamps

Drive #1 :
WD800JB-00JJA0. From the picture, U1 is burnt. What's on the IC body is molten/burnt foam from the protector foam pad. From my searching on this site, U1 would be the motor controller.

Drive #2 :
WD1000BB-00CCB0, U1 and Q1 are both burnt. Although I couldn't confirm this by searching the forum, my guess is U1 would be the motor controller here as well, judging from the traces that go to the motor contacts. As for Q1, I would guess it's a transistor but I have no idea of what it actually does.

My questions :
1- What are the odds that the preamps on the heads might be dead? That would mean a PCB swap + ROM transfer using an identical PCB number would not be sufficient.

2- Is there any way to find out if the preamps are good, hopefully using a multimeter, or even an oscilloscope?

Drives details:

WD800JB-00JJA0.
Date: 28 nov 2004
DCM: HSBHYTJCH
PCB: 2060-001292-000 rev. A
Thailand

WD1000BB-00CCB0
Date: 20 Aug 2004 M
DCM: RSBABT2CH
PCB: 2060-001092-006 rev. A
Malaysia


Thank you all in advance!

After researching more, I found a few posts about checking the preamp.

1- Pins to probe

I found the below picture in a thread on the WD800 over at a russian forum. The pinout in the picture seems to make sense if I refer to N.C.'s post from 2008, here.

Q: Anyone can please confirm the pinout in the pic?


2- Readings

As per pepe's 2007 post here, the negative lead to the -5V should read infinity, or else the preamp is dead.

Q1: What readings on the +5V pin are expected of a good or dead preamp?
Q2: Does it make any difference if I ground to the ground on the heads connector or to the chassis through a PCB screw hole, i.e. potential damage to the preamp or else?


Next I'm going to verify what current my DDM uses for diode measurement, after reading dobrevjetser's warning about measuring current and preamps here.

I'm learning a lot thanks to you guys and this great forum. Hopefully the preamps are gonna be ok and I can go ahead and buy the PCBs and finally retreive the data.

Thank you all very much in advance.

A typical DMM uses about 0.1mA on the 200 ohms range and 1mA on the diode test range. The open circuit voltages would be 0.5V and 3V, respectively.

In short, this widespread fear about preamp damage is misplaced, as was shown in that "Samsung" thread. It's just another urban legend that has taken root in the data recovery community. You just need to take the usual ESD precautions, as the heads are easily damaged by static discharge.

As for the preamp connector pinout, it looks correct to me. The supply rails are normally identifiable by the larger traces, as are the two voice coil traces at the other end. You can easily confirm the +5V and ground pins by testing for continuity with the Molex power connector. AFAICT, on the WD800JB board the -5V supply should come from the anode of diode D4, whereas on the WD1000BB board it appears to come from the anode of D7.

I notice that on the WD800JB PCB, the transistor at Q8 appears to be blistered. I believe Q8 is the pass transistor in a +3.3V linear regulator. If so, then you would have to be concerned about any device that is connected to this rail. Unfortunately this includes the flash memory at U12. To this end I would test for a short between ground and the emitter of Q8.

BTW, the following articles may be of interest to you.

http://www.users.on.net/~fzabkar/HDD/HD ... Flash.html
http://www.users.on.net/~fzabkar/HDD/HD ... ntrol.html
http://www.users.on.net/~fzabkar/HDD/HDD_Preamp.html
http://www.users.on.net/~fzabkar/HDD/HDD_RAM.html
http://www.users.on.net/~fzabkar/HDD/HD ... annel.html
http://www.users.on.net/~fzabkar/HDD/HD ... _Regs.html

Tutorial - how to locate the basic test points of an unfamiliar HDD PCB:
http://www.users.on.net/~fzabkar/HDD/Tu ... 0411N.html

This thread shows the test points for a 2060-001113 PCB:
easy-faultdiagnostic-troubleshooting-must-try-t12319.html

_________________
A backup a day keeps DR away.

Wow thank you for the links to your pages, lots of info!

For the record, my DMM is an Equus 4230, spec says 2.8V open circuit voltage.

Sadly didn't have time to investigate yesterday, I'm at work right now but I will take the time to do it later today and will report results.

Alright, here are the results.

WD1000BB
-5V
red to GND, black to -5V : infinity
black to GND, red to -5V : 586
+5V
red to GND, black to +5V : 660
black to GND, red to +5V : infinity


WD800JB
-5V
red to GND, black to -5V : infinity
black to GND, red to -5V : 621
+5V
red to GND, black to +5V : 716
black to GND, red to +5V : 1330 ... hmm that one doesn't look right, does it?

As for Q8. If I'm not mistaken the emitter is the top left pin on the picture (the one right next to the "Q8" designator), then there is continuity between the emitter and ground . Is there anything I can check on U12 itself ?

At least I'm getting somewhere!
Thanks

Mmm, I double checked my readings and it seems I'm now getting a reading of about 1000 while probing black to GND, red to +5V on the WD1000BB. Maybe the probe was not contacting the pin correctly the first time around.

One other thing I noticed, the non-infinite readings (i.e., all except red to GND, black to -5V) seem to drop over time, is this normal?

I don't know what to expect for your -5V and +5V resistance measurements. However, it's clear that there are no short circuits or excessively low resistances, so it looks good to me.

However, if the emitter of Q8 (which you have correctly identified) is shorted to ground, then this bodes ill for U12. I would confirm your result by measuring the resistance between pins 4 (Gnd) and 8 (Vcc) of U12.

_________________
A backup a day keeps DR away.

I get an initial reading of 1.0 ohm between Vcc and Gnd on U12, and then the reading slowly drops to 0.9, 0.8, etc..
Does that mean U12 is completely shot?

Power on each PCB by itself (if you can) and if the discrete components below the motor ICs are blistering hot (particularly with the WD800JB), then the preamp(s) are shot.

Looking at the damage on the PCBs, it doesn't look good.

_________________
http://www.datasaversllc.com

Your measurement confirms that something is shorting the Vio supply. The culprit could be any or all of the serial flash memory at U12, the hynix SDRAM, or the Marvell 88i6540-LFH MCU. I would carefully desolder and isolate pin #8 of the flash IC and retest it for shorts.

If the IC tests OK, then you may have a chance to transfer it to a donor PCB.

The following board supplier includes a transfer service for US$10:
http://www.hdd-parts.com/firmware-transfer.html

_________________
A backup a day keeps DR away.

Yes, what you're saying makes perfect sense. I'm going to look for a local shop with a hot air gun.
Besides, I had contacted HDD-parts already and was going to send them my WD1000BB board this weekend.

Again, thanks for all the great help!

Hi guys. I finally got the replacement WD1000BB board with the copied firmware, from HDD-parts.
I noticed the replacement board revision (-007) differed from the original (-006), and that the chip at Q1 was missing so I fired an email to HDD-parts, who responded that the boards were compatible.

So I screw the new PCB down to the drive, install and boot. The disk starts spinning and then after maybe 7-8 seconds, I start hearing 4 short clicks... then silence, then 4 short clicks again, then silence again. After a couple more clicks, I shut the machine down in fear that the drive was dying.

I searched and found the "Noise related to PCB in WD HDDs (Final Edition)" sticky, and now I'm worried
Should I try to boot again and see if it still clicks or would that just destroy the drive further?

I tried placing the jumpers in safe mode, but the drive wouldn't be recognised by the BIOS.

I put the jumper back to the master setting and powered up, with the MHDD boot CD in. I tried accessing the BIOS but at that point either the machine had frozen or was taking a very long time to get to the BIOS, at which point I shut it down (didn't want the drive to spin for long for nothing).

Does that sound like a PCB or a firmware problem?

Assuming the board is compatible and rom transfer has been done correctly, clicking - spin down - spin up - clicking - spin down means most probably preamp is dead.

_________________
http://www.northwind.gr
SandForce SSD Recovery
Ransomware Reverse Engineering - NoMoreRansom! partners

There is a very high probability of preamp failure.

If so, you've exhausted the options for DIY recovery. Time for pro help if you need the data.

_________________
http://www.datasaversllc.com

@mach128x:

Looking at the above work done by other members:

a) you didn't reply with the results of the recommended test by jono-ats to check for one symptom of preamp damage.

b) you didn't reply with the results of the recommended test by fzabkar to check whether your original U12 was damaged and causing the shorted Vio on your original board

So in addition to the possibility of a damaged preamp, it also seems possible that your lack of success might be because you've transferred a faulty U12 onto the new PCB.

@Vulcan:

- Which components would that be, C42 to C45?
- U12 is on the WD800JB which I haven't tackled yet, in my latest posts I have focused on the replacement PCB and rom chip swap (U2) of the WD1000BB.

Thanks for explaining - the mixing of 2 drives in one thread had confused me about which problem was currently being dealt with, but I can see this now after re-reading. I'll leave you to it.