Before attempting to move the ROM chip to a donor PCB, I'm hoping to figure out exactly what chip is bad (as it may be a simpler fix than moving the ROM).

When powered on using an external power supply, the PCB and PSU both "tick" for about 5 seconds. Green LED on PCB dims with each "tick" then shuts off completely after 5 seconds.

Looking for some guidance (fzabkar?) in determining where the problem is occuring.

Thanks in advance for any help!

Brian

_________________
$300 Data Recovery - Los Angeles Data Recovery

I haven't been able to make sense of everything, but I've marked those voltage test points that I would check.

It appears from the datasheets that the MCU, flash memory, and SDRAM require 3.3V Vio. I expect that the Vcore supply would be around 2V +/- 0.5V.

The read channel IC appears to require its own 2 supplies.

There is also another switchmode regulator, possibly a negative supply for the preamp.

The area that I've marked as unknown appears to be some kind of charge pump.


Product Manual:
http://www.seagate.com/files/support/di ... csi_pm.pdf

88C7500M-LAE1, Marvell, Read Channel transceiver (PHY):
http://web.archive.org/web/200211251503 ... 8C7500.jsp

MC34072, High Slew Rate, Wide Bandwidth, Single Supply Operational Amplifier:
http://www.unisonic.com.tw/datasheet/MC34072.pdf

NJD2873T4G, ON Semiconductor, NPN Power Transistor, 2A, 50V:
http://www.onsemi.com/pub_link/Collater ... 73T4-D.PDF

MT48LC2M32B2-6G, Micron, 512K x 32 x 4 Banks, 3.3V, SDR SDRAM:
http://www.micron.com/~/media/documents ... _sdram.pdf

M29W400DB, STMicroelectronics, 4Mbit (512 Kb x 8 or 256 Kb x 16), 2.7V to 3.6V:
https://www.micron.com/~/media/document ... 9w400d.pdf

FDC655BN, Fairchild, Single N-Channel, Logic Level, PowerTrench MOSFET, 6.3A, 30V, 25mohm Rdson:
http://pdf.datasheetcatalog.com/datashe ... 0854_1.pdf

_________________
A backup a day keeps DR away.

Does the PCB on its own cause the PSU to tick?

_________________
A backup a day keeps DR away.

Yes. Without being connected to drive, still ticks. Donor PCB on patient drive spins up and sounds normal (but won't ID, since ROM not original). We are trying to walk-through and measure points above. Thanks for your help fzabkar!

_________________
$300 Data Recovery - Los Angeles Data Recovery

ISTM that there may be a dynamic short that appears only when the PCB tries to switch on some part of the circuityr.

I would start by measuring those test points on a working PCB. Once you have that information, I would test for shorts at the outputs of the identified onboard supplies. I would need the voltage measurements before I could proceed.

_________________
A backup a day keeps DR away.

I should add that some PSUs will tick if they are lightly loaded. A switchmode PSU requires a minimum load to regulate properly.

_________________
A backup a day keeps DR away.

Brian ,
Its Too Easy To Remove The ROM And solder To Another Donor PCB Without All Such Hassel .Kindly Do The Following [ Let Me Know if you need a soldering trick tutorial " i was planning to make a video soon ] .

1 : Remove ROM From Donor/Patient With Hot Air
2 : Clean The Pads Of the Donor With Wick where ROM is To Be soldered [ So that they are leveled to pcb level ]
3 : Do the same with flux and soldering iron to legs of ROM [ Just with Iron Not With Wick ]
4 : Now the legs of the patient rom + donor pcb is leveled
5 : Brush Some Flux to the Pads of Donor PCB And Place The ROM On It
6 : Then Clean Your IRON tip And put little solder on it and slide it by touching all the pins slowly [ With This trick you can have factory grade soldering in the ROM try it
7 : Let me know if you experience any issue might be i can make a video of the same

_________________
Regards
Amarbir S Dhillon , Chandigarh Data Recovery Labs [India]
Logical,Semi Physical And Physical Data Recovery
Website-> http://www.chandigarhdatarecovery.com

@fzabkar:

Ok, results in Red text are from "good" PCB. Any results that differ on "bad" PCB have Yellow highlight.

It's possible I made things worse while testing, because now the light doesn't come on at all. Arg! If I did make it worse, I think it happened when I was testing the upper-left chip in the "unknown.jpg" (which had 12, 17, 17, 22 volts on the "good" drive). At first these chips had voltage, then suddenly nothing - so I had to re-measure them all, and they pretty much all went to 0.

@Amarbir:

Thanks for the tips! I actually have some flux and a soldering iron on the way just for the reason you mentioned below (I saw almost exactly what you are describing in a video tutorial). We have never needed an iron as the hot air gun has worked well, but it does seem to be crucial to get rid of excess solder when used with flux + wick on these big chips.

_________________
$300 Data Recovery - Los Angeles Data Recovery

ISTM that the problem may be a shorted 12V TVS diode. Did you actually test it?

PC PSUs will shut down and stay down in the presence of a short on any of their outputs, but other supplies will sometimes hiccup when driving into a short. Yours may be the latter type.

_________________
A backup a day keeps DR away.

It tests good and to be sure, I swapped with "good" drive's TVS and problem remains (dead).

_________________
$300 Data Recovery - Los Angeles Data Recovery

ISTM that there is a 3A fuse (?) at the top left corner. Is 12V present on the connector side, or does it hiccup?

What resistance do you measure between ground and the load side of the fuse? Use the 200 ohms range.

_________________
A backup a day keeps DR away.

Not sure how to check resistance, but there are different readings for good/bad pcbs...

_________________
$300 Data Recovery - Los Angeles Data Recovery

Measure the resistance of the fuse, and then measure the resistance of the 12V TVS diode in-circuit. A good fuse should measure close to 0.0 ohms on the 200 ohms range.

If the diode tests as a short, then there will be a short circuit on the +12V supply. Since you have replaced the diode, this means that the short will probably be due to the T3, 35V, 15uF capacitor (tantalum ?), the motor controller, or something in the charge pump (VCM boost) circuit.

_________________
A backup a day keeps DR away.

If you confirm that the fuse is open, as it appears to be, and if you also confirm that there is no short on the load side of the fuse, then simply replace the fuse and give it a go. Perhaps the fuse failed due to fatigue. You could connect an ordinary glass fuse with lead wires.

_________________
A backup a day keeps DR away.

Connect the leads to the meter in the same way as for your voltage measurements. Then select the 200 ohms resistance range. Place the probes on either side of the component being tested. Make sure that the PCB is powered off during these tests.

_________________
A backup a day keeps DR away.

Unless I'm not understanding my multimeter correctly, manual ranges only include 6, 60, and 600. I did get a reading of .3 on 600. Anyhow, I moved the fuse and the problem remains. I want to thank you for your help fzabkar, but I'll be stopping efforts on this line of inquiry due to speed up this recovery. Instead, I'm having a friend at another DR company move the chip for us.

@Amarbir I did practice with your techniques and they worked but I don't feel 100% confident and cannot risk my customer's data, so I'll keep practicing and try that in the future.

_________________
$300 Data Recovery - Los Angeles Data Recovery

Do you mean "moved' as in replaced, or do you mean "removed"?

If the fuse reads 0.3 ohms, then it is good. That then begs the question, how can you measure 12V on the incoming side and 0V on the load side? If these measurements are correct, then the component could be a polyswitch, ie a resettable fuse. In this case there would be a short circuit on the load side. That's why I asked you to measure the 12V TVS diode in-circuit.

AISI, the problem has a trivial solution that doesn't warrant reaching for the blowtorch.

_________________
A backup a day keeps DR away.

"Moved" as in moved the good fuse from the good drive to the bad drive.

_________________
$300 Data Recovery - Los Angeles Data Recovery

Sir ,
Will Shoot That Video On Sunday Or Today At Office

PS : Forget all of this just buy a new pcb and swap rom thats all

_________________
Regards
Amarbir S Dhillon , Chandigarh Data Recovery Labs [India]
Logical,Semi Physical And Physical Data Recovery
Website-> http://www.chandigarhdatarecovery.com