I have this 1 TB Desktop model and the power receptacle hasn't worked right since day one. I'm out of warranty and now it looks like the pcb has some cooked spots. TO me this looks like where the power attaches to the board. The red/black solder points. It smells liek cooked trons too.

I've attached a picture of the board and close ups of where I think the damage is located.

Can I get a replacement board?

Are there repair kits availabke?

Is it possible to mount this and copy the data with out rebuidling the RAID 0?

Thanks

S

Other than the smell, what are the symptoms?

Have you attached the two drives to a SATA port on your computer's motherboard, and was BIOS able to detect them with their correct model number and correct capacity?

Do you have a multimeter?

_________________
A backup a day keeps DR away.

Thank You

I think one drive may be in bad shape too. The other drive has partition information available however Windows wants to format as soon as I plug it in to a Windows box. I havent tried Linux yet. I'm using WINHEX to recover data.

Yes I have a Fluke 77/AN.

I would think that WinHex should be able to reconstruct your RAID, assuming that the two drives are logically intact. The reason that Windows wants to format the drive that has your partition information is that half of the file system is on the other drive. This means that the boot sector is not where the partition table says it is. Furthermore, the partition table was built for a 1TB drive, not 500GB.

The second drive has no partition information, so it would be seen as uninitialised.

In Linux you could use mdadm to reconstruct your RAID as a software RAID. In Windows you also have DMDE and RAID Reconstructor.

As for the PCB in your enclosure, does it spin up both drives, or is it dead? If it's dead and you wish to troubleshoot it, then we would need a higher resolution photo. It would help also if you could identify any chip markings which are not readily discernible in the photo.

_________________
A backup a day keeps DR away.

Here's some more pictures per your last request. Please let me know if you need more. Thank you for helping. I'm standing by for direction.

S

You didn't answer my question regarding the symptoms.

However, I see that one of the two APM4532 dual MOSFETs (Q9) is visibly damaged. AIUI, these devices are controlled by the Oxford Semiconductor bridge IC and they switch +12V and +5V power to the drives.

I suspect that each package handles one particular drive, in which case I suspect that only one of the two drives spins up. Am I correct?

Assuming that's the only problem, you could try to source a replacement chip (or an equivalent), or you could hardwire the circuit so that the drive spins up automatically, and remains on, whenever power is applied.

APM4532 / APM4532K, Anpec, Dual Enhancement Mode MOSFET (N-and P-Channel), N-Channel 30V/5A, P-Channel -30V/-3.5A:
http://www.anpec.com.tw/ashx_prod_file. ... M4532K.pdf
http://www.datasheetcatalog.org/datashe ... 0443_1.pdf

_________________
A backup a day keeps DR away.

I havent applied power since this unit started smoking. I wasn't planning to apply power until the unit is repaired.

I would like to power with a i/o sw, not from the same plug and pray the way I did becore if that's possible. So if a power switch is cheap and easy to install I would prefer that. This way I plug in the drives and power with a I/o sw to use in case of bad results. I'm not too worried about the cosmetics.

I can't tell which drives spun up since I didn't turn back on. I can tell you one drive was zapped including the data. Both spin up now when connected.


What's the best way to recover data on RAID 0?

I replied but it hasn't posted yet.

Once the power went I didn't turn the unit back on. One drive also was zapped and it looks like one has data. I wouldlike to bypass the power and connect a i/o switch to control manually. what is the best way to do this and what power supply should i use?

Thank You

You say that one drive is "in bad shape". I see no evidence of that. In fact you didn't tell us whether BIOS had detected it with its correct model number and capacity, in which case the drive would likely be OK.

In any case, I'm thinking that the best solution would be to replace the faulty chip. I say this because I don't know how the bridge IC spins up the drives. It could be that it implements staggered spinup, in which case it would spin up one drive, wait for it to become ready, and then spin up the second. In this way it could minimise the startup load on the AC adapter. This is only pure conjecture, though. Instead it could be that the adapter may be able to tolerate a short overload.

Anyway, you would still need to remove the IC switch in order to run wire links between its inputs and outputs. You may as well replace it. I would visit the Digikey or Mouser web sites and perform a parametric search on dual MOSFETs in 8-pin packages, one N-channel, the other P-channel.

For example, here are three parts that are used in other WD externals:

AO4616, Alpha & Omega Semiconductor, 30V Complementary MOSFET, N-Channel 30V 8A, P-Channel 30V 7A:
http://www.aosmd.com/pdfs/datasheet/AO4616.pdf

AO4620, Alpha & Omega, Complementary Enhancement Mode FET, N-ch 30V 7.2A, P-ch -30V -5.3A:
http://www.aosmd.com/res/data_sheets/AO4620.pdf

AO4619, Alpha & Omega, Complementary Enhancement Mode FET, N-ch 30V 7.4A, P-ch -30V -5.2A:
http://www.aosmd.com/res/data_sheets/AO4619.pdf
http://aosmd.com/res/markings/AO4619.pdf (marking description)

I suggest you compare their specifications and pinouts against your APM4532 device. Perhaps one or more will be a suitable replacement. Let us know what you find.

Here are links to Digikey and others:
http://www.google.com/search?q=AO4616+O ... OR+digikey

_________________
A backup a day keeps DR away.