Hi,
I have a pair of Seagate 80Gb drives in raid0, bought at the same time from the same shop. They have both simultaneously died, at the exact same moment, and no longer spin up. They have almost identical PCB's, well the codes are all the same but a couple of compontents are different manufacturer.

7200.7
80Gb
ST380013AS
p/n 9w2812-301
HDA p/n 100275523
Config syy-06
Firmware 3.18
Date code 04247
Site code AMK

Nothing appears burnt or funny smells.
With power, the processor gets warm after half a minute. I can measure 5/12V across the protection diodes, so presume they are ok? They do not seem to do what I expect without power though on multimeter, this normal?
I cannot find any compatible boards on eBay - can anyone suggest things I can test with multimeter on the board or nything else??
Many thanks for any help or suggestions!

Yes, you should measure 5/12V across the diodes, but not across the fuses. Make sure you don't mix up the two. Measuring the diodes with the power off requires a Multimeter with a diode function check (Most do). Check one way, then reverse the leads. It should read around 0.5 one way, and nothing the other.

Have you tried one or both in another computer? How close are the voltages to 5.00/12.00? Have you hooked up the drives to a terminal? Are these in a computer, or a NAS enclosure?

Hi , thanks for reply, have tried with two different psu's and an external 12/5v supply from a USB<>IDE adapter, both drives dead on all.

Measured the 12/5v across a diodes marked 'QE RSL' and other on one marked 'FT3 PV46' - dont see any fuses? They are largest two black components nearest the sata power plug, with white stripe on one end (cathode)

QE 0.643 correct way , and 1.483 other way (presume this should be off scale?)
FT3 0.801 correct way and off scale other way after flashing 1.322 for a fraction of a second.

QE 0.620 correct way, 1.474 incorrect way (presume this should be off scale?)
FT3 0.791 correct way, and off scale other way, after flashing 1.285 for fraction of a second.

That sounds right. Have you tried powering without the SATA cable connected?

Or maybe, you just have crappy crappy luck....

Yeah that is powering on with no SATA connected.

I should perhaps add that they were originally connected to the raid controller on an EVGA 650i board. This board died, I think because the waterblock came off the northbridge and it fried (my mistake the tube got caught and pulled it slightly to one side so it rocked off the chip). However I tried the drives on an Intel board with raid controller, but it saw them as two separate Seagate drives. I figured this was normal as the Intel raid controller presumable couldnt see the raid set. However they did seem fine at this point. I then removed them and a few days later took them to a friends who had a similar board to get the data off them. When I arrived at his place they were both dead. They were both in Scythe silent boxes so we at all times protected and impossible to have bashed or shorted etc. I can therefore only assume plugging them in /out somehow killed them due to power issue. I imagine therefore a component is frazzled from unusual voltage/current, and nothing major has happened?

One of the things I can think of that fits the symptoms is Static Electricity. While the boxes would protect the PCB on the drive, it wouldn't protect the connector. If you live in the Northern hemisphere, it's probably warm where you are, and likely humid, so Static shouldn't be much of a problem.

Perhaps your friend's PS or RAID controller fried them.

Have you tried them back in your computer with the Intel board, with the same exact setup that let you see them individually?

Unfortunately, if you measured proper voltage across the protection diodes, it means that the diodes are not shorted, and that the fuses are OK. Those parts are the non-major components that often fry. If these are good, the problem is a bit more difficult.

Have you tried putting the drives up to your ear when powering them? Do they make ANY sound?

I was very careful against ESD, i'm almost 100% sure it wasnt that.
Tried on 3 different power sources and same.

I have just brought back a logic board from a 120Gb 7200.7 from work, its quite different in shape and P/N's are different but I figured it would be enough to spin it up to at least make sure motor is working etc. It indeed span up and clicked abit so I powered it off, I presume this is normal if its the wrong p/n and firmware.

I then put the old board back on. Powering it up there is a very feint noise of some sort. Powering it off it makes a click, a very similar noise to head parking. I presume this is as a result of the other board spinning it up and leaving it in a different state?

I am also attaching 2 pics.
1) while inspecing the board on one , it appears during manufacture two capacitors were stuck to the same part of the board? they are both firmly attached, I thought to begin with one must have come from somewhere on the board but appears not. Any ideas?!
2)second one is of the two diodes I tested.

Hi, chud

I read the original post, and i can say, this was a spike, comes over the PS. (something like ESD issue)
The seagate's preamp is really sensitive for voltage rising.
I suggest, to check the two preamp for short circuit, before trying another pcb.
(maybe only damaged, but not shorted)
If you are lucky, the eeprom stay OK on the pcb, but another parts are too hard to test.

I think you need to find the closest pcb, and swap the eeprom, and/or the MCU, if the preamp is OK....

Best Regards,
Janos

Hi Janos,
Thank you for your reply, can I test the preamp this way without opening the drive? I thought it was part of the MHA?
If it span up using another PCB, and made abit of a repetative ticking noise, do you think this means a correct PCB may fix it?
Many thanks,

Any ideas why 2 caps on one pad, centre of image, between Winbond and ST chip (please click on it to see bigger) ?? Anyone seen such a thing before?

This is only the manufacturer's solution.
Maybe price....

Janos

You need to use oscilloscope or special programs like pc3000 to test the preamp.
Don't open the drive!



Maybe, but can be a broken preamp too...

Janos

On the other drive is only one cap in that place, looks like there pick & place machine screwed up making it?

It could be. However, manufacturers often make design changes after a PCB has been designed, and try their best to fit the part on the existing board, instead of having a new design made up. Back in the old days, I've seen wires added, and other parts glued on.

Either way, it looks like a low value capacitor, and likely won't hurt anything, especially if the drive did work at one time.

The drive no longer spins up right ?? I have pcb which is compatible with the one u are looking for but u will have to change rom .

PM me if u need the pcb with ur location .

Hi rameez, have sent PM.
Am I correct in thinking the ROM chip to remove from my board is 8 legged one;
25P05AV
ST 9330B
its a 64K flash chip right?

Hi,

This is the EEPROM chip on the pcb, but you have no rom problem, if i remember right.
Why do you want to remove this?

Janos

could someone please clarify what is where on these Segates? I am now completely confused!!
firmware location, is it on: ROM chip (where is this on PCB?), 25P05AV flash EEPROM , on the actual platters ?
v3.18 refers to the version stored in what location?

I am currently under this impression; there is a ROM chip that with 'some' code that then reads the firmware from the drive platters. Does 3.18 refer to the info on the platters or this ROM chip. Where is this chip? What does the flash EEPROM have to do with this, if anything?

Many thanks to anyone who can give a definative explanation!

3.18 refers to the rom code .

Hi,
Well incase anyone ever stumbles upon this thread I thought I should post the outcome.
I managed to track down a logic board from SafeOnComputer, from eBay. He actually sent me two, one with almost identical codes, but with an Agere controller chip and one slightly different but with an ST chip. The ST chip one had a similar but different p/n, (I will try and edit this post with the details if I can find the bit of paper I scribbled it on before sending them back.)

Anyway, the Agere one span up but didnt detect in BIOS, the ST one (after resoldering on a tiny component that came off in transit!) detected and seemed to be ok.
To cut a long story short, I eventually used Raid Reconstructor to sector image each drive with this new logic board on, then mounted them in Captain Nemo as a single recontrcuted drive. I got the entire data set back (160Gb) bar 3 files. These all >1Gb files but were 0 bytes - so at least they were obviouusly not right rather than corrupted. Amongst everything I had 10Gb of RAR files to check that passed CRC test. All other things I could find appeared fine as best as I could test. I was almost 100% sure integrity had remained therefore.

Initially after trying the new board on a drive, I had tried replacing the old board, but it was still dead. This seemed quite normal.

Now the interesting part. After I had imaged both drives, I accidently connected the first one back up with its original dead logic board on. To my amazement it was working. I then put the old board back on the second too and that worked?! Now both drives were alive again. It appeared this only happened after they had both been hooked up with the replacement board for a few hours while sector copying (as I had tried changing to old pcb at the begiining after swapout and was still dead)
Anyone can explain this???

Anyway I bought a PCI Sil3112 raid card to try them on with the old pcb's. This was because my old nforce board used a Sil controller, and my new one was ICH9R intel. (Incidently raid recontructor uses drive on any channel so you can get data from drives from different raid controller, without needing any raid controller). They did pick up on it as a valid raid set, unfortunately they didnt work so windows could see them as a drive - it appeared the partition was damaged somehow. I spent quite a while trying to fix it but could not.

Just for completeness in the post - other things I did - I imaged the two sector copies back to a pair of 120gb drives, these also picked up on the raid controller but with bad partition. (probably not surprising as different size) - i used GetDataback for NTFS but didnt get much from them.

I would like to say a big thank you to SafgeOnComputer and www.runtime.org (& rameez who offered me a PCB but due to delay had allready got one on order from SafeonComputer -although that did cost me $59.99)