Seagate ST3000DM001 spins up, DRD&DSC ok, but ID causes BSY

[ketil]

I have a 3TB Seagate ST3000DM001, PCB 100687658, SATA drive from an external USB drive enclosure. Here's a picture of the PCB:

https://drive.google.com/file/d/0B947HP ... sp=sharing

When I power on, the disk spins up, and reaches DRD & DSC ok. No special sounds. If I try to ID the drive, it just gives up and the drive gets stuck in BSY state. I have to power off and on to get out of BSY state, reset doesn't help. I unscrewed the PCB from the drive and connected it to PC3000, and then it quickly reaches DRD & DSC, and then drive ID is no problem. But in this case, the size is 3.86GB, which is wrong because the drive is really 3000GB. If I start generic F3 utility, I can watch what happens on the COM port. Then I power down, reconnect the drive, and power up again. Over the COM port I can see that the drive powers on and spins up ok. Again, if I try to ID the drive, it hangs in BSY state and I have to powercycle the drive. I tried increasing timeout for ID from 10 to 120 seconds, but it still doesn't work. I tried to test heads, but the command seemed to fail.

I was able to download the ROM when connected to PCB only. I've also tried editing RAM and downloading service area, but it always fails.

When the drive is connected to a Linux computer with USB-to-SATA connector, and the drive is recognized as a 3TB drive, but after a while it is offlined, "not ready after error recovery".

Any suggestions for recovery?

[Nick_TS]

Hello Ketil,

Can reach terminal when PCB is setup onto HDA?
If yes then send ud terminal log.
Also tell us what type of terminal adapter do you use?

[ketil]

I get serial access, but only if I start the PCB detached. Then while the utility is open, I can power down, connect PCB to HD, and power up again. Now I see serial output, and everything looks good, no error messages. I can send you the output tomorrow when I have access to the drive, if it's interesting. If I try to ID the drive from the utility, it also hangs in BSY state again, and I have to powercycle to get it out of BSY mode.

[AJ2008]

Probably it has some SMART error. You will need to clear SMART over terminal

[ketil]

Do you mean using a procedure something like:

/1
N1

followed by something like:

m0,2,2,,,,,22 or a variant?

Or do you mean to follow the 'Solution for the "HDD Lock (LED: 000000CC)" problem' method from the F3 manual?

[AJ2008]

Do you mean using a procedure something like:

/1
N1

followed by something like:

m0,2,2,,,,,22 or a variant?

Or do you mean to follow the 'Solution for the "HDD Lock (LED: 000000CC)" problem' method from the F3 manual?

N1 will clear smart, but sometimes is not possible without 'm' command first, but in this case you have the risk of loosing entries from non resident tables.

If 'N1' alone does not work or gives you LED error, check with V40 if you have entries that need to be preserved before you attempt any other methods for it

[ketil]

N1 will clear smart, but sometimes is not possible without 'm' command first, but in this case you have the risk of loosing entries from non resident tables.

If 'N1' alone does not work or gives you LED error, check with V40 if you have entries that need to be preserved before you attempt any other methods for it

V40?

I'm willing to try this, followed by Translator Regeneration, if that's my only option. I can't read anything from the drive, it just reaches BSY state any time anything is attempted.

Can reach terminal when PCB is setup onto HDA?
If yes then send ud terminal log.

Here's the terminal output when I power on the drive:

Code: Select all

Boot 0x40M
 Spin Up
 Trans.

 Spin Up
SpinOk
(P) SATA Reset


[ketil]

I tried V40, and that list was empty:

Code: Select all

F3 T>V40
Nonresident GList                    0 entries returned
Total entries available: 0
          PBA           Len  Flags  Phy Cyl Hd  PhySctr    SFI



so I guess m2,.... is ok.

[AJ2008]

Yes. Or at least it wont make it worse, but this depends if there was a NRG previously that is somehow cleared.

[ketil]

Thanks for the help, after

Code: Select all

/1
N1

then power cycle, and then

Code: Select all

m0,2,2,,,,,22

followed by another power cycle, I can now get the drive to ID fine. I was able to read about 5 million sectors, and now I get UNC on all sectors, so I guess I have to try translator regeneration. Will try that and then try Data Extractor again.

[AJ2008]

I was able to read about 5 million sectors, and now I get UNC on all sectors, so I guess I have to try translator regeneration.

I guess NRG was cleared previously. Try translator recovery, not regen

[ketil]

Actually, I jumped to a conclusion too soon. It seems to be some bad sectors only. I disabled the current head briefly while I got errors, and it started reading successfully again at the next head. Now I'm at LBA 40 million, and getting mostly success but also a few failures (with some noises from the drive).

I didn't realize translator regeneration and recovery were two different things. I guess I have to read the manual to find out about that, can't see recovery anywhere in the menus.

[AJ2008]

I tried to add attachment here showing location of 'Recover Translator' but board wont let me add it:

"Sorry, the board attachment quota has been reached"

but it is located under:

Work With Service Area -> Translator -> Recover Translator

hope this is useful

[ketil]

OK, I see it now! Thanks again.

[ketil]

I've been working at this drive some more, and I got lots of trouble with the drive after reading a little bit of data. After reaching LBA 40 million or so, DE just tried to read each single sector forever. I tried playing with settings, and changing to read using active utility, but nothing helped.

So I decided to clear SMART again and try once more. This time when I ran V40, there were 58 defects on the list. I saved the defects with PC3K, and did the same procedure to clear SMART. After doing that, there was immediately 78 defects in the NRG. Strange, but ok. I just left these 78 in there.

I still had trouble with my DE task, so I ended up creating a new one. This one read everything much better (using UDMA-133), I had more trouble and errors with the first DE task. But again at LBA 40720238 I got some trouble, and everything after that failed. So I figured this was a defect. I ran the "Single LBA-position adjustment" and added this sector there. After thinking a bit, PC3K said:

Code: Select all

Start scanning...
Checking for recovery, LBA = 40720238 - 0 = 40720238
Access gained, LBA = 40720238!

That looked good! So I started acquisition again, but it still failed. Then I changed DE reading mode to use Active PC3K utility, and things are working well. So far I'm at LBA 4813000, reading from active utility is sloow... Do I have to use this reading mode after adjusting LBA, or is there a chance I can change back to UDMA-133 at some point?