Hello,

I'm new at this and forcing myself to learn as a result of not having good backups. The data is important to me but not willing to spend and want to learn as my punishment. Yes, backups are important. This particular machine backups had to be manual, well, that will change.

Here's the details I'd really appreciate any help this forum is willing to provide.

Hello, I've recently experienced drive clicking upon boot and the drive is no longer recognized in the laptop and not visible in an external USB sata interface.

I had a lot of serial interface trouble and ordered a second adapter. But I do have an arduino UNO, the only thing last Saturday I could find at radio shack.

The first approach was using a held reset pin on an Arduino UNO, direct to RX TX TX RX didnt work and I think the resistor voltage divider wasnt close enough to satisfy the TTL serial circuit in the drive.

I finally got the second TTL adaptor and it too had initial trouble. For me it wasnt baud rate but a common ground problem. Once I grounded the TTL adapter to the drive serial ground and had a common ground I finally got clean output from the drive.

Enough on the TTL background... Its pretty well documented, some drive require very close voltage tolerances before the serial port will work.

I have the controller PCB and motor connections connected.

I open HT at 38400 8 1 n with the serial connected and when I apply sata power here's where I'm at:

I get either:

Rst 0x10M
LED:00000067 FAddr:003400D8
LED:00000067 FAddr:003400D8

or

Rst 0x10M
LED:00000067 FAddr:0000987C
LED:00000067 FAddr:0000987C

suggestions?


ctrl-z produces nothing...


I have 5 donor drives on the way from a working used lot with close date codes and will scrutinize the PCBs to see if there are any differences. From a great about of novice reading, I think there's only a 25% chance it has anything to do with a PCB error.

I read if it gets to that point the flash will need to be moved from the host to the donor PCB or the donor PCB wont have the correct adpative codes...
Does the adaptive data include head detail/anomolies or genreally not?


The drive when power is applied clicks 12 times and produces the above update.

The spin up sounds good and the drives continues to spin after the clicking stops.
The drive isn't recognized by BIOS and not on an external USB sata controller either.

I'm trying to get time in the clean room to open and look inside the drive.

Is this drive a likely patient for donor heads?

Is the serial terminal possibly locked, or anything else I can try in terminal?

--------------------------------
Thanks for community support
Paul

1) headstack damaged

2) rom damaged

3) pcb board damaged

it depends on the error code as it can be anything

im looking at 1

The drive spins up, so it can't be a ROM problem.

_________________
A backup a day keeps DR away.

corrupt some part of the ROM and the drive will spin up perfectly but won't init, and possibly with hiss and beep...

Isn't the ROM checksummed, and wouldn't a corrupt ROM fail a checksum test during the drive's POST?

_________________
A backup a day keeps DR away.

Looking around I'm not seeing any details on a checksum halting the spin up. Certainly many messages on if the controller is totally non functional the drive wont spin up and around a 20% with a spin up it may still be somewhere in the controller.

With all the research I'm really hoping its not the ROM as there no way I can rebuild a ROM with my current capability.

I could handle moving the flash from the original to a matched donor board.

But see some detail that indicates that to can cause problems.

A clean room head swap I think I can handle with some practice and a head tool.

I'm willing to put the time and practice needed to continue this effort.

Thanks for all the current replies and look forward to what other are thinking as I work through this learning experience.

Thanks,
Paul

Heads + media damage (most likely).

Concensus seems to be building around heads.

Thanks for the ongoing responses.

Best Regards,
Paul

With output coming from the terminal is there anything I can send to the drive to get more information?

Control-z doesnt result in a prompt?

If its the heads, I've been able to secure time in a class 100 environment.

With the 5 drives I have coming, Is there any way from the date codes to determine the most likely compatible head stack?

They are all from Site: WU and range in sate code from 10445, 2 x 10452, 11352, and an 11111, with the original drive being 11026.

I havent seen the PCB up close yet to see if any match my current drive.

I also want to rule out any incompatibles if there are any of those just to get some practice looking inside the drive and before I go into the clean room.

Looking closely on the PCB side of the drive there's a 100585743 on 4 of the drives and 1 is 100585745 near the square scan code. Of course my original drive would be a 10058575. Not sure what part that numbers refers to, but I only have one of those in the lot.

and the date codes are a little different.

all the PCBs are same part number Rev E


Looking at one of them in terminal:

Rst 0x10M
(P) SATA Reset

upon pressing ctrl-z I get:

ASCII Diag mode

F3 T>

Much healthier than my drive.

I practiced on one of the drives. Completely removed the heads and replaced that back into the same drive.

The custom head tool fit perfectly. I need to get a couple of different shaped tweezers to help w fine grained placement ability.

The drive passed boot and reset and came online on a USB controller and windows was happy the drive was healthy.

I wasn't in the clean room. Its my practice drive.

I was able to read the drive for a short time. Windows 7 is now telling me there's a problem with the drive.

I had to leave but I'll be back at it later.

Could simply air exposer alone cause the drive to error out within minutes?

As I said this is intended to be a scrap drive and had no data on it so I can practice.

I was very pleased the drive was recognized and of relative health give the open air surgery.


I'm going to pull the drive back apart when I get back to the office look closer at the drive. I had to run out for a while.

Yes certainly, particles invisible to most eyes can, and probably will, get between head and platter and destroy head. Think of a supermarket trolley and a tiny stone under the wheel on a VERY clean floor... remember the head does not run on the platter, but on a very small cushion of air, so the head needs to be positioned exactly right and nothing in the air gap.

Once you pull the drive apart, look at the heads under a microscope. It may not even be the heads, it could be as simple as the cover not screwed down to a proper torque, too tight or loose, even though I saw a thread about this recently, it is a possibility.

I like to think of tests with no controls as tests with no results.

I go into the clean room tomorrow.

Any tips that might be helpful?

This being my 1st drive I cant say I'm comfortable but be cautious to ensure familiar with the internal of the drive, the magnets and how things can move quickly if you lose track of the magnetic fields and its impact on anything that comes within its grip.

I have a question on the heads and the adaptive firmware in these drives.

Is there any significant difference in the heads that impacts any of the adaptive data?

I have 3 donor drives to work with.

Would a torque problem suddenly present itself on a drive that's been functioning fine for a couple years suddenly just appear?

I wouldn't consider the first drive a test but rather gaining experience inside a real drive to get very intentional on movements and the parts and sequence of removing the replacement heads from the donor and then installing them back into the drive.

If you pulled it apart, yes it is a possibility because you had to re-torque the screws. Just screwing them in is re-torquing them regardless of itf you measured it or not with a torque wrench.

well, I'm out of the clean room. I was finally able to view the platters on the original drive. The top platter is very unhealthy and heavily scored at one location about 1/2 way into the platter. The score is about the width of the head, for whatever reason, the drive doesnt travel but it took a drive into the platter.

I went ahead and replaced the heads anyhow and put the drive back together.

The drive with the donor heads spins up:

terminal reports:

Rst 0x20M

with Ctrl-z to do get to the prompt

F3 T>

The drive doesn't boot, nor is it recognized by BIOS.

Many thanks to the comments I've received.

I'd appreciate any comment or suggestions at this point in the process.

Maybe, drive BSY?

As I indicated, this is a learning experience and I'm not expecting to get any data back from the drive. If I do at some point great, but this is to further my knowledge into the depths of drive internals.

I back to the office, looking at the serial output from the drive:

Rst 0x10M
LED:00000067 FAddr:0000987C
LED:00000067 FAddr:0000987C

Any Comments would be greatly appreciated.

Heads traversing the damaged areas destroys them.

Unfortunately, game over.

_________________
http://www.datasaversllc.com

How dare the damaged platter eats the donor heads....


Well, A big thanks to everyone that's participated in this thread.

It's been a big learning experience and although I didn't retrieve my data from the drive. I learned a little about TTL, took on a side community project to make use of the initial Arduino I tried to get TTL to work on, a little something amount modern drive serial command and what drives look like on the inside.


As soon as I opened the drive in the clean room and saw the heavily scored platters, I knew then the game was very likely over.

RIP my damaged ST9250410AS... Zzzz...