NC,

You bring up a very interesting point about head alignment. Since this is a single-platter drive I assume the alignment referred to is lateral... or the position of the platter on the spindle.

If this is the case, I'll need professional help.

I doubt the drive has been 'clipped'. I know Tivo is using a proprietary file system or partition system that linux cannot recognize (ext2, ext3, FATnn, NTFS). The Tivo drives I have imaged successfully all report the full capacity of the drive. Why it's showing half capacity is probably due to a corrupt system area.

-Bit

What was so important? Nothing really. It's a challenge and I love challenges. Some have called me a Renaissance Man because of all my interests and hobbies extending from programming micocontrollers, being a single-engine land pilot, to expert long-range hi-powered marksman to building a cabin amongst other things and the list is long. This happens to be my latest.

I suspect the tools you speak of are in the thousands of US dollars to acquire.

-Bit

... or the alignment is lost, and only one head can be used.....

Excellent!!! That makes the most sense so far.

The original drive reported half capacity before it died completely and before I opened it up. It may be that a head or preamp went bad and flooded the bad block table (on the drive) on one side of the drive... thus the calculated size is half it's original capacity.

my ar*e. If one head fails or gets disconnected, the drive clicks furiously or make the maxtor typical sound. Cut one head and try to init one of the donors, if you can. At present it is necessary to rebuild the drive , check sa integrity and then diagnose surface.

Okay BlackST.

Why do you think the original un-opened drive reported half its capacity before it failed completely? hmmm, that probably should have been my original post .

Now I'm thinking about building some hardware.

This would be an IDE switch. It would switch the 80 pin cable connected from one drive to another. I suspect we're looking at cmos levels not ttl levels for the out/inputs.

The idea here is to bypass the hot-swap spin-down which reads the SA after startup.

The process would be as follows:

1) Connect both drives to the imaging computer so both spin up on boot with the IDE switch installed.

2) After boot, use ddrescue to initiate imaging and before activating, switch the 80 pin IDE switch to the image drive.

Think this will work?

This is quantum technology.

This will work only on a good working drive, and this way you can only skip the drive recognizations, and therefor you will skip the transfer selection settings wich is important for the drive usage.

Additionally, the halfly initialized drive have no inicialized SA, the translator is not working, and reports AMNF for every sector access command.

in one word, this will not work.

Additionally: without the head alignment, you can't do nothing.
I have never seen any SA problem on this model, i am 99.99% sure, the SA is good.

Janos

Now I'm a bit upset that none of the videos I watched mentioned head alignment problems, even for a single platter drive and that it was easier to swap platters than to swap heads for a single platter drive. Oh well, live and learn I guess.

So, is there any way I can align the heads myself?

Well, i am using special tools wich is designed by myself for this purpose.
I think you can't do nothing on this one.

If you think, i can help, but only in my lab.

Janos

I think there is some misunderstanding regarding head alignment on this drive.

This is not a Western Digital drive where the heads fit loosely on the shaft when the cover is removed. The head shafts are of one piece, rigid and impossible to align.

The only alignment I can see is the lateral positioning of the platter which looks like about 0.001" on the spindle. I'm not sure if that is enough to cause an error, but not out of the realm of possibilities.

Given this criteria, do you all think head alignment is still a problem?

The platter gap, what you can see, is multiple of size of one track on modern drives so removing platter from 500G-1TB drive is a nightmare by re-aligning.
Fortunately this model have only 20GB /side density this is quite low, but anyway, can make some problem...
(maybe calibration problems, speed issues, some servo-lost, unstable writes, and so on...)

Your actual problem is the head alignment wich is unique on this drive, the problem is similar to WD models.

Janos

If I get the platter perfectly centered on the spindle, will that help with head alignment? Is 'centered' the factory specification?

I have indicators that will do this down to 0.0001"

Or, do I need a microscope to do this, if so, how much power do I need?

I estimate the track spacing to be about 0.0000218253" at 54,982 tracks in 1.2"

Problem is that for current drives the positioning in the factory has a bigger tollerance than the track-width.
So even if you center the platter perfectly, this is not a guarantee the tracks are centered since they are written after the platter is attached to the spindle in the factory.

I have no idea if this is also the case for the Quantum you are working with, but a microscope will not help you; rather a DDA (disk drive analyzer).


Dobre

_________________
Murphy was an optimist

Datarecovery in Belgium, Holland, France and Germany
Datarecoverytools http://www.drtools.eu

To cut a long story short - the keyword for these puppies is ADAPTIVE FORMAT. If you really think it was HEADS, without know how and HW tools - not mechanical - you won't get the drive ready. It took me about 1 year of fight , having all I would need, to undestand and work with the 540 and 740 family. Good luck and happy tinkering.