HGST Z7K500-500

[Zorb]

Am I missing something or would it make more sense to swap the platter than the heads on this drive? It seems to have a lazy head, alternating between reading great and reading horrifically slowly but with no errors, about every quarter million sectors. Imaging head by head gave me a perfect read of one surface, with the other so slow that it will probably take over a month.

The datasheet I just grabbed shows only a single platter. Just for shits, I just tore apart a semi-broken (recovered) Z7K320-320, and found that the platter can be removed and reinstalled repeatedly (did it four times) and the drive still works. To top it off, I did this at my desk with some incredibly basic hand tools because I haven't finished pulling my cleanroom gear out of mothballs yet! I haven't even touched the physical side of a hard drive project in a while and this was still insanely easy.

Anyone want to give me some reasons I would rather swap the heads if a media swap is so much less of an apparent headache? This seems kind of like the old Maxtor N40P, which only used the bottom platter surface anyway, and a monkey could swap the platter in five minutes.

[Zorb]

This worked perfectly. Swapped the platter and PCB with the heads securely on their ramps. Now, I have a couple of ideas to refine this process. Approximately how far inward is the outermost track from the edge of the platter? I have a tool idea and I want some input in order to refine it a bit.

I used aluminum tape to grab the edge of the platter in four locations, then used a tool to grab the center once it was sufficiently lifted off. It worked, but felt more precarious than I was really happy with. I am working on a design for a metal or plastic tool to do this job very securely.

[pepe]

As a rule of thumb i never move platters if there is nothing wrong with bearing or other significant indication. Just because if anything goes wrong, MHA is replaceable, but the platter?
I am saying this while moving platters all around on a weekly basis but only when it is absolutely required by the recovery process.

[fzabkar]

https://groups.google.com/d/topic/datarecoverycertification/1xh-hbJrocE

[pepe]

yeah.
i swapped pcb on an ssd with sandforce last week. worked like a charm.

[Zorb]

The post on Google forums is beyond idiotic. No sane individual would expect a platter to work if reversed. This person should be Westinghoused for his stupidity.

Also jokes aside, data recovery shops shouldn't even play with Sandforce based SSDs, if only as punishment for stupid users buying them.

I found a single platter substantially easier to move than the head assembly. If I can figure out what I need to know to build this tool, it will make it substantially less risky than it is now, and much less risky than moving the head assembly.