HGST HUH728080ALE604 Heads case

[einstein9]

Dear experts,

I got a G-Speed Studio 32TB (4* 8TB) case with one drive issue (heads) typical mistake of RAID0
if anyone have an experience with such cases let me know by PM with the estimated time/cost for this case.

Hope someone here did work with those drives.

Thank you

[einstein9]

More details about it just in case

Attachments

[einstein9]

no one yet?

[pcimage]

I don’t know anybody who wants to touch these things

[northwind]

I believe the main problem is that even if you solve the He problem, and even if you manage to successfully open and close the lid and change the heads (which is VERY challenging), the FW is signed and you have no access to SA.

[pcimage]

I believe the main problem is that even if you solve the He problem, and even if you manage to successfully open and close the lid and change the heads (which is VERY challenging), the FW is signed and you have no access to SA.

Exactly so

[Doomer]

I tried to do this model a couple of times, was not able to complete it successfully
First one developed SA issues which I was not able to solve w/o a loader which supposed to be signed(and I don't have it)
Second one wouldn't start because He sensor kept complaining about low He volume.

I was able to start imaging previous model (6TB He) but it MD-ed pretty quickly

[einstein9]

I don’t know anybody who wants to touch these things

looks so really...

I believe the main problem is that even if you solve the He problem, and even if you manage to successfully open and close the lid and change the heads (which is VERY challenging), the FW is signed and you have no access to SA.

I tried to do this model a couple of times, was not able to complete it successfully
First one developed SA issues which I was not able to solve w/o a loader which supposed to be signed(and I don't have it)
Second one wouldn't start because He sensor kept complaining about low He volume.

I was able to start imaging previous model (6TB He) but it MD-ed pretty quickly

Doomer, when you tried the 6TB, can you tell me how many donors you used? & % of reading you got? out of those donors?

I have 4 donors here (willing to proceed with this case) but wanted to know know if 4 is enough or not based on your experience?

Thank you all for sharing guys..

[northwind]

I might be wrong, but it's not a matter of # of donors. What I tend to believe is that without He properly infused inside the drive, the heads will cause Media Damage pretty quickly. If someone knows otherwise, I'd like to know too.
And, as I said, this is just the tip of the iceberg.

[Doomer]

Doomer, when you tried the 6TB, can you tell me how many donors you used? & % of reading you got? out of those donors?

I used one
If you just change heads it's not gonna work (even if you try 4 different heads), that's something to R&D for you
And I put He inside before I powered up the drive (it won't work w/o He, I tried that too, killed a donor during the process). You'd have to R&D the process.
I don't remember how much reading I've got but it was not enough

[fzabkar]

If you just change heads it's not gonna work (even if you try 4 different heads)

I guess this means that the donor's head adaptives need to be patched into the patient's ROM?

[fzabkar]

I have 4 donors here (willing to proceed with this case)

If you upload the 4 ROMs, I can try to make sense of the adaptives for you. I have written a tool which extracts these components from the ROM.

[einstein9]

I might be wrong, but it's not a matter of # of donors. What I tend to believe is that without He properly infused inside the drive, the heads will cause Media Damage pretty quickly. If someone knows otherwise, I'd like to know too.
And, as I said, this is just the tip of the iceberg.

I know that, i was asking to see the chances i have here with this case.

Doomer, when you tried the 6TB, can you tell me how many donors you used? & % of reading you got? out of those donors?

I used one
If you just change heads it's not gonna work (even if you try 4 different heads), that's something to R&D for you
And I put He inside before I powered up the drive (it won't work w/o He, I tried that too, killed a donor during the process). You'd have to R&D the process.
I don't remember how much reading I've got but it was not enough

I know that those drives are made in Japan, when they do things they do it properly (really well).....
Something else here, I have a Japanese friend (DR field) who did a lot of similar work with good successful rate of reading the diff. between me & him is time (he did not but i do now here with this)

I downloaded this video here from hddsurgery https://www.youtube.com/watch?v=ANMtvYnI1gQ and saw it maybe 50 times along with the notes from my friend too
and i think i have a better chance here.

I have 4 donors here (willing to proceed with this case)

If you upload the 4 ROMs, I can try to make sense of the adaptives for you. I have written a tool which extracts these components from the ROM.

Thnx, but i think i can do that too,,, will upload it later 4u when the work is done (need some time here).


Thank you again Doomer & northwind for the info/feedback here

[DR-Kiev]

Are you sure , failed disk has heads issue? We are receiving them with most common problem -"motor stuck", which is different story, and have greater channces to be solvable.

[einstein9]

Are you sure , failed disk has heads issue? We are receiving them with most common problem -"motor stuck", which is different story, and have greater channces to be solvable.

yes iam

will proceed in few days with experts help (not in DR) here

[fzabkar]

Just FYI, the current ATA standard now has commands for non-destructively testing read elements. There is also a command for depopping heads.

If your drive supports it, and if you are prepared to sacrifice it for the sake of testing, then depop one head and see how the ROM is affected. Alternatively, dump the ROMs from two or more drives from the same family, but with different numbers of heads, or different head maps. Then the adaptives should start to make more sense.

[fzabkar]

HGST's helium drives support the new ATA standard's Rebuild Assist feature set. This allows the drive to test and report the status of its physical elements. AIUI, the Rebuild Assist log can then be used to disable selected heads via a bitmap. Any read command issued to a disabled head would immediately report an error, without attempting an actual read.