Cross Compatible WD Families

Hello all,

Just to add to the Diablo3D / Diablo3S compatibility:

I've just been working on a failed Diablo3D. It was using a head-map of 0145 and I was struggling to find a suitable donor. I tried using a Diablo3S, with suitable DCM and microjogs. It provided some unusual results. It was possible to initialise the HDD to RDY, however, SA access was intermittent. If I blocked access to the SA (any method) then I could read Copy 0 modules. However, all the other heads were unresponsive. There was no write functionality for Head 0. If I allowed access to the SA the HDD would ID, but take a long time to read modules, sometimes it would click, and spin-down, then up then re-calibrate. Generally, it behaved in quite an odd way. I gave up, and swapped the heads back into the donor, and the heads still worked OK.

In the end, I used a 3TB Disablo3D which used 6 heads. With some physical alteration to the head assembly to account for the missing platter in the patient, it was possible to initialise the patient HDD on the first power on and read normally.

Best regards,
John
 

Jared

Administrator
Staff member
Sometimes heads, even of the same family, are a close enough match to read SA but not the user area. So I'm not surprised by your findings. I've had very similar things happen with same family, same model drives when just the date was a little too far off. I'd bet if you found a Diablo 3S with a closer date it'd work just as well as the 3D one.
 

Jared

Administrator
Staff member
Has anyone ever found a family that is cross compatible for heads with Helios?

I've got one here now I'm quoting and the only donors I'm seeing are all $300 and up, or have the wrong DCM (which from what I'm seeing on Donor Drives seems to indicate a different preamp vendor). I'm noticing Shasta 3D looks very similar and has same number of heads as well as a similar time period of manufacture. The aerial density is a couple hundred different, but that's not always a deal breaker.

Did anyone try this? Otherwise, I'll have to either shell out the money or hope that a different preamp vendor drive will still work.
 

Blizzard

Member
Jared":3cf1p7fe said:
Has anyone ever found a family that is cross compatible for heads with Helios?

I've got one here now I'm quoting and the only donors I'm seeing are all $300 and up, or have the wrong DCM (which from what I'm seeing on Donor Drives seems to indicate a different preamp vendor). I'm noticing Shasta 3D looks very similar and has same number of heads as well as a similar time period of manufacture. The aerial density is a couple hundred different, but that's not always a deal breaker.

Did anyone try this? Otherwise, I'll have to either shell out the money or hope that a different preamp vendor drive will still work.

Yes, I have at least 2 examples of using Shasta 3D for Helios donor here.
 

Jared

Administrator
Staff member
Thanks, I hadn't looked at your list in a little while. I added that pairing as well as the McKinley / Venus pairing I see there now.
 
HI

any Cross Compatible WD Family with Espirt ???
tried Europa it is compatible
but I need another family, running out of Donors & time to order a new one
any suggestion much appreciated
 

Jared

Administrator
Staff member
abedalkareem":3e414z4x said:
HI

any Cross Compatible WD Family with Espirt ???
tried Europa it is compatible
but I need another family, running out of Donors & time to order a new one
any suggestion much appreciated

Zephyr, Mariner, Dolphin, and Mariner 5 4K families have a really close aerial density and the right number of heads, but given that Espirt is a WD black I doubt it'd work.
 

jcarlosmor

New member
Jared":1e27ca9x said:
Confirmed! Sabre 53 and Sabre 58 are cross-compatible for heads. :D

For Sabre 53, what about replacing deffective double headstacks (with both upper or lower heads), with good single headstacks (with only upper or lower head).

It appears that most of Sabre 53 of the same capacity (40 GB) were manufactured with both options of headstacks at assembly plant, maybe for headstack availability at the time.

If we assume the media platters were the same, that would mean that in double headstacks one of the head was disabled. Lacking of specialized data recovery hardware (PC-3000 and similar), is there any way to electrically test which head was disabled? I had measured all of the pins of the headstack connector, but I am sure that the signals are fast enough to show even for a 300 MHz or lower bandwidth oscilloscope. I can measure a square wave at about 24 KHz on some of the pins, but I cannot get any signal activity (other than fixed voltages) at the remaining pins.

The marvell chip datasheet also is not available to common public, so I have no idea what the other chips pins is worth to check. Measuring at the input of the preamp is not a good idea also, also, since I suppose that the heads are some kind of differential devices, and that the preamp converts these signals to single-ended signals referenced to ground. (is at the output of the preamp, that is, the connector between headstack and logic PCB that I am doing my measuring).

Regards.
 

Jared

Administrator
Staff member
The head map used on each WD drive is stored in the ROM. But the format and location of the head map will vary depending on the family of the drive. I've reverse-engineered most of them myself to be able to manually read the head maps, but a tool like PC-3000 certainly can pull up the information a lot faster.

As to your suggestion of using single-head headstacks to replace double ones, that generally will never work. You can use a headstack with more heads than needed, but not less (typically). I say typically, because with the right tools and know-how it is possible to start most WD drives off of just one head, and it can be pretty much any head. However, for the purpose of what you're suggesting, it won't work.

Most drives will check each of the heads they are expecting to find during the initialization process. If it's expecting a head that isn't there, it'll just click and give up. Maybe if a head is actually disabled in the firmware it could work, but generally if they are manufacturing drives with missing heads, that same series won't also have digitally disabled heads. Usually, it's one way or the other. Early on (in the 40Gb days) platters were being tested and servo'd outside the drive, so they knew what head config to install. Later on, they started doing all the servo writing and testing with the drive fully assembled and then began to digitally disable heads rather than install headstacks that have them physically missing.
 
Top