Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
March 3rd, 2026, 10:53
Dear forum members,
I’m dealing with a Samsung SP1604N PALO FS 160GB and could use some advice.
A head swap was performed in a cleanroom. Unfortunately, the drive’s SA is damaged.
I have several SP1604N 160GB PALO FS drives available as donors. Some have firmware 100-24 and others 100-31. The patient drive originally has firmware 100-24.
I performed a hot swap using a working donor with the same firmware (100-24), but I did not first transfer the patient’s SA modules (I do have a backup of the patient’s SA from head 2) to the donor before hot-swapping, That was a mistake.
At the same time, I was experimenting with SHTR and MRT. On a test drive, I used SHTR to load burn files:
* TJ32A_F.dn
* then TJ32A_B.dn3
After that, I wrote the patient’s SA modules to this drive and copied the memory from the test drive. Since the sleep function does not work with the B.dn3 burn file, I mounted that PCB onto the patient drive, started it and loaded the memory into it. This actually worked.
Later I noticed that even without manually loading memory, the drive would eventually spin up by itself after a lot of clicking when using this burn file PCB, unlike using the normal firmware with the PCB originally from the patients drive.
This makes me think that this burn file may be skipping parts of initialization. I also suspect that not all SA information is being properly loaded when using this burn file.
Current situation:
In MRT (PIO mode), I can now read data.
However:
* First read attempt gives around 10% success (light green).
* I also get dark green sectors (no CRC error but containing garbage).
* If I retry a dark green sector multiple times (sometimes 5+ retries), I can eventually get valid data.
My questions:
* Are the dark green sectors more likely caused by a weak or partially damaged HSA?
* Or could this behavior be due to incomplete SA module loading (because of the burn file)?
* Does anyone have experience with physically damaged SA on head 0 and 1 specifically on this model?
Tools available:
* SHTR
* MRT (no head map support for this drive)
Any insights or experience with this family would be greatly appreciated.
Thank you.
March 7th, 2026, 11:10
EIDE hard disks are a headache as traces run all over the place.
Hate to see people assume hard disks can run forever. Hardly. This is why backups are known from the mainframe shops.
Powered by phpBB © phpBB Group.