I want to clarify some information about how low level (or "middle level") formating fixes bad clusters. Experiences, pointers or information that you know is correct are welcome.
Hard disks throws bad clusters all of a sudden. And this continues to increase. We all know that a low level format can *sometimes* fix a hard disk throwing bad clusters in this way. And we are able to use the hard disk without problems for a period of time on the same machine (much longer than the rate it was giving bad clusters). Consider a large hard disk with a number of platters and lots of bad clusters (more than what could be replaced by the spare disk space during a low level format).
(1) Can anyone explain why bad clusters appeared originally in this senario
(2) How did a low-level format fix it making it more "stable" than before.
I am sure there are a few of you who had a similar experience. You think the disk is really dieing, but then with a lot of luck you are able to revive the disk.
1) Typically due to rough handling.
2) Modern harddrives can remap some bad sectors automatically. Remapped sectors can sometimes be spotted by programs like HDTach, that makes a transfer speed graph of the drive. A large downward going spikes indicates that the drive had to go somewhere else for the sector.
Bad sectors that are not remapped can be ignored by the OS. During a format of the drive, the entire cluster that contains the bad sector is marked as bad. For FAT32 this is done by writing a reserved value in the FAT table.
A lowlevel format is not possible without special software. At some point in time several PCs had the option in the BIOS. Im not sure why it was removed. Perhaps because of vendor specific implementation of the format track command.
I expect that the purpose of a lowlevel format would be to attempt reclaiming of remapped sectors.
Some suggest a loss of maganetism on the hard disk could resut in bad sectors. How (and why) likely is it that a hard disk looses magnetism.
Could the location of the speaker inside the box cause bad sectors? And would a low-level format be able to fix this permanently?
In my initial post I was trying to findout if anyone were able to recover a hard disk by doing a low-level format (meaning, bad clusters weren't being caused because of physical damage on the platters).
Bad sectors can be "achieved" by a number of means:
A. Bad handling of the drive.
B. Physical flaws on the surface of the drive.
C. Particulate on the surface (dust)
D. Problems with thermal expansion, lack of hardware compensation/calibration
E. Problems with the hardware/firmware.
Low level formats mark all bad/dubious sectors as bad, meaning that they can never again be used. They are now essentially dead spaces on the drive.
My 60GXP went like that. It would operate normally for a couple of months then suddenly throw up a handful of bad sectors. If i was unlucky the OS died. If i was more fortunate i only lost a bit of data or the bad sectors occured in blank space. The only solution was to do a DFT reformat.
Unfortunately it just kept happenng every 2-3 months. After the third instance i had lost all faith and patience in my 60GXP and RMA'ed it.
BOTTOM LINE: If you start getting bad sectors its unlikely you will stop getting them. In the long term its better getting the drive RMA'ed or replaced.
A magnetic field can corrupt data on the drive, but it doesn't cause permanent errors. If an external field is not present, the drive just doesn't suddenly loose magnetism. This is extremely unlikely and would indicate problems during writing.
In order for a magnetic field to change data, it should be pretty strong. I don't think the field from a speaker is strong enough.