I have heard of this technique, but havent used it yet, however drives fail for all sorts of different reasons and chilling a hard drive may work for a hard drive which warms up and then operates outide of normal operating tolerances to fail.
All drives I have had, failures have been caused by different causes, such as:
- The head striking the platter and scratching ( such as when my office chair decided to raise itself a few seconds after getting out of it and the arm of the chair slammed the underside of the desk sending a shock wave to the hard drive and causing the clunk of death )
- The drive motor needle bearings after 8 years of use 24/7/365 in a NT4 file server went from the whining noize of the Fast RPM SCSI drives of the time, to what started at about 6 years into its life a cyclic whine harmonic to a loud groan and not able to spin to the correct speed to be read.
- Had an early Maxtor 500GB SATA drive that would run fast and then make a Screetch/PING noise and then go out to lunch for about 30 seconds and then start working again. Sent this one back to warranty replacement.
- A customer brought in a system that the hard drive smoked the controller board in it. They plugged both the SATA power as well as the P-connector into an early SATA drive that had optional power input, but of which had a warning in small print never to plug in both that they didnt see.