Archived from groups: alt.comp.hardware.pc-homebuilt (
More info?)
Others are confronting Mxsmanic's speculations.
Speculations? Yes. He cannot provide any technical reasons,
numbers, or citations for his claims. Somehow he just knows
One lowest number for power cycling I ever saw was on an IBM
disk drive - 40,000. That means power cycling seven times
every day (including holidays) for .... 15 years. Yes, power
cycling is destructive. And then we replace the Daily News
tabloid version with reality - the numbers. Seven times every
day for *15 years* means, for all practical purposes, that
power cycling is only destructive in myths and half truth
reasoning (also called propaganda).
Bearing failure is a significant reason for failure.
Bearing failure is due to hours of operation - not due to
power cycling. As in most myths about 'destructive power
cycling', the purveyor does not even provide manufacturer data
sheet numbers. Hours of operation is often a reason for
failures - not power cycling, not mythical transients, and not
the so many other things promoted by junk science reasoning.
Note reasons provided by Mxsmanic. He observed things. He
did not bother to learn why things failed. Did not repair
failed component inside the drive. Did not learn what
manufacturers numbers said. Did not demand numbers. Failed
to first ask so many "Whys?" Instead he just knows only from
personal observation followed by personal speculation.
Definitive conclusions based only upon a hypothesis or
speculation - and no numbers - is classic junk science
reasoning.
Mxsmanic wrote:
> JAD writes:
>> your not serious? bearing failure makes up 80%+ of hard drive failure.
>
> All the failures I've encountered on disk drives have been media
> failures or head/actuator failures. Bearings will eventually fail,
> but usually not before something else fails. And if one starts and
> stops the disk regularly, one nearly guarantees that something else
> will fail first.
>
>> Although a 'load' is put on the PSU during spinup, I
>> don't think that out weighs the 24/7 spinning of the bearings.
>
> More importantly, a load is put on the media and the heads, and they
> will start to wear out after enough cycles.