Talk:Head crash

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Why do we need an article dedicated to head crash when we have the more general Hard disk failure article?

84.215.158.45 04:36, 14 October 2007 (UTC)[reply]


What sort of conditions increase the probability of a head crash occuring? Analogously, what steps can be taken to reduce the chances of a head crash? This year I've overheard more and more people mentioning, "ahh my hard drive crashed, I have to get a new one" more than I've ever heard it (and I've spent most of the last 6 years of my life in computer labs) -- is there any empirical evidence of hard drives, in general, becoming produced less and less reliable? This doesn't seem to be a topic covered very much on the web, other than what to do after the fact, so this article could use some expansion. --69.234.185.170 06:07, 20 Apr 2005 (UTC)

Generally the biggest cause of head crashes in PC and similar hard drives is moving or dropping the drive when it is powered on. Drives don't like sudden movements when they are spun-up and sudden impacts can cause the heads to dig into the platter surface, damaging the magnetic coating. IMHO most 'head crashes' people claim to have suffered-from are more likely to be soft failures, modern EIDE and SCSI drives being extremely reliable, providing they are handled correctly. The drives are assembled in clean rooms so dust is not usually a factor providing the drive case hasn't been opened. Although many will end up in consumer PCs these drives are manufactured to business (i.e., professional use) standards and providing the drive doesn't fail within the first few weeks, most will easily last five years, the drive generally becoming obsolete due to drive capacity/speed advances, (i.e., people replacing the drive with a bigger or faster one), before actually wearing out.
BTW, I used to work on Honeywell/Bull mainframes and the Control Data/Data General MSU4551 removable platter hard disk drives occasionally had proper head crashes. One could always tell when, as there used to be a distinct smell of burning around the drive. The result of this was having to get an engineer to replace the individual head assemblies, and then use the latest backup disk pack and re-run any jobs that hadn't been included in the backup. Ian Dunster (talk) 10:44, 21 September 2008 (UTC)[reply]

What is the probability for head crash in case of normal HDD usage (according to the manufacturer's specifications) in server environment (several months non-stop online)?

Is it more probably a meteorite to destroy a HDD than a head crash in case of normal stationary usage for period about 6 years?

A link to an article (from an authoritative non-biased source) with such information will be valuable.

"Since most modern drives spin at rates between 5,000 and 15,000 revolutions per minute, the damage caused by a head crash can be disastrous."

A few bad sectors (not caused of head crash) over very important files can also be "disastrous". In how many of the cases the physical damage is considerable (more than 5% of the magnetic surface is damaged)? (reliable statistics data will be valuable)

Maybe it is not a good idea to listen music from HDD based MP3 player while skiing/snowboarding. How severe is the problem for the stationary (desktop/server) users?

According to this article some HDDs have protective layer on the platters. Also (from the same source) it is more probably a head crash to occur in old HDDs despite the higher floating height of the heads. --

120km/h a little abstract[edit]

>At 7,200 RPM the edge of the platter is traveling at over 120 kilometres per hour (75 mph) This is quite abstract for such a small device. Why not write "~33m/s". That will give people a better sense of how fast the disk is spinning. —Preceding unsigned comment added by 91.32.126.131 (talk) 00:22, 11 February 2011 (UTC)[reply]

Really? I have no idea how fast "~33m/s" is. However, "over 120 kilometres per hour (75 mph)" actually means something to me. —Diiscool (talk) 14:37, 11 February 2011 (UTC)[reply]
And you can totally imagine how fast a small disk is when it is spinning at 120km/h. No, I don't think so. You know how long a second is, you know how long 33m are. But I guess you don't even know what 120km/h means. You're just saying that because you can read your speed meter in your car. —Preceding unsigned comment added by 91.32.126.131 (talk) 15:29, 11 February 2011 (UTC)[reply]
I think the speed of a car is a reasonably good comparison. What can I compare 33 m/s with? The speed of storm wind? That's far more abstract. -- intgr [talk] 21:28, 11 February 2011 (UTC)[reply]
You imagine making a leap of 33m in a second. That's pretty reasonable. Other units are rather abstract. Ask some people how fast 120km/h is. You'll get the answer "well, my car goes that fast". I hope you know how subjective the perception of speed is in a car. Drive 200km/h for 10 minutes and slow down to 100km/h. You'll feel like you're driving 50km/h. Furthermore, we're not talking about a car, we're talking about a small static object that has spinning disks in it. Smaller units make much more sense for that. 91.32.126.131 (talk) 21:45, 11 February 2011 (UTC)[reply]
So we just use both figures, and while we're at it we can include 100 feet per second. Incidentally back in the days when disk drives looked like washing machines, the 14" platters spun at 3600 rpm for an "edge speed" over 100 mph or around 150 fps. I'll leave it up to you guys to do the metric conversions there :) Nibios (talk) 19:14, 21 May 2011 (UTC)[reply]