Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I have the CM690 II case, and I'm a bit worried about my HDD temps. It's quite hot in my country now (Croatia), so my drives are quite hot at about 45°C, while the rest of the system seems to be coping nicely.

My question is, should I invest in some more fans or are these temperatures "acceptable"? I have room in the case for a HDD fan and two more case fans below the drives, but I'm unsure what exactly to do in this situation, and how useful would these fans would be (I don't really have much experience with hardware and/or cooling systems).

Also, are there any cooling alternatives/additions to fans which would work well with CM690 HDD cage?

share|improve this question
2  
I would not be concerned with that temperature. –  Shinrai Aug 6 '12 at 20:41
    
What does the manufacturer of your HDD report as safe operating temperatures? This is commonly available info on most/all manufacturers' sties. –  Ƭᴇcʜιᴇ007 Aug 6 '12 at 20:43
    
@techie007: WD specs say 0 to 60, but the question is whether this is a good long-term decision to make or not? –  dr Hannibal Lecter Aug 6 '12 at 20:47
    
You should probably buy a face mask and a straight jacket as hardware, for starters. Just joking. Your screen name and all. It cools down the temperature or it gets the hose again! –  allquixotic Aug 6 '12 at 20:47
    
@allquixotic: Don't make me find out where you live! :) –  dr Hannibal Lecter Aug 6 '12 at 20:50

1 Answer 1

up vote 3 down vote accepted

A very large empirical study in Google's datacenters was conducted about the reliability of hard drives, and the study analyzed temperature as one of the variables that affected mean time between failure (MTBF), expressed in the article as Average Failure Rate (AFR). The paper is an interesting read, and if you study it carefully, you will gain some detailed insight into how various factors affect the expected lifespan of a hard disk.

The paper.

However -- if you're looking for an unscientific answer, any temperature within a few degrees of the cited recommended operating temperature -- 0 to 60 C as you say -- should be "fine". You can expect typical failure rates with temperatures within the recommended operating temperature of the drive. You can expect steadily increasing failure rates with temperatures outside of the recommended operating temperature, either at the high or low end.

Since you're well within the operating temperature recommendations, I wouldn't worry about it.

A very salient passage from the paper suggests that temperatures near the high end of the recommended operating range are not high enough to cause a measurable increase in failure rates:

In the lower and middle temperature ranges, higher temperatures are not associated with higher failure rates. This is a fairly surprising result, which could indicate that datacenter or server designers have more freedom than previously thought when setting operating temperatures for equipment that contains disk drives. We can conclude that at moderate temperature ranges it is likely that there are other effects which affect failure rates much more strongly than temperatures do.

I can also correlate these findings with my own personal experience. I have had a dedicated server in a hosting company's datacenter for over two years, and the four HDDs have been on 24/7 since it was installed brand new.

They are bog standard SATA 3GB/s 2TB Seagate hard disks -- not Enterprise-grade. I saw the temperatures in SMART and asked the datacenter about it, but they said that all their drives run at or near that temperature due to the design of the airflow in their datacenter, and that they still have very good MTBF for that model of disk. They told me that the RAID mirroring would protect the data, that they monitor the disk health and will replace a broken disk ASAP, and that they usually get disk failure earlier because of many head parking cycles or because of manufacturing defects, NOT because of temperature.

The disks run at 58 C and not a one of them has failed yet, nor does SMART display signs of pre-failure.

share|improve this answer
1  
Interesting info..can you think of any reason why did they decide to make the sensor indicator in GNOME red at 45? If Google's research is true, then this seems quite arbitrary..? –  dr Hannibal Lecter Aug 6 '12 at 21:17
1  
Yes, it probably is quite arbitrary, and based upon outdated metrics about what is a "typical" acceptable temperature range. If your manufacturer says 0 to 60, then 45 is only in the 75th percentile of the range, which is equidistant between "ideal" (30 C) and "the high range of normal" (60 C). FYI, hardware in general has been becoming more tolerant of high temperature in recent years. Graphics cards can sometimes tolerate 90 to 100 C now, whereas that would instantly melt a card from 10 years ago. –  allquixotic Aug 6 '12 at 21:21
1  
New alloys, solid caps, and other such electro-mechanical and materials science developments have enabled this new, higher tolerance of heat in electronics. Heat is still definitely a problem and you can't just ignore it totally, but the acceptable ranges for stable operation have been increasing across the board. –  allquixotic Aug 6 '12 at 21:23
1  
Well, it seems like I'm going to sleep well tonight, thank you for all the info. TIL! Edit: Also, I will not be hunting you down..so..you're welcome! –  dr Hannibal Lecter Aug 6 '12 at 21:26

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.