• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

HD temps, hotter != more failures

Elixer

Elixer

Lifer
Another interesting read from Backblaze guys http://blog.backblaze.com/2014/05/12/hard-drive-temperature-does-it-matter/#more-5637

Hard Drive Temperature Takeaways

Overall, there is not a correlation between operating temperature and failure rates. The one exception is the Seagate Barracuda 1.5TB drives, which fail slightly more when they run warmer.

As long as you run drives well within their allowed range of operating temperatures, keeping them cooler doesn’t matter.
Click to expand...
Which basically means as long as you are within the temp range of whatever they rate the HDs (WD Reds are 0C - 70C!, black, blue, green 0C - 60C, purple 0C - 65C Toshiba 0C-70C, Hitachi 5C - 55C, Seagate (newer), 5C-60C Seagate older, 5C- 50C), you will be fine.

I guess, this also explains why most all exernal HDs that come from the makers have no fan.
 
looks like their data doesn't go above 38C or below 14. aka invalid generalization

almost nothing interesting happens in the middle of operating ranges general, it's only when you approach either extremes.

and for home use, it'd be most interesting if you can lower airflow across the drives such that the drives stay 40-50C or near the upper bound of their operating range.

more interestingly, they should compare it to industrial temperature hard drives (operating range -30C to 85C)
 
Last edited:
phis6 said:
I think that is because of transport and handling and not with temperature issues.
Click to expand...

As per Elixer's point, the only time I saw a failure with an external drive in boxes used by my extended family was a unit left running 24/7 and weeks at a time. I mentioned in another recent post and thread that I measured 47C for a drive in my aluminum box -- which does include a fan.

Depending on how the external drive is powered, there are also other factors, like loose USB or eSATA connections that get bumped accidentally during intensive write operations.

Personally, I just made the decision that any single external drive box with its own power supply would only be used for backup, so I would dismount one and power it down when finished.
 
Ayah said:
looks like their data doesn't go above 38C or below 14. aka invalid generalization

almost nothing interesting happens in the middle of operating ranges general, it's only when you approach either extremes.

and for home use, it'd be most interesting if you can lower airflow across the drives such that the drives stay 40-50C or near the upper bound of their operating range.

more interestingly, they should compare it to industrial temperature hard drives (operating range -30C to 85C)
Click to expand...

This. Laptop drives typically reach 50c, and they do fail due to temperature.
 
Phynaz said:
This. Laptop drives typically reach 50c, and they do fail due to temperature.
Click to expand...
They fail from being *over* the temperature that they are rated for, or *under* the temp they are rated for.
If you run them within their specs, then, temp don't play a issue for most of the drives, as noted.
 
Last edited:
Elixer said:
If you run them within their specs, then, temp don't play a issue for most of the drives, as noted.
Click to expand...

you cannot conclude that from the backblaze data. all you can say is that temperatures don't seem to matter between 14C and 38C as a general statement and more restricted ranges for the other graphs.
 
Yeah I wonder if their conclusion is too much, and not supported by their data.

How can they make conclusions about temperatures that are much higher than the range they observed/tested? They conclude that drives are "safe" if below the recommended specs, and yet their data did not test drives that warm.
 
Ayah said:
you cannot conclude that from the backblaze data. all you can say is that temperatures don't seem to matter between 14C and 38C as a general statement and more restricted ranges for the other graphs.
Click to expand...

I wasn't basing it on only Backblaze's data, I was also looking at Google's.
They seem to have come to the same conclusions.
 
All I know is that I had a 75GXP IBM IDE HDD, in a mobile rack without a fan, and I started losing sectors when the drive hit 50C.
 
From the main graph it looks like Seagate's temps cluster around 21/22C and higher failures are noted around >25C. Hitachi's drives cluster around 26C.

I think Seagate did the right thing by placing their sensor at the right place where reported temps can give advanced warning on failures. While Hitachi placed their temp sensor in another spot which gives no predictive ability.
 
VirtualLarry said:
All I know is that I had a 75GXP IBM IDE HDD, in a mobile rack without a fan, and I started losing sectors when the drive hit 50C.
Click to expand...
That one is the outlier and so bad that they had a class action lawsuit and one of the big reasons IBM sold the division (once they built trust back up). Great drives, the fastest out there, and IBM was always the first kid on the street sometimes for a year with the new capacity. But those glass platters were a mess for them. They expanded and contracted so easily.

That said I had a 45GXP, 60GXP, and a 120GXP and all of them fell into the "DeathStar" groupings. None of them with issues.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top