There is some discussion of hard disk cooling fans in the thread SATA Versus SCSI (Reply #18-#26).  To highlight the subject, however, I am starting a new thread.

Below is information on hard disk fans that may be of interest:

• There are several models of the bottom-mount cooling units available, distinguished by the number of fans present (one or two), the CFM air flow, and the noise level.  A selection of options may be found at Micro Center and Performance PCs, for example.
  
• See this article for a good overview of disk fan installation.
  
• If you have an available 5.25" bay, there are disk cooling units that combine a heat sink with an intake fan – e.g., StarTech HDDPRO Hard Disk Cooler.

For those who are installing a disk fan, here is a request:  could you kindly monitor and measure the temperature of the disk before and after the installation of the fan, and report back the improvement observed?  For software tools to retrieve and display the hard disk temperature, please see Monitor Hard Disk Temperature.

I'm probably one of the greatest proponents of hard drive cooling.  I've tried the ones with two tiny fans that sit in a 5.25" gizmo of one kind or another.  Both with the fans in the front and the fans in the rear.  And, some with air filters that plug up with dirt and never seem to get cleaned out. :-?
All those things that I wasted my money on are history.

I even spent $49 for a cooler that seriously wet the bed.  I think it was called a "Hermanator".  
Go Figure!

As for hard drive temperature measurements, I use a digital temperature sensor.  
The first digit of my left hand.  ;D ;D ;D
If I lay my finger on the drive and it makes me say "Ouch!!!!!!", it's too hot.

Just for grinns, I just tested (digitally)  the temp of my two SATA HD's that have been on now for about six hours.
There is NO detectable temperature there at all.  Both are essentially at room-temp.

Here's my backup drive, with 5.25" bay spacers and VIO 'two-fan' cooler, mounted in an upper drive bay.  
http://img.photobucket.com/albums/v235/El_Pescador/DrWho2006.jpg

If this pic is too large for this forum, maybe some kind soul (moderator) will shrink it for me. ::)

At just $4 or less for the VIO two-fan coolers, keeping hard drives cool should be well within the price range of almost anyone.  I order them six at a time, to keep shipping costs down.

Cheers,
The Doctor  8-)

DrWho2006, a few questions:

• The fan referenced in Reply #1 is blowing air up (toward the bottom surface the hard disk drive) – correct?
  
• Does the extra height added by the fan require that you use two 5.25" bays to hold one hard disk drive?
  
• Is the speed of the fan adjustable (low/medium/high)?  If so, it is done through software or through a switch on the fan?
  
• What is your qualitative description of the noise added by the running fan?

Thank you.

DrWho2006 wrote on Apr 22^nd, 2007 at 11:28am:

I'm probably one of the greatest proponents of hard drive cooling..... As for hard drive temperature measurements, I use a digital temperature sensor.   The first digit of my left hand.  ;D ;D ;D .... If I lay my finger on the drive and it makes me say "Ouch!!!!!!", it's too hot. Just for grinns, I just tested (digitally)  the temp of my two SATA HD's that have been on now for about six hours. Both are essentially at room-temp. At just $4 or less for the VIO two-fan coolers, keeping hard drives cool should be well within the price range of almost anyone.  I order them six at a time, to keep shipping costs down.

I got the fans from CWC group, installed them. I could touch the top of the drives and the temp of the drives feels about the same as room temp. In the past, before the fans, they used to be real hot. Now I am a believer in the fans which cost less than $4 plus shipping.

Here are my other comments on the fan package.

1. The screws that came with the package were of wrong size and did not fit the bottom of HDs.
2. There were no offsets in the package. Fortunately I had some of them in my tool kit.
3. The power connector is long enough and nice.
4. The model I got was Xper Hard drive cooler model HC-350.
5. CWC was very responsive. Shipping prompt. When they did not hava a particular color in stock, they called me immediately and everything was taken care of. I also inquired about the ball bearing vs non ball bearing versions. Was told they have been shipping these coolers since 1990 and have had no problems. Would order again with the company.

Overall, I think you have the best bargain for the money. Have had HD failures in 1-3 years and am now convinced they are due to heat.

Note: To cut down shipping costs, order them at least a doz by pooling with your friends if necessary. The local store is selling similar fans for $10.00.

*  [smiley=2vrolijk_08.gif]

It occurs to me that when dissipating heat from a hard disk drive, the heat doesn't just disappear – it goes into the main chassis housing the other PC components.  Therefore, wouldn't it be wise to also increase the RPM of the rear exhaust fan when using a hard disk drive cooling fan?

P.S.:  Rama, perhaps you too could look at the questions in Reply #2?

Pleonasm wrote on Apr 22^nd, 2007 at 2:17pm:

DrWho2006, a few questions: The fan referenced in Reply #1 is blowing air up (toward the bottom surface the hard disk drive) – correct?

Yes.


Quote:

Does the extra height added by the fan require that you use two 5.25" bays to hold one hard disk drive?

I have installed them in 3-1/2" bays. It took some extra space; almost one bay. Since I had extra bays there was no space problem.


Quote:

Is the speed of the fan adjustable (low/medium/high)?  If so, it is done through software or through a switch on the fan?

No switch on the fan to adjust the speed.


Quote:

What is your qualitative description of the noise added by the running fan? Thank you.

No noticeable increase in noise. Could not even tell there were two additional fans without looking at them.

*  8-)

Rama, if you do not have an extra and empty 3.5" bay, will the fan work in a single 3.5" bay – or, is the increase in the vertical height too much to fit into a single 3.5" bay slot?

Thanks.

Pleonasm wrote on Apr 22^nd, 2007 at 3:53pm:

Rama, if you do not have an extra and empty 3.5" bay, will the fan work in a single 3.5" bay – or, is the increase in the vertical height too much to fit into a single 3.5" bay slot?

As the fan is attached to the bottom of the HD, each HD needs at least 3/4" additional space (without the offset) and 1 " space with offset,  hence would not fit in a single bay.  However, even with a single bay, if there is space at the bottom of the bay, then a fan can be added.

We may want to explore if there are any skinny fans which when added to a HD will fit into a single bay. On the other hand, I have always tried to avoid installing drives close to each other since additional space in between drives helps circulation (in all my cases, the space has not been a problem with 2 HDs.)

Hope this helps.

*  :)

Rama wrote on Apr 22^nd, 2007 at 5:03pm:

As the fan is attached to the bottom of the HD, each HD needs at least 3/4" additional space (without the offset) and 1 " space with offset,  hence would not fit in a single bay.

The lack of space above/below the HD is the reason I have switched to the Antec Spot Cool fan.  I can mount it with a screw to some empty space in the enclosure, and with the flexible arm I can point it to where I want it.

http://i24.photobucket.com/albums/c5/ghostimage/SpotCoolFan.jpg

I worked through many different types of coolers and ways of mounting drives till I came up with the method I now use.
The questions about noise, etc. have already been answered.  Cool operation with virtually NO noise.
I just leave the side panels off of my tower, so heat buildup in the case is really NO problem.

And at less than $4, it's sure not going to break the bank.

When you start adding things to a PC that the original designers never counted on, you sometimes have to get creative.  When mounting the drives+coolers in the lower drive bays of a mid tower, you sometimes have to bend out the little tabs that the drives normally sit on.  Again, NO Biggie!
http://img.photobucket.com/albums/v235/El_Pescador/AinstalledcooledDrive.jpg
Main HD in a lower drive bay, with tabs bent out.

When there are NO drive bays, like in a little DELL, there are still ways.......( get inventive )
I'm shipping a rebuilt Dell to Illinois this week, that I did major revisions to.
The original drive failed (burned up) because of little or NO drive cooling air reaching the bottom of the drive, where all the heat is generated.

Here's the HD bracket for the Dell, providing very poor circulation against the bottom of the drive.
http://img.photobucket.com/albums/v235/El_Pescador/BHDProject1.jpg

So I cut out the grill over the drive:
http://img.photobucket.com/albums/v235/El_Pescador/CHDProject2.jpg

Then I installed the 1/4" brass standoffs and then mounted the cooler to the drive.
http://img.photobucket.com/albums/v235/El_Pescador/DHDProject3.jpg

Then returned the whole drive + cooler + drive bracket, back into the case.
http://img.photobucket.com/albums/v235/El_Pescador/EHDProject5.jpg

Now this new hard drive will stay as cool as a cucumber and should last a long time.

Even laying the drive upside down in the bottom of the case is completely acceptable, in cases where there is just no room anywhere else.  A friend in Arkansas did that and secured the drive with pink duct tape.  I've got the picture someplace.  ;) ;D ;D ;D
Here's my own version:
http://i54.photobucket.com/albums/g90/DrWho1943/NEW.jpg

How to cool a hot running external HD:  (Gimme a smile!)  ;)
http://img.photobucket.com/albums/v235/El_Pescador/FextHDfan.jpg

I'm sorry this post ran so long, but there's just so much to say and so little time.

Doc  8-)

Ghost4me, I see why you prefer the Antec SpotCooler fan.  A few questions:

• Exactly how does this fan attached to the motherboard?  Is the mechanical connection “secure” or is it somewhat “flimsy”?  Is it easy to physically remove the fan if necessary?
  
• How long are the wires that connect the fan to the motherboard fan connector?
  
• Can the blue LED on the fan be turned off (or disabled)?

Thank you.

* * * * * * * * * *
DrWho2006, thank you for your excellent posts to this thread.  I am confident that they are (and will be) appreciated by many.  (Keep 'em comin'!)

An excellent article about the Antec SpotCool fan is:  Antec SpotCool Flexible Fan System Cooler Review.

I had never seen one of those little single fan coolers.

Velly Intellestink!

Going back to Rama's questions.
I put the backup drive in the 5.25" bay just for the heck of it
and to relieve congestion in the lower bays.
NOT, because that was the only place I had to put it.

I had the space and I had the 1" side spacers, so why not?
Because of where the mounting holes are in the case for the spacers to bolt onto,
YES it does take up two 5.25" bays, with the fans attached.  Again, no biggie!
With so much air space all the way around the drive, even if I someday decide to put
my side panels back on.....this drive will stay nice and cool.  ;)

Doc  :-?

A few accessories that might be of interest to readers of this thread:

    Fan Power Extension Cable
    Fan Power Splitter Cable

Pleonasm wrote on Apr 23^rd, 2007 at 10:14am:

A few questions: Exactly how does this fan attached to the motherboard?  Is the mechanical connection “secure” or is it somewhat “flimsy”?  Is it easy to physically remove the fan if necessary? How long are the wires that connect the fan to the motherboard fan connector? Can the blue LED on the fan be turned off (or disabled)?

The review link which you posted is very complete with good pictures.  The fan is attached to any empty hole in the case with a standard slot screw.
The wire is about 6-8 inches.
No, you can't turn the blue LED off, but it's not noticable, unless you have a clear pc box.

I have mine pointing to the two hard drives, which it cools nicely.  I've used hard drive fans (pictured in this thread) but as mentioned they take up room.  The Antec fan is quieter than any other HD fan I've used.

I have installed 6 fans in 3 machines and can see the HDs are cool to touch compared to they being hot before. Hopefully my HD failures should come down. Time will tell. When I installed in the third machine, it was very quick and am becoming a expert!

Two issues some may face are (1) bending a tab or jiggling the drive into the HD bay (already addressed by Dr.Who) and (2) buying the brass 1/4" standoffs. The local computer store had a Ultra (www.ultraproducts.com) Multi-purpose Screw Kit - Part # ULT31338 (also No. 9846909683) with barcode 2276931338. The package which cost $2.99 retail contains the following:

1. Metal Speed Thumb Screws (10).
2. Computer Assembly Screws Hex/Phillips 6-32/5mm (24)
3. Computer Assembly Screws Hex/Phillips M3/5mm (24)
4. Brass Motherboard Standoffs 8.5mm (15)
5. Brass Motherboard Standoffs 6.5mm (15)

I think the price reasonable for the use at hand.

I have couple of more computers that need to get fans installed. Plan to talk to the company and find out the difference between various models and order a dozen.

*  8-)

HI Again,
For us who have built many systems over the years, having a good supply of those little brass spacers is not a problem.  An ample supply comes with every new case,,,,, enough to mount even the largest 'full sized' mother boards.
What is a problem is that many case manufacturers use the spacers with fine threads, which will NOT screw into the holes on the bottom of the drive.  And yet some makers use little shiney metal clips and not the brass spacers at all.  :'(

So I'm quickly running out of the 1/4" brass spacers that have the course threads on the male end.

I don't see that as a problem though,,,, but just another inevitable fork in the road.
Sometimes we have to look beyond the obvious and whats right in front of us.

The whole idea is to get some space between the fan and the drive to avoid the "Air Dam" effect that occurs when the fan is too close.  Anything about 1/4" thick will do nicely.
Rubber spacers, or thin strips of balsa wood, or stacks of little washers and longer screws? ;)

The day I finally run out of the 1/4" brass standoffs is just another day to "Get Creative".  

You'd be surprised what I do around here with a stack of beer coasters from my favorite watering hole.  ;)  Cut into narrow strips and glued together, I'm sure they'd make fine spacers. :-?

Get creative, be happy and stay COOL!

Yous guys are the greatest!

The Doctor  8-)

First I thought of being creative about spacers, growing up in an environment where we made our own toys when we were children, then wanted to give a try at the local electronics store and lucked out finding the brass standouts at a reasonable price. I am sure one can get them at any shop which assembles and repairs computers. Would like to know from others what creative approches taken as a substitute to brass ones.

*  ;)

Rama,
Let me start by saying I LOVE your little saying under your name.
I used to have in my own Signature line, "The only bad backup is the one you decided NOT to make".

I was allowed to paw through the junk hardware box at "Cheap Guys Computers" one day, but only came up with five spacers of the right length and screw thread.
The male end of the spacer absolutely must have course threads in order to properly screw into the holes on the bottom of the hard drive.  The female side doesn't matter as long as you have compatible screws.  

Why the coolers are coming with 4 fine threaded screws is beyond me, if they intended for the coolers to be screwed directly onto the drives.  The threads just don't match. >:(  Someone else had pointed that out in this forum.  Another one of life's little mysteries. ::)

I see on different web sites like the one previously posted, that there are many different types of two-fan HD coolers.  Prices range from $9.95 all the way up to $19.95 for essentially the same thing.
I was elated to find the VIO coolers discounted down to less than $4 ea.
It's kind of like getting a "Big Mack" for $.25. (twenty five cents) ;) ;D ;D ;D

They all DO require some annual or semi-annual maintenance, like cleaning and oiling, but hey......
isn't that better than loosing hard drives and all the hassle that goes with that?

I've found that just one drop of fine machine oil, injected through the seal on the back of the fan motor once or twice a year is all that's required to keep those little guys running smoothly for years to come.  When I say "Injected", I mean with a hypodermic needle.  
I tried an insulin syringe, but the needle is too fine for the oil to freely flow through it.  You need something larger.  I use a syringe that came with an Ink Refill kit.  Works perfect!

Thus endeth "Fan Maintenance 101".  ;) ;D ;D ;D

The Doctor  8-)

Last week I ordered the fans from CWC-Group. I wanted to order a dozen more today and was not happy with their high shipping charges relative to the weight of the shipment.

Looking around, I found www.outletpc.com have decent pricing and reasonable shipping charge, and ordered the fans even though they did not have the variety as CWC. I even added a few more items to the shipment for a very marginal increase in shipping cost. This seller is worth looking if you are buying fans and other computer items.

*  ;D

The temp after installation of fans on IDE drives runs at 92 degrees F and before it was hovering around 105 - 110 degrees F. Most noticeable was the how cool they are to touch whereas they were very hot before. I used a digital temp gauge to check the temp. Hope this helps.

*  8-)

It appears that the OutletPC has some good prices on certain items.
Some items I can get much cheaper right here locally.
The price on the little coolers is good and if you can save a few bucks on shipping.....GO FOR IT.

Get the best deal you can, I allus say.

The Doctor  8-)

The reader of this thread may also find the following articles of interest:

• ”Information about fans” and “Hard Drive Cooling” at The Heatsink Guide
  
• Temperature Limits and Drive Cooling

In addition, note that there exist hard disk cooling fans that require a minimal amount of vertical space when mounted:

• Thermaltake Hard Drive Cooler A2376:  height of 15 mm (0.6”)
  
• Hard Drive Cooler for 3.5" HDD:  height of 11 mm (0.4”)

Could fans such as these avoid the problem of using two drive bays to accommodate taller models such as the VIO Xper 3.5-Inch Hard Drive Cooler (0.75” or 1” with stand-offs)?

To place the entire issue of hard disk drive cooling in context, consider the insights that Google Labs produced in its recent study:


Quote:

Contrary to previously reported results, we found very little correlation between failure rates and either elevated temperature or activity levels. …failures do not increase when the average temperature increases.  In fact, there is a clear trend showing that lower temperatures are associated with higher failure rates. Only at very high temperatures is there a slight reversal of this trend….

Source:  Failure Trends in a Large Disk Drive Population

In 43 years as a field tech, I've learned to never rely on 'lab results'.
Labs are sterile, over controlled atmospheres where anything can be made to happen.

I worked with some of the best electronics engineers in the country at the "Caterpillar Research Labs" and I had to rework almost every circuit they came up with.  What works on paper, or values set using a slide rule, just don't work in practical application, like on a bulldozer working in a steaming hot swamp or above the arctic circle.  

So please don't tell me I'm wrong, based on some lame lab results.  OK?

I've worked on literally thousands of pieces of computerized equipment in so many different environments, you wouldn't even believe it.
One Rule I've found from 43 years of field service, is that:
1.   HEAT KILLS
2.   and lack of proper maintenance can take you right back to rule #1.

Stay COOL, Stay SAFE!  

The Doctor  8-)

DrWho2006 wrote on Apr 28^th, 2007 at 11:54am:

In 43 years as a field tech, I've learned to never rely on 'lab results'. Labs are sterile, over controlled atmospheres where anything can be made to happen. I worked with some of the best electronics engineers in the country at the "Caterpillar Research Labs" and I had to rework almost every circuit they came up with.  What works on paper, or values set using a slide rule, just don't work in practical application, like on a bulldozer working in a steaming hot swamp or above the arctic circle.   So please don't tell me I'm wrong, based on some lame lab results.  OK? I've worked on literally thousands of pieces of computerized equipment in so many different environments, you wouldn't even believe it. One Rule I've found from 43 years of field service, is that: 1.   HEAT KILLS 2.   and lack of proper maintenance can take you right back to rule #1. Stay COOL, Stay SAFE!   The Doctor  8-)

1. I agree with your observation. Everyone knows high temperature is not good for printed circuits. If they are, then we will not see any cooling fans in computers. Most today's computers will shut down if the temp goes high.
2. From a commonsense point of view a lower/room temperature cannot harm printed circuits. Bottom of HDs have a printed circuit which controls all HD activities.
3. Re lower temperatures, the only comment I have heard is that in very low temperatures, HDs may have difficulty in starting to spin. Many of the current HDs use a fluid suspension instead of ball bearings. At very low temperatures, the fluid may freeze and and drive may not spin.
4. May be others with personal experience want  to add their experience.

*  8-)

What most folks who have not worked in the Electronics Industry, don't know, is that home computers are made of what we might call, "Hobby Grade" components.
They are made up of the lowest quality components from the lowest bidder.
Witness, the bulging capacitors on many motherboards.

They will only operate at temperatures comfortable to the operator.
If you're not comfortable at 120°F., then your PC won't be happy either.
Adversely, if you're not comfortable at 32°F, your computer won't be either.
For any home computer to 'live long and prosper' it should be kept at the same
temperature as an operator sitting there in a short sleeved shirt. :-?

The last HD that failed on me was several years ago, before I started using two-fan coolers on every drive.  When I reached in to remove it, it was like removing a fresh baked potato from a hot oven.
Ouch!!!!!

To say that it was "Too Hot To Handle", is a gross understatement.

I was setting up a new HD on my bench a while back.  It was sitting upside down, with the motor hub and logic board facing up.  After partitioning and formatting the drive, it was "Too Hot To Handle".
I stuck on one of my Two-Fan coolers and within just a few minutes the drive was back down to room temperature.  

The shock factor, of a circuit board getting that hot and then cooling off when the drive is shut down can (in time) rip the circuits apart.  Keeping it always at room temperature eliminates all that expanding and contracting from damaging the drive.

Cheers!
The Doctor  8-)

Pleonasm wrote on Apr 28^th, 2007 at 9:16am:

To place the entire issue of hard disk drive cooling in context, consider the insights that Google Labs produced in its recent study:

DrWho2006,

You wouldn't get away with that "personal experience" argument in Medicine these days. Evidence based treatment is demanded. Why should electronics be any different?

DrWho2006, although the results referenced in Reply #23 were produced by Google Labs, they are not "lab results":  rather, they are based upon actual usage of "more than one hundred thousand disk drives" that were "deployed in several types of systems across all of Google’s services."   The disks examined are "a combination of serial and parallel ATA consumer-grade hard disk drives, ranging in speed from 5400 to 7200 rpm, and in size from 80 to 400 GB."  If that does not qualify as "real world," then nothing does.

To interject some moderation to the discussion, I do not believe that anyone would disagree that temperatures outside of the design specifications stipulated by the hard disk manufacturer are harmful.  However, the axiom "if it isn't broken, then don't fit it" also applies here.  It would wise for a PC user to first monitor their hard disk drive temperatures, and if they routinely fall within the manufacturer's specifications, then adding a disk fan is not necessary – nor would any benefit arise from doing so.

P.S.:  DrWho2006, I would appreciate your viewpoint on the "low profile" (minimal height) fans described in Reply #22.

DrWho2006 wrote on Apr 28^th, 2007 at 2:24pm:

What most folks who have not worked in the Electronics Industry, don't know, is that home computers are made of what we might call, "Hobby Grade" components. They are made up of the lowest quality components from the lowest bidder. Witness, the bulging capacitors on many motherboards. They will only operate at temperatures comfortable to the operator. If you're not comfortable at 120°F., then your PC won't be happy either. Adversely, if you're not comfortable at 32°F, your computer won't be either. For any home computer to 'live long and prosper' it should be kept at the same temperature as an operator sitting there in a short sleeved shirt. :-? The last HD that failed on me was several years ago, before I started using two-fan coolers on every drive.  When I reached in to remove it, it was like removing a fresh baked potato from a hot oven. Ouch!!!!! To say that it was "Too Hot To Handle", is a gross understatement. I was setting up a new HD on my bench a while back.  It was sitting upside down, with the motor hub and logic board facing up.  After partitioning and formatting the drive, it was "Too Hot To Handle". I stuck on one of my Two-Fan coolers and within just a few minutes the drive was back down to room temperature.   The shock factor, of a circuit board getting that hot and then cooling off when the drive is shut down can (in time) rip the circuits apart.  Keeping it always at room temperature eliminates all that expanding and contracting from damaging the drive. Cheers! The Doctor  8-)

In the last ten years, none of my consumer grade drives lasted more than 3 yrs even with minimal usage and more often they failed in little over a year. Since the cost of fans is so minimal, it was worth a try for me. Let me revisit the issue in about a year to see how my drives are doing. If by that time I have not had any HD failures, it would be a reasonable proof for me. If others are quite happy with their systems now, then they can decide what they want to do if they run into premature HD failures.

*  8-)

Rama, I agree that the addition of a disk fan is “cheap insurance” – whether or not it is actually needed, or whether it delivers any benefit in terms of device longevity when the drive is already operating within the manufacturer’s temperature parameters.  Stated differently, there would appear to be little downside risk from using a disk fan (if properly installed and positioned).

Of course, a PC user should assess the thermal state not only of the hard disk drive, but also of the other key components and adjust the cooling strategy as may be required (e.g., enhancing air flow through the case by increasing the rear fan exhaust RPM, etc.).

Pleonasm wrote on Apr 30^th, 2007 at 12:20pm:

Stated differently, there would appear to be little downside risk from using a disk fan (if properly installed and positioned).

Pleonasm,

That would be an interesting study. Whether cooling fans actually increase the life of a HD or paradoxically decrease the life of a HD. The Google study indicates low temperatures may be a problem.


Quote:

In fact, there is a clear trend showing that lower temperatures are associated with higher failure rates.

Brian wrote on Apr 30^th, 2007 at 4:42pm:

Pleonasm wrote on Apr 30th, 2007 at 12:20pm: Stated differently, there would appear to be little downside risk from using a disk fan (if properly installed and positioned). Pleonasm, That would be an interesting study. Whether cooling fans actually increase the life of a HD or paradoxically decrease the life of a HD. The Google study indicates low temperatures may be a problem. Quote: In fact, there is a clear trend showing that lower temperatures are associated with higher failure rates.

We need to look at actual numbers regarding temperature. In some of the southern states, what is lower is perhaps what is higher in the NW.  Again it is a good project for anyone intersted in making a study.

Another way to look at is for a manufacturer like Seagate who offer 5 yr warranty on consumer grade drives  to analyze the return rates due to HD failure and time elapsed between the purchase and return along with the region of the country where the returns take place. I suspect HD manufacturers may treat such a study as proprietory confidential info and may not publish them, especially if the drive failures are frequent in 1 - 3 yr range.

mkr

Brian, you are right:  based on the Google research, there is an increase in the AFR (annualized failure rate) of hard disk drives from about 2% at 30C to 9% at 17C (see Figure 4 in the Google research paper).  Note, however, that the “low temperature effect” is generally present only during the first three months of the hard disk drive lifecycle, after which it is ameliorated quite a bit (see Figure 5).  The temperature range of 30C to 40C appears to have the lowest associated failure rate across the entire lifecycle (although that AFR is not significantly different from the 40C+ range, as judged by the standard error bars in Figure 5, in general).

Rama, “actual temperatures” are in fact reported in the Google research paper (see especially Figures 4 and 5).  Your idea about correlating the AFR with geography is interesting, but it wouldn’t account for whether or not the unit is in an air conditioned environment, for example, where the ambient temperature is controlled.

As a practical matter, I do not see any reason for a person to use a hard disk drive fan in the absence of experiencing temperatures which are outside of the operating limits established by the manufacturer (e.g., 0C to 60C by Seagate, in general).  In the spirit of “don’t fix it if it is not broken,” the first step – in my opinion – would be for a PC user to monitor the hard disk drive temperature, and only then make a decision to install a cooling fan if circumstances justify it.

Hey all, I understand the concept heat = excessive wear, of course. What I'm curious about is what exactly is a safe operating temperature to run a hard drive at? I just bought a 400GB Western Digital Caviar SE16 drive last week and I checked the stats just now - says the operating temperature is 0 - 60 C. Of course, I imagine it's better for it to run in 30's?

http://www.westerndigital.com/en/products/productspecs.asp?driveid=300

Couldn't find a detailed spec list about thermal wear like I have been able to with Intel and AMD processors before. Thanks a lot!

Baseem, if Western Digital says that the acceptable operating temperature for the hard disk drive is between 0C and 60C, then anything in that range is – by definition – acceptable.  I doubt that anyone on this forum has engineering credentials that exceed those of the professionals at the manufacturer, so why not simply accept the assertion as it is stated?

Pleonasm, I agree that one should always accept the manufacturer's specs. However, I was confused if this was the operating environment, or the operating temperature of the drive itself. Granted, 60 degrees Celsius would be a pretty extreme temperature to expect a computer to run in, but it's not clear what is meant in the specs list.

The speedfan program that was suggested to me in the other thread I started about DBAN & hard drive life worked great. I was able to read my 120GB HDD's temps. It displayed 24C when I first powered that machine on this morning, but it never got above 35C at any point in time (monitored it in the system tray) which leads me to wonder if I even need a hard drive cooler at all. Any thoughts?

Baseem, Seagate references the hard disk case temperature (provided by an internal disk sensor) when discussing “operating temperature” (usually 0C to 60C).  Seagate also discusses ambient temperature (usually 5C to 55C), which refers to the temperature of the room in which the PC is located.  I suspect that the same terminology applies to Western Digital, too.

The downside risk of using a hard disk cooling fan would appear to be minimal.  However, based on your prior post, it doesn’t seem that you need to take any action.

DrWho2006 wrote on Apr 22^nd, 2007 at 8:22pm:

I worked through many different types of coolers and ways of mounting drives....

I got a Seagate ST328040A Barracuda 7200 rpm 28.5gb hard drive from a friend and when I formatted the drive, after 30 minutes it was very hot even to touch. This was in spite of it laying open on my desk on its back. Wondered if there was something wrong with the drive. A google search revealed the reviewers found the same thing. They had to wait for atleast one minute of cool of before handling the drive.

So I installed a cooling fan with a 1/4" stand off. Lo and behold, the drive is running cool; no more than a degree or more from room temperature.

In the computer case where I had to install, the harddrive had to go into a l_l shaped cage with two small holes in the bottom of the cage. So I drilled holes in the bottom since i did not have tools to cut off a large square opening which would have provided better air discharge. It is running cool and I do not have to worry about overheating inside the computer case.

No I am convinced of the effectiveness of the cooling fan and I do have some extra ones handy in case I need them.

Rama

Rama, according to the Seagate website, the operating temperature of the ST328040A hard disk drive is 0C to 55C.  If, for example, your drive was at 50C (122F) it would be both “hot to touch” and operating within its design specifications.  In other words, your observation that the drive was “hot” doesn’t necessarily imply that it was running outside of its design tolerances - and, your success in cooling the drive doesn’t imply that that it will last longer or run more reliably.

Now, if the temperature of the drive was in fact above 55C, then clearly corrective action would be required.  Did you actually measure the temperature of the drive (e.g., using SpeedFan)?

Pleonasm wrote on Aug 6^th, 2007 at 9:36am:

Rama, according to the Seagate website, the operating temperature of the ST328040A hard disk drive is 0C to 55C.  If, for example, your drive was at 50C (122F) it would be both “hot to touch” and operating within its design specifications.  In other words, your observation that the drive was “hot” doesn’t necessarily imply that it was running outside of its design tolerances - and, your success in cooling the drive doesn’t imply that that it will last longer or run more reliably. Now, if the temperature of the drive was in fact above 55C, then clearly corrective action would be required.  Did you actually measure the temperature of the drive (e.g., using SpeedFan)?

No, I did not measure the temp using speedfan.

Personally I would rather run the computer at room temp. Since it was a free HD, it really did not matter if it fails in say 10 days. I have a backup!!! All I need is to put in new HD and I am back in business. What a comfort you have with Ghost!

*

When you touch your running HD and have to immediately pull your blistered finger away with a resounding "Ouch!" (or much worse verbiage) it's obviously too hot.
Nothing in a computer should ever run that hot.

Any electronics designer who has to use "Hobby Grade" components, like our home computers use, will tell you quite frankly that "Heat Kills".

Why else would we have CPU coolers, Video chip coolers, PSU coolers, case fans, on and on.
HD coolers are just one more step in the effort to keep our PC's alive just a little bit longer.

Electronics are damaged when they heat up from room temps to HOT time after time.
The expanding and contracting literally RIPS them apart.

Google? They are not electro/mechanical engineers and I sure wouldn't put much stock in anything they say. :P

Hard drive coolers are not Refrigeration devices.  They don't CHILL a drive, but just blow away excess heat. (it can't go below room temperature)
Keeping a drive at room temps has NO downside.  They will last longer and your valuable data will be better protected.

If room temperature was damaging to a drive, they would come with built in HEATERS.  :P

The last Deathstar drive that squalled, balled and stopped on me, (sans cooler) was so hot when I reached in to remove it, that it literally BURNED my fingers.
I had to wrap a rag around it to take it out. ( I couldn't find my Oven Mits)

Since that day, I've had two-fan coolers mounted under every drive I've owned.
That same technique is now being used (with the 1/4" standoffs) by hundreds of PC users world wide.  
One guy wrote me to tell me that his 10,000 rpm Raptor drive that used to run scalding hot, now runs cool with the installation of the little cooler.  

Here's one guy that took my cooling tips to the extreme!

http://i54.photobucket.com/albums/g90/DrWho1943/redblue.jpg

The inlet and exhaust fans are obvious. What's not so obvious is that the RED fan is actually mounted in a clear plastic Left-Side case panel.  And the two blue fans in the base are sitting on top of the HD, taped to the base plate with duct tape.  ;) ;D ;D ;D

When he sent me this picture, my only comment was, "COOL!"

Y'all have a great day now, Y'hear?
Shadow  8-)

Today I had a brief talk with an Hard Drive designer who has been in the industry for several decades. During the discussion about the Google's recent research, it was pointed out that Google tested consumer grade drives used on servers. Secondly, server grade drives are better built to deal with heavy input and output and also non stop operation.

On the other hand, as anyone who has dealt with electronics know that heat kills. Also the drive heads are floating on the drive platter almost touching them. Hence any expansion due to heat is likely to increase problems.

So using the above feedback, I am going to continue to use hard drive fans (they are cheap)  and keep the drives cool. This also reduces the overall temperature inside the computer case, especially the ones which are closed. The problem manufacturers face with additional cooling fan is the consumer complaints about the additional noise; typically some have their computers next to their beds and complain about the noise bothering them when trying to go sleep. Also anyone who is concerned with drive failure in business environments, should consider buying server grade hard drives as they are better built and of course they cost more.

*

Rama, it is true that the Google research is based upon consumer-grade hard disk drives used in a server environment.  But, doesn't that make their findings even more solid?

If the temperature of a consumer-grade hard disk drive (while operating within the manufacturer's design tolerances) isn't highly correlated with failure when used in a server environment, then one would expect the finding to be all the more descriptive of a less demanding home environment, no?

* * * * * * * * * *
TheShadow, if the temperature of your own hard disk drive exceeded the manufacturer's design tolerances, then certainly it was proper for you to take corrective action to cool the hardware.  It doesn't logically follow, however, that installing a cooling fan when a hard disk drive is operating within its tolerances has any benefit.

By the way, the primary author of the Google research is Eduardo Pinheiro, who holds both a PhD in Computer Science from Rutgers University as well as a BS in Computer Engineering.  The second author of the paper, Wolf-Dietrich Weber, holds a PhD in Electrical Engineering from Stanford University.  The third author, Luiz André Barroso, has earned a PhD degree in Computer Engineering from the University of Southern California and holds both B.S. and M.S. degrees in Electrical Engineering.  Therefore, you may wish to reconsider your assertion that Google personnel "are not electro/mechanical engineers and I sure wouldn't put much stock in anything they say."

All the points are well taken. Having faced with hard drive failures, after 1 or 2 years of use, I am more inclined to invest a couple of dollars on a fan even if there is a 10% probability it is going to help.  I would leave it to the readers to decide what they want to do.

*

A PHD doesn't make someone a Thermal Dynamics Engineer.
Sometimes I think that the more they learn, the less they know.  ::)

The "Ripping" effect of electronics expanding and contracting is well known in the engineering community and many techniques have been developed to combat it.

Running a COLD drive, is more apt to occur in Minnesota than here in Florida.
While a drive getting HOT is more apt to occur here than in Minnesota.

In either case, when the drive is under load, it WILL heat up.  A hard drive is made up of several dissimilar metals, plastics, glass (PC Boards), etc.  These materials DO-NOT expand or contract at the same rate.  Ergo, the "ripping" effect of heating and cooling.

Preventing most of that cannot help but extend the life of any drive or other electronic components.  And at less than $5usd for a little drive cooler, I don't see all the hate and discontent this topic has stirred up.  Except maybe that some folks just like to argue and it doesn't matter what the topic.

What really makes me laugh, is the number of people who buy an Intel CPU, over-clock it and then complain that its running HOT.  Then the fight starts over which CPU Cooler is best.
Those threads can go on and on for page after page and there's really no one answer to it.

My AMD CPU is running at the manufacturers recommended speed and is cooled by the Stock cooler that came boxed with the CPU.  And the neatest thing of all is that I can lay my finger on the heatsink at any time and detect only slight warmth.  NEVER HOT!
My first wife ran hotter than that CPU. ::) ;D ;D ;D

BUT, and that's a really big But, I remove my CPU fan and heatsink at least once a year and scrub the dirt out of the heatsink and clean and oil the fan.  Likewise, every fan in my PC (all nine of them) gets a good cleaning at the same time.  Just a little maintenance goes a long way to keeping a PC running cool and happy.

Be COOL and be Happy!
The Shadow  8-)

TheShadow, my request is simple:  provide credible evidence that “Preventing most of that [heat] cannot help but extend the life of any drive.”  As self-evident and commonsensical as the statement seems, the actual data collected and analyzed by the Google researchers suggests that the assertion is largely false (for cases in which the hard disk drive is operating within the manufacturer’s design tolerances).  I admit, this is counter-intuitive and surprising:  but data are data.

Remember, the research reported by the Google personnel isn’t an “opinion,” but is based upon the systematic examination of a huge number of actual, observed hard disk drive failures.  It is not "theory."  All joking aside, I do hope you agree that the credentials of the Google researchers are beyond reproach.  To call these individuals “experts” in computer science and engineering would be an understatement.

:)

I'll put my 43 years of electronics experience, in the field and in the research lab,  up against those egg-heads any day of the week.

And, Oh yes, I'll continue to install two-fan coolers with 1/4" standoffs on all my drives.

You just follow the PHD's and I hope you don't loose too many HD's to overheating.  ;)

Of course, you won't really know it died to overheating, it will just die and you won't know why.

I hope there's no hard feelings!  :-?

Shadow  8-)

The Shadow, please remember that the Google results are based upon an examination of "more than one hundred thousand disk drives".  It isn't an "egg head" theoretical, proscriptive argument; rather, it is descriptive research.  The Google personnel are describing exactly what they observed, not stating what should be based upon thermal dynamics.

These findings suggest that there is little advantage to installing and using a hard disk drive fan when the unit is already operating within the manufacturer's temperature tolerances.

:)

Let me add another factor. Some drives operate very hot. While the drives may themselves be within allowed temp range, the heat that is dissipated within the case may have some effect on the rest of the electronics. Everyone knows electronics does like heat. So depending on the tolerance of the rest of the electronics to increased heat, there will be correlation between the heat and problem with rest of the components inside the computer.  Personally, I like it cool with all electronics, unless it is unavoidable are expensive or inconvenient to control heat.

mkr

[quote author=685E7B44431E1C1C1A2C0 link=1177176215/1#1 date=1177259297]I'm probably one of the greatest proponents of hard drive cooling.

I had installed fans in all my hard drives about a year ago. Today when one of my computers failed to boot, when I removed the HD, I found the two fans both stuck and not working. After some time power on, they got hot. Looks like the lubricant dried up. I replaced the fan.

The moral of the story is; check all your fans atleast once a year. You can use a dentist mirror or some other means to inspect them. One of the users here mentioned that he lubricates the fans annually.

Rama

In my working life, I worked for four of the worlds largest corporations that design and build electro/mechanical devices.  I've worked on everything from a small manual cash register to a D-10 CAT and every machine I ever worked on requires Maintenance.

Computer maintenance is as simple as cleaning out junk files and refraging the HD on a regular schedule.  Or it can be more proactive, like cleaning and oiling your fans and blowing out the dust bunnies on a semi-annual schedule.
Maintenance just comes naturally to me, after doing it professionally for so many years.

After adding another three air intake fans to the front of my PC, I now have 15 fans in my system.  It just doesn't bother me at all to check them from time to time and keep them clean and running at top performance.  
PSU fans, scavenged from defunct power supplies, make really good case fans.  Besides being FREE, they are usually quiet too.  So my PC, besides being nice and cool, is nice and quiet too.  It sets right here beside me on my desk and I don't mind the purring sound it makes, at all.
My PC gives me so much pleasure and comfort that I don't mind at all giving something back in the way of TLC.

This is my "Baby" after I added an intake fan to the front panel.  With the red eye looking at me, it reminds me of the "HAL 9000". ;)
http://i54.photobucket.com/albums/g90/DrWho1943/Cyclops.jpg

Stay Cool!
The Shadow  8-)

That is a neat way to install the fan and the air flow would be excellent. Personally little purring does not bother me. BTW, when I inspected another machine, I found the fan dead and replaced it. Your idea of using the ps fans is neat.

*  :D

Recently, for the first time since building this system, many months ago, I've installed both side panels.  After several hours of operation I noticed that the front part of the side panels was warm to the touch.  That's heat that should be blown out the back, not hanging around inside the case and warming up the side panels. I installed the red fan in the upper front panel as seen in the picture above.  That took care of the heat being generated by the HD that sits right behind those upper drive bay faceplates.
But there was no air-flow to move out the heat being generated by the two HD's that sit in the drive bays in the lower front of the case.
To facilitate that, I added two more case fans to the steel frame, right behind the plastic front panel.  Again, I went to my box of recovered PSU fans.
This picture shows my 3rd hard drive in the upper drive bay, mounted with spacers and the place, in the lower case frame, for two more fans, provided by the case designers.
http://i54.photobucket.com/albums/g90/DrWho1943/FrontOpen.jpg

I didn't like all that grill-work that would block the airflow into the two case fans, so with end-cutters in hand, snip, snip, I removed the grills.
http://i54.photobucket.com/albums/g90/DrWho1943/Ready2Blow.jpg
Now the two new case fans can breath more efficiently and blow more cooling air back over my two HD's and move their heat into the back of the case where a large exhaust fan can remove it.
Now, with the case reassembled, everything stays at room temperature.
This added airflow also increases the efficiency of the CPU and GPU coolers.
In my 44 years of working on equipment that had the propensity to overheat and die, I've developed the same mantra as so many other tech's and engineers, "Heat Kills" (electronics).
If that were not true, then NO CPU or GPU would have a cooler affixed to it.  And, NO PSU would have one or even TWO fans installed.

I don't understand the 'Nay-Sayers' that would come on a HELP forum and tell people to just let their PC's run as hot as possible.  That's counter productive to say the least and just plain "Destructive" to say the most.

Windows in their Power Management app, has the ability to shut down a HD that's not being used for XX number of minutes.  Setting that to a low value for a system that seems to be running hot, can greatly reduce the heat that's coming from the HD's, at least.  I set mine to 10 minutes.
For the little "Mini" PC's, that can be exceptionally critical, since they have virtually NO airflow anyway.

As my grade school principal used to say, "A word to the wise is sufficient".

Y'all have a great day now, Y'hear?

Happy Halloween! http://i54.photobucket.com/albums/g90/DrWho1943/Flyingbat.gif

Rama wrote on Oct 23^rd, 2008 at 3:56pm:

[quote author=685E7B44431E1C1C1A2C0 link=1177176215/1#1 date=1177259297]I'm probably one of the greatest proponents of hard drive cooling. I had installed fans in all my hard drives about a year ago. Today when one of my computers failed to boot, when I removed the HD, I found the two fans both stuck and not working. After some time power on, they got hot. Looks like the lubricant dried up. I replaced the fan. The moral of the story is; check all your fans atleast once a year. You can use a dentist mirror or some other means to inspect them. One of the users here mentioned that he lubricates the fans annually. Rama

Last few days I noticed noise coming from the computer. First I thought CD was failing and even after disconnecting the power, the noise reappeared after a day. It appeared to be coming from hard drive. Usually when the bearing in HD fail, you start hearing noise. Got real concerned because it was the main production machine. After removing the HD, testing showed the noise was from the cooling fan. Few drops of watchmaker's oil (which does not gum up) in the bearings quietened the fans.

Moral of the story: Periodic lubrication - once a year, is a must to avoid failure of fans. Learnt hard way!

Done anyone know of hard drive fans with high quality bearings which does not need any maintenance?

Rama

Quote:

I don't understand the 'Nay-Sayers' that would come on a HELP forum and tell people to just let their PC's run as hot as possible.  That's counter productive to say the least and just plain "Destructive" to say the most.

I am not aware that anyone in this thread has expressed such a viewpoint.

If a PC is running outside of the manufacturer’s temperature tolerances, then corrective action should obviously be taken; if not, then none is obviously needed. For example, I am running three SAS 15K RPM internal hard disk drives in a HP xw4600 Workstation, and they only get ‘warm’ (not ‘hot’) to the touch, cooled by the fans installed in the case by the manufacturer. In this situation, there is simply no advantage to ‘fixing’ something that isn’t ‘broken.’

On a more general note, the BIOS in my workstation has a setting to adjust the default rate of the system fan (located in the rear of the chassis) when the processor is in idle. Some users may simply find it beneficial to increase such a setting to create more overall airflow in the case if needed, rather than installing a new fan in the PC enclosure.

Rama,

Quote:

Done anyone know of hard drive fans with high quality bearings which does not need any maintenance?

All the little $3 fans have sleeve bearings, just like the fans in most PSU's.
They are impregnated with oil during manufacture, and that's it.  When that oil dries out, they begin to HOWL!

Most of the supply houses that sell the cheap fans, also sell the expensive ones with 'Ball Bearings' in them.  Much higher price and much higher life expectancy.  Ya gets wot ya pays for. ;)

I just did my semi annual maintenance on my own PC when my GPU cooler was moaning.  I found several cooling fans on my HD's were in bad need of maintenance or replacement.

I replaced two of them and serviced the rest.  The GPU cooler just needed some TLC. (clean & oil )
Months ago, the little Stock 40mm fan on my GPU heatsink stalled and fried.
I replaced it with a 50mm CPU fan from an old K6 Heatsink.  There was no way to screw it down, so I secured it with HOT GLUE.  
Hey!....It works great and my GPU has never been so cool! ;)
http://i54.photobucket.com/albums/g90/DrWho1943/GPU_Cooler.jpg
Sorry for the blurry picture.  I don't have a close-up lens on my digital camera.

When that little fan finally gives out..... I've got a half dozen more just waiting to go back to work.  I save everything like that, , that I scavenge out of old computers.  Waste not....want not! (who said that?)

Like, all my new case fans are out of defunct PSU's.

By the way, I just saved a motherboard, by replacing the ten bulging and leaking capacitors, on the 5v rail.  It took me about an hour, all told, but it was well worth it  to save a motherboard in a nice little PC.
With a sizeable ram upgrade, a new HD and a bigger and better PSU, that little PC has a new lease on life.  VOILA !!!

I do wish y'alls a very good 2009!
The Shadow  8-)