HDD cooling

honegod

honegod

[H]F Junkie
Joined
Aug 31, 2000
Messages
8,327
I am at the HDD mount / cooling stage of my build.
I have seagate 4tb 2.5 drives shucked.

In the USB external case each drive got quite warm, reasonable @ 0 airflow.

Implying that heat needs to be dissipated.

So... where is the heat being generated ?

The bare drive with prototype copper rubber mount brackets bolted to the sides only got slightly above ambient.

But the whole circuit board on the bottom got noticeably warmer.

So... Is that heat being generated BY the circuit board
or is the board merely transferring heat produced by whatever is UNDER the board ?

The Question = anybody remote mounted the drive circuit board and Knows what could use a heatsink in there ?
 
So... where is the heat being generated ? -> use the finger test and start touching everything (note, this doesn't apply to all life situations :D)

anybody remote mounted the drive circuit board and Knows what could use a heatsink in there ? -> no, but if the drive didn't come with why, why worry about it at all?
 
Drives are built to a specific size & shape to be mechanically interchangeable.

Since I am in the process of designing and constructing my harddrive mounting I am not so constrained.

If i desire to put the board waaay over there, with the mechanicals over here I, unlike the industry, am free to do so.

Yah, I did an open air 30gb write and caressed the drive all over.
Since fingertips are less temperature sensitive than lips I then smooched the drive all over.
Only the circuit board was noticably warm, no hot spots, an even warm over the entire surface.

No hot spots implying no one or 2 things getting hot & that heat spreading through the entire board.

More like some hot things UNDER the board spreading the heat by convection to the underside of the board.

Hence wondering what is under there getting hot.
 
well that seagate will die probaly...I do not know....still got a 4 year old WD blue and literally still nothing wrong with it....
 
Batboy88 said:
well that seagate will die probaly...I do not know....still got a 4 year old WD blue and literally still nothing wrong with it....
Click to expand...

Heck yeah the magic smoke WILL escape.

4X4 tb drives in raid 1 as placeholders until whatever comes after M.2 hits 16tb @ $100.

The mounting is easy to hotswap, tooless, everything that actually touches the drives is 99.99 copper for heatsinkyness, fully enclosed in a duct leading to the big hole where the powersupply was supposed to go,
( by everything I mean, I could NOT find copper allen screws to screw into the mounting holes, but I DID find 3mm copper rod, a m3 die and vise grips = copper allthread.)

So... The time is now to figure out this heat source thing, solution, so I can get started on making the tooling to make the parts to get the drives in my box
 
honegod said:
Heck yeah the magic smoke WILL escape.

4X4 tb drives in raid 1 as placeholders until whatever comes after M.2 hits 16tb @ $100.

The mounting is easy to hotswap, tooless, everything that actually touches the drives is 99.99 copper for heatsinkyness, fully enclosed in a duct leading to the big hole where the powersupply was supposed to go,
( by everything I mean, I could NOT find copper allen screws to screw into the mounting holes, but I DID find 3mm copper rod, a m3 die and vise grips = copper allthread.)

So... The time is now to figure out this heat source thing, solution, so I can get started on making the tooling to make the parts to get the drives in my box
Click to expand...

It's actually Toshiba that has a large drive wouldn't mind having, huge cache man....

but idk seagate has them too.
 
Batboy88 said:
It's actually Toshiba that has a large drive wouldn't mind having, huge cache man....

but idk seagate has them too.
Click to expand...

2.5" drives
Seagate 4tb external USB portable.

@ $100 on sale, often.

15mm thick though, so they do not fit almost EVERY hotswap bay, but still less tall than 3.5" , way shorter, and about the same width WITH my heatsink/isolater mounts in place.
 
Most of the heat a spinning hard drive produces is from the swirling air inside generated by the platters. The friction of the platters spinning at 5400/7200/10K/15K rpm heats up the air inside. This heat is transferred to the external aluminum drive case surrounding the platters.
That is why some manufacturers use helium inside their drives. It's not to make them lighter, but rather to reduce friction to generate less heat.
Cooling the entire hard drive case is what keeps the internals (platters) cooler and prolongs their life.
Generally, the same airflow that is keeping the case cool is also keeping the electronics cool.
 
Azrak said:
Most of the heat a spinning hard drive produces is from the swirling air inside generated by the platters. The friction of the platters spinning at 5400/7200/10K/15K rpm heats up the air inside. This7the heat is transferred to the external aluminum drive case surrounding the platters.
That is why some manufacturers use helium inside their drives. It's not to make them lighter, but rather to reduce friction to generate less heat.
Cooling the entire hard drive case is what keeps the internals (platters) cooler and prolongs their life.
Generally, the same airflow that is keeping the case cool is also keeping the electronics cool.
Click to expand...

I like the primary heat source being the 'viscosity' of the working fluid inside, I can see helium, under a light vacuume even, being less efficient at converting mechanical energy into heat.

What I was primarily thinking about was the bits that convert electrical energy into mechanical motion, like the motor that has to create the 'air' motion by spinning the discs,and the servo bit that yanks the RW arm to & fro against the friction of the air.

The platters and arms would have to share, with the air, all the heat produced by air friction.
And the only thing available to cool the platters and arms would be the air inside the drive.
The same air that is being used to GENERATE the heat that they need to get rid of.

So it sounds like I need to completely fin the top cover to draw heat out of the working fluid.
fins on all sides to extract heat from the body
And DEFINITLY yank the bottom circuit board to access the biggest area that has hot fluid on the other side.
Just moving the board a little to allow SOME airflow to the bottom of the drive case would HAVE to help.
 
honegod said:
would HAVE to help.
Click to expand...
Help what?
Keep it cooler? Prolong the device life?
I fail to see the ultimate goal you are trying to achieve both here and in the fan thread.
The level of effort you are seemingly willing to put in to reduce temperatures is far beyond what would ever be noticeable to the lifetime of the device being cooled. This is why you don't see hard drives with fins all over the top and bottom covers. The extra surface area doesn't make enough of a difference to the lifetime of the product and only adds manufacturing cost and design complexity.
Drive manufacturers already assume that the drive will be in some sort of airflow stream to keep it within normal operating temperatures and they design the whole hard drive with that in mind.
http://knowledge.seagate.com/articles/en_US/FAQ/193771en
A couple degrees of temperature reduction through device modification isn't worth the extra effort of money it would cost to implement, plus you risk damage to the very thing are trying to protect.
If you are doing it just because you can, well... fair enough. Everyone has a hobby or passion. But it's just a 4TB hard drive.

I take it you are a min/maxer type of person, based on your other thread discussing overvolting fans. Nothing wrong with that, either.
Yes, putting fins on a hard drive will reduce the inside temperature by a degree or two, as will moving the circuit board to a new location.

As a side note, I'm fascinated to know how you will be extending those flex cables that go from the board to the inside of the hard drive case while keeping reliability the same as stock.
 
Azrak said:
Yes, putting fins on a hard drive will reduce the inside temperature by a degree or two, as will moving the circuit board to a new location.

As a side note, I'm fascinated to know how you will be extending those flex cables that go from the board to the inside of the hard drive case while keeping reliability the same as stock.
Click to expand...

The initial concept is that i need to build a harddrive bay to accommodate four 15mm tall 2.5" drives.

A design consideration is cooling the drives.

So... I need to find out about what exactly needs cooling, how best to cool it, design a cooling solution i can construct, THEN finalise the mounts to best utilise the cooling setup.

Since my googling Sux i came here, the best place on the planet to find someone who had already been down this path, to get up to speed.


I also look forward to gaining skillz at remote board mounting
 
If they are still within spec you don't need to do anything. Did you ever look at the actual temperatures and see?
 
Still haven't been in cases in years...drives don't get as hotttt and neither does the psu...a few psu's like seriously got pretty warm/hot to the touch all closed up in the case....and everything else is cooler too. you're extending the life of you're shit man lol
 
Azrak said:
Help what?
Keep it cooler? Prolong the device life?
I fail to see the ultimate goal you are trying to achieve both here and in the fan thread.
The level of effort you are seemingly willing to put in to reduce temperatures is far beyond what would ever be noticeable to the lifetime of the device being cooled. This is why you don't see hard drives with fins all over the top and bottom covers. The extra surface area doesn't make enough of a difference to the lifetime of the product and only adds manufacturing cost and design complexity.
Drive manufacturers already assume that the drive will be in some sort of airflow stream to keep it within normal operating temperatures and they design the whole hard drive with that in mind.
http://knowledge.seagate.com/articles/en_US/FAQ/193771en
A couple degrees of temperature reduction through device modification isn't worth the extra effort of money it would cost to implement, plus you risk damage to the very thing are trying to protect.
If you are doing it just because you can, well... fair enough. Everyone has a hobby or passion. But it's just a 4TB hard drive.

I take it you are a min/maxer type of person, based on your other thread discussing overvolting fans. Nothing wrong with that, either.
Yes, putting fins on a hard drive will reduce the inside temperature by a degree or two, as will moving the circuit board to a new location.

As a side note, I'm fascinated to know how you will be extending those flex cables that go from the board to the inside of the hard drive case while keeping reliability the same as stock.
Click to expand...
#TSRH

You're wasting your time with this. Just set up a stack with at least 1/2-3/4" separation (13-19mm), and use one fan to pull in fresh air across them. If you want to maximize airflow cooling, then don't place them in the "eye" of the fan output (where the hub is), and keep them in the blade wash area.

Otherwise, you're just pissin' in the wind ...
 
idk if it makes much a difference...might have a better chance of killing ssds writing to them excessively....idk the mushkin saying like something crazy and getting up there like 307gb total writes and it's still ok lol. I wouldn't worry about it honestly...still have old HDDs that work too....
 
It sounds like Plan A is best then.

A, consists of lapping the sides of the hdd to get to clean flat aluminium, a thin coat of noctua paste, and bolt on the, copper, vibration isolater mounts.

Facing the 200mm intake fan, surrounded by a duct sealed to a big open vent at the back of the case.
 
honegod said:
It sounds like Plan A is best then.
Click to expand...
Yeah, plan A is good except for the part about lapping the drives and putting thermal paste on them. The only meaningful effect that's going to accomplish is voiding the hell out of whatever warranties your drives might've had. Any fan blowing over the drives is entirely sufficient on its own to keep them cool.
 
honegod said:
It sounds like Plan A is best then.

A, consists of lapping the sides of the hdd to get to clean flat aluminium, a thin coat of noctua paste, and bolt on the, copper, vibration isolater mounts.

Facing the 200mm intake fan, surrounded by a duct sealed to a big open vent at the back of the case.
Click to expand...
We'll, if you're gonna ignore our advice, at least do a build log & don't forget to cover the failures too. You're DEFINITELY gonna brick at least one of those drives.
 
You must log in or register to reply here.
Back
Top