Peltier cooling device

[Pixels]

Out of interest......what would be the considered upper and lower operating temperature range of a standard specification modern hard drive?

Samsung quote 0 - 60c operating.
But temperature is not the issue - relative humidity is the issue. If that reaches 100% then dew will start to form.
Cold does not necessarily = dew

As to whether dew can form inside a drive, it may depend on the relative humidity that the drive has been operating in, which over time will effect the relative humidity inside the drive.
For those that live in areas of high humidity, cooling a drive may well push the relative humidity to 100%. To that extent, I'll jump to BlackST's defense.
Consideration should also be given to the small internal volume of a drive, relative to the surface area of all the components.
Except in extreme cases, there may simply be not enough water vapour present to form droplets of any significance and in any event, would be forming on the internal surfaces of the cover and frame first as these would be cooling much faster than anywhere else.

I will pick BlackST up on one point though, it's not the warming of the drive that's causing dew to form, it's the warmer air coming into contact with the colder surfaces, causing that air in contact with the colder surfaces to reach 100% RH and dew to form. But maybe that's what you meant - it's all "relative"

[ktheking]

Thanks Pixels. Yours was the first scientificaly based anwser I've gotten.
Indeed humidity level is a key factor.
Getting this cooled in a dry environment will be the only to give you a second chance of trying to recover the data from the drive.

For the other responses : Yes ,I agree , changing the faulty equipment is a better way of dealing with the problem altogether.

But my main purposes are to keep costs low,speeding up the process and rendering a general solution. As pointed by Scott in one of his presentations : 80% of recovery can be handled by software. Adding 5% by solving it with cooling solution makes it an impressive 85% without having spent 10K on equipment.

So when building my home made solution I should consider air humidity.
And also at what rate the disc will get back into room temperature.
If the disc will go at a slow rate back to room temperature,no condens will be formed either.

Thus some kind of formula should exist for condens forming containing : humidity level,start temperature,end temperature ,time and surface.

http://us.kaeser.com/Online_Services/To ... efault.asp

Another way alltogether (simpeler,better) would be to have a complete cooled environment in which the drive works,so that the air floating over and trough the disk is the same outside as inside.Making no temperature difference,thus no condensation (all humidity in fridge will condens on the coolest part in the fridge,which is the cooling module).
When shutting down the fridge the coolest part will always ,remain the cooling module. Adding a big chunks of metal,will make sure nothing will be on the disk.
Playing around with one of those mini fridges ,in which you drill a hole to pass the cables would be the best solution. Those will bring also the temperature back to room temperature in a very slow rate,so that no condensation is formed,and a different method can be tried.
Adding a outside temperature reader will be needed also.(to see when you can take out the disk out of the fridge again).

Kind regards,

K.

[BlackST]

Clarification : the mechanism is exactly what Pixels stated : warmer air coming into contact with the colder surfaces. People usually freeze the drive and immediately turn it on (btw... think about the PCB too...) otherwise the method would be useless. I have seen too many disasters to give it even a "thought of trial".

[Doomer]

I guess I have to teach physics here too

Some acronyms
AH - Absolute humidity
RH - Relative humidity
DP - Dew point
AP - atmosphere pressure


1. DP occurs when colder surfaces contacts with warmer air if AH of air is high enough to produce dew. Physics of this process: warmer air contacts with colder surface and gets cooled down near the surface, RH tries to balance itself which drops AH in cooled air causing DP. DP is only possible if air gets cooled down fast enough. DP is not possible when a surface is warmer then air around
2. Let's see what happens when we put drive in the freezer: metal parts get cooled down then air inside gets cooled down (there is a small possibility of DP in form of ice appearing on metal parts inside of HDD if drive gets cooled down very fast but I don't think it's possible in normal freezer in liquid nitrogen more like it). When air inside HDD gets cooled down two processes happen: (a) Air loses AP (b) AP gets balanced inside and outside HDD causing some air to come out of HDD. So there is a major conclusion: when drive frozen AP of air inside is not enough to produce DP and pressure inside HDD equals AP
3. Now let's see what happens when we take the drive out of freezer: obviously DP occurs and dew appears on outside parts of HDD. But air inside is still cool and with low AP. The only way for dew appear inside HDD if we put some warmer air inside. There is NO WAY air gets through the breathe hole inside HDD UNTIL drive is warm enough because pressure inside and outside of HDD equals AP at the moment we take the drive out of freezer
4. Some other facts. Metal has higher temperature conduction then air, so the only way for air inside to get warmer is to get warmed up through metal parts, so metal parts inside HDD will always be warmer than air inside when we take HDD out of freezer. If we power up the drive there is even lower chance for DP because on air convection and vaporize processes


As you see there is only one way for DP to occur inside HDD - in the process of freezing the drive, so if you take really hot drive and put it into really cold freezer and cool down metal parts very fast and drive is not spinning at this moment all that together could cause DP and some ice may appear inside but temperatures should be extreme

Last edited by Doomer on April 12th, 2010, 11:08, edited 1 time in total.

[BlackST]

As you see there is only one way for DP to occur inside HDD - in the process of freezing the drive, so if you take really hot drive and put it into really cold freezer and cool down metal parts very fast this could cause DP and some ice may appear inside but temperatures should be extreme

Usual freezer temp. = -20 , drive after running for a while while trying to pry data out and reading internet stuff, about 40 (all deg. C) or even more. Delta = 60 deg. C. Is it sufficient ?
NB. the mass of a 3,5" drive maybe is somehow "big", what about a 2,5" ?

[Doomer]

That's a "maybe" here
But physics is different
Dew may appear not because somebody have taken drive out of freezer but because one put it in
That's my point

So if you want to cool down you drive don't forget to spin it while freezing

[guru]

I'm going to put a disk in the freezer and take pics.. Any choice of HDD ?

[Doomer]

I'm going to put a disk in the freezer and take pics..

I'm wondering how you squeeze camera inside HDD w/o opening it
Because if you open it the whole process will be different

[zebong]

i am reading this post from the 1 day , and is very interesting

[dobrevjetser]

When air inside HDD gets cooled down two processes happen: (a) Air loses AP (b) AP gets balanced inside and outside HDD causing some air to come out of HDD. So there is a major conclusion: when drive frozen AP of air inside is not enough to produce DP and pressure inside HDD equals AP

Doomer,

I would think that, since hot air that is cooled down takes less volume, there will be underpressure in drive when it is cooled down, so AP gets balanced by air going INTO the drive.


Dobre

[dobrevjetser]

@KtheKing


80% of drives fixed with software is only possible if you refuse a lot of drives.
About half the drives that come in here are fallen drives or drives with headfailure.
Anyone can explain me how to fix these with software


Dobre

[Doomer]

Doomer,

I would think that, since hot air that is cooled down takes less volume, there will be underpressure in drive when it is cooled down, so AP gets balanced by air going INTO the drive.
Dobre

Yes you actually right
Air goes INTO the drive and air inside freezer has very low AH, so it makes chances for DP inside HDD even lower

[Alexii]

Freezer : in 1 hr. or less you have dew growth inside when you take it out...

This would be incorrect
Physics doesn't work this way

PS: Strange thing is I discuss the same topic at least once a year
PPS: more details here - http://forum.ixbt.com/topic.cgi?id=11:22373 (sorry it's in Russian )

Very intertaining read =)

[dobrevjetser]

Depends if your freezer is a standing model or a box model.
Box model is no problem. Cold air remains below.
If its a standing model, and you are not doing it fast, all air in the cooling department can be replaced with air coming from outside the freezer, with higher AH.
Depends also if air in hd is cooled down faster than the air in the freezer compartiment.
Depending on what air is cooled down faster, air could go into or out of the drive.
At this level, we are beginning to talk about probabilities


Dobre

[Doomer]

If its a standing model, and you are not doing it fast, all air in the cooling department can be replaced with air coming from outside the freezer, with higher AH.

Now we talking about different pressures then

BTW when we take drive out of freezer air will go OUT of the drive which will not bring warmer air inside HDD -> no DP inside

[BlackST]

I have an interesting experiment to do. I'll discuss it behind "locked door" tonight , Rome time . Stay tuned...

[craig6928]

try dry ice works great with drives

[Pixels]

Doomer, whilst what you say about AP and AH are true, they are of much less concern than the change in RH for a given change in temp.
Over a temp range of 30 - 0c, there is about 10% change in AP, but about 80% change in RH.
For this same reason, DP and the formation of ice or frost [if DP is 0c or less] is largely independent of rate of temp change.
A 10% change in AP has only very minor effect on DP.

It really, [simply put] depends on what the initial RH is inside the drive which will determine the DP or FP temp.

[guru]

thinking of using a WD raptor with clear lid ;o)

[Doomer]

It really, [simply put] depends on what the initial RH is inside the drive which will determine the DP or FP temp.

That's why there is silica gel granules in the breathe filter and sometimes in recirculation filter
BTW I attached DP table for everybody who may be interested in it

Attachments

Dew point table

DP.jpg (109.64 KiB) Viewed 8837 times