Thermoelectric notebook cooler (Peltier)

How this Intel Centrino 2 notebook overclocked to 4GHz would work? Can i make a custom one?

 

Looks like this cooler connects to the back vents directly, then uses fans to blow hot air from and out the exhaust in the back. Should be relatively easy to make, Really unnecessary though, a fan underneath blow cool air in would do a good enough job

 

TEC is pretty easy to make yourself. Made one many years ago while I was in undergrad (electrical engineering) but it didn't work well lol. Good thing I was testing it on a cheap mobo/Athlon CPU.

 

I'm a pretty firm advocate of just getting a lapdesk/notebook pad that gives some decent ground clearance and has a fan will cool your machine off enough. If you're getting overheat problems while gaming, then you've OC'ed it too much, or you need repaste your CPU/GPU. If you're going to lug around a massive fan setup like that, you might as well just tote around a desktop with handles for gaming.

 

not to mention that if you plan on using a peltier on your lap then remember that while one side(your machine) will be chilly, the other side(your privates) are going to be scorching hot.

 

It's too bad not one peltier laptop cooling pad product exists. I guess one could aim a window air conditioner at your laptop and call it good. Or, if you have central air, put your laptop over the register.

 

repasting with IC Diamond is probably one of the best things to do. It dropped by temps 10-20C depending on load Also drilling holes in the bottom of the case will help a lot

EDIT: are there any how tos?

 

Yes.. people often forget the other side of the plate is hot as can be. If anything you have to have superior cooling to run a Peltier system.

 

Let me explain what that is, it's a prototype. It was never mass produce. It was a coolitsystems and intel project. The notebook had overclocking features and the product was made to demonstrate those capabilities. You can see in your picture that the cooler is on a Intel Developer Forum demonstration.
It was an OCZ whitebook aka Flextronics (Arima) W840DI aka Alienware M17.

Basically it's just a custom built notebook cooler with fans and peltiers. The fans matched the notebook exhausts.
Of course most links are all dead now.
http://www.coolitsystems.com/index....-intel-laptops&catid=2:press-release&Itemid=3
http://forum.coolitsystems.com/index.php/en/home/6-cool-talk/4666-mtec-docking-station.html

I tried to buy it several times and contacted coolitsystems, intel, ocz, flextronics and many others but the product was never commercialized.
Imo it was a very good idea but anyway you can build one easily. Just add some tec units to a notebook cooler.

Here's the old link:
http://forum.notebookreview.com/ocz/364912-ocz-whitebook-cooler.html

press release1:

press release 2:

Looking back, 140$ is really expensive. Costume made for a specific model. including the rear side where current notebook coolers don't ever cool. It won't work properly on other models.
Basically it's a box with holes on the bottom and rear coinciding to the notebook fans. The fans push the air from where the notebook pulls and extracts the air from where the notebook expels the hot air. you can see in the rear that there's prominent part, that's to coincide to the rear fan without blocking the rear I/O ports. It should also have a left side where there's the cpu and northbridge coolers and fan but it just has a bottom fan for it. It didn't had an on/off button or anykind of speed variation. It's just plug and play like other coolit products.
Funny thing the notebook had some design issues when using some cpu's. Sometimes it was required to use a copper shim to make a good contact. Problem discovered by this forum. Also had some issues with outdated ATI drivers.

I'm really tired of seeing those IC Diamond "miracles". There's no way whatsoever that a thermal compound can drop temps that much. What happens all the time is that the stock thermal pads are badly installed and there's a huge deficiency. IC Diamond or any other thermal compound will drop the temps by that much. Some are better and cheaper then IC Diamond, some are not. also if you have a good thermal compound already correctly installed there's no way you'll see any significant drop in temps by using IC diamond, actually the improvement can be 0 or less then 0.

 

whatever...the 20C difference is because of old cheap compound. But I guarantee you that a well placed cheap low quality compound will be at least 10C worse. It is common knowledge.....plus physics....

 

Care to share your miracle benchmarks with pics and test methodology?

Don't bother.



Even mayonnaise is less then 5C worse then IC Diamond. I don't see how you do those miracles of 20C difference. You must be really special.

And it has some disadvantages vs other compounds, it can be much more expensive, it's not suitable for dry ice or similar, it scratches off the cpu due to the "polishing" effect voiding cpu warranty.

 

yea...thats a fake. I run folding at home v7 with cpu and gpu. I had to run my GTX 260m at 300 750 500 for it to stay around 80C with my modded cooler (I think it hovered around 76-80C when room was cooler and then up to 82C or something). I put diamond on it and i can overclock to 580 1450 950 and it runs at....hold on turning it back on....oh it idles at 46C cpu and 60C GPU depending if it gets a little colder at night it is 56C for gpu. Now its running 55-58 CPU and 74C gpu with overclock. Also the case fan wont run faster then 3370rpm. It would run at 4500-5500rpm with the underclocks with the old paste(it spins up at about 82C). If I ran this with the stock clocks it would run up to 85-90C with old paste with cooler. Without cooler it would hit 95-100C...i wouldn't let it run any longer then a few mins before i turned the cooler back on because thats way to hot. Note this is how my G51 has always been. I also have the back plates drilled to no end Let me turn my cooler off and see how it runs. I remember it getting to 82C then fan spins up and cools it down to 72C and constantly does that the whole time because the fan wont spin faster then 3370 until it breaks 82C. (note this was when the room was a lot colder)

Well it is constant at 65C cpu and 82 GPU with 4500rpms. My room was at least i think 10-15 degrees F colder last time i did this so i assume that is why it is not cycling between 72-82C. I put the AC unit in the window a little while ago because I thought it was going to get hot but Chicago's weather is so finicky lately. I put it in early so my room was freezing the other day when i did this. I had to where sweat pants and a hoody and i was still cold. Now I am wearing shorts and a white t-shirt and I am comfortable. I don't have a temperature gage so I can't really tell what the temp is...sorry.

Sidenote the cooler is a rosewill cooler with 3 92 mm fans that I replaced with vantec tornados running them at 5V with 2.7amps? I think. They run at like 40% capacity i think. So maybe a total of 120-150CFM blowing on my laptop. I will eventually take photos and create a thread with everything I have done. I still need to add heatsinks.


Also before I switch to IC Diamond I would play games (with overclocks) with the cooler off and it would run at about 90-100C depending on what is happening in the game. This is much less of a load compared to running fold at home full speed but most games dont use much cpu. Now with folding at home without cooler runs at most 82C Also note when I was playing games the case fan runs at 5500-6000rpm so now it runs at 82C in a warmer room at only 3370rpms. That is an easy 10-20C if it would run at 5500-6000rpms it would be cooler then 82C. Ask anyone with a G51J that it will hit 90-100C easy even with back plates drilled. Now it wont break 82C


EDIT: if that is not a fake then they are testing that on a desktop with a very good cooling system. Note 32C is like 90 degrees F so that is only slightly higher then the common room temp of 70ish degrees, which is ~21C. Thermal compounds shin at higher temps. So if you have a very good cooler...aka a water cooler or a heavy duty copper heatsink with a beefy fan they will all run that 32-40C. Especially if your comparing idle. If you have a water cooled heatsink the quality of paste really doesn't matter because the heatsink is so good. Now when you work with laptops with small heatsinks the higher quality paste matters more because the temps normally run much hotter. I am talking about numbers in the 60-100C range, which is 140-212 degrees F. If you ever went to school or had a good science/physics class you would/should have learned that heat transfers faster at a greater difference. So we can deduce that a 20-50 degree F difference from room temp is not going to very much between thermal paste. A 70-142 degrees F difference between room temp will show the difference from a thermal paste to a much great level. Also you could go as far as comparing the % difference in each test. I am decent at math but I think you would do something along the lines of say the best score is 11C off of room temp and the worse is 19C. You can see a ~42% increase in cooling efficiency that way or is it like ~72%? I am not exactly sure how you want to compare that. I forget exactly which way you want to divide numbers and what not. For me I can't quite do that because I am not sure what my room temp is....I would say from 50-60 degrees F depending on when I did this but I can't be exact. Also reason why mine is such a drastic increase is because my room temp is much cooler then theirs....assuming they follows the standard protocol of ~70 degrees F for room/base temp. My numbers are completely realistic because it is not just happening but because you have not included all the variables into the equation. This is really quite simple. It is basic science you learn in school/practical experience. A pie cools quicker in the window instead of an oven(off of course)

EDIT:
http://en.wikipedia.org/wiki/Laws_of_thermodynamics
http://en.wikipedia.org/wiki/Heat_transfer

 

care to share your pics and scientific test methodology? Do you have any proof to support your claim? because any serious review around shows you are wrong. do you work for ic diamond?

But yeah you must be one that are right and every serious review is wrong.



arctic silver 5 vs ic diamond only 2C difference.
IC Diamond 7 Carat Thermal Compound vs Arctic Silver 5 - Icrontic

proof that IC diamond scratches and corrodes the cpu:
ic diamond 7 abrasion ? - Overclockers Forums
Yeesh IC Diamond 7 is some crazy stuff - AnandTech Forums
Yeesh IC Diamond 7 is some crazy stuff - AnandTech Forums

also is hard to apply, extremely thick. and expensive.

I never said it wasn't good. It's one of the bests. top10 I would say. I said it doesn't make miracles, isn't the best, there's cheaper options with better performance without serious cpu damage issues. and 20ºC difference is a miracle. and if anyone believes in 20C difference it has to believe in better then that because there is better compounds. 20C difference won't happen on Earth.

 

This is getting good!

 

the only way ive ever gotten a 20c drop in 10 years of desktops/notebooks is by not seating the heatsink properly.

literally about 30 computers all high end for their time... (i upgrade multiple times a year, im retarded)

i own MX-2,MX-3,MX-4 and ICD 7. Arctic ceramique, AS3 and AS5 is also in my stash. ive used all of them. none of them make a massively noticeble difference.

none of them saved my Fermi G100 from hitting 100c in furmark @ high volts w/ high clocks 100% fanspeed ....


ICD isn't a miracle but based on its material it has the highest theoretical potential.

 

We are doing a series of video's on IC Diamond one of them deals with compound application and removal. In the segment on removal we generate an oxide layer on copper that is only 100 Angstroms (a hydrogen atom is about 1 A in diameter, a carbon atom is about 2 A in diameter) or 10 nanometres thick (1% of a micron) Your typical laser etched lettering is approx. 200 to 500 % thicker @ 2 to 5 microns.

In this extreme example we load at 50lbs then remove the compound with Zero damage to the oxide coated layer.

We then demonstrate the delicacy of the coating by taking a soft Kleenex and polishing it to a bright copper shine in a few seconds and then follow with a second example using a bare finger to get the same polishing/ oxide coating removal with a few seconds also.

With proper removal/application IC Diamond is less abrasive than a soft cotton cloth, Kleenex or even a bare finger.....All it takes is a decent solvent and a minute or two of your time to do.

I'll have the Video up in a week or so.

 

whatever man...i know what is happening to my laptop is accurate and you keep quoting extremely low temps. Higher it is the more it will show. I don't need to waste my time arguing with you. Also you haven't stated anything about those reviews....all i see is numbers and no details on what tech they were using to test this. I gave you all the info I have and you give me these half completed charts.


Also the abrasive parts....nearly all thermal pastes are that way according to your forum links....so??? Who cares if almost all of them are abrasive lol

Lastly that link on AS5 and IC diamond 7 is dumb. First you say its hard to apply...have you looked at IC website to see how you apply it??? It is easy to apply and is not difficult. Granted it does come out of the tube a little tuff but are you serious?!?!? Also that link from the AS5vs IC diamond is so stupid....I mean the review is an idiot. He is talking about heating up the thermal paste to make it easier and spreading it evenly over the whole area lol....he must not have read IC website on how to apply lol.

/\
|
lol too funny
Application