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

Designing a Heatsink

T

The Pentium Guy

Diamond Member
At a school I'm going to (I'm going to 11th grade), I'm asked to undertake a heavy research project (this is a technical school, mind you).

However, this isn't one of those "make your own volcano"-type research project with the stupid vinegar/baking soda. Kids who have "succeeded me" (for the lack of a better phrase) have invented stuff like a device that police can use to slow down cars automatically. Lots of interesting stuff.

So I'm asked to invent something.

First a quick note - I'm not asking for "homework help" or anything like this. I have till August 22 to decide on the project - and since I have a strong passion (or devotion) for hardware, I felt that I could do something hardware related.

I'm going to research different materials for heatsinks, and probably create a new type of heatsink (ex: Heatpipes are an interesting technology --> how about a "new" idea like that?). Weight, material strength, surface area, noise, etc will be taken into consideration as well.

Well, here's the thing I'm worried about: Would this be feasable? I would have to contact some sort of manufacturer that does metal crafting right?

Again, I'm really not asking for homework help, just wondering if this is feasable and if this is a project that I should undertake (being a highschooler).

Cliffnotes:
-Is it feasable to design a heatsink? I'd have to go through steps such as contacting a metal-crafting manufacturer or something of the like.

Edit: Perhaps I should broaden the subject to "Designing a CPU Cooler" as opposed to "Designing a Heatsink". A cheaper alternative to phase perhaps? I've got about 6-7 months to do research on this (the kids I talked to who went there said they had long, sleepless nights doing research, heh).

-The Pentium Guy
 
What should I say - a CPU cooler is a big piece of metal that stays on the CPU. You could be able to find a way to make it small in one direction, or something - but it won't be such a big thing.
What I could suggest would be to have a phase-change cooler with water... Just think about it, it might deserve a long look. Even if you make it as a proof of concept and it works, it will be something.
Hmmm... metalworking, pipes and so on would be needed.
 
I was thinking about Phase with water. Phase is: Liquid at room temp, Gas at cpu temp. Right? Adding a solute to water (anything that increases its molality) would cause it's boiling point to go up. Fine tuning it by creating a low-pressure environment (not the case, the cpu cooler itself) would cause the water to evaporate even faster.

Not sure. This sounds VERY interesting. Thanks for the idea!
 
The problem is that in order to be able to design such a thing you more or less NEED to use computer simulations (at least if the idea is to design something that works well), if you have software like e.g. Femlab available that is doable but otherwise it is hard; writting something like that from scratch is almost impossible unless you are a FEM expert.

Take a look at http://www.comsol.com/ for some more info about Femlab.
It would probably even be possible to use the multiphysics capabilities of Femlab to simulate e.g. heatpipes or something similar.

Btw, Femlab is relatibely easy to learn if you understand the basics about the math that is being used; it is not as complicated as it looks.


 
Wow this is free? I'm definately checking this out once I get back home (I'm currently in india right now, on vacation). Thanks for the help guys, you've given me some confidence.

Only problem I'm left with is how to manufacture this 😉. Perhaps me trying to learn how to solder is the cheapest solution, heh!
 
Well interesting

However designing a heatsink is too easy
The characteristic of a heatsink only depends on material, size and surface area

You just need the best conducting material (Silver > Copper > Aluminium)
Silver is a bit expensive and impractical characteristic (Softer)
So you'd stick to copper.

And you just need a design with as much surface area as possible. So you make it quite big while being not too heavy and expensive. With as much fins, cuts to increase surface area as possible.

That's not too challenging
Designing a CPU heatsink however, makes you to take ALSO into account airflow, the CPU fan's mounting, the weight of heatsink (not to damage the CPU die)
 
Originally posted by: The Pentium Guy
I was thinking about Phase with water. Phase is: Liquid at room temp, Gas at cpu temp. Right? Adding a solute to water (anything that increases its molality) would cause it's boiling point to go up. Fine tuning it by creating a low-pressure environment (not the case, the cpu cooler itself) would cause the water to evaporate even faster.

Not sure. This sounds VERY interesting. Thanks for the idea!
Click to expand...

:thumbsup:
Yes, something like that. Play with the pressure so you could use some strange liquid. Or you might choose a different agent and a different pressure.
You might think at using ethilic alcohol (ethanol) as its boiling point is at some 76 Celsius (I think). Find different liquids, research each other (volume as gas, heat taken by evaporation, extra heat transported from extra heating (if any). You'll find some liquids working better than others.
Also, you could use a presurised liquid (instead as - with water - depresurised one). See how much could you pressurise the heat pipes, and how much it will help at keeping the CPU cool

Oh,you might want a special core block that is able to withstand the pressures/depressurisation. Not sure if normal water blocks will survive
 
You could do something with phase-change. There are some good guides on how to build them. Also, I'd use propane. It's not expensive, and cools almost as good as R402.
 
Take into account that propane is highly flammable (but ethilic alcohol is too). I would be much more afraid of propane than of alcohol
 
Originally posted by: Calin
What should I say - a CPU cooler is a big piece of metal that stays on the CPU. You could be able to find a way to make it small in one direction, or something - but it won't be such a big thing.
What I could suggest would be to have a phase-change cooler with water... Just think about it, it might deserve a long look. Even if you make it as a proof of concept and it works, it will be something.
Hmmm... metalworking, pipes and so on would be needed.
Click to expand...
Phase change with water in a computer will never work unless you can somehow operate at low pressures, which would be extremely difficult. You're already in serious trouble if your CPU heats up to 100°C to boil off the water.

Originally posted by: f95toli
The problem is that in order to be able to design such a thing you more or less NEED to use computer simulations (at least if the idea is to design something that works well), if you have software like e.g. Femlab available that is doable but otherwise it is hard; writting something like that from scratch is almost impossible unless you are a FEM expert.

Take a look at http://www.comsol.com/ for some more info about Femlab.
It would probably even be possible to use the multiphysics capabilities of Femlab to simulate e.g. heatpipes or something similar.

Btw, Femlab is relatibely easy to learn if you understand the basics about the math that is being used; it is not as complicated as it looks.
Click to expand...
:thumbsup:
If you have access to a combination of FEMLAB and MATLAB, you can do this pretty easily. Problem is, they're probably out of your price range, so you'd have to rely on your school to foot the bill.
Originally posted by: AnnihilatorX
And you just need a design with as much surface area as possible. So you make it quite big while being not too heavy and expensive. With as much fins, cuts to increase surface area as possible.

That's not too challenging
Designing a CPU heatsink however, makes you to take ALSO into account airflow, the CPU fan's mounting, the weight of heatsink (not to damage the CPU die)
Click to expand...
This isn't necessarily true. Geometry will also affect the amount of heat transfer and would be a very good basis for the project. Of course, to make a complex geometry will dramatically increase the difficulty and cost of the project, which is why heat sinks are made with such simple geometries now. 😛

If you're interested, I designed a water cooling system that I planned on building myself, but now it doesn't look like I'll have the time to do it. It's probably not involved enough to be your entire research project, but it might be worth looking at to give you an idea anyway. I abandoned it because the guys in cases and cooling hate science, preferring misinformation. 😛

Here is a short writeup on it. I also neglected to define my terms in the write-up, so here they are:

rho (looks like a P) is water density (~1000 kg/m^3)
Cp is the water heat capacity (~4.18 J/g-°C)
Q is the volumetric flowrate (300 gph)
T0 is the input temperature (chosen as 25°C)
T1 is the output temperature (calculated as ~25.2°C)
Pcpu is the CPU's power consumption (and, therefore, heat generation) rate (chosen as 150 W, but I don't know how close to reasonable this is)
 
Something that Police can use to slow down cars is something that anyone can appreciate. Do you think that a heat sink is the same in that respect?
I mean do you think the people who are going to grade your assignment understand the challenge or even the need for a heat sink?

I don't mean to burst your bubble. But, let's take a step back and think about what you need here. I am sure you know that there are engineers in companies whose only job is to do what you are thinking of. Designing a better cooler. I am sure that you can learn from them if some day you decide to become one of those engineers. But, right now, you want o pick a project that you can learn from and on the side can get a good grade for. Let's be realistic.
 
I would think you'd need to do simulations as well as have access to a machine shop to build and test your designs (almost all physicists are at least competent machinists).
 
Originally posted by: Navid
Something that Police can use to slow down cars is something that anyone can appreciate. Do you think that a heat sink is the same in that respect?
I mean do you think the people who are going to grade your assignment understand the challenge or even the need for a heat sink?

I don't mean to burst your bubble. But, let's take a step back and think about what you need here. I am sure you know that there are engineers in companies whose only job is to do what you are thinking of. Designing a better cooler. I am sure that you can learn from them if some day you decide to become one of those engineers. But, right now, you want o pick a project that you can learn from and on the side can get a good grade for. Let's be realistic.
Click to expand...
He can learn from this and designing it isn't overly difficult, given the right tools. Heat transfer is analagous to mass and momentum transfer, so learning these fundamentals would definitely give him a leg up if he were inclined to go into any sort of engineering discipline.

If you really want to look at something more complex, you can stick with one heat sink design and make a variety of fans. Hold the RPMs constant and change the impeller (blade) design. The difference you'll see is amazing, and the literature is readily available for different applications (see agitation in particular). If you need to solve the math surrounding the system, it will be more complex, but with simulation software it can still be done. I can tell you how to get an approximate analytical solution around a simple fin geometry as well.
 
Originally posted by: CycloWizard

Phase change with water in a computer will never work unless you can somehow operate at low pressures, which would be extremely difficult. You're already in serious trouble if your CPU heats up to 100°C to boil off the water.
Click to expand...

I don't think the goal of the research project is to create plans for a product ready-to-produce, ready-to-sell, cheap to produce and wanted by everyone. In this regard, I think I would admire something that is out-of-the-box thinking, strange use of very normal items and so on. Would it be hard to make a phase change cooler based on water? I bet. But this makes half the interesting part. Also, take into account that if the pressure increase, the boiling point rises, so there will be a very hard limit at the heat the system can accept. A proper design will take into account this, and other things. This will be heavy research.
 
Thanks for the encouragement guys. I feel this may be a good research project for me - we're asked to invent or improve something and make it a challenging project.

If you have access to a combination of FEMLAB and MATLAB, you can do this pretty easily. Problem is, they're probably out of your price range, so you'd have to rely on your school to foot the bill.
Click to expand...
Ah dammn. I thought FemLab was free. My school's got a lot of engineering software anyways (it's owned by WPI), when I go there (end of august) I'll ask if they have it with them. I have an uncle that works in MatLab (I think, I could be confusing that company with something else) - so I have multiple ways around this stubmling block.

Here is a short writeup on it. I also neglected to define my terms in the write-up, so here they are:
Click to expand...
Interesting - I read through this but I haven't understood it fully yet. I think I should pick up a Thermal Engineering book from amazon to help me out with this stuff. At least now I have a brief idea of what to take into account for while doing this.

Hold the RPMs constant and change the impeller (blade) design.
Click to expand...
You've given me an idea - earlier I was planning on making it so that a Panaflo or some fan can be mounted on the heatsink. If I have time, I'll look into building a fan into there. You've got a good point though: some fans concentrate air into a certain point (most ideal for cpu/vga heatsinks, and some fans blades spread the air all around (most ideal for room fans, case fans, etc) - I'll look into this. Fan mounting is another issue too. I'll have to decide whether to mount it on the side or on the top, or even 45 degrees. These are the type of things I'll be looking at in months to come.

Depends though - if I go the phase change route, I probably would be looking at a totally different set of ideas. I'll decide once I put my thoughts down and think.

Phase does sound cooler than air 😉.

Thanks guys.
-The Pentium Guy
 
Sounds like a good project, especially since u can bring marketing and stuff into it (good for marks 🙂)
Look for DIY Water Cooling / CPU Blocks, basically u jsut get a slab of copper, cut it, drill it out and ur done, its good fun, i did it for my old T-Bird.
 
I'm sorry, I did not mean to discourage you. I am glad that it looks like it is a good project after all.

Your uncle may be at Mathworks (company), which wrote Matlab (product/software tool),
or, Mathsoft (company), which wrote Mathcad (product).

Either is a very powerful and popular tool, and can be used for simulating your project.
 
Yeah, it was Mathworks actually 😉 - that's a good thing.

You didn't discourage me by the way Navid, in fact you helped me out a bit with this:
"I mean do you think the people who are going to grade your assignment understand the challenge or even the need for a heat sink?" - After realizing that not everyone is a hardware freak 🙂)), I think I should explain the purpose of this cpu cooling kit.

Actually, I heard from my peers that our other science teacher is a "Visiting Scholar" (I think that means that she voluntairly teaches at our school) who works at Intel (heh, let's not turn this into an Intel/AMD flamewar), so she should be pretty knowledgable about this.

-TPG
 
Looking at this thread, I didn't even imagined so much work/research is put into one phase change cooler... or even a normal heat sink/fan.
Looks like you have a good bit of research and work on your hands
 
Originally posted by: Calin
I don't think the goal of the research project is to create plans for a product ready-to-produce, ready-to-sell, cheap to produce and wanted by everyone. In this regard, I think I would admire something that is out-of-the-box thinking, strange use of very normal items and so on. Would it be hard to make a phase change cooler based on water? I bet. But this makes half the interesting part. Also, take into account that if the pressure increase, the boiling point rises, so there will be a very hard limit at the heat the system can accept. A proper design will take into account this, and other things. This will be heavy research.
Click to expand...
Oh, I'm not doubting it would be impressive. However, there's a reason water isn't used for computer phase change: its boiling point isn't even close to the operating temperature of a processor. The operating temperature of the processor will be very close to whatever the boiling point of the substance is. Without checking the books, I can say with reasonable certainty that he would need almost a complete vacuum to actually do this with water and achieve any sort of success. Since the water needs to be pumped, this is pretty much impossible.
Originally posted by: The Pentium Guy
Interesting - I read through this but I haven't understood it fully yet. I think I should pick up a Thermal Engineering book from amazon to help me out with this stuff. At least now I have a brief idea of what to take into account for while doing this.
Click to expand...
There is a free heat transfer book available online called "A Heat Transfer Textbook" by Lienhard. If you look around for a while, you might be able to find the download for it (probably gutenbergproject.com or some such). If not, PM me and I can send it to you.
You've given me an idea - earlier I was planning on making it so that a Panaflo or some fan can be mounted on the heatsink. If I have time, I'll look into building a fan into there. You've got a good point though: some fans concentrate air into a certain point (most ideal for cpu/vga heatsinks, and some fans blades spread the air all around (most ideal for room fans, case fans, etc) - I'll look into this. Fan mounting is another issue too. I'll have to decide whether to mount it on the side or on the top, or even 45 degrees. These are the type of things I'll be looking at in months to come.
Click to expand...
You'll see that the field of impellers is pretty well-developed. There are correlations (most of them proprietary, but I've done some work in this field so I actually have data that I'm willing to share) for the amount of thrust that a fan develops at different rotational frequencies for a given fan geometry.
 
I kind of glanced through this thread, but I always thought that heatsink design required a serious amount of thermodynamic fluid simulations which would in turn require the understanding of concepts that are beyond the typical grade 11 student. Maybe I'm completely off base here, but I always thought that modern heatsinks required more design than just welding a large number of copper fins onto a base, especially for using something like phase change cooling or heat pipes.
 
Raynor,

It all depends on how good he want it to be. A real simple (read inefficent) heat sink is a block of copper/alumium with a fan. However, if you want to achieve something more impressive then you are looking at CFD (computation fluid dynamics) and worrying about air recirculation, case temperatures, and a multitude of other issues (not to mention investigating alternative materials/costs of manufacture/etc).

So relax... let a kid have fun. Otherwise he will need to build a workstation, buy a copy of Fluent or CD-Star and model a case in 3D. Then he can figure out the effects of various fan configurations, hot spots and temperature generators in the case, and the effects of plugging (etc) over time.

"I wanna build a heat sink" Sure... no problem...

"I want to build the most efficent heat sink in the world for cpu.." A little more work

"I want to build the best (cost/efficency/size/etc) heat sink in the world for cpus." That might take a while.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top