Heatsinks

[SciManAl]

clear

[clocker]

OK, I'll bite.
Define "passive".
Does this mean no pump or no active cooling (i.e. fan), or both?
Is there fluid involved?

Check this out...a beautifully executed "passive system".

[SciManAl]

clear

[lynx]

As you are obviously aware, maximizing surface area is the desired objective. The mistake many designers make is in assuming this only applies to the metal/air interface. However, it is equally important to make the most of the coolant/metal interface in your radiator design, but without restricting coolant flow.

One of the best ways to effect this is to feed the coolant into a manifold, from which you can split the flow into several small tubes. At the other end the tubes would combine into another manifold which would lead the coolant back to the heat source(s).

The result is that if you split the flow into 4 pipes of half the diameter (thereby keeping the same cross-sectional area), you will double the surface area between the coolant and the metal. This is similar to the effect you were talking about with flattened tubes, but I think you would find it easier to drill round holes into any fins rather than try to make fins to match your flattened tubes.

Having said that, there must be any number of radiator manufacturers who would be able to meet (and exceed) your requirements with off-the-shelf products at a price which would be much lower than you could hope to match by making them yourself. It depends whether you want the satisfaction of a totally self-built product or just one that works.

[SciManAl]

clear

[clocker]

Al, I assume you are referring to the Innovatec passive radiators that I was interested in. I never actually got them... I now think that the concept is flawed.
The problem with passive air cooling is that air is a great insulator. Simply trying to dissipate great amounts of heat into still air is a losing proposition. Try removing your fan from the heatsink on your chip for an exciting ( and potentially expensive) demonstration.

I have seen ( can't remember where ATM) a radiator setup that consisted of long lengths of copper tubing that were laid out on a concrete floor. As the concrete was always shaded ( and thus cool) it provided an excellent medium to transfer the heat from the pipes.
The builder of this setup had the pump and reservoir in the garage also with only two waterlines extending into the house, so the net result was a very quiet and effective passive system that was fairly cheap to build.

It would also provide you with plenty of opportunity to exercise your soldering skills too.

[SciManAl]

clear

[_John_Lennon_]

Originally posted by clocker@10 February 2004 - 21:48
Al, I assume you are referring to the Innovatec passive radiators that I was interested in. I never actually got them... I now think that the concept is flawed.
The problem with passive air cooling is that air is a great insulator. Simply trying to dissipate great amounts of heat into still air is a losing proposition. Try removing your fan from the heatsink on your chip for an exciting ( and potentially expensive) demonstration.

I have seen ( can't remember where ATM) a radiator setup that consisted of long lengths of copper tubing that were laid out on a concrete floor. As the concrete was always shaded ( and thus cool) it provided an excellent medium to transfer the heat from the pipes.
The builder of this setup had the pump and reservoir in the garage also with only two waterlines extending into the house, so the net result was a very quiet and effective passive system that was fairly cheap to build.

It would also provide you with plenty of opportunity to exercise your soldering skills too.

I remember seeing on some forum something similar to that as well.

Except he ran the 30 feet of copper tubing, outside his house and under ground, in one nice big loop.

It runs at about -5C or so I believe. But he has a Pelt on it with those temps.

[clocker]

Yeah, underground would be excellent.
Depending on where you live and what the frost line is, I should think that 3-4 feet deep would do nicely.
The Pelt seems like a bit of overkill to me... I could live with +2-5 deg, I think.

[_John_Lennon_]

Originally posted by clocker@10 February 2004 - 22:37
Yeah, underground would be excellent.
Depending on where you live and what the frost line is, I should think that 3-4 feet deep would do nicely.
The Pelt seems like a bit of overkill to me... I could live with +2-5 deg, I think.

The pelt he already had, and just wanted to see what temps he could get with it.

I dont think he kept it on though, if im not mistaken he gets low teens to single digit temps, in C all day, all night.

After all, the group take a long time to asorb heat, and a long time to dissipate it if im not mistaken.