Heatplane instead of heatsink?

[rif]

I've been reading the PLH posts and docs, and an "opportunity" re power JFETs dissipation abilities is mentionned a few times.

Wouldn't the use of a heatplane instead of a heatsink greatly help? Of course you still need to transfer the nrg to ambient surroundings, but the heatplane would very efficiently move it away from the power JFET.

just a thought.

[azira]

Intuitively, I would say that a heat plane would not work as well as a heat sink. Note that I am not saying that a heat plane would not work at all.

A heatsink is fins attached to a slab of metal. That slab of metal is essentially your heat plane. So then you can consider a heatsink as a heat plane with fins which should always out perform a simple heat plane.

Of course there is a matter of size. At some size, a heat plane would be able to perform as well as a given heatsink. But that size is considerably larger (think relative surface area).

Perhaps what you might be thinking of is a heat spreader (such as used on many CPU coolers). Many have hunks of copper embedded in an aluminum heat sink becuase the copper is more efficient at sending that heat around while the aluminum is better at radiating it.
--
Danny

[rif]

Maybe I didn't use the correct term -- I'm thinking of something like a heatpipe, but in a plane instead of a pipe.

Basically a heatplane would be a hollow plane of metal. Inside the hollow is a wicking material and typically a phase change substance. The heat is transported very efficiently by the phase change substance -- instead of relying on diffusion of heat in a solid metal to carry energy, it relies on a mobile liquid/gas.

I found a link:

http://www.tsheatronics.co.jp/englis.../01_carry.html

[Mr Evil]

Heatpipes don't actually dissipate heat, they only move it from one place to another; you need a heatsink on one end of them anyway. They are used because it is not always possible to mount a large enough heatsink on the heat source itself. A heatpipe allows the heatsink to be located somewhere convenient, distant from the heat source.

Unless you cannot mount the transistors directly on a large enough heatsink, because of space constraints or whatever, then using a heatpipe (or 'heat plane') will not do any good.

[rif]

Yes, they only transport heat -- you still have to get rid of it.

But the idea is to get the heat away from the active device so that it does not overheat. You then dissipate the heat to the ambient environment where temperature rise isn't as important.

The fundamental thing to remember is the difference between heat and temperature. Move the heat away from the active device as fast as possible to keep its temperature low.

http://www.tsheatronics.co.jp/englis..._ex/index.html

[jackinnj]

i have a few dozen of these -- with a little ingenuity you could fashion this into a heatpipe -- 4 chambers and sufficient wall thickness to allow you to tap for 6-32's or 8-32's -- anyone into welding aluminum?

[Mr Evil]

Originally posted by rif
Yes, they only transport heat -- you still have to get rid of it.

But the idea is to get the heat away from the active device so that it does not overheat. You then dissipate the heat to the ambient environment where temperature rise isn't as important.

The fundamental thing to remember is the difference between heat and temperature. Move the heat away from the active device as fast as possible to keep its temperature low.

http://www.tsheatronics.co.jp/englis..._ex/index.html

Unfortunately it doesn't work like that. If you insert a heatpipe between the heat source and sink you are adding thermal resistance and thus the temperature of the heat source will increase. The example on the page you linked to shows using a heatpipe in place of a plain metal plate, where you will see an improvement because a heatpipe has (hopefully) a lower thermal resistance than metal, but it's still more resistance than if you attached the heatsink directly to the heat source!

You could integrate a heatpipe into a heatsink, where the heatpipe distributes heat to the fins instead of a metal baseplate. That would be a more practical application for audio amps. Indeed, some commercial heatsinks do just that.

[rif]

Originally posted by Mr Evil

You could integrate a heatpipe into a heatsink, where the heatpipe distributes heat to the fins instead of a metal baseplate. That would be a more practical application for audio amps. Indeed, some commercial heatsinks do just that.

That is what I meant to communicate -- guess it didn't come across that way. I always intended a traditional heatsink at the other end -- and by moving it away from the heat source, it frees up some design constraints.