Steve Eddy said:
To further confuse people:
Metals are either ferromagnetic, antiferromagnetic, or paramagnetic when referring to bulk properties, when the electron spins are all oriented in one direction(fm) when they are locked in equal and opposite direction (afm) or when spin orienations are random ie their sum is ~0. I believe both Al or Cu qualify as paramagnetic. There may be temperature below which they become antiferromagnetic but am not sure.
Now when one introduces nuclear spins in the picture (ie if one is sitting in a 24 telsa magnet for example, then the paramagnetism-diamagnetism distiction start to matter. A paramagnet is by def something that tends to get toward the magnetic field and diamagnet is something that tends to move away from it. being repelled from the magnetic field. The parameter that quantifies paramagnetism and diamagnetism is called magn. susceptibility I guess the positive or negative signs of susceptibility describes a diamagnet vs. a paramagnet. Don't quote me on that since I haven't checked the handbook of phys and chem in a while.
I believe that in our amp we can safely limit our worries to bulk metal properties. I hardly think your transformer stray flux will amount to gausses.
Peter,
on a more practical aspect, I believe Cu is softer than Al or maybe they are about the same, however, it's not always the case that a softer material is easier to handle. Wood is an example, the temperature of the tool will probably be higher when drilling wood than when drilling Al.
Maybe the problem with the cutting tools and Cu is that Cu tends to bind and coat the cutting edge of the steel tool making it dull.
Just a thought.
Hi Peter,
Copper comes in several verities. But it is rather difficult to machine. It’s rather tough, as do some flavours of bronze. For proper machining you need very sharp tools with the proper cutting angles.
Copper comes in several verities. But it is rather difficult to machine. It’s rather tough, as do some flavours of bronze. For proper machining you need very sharp tools with the proper cutting angles.
..it is rather difficult to machine..
Cu work hardens quite rapidy, and obviously just at the wrong moment ( and place ) ..... i.e. just as you cut it, and just under the cutting tool 🙂
there are ways to anneal it to make it softer again ( heat to cherry red ..... ) but thats not really practical for delicate electronics work, and it does oxidise the surface.
and as soon as you start to cut or work it..... guess what happens 😉
ray
Cu work hardens quite rapidy, and obviously just at the wrong moment ( and place ) ..... i.e. just as you cut it, and just under the cutting tool 🙂
there are ways to anneal it to make it softer again ( heat to cherry red ..... ) but thats not really practical for delicate electronics work, and it does oxidise the surface.
and as soon as you start to cut or work it..... guess what happens 😉
ray
since copper is much more expensive than aluminium, why bother?
You get a much larger aluminium heatsink this would have a much lower thermal resistance than a equal-in-cash copper heatsink.
/rickard
You get a much larger aluminium heatsink this would have a much lower thermal resistance than a equal-in-cash copper heatsink.
/rickard
Solid Snake said:
You want to build a heat exchanger to air, so you need lots of surface area, and rapid thermal conductivity from device to heatsink surface. The choice of material is based mainly on economics. Copper is much more expensive than aluminum, but does have much higher thermal conductivity. (Copper is much heaver, softer and corrodes easily, so is less advantageous on those grounds too.) So people just use twice as thick an aluminum plate to achieve similar thermal conductivity in a lighter heatsink.
Copper is now used selectively in CPU heatsinks now, as a heat spreader under the CPU, and that is a very good compromise, giving high thermal conductivity where there is the most heat to move. It is also used in the lead frames of power transistors (Kovar alloy), with tin plating for corrosion resistance. If you look at the cut edge of many TO-220 devices, you see copper - the die is on a copper block for heat spreading. Many metal TO-3 packages have a copper heat spreader under the die too. Again, a very sensible selective use of copper.
Incidentally silver has even higher thermal conductivity than copper, and would be the purist's choice for heat spreaders. It is of course used in the highly effective thermal compounds for CPUs, such as Artic Silver 3, and makes a dramatic difference using just a thin film of silver paste. (Since it is electrically conductive, it might not be wise to use it on audio gear, unless you are very careful with it. ) I have also seen silver loaded elastomeric insulators for transistors, efficient but expensive.
Probably our heatsinks would be more effective (at minimal additional expense) if the devices mounted to copper heat spreaders on the aluminum heatsink. One problem is bonding the dissimilar materials, copper and aluminum. You might have to do a pressure weld, the way US coins are made today, so it would have to be a high volume part to amortize tooling cost.
Good question, and you probably are hitting on something we will see in future heatsink products.
- Status
- Not open for further replies.