In reply to Paul's message of Mon, 9 Oct 2006 18:23:27 -0700 (PDT): Hi Paul, [snip] >That's basically the idea. The broadband UHF radiation >varies from material to material. The peak wavelength >of such radiation is considerably lower for >electrically conductive materials such as iron for >obvious reasons. Under normal core usages the >radiation is incoherent, but if you magnetically flip >all the spins at relatively the same time you then >have a coherent pulse; i.e., tons of potential energy.
I suspect that if you place the entire device in a cavity designed to resonate at the emitted frequency, then the resultant standing waves in the cavity will help cohere the microwave emission. This is analogous to he electron bunching effect in magnetrons. Essentially, you have a "forced" oscillation, where the forcing energy is supplied by the device rather than from an external source. BTW, I think that if you have "single atom domains", you will find that all the domains are effectively the same size, and consequently that should all emit at a single frequency (provided that the magnetic field is uniform across the device), or have I misunderstood? > Note that in such a case a great deal of the energy >is not absorbed by the material because it's in a >state of flipping to saturation. Furthermore, by tapping the microwave radiation and rectifying it, you are "sucking" it out as it were. Not exactly true, but what you normally have is an equilibrium where energy is being both emitted and absorbed. By attaching a "one way valve" to the energy flow, you ensure that the equilibrium constantly shifts in the direction of outflow. It's also what you are trying to do by operating near saturation. In fact before the invention of diodes magnetic saturation was widely used as a crude means of rectification. > > > >> You mention small domains as being >> advantageous. Could this be attained by reducing the >> density of >> the active atoms? IOW could you simply use a >> compound that is >> essentially an insulator, with say only one active >> atom among ten >> "insulator atoms"? That would appear to result in >> domains >> comprising single atoms. > > >That's an interesting idea. It should work. It would >probably decrease the materials saturation and >permeability. Your idea is somewhat similar to >nanocrystalline material. One of your ferromagnetic >atoms surround by insulation could be single crystal. >A single nanocrystalline material is one large >ferromagnetic crystal. Well, large as in a dozen or >more atoms in diameter. > Use of a diode may mean that saturation is no longer so important, and as to permeability, a loss in this regard, would just result in a smaller power output, if have understood correctly. However I don't see much use (yet), for 50 MW in the average home, so a drop in power output to say 10 kW shouldn't really be a major problem. What you really need to look at is power density, i.e. power/kg of mass, then compare that to other power sources such as e.g. an internal combustion engine. Since you are going to be processing considerable power, your device will either get very cold very quickly, or you need to ensure rapid transport of heat to it. The latter is relatively easily done by making the active component of your device a flat plate, and binding it to a thin metal plate which forms one side of a water pipe. Then water flowing through the pipe can carry heat to the device. (This has the advantage of also providing chilled water for air conditioning in the summer). The pipe needs to constructed from a metal that won't interfere with the operation of the device (copper?). However the device itself also would ideally be a good thermal conductor, while at the same time being a good electrical insulator. The only substance I know, that comes close is BeO. (highly toxic BTW). Regards, Robin van Spaandonk http://users.bigpond.net.au/rvanspaa/ Competition provides the motivation, Cooperation provides the means.
