The Kozeka Principle Part I

Primary references:—

http://www.kedroncorp.com/abstract.html [now only available archived; e.g. at https://web.archive.org/web/20080615010715/http://www.kedroncorp.com/abstract.html], and search http://peswiki.com/index.php/Main_Page for "kedron".

Kozeka principle:— a difference is claimed between the energy delivered
when two magnets attract horizontally, and the energy required to separate them vertically

Summary of operating principle

In 2007 Dr Kenneth Kozeka claimed that there was a difference between the energy delivered when two magnets pulled together horizontally, and the energy required to separate them vertically. He claimed that two NdFeB38  3/4 inch cube magnets "... are capable of generating 7.46 inch-pounds (work) when they pull themselves together “sideways” in the horizontal plane. It takes only 6.56 inch-pounds to pull the magnets apart along a vertical path that is perpendicular to the path they followed when they came together. This leaves a .90 inch-pound net-yield of mechanical energy (work) which can be used for example to turn an electric generator."

Kozeka's graph of attraction force vs distance. The units for the axes are not stated

I decided to model Kozeka's idea exactly as described above (except that I used NdFeB39, which was already in my program's library, instead of NdFeB38.) Ansoft Maxwell v11 was the magnetostatic modelling program used. The results when plotted out gave the graph shown below.

My results for attraction force vs distance

Integrating the attraction force curves to find the energies gave negligible energy difference between them, i.e. 0.6957J = 6.157 inch-pound versus 0.6929J = 6.132 inch-pound. (The difference is well within the 1% energy error I used to define "solved" force calculations).

Further Investigation

These results showed one potentially interesting feature: although it is hard to see in my graph above, the horizontal force goes slightly negative for a while. I decided to investigate this further, with larger but thinner magnets.

Modelling of two 96 × 60 × 8 NdFeB45H magnets

For the two 96 × 60 × 8mm NdFeB45H magnets shown above, this effect becomes much more prominent (note how the blue horizontal force curve goes significantly negative beyond about 61mm of separation):—

Attraction force vs distance for two 96 × 60 × 8 NdFeB45H magnets

Can the "energy well" be avoided?

Once again, if the magnets are brought in from "remote" positions, there is no net energy difference. But, if it were possible to start the horizontal movement with the magnets separated no more than about 61mm, and to keep returning to such a starting position with little or no energy penalty, there would indeed be a large excess of energy to be gained. Unfortunately, it seems to be impossible to avoid the energy "well" involved in bringing the magnets together to this desirable starting position. I have looked at repelling instead of attracting magnets; at movement along each of the three orthogonal axes; at diverting the flux of one or both magnets into steel "shields" over part of the operating cycle, etc, but have never found any way of avoiding this energy "well", to the extent that any significant excess energy can be gained.

A version of the Kozeka principle with the magnets enclosed within low-hysteresis
steel shields over the energy well until they have reached the desirable starting position.
However, no net energy gain has been found with this approach.

Next time I'll look at versions of the Kozeka principle where, in one case, a significant energy gain is indeed achieved (but it is probably only a "one-off" gain).