Hi Ion, F6LT, all,
As I mentioned, I'd rather not focus on any particular version of this network of concepts as 'the one', because it leads to excessive disputation. This is exactly the situation I was referring to. It's easy to latch on to a specific example, and find holes in it, and of course some holes may exist, and that particular idea may not work. My focus is not "does this device work?" but "is there a principle here that can be applied to overunity?" These are not the same questions.
I've only just begun to proliferate examples. Each of them has elements that are different from the others, but each support the concept that secondary fluxes can be made to appear to be nullified so that they don't load the primary. Or another, and perhaps more fruitful way to think of it, reactive power can be converted to real power.
I included Alek as a recent example which MAY overcome issues of increased magnetizing current due to gaps, but he may not be correct, of course. I think Jensen is on solider ground. It doesn't appear that Thane Heins gets it, since without gaps, his device is a type of 'load sharing' transformer where all secondaries still load the primary. There appear to be a lot of variant designs, so I'm not sure about that.
I agree that an absence of loading in a transformer can happen because the core goes from more saturated to less saturated. It may also occur just because the transformer is so inefficient that the load is hardly detected. The situation in some of these devices is somewhat different from that. Take a look at JLN's test of the two secondary loading, without consideration of overall efficiency. I'm using this as an example because it is already on the net, but I've seen the same exact phenomenon in tests with four secondaries, where loading each successive secondary increases the power output while reducing, or maintaining the same, power input. There is also a detailed patent from inventor Jim Murray that shows this same effect in a rotary device. Etc. Etc.
http://67.198.37.16/mirrors/jnaudin.free.fr/2002.06.03/html/smep11.htmIn this example with two secondaries, with B1 having a 10 ohm load, and B2 open circuit, the power input is 124.04 W. When B2 is shorted, the power input drops to 114.38. It seems pretty clear to me that, whatever else is going on, some reactive flux from B2 is now active flux at B1.
The white crow principle of William James says, 'even if only one true example of something exists, then that thing exists'. If only one example where reactive power can be converted to real power exists, then that principle exists. If you find a problem with these particular examples, I can show you more if you want.
The argument that 'if something like this was true, people would have discovered it early on' although in many cases true, is not a scientific principle, any more than if I were to state, 'people have observed this over and over, so it must be true.' Neither one has anything to do with testing a hypothesis but are a priori arguments from common sense. But common sense can be wrong, and scientists do occasionally discover something in plain sight, for instance, the discovery in the last few years of a new organ in the human body, the interstitium.
Your own statements show that if this 'reactive to real power' conversion were to actually occur, most engineers would attribute it to reduction in core loss. "Seen it before, move on". And, as I've already mentioned, the efficiency of most small transformers does not reach the levels where the effect would stand out. Using three separated coils on a transformer is not common practice commercially because of increased leakage flux. Most of the patents (indeed from the late 19th century on) that use a gapped EI core are welding transformers where the center leg is controlled to control the current. There is no practical usefulness and some inefficiencies in such a design otherwise, so it is not really common practice that I've seen. So the idea that people throughout the history of transformers have been observing the power output from two secondaries on the legs of a gapped EI core looking for some anomaly is not part of the historical record. Jensen also points out that the gap must be very precise in order to avoid losing the excess energy in the gap. So, historical welding transformer inventors, without having Jensen's rationale, would not have been designing transformers with a specific gap tailored to the particular transformer at hand and, in any case, would not be driving the transformer from the middle, but from one of the legs to the opposite leg.
But these are all a priori arguments too. Only tests would show.
Finally, the situation is not as simple as it seems. What is 'flux nullification' or cancellation, for instance? Obviously when two opposing fluxes are superimposed in a high mu core, there will be a portion that leaks out of the core, and a portion that remains in the core. Half way between those coils, a test coil that does not intersect leakage flux will not show (much of) a voltage. But is there an absence of magnetic energy in the core? No-- in a simplified model, each domain is subjected to equal and opposite MMFs and thus don't align with either flux, but the magnetic energy is still there, as 'stress'. You can DC power two opposing coils on a core, and when you shut off the power, there will be a back emf spike from both coils. The energy that was invisible before can now be partially recovered.
These are all basically interference phenomena, and the cases of 'total positive interference' and 'total negative interference' are not covered well in the literature on these things-- the assumption being that positive and negative interference must balance out in any particular situation. However there are concrete examples where this is not true, for instance in nano optics, where light shining through micro holes that are less than a wavelength apart has an increase in light intensity on the other side. I believe all these phenomena are the same, whether flux or light. But that is just a belief.
I could go on (and on :-) but if there is no consensus that these devices could be gainful, I'd be happy to move on to parametric devices, which have enough complexity for anyone :-) My brief is not to defend any particular device, or concept, -- or even the concept of 'overunity'!-- but to increase efficiencies and reduce losses in electrical equipment by using little known historical data. That is certainly doable.
Regards,
Fred