AC-to-AC, PM parallel path concept.

[nwman]

Like some of you I am on the hunt for the Holy Grail a permanent magnets OU transformer. Most of you reading this have a lot more knowledge and experience in this field then I do so I will try to keep it simple.

It seems there are a lot of inherent problems with a transformer that incorporates permanent magnets [PM]. Most of us know what these problems are from Lenz to polarity saturations and DC pulse characteristics. In simplest terms if you can make the primary coil(s) run off AC and some how have the secondary coil(s) collect an AC current it should overcome the primary flaws of this concept and operate like a normal AC transformer. Of course the question is how can you do this? There are many great idea out there but “none” are “conclusively” working. If there was then none of us would be here. Moving on.

So I started to think. Scary I know.  I looked at all the PM magnetic flux control designs that have a primary coil that can run off AC. I looked at the simple parallel path concepts shown below.  As you alternate the polarity of the control coils the PM flux moves from one side to another. From numerous reports it takes relatively little power to redirect the PM flux. The general consensus is that it creates 3(+) times the magnetic field density then what the coils can produce alone. So “they” thought to put secondary coils around the side connecting bars to collect the flux. For what ever reasons this has yet work. Well you still run into Lenz because the secondary coils are being fed a DC current. So I looked at it and tried to figure out how I could use the same configuration but get AC to be fed into the secondary coils.

[nwman]

Long story short I simply asked myself what would happen if I just attached an identical configuration to the other

side of the connecting bar [A]? But with the polarity of the PMs flipped [ Shown below]. So now if you connect all the control coils together in the right order you should be able to alternate the current in the control coils in a why that make the PMs flux alternate from one side to another in each side. Now in one polarity the connecting bar [A], with the secondary coil, will experience one direction of polarity [Config 1].

When you switch the control coils polarity then the flux switches sides in both sides and thus the direction of polarity in the secondary coil [A] alternates [Config 2]. As long as the fields behave in this manner you should be able to input AC and pull out AC.

[nwman]

So my idea is to simple take the configuration shown and bend it into a loop [shown Below ‘Top View’]. This way the ends are closed and everything is compact.

[nwman]

Now I would guess most of you would say that the flux line would rather jump across to each other then to switch sides. Though I haven’t tested this exact configuration yet I do believe that the path of lease resistance will be to act like described above. From just playing around with a few configurations with magnets and cores that I have I think that when the EM control coils turn on they will create a path of less resistance then the path between the magnets themselves. If the PMs are closer to the control coils then they are to each other then it would be a better path. Also there is no air gap between the PM and their control coils and there is two air gaps between the magnets. Plus from videos shown on YouTube.com as one side’s field gets stronger the other side gets weaker.

In conclusion, I plan to building this. However before I dump a few hundred dollars again into a build I would like to get feed back and see where I have gone wrong. Or if there is a cheaper way of testing these principles. Or if I have over looked some Law that would prevent this from working.

Thanks,

Tim

[BEP]

@nwman

Please forgive my butting-in on a subject I generally avoid. I have enough experience with these devices to concentrate on other efforts.
I consider the following true on my bench. I'll stop before proclaiming them true for your bench. I don't know for sure. They are sure enough for me. I'll use conventional terms as they should be enough to get the idea across.

The are many problems in using a PM as part of a transformer. Not the least is the fact conduction is a part of this even though there is no conventional 'flow'. Nonferrous magnets will not 'conduct' in the needed way. Most 'problems' I've seen posted seem to be nothing more than excuses.
The idea that a flux can be completely redirected or turned off/on is crap without parts physically moving.
You cannot reduce a connected Mag force, you can only change the area covered - compress or expand it while the circuit is near-field. Once it is far-field then you can do most of the things done in an electrical circuit.

So the only way to do what these devices claim to do is to change the magnetic circuit connection from near to far field and back.

Sounds even more nutty? Not at all. 'Load-Bank' is term I am very familiar with. You have common resistive types - just a whopper of a decade box, not much more. Then you have inductive load banks.

The majority are fixed transformers that may have winding taps and they may be switched in and out of circuit to apply the correct reactive load. Then you have variables. Conventional ones with a knob and a wiper to vary the inductance and those very expensive ones that have a set of control coils. They are simply magnetic amplifiers with a fixed load. The gate coil controls the amount of the input power allowed to reach the load.

The main principal behind all the inductive load steps, except the ones that work like a Varactor, is the what is going on in the core.

The core is not a continuous repeated iron layer upon layer. In calculated placement are interveining layers of aluminum, copper or others.

The ones with copper/brass layers have a coil covering the whole branch of the core. When energy is applied to this coil it creates a 'connection' across the diamagnetic layers. So during the cycle the overall core conductance actually varies.

These were neat to understand but the ones using paramagnetic material are even more fascinating.

Same core setup except periodic layers made of aluminum:
As mag field approaches aluminum the aluminum acts ferromagnetic but only during the approach and removal. In short: It can act as a switch.

So using Al as a junction splice or corner overlap in laminations can dramatically change the hysteresis curve.
If you wrap the area of the Al junction with a coil you gain some control over magnetic conduction, at that location.

A lot of hot air above but high perm metal cores or not you must have strong variations in that core with controlled fields over those variations or you will never have much use for the device.

Slicing a gap and stuffing Al in it will not work as mag fields simply expand and go around. Putting an insulator in that gap will do nothing but prevent the full strength of the 'on' state.

Oh well, I'm done. If you can make sense of the above then also understand I've never achieved more than a bistable magnet. Sort of a magnetic flipflop. Speeds were too slow to be of any use for anything. On the ones that would change at useable speeds they weren't wanted after the first test.