AFA Alternating Flux Amplifier

[dieter]

Project AFA Alternating Flux Amplifier


Hello everybody, this project was initiated by the investigation of the Figueras enigma: It seems to me, nobody really knows the secret of the Figuera generator, and after some rather blind trial and error tests I became a little tired of digging in the darkness. Then I became aware of an idea that was like a revelation: What if the middle , secondary coil is no coil at all, but a permanent magnet? That would have explained everything. I soon found several patents about this principle: guiding a magnetic path, make it switch paths to get a max change of the Field with little input energy.


But could this be real? Did anyone ever test these patents? Or is it just another fake thing that none ever managed to replicate successfully?


I quickly decided to do some fundamental, simple tests, and I have to tell you, this was totally encouraging. I posted the following test results already in the figuera thread, but the design  with petmanent magnets makes it kind of off topic  there, so I thought I should start a new Topic.


So these are the Results of the first principle check...

To those who said I should post results of my permanent magnet version, Alternating Flux Amplifier AFA, here's some information.


I do now definitely know that PMs are a real option.
I made a simple setup: a steel core 5mm, length about 15cm. In the center there's a coil with a fine wire and many turns, probably thousands. It has a high impendance, but picks up small inductive fields pretty well. It's winded on a plastic tube that fits nicely on the 5mm core.


On both sides of this coil I got two low impendance coils. They are smaller, less turns, thicker wire, maybe 0.3mm.
They don't fit very well as their core diameter should be about 10mm, so they are loose. They are connected in parallel, with the halfrectified pulse of a 50 hz 12vac supply source, so these pulses are the positive part of the sinuswave only, with 50% idle time between pulses. The output of the middle coil is rectified by 4 diodes.
Note: my voltmeter may show silly values, but this is about the diffrence between a setup with and without magnets.


Voltage was measured without a load:
With two Neodymium Magnets attached to the ends of the core: 225 vdc


Without the Neodyms: 140 vdc


With the Neos with wrong polarity attached: less than 140, maybe 90 vdc


The Amps were measured in parallel [EDIT: IN SERIES!] with a 1.8 kOhm Resistor load, the voltage dropped almost to zero, but this is only about the diffrence between magnets and no magnets:


With Neos: 260 DCmA
Without: 200DCmA


Then I used little steel rods instead of the Neos, to see if it was only a higher inductance that caused the gain, but nope, voltage and ampere were lower than without anything attached.


This test has clearly shown: attaching 2 Magnets to the ends of a dc pulsed core increased voltage and ampere, if attached with the correct polarity. Using the wrong polarity, even only on one of them, but especially on both, decreased the output substancially, by more than half.


Also interesting: a test with a resistor between supply and setup , so the whole thing would not run with the full power of the 7.5 VA supply, reversed the result, adding the neodyms reduced the efficiency in general. This means, the gain by PMs requires a certain core saturation in relation to the PM strenght and core size.


Conclusion: Permanent Magnets can truely generate electric current! OU seems to be only a matter of proper implementation of the inductive coupling.

I will attach an image of the simple test setup.


I envite all of you to join this project, because here we got some really interesting test results already, no need to decipher a secret, the theory is simple and logic, yet obstructing maxwells theorem etc., said to be impossible by "accepted theories".All that is now required is some good knowledge in building efficent transformers with close to zero losses... once that works, everything from the permanent magnet will be gain!

[dieter]

Here's also my first drawing of the idea. This one would be more effective than the simple test, since it does not only  block the permanent field, but switches the path of it , making use of it at any time.
It may be a bit harder to construct, unless an iron torroid is used (contrairy to the assumed double C core), which on the other hand makes the winding a lot of work.


There is also the controversal issue of pathseeking and nonpathseeking flux involved, that I want to discuss with you later.


It would be great to see some of you verifying my results.

[gyulasun]

Dear Dieter,

First I did not want to comment your above asa_test1 setup and then later I have noticed you edited your text  "The Amps were measured in parallel [EDIT:in SERIES] with a 1.8 kOhm resistor load",  i.e. you eventually used the Ampermeter in series with the load resistor, at least I HOPE!

Now what I think is happening in your setup is that your input half waves establish a DC magnetic bias on the B-H curve, and the half waves 'ride' on the DC bias: this means that actually two magnetic fields interact with each other in the rod, and I suspect that when the correct permanent magnet poles are not present at the rod ends,  the rod may get saturated at the positive peaks of the input pulses because the DC bias can shift up the core's magnetic operation point on the B-H curve where the B induction already reaches the saturation level of the core where the permeability is reduced, hence the induced output becomes less. 
AND when you attach the magnets with the "correct" polarity, the magnetic operation point of the core gets shifted down into the 'more linear region' of the B-H curve,  i.e. the current peaks cannot reach the upper saturation area of the core hence the permeability does not get reduced and the induction can increase.

You can check this saturation problem in two ways. You could either

1) change the input voltage polarity of the half waves by swapping the input wires of the primary coils at the diodes and see whether changing the permanent magnets polarity also to the opposite may or may not cause again a similar output voltage level or
2) you may use a high value capacitor in SERIES with the input to block the DC level of the half wave pulses, say you connect the positive leg of an 1000uF 35V electrolytic capacitor to the cathode point of the upper diode and you connect the negative leg of that cap to the common point of the input primary coils where earlier the cathode of the upper diode was connected.

One more notice on the measured current, I quote your text with the measured currents:

" The Amps were measured in parallel [EDIT: IN SERIES!] with a 1.8 kOhm Resistor load, the voltage dropped almost to zero, but this is only about the difference between magnets and no magnets:

With Neos: 260 DCmA
Without: 200DCmA "

just considering the 200mA DC current via the 1.8 kOhm resistor, the voltage drop across this resistor should have been 0.2A*1800 Ohm = 360 V ! and this means that you should have been able to measure a similar order of voltage across it but you found the voltage dropped almost to zero, as you wrote. Sorry but I have to ask whether you correctly used the voltmeter and the ampermeter i.e. in the voltmeter mode your meter was in parallel with the 1.8 kOhm and in the ampermeter mode your meter was in series with the 1.8 kOhm?  There is too much difference between some hundred Volts and an almost zero voltage level...  and it may be worth clarifying it first of all to yourself. (Of course the resistor should have dissipated 0.2A*0.2A*1800 Ohm = 72 W power and would have got burnt in seconds if the 200mA current had flown indeed.)

Because I assume you correctly used the meter, it is the meter which fools the measured data, especially in current measurement mode as I assume, so be careful if you really wish to measure a pulsing DC current with it, even though in this case the full wave rectified output voltage also had positive half waves, not missing in every second half wave as at the input. You could clean up the output at the 4 diode DC output by using a puffer capacitor across its DC output and load the cap with the 1.8 kOhm.

Greetings,  Gyula

[dieter]

Hi Gyula
Thank you for your attention. The Voltage was measured without a load. There was not 225v * 0.26A. I was only testing if both increased, and they did. The efficiency of the setup in terms of loss during transformation is poor, for several reasons, but was enhanced to 209% by the Magnets. I already know that this may be caused by the increased flux density due to nonlinear permeability, or as you said, if I got you right, a DC Offset. Whatever it was, it increased the output. Whether or not this could increase the output of a system that already runs at high efficiency (and go beyond 100%) is an other question and actually not the goal of this project. It was a test to see how the repetive absence of a static B Field performs in a pulsed core. It certainly already proofs that the statement "static permanent magnets cannot increase output" is wrong.


The AFA is based on a much smarter thought, as you can see in the diagram. As a pulse in one coil does not only block the path of the pm, but redirects it trough the other coil, we need only 1 force unit for this pulse, but we get 3 force units in the induced: 2 from the pm and one from the pulse em, cause they all want to go to their south pole and they can flow in parallel. Well, it's in the diagram. Unmentioned yet is the back-emf of the collapsing field, that is adding to the amplitude negatively (adding to the -1 pulse), so we may end up with an alternating current like -4,3,-4,3,-4,3, with an input of 1.


What I call AFA is also named "magnetic frame" in some patents and appearently I am not the first one with this idea. I just want to build a working device.
The challenge lies within the efficient construction. High Permeabillity is needed.


Frequencies and Mark/Space/waveform variations may be tested, although it would be cool to build it with common materials.


Thanks for your suggestions. When it comes to volt and amp meters, I usually fail, so the goal is to close the loop. That would make any measurement unneccessary. Nonetheless, I may have to buy a new meter the sooner or later. Who said free energy is for free?   

[dieter]

The Secret Life of our Fridge Magnets

Magnet Flux Study, by D.Marfurt


(As I promised) I want to show you some experiments you can do in order to understand Magnetical Flux and what I call "pathseeking and non-pathseeking flux". With a basic understanding people usually don't know much about the magnetical behaviour(s) in the field of flux paths.

I refer to the illustration below.

1.)
Just an iron rod, one magnet attached ontop and a nail at the bottom. The nail is attracted. This is a typical scenario we know since we were kids.  If I lift up the bar in the air then the nail keeps on sticking, it does not fall down. The strong Neodymium magnet is searching for a way to a contrairy pole in every corner of the iron rod, keeps on searching even in the nail. The "B-Field" or magnetic strenght is distributed evenly accross the entire iron mass that is in contact (although, it is getting weaker with distance due to limited "magnetical conductivity" or permeability, depending on material properties, which is rather low in iron, but still 200 times higher than in air). So the magnets southpole is searching for a north pole, this could be its own north pole, or the one of an other magnet. This is "pathseeking magnetical flux".

2.)
Now, while I hold this in the air, I carefully stick a second magnet to the iron rod, with the north pole towards the iron. (make sure not to push the rod, use your fingertips to slowly get in contact). As soon as the 2nd magnet sticks there, it will build a flux path with the other magnet and the nail drops off, falls down to the ground. A tiny bit of remanent magnetism my reside in the
bottom end of the iron, but usually not enough to hold the nail. Note: interestingly the vanishing of remanent (remaining) magnetism sometimes takes som real time, like a second or so, you may experience such a delay in the nail drop off. So now these two magnets hav   e found a path to a contrairy pole of equal amplitude, which seems to make them happy in that they no longer search for a path in every corner of the iron. They use the shortest path to eachother and distribute only in path diameter if the iron's saturation forces them to do so. Because, the iron can "take" only a certain amount of field density. Therefor, a core can be too small but it hardly can be to big, unless saturation is what you need. So now the nail will not (or barely) stick anymore. Not only at the bottom of the rod, on the entire rod it will do so, except for the area very close to the magnet, because there we got oversaturated iron.
This is "non-pathseeking magnetical flux".

3.)
Now I take a third magnet and stick it to the rod, as seen in the picture. Now we got two south poles, in need of a north pole. The left side magnet now has to serve for two contrairy poles. It isn't absolutely clear how they share this pole, but obviously they are not satisfied, so they continue pathseeking. The entire rod is now attractive to the nail again. For the sake of simplicity this is separated in bottom and top region in the illustration. It may or may not be the case that the two south poles share the entire iron mass. Remarkable is the fact that any imbalance between a north and south pole field strenght can result in the transition from a non-pathseeking to a pathseeking stage.

4.)
I take a 4. magnet and carefully get closer, as seen in the picture. As soon as I am about half an inch away from the rod, the nail drops off. Again, we have built a path, or in this case obviously two paths. This balance between the four poles puts the scenario into a state of non-pathseeking again.

5.)
When I finally stick the 4. magnet to the iron rod, the bottom part becomes oversaturated and the nail will be attracted again. So it seems, due to the lower distance from magnet 3 to magnet 2 (compared to from 1 to 2), magnet 3 "consumes" some of magnet 2's south pole, even if it has a further north pole (4) at the bottom. Probably contradicting is the fact that the nail will not stick at the top part when attached to the eft side of the rod, but will actually stick when attached to the right side, which indicated that there is a full path between magnet 1+2 and the iron rod in this upper part is not saturated.

Please note, such materials like the ones used here have rather strong remanence features, so eg. the nail will remain magnetized slightly and in some configurations it will stick due to its own remanent magnetism. But you should see the diffrence between a barely sticking remanent nail and a rather strong effect of a pathseeking Neodymium magnet. In doubt, restart the experiment at (1.), this will reset things accordingly.


In Picture 2 I am using a thicker iron rod that gets less saturated. It works about the same but one interesting thing is, when there are obviously two parallel paths (as illustrated), unlike in the (5.) stage of the previous experiment, the nail will in fact  drop off.


So, the essence of this study may be: there may be two flux paths in the same monolithical part of a core and as long as there is no oversaturation, they may coexist without canceling eachother out. This could mean, in a typical core the fwd MMF flows on a inner path and the back MMF flows on a outer path. This could mean, in a Thane Heins Bitorroid system the two back MMFs may not compensate eachother, but flow side by side in the same outer core.

The Secret Life of our Fridge Magnets ... remains a Secret, at least partially.


Regards