Great theory...

[dieter]

I want to show you an idea that works great in theory, but obviously not in real life..,


Imagine a EI core, one coil on each of the 3 legs. The middle coil be the primary. The left coil has a LED as a load, the right coil has no load at all. There is 12V 59Hz in the primary. I get eg. 2 volts and 7 mA trough the LED. Now I short circuit the right coil. Suddently there are 54 mA and 2.6V flowing trough the LED.


This means, when the right coil was unloaded, 80% of the flux went trough it because the prim. has seen the easier path, because on the left side the LED load reduced the permeability. After short circuiting the right coil, suddently the left side became the side of the easier path, resulting in most of the flux flowing trough the left coil, causing induction and increasing the current for the LED like 7x.


Based on this observation I concluded that, I may replace the primary by a Permanent magnet and add an alternating switch to the secondaries: this switch will connect a load repeatedly and alternating to the right coil and to the left coil, always leaving the unconnected coil unloaded.


This alternating state of (passive) load connection should now be capable of forcing the flux path of the magnet to the currently unloaded coil and hence cause an ongoing flux path switching. The induced current may be high, depending on magnet, core, coil etc., but has absolutely no relation whatsoever to the connection switcher.


That's about it, my nice idea, too bad it just didn't work. Can somebody explain me why???.


Thanks

[PiCéd]

Permanant magnet is a force, if you want energy you must move this force.

[dieter]

You may have not understood what I wrote, nor magnetic paths in general, no offence.


Regards

[forest]

core saturation ?

[dieter]

Thank you forest,


As indicated, I already made a quick test, As I had no core that would work with two coils and a PM, i decided to use the 3 coil transformer and put a DC current on the middle coil, simulating a permanent magnet PM. I also added a 4700 uF cap to the supply to smooth any remaining ripples of the already smoothed , rectified dc.


I used a commutator kind of thing to switch the connections, using brushes and a modified computer fan. This fan had its own wall supply.


I used several resistors to test various saturations caused by the "permanent magnet", output was not affected, a half a volt at less than 1 microamp. I was getting the same without any current in the middle coil ...  , although only when the commutator was in action.


But I was thinking a little more about it. I yet still see no reason why it cannot work, but there may be a reason why it didn't start:


The permeability of a core side will be reduced only when a current flows in that side's coil. But such a current will flow only when there is a change in the magnetic field. Which is not happening because , as already stated, there is no current flow...


Long story short: if this works, it must be kickstarted somehow.


Regards