Shorting coil gives back more power

[i_ron]

Konehead,

>>Also this summer Gyula gave me excellent bidirectional mosfet circuit, where mosfets >>connect at the source and gate so they switch AC this works very great.

Can you post this circuit here?

GL.

GL, it was posted on page 11, #150

Ron

[i_ron]

Hi Doug, and All,

The simplest "peak finder" that I could find was this one...

http://www.8051projects.info/blogs.asp?view=plink&id=198

Just use an extra coil as you were doing and with one transistor... voila... zero crossing. With the coil adjustable one just moves the coil until it sits squarely on top of the sine peak.

Here is what the circuit looks like, the diode D2 separates the voltage regulator from the transistor sense part of the circuit...

Q1 puts out a positive logic signal so a 555 won't work, I used a CD4047 for the one shot. The one shot then triggers the CD4093. R5 is a pot to set the pulse width.

But minimal parts!

Second pic is my RV setup

Third pic is what the signal looks like, but not shorting the coil in this shot

Ron

[guruji]

Magnacoaster used a pulsed coil of not that many turns about 100 is enough as seen in his early photos. He used a laminate steel core like transformer strips then applied 5 powerful neos one end and 1 the other. This magnet layout is VITAL.

It creates a Bloch wall which sits about 2/3 across for the length of the core. The edge of the coil sits right over the bloch wall. So you need it to slide along so you can find the sweet spot.  When the coil is pulsed the magnetic field is already on the Bloch wall and acts as a pivot point. You can not do this any other way when using such powerful neos to fight against this magnetic field would need like 400 amp pulses so you hit it right on the Bloch wall. This is called  asymmetric magnetic modulation. When the field collapses the neos flux over shoots and generates more power out of the coil than what you put into it. Now coil shorting will help here a lot. I dont know what he is doing now but if you hit the top of the rebounding neo spike with  short you can almost certainly increase the voltage perhaps 5 times more then syphon off ONLY the HF part of this signal will prevent loading to the driver. In practice you would put blocking diodes going into the coil to prevent BEMF reaching the driver then use about 0.01uF in series each side of the coil to syphon off only the HF and use as high pass filter. Then add a HF balum traffo and impedance match down before going to FWBR then into a dump cap.

Having not built it that is all i can tell you but if i was to build any of this stuff i would do this rather then piss about with bedini wheels and spinning magnets.

I also know he was using high side fet triggers because when they blew out short circuit it almost always melted his coils:)

Hi Bolt can you please post a diagram of this?
Thanks

[popolibero]

Guys, I don't know if you think it's off topic but I have a few magnacoaster pics to have a visual of the setup both mechanical and solid state, can't see the magnetic polarity of course, but still...

Let me know if you want me to post them here.

Mario

[Feynman]

It creates a Bloch wall which sits about 2/3 across for the length of the core. The edge of the coil sits right over the bloch wall. So you need it to slide along so you can find the sweet spot.  When the coil is pulsed the magnetic field is already on the Bloch wall and acts as a pivot point. You can not do this any other way when using such powerful neos to fight against this magnetic field would need like 400 amp pulses so you hit it right on the Bloch wall. This is called  asymmetric magnetic modulation. When the field collapses the neos flux over shoots and generates more power out of the coil than what you put into it. Now coil shorting will help here a lot. I dont know what he is doing now but if you hit the top of the rebounding neo spike with  short you can almost certainly increase the voltage perhaps 5 times more then syphon off ONLY the HF part of this signal will prevent loading to the driver. In practice you would put blocking diodes going into the coil to prevent BEMF reaching the driver then use about 0.01uF in series each side of the coil to syphon off only the HF and use as high pass filter. Then add a HF balum traffo and impedance match down before going to FWBR then into a dump cap.

Wow bolt ,great description.  Does this have anything in common with Groundloop's solid state orbo or Naudin's 2sGen?    It sounds like modulating the hysteresis curve using neodymium magnets.

Naudin 2SGen
http://jnaudin.free.fr/2SGen/indexen.htm

Also lol @ 'piss about with Bedini motors' hahaha