user TURBO?s replication of Steven Mark?s TPU ?

[Vortex1]

This is a good idea to try. I would add some 10 volt zeners from gate to source to protect the fet's and some resistance in series with the capacitors to limit loading effects. The resistance, combined with the gate capacitance will allow frequency tuning.

The capacitors are not needed if the resistance is added. Then also some biasing pots to get the Fet's in range.

This circuit will determine if a magnetic field is massless or has some mass and thus inertia. Bibhas De has some things to say about this.

Be wary of heating effects in the Fet's which will change the operating point and cause a false speedup effect. You may need to insure a constant temp on the fets to separate cause and effect.

To allow disconnection of the internal FET diode, a separate diode can be added pointing to the drain (coil-diode-drain) This will unclamp the flyback emf on negative swings.  The dotted diode is called the "parasitic" diode. It is inherent to the fabrication process of FET's wafers. It will clamp the free oscillation of your coil unless the disconnect diode is provided (coil-diode-FET).

I have often used this circuit in coil testing with feedback. It is safe for your FET's

Sorry for the hand drawing, I wanted to get this up quickly before Fet's were needlessly destroyed.

If you want to go like hell, reduce or eliminate the frequency resistors. I like to start slowly at first.

good Luck Tomas

[pese]

From that , you can create an nice FET-Power Inverter.

Need only one coil for output ,
or even only diode plus condensor.

Nothing to create FE.


Oszillator that are running round
(over 3 transitors ) you find at NAUDIN
(coreless , but even with Coils it will work.
But , it produce only HEAT . ( = LOSSES)

[wattsup]

@all

I am in the process of making my second TPU, but I have been looking extremely carefully at all the videos, literature, etc.

My main problem is the with collectors as prescribed by some of you guys. It seems that simply bunching horizontal collector coils together in the center, then wrapping it with the control coil is not to efficient. Anything under the outmost collector wires are to far away from the control coil transfer intersections, since there is no center iron core for induction to go through the complete collector bundle before reaching the iron core. It would seem to me that all of the collector wires must be in contact or intersecting with the control coils, at all times, for there to be maximum effect. The following is taken from the SM pdf and is part of a report done by Dr.ROLAND SCHINZINGER who was present at one of SM'S video demos.

"After the test the inventor cut the toroidally shaped device into segments (though not the controller box located at the center of the device). These samples consisted of an array of circumferentially arranged coils and wires grouped around a core made of a cork like substance."

This seems logical in order for the collectors to be on the outside of the cork type center, allowing for greater exposure of the collector to the control fields. Sm says what happens when a magnetic field travels over A wire, not a bunch of wires.

Also, the coil levels were more rectangular then circular, keeping with the idea that this maximizes more sidewall motion and less north/south poles, which favor the creation of the projected magnetic field toward the TPU center.

I have other reservations but these are the most pressing at the moment. If anyone has any concrete evidence that the current 3 level TPU form as worked on by most of you is the right way to go, and not just by getting kicks (I got kicks also with just simple 5 VDC pulsing), please advise.

[giantkiller]

Here's your magic bullet...
Make a TPU you can tune and forget all this diatribal bs.

--giantkiller.

[otto]

Hello all,

this weekend I bought a lamp wire for my collector coils. I wound 21 turns. That are 7 turns in 3 layers. Over this collector coil I wound my control coils, again in 4 segments. Each segment has 85 turns, wire diameter is 0,65mm. My collector coil is 7,70m long and each segment of the control coil is the same 7,70m long. That means the ratio is: collector coil:control coils = 1:4.
Then I was thinking about the TPU: we have 3 collector coils + 3 control coils in segments + a coil wound all over the others, the feedback coil.
I wanted a really big signal so I wound over the segments an additional coil. This additional coil has 85 turns. I feeded the signal back. Now I have with only 1 collector coil + 1 control coil, in segments + 1 feedback coil = over 400 V kicks.

It will really be interesting to see when such two coils adds their signals. Not to mention what will happen when there are 3 of such coils!!

What I know for sure:

1. Number of turns of the collector coil is NOT important!!
2.Wire diameter, number of turns of the control coil is NOT important!!

THE IMPORTANT IN THE TPU IS THE FEEDBACK

Otto