Replication of Mini Radiant Exciter circuit of Nelson Rocha

[Zephir]

So you completely missed the large wire-wound resistor.. Certainly you are aware these type of resistors are wound in a bucking coil configuration.

You know, you can be ingenious as you want - but there will be always some 4-years old Asian kid at YouTube, which would handle it better.. :-) And these really smart kids have their overunity device in the kitchen already. So that thank you for your remark: I'm fond of bucking coils, but I would need to get familiar with this type of resistors more. I already dismantled/broke few of them, but they always contained just a single coil. Also, these resistors usually have high resistance and energy loses, as they weren't designed to serve as a component of resonance circuits: their resistance would probably beat the effects of their inductance. Could you link the NelsonRocha's resistor from some catalogue? It should have at least three legs, if your theory is correct. Also this resistor should be present in another NelsonRocha's circuits.

[Vortex1]

The circuit as it stands is a basic blocking oscillator (aka JT) with a high gain Darlington in place of the standard single transistor used in such circuits. Due to the high gain of the TIP122, the circuit may be very prone to several parasitic oscillation points depending on the bias and other factors.

The use of the 4700 pF capacitor on the secondary of the saturating transformer T2 partially rings the secondary making it somewhat resonant.

I have redrawn the original N Rocha circuit that includes the components inside the TIP122 and the possible saturable biasing core of Zephir's hypothesis.

Here is the first draft, drawn with conventional input on left, output on right, rails at top and bottom for ease of reading.

Since the bridge rectifiers were not identified, I left them with "?"

Any errors spotted I will correct in the next release. ?

From the waveforms I've seen thus far in others replications, the circuit can be unstable (depending on bias setting) and will tend to bounce between several operating modes at the slightest external interference or synchronization with ambient externally generated frequencies or static magnetic fields large enough to couple to the core.

Regards

Edit: drawing revised to NR ver 1.01

[Zephir]

Hi Vortex1 - and thank You very much for your circuit and detailed description of it.
Now I think we have state-of art formulation of the problem and we can start with its analysis.

From your scheme follows clearly, that the primary of T1 transformer is directly connected to +Source of DC circuit.
Which is strange, because such a transformer would suffer with high DC bias current at the case of failure of oscillations (it's primary of Tesla coil with no reactance for DC current). What's worse, it can work only when the power source will get shorted with Q1 Darlington transistor - with consequences for all components in the circuit.

It could explain why @Dog-One's replica stopped work so soon and maybe it could even explain the purpose of 20 W resistor, which @Dog-One pointed above and which isn't actually labeled in yours/Mr. Rocha's diagrams. Constant voltage source cannot be shorted with transistor without some protection or punishment. Anyway, this would be very unusual and ugly design.

[Dog-One]

From your scheme follows clearly, that the primary of T1 transformer is directly connected to +Source of DC circuit.

Maybe take Vortex1's image, draw a red path that shows this short.  Or not, because there is no short.
This circuit only draws a few milliamps of power.  How would I know that?  I built it and tested it.
Certainly not Vortex1 and I's first rodeo.  This is why I stressed in Grumage's thread that everyone
should build this very simple circuit--it has basic electronics as well as some more advanced concepts.
I guess we need to twist your arm Zephir.  You can't run until you at least learn to walk.

[Vortex1]

I would like to add that T1 is being charged with current just like T2 when Q1 turns on.

In a normal blocking oscillator the current through T2 would reach a point where the core would saturate, therefore base drive of Q1 would go to zero, and  Q1 would turn off, then the cycle would repeat.

Because Q1 draws so much current as it is driving essentially two transformers in tandem, it was necessary to use a Darlington TIP122 which has very high current gain. In this way the blocking oscillator could run normally and not run out of base drive current.

Regarding the 15 Ohm 20 Watt resistor. If one were to use such a resistor near it's rating one would have to anticipate a DC or RMS current of 1.155 Amps running through it and  expect a Voltage across it of 17.32V. This seems ridiculous to waste so much power to light a tiny LED.

Also regarding Dog-One's comment about bucking windings, all of the ones I have taken apart are straight wound types. I admit there may be special non-inductive types that use a reverse winding, but this resistor does not look like that type.

Usually these cement cased resistors have cement fill on one side (the side facing the PC board). It is easy to hollow out the cement filled portion and there would be ample room to put a long battery or a stack of button cells. I'm not saying this was done however for future reference for anyone dishonest who wants to make youtube money it would be a good hiding place, since there is no apparent engineering purpose for such a resistor in this circuit except by a wild stretch.

http://mhchip.en.alibaba.com/product/1569337369-801402376/20W_15R_5_Ceramic_cement_power_resistor_20W15RJ.html#!

Regards