@gyulasun,

45kHz is a typo. The maximum OU is observed at 450kHz -- recall the graph in http://www.overunity.com/index.php?topic=8411.msg242764#msg242764. At 45kHz we have underunity. By the way, thanks for alerting me about the redundancy measurement of the voltage.

@gyulasun,

Checked it with 0.001uF, 0.0047uF and 0.01uF. This is all I had at hand. No effect. 45kHz seems to be the cutoff frequency for the switch in phase shift.

@Omnibus,

Can the attached audio transformer be used?

Also, can you clarify in your drawing on how you connected your transformer?
It will help a lot if you draw some dots in the transformer and label primary and
secondary. Did you put the function generator input to primary or to secondary?
What was your input voltage and duty cycle at 450KHz where the COP was
largest? What was the Henry value of your extra coils?

Groundloop.

@Groundloop,

My studies show that it is quite likely that every transformer should exhibit OU at certain characteristic range of frequencies. Would be interesting to see if you would observe OU with your transformer. The extra coils are of several nH calculated inductance and it was interesting that adding more or having only one or none affected adversely the effect. The effect without the coils or with more extra coils is still there but is less. The signal from the function generator is applied to the primary coil of the transformer. As you can see from the graph shown here: http://www.overunity.com/index.php?topic=8411.msg242764#msg242764 , for the transformer studied the OU effect appears in the range of frequencies between about 200-700kHz, reaching its maximum values at about 400-450kHz.

@Omnibus,

Thank you for taking time to answer my questions.

I plan to build something like in the attached drawing.

Groundloop.