I was looking back over Naudin's Parametric Power Generator effort from '97, given a little bit better understanding of the problem. As I understand it now, his setup could have worked pretty well. In fact, at the frequency he is driving it, if it had actually worked it probably would have self destructed almost instantly.
I just see one glaring problem. If he is actually changing his inductance from 0.13H to 0.05H as is claimed, his resonant frequency should be changing from ~14kHz to ~22.5kHz from one part of the wave to the next, resulting in a very distorted output wave. His scopeshot appears to be a perfect sinusoid ...meaning... the parameter isn't really changing. Until 1/2dLi^2 exceeds 1/2Ri^2 T/2, you'll get some resonance, but it will not grow to the levels needed to produce useful power.
I suspect that some of the highly efficient motor/transformer designs out there are dancing on the edge of saturation, and in doing so, are flirting with parameter variation power. Done wrong, it would be hit and miss...but mostly miss. In the cases when folks stumbled on the correct answer, the designs probably blew up within seconds, or faster. That is what this predicts. It is necessary to have a closed loop feedback control to keep it operating in a useful range, yet I'm sure that most are not thinking in those terms.
I just see one glaring problem. If he is actually changing his inductance from 0.13H to 0.05H as is claimed, his resonant frequency should be changing from ~14kHz to ~22.5kHz from one part of the wave to the next, resulting in a very distorted output wave. His scopeshot appears to be a perfect sinusoid ...meaning... the parameter isn't really changing. Until 1/2dLi^2 exceeds 1/2Ri^2 T/2, you'll get some resonance, but it will not grow to the levels needed to produce useful power.
I suspect that some of the highly efficient motor/transformer designs out there are dancing on the edge of saturation, and in doing so, are flirting with parameter variation power. Done wrong, it would be hit and miss...but mostly miss. In the cases when folks stumbled on the correct answer, the designs probably blew up within seconds, or faster. That is what this predicts. It is necessary to have a closed loop feedback control to keep it operating in a useful range, yet I'm sure that most are not thinking in those terms.
Comment