|
Back on topic in a serious vein now. Here is a simple model of what I take is (or has been at last report) the Ainslie circuit, with the basic waveforms that would be expected under "classic" analysis.
I'm showing one single pulse here, at around 10% duty cycle on-time for the MOSFET gate. I used a model of the IRF740 Hexfet. The values are not particularly important. Over a monte carlo range of various values for the gate resistor and load inductance, taken in small increments, no huge deviations from this basic waveform were seen.
This proves nothing but will serve to gain a basic understanding of how the circuit would normally be expected to work (without zipons or radiant energy or a circuit layout containing large parasitic L's and C's or feedback into the driving pulse generator or incorrect probe/ground placement that could cause actual or perceived continuous oscillation).
Of significant note is the relationship between the voltage and current waveforms and the fact that the current waveform does indeed go negative for a brief time as the first tall voltage pulse falls back toward zero. The simulation clearly shows two salient facts:
1. The body diode of the MOSFET does indeed turn on. This is the period shown on the voltage trace as -1V and, during this time, the inductor energy is indeed charging the battery.
2. The mechanism that drives the body diode to turn on is quite simple and easily demonstrated. As the inductor voltage collapses after the first tall spike, the output (drain-source) capacitance of the MOSFET (Coss) is in series with the inductor and the MOSFET channel is off. As the voltage at the drain rapidly drops back toward zero, the charged Coss forces the body diode to turn on and the drain voltage goes negative limited by the forward voltage drop of the body diode.
I verified this theory by adding an extra discreet capacitor of 1nF from drain to source of the MOSFET and observed a corresponding increase in the negative-driven current spike. (not shown)
Once some of the Load Inductance and the Coss charge energy is delivered to the battery, the drain voltage falls to zero and the Coss is discharged but there is still energy left in the inductor and the drain voltage rises again, its amplitude limited now by the energy that has been taken out of the series LC resonant circuit and delived back to the battery through the body diode.
This process repeats as the series rresonant LC tank rings down. After the first couple of cycles, no further energy is returned to the battery as it can be observed that the current stops going below zero despite continued ringing.
Note that the area within the negative-going current trace is only a very small portion of the forward energy (above zero) trace area. This is because only a small portion of that energy coming from the battery is actually stored in the inductor and the Coss as compared to the energy being dissipated in the load resistance.
|
Author
Topic: The Rosemary Ainslie Circuit (Read 459809 times)
