The Black Art Of Plate Supply Regulation Explained includes resonant choke input

[W2VW]

In this world there are several Black arts. Rebuilding automatic transmissions, figuring out women and building high Voltage power supplies that don't suck.

Many thanks to W2NNNNNBC for the link:

http://www.dealamerica.com/RCPS/Resonant.htm

[N9NEO]

Interesting work Dave. 

I would expect the input choke type power supply could get into trouble if the input choke and the filter capacitor were resonant at 120Hz (A 120Hz resonant frequency wouldn't be too difficult to land upon accidently by just helter skelter grabbing parts out of a junk box.) With a light load the series resonant circuit formed would get you some big 120Hz voltage on the output capacitor.  Enough to fail it.

The critical value of inductance that the author mentions is the minimum value of inductance required to keep the current in the inductor continuous.

Switchercad would be a good program to model a HV power supply.

73
Bob

[W2VW]

Yes, the cap and choke will see some mighty circulating Voltage. I had heard the same stuff about avoiding resonance but not much else useful until now. I only know of 1 A/B so far which was a success after cooking one 6 kV cap. More to follow.

[k4kyv]

I think the Collins KW-1 uses the resonant choke approach.  When building your own, the problem is finding the proper value resonating capacitor with high enough voltage and current ratings.

One factor left out in the otherwise excellent treatise on power supply filters is dynamic regulation.  While the static regulation of the choke input filter may appear excellent when an analogue voltmeter is placed across the output of the supply, if you monitor the output voltage with an oscilloscope you will notice wide swings in instantaneous voltage when the power supply is subjected to a varying load such as a class-C cw amplifier, or a class-B modulator or SSB linear.  These voltage swings can be reduced by increasing the capacitance of the filter capacitor immediately following the input choke.  However, there is a limit to the amount of capacitance you can use, while maintaining a reasonable inrush current to the capacitor.  Too much capacitance may blow fuses whenever the power supply is turned on unless a step-start circuit is used.  I have found that somewhere between 25 and 50 mfd is the best compromise between dynamic regulation and inrush current, and this depends on the ratio of current drawn by the load and output voltage, or in other words, the effective load resistance..

I do not know if the resonant choke filter offers any advantage over the conventional choke input circuit, in terms of dynamic regulation.