Hi Chief,
This appears to be just the Russian name for a variable capacitor/ varactor/ varicap.
Here's a Russian patent where the term is used, and refers to a variety of different varactors as varicondas...
https://patents.google.com/patent/RU2550090C2/enI spent a good amount of time reviewing many parametric amplifier patents over the last week, and couldn't eliminate the possibility of excess energy operation.
It's certain that the reverse bias on the varactor does charge it as a capacitor by displacement current, and this charge is mixed with the signal at the initial stage.
And it's also certain that this charge is then carefully isolated from the signal, usually by dissipation in a tank circuit at the pump frequency. The charge used to change the C is not added to the signal but is wasted. Designers were very careful not to let this the pump energy enter the signal path, since their focus was on lowest noise and distortion possible.
You can see an example of this in the attached patent, which is simpler than most designs. In Fig. 1, the pump energy is sent into the varactor circuit by winding 18 and circulates in loop 1 containing the varactors, not passing beyond node B into the signal tank 10. The pump energy is dissipated in loop 1.
In theory it would be possible to reuse some of the pump energy by reducing extraneous resistance in loop 1 and the pump circuit (for instance resistor 16), paralleling varactors, or more radically, somehow making sure the varactors never see a forward bias, which is the main loss in loop 1. Then possibly putting a useful load in this loop.
If the pump signal is at twice the frequency of the signal tank 10, then the amplifier becomes a 'degenerate' oscillator, and the signal source can be eliminated and inductors 6a and 6b simplified. Spontaneous oscillations will then form in 10 if the load at node 14 is correct.
Then the question becomes, how much power can the varactors provide at node 14 ? This has a lot to do with the C ratio of the varactors. All previous experiments have been done with varactors with C ratio of .5- 2.5, the range usually used by the parametric amplifier designers-- but the hyperabrupt tuning varactors can have C ratios of as much as 22, over a bias V of for instance 15-30 V. These type are never used in amplifiers because the signal will of course be highly distorted, but this doesn't matter if we're just trying to get as much power as possible.
The experiments that Roberto Notte and I did using another patent (from Gunn) did appear to generate excess energy, although it didn't get self running due to some impedance issues. That circuit used varactors with a C ratio of 2.5.
Fred
feature is available which provides a place all members can chat, either publicly or privately.
Author
Topic: Parametrics, Noise coherence, and Switching (Read 29109 times)
