Rerurn to Romy the Cat's Site

drdna wrote:

What specific objectives did the air caps address in the circuit? You refer to the tempted virtue sound but I am unsure as to the goals that you were trying to achieve with this circuit alteration?

 I what wrote about the air caps in the thread: “Where are our good phonostages?”

The genuineness of this Paravicini circuit is that he use feedback to plate – or to vey high impedance – and in this situation it revilers very small values of capacitance to write the necessary RIAA curve. In fact the curve is written not by the feedback but by the Miller capacitance of the tube. The 12AX7 was designer as audio tube, it is a radio-frequency rube  and therefore is has very high Miller capacitance, the capacitance the otherwise conceded as parasitic but the smart Paravicini put it actually in the beneficial use. So we have a circuit with “hidden RIAA”.

Now Paravicini adds the very light touch of capacitance into feedback to make the Miller capacitance to work what it shell work. The values are very small s the capacitance in feedback sort of multiply the tube’s own Miller capacitance. The series capacitors in feedback, regardless of value and types are very bad. In fact a feedback cap is the more auditable them loading, coupling, cathode or any other cap. The coupling cap for instance is charged with DC on both sides and the AC. What cap is? It is the sequence of conductive and non- conductive surfaces where non- conductive dielectric acts as an abrupt termination of AC flow. However, the dielectric is not a perfect dielectric and has some free electrons in itself that can care the signal.  This abrupt termination induces a microscopic force in the dielectric that make the dielectric to move, and by virtue of it creating variable reflections of transmission termination. So, that coupling cap has already charged or pollarded with DC on both sides and the cap’s conductors have already biased or physically oppressed the cap’s dielectrics. In case of cap sitting in feedback there is no DC (in second cap) and the dielectrics are not damped. Furthermore the re-polarization of dielectrics does not happen at non-auditable frequency but at very much auditable frequency – the frequency of the correction. I do not even mention that dielectrics experience different behavior of  electro- mechanic induction at different frequencies…

 So, how to create a better condition on this situation? The concussion is to get rid of dielectric all together. The non-dielectric caps: air or vacuum are very large for a given value but in this Paravicini case the values of the caps are so small (100pf and 300pf) that the air caps could be called for the duty.

I do not say the in the “End of Life Phonostage” the air caps is something that made all sound. The ET SU-2 transformer I think is the very heart of this Phonostage and 80% response for the corrector’s sound. Still, I think the air caps were very contributive as well. I did not listen the “End of Life Phonostage” with mica caps but I did listen and did “evaluate” the vintage mica, the polystyrene and air caps in context of my 7788-7721 phonostage.  (It uses the absolutely the same Paravicini’s ides of feedback only uses7788 pentode in input and has just two stages). The result was very superbly convincing and perfectly proved the theory about the fundamental advantages of air caps.

Rerurn to Romy the Cat's Site