Rarely is any topology fully optimized, since few designers consider noise reduction possible without the aid of heavy feedback or without brute force techniques, such as chokes or regulators. So how can we fairly judge a topology unless it is completely and competently  designed? Sadly, we can't.
     Do not forget that the techniques outlined here can be applied to a wholly solid-state amplifier with a few modifications. So is there anymore to be squeezed from this topic? I am sure that there is, but we will wait to see what the e-mail brings before returning to it.
     (I had hope to include an explanation on how to design a 4 watt OTL amplifier in this article, but it will have to wait until next time.)

                                  //JRB


Reference
    "Analyses of Drivers for Single-Ended Push-Pull Stage," Hiroshi Amemiya, IRE Transactions-Audio, Sept-Oct 1955

    The circuit above shows the enhancements made to the amplifier. As I now look at it, I see that the 5965 might be a substantially better choice for the long-tail phase splitter, as it would provide both more gain and a larger voltage swing. But then this remake was not meant to definitive (as I wanted to retain as much of the original design as possible).
    Which of the above modifications yields the most results? Balancing the drive signals comes first and is followed by the noise canceling techniques; and although it is last, using a different input tube is needed to run a higher idle current and would have to be implemented. 
 
Conclusions
     I remember seeing a circuit that purposely added hum and limited the bandwidth of solid-state amplifiers in order to reproduce the golden sound of tubes! That tube circuits often hum is as unfortunate as some great wines often smelling of sulfur dioxide, but only a fool imagines that sulfates make a grand cru.

< PREVIOUS

NEXT >