Rescued from a friends cellar, and making sweet music again!

I was fortunate to get a job in a company which built valve amplifiers. Fuelled by my curiosity about the "valve sound" I experimented with some single ended and push pull valve circuits.

This amplifier was built out of old bits and was a test bed for various ideas. It ended up using 4 EL84s in parallel push pull mode, a (very) long tail pair phase splitter and a single triode first stage.

One of the little luxuries of designing stuff from scratch is you can try anything. This amplifier has hybrid fixed and self bias for ease of use, and to get away from using cathode resistor bypass capacitors with out sacrificing to much gain.

Whilst acknowledging the advantages of the push pull output configuration, it was whilst working on this amp I began to become convinced of some apparent fundamental limitations of PP. I like to run amps open loop when possible before adding feedback, so I can hear what is going on inside. What I could hear apart from distortion and high output impedance, was a lack of fine detail, and a slight grungyness. I had noticed this on other (pp) amplifiers as well. Adding (secondary) feedback around this muddle tends to mask it with that "nice" sweet sound, so typical of valve amplifiers. The problem is, once something has gone, you can't get it back. We don't like that.

The low output power is a result of the lower than optimal HT voltage, not the best match of output transformers, and class A operation. If I could find enough EL86s it should really sing....