2 thoughts on “Mono Amp: EF37a driver (part II)”

  1. nice work, man! the screen voltage can greatly affect the total swing before clipping and the harmonic spectrum… with pentodes the gain generally increases with screen voltage, but so does odd order harmonic distortion. every pentode has a handful of sweet spots for every operating point, because of this it can be worth your while to feel around a bit. i find a good start for bigger swings is a screen voltage around half the plate voltage. but for preamp use i will go a bit closer to the plate for more gain/gm. the current source loading will always give a triode like spectrum to begin with, but as soon as the diode line comes into play there will be blood. you really hang on to the mu follower circuit. i appreciate that. i never saw it as elegant… or thought it was better than an optimized buffer. but yours has really evolved! that little bit of compliance is everything! keep up the great work!

    1. Thanks JC, much appreciate your feedback and insight, you should be writing a book on this stuff!

      I figured out whilst adjusting the screen voltage looking at the FFT the behaviour on the odd (H3 primarily) distortion. There is a sweet point indeed and a trade-off when looking for gain. This is the point where I think higher-gm pentodes will perform better than the EF37a. However, I’m in love with this pre-WWII pentode and used it a lot before. In this case the EF37a is running at 235V to give enough headroom (when operated with +400V at the CCS) whilst the screen voltage was reduced down to 112V so exactly as you say. I tuned this by looking the FFT plot at 200Vpp swing.

      I need to experiment more, I guess as you say the topology is perfect for a preamp.

      Have a Merry Christmas!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.