Claus Byrith 4-30 Questions

[Mike L.]

Hello everyone,

I've searched the forums for an answer to my questions. There has been a lot of discussion about the Claus Byrith 4-30 amp design found here http://www.lundahl.se/claus_b.html but no one seems to have addressed my specific concern.

I'm currently laying out my chassis. I've decided to hand wire rather than make the circuit boards. I was first using the "final" schematic presented on page 19 in the first section of the paper. At a later time, I was revisiting possibly making the circuit boards as presented in Appendix IV. When examining the schematic contained in Appendix IV that is used in the circuit boards I found that several new components have been included. Specifically capacitors C3 (220uF) and C4 (220nF) have been added to the feedback loop, and capacitor C2 (47pF) and resistor R4 (4K7) are bypassing the plate load resistor on the EF86. I can find no reference to these additional components in the text of the paper. Could some kind and knowledgeable person explain why these items might have been added?

Thank you.
Mike L.

[ray_moth]

C3 and C4 have not really been added to the feedback loop, they are paralleled to serve as a bypass cap for the cathode bias resistor - something he said he wouldn't do! Evidently he decided that he wanted cathode bypass after all, but forgot to correct the earlier disign details. Some people like to parallel a large cap (220uF) with a smaller cap (220nF) to make a better composite cap, although usually the smaller cap is 1% of the big cap, which would make it 2.2uF not 220nF.

C2 and R4 shouldn't be there, according to the text . Claus reckons he didn't need them once he had converted the EF86 to a triode and dropped its plate load from 100k to 47k (R3). That seems unlikely to me, though, because there's almost certainly going to be a peak in the high frequency gain with the global NFB loop, and you need that little Zobel network to correct it. Maybe Claus discovered this and changed his mind later, but forgot to correct the earlier disign details.

[Mike L.]

Thanks very much for the explaination. I guess I'll go the hand wire route. That way it will be easier to add/remove these questionable components and listen for sonic differences.