Caps in poweramp stage

[Bernardduur]

Hey all!

I am quite new to the amp building process but build already quite a lot. Now I try to understand some stuff........

I build a poweramp stage with 2 EL34's, 220k's for the bias and a 82k / 100k before that. In that stage there are 2 caps. What is the role of these caps and what is the effect of the size of em on the final sound?

Next, on some schematics I notice a small cap (47 pF) over the 82k / 100k section. What is the role of that cap?

Thanks!

[jrfrond]

The 47pF cap across the phase-splitter plates are there "just-in-case" there might be an oscillation or RF interference problem in that stage. It is probably NOT needed, but since it really doesn't affect the audio spectrum, I'd just put one in and forget it. Make sure your cap is rated 500V or better.

The other two caps couple the phase-splitter to the grids of the power tubes, and are called (duh!) coupling caps. Depending on the amp and how it is used, you can do some final low-frequency shaping with these caps. Typically, as in BF Fender amps, the value is .1uF, but for some designs where the power amp will be driven hard, you might want to drop that value. Many Marshall amps dropped the value to .022uF because of this. The lowered value can decrease "rolling" bass frequencies (intermodulation distortion) during power amp overdrive, and also prevent a condition known as "blocking distortion", whereas the coupling capacitors are fed signal faster than they can discharge it, and it drives the power tube grids more positive into a condition called "cutoff", a REAL nasty-sounding problem. The chosen value of these caps is a balance between optimum low-frequency response and absence of blocking and intermodulation distortion. If the amp is to be played clean, or is a bass amp, these factors typically are not issues.

[Bernardduur]

Thanks!

That was exactly what I needed to know............

[pdf64]

Hi John, for the sake of clarity, did you mean to say that blocking distortion was due to the power tube grids being driven more negative (rather than positive)? Thanks - Peter.

[jrfrond]

No, I specifically stated "positive" in my post.