DCAudioDIY.com

I remember a discussion quite a while ago about the cathode cap being in the signal path and that it makes a big difference the sound quality. Some think the signal path is from grid to plate, but the cathode supplies the path for that signal. In my very limited experience, I see most cathode caps are around 470uf and are electrolytics, probably to save space and money.
Is the high ESR of electrolytics the problem. I see that the ESR of cheap electrolytics can be quite high. The amp I am working on uses a KENET 1000uf with a ESR from .075 ohms to 0.1ohms depending on frequency. This seems to be a very high capacitance and low ESR for this position. I kind of remember someone saying one can get away with a capacitance as low as 50uf poly as a cathode cap.
Is this KENET worth replacing with a lower capacity poly cap?

BTW you can flame all you want; I expect that those with far more knowledge and egos to do so.

A triode has two inputs (three if you count the heater): the grid and the cathode. The gain from the cathode is slightly higher than that from the grid. A signal (or distortion of it) present at the cathode will definitely have a sonic impact. How much? Who knows? The newer low ESR Os-Con, etc. types have very low ESR and are a good choice for this application (I think).

Will a parallel film cap at this position improve things? Who knows? It is commonly done. I'd suggest that a test article (say, a simple line stage) would be a good tool to try and hear differences (if any).
Of course, a high resolution system in a good room are essential, lest any minute differences be muddled over.

Stuart

I have always recommended using a very large cap for this application because you want the lowest impedance possible. It is true that as the capacitor size increases, its resonance frequency decreases, and this is the point where the cap becomes inductive rather than capacitive. For audio frequencies you are not likely to have this occur, but it can be a rational for bypassing. If bypassing with a smaller cap consider one that is 1/10 to 1/100 the value of the main cap. I would use at least 1000 uF for the main cap, and I would be happier with 10000 uF.

Hi Paul,
Is this new project a power amp or preamp?
A lot of the "classic" circuits don't have super large caps in this position, because bass performance was severely lacking in both the speakers of the era and/or the output transformers of the power amp output stage. So, a 200uF cap seemed fine at the time.

Whenever I think about paralleling different caps (bypassing....) I rewatch this video https://youtu.be/TpXvac1Y3h0?si=cl5CsJKrkkoQDPOA and then promise myself that i WILL look for all the resonances.
I love this topic BTW !!!

Bruce.

Thanks for the gentle responses and I will consider all suggestions.
I learned some things and did not realize that larger the capacitance the better up to a point.
A friend of mine who designs and builds 300B SET's does not like electrolytics in this position.
I do understand that one can lose bass response if this cap is too low.
I just wanted to confirm that the lowest ESR is beneficial.
I am working on a Japanese 6550 SET kit that can handle up to KT170's. I am curious about the debate that the KT170 sounds better than the KT88. This amp has switchable bias and plate voltages

_________________
Walt

A rather compact high value PP film capacitor choice for power tube cathode bypass are modern DC Link capacitors. The values of 400VDC rated capacitors have values extending well above 100uF. I tend to use these in my power supplies, but people have used them in speaker crossovers and cathode bypass applications. Bypass with a smaller film like you would do with electrolytics.

BTW, agree with using fixed bias where possible over cathode bias, it avoids the need for cathode bypass caps for a low impedance AC path to ground since the cathode is at ground. Of course, that is not always practical and needs a reliable and low noise means of applying the negative bias voltage.

David

David and Charlie turned me on to DC Link caps in power supply use. The improvement over electrolytics there is not insignificant!

I had thought the same about using them (bypassed) in large value speaker crossover applications. But there, as well as in the cathode circuit makes REALLY good sense.

I'll also agree about most common applications where grounding the cathode means that a negative grid supply is needed. The complexity of that, combined with the caps needed there to ensure quiet operation, are probably worse sonically than a really good cathode bypass.

Stuart

The simplest solution is to go with solid state in the 21st century -

https://www.stereophile.com/content/buc ... -amplifier