Best Way to Reduce Hum?

[FourT6and2]

The amp in question uses partially bypassed preamp cathodes for voicing. This is a good case for DC heaters. Even with higher gain designs I've done well with typical AC filament wiring by fully bypassing the cathode of first gain stage. The trend of using partially bypassed cathodes for guitar amp voicing has been taken out of context IMO. I do design with partially bypassed preamp cathodes, but I've gone back to some older designs and revoiced them with a fully bypassed first gain stage and a smaller coupling cap. Or, when the coupling cap and bypass cap create a shelving effect I've had to get a little more creative. But IMHO you can reduce the potential for filament hum (all other things being ideal) by just fully bypassing the first gain stage and omit the need for DC filaments. YMMV.

I'd love if you went into ore detail with this, please. I'm still learning and I'm not really an electrical engineer or designer or anything. Just a DIY guy who is teaching himself as he goes. Fully bypassed cathode vs partially bypassed on the first gain stage? In looking at V1a/b, they both have 0.68uF/820R cathode bypass arrangements (so does V2a). V1a has a 470R plate resistor and V1b is 100k. This arrangement, I'd think, is part of what makes this amp sound the way it sounds. But how does it dictate whether or not to use DC Heaters?

What I mean is, the partially bypassed cathode arrangement for each stage is part of what makes the amp sound the way it does. If I fully bypass the first stage, it will no longer sound the same...

[Chuck H]

And I mentioned that because your amp is voiced that way (with the partially bypassed cathode) that DC heaters on the early stages were not a bad idea. Partially bypassed simply means that the cap value bypassing the cathode circuit is too small to pass all audio into the LF. But it also means that any noise on top of the cathode, like filament hum, if present, can also get into the audio path because it's not shorted to ground via the cap. Since noise on the first gain stage will be amplified most it's a good idea to design that stage with low noise in mind.

Partially bypassing a preamp cathode has a shelving effect on gain at the pass frequency. The effect varies with impedance and, of course capacitor value. It can be duplicated with high pass filter circuits instead allowing the first stage cathode to be fully bypassed. Then again the HP filter adds series resistance and, therefor, shot noise. So in that case you have to pick your nits. Or, in the case of your amps design, employ DC heaters.

I wasn't actually suggesting a change to your amps circuit. Just making an observation out loud that partially bypassed first gain stages have become common in guitar amp designs and then DC filaments. I know that tube guitar amp design isn't something you can learn in a class. I'm certainly not formally educated in electronics but it just occurred to me that in the hands of guitar amp designers fundamental things like fully bypassing the cathode of the first gain stage get overlooked and then noise, like filament hum needs to be dealt with by using additional circuitry. Of course your design would sound different if the first stage cathode were fully bypassed. But a design that starts with a fully bypassed first stage cathode could be voiced for the same tone in the end game and it's possible that DC filaments wouldn't have been considered necessary.