I put the filter cap can back in the amp. Even with the stock circuit, it sounds better - seems to have more bass and note definition. It wasn't humming, but I replace filter caps on old amps and radios as a matter of course.
The original circuit is definitely tilted toward cutting bass and mids. It's super clean and has a lot of chimey high end. I'm going to change all that...bwhahahahaha. It sounds good as is, but I'd like more gain and some thump in the bass (as much as an 8 inch speaker can thump...).
I'm going after two areas to hot rod the amp.
The first is to raise the voltage on the preamp tube. The two plates have 106 and 150 volts on them. A little more voltage will increase both gain and headroom. To that end, I changed the two 220K plate load resistors that go to the preamp to 100K. I'm now seeing about 195 volts on the first stage, and 205 on the second - right in the sweet spot for good gain and headroom.
Tone-wise, I'm going to change the values of the capacitors in the signal chain.
The main culprits are the three .01 uF caps. That value limits the amount of bass that can pass through the circuit. Basically, the higher the value, the more midrange and bass that is passed. There is a limit - too large of a value and the amp will sound muddy, or it 'motorboat' - it will oscillate at low-frequencies, which we definitely don't want.
There's one cap from the inputs to the grid of the first triode in the 12AX7, then another from the plate (output) of that stage to the volume control, and then a third is the coupling cap from the preamp to the power tube (6V6).
When this amp was designed, distortion was a thing to be avoided. The lower value caps were a way to help keep the tone clean. And I'd venture to guess that the designer was also concerned about putting too much bass through that puny output transformer as well.
But I'm not afraid. I will boldly double the values of the coupling caps!
The shot above is some of the experimentation. I went up to a value of .02 uF on all the caps, and played around some with some different types of caps. You can also see the red alligator clip - it's grounding a 1.5K cathode resistor on the first stage. The stock circuit had no resistor there at all. This is pretty much standard stuff - Fender used these same values in tweed amps.
There's another K-40Y going to the volume control. It's actually not soldered in yet in this picture. That's so you can see the 1.5K resistor and 25uF bypass on the first triode of the preamp.
To the right is another Russian PIO cap. I'm not sure what these are called, but I really like them. It's actually the first time I've used one. This one is the coupling cap to the 6V6. A little more 'chime' to the high end on these. Maybe not great for a hi-fi amp, but very nice for this little amp. I had fun comparing the different types. I almost put an Orange Drop for the 'middle' cap but decided to go with the PIO. The PIO sounded a little smoother in the bass - the Spragues are really clear and tight sounding, but I decided against them for this amp.
I have no idea what I'm talking about.
The "P8U" is the model. The "C" is for a ceramic magnet.
Note the hand-work on the bevel cut.
I also put a modern AC cord on it - you can see that to the left.
The old speaker wires were really cheap looking so I also put some modern cloth-insulated cable in there. Looks pretty retro I think.
The amp now has a lot more gain, and has clean chimey highs, nice mids and surprisingly solid bass. It's nice and clean up to about "5" on the volume, then it starts to break up more and more. It's not a distortion machine, but it gets a real nice cranked rock or heavy blues tone.
UPDATE 12 Nov 2013: I played lap steel (Oahu Diana) through it last night. Sounds good! It may become my dedicated lap steel amp.