There are several threads on DIY amps such as the Bottlehead Crack, it is easy to get a collective interest in any kind of project here, people experiment and discover all kinds of upgrades. I cannot talk intelligently about it till I see a schematic but I like solving things.
I can definitely send you a board, but getting the kit together is a bit too much of a distraction. I'll finish the BOM and the schematic and do a wrap-up chapter, but I don't think I'll be doing any further development.
So why am I letting this go?
1. It accomplished its primary objective, to be a neat-looking coaster.
2. It does operate as an amp, though it doesn't meet all of the performance criteria defined in the first chapter.
3. I have lots of other things to do, and fixing it--for real--will require significant changes, including some that will make it less suitable for a DIY project.
What do I mean by "significant changes?"
Well, the primary thing this amp needs is more voltage on the tube. That's the big differentiator between it and the original Vali. Vali had 2x the amount of voltage to play with. A 60V B+ is way, way, way better than 30V. Period. You ain't changing tube physics. The 6088 wants a lot more volts than we can give it in the coaster. The 6418 is closer to ideal. I haven't been able to find a subminiature tube that runs on lower volts, or not one that is gettable, at least.
And that's it. We can add more volts or find a tube that works better at lower volts. That's pretty much it for significant performance increases.
- Adding more volts is out, since you don't want 80V running around on a bare board without a chassis.
- And, like I said, I haven't found another tube that works better at low volts.
"But what about more feedback, or changing it to just a follower?" some of the experienced tube guys might ask.
- Feedback? It only has about 12db of gain. Not much feedback to play with here. Plus, then it isn't "no overall feedback." Ha!
- Follower? It's a direct heated cathode. Have fun with that one.
So, there you go. Sure, some small improvements can be made. But the big wins would be in things we don't want to do, can't find, or don't work with this tube. The best bet would be to find a subminiature tube that works at lower volts. You can start here
http://pw2.netcom.com/~wa2ise/radios/penciltubes.html and search for "s:15", confirm the tube has a 1.2V or 1.25V heater, and see if there are any tubes you can find. Then it's a matter of running them through the board, testing different plate loads, and seeing how they do.