first audio test

So I hooked the amp up to a tiny 8 Ohm speaker (computer type, must be about a 2" driver).  And good news is, there's audio, and the basic systems seem to work as intended.  There are a bunch of caveats, though.  Which is to be expected, I'm just happy there was sound at all, given that this is a total redesign of the circuit, never before tested in any form.

Both preamps work, but the AB763 is noticeably louder than the VTX.  It's awfully hard to tell, with this tiny speaker, how loud the real amp will be.  Maybe not nearly loud enough.  Might be, at full volume, that it's just putting out about 1W or so.  Or maybe it's just due to speaker inefficiency, etc..  I think I'll have to rig up a way to connect the real 12" speaker, before putting everything back in the cabinet, because I might have a serious problem with signal levels in this new circuit.

The new AB763 chassis needs to be grounded.  I put a screw hole for a ground strap, but I haven't connected it yet, and it needs it.

At full gain, the AB763 goes into oscillation or saturation (I assume) when the treble control is set too high: there's a maximum point, beyond which the sound cuts out.  I've expected I might have to add some 10pF "taming" capacitors between plate and cathode on the preamp tubes, and this is the sort of reason why.  I welcome the chance to address this issue here, because I generally run into this problem with all my high gain tube amp designs, and I need a general solution (other than don't turn the treble up all the way) before I can consider them "ready for market".

The nfb circuit seems to do nothing audible, in or out.  The presence and depth controls have no audible effect.  I think I'm just way off in terms of dB in this loop, with my novel concept of running it into the cathode side of the cathodyne phase inverter.

The VTX preamp, although it seems to have a much lower signal level than the AB763, still bleeds into the output when the preamp select switch is set to cut it out.  It's like the contact resistance on that switch is too high; which is very possible, these Chinese switches are terrible and I had to be really careful soldering to them, too much heat would start to melt the plastic around the lugs and they'd shift position.  I was afraid I might find the switches completely non-working; they seem to work, but problems with solidity of contact would not be a surprise.  I may need to order some entirely different (and presumably more expensive!) switches...

As mentioned, I need to increase the brightness of the blue LED in the RGB pilot light.