I had been steering clear of the 300B because I feel that the prices being charged for this tube are outrageous.  Especially so, when you realise that a decent version of the superior (IHMO) 845 is $44.  Still, I was curious:

300B SE 001.jpg (127492 bytes)        300B SE 002.jpg (144898 bytes)

This amp uses a single 5842 for gain and drive, the drive stage being choke loaded.  The output stage is fixed bias.  This keeps the signal path very direct.  The power supply uses a GZ37 (it can use a GZ34 also) for slow turn on, which feed a C-L-C filter.  Each output stage is fed from it's own regulator.  Each regulator uses one section of a 5751 as an error amplifier and  a mosfet as the series pass element.  The drive stage takes it's power from each respective output stage regulator. This power is already very quiet and stable, so all that is required is to buffer each drive stage from the output stage using (in this case) a mosfet, the gate being held at the required voltage by a gas tube and RC noise filter (the gate requires very little current).  The power supplies are bypassed using a combination of both polypropylene and paper oil capacitors.  The 300B filaments are fed using DC current regulation: 

The advantages are:

1/  There is no switch on current surge to challenge the fragile (and expensive) filaments

2/  The resistance 'looking back into' a current regulator is much higher than that of a voltage regulator.  This mitigates the ac short circuit that a voltage regulator presents to the end-to-end signal which develops along the filament as a consequence of the bias gradient* which, I suspect may be one reason why direct heated triodes energized using current regulation are considered to sound better.  To further increase the impedance "seen" by the end-to-end filament signal, I couple the current regulator to the filament using filament chokes.

*  The bias gradient results from the DC voltage applied to the filament.

300B OPT Potting.jpg (61139 bytes)I wound my own output transformers, the black box behind the caps on the right, visible in the picture above contains both output transformers similarly to the 845 amp. The picture shows the pair of double C core output transformers (as the potting wax sets) which are mounted side-by-side, clamped between two oak end plates.

300B SE Schematic:

I am using the Sophia branded TJ perforated plate 300Bs.  To my ears and those of my friends, these are as good as new production 300Bs get but not as satisfying as the 845.  The regulated power supplies make this amplifier fairly easy to adapt for other tubes.  I have done temporary adaptations to try both 45s and 50s.  The 45s are unquestionably superior to the 300Bs but the 50s are another experience yet.  They have all the speed and apparent heft of the 845s as well as the resolution, imaging, tonal balance and musicality of that tube (845).

Recently, I re-worked the bias networks to allow me to plug in a pair of NOS 50s I have (the filament current is the same as the 300B).  Since this amp uses fixed bias, this might be a little risky.  I replaced the grid resistors with chokes (Lundahl LL1670s) and during the bias re-work, reduce the DC source resistance of the bias voltage.  The tubes seem (so far) to be perfectly stable.  A factor in mitigation might be that the plate voltage is 400V, 50V below the maximum.  These tubes sound amazing!  The 300B cannot compete and I regret to admit that it surpasses my 845 amp.