improvements on 12AX7 12AT7 EL34 schematic?

[zintolo]

@THD+N
I think you are looking at the original schematic, not latest updates.

It was simply an hybrid cascoded LTPI with cathode followers driving the output tubes using UL and conventional nfb.
It sounded fine but not great.

 

[Zung]

Another layout suggestion - easier to keep signal, output and power supply seperate.
M&A TVA-10

That's the one well known for its strong affinity for self-immolation

 

[Francois G]

Care to explain?

 

[zintolo]

I am considering what is the limit for the current of the phase splitter and the capability to drive the grids of the amp.

The slewrate of the amp, considering the bias at -40V, so 80Vpp needed of swing, at 50 kHz is:
2 x pi x f x V = 2 x 3,14 x 100 kHz x 80 = 25 V/us

Anyone knows the input capacitance of a EL34 with 23 and 40% UL?
Cg1 = 15 pF
Cag1 = 1 pF
but what about the Miller capacitance when the screen is not at a fixed voltage?

C = i x dt / V
dV/dt = 25 V/us

then, if it was a pentode:
15 pF = i / slewrate

so
i = 15 pF x 25 V/us = 375 uA would be needed.

Considering that with 40% UL g2 swings about -2.3 times g1, am I on the safe side considering twice the pentode input capacitance?
That would lead to 750 uA to have the right slewrate, and it would still be fine.

What is a good rule of thumb? To have 2 to 3 times the needed current?