Replacing the cathode resistor with a current source in SE ?

[OneyedK]

(1) sorry, but I don't see how this is relevant? No matter how you decide to bias, you still have to couple with previous stages.

(2) Negative power supply can be rather simple, since the current needed is almost nothing, unless you're talking about using an interstage transformer, wich brings us back to (1).

(3) Without the transformer (again) the LPF formed by Rg1 and the miller capacitance of the tube will get rid of most noise, not that I want to say that we have to design a negative power supply without care!

But hey, just my 2 cents, maybe I'm looking at it the wrong way
Enlighten me please...

[kevinkr]

In most applications the bias supply is completely divorced from the grid circuit by the grid resistor. As the previous poster pointed out most tube amplifier circuits are RC coupled. In the case of an IT application it is a bit more complex, but honestly speaking I have never found large value cathode bypass caps no matter who makes them to be particularly transparent.

I use fixed bias in just about every power stage I have ever designed, and find the inconvenience of an additional supply inconsequential.

Battery bias in some cases is convenient, in other cases one can use LEDs in the cathode circuit for what is effectively fixed bias operation. (See SY's red light district, etc. for examples.)

[rdf]

Originally Posted by kevinkr

Battery bias in some cases is convenient, in other cases one can use LEDs in the cathode circuit for what is effectively fixed bias operation. (See SY's red light district, etc. for examples.)

Possibly a better option is to current source the LEDs and put them in the grid circuit in series with the grid resistor. Since the dynamic signal current through them is infinitesimally smaller, so is any impact from non-linearity.