[Amps] Tuned input and 4CX1500B cathode impedance

[Paul Kelley N1BUG]

Fellow amp addicts,

I have been building amps for years but I am on somewhat new ground 
here.

I am trying to estimate the drive impedance of a 4CX1500B in order 
to design a tuned input with minimal guess work. I understand the 
impedance varies throughout the input cycle, but there must be some 
value which can be used to at least get close on the tuned input 
network. How does one arrive at that value? I have read that it can 
be derived from tube data sheets, but I fail to understand how. I am 
sure I could experimentally derive a reasonable input network but I 
would rather learn something in the process, if possible.

The tube is cathode driven with grid at RF ground and screen at RF 
and DC ground. While the Eimac data sheet only gives operating 
conditions for grid driven service, under direct inter-electrode 
capacitances it does specify 38pf Cin "with grounded grid and 
screen". That would seem to apply to my cathode driven arrangement. 
I still don't know how to arrive at a figure for the real part of 
the impedance.

I have come up with a somewhat wild guess that it is about 100 ohms 
shunted by 38pf. This is derived from SWR measurements directly at 
the amplifier input with real operating conditions at frequencies 
between 1.8 and 50 MHz, using 1) no input swamping resistor; 2) 50 
ohm resistor across the input; 2) 100 ohms across the input. From 
those measurements I can seemingly justify my 100 ohms shunted by 
38pf hypothesis, but I would be lying if I said I had much 
confidence in this.

 From this work it is apparent that having a 50 ohm resistor across 
the input is desirable in that 100W is perfect to drive it with that 
in place. Is there any reason why I shouldn't use a pi network tuned 
input with the 50 ohm shunt resistor on the tube side of the 
network? I don't recall ever seeing a "swamping" resistor in 
combination with tuned input.

I will appreciate any guidance and education I can get here.

73,
Paul N1BUG