Question about input networks pos & neg

[AMLOVER]

Hi,

I am wondering what are the pos and neg of each of the two usual input networks for  grid driven tetrodes and why the network type 2 called parallel with linked input is the most used. I have used both with some success but still can't understand the pos & neg of each one.

All the opinions appreciated.

[W3RSW]

There is no positive or negative in a DC static sense for the alternating current, RF (radio frequency) coming into either circuit.

In both cases the DC bias connection point will be negative with respect to ground since you are grid driving the tube.  You need to look up and study biasing an amplifier by class, A, B, AB and C , depending on the operating service, but all have a negative grid with respect to the tube cathode and ground.
 
In case two ( lower diagram) the DC bias travels through the secondary winding ( under the primary, outside coil as your circuit is drawn) thence to the grid.  The RF is isolated from the DC bias since the coils are not connected.  Only RF can pass across the coil windings setting up an EI field.

In case one  (upper picture) the capacitor right before the tube grid isolates the DC to the grid and the bias travels through the grid resistor directly to the tube grid.

Both diagram's bias points will have negative, resting DC voltage with respect to ground at the tube socket.

Defining the tube elements:  Cathode ( bottom of tube symbol, actually shared with filament heater voltage in your diagram). ; grid , (dotted line in middle; )  anode or "plate" as top symbol.  The filament/cathode will be at ground, not shown in your diagrams.

Ok , the upper circuit (pi network, called a pi because the coil and two variable caps look like Greek letter pi) is most used in more modern amplifiers, particularly linear amplifiers that you've commonly seen in class AB, such as an Acom or Alpha.  The output networks of these amplifiers are also pi nets.

The lower circuit (coil coupling) was the most commonly used to feed the final stage of an amplifier up through the 1940's ; many were plate AM modulated and coil coupling is still used in interstage, driving stages of tube amplifiers.  

[w4bfs]

a great question !   which is better ?   both have the potential to get the job done so which way do ya go ?

look at an example:    a 4-400 pa stage in class C operation

(1) what is the driven grid minimum input impedance ?   from the Eimac data sheet:

290 Vpeak grid drive at 12mA grid current sets the minimum input impedance.  the 290V peak divided by 1,4165 yields about 200V average swing so    200 / .012  =  16,666 Ohms and P = 200 x .012 = 2.4W

most of the time in practice a swamping resistor is used for stability so lets assume a 20k resistor is used which lowers the input impedance to around 9k ....ymmv

so we are attempting to raise the impedance from 50 to 9k Ohms or times 180 .... this is at the extreme end of the range of a pi-net (ckt 1) ... you may find the real world  LC values vanishingly large/small

ckt 2 is in its element with this problem because the impedance is raised by the square of the turns ratio so  if a turns ratio of 14 were used then the other needed C values are readily applied

[W3RSW]

Also you've labeled the bias as "safety bias voltage."  It is really the operating voltage necessary for the tube to properly even operate.  It's only 'safety' In that without it cathode to anode current would be excessive unless the tube was designed for zero bias such as an 811, 572b, etc, and even these require a small amount of bias for best linearity at higher plate voltages.

[AMLOVER]

BFS,
you made it so clear to me that I don't need to read it twice.
Large input impedances need parallel input networks and the transformation is easily made by the coils factor. Smaller ones could be held by a pi network as far the transformation to 50 ohm is in the real world materials' rates...
Thank you for enlightening my darkness...

RSW,
Thank you, too.