All Triode Oscillators have Miller Capacitance.
And a 4-125 is a Tetrode, not a triode. Therefore, the Miller Capacitance is much much lower than the miller capacitance of a triode.
And as always, "All Generalizations Have Exceptions".
Would you like to use the 4-125 as a triode:
1. Ground the Plate
2. Connect the Screen to an output transformer primary, and B+ to the other end of the primary.
The screen, if used as a plate of a triode, can dissipate 20 Watts (even a little more Watts, because the plate is now cold).
u, mu (G1 to G2) = 6.2
The filament takes 32.5 Watts.
Now we are using the 4-125 as a triode that has more power in the filament than in the "Plate" (Screen).
I do not have a rating for the G1 to G2 Transconductance, so I can not calculate the Screen's "Plate Resistance".
And a 4-125 is a Tetrode, not a triode. Therefore, the Miller Capacitance is much much lower than the miller capacitance of a triode.
And as always, "All Generalizations Have Exceptions".
Would you like to use the 4-125 as a triode:
1. Ground the Plate
2. Connect the Screen to an output transformer primary, and B+ to the other end of the primary.
The screen, if used as a plate of a triode, can dissipate 20 Watts (even a little more Watts, because the plate is now cold).
u, mu (G1 to G2) = 6.2
The filament takes 32.5 Watts.
Now we are using the 4-125 as a triode that has more power in the filament than in the "Plate" (Screen).
I do not have a rating for the G1 to G2 Transconductance, so I can not calculate the Screen's "Plate Resistance".
Last edited:
Miller effect does not work inside the tube; it presents a low impedance outside the tube. However, add in some parasitic reactance and a little transit time delay and Miller 'capacitance' becomes a complicated impedance which could (under certain circumstances) include some negative resistance.
DF96,
If I understand “Miller Capacitance” correctly, it is like this:
Triode:
Consider a triode that has 1 pF of capacitance from control grid to plate (Cgp).
Connect the triode in a circuit that has a gain from control grid to plate of -49 (the minus sign indicates the opposite phase). For simplicity, lets drop the - sign, and just call the gain 49.
But if we were to measure the charge current due to the changing voltages, we get something
like this . . . The charge current is proportional to the capacitance, and to the voltage change,
Cgp (1 + 49). The charge current is the same as if we were effectively driving a 50 pF capacitor.
“Miller Capacitance”, indeed.
Perhaps a poor choice of a word for an effect, it is not a real capacitor.
“Miller Charge Effect” might be a better choice of nomenclature.
Tetrode:
There is a Screen that is inserted between the control grid and the plate.
In Tetrode operation, the Screen is at AC ground and RF ground (as best as we can bypass it to ground, and as low of inductance lead(s) as the tube is designed to have).
Consider a Tetrode that has 0.05 pF. Cgp = 0.05
Set the circuit so that the gain again is -49 (call it 49).
0.05 pF (1 + 49) = 2.5 pF.
The Miller Charge Effect is as if we are driving 2.5 pF.
The 4-125 Cgp is 0.05pF.
The Miller Charge Effect is much less than for most triodes.
But if you wire the 4-125 as a Triode (connect screen to plate), then the Miller Charge Effect is much more dependent on the capacitance of the control grid to the screen, Csg.
Triode and Tetrode:
In the cases above, we also have to consider the capacitance of the control grid to:
Filament leads and the Cathode.
Negative Resistance:
The Kink effect is well established for Tetrodes and Pentodes.
Letting the plate voltage swing far less than the screen voltage is the region where the kinks, if any, occur.
Kinks are negative resistance.
If I understand “Miller Capacitance” correctly, it is like this:
Triode:
Consider a triode that has 1 pF of capacitance from control grid to plate (Cgp).
Connect the triode in a circuit that has a gain from control grid to plate of -49 (the minus sign indicates the opposite phase). For simplicity, lets drop the - sign, and just call the gain 49.
But if we were to measure the charge current due to the changing voltages, we get something
like this . . . The charge current is proportional to the capacitance, and to the voltage change,
Cgp (1 + 49). The charge current is the same as if we were effectively driving a 50 pF capacitor.
“Miller Capacitance”, indeed.
Perhaps a poor choice of a word for an effect, it is not a real capacitor.
“Miller Charge Effect” might be a better choice of nomenclature.
Tetrode:
There is a Screen that is inserted between the control grid and the plate.
In Tetrode operation, the Screen is at AC ground and RF ground (as best as we can bypass it to ground, and as low of inductance lead(s) as the tube is designed to have).
Consider a Tetrode that has 0.05 pF. Cgp = 0.05
Set the circuit so that the gain again is -49 (call it 49).
0.05 pF (1 + 49) = 2.5 pF.
The Miller Charge Effect is as if we are driving 2.5 pF.
The 4-125 Cgp is 0.05pF.
The Miller Charge Effect is much less than for most triodes.
But if you wire the 4-125 as a Triode (connect screen to plate), then the Miller Charge Effect is much more dependent on the capacitance of the control grid to the screen, Csg.
Triode and Tetrode:
In the cases above, we also have to consider the capacitance of the control grid to:
Filament leads and the Cathode.
Negative Resistance:
The Kink effect is well established for Tetrodes and Pentodes.
Letting the plate voltage swing far less than the screen voltage is the region where the kinks, if any, occur.
Kinks are negative resistance.
Last edited: