Miller capacitance is a subject that is associated with any tube or transistor. In layman’s terms, it affects the circuit as a low pass filter. It does this because as it amplifies the tube is ‘fighting’ against the grid to anode capacitance. (Meaning as it is amplifying the grid to anode capacitance is charging and discharging similar to an astable oscillator.)
Millers capacitance = the grids capacitance times the anodes capacitance. This is what Millers capacitance is! (all interelemental capacitances mentioned need to include stray capacitance
Miller’s capacitance is normally associated with triodes but applies to all valves. It is ore noticeable in the triodes though (because pentodes and tetrodes have screengrids which ‘sheilds’ the valve.). Which is unlucky for all us audio designer builders (unless you use the low pass filter to your advantage
) because as most of you know the preamp tubes are almost always dual triodes.
Right this is how it affects the audio circuits. I said in layman’s terms Millers capacitance affects low pass filtering. This is actually only ½ the story. Millers capacitance does have an effect on low pass filtering but it is not the be all and end all! Really its up to the input capacitance that decides it!
Input capacitance is measured by the sum of grid to cathode capacitance plus the grid to anode capacitance times thee stages gain + 1.
If you were to write this as a formula it would be:
GK=Grid to cathode capacitance
GA=Grid to Anode capacitance
A=Gain of the stage
As the more knowledgeable of you should know. All these capacitances are really tiny, so they wouldn’t have much effect on the overall outcome right? Wrong! Because this is times by the gain of the stage it can have huge affects.
“Yeah so what if it makes a load of numbers change it’s the sound that matters” Most of you will be screaming at me after this brief theory lesson will say! It is true so lets take a few examples. Right lets take the good old 12AX7 OK? GK=1.6pf GA=1.7pf (I looked at tube data sheets for those I am not that much of a robot!!) lets use a gain of roughly 52 and a stray capacitance of 0.7pf on both GK and GA.
InC=2.3+(52+1)*2.4=127.1pf of input capacitance.
A 12AY7 has a GK=1.3 and a GA=1.3 (keeping with the 0.7 stray and 52 gain) would equal:
of input capacitace. As you can see the 12AY7 could be replaced instead of a 12AX7 for a quicker low filter cut off! (although there are other considerations to think about).
But do you know what? It isn’t that simple (what a surprise!) because input capacitance by itself does not affect low frequency filtering, as the more knowledgeable will know it is reactance which will decide frequency cut offs.
The formula for this is:
So lets keep a constant of 68k for the input resistance (standard value for an input resistor) So our 12AX7 example would equal:
=18.4kHz (1 dp)
Where as the 12AY7 example would be:
=21.7kHz (Again 1 dp)
This means that the 12AY7 example will cut off more of the low frequencies than the 12AX7 example. But as you can see there are many variables and chances to experiment! But also take into consideration (as microdmitry said) all valves are different!
This is my interpretation of the Miller effect micro please inform me if this is right or wrong! Also take into account I am 15 yrs old so don’t flame me for a few mistakes
Thank you, everyone for keeping this thread with intelligent points and arguments. I will (try) and inform you of all the different factors in choosing tubes!!!
Also anyone who has done any real research into the interelemental capacitances of different makes please state them. :P