I am trying to figure out the right input resistance to use to roll off high frequencies above 8Khz in a preamp 12ay7 tube. I need the total capacitance, Miller capacitance, same thing right?
Values added for 12ay7:
Ct = (gk=1.3pf + stray) 27mu*(gp=1.3pf + stray)
I found article with a 12ax7 showing that the stray capacitance is 0.7pf.
Would it be close enough to just assume they are the same?
How would I find the stray capacitance if not?
Is there any chart showing all tubes miller capacitance?
Values added for 12ay7:
Ct = (gk=1.3pf + stray) 27mu*(gp=1.3pf + stray)
I found article with a 12ax7 showing that the stray capacitance is 0.7pf.
Would it be close enough to just assume they are the same?
How would I find the stray capacitance if not?
Is there any chart showing all tubes miller capacitance?
New valve/tube to me somewhat like an ECC83-
gk-1.6pf
ak=0.46pf
ga=1.7pf
anode (plate) differs slightly in capacitance--ak depending on the manufacturer.
My figures are all well known UK/European manufacturers no longer in business.
gk-1.6pf
ak=0.46pf
ga=1.7pf
anode (plate) differs slightly in capacitance--ak depending on the manufacturer.
My figures are all well known UK/European manufacturers no longer in business.
When in doubt, build one out on a breadboard and put a 1M pot in series with the grid of the 12AY7, then measure the output until you get what you want.
I would also recommend looking at the Sallen-Key topology since you didn't specify how steeply you'd like things to roll off.
I would also recommend looking at the Sallen-Key topology since you didn't specify how steeply you'd like things to roll off.
The internal capacitance is insignificant compared to the cap you need to create a low-pass filter on the grid. It takes 20K R and .001uF to create the 8Khz corner frequency.
So don't bother with trying to create a low pass filter with the tube's capacitance, just build a standard LPF instead. Wont I lose a lot of signal that way? The 12AY7 is a low mu tube to begin with but it matches the in/out impedance I am looking for, 12ax7 wont cut it. Maybe a 12at7? I think the two are probably fairly interchangeable, I could do some experimenting.
Now that I think about it the resistor needed with pico-farads would be a far worse problem. Thanks.
Last edited:
Great Ideas. As for the slope, not so sure really. The reason for this is I want to have a switchable rolloff on a phono preamp. 16rpm records do not record frequencies above 8K. Above that will just be noise. The ceramic cartridge will be going through a Mosfet stage first, but the 12ay7 will still have a high input impedance.
A 10 or 12 band EQ would be a lot more fun and usefull that a permanently fixed low-pass filter built into the pre,....
Stray capacitance is circuit and layout dependant. Miller is the parasitic grid to plate capacitance multiplied with the gain of the tube. You will only achieve gains close to mu using a choke or CCS on the plate. So your gain is likely 25-30. 12AY7 has Cg-p of 1.3pF. The miller C is then roughly 30pF. The miller C varies slightly from tube to tube b/c gain varies slightly as well as the Cg-p.
So basing your low pass filtering on miller is ok for elimenating high frequencies but not for making a precise filter. You will need a very large resistance and then you have a noise generator in your signal path.
Making a LP filter with sensible values will not attenuate your signal other than above the cut off frequency but that is what you want.