Simple 12ax7 preamp has no high end (LTSpice sim)

[rectifryer]

I have, erm, bastardized various simple preamp designs via LTSpice trying to get something very balanced but with over 65 db gain. I have seen some cathode follower designs that are nice but are low gain. Unfortunately my "design" has terrible high end. What is causing this in my circuit?

I am working with a transformer that only puts out about 120v rms. I am assuming it will be a little less after I run it through a rectifier circuit.

 

[12afael]

the grid stopper are too high. it produce a low pass filter together with the parasit capacitance of the tube. grid to plate // grid to anode. try with a value from 1K to 50k.

also you can use u for micro on caps and K for kilo at resistors at ltspice.

 

[volker]

• Use a doubler with your transformer, this will put you around healthy 300V.
• Both tubes need a grid resistor to ground.
• If your intended use is a mic preamp, you will need a stepup transformer.
• A bare 12AX7 won't drive much of anything. The cathode follower you mention would be a good start (while voltage gain is ~1, it has current gain), preferably with a different more powerful tube.
• You need a way to control gain, either with a pot between the stages or via feedback or some kind of pad in front or after the amp.

 

[ruffrecords]

As has already been said the grid stoppers are too high. In fact with a 12AX7 you don't need then at all unless you deliberately want to roll of the top end as in a guitar amp for example.

Cheers

Ian

 

[rectifryer]

I implemented your suggestions to the best of my understanding. What's hilarious is I was using existing values in existing designs :/

The circuit is considerable more balanced, but it still drops 4 db of response from 10 hz to 20khz.

Is there some sort of interaction between triodes that causes this or is this just a cumulative miller effect? Even if I remove the grid resisters entirely, it doesn't seem to make more difference that what the current response graph shows. I will probably remove them for simplicity.

 

[12afael]

the grid of the second stage is floating it will not work that way, it must be referenced, usually to ground. it look more like guitar preamp than a mic preamp, 1M at the input is too high unless you have a 200 to 1M transformer.

 

[ruffrecords]

What is left now is what is known as the Miller effect. The anode the grid capacitance of the second tube is magnified by the tube gain and appears between its grid and ground. This forms a pot divider with the output impedance of the first stage which give you your treble loss. There are two things you can do to improve this. First decouple the cathode resistor of the first stage with 220uF. This will reduce the output impedance of the first stage and move the hf loss further up the spectrum. Secondly you can reduce the value of the first stage anode load. This also reduces the output impedance of the first stage.

Cheers

Ian

 

[rectifryer]

Thanks alot guys.  Ian, your 6CG7 preamps look pretty awesome.