Tweet
I am planning to build a 2x 6V6, paraphase PI amp. And was using as guide, for initial schematic and components values, the Gibson GA-6 (interim model, i.e., 12AY7 preamp & 6SL7 PI) and the Rickenbacker M-11 with 6SC7 preamp & PI.
The theory of the paraphase PI operation is pretty simple:
The signal to be amplified, goes through a tube driver to the push-pull output tube "1".
From this same output, a signal is derived, and goes to a resistive divider (like a fix potentiometer), reducing this signal voltage. This has to be done in such way, that when this signal is amplified by the other tube driver, it produces the same output of the first tube driver (and opposite phase), before it is fed to the push-pull output tube "2", in order to have a balanced push-pull operation.
So, the first requirement here, is that this voltage reduction, is equals to the inverse of the tube driver gain, so the second tube driver produces the same output of the first one.
In other words, Gain = (Rg + R_didv) /Rdiv… So far, so good.
Also, in this PI design, if the first tube driver goes into cut-off, then the second driver also stops, because it is fed from the first tube. For this reason, it is a good design practice to bias these as linear as possible and with maximum headroom (Class "A" operation).
However, both of these designs appear to be violating these two requirements, and thus generating (from forums limited information) very unpleasant distortion.
THE QUESTION IS: does anybody there with hands-on experience on any of those amps, to elaborate further on my observations?
Thanks!
The theory of the paraphase PI operation is pretty simple:
The signal to be amplified, goes through a tube driver to the push-pull output tube "1".
From this same output, a signal is derived, and goes to a resistive divider (like a fix potentiometer), reducing this signal voltage. This has to be done in such way, that when this signal is amplified by the other tube driver, it produces the same output of the first tube driver (and opposite phase), before it is fed to the push-pull output tube "2", in order to have a balanced push-pull operation.
So, the first requirement here, is that this voltage reduction, is equals to the inverse of the tube driver gain, so the second tube driver produces the same output of the first one.
In other words, Gain = (Rg + R_didv) /Rdiv… So far, so good.
Also, in this PI design, if the first tube driver goes into cut-off, then the second driver also stops, because it is fed from the first tube. For this reason, it is a good design practice to bias these as linear as possible and with maximum headroom (Class "A" operation).
However, both of these designs appear to be violating these two requirements, and thus generating (from forums limited information) very unpleasant distortion.
THE QUESTION IS: does anybody there with hands-on experience on any of those amps, to elaborate further on my observations?
Thanks!
Comment