basic feedback question

[mikepukmel]

I never quite got the signal flow through negative feedback. For the simplest example (I think) take a champ. Read through some vacuum tube books, and managed to understand why signal is inverted at the plate relative to the input grid. And signal is in phase between input and output of the output transformer.

If that's the case, then a positive half cycle at the input flips to negative after the first half of the preamp tube. The signal at the cathode of the second half of the preamp tube is in phase with the grid so its also negative half cycle. at the plate of the second half of the preamp tube, the signal is positive again. Then flips one more time going through the output tube.

Why would the speaker signal be a negative feedback if tied to the cathode of the second half of the preamp?

[Helmholtz]

[QUOTE=mikepukmel;n913639]
Why would the speaker signal be a negative feedback if tied to the cathode of the second half of the preamp?
[QUOTE]

The phase relations in your drawing are correct. As the second preamp triode is within the global NFB loop, I consider it part of the power amp.
The signal at its grid is in phase with the speaker output, so the triode grid is the non-inverting input of the power amp.

The cathode of the triode is used as input for the NFB signal. This means that here the tube is used in common grid amplifier mode, where the amplified plate signal is in phase with the cathode input signal. In other words, grid and cathode input signals produce opposite phase plate signals.
This makes the cathode of the triode the inverting input of the power amp.

Reversing the leads of either OT primary or secondary will turn NFB into positive feedback and will most probably result in oscillation (howling).

It is not always clear from the schematic if the OT inverses the phase or not, but you were lucky .

[loudthud]

What always confused me is the talk in textbooks about the feedback being "out of phase" with the input. In non-inverting circuits, the feedback is actually in phase with the input.

As Helmholtz points out, the Grid of V1B acts as the non-inverting input and the Cathode acts as the inverting input. Think of how the amp operates with the feedback disconnected. The amp responds to the difference in signal between the non-inverting and the inverting inputs. Without feedback, the inverting input is not used. Feedback reduces the amount of signal between these two inputs. If the feedback increased the amount of signal between the two inputs, that would cause the amp to oscillate.

Check this thread:

https://music-electronics-forum.com/...orrect-phasing

[mikepukmel]

[QUOTE=Helmholtz;n913644][QUOTE=mikepukmel;n913639]
Why would the speaker signal be a negative feedback if tied to the cathode of the second half of the preamp?

The phase relations in your drawing are correct. As the second preamp triode is within the global NFB loop, I consider it part of the power amp.
The signal at its grid is in phase with the speaker output, so the triode grid is the non-inverting input of the power amp.

The cathode of the triode is used as input for the NFB signal. This means that here the tube is used in common grid amplifier mode, where the amplified plate signal is in phase with the cathode input signal. In other words, grid and cathode input signals produce opposite phase plate signals.
This makes the cathode of the triode the inverting input of the power amp.

Reversing the leads of either OT primary or secondary will turn NFB into positive feedback and will most probably result in oscillation (howling).

It is not always clear from the schematic if the OT inverses the phase or not, but you were lucky .

Thanks Helmholtz! Well, wasn't so lucky the first power up, it was out of phase and did make awful noises. *** I think I got it!

[mikepukmel]

What always confused me is the talk in textbooks about the feedback being "out of phase" with the input. In non-inverting circuits, the feedback is actually in phase with the input.

As Helmholtz points out, the Grid of V1B acts as the non-inverting input and the Cathode acts as the inverting input. Think of how the amp operates with the feedback disconnected. The amp responds to the difference in signal between the non-inverting and the inverting inputs. Without feedback, the inverting input is not used. Feedback reduces the amount of signal between these two inputs. If the feedback increased the amount of signal between the two inputs, that would cause the amp to oscillate.

Check this thread:

https://music-electronics-forum.com/...orrect-phasing

Thanks Loud, reading . . .

[mikepukmel]

THis is the part that was staring me in the face, but didn't 'sink in':

"In other words, grid and cathode input signals produce opposite phase plate signals." Thanks. OK its sinking in down through the sand.

[Helmholtz]

"In other words, grid and cathode input signals produce opposite phase plate signals." Thanks. OK its sinking in down through the sand.

Another way to look at it is using the fact that a tube is controlled by the difference between grid and cathode voltage. This holds for DC as well as signal (AC) voltages.
So the plate signal is the amplified (and inverted) difference between grid and cathode input signals.
Means that in order to achieve NFB, the feedback signal fed to the cathode must be in phase with the grid input signal.