Help understanding tube preamp output levels

[pdf64]

I think your schematic is fine. The reverb level control shouldn't affect the dry signal level - I assume that your reverb board only puts out the effected signal, that there's no dry signal mixed in with it? I don't understand what mechanism could cause a resistive potential divider to affect the tone, providing that reasonable precautions are taken to prevent excessive capacitive coupling between the input and output, and the output and ground.
Most any power amplifier is not going to require any appreciable current input, except under heavy overload conditions maybe, and then only to drive (input protection) diode clippers; they generally operate as voltage amplifiers, with high current output capability. Peter.

[armstrom]

Thanks for the feedback. The reverb level won't affect the dry signal directly, but in the schematic I posted the summing amp basically cuts each signal in half before summing them. If the signal voltage after the input buffer were 1V then that 1V signal would be split and one leg fed directly into the final summing amp (dry signal) and the second leg would pass through the reverb circuit and then into the summing amp.
So, if the summing amp is doing the following:
(Vdry * 0.5) + (Vrvb * 0.5) = Vout then
Vout = 0.5V + 0.5V = 1V
(assuming reverb level is all the way up).

If I reduce the reverb level to 0 (short to ground). Then my output voltage would only be .5V (half the dry signal).
Vout = 0.5V + 0V = 0.5V

I can't think of a good way to make a summing amp work. If I try to keep the dry signal at full strength regardless of the reverb level then I could end up with twice the output signal. I've attached a new schematic... this is the direction I'm leaning right now. (only changes are right at the last opamp)

[pdf64]

I think your previous schematic is better as the signals are getting mixed in a controlled manner - op amp outputs have effectively zero source resistance, so that 2k2 will be getting grounded via the coupling cap, creating a frequency dependant mess. If you're worried about the dry signal level cut then change to 2 x 20k to 2 x 10k in the 1st one. Don't worry about the reverb signal being the same magnitude as the dry level as it would sound ridiculously cavernous, no-one would turn it up that high. If they did, just turn the master down to restore unity gain. Peter.

[Steve Conner]

My dynamic range issues were due to integrating the board into a non-master volume amp. The reverb return mixer connects straight to the power amp with no master volume in between.

This means that the output of the reverb unit has to be given enough gain to allow decent reverb, without clipping the input, when the amp is roaring at a gig. Now at practice volume (since the volume control is before the reverb send) the reverb board is operating with a minuscule input signal that only toggles about three bits, and the result is a trashy digital sound.

A master volume, or dual gang dwell pot with one gang before the reverb board and the other after, would have helped. But I found that a tube-driven spring reverb pan could cope with the dynamic range fine, using just a dwell control on the recovery amp, and sound great at any volume. Springs have no least significant bits, and the tube driver clips gracefully to extend the dynamic range at the other end.

I used a 12AX7 anode follower as the return mixer. Your circuit is wrong, you need to configure the last op-amp as the silicon version of the anode follower, called a summing amp (google it)

And yes, I suppose I can offer advice by e-mail as long as you promise to tell me how it sounds when it's done. steve at scopeboy dot com