Princeton Reverb Mod

[Fletcher Munson]

If I were to strip out the reverb from a blackface Princeton circuit (AA1164) at the points indicated, should I remove the 3.3M mixing resistor? Or would that increase the gain going to the third stage too much? Could I replace the mixing resistor with a lower value or even a pot to provide a gain control?

[ric]

If I were to strip out the reverb from a blackface Princeton circuit (AA1164) at the points indicated, should I remove the 3.3M mixing resistor? Or would that increase the gain going to the third stage too much? Could I replace the mixing resistor with a lower value or even a pot to provide a gain control?

[ATTACH=CONFIG]49017[/ATTACH]

If you....

[Fletcher Munson]

Let me try to re-phrase my question: If I were to strip out the reverb from this circuit, what would I need to do to retain the original tone and gain structure?

As I understand it, the Reverb pot serves as the grid reference to ground for the third gain stage. So I'd have to replace that with a grid leak resistor of some value (1M?).

But what about the 3.3M mixing resistor (plus 10pf cap)? Strip it out or leave it in?

[Jazz P Bass]

These amps are easy enough to work on.

Do the work, answer your questions.

[Leo_Gnardo]

If I were to strip out the reverb from this circuit, what would I need to do to retain the original tone and gain structure?

Easiest way: unplug wires to the reverb tank.

DONE!

If you feel you must do more, remove the reverb drive tube too. A shorting RCA plug to the reverb return jack wouldn't be a bad idea. Finally don't turn up the reverb knob, or move the wire on its wiper to the ground tab so you don't accidentally dial up some hum/buzz/noise. Then you can't argue the fact that you have the exact same circuit, only the reverb is missing.

Should you wish to restore reverb in the future, it won't be any brain strain at all, just a couple minutes work.

[eschertron]

Of course, Leo has the right answer. Kudos!

I just happened to be looking over this schematic today, comparing it to a current project. So I did the maths on that 3.3M resistor. With the trem pot all down, there's a 470k resistor that forms the voltage divider with the 3.3M. About 18dB of attenuation with the 3.3M/470k string. Replacing the 470k with the 1M? resistor in your drawing gains you about 6dB into the next stage. Whether you want that or not is between you and your ears. Or, between your ears.

[Steve A.]

If I were to strip out the reverb from a blackface Princeton circuit (AA1164) at the points indicated, should I remove the 3.3M mixing resistor? Or would that increase the gain going to the third stage too much? Could I replace the mixing resistor with a lower value or even a pot to provide a gain control?

[ATTACH=CONFIG]49017[/ATTACH]

If you replaced the 3M3/10pf reverb splitter in the Princeton Reverb with a jumper you would be on your way towards building one of Ken Fischer's pre-Trainwreck prototypes according to some reports. Back in 2001 I drew up the following "fake" schematic to illustrate the similarities between the PR and Trainwreck preamps if you replace the reverb splitter with a jumper.



Attached PDF file:
wrek_pr.pdf

.

Steve A.

P.S. And yes it would have gobs of gain, to use the technical term...

Here is the Princeton Reverb AA1164 schematic ...

.

[Chuck H]

All that^^^^^^^^^^^^^

Or you could just turn the reverb down and leave it there

[Leo_Gnardo]

If you replaced the 3M3/10pf reverb splitter in the Princeton Reverb with a jumper you would be on your way towards building one of Ken Fischer's pre-Trainwreck prototypes according to some reports. Back in 2001 I drew up the following "fake" schematic to illustrate the similarities between the PR and Trainwreck preamps if you replace the reverb splitter with a jumper.

Yep, that ought slap the third triode plenty hard. A nice variation would be a 12AY7 in the first tube socket. May have to give this one a go!

[Fletcher Munson]

Of course, Leo has the right answer. Kudos!

I just happened to be looking over this schematic today, comparing it to a current project. So I did the maths on that 3.3M resistor. With the trem pot all down, there's a 470k resistor that forms the voltage divider with the 3.3M. About 18dB of attenuation with the 3.3M/470k string. Replacing the 470k with the 1M? resistor in your drawing gains you about 6dB into the next stage. Whether you want that or not is between you and your ears. Or, between your ears.

What I think my ears want to hear is a pot that sweeps between "vintage Princeton Reverb" and "gobs of gain." Perhaps a simple volume/gain control before the third gain stage?



Starting with 160V for the voltage source, I get 19.9V output voltage from the 3.3M/470k voltage divider. Using a 3MA pot and a 430k fixed resistor, I get 20.0V output voltage. So that should be pretty close to "vintage." When you sweep the pot the other way, you should get "gobs of gain." Probably too much.

A huge thank you leo_g, eschertron, and steve a. Much appreciated.

[eschertron]

If you don't happen to have a 3M pot, use a 1M and divide the 470k by three-ish to get a similar effect. Or keep some fraction of the 3.3M on top of the pot, paralleled with a cap, so the fixed 'brightness' isn't totally lost. The sky's the limit

[Steve A.]

Yep, that ought slap the third triode plenty hard. A nice variation would be a 12AY7 in the first tube socket. May have to give this one a go!

I need to do a more thorough comparison of the two circuits to see how Ken Fischer kept the volume down to a reasonable level in adapting the basic PR architecture.
BTW I seem to remember that the current owner of that pre-Trainwreck prototype who lived in Hawaii made a few posts at AMPAGE about the amp.

Steve A.

[Steve A.]

If I were to strip out the reverb from a blackface Princeton circuit (AA1164) at the points indicated, should I remove the 3.3M mixing resistor? Or would that increase the gain going to the third stage too much? Could I replace the mixing resistor with a lower value or even a pot to provide a gain control?

[ATTACH=CONFIG]49017[/ATTACH]

Hmmm.... perhaps you should check out the schematic for the non-reverb 1964 Princeton amp.


http://www.thevintagesound.com/ffg/s...a964_schem.gif

Here is a cropped and scaled down version of the full schematic for in-line display here:



Here is the full schematic as an attachment (it does not seem to work. Download file from link above.)





Steve A.

[Chuck H]

Here you go

The green traces in the graph represents a stock PR second gain stage at the bottom trace (including the effect of the 10p cap) and a Trainwreck at the top. I used these as marker goals. The purple traces represent the circuit shown with the bottom being the 1M pot at full rotation one way and the top trace at full rotation the other.

An interesting thing happens by simply adding series resistance. You can change the frequency knee of the coupling cap due to circuit impedance. This actually allowed for closer approximation of both adjustments with less than .2dB loss at 82Hz in PR mode and a 16dB cut in TW mode much like the TW circuit.

On the same lines there are obviously other differences between your amp and a TW design. Like your third stage not being cold biased and un bypassed, the PI being a cathodyne rather than a LTP and the tone stack being very different. Not to mention plate voltages. But here you go anyway. This circuit has the advantage (?) of using a much more readily available 1M pot as well as representing the low TW AC loading in the higher gain mode while correcting to a higher AC load in PR mode.

Submitted for your consideration.

[Fletcher Munson]

Really impressive Chuck. Thanks for doing this! Are you using LTSpice or... ? I was just trying to digest your earlier version when I refreshed the screen and saw that you'd refined the circuit. The new version is nearly perfect in PR mode, and should be controllable (and pretty damn interesting) in TW mode.