Reducing gain on Bride of Zen?

Increasing R108 (and R208, if that is the right-hand channel) would do it. You could also decrease R104/R204. Or some combination of both.

Increasing R108 would tend to decrease your bias current as well. If you want to maintain the 40 mA bias, you should bump up the gate voltage accordingly using P102.

Alternately you could also put pads on all of your inputs to a greater or lesser degree (whateve it might take to get all of them at roughly the same input level) and put your volume control at the output. You'd get the added benefit of having all the volumes be roughly equal when you switch from one to another.


First things first: I would put it before the cap.

I'm planning on rebuilding mine with fixed attenuation for each input. That way I won't have to fiddle with an extra knob, or add the extra expense of another variable attenuator.

On the plus side all of my inputs will have close to the same amplitude going into the amp so I won't have to worry about remembering to turn down the volume before selecting my CD player after listening to the tuner.

That's all personal preference, of course.

I suppose I could get by with a 3 position bat-toggle switch labeled "mute", "about right" and "loud". :D

The plus side to having two variable attenuators is that you can vary input level and the output level at will.

Hope that helps,

If I have the same problem, I would go for the fixed attenuator as in the figure. I could test R0 with different sizes. :yes: unless eLason or Mr Nelson Pass says :no:.


  • fix_att.jpg
    5.6 KB · Views: 554
No frowny-face from me. :)

Yochim - You could use little linear-taper pots for each input if you like. That would save you from having to unsolder/resolder resistors in a fixed attenuator if you decide later to change your source gear around. (Or just want a little more or less loudness.)

If audiophile nervosa sets in, rest easy that such pots have been used in at least one well-thought of solid state linestage that I've peeked under the hood. :) (Not sure if I should name names, so I won't.)

At any rate, you can measure the resistance from the adjustment pin to ground and divide it by the resistance of the pot as a whole to get the fraction of the amplitude you will see. (This is presuming that you hook the pot up as a simple voltage divider where the signal comes in "on top", the "bottom" is connected to ground, and you tap the signal from the wiper pin.)

You could also clip the ground lead of your multimeter to ground, and measure the voltage at the wiper pin until you get it just where you want it.



2013-01-03 3:28 am
NFB in the Bride of Zen?

Sorry to resurrect this old thread, but perhaps it is better than opening yet another one on the BOZ. This thread is somewhat related to my question on the BOZ, so here goes (apologies if this has already been addressed elsewhere).

In Nelson's paper for the BOZ, I quote:

The Zen preamp circuit has a much easier job in this regard, as the load attached to the Drain of the MOSFET will be resistive over the audio frequency range, allowing for intrinsically flat response. We will have much less reason to employ feedback in this circuit, and so we will not.
I am perplexed by the last part of this statement, as clearly the BOZ uses some current-series NFB due to the presence of the unbypassed R108 source degeneration resistor. Simulation indicates about 6dB of NFB.

Last edited:


2013-01-03 3:28 am
What do you mean by "overall feedback"? I suspect you mean "global" feedback.

Recall the BOZ uses a single gain stage...

Since it is possible to design a single gain stage with no local AC feedback, NFB is present regardless if it is locally or globally (multi-stage) derived.

Well, that's my 2 cents anyway. ;)