Bainzy Mod

Ever wished you could try all those gain stage mods of the 80’s, but hated the way they were always permanent or made the amp sound really different when switched off? After thinking this I went about designing an extra gain stage mod that utilises one half of an extra 12AX7/ECC83 tube, or both halves if you want more gain. The key aspect of the mod is that when switched off it is 100% out of the circuit; totally bypassed. The reason it’s called the “Bainzy Mod” is because that made it easy to remember which mod it was on forums, by naming it after my forum nickname.

How Does It Work?

The basic premise of the modification is that it’s 100% switchable, so first of all you get a DPDT on/on switch and mount it either on the back panel of the chassis, inside a spare input jack hole, as a push/pull switch on the back of a volume pot, whatever – it’s totally up to you. Then you need to replace the 1M pot that’s used for Channel II’s volume control with a dual 1M pot, again a log/audio taper one. You then get your extra tube, install it, either socket mount the components or mount them on a new piece of turret board material (recommended).

Now, when you flick the DPDT switch to engage the extra stage, the signal from Channel I is tapped from between the 470k mixer resistors and Channel I’s volume control, and is sent to the input of the extra gain stage, and goes through the extra triode and components until it goes back to the mixer resistors. Since we’re not using Channel II right now, we can use that dual 1M pot to act as a gain control for the brand new extra stage. Therefore we wire up one part of the pot as Channel II’s volume control, as usual, and the other part of the pot as the extra gain stage’s volume control. Although these turn up the volume of both channels at the same time, it doesn’t matter since you won’t be using both channels at once at any time. You can still jumper the channels with the mod disengaged, but you wouldn’t want to do it with the mod engaged – it doesn’t sound particularly good.

When the gain stage is off, the signal path of the stock circuit is completely restored, and the input grid of the extra stage is grounded.

The Diagram

Bainzy ModClick the thumbnail to enlarge the diagram. Please don’t reprint, post or copy the article or the diagram without seeking permission from this site, but linking to this web page is totally fine.
This diagram was designed to be an aftermarket modification to 1959 or 1987 type Marshall amps, so the mod is laid out in a way that’s not really the best for curbing extraneous noise. To make the mod quiet, you can do one of two things:

1) Rewire the mod so that Channel I’s first stage is wired to the new tube, and the extra gain stage is wired to the left half of V1. In other words, flipping the two stages round. Therefore the signal doesn’t dart back and forth across the preamp and is much less likely to pick up any noise.

2) Shield the signal carrying wires extensively for the extra gain stage with shielded cable. In the unlikely event that this isn’t enough, you can shield some of the wires for Channel I’s stock first stage too.
The Values

Now that you’ve got the diagram copied and the mod finished, here comes the fun part. You can tweak the values of the extra gain stage endlessly, if you want you can even wire it up so the amp becomes an almost exact copy of a 2203 or 2204 when the mod is engaged. The values I’ve given you to start off with (2.7k/.68uF cathode, 22nF coupling cap) should give you a good sound from the get-go, but it’s much more rewarding to find a value combination that you really like and feel it suits the rest of your amp best. Here’s some I’d recommend trying:

  • .0022uF coupling cap instead of the 0.022uF capacitor
  • 3.3k/100uF cathode instead of 2.7k/.68uF
  • 10k cathode resistor, no bypass cap (like a 2203/4)
  • 500pf capacitor over (in parallel with) the 470k resistor
  • 240pf capacitor over (in parallel with) the 470k resistor

In addition, you can replace the gain control for the extra stage with a fixed set of resistors, so it’s at a set level of gain boost. This also removes the problem of having to turn up Channel II’s volume control as in some amps this may cause a very slight change to the sound as the signal bleeds through into Channel I. The way I do this is remove the wires from the section of the dual 1M pot that the extra gain stage is using, and put a 470k resistor between the two signal wires. Where the dark blue signal wire touches the 470k resistor, I connect a 68k resistor, and send that to ground. You could also add a bright cap over the 470k resistor, for example 250pf or 500pf. If you wish, you can now go back to using a single 1M pot for Channel II’s volume control.
More Gain

If you’re after even more gain, you could try wiring up the extra tube in parallel, thus utilising both sides of the 12AX7/ECC83. Simply join pins 1 and 6 together, pins 2 and 7 together, and pins 3 and 8 together.

You could also repeat the ‘Bainzy mod’ circuitry, adding another DPDT switch and using a fixed resistor for a fixed gain amount, to wire up another stage using one side of the extra 12AX7/ECC83 for each added gain stage.
What Tubes?

You could use any 12AX7 (ECC83) theoretically on this mod, even a 12AU7 or 12AT7. I managed to get the best results by using the JJ/Tesla ECC83S, which actually sounded better to my ears than any of my NOS RFT, Mullard, Sylvania or Valvo tubes which was a nice surprise.

click here for a soundclip example of the mod in use, set with values for a compressed lead tone!

Copyright © Richard Baines 2006

Written by 

One thought on “Bainzy Mod

  1. Nice. I am trying to work out how to lay this out per your suggestion #1 “Rewire the mod so that Channel I’s first stage is wired to the new tube, and the extra gain stage is wired to the left half of V1. In other words, flipping the two stages round.”

    In doing so and looking at the signal path, I wonder why not tap the signal after the summing resistors? It seems that would have the same effect, but would allow the user to boost either channel and/or the jumpered signal from both channels.

    Thoughts? Thanks.

Comments are closed.