470K attenuator resistors...ever messed with them?

[Matthews Guitars]

I just had an interesting modded amp across my bench for a while. It's a blackface Fender Bassman, 1965 by the serial number, and it had some simple mods done that made it a real screamer, but it could have been reverted to dead stock condition in 15 minutes if you wanted to.

The drawback is that it was SO hot that you sacrificed almost all of the clean headroom.

As I investigated it, to find out what the mods were and how they worked, I soon found a single resistor change that was responsible for most of its impressive level of drive.

The 470K attenuator resistor on the grid of the second preamp tube had been dropped to just 22K.

Marshalls share that feature.

There's a 470K attenuator between V1B and V1A on a 2203/2204 and other models that share the preamp/master volume circuit, and two resistors of the same value that serve the same function in a 1987/1959.

Have any of you played with reducing these resistor values for easy gain boosts?

What were your results and conclusions? I realize that if you go too low, blocking distortion is likely to occur.

Of course it's easy enough to do this mod. But I want to hear your opinions before I even bother to try it.

 

[neikeel]

On a 1987/1959 they are more a voltage divider, there should be similar arrangement on a 2203/4 plus the treblepeaker. It’s like swapping a 1meg pot for a 50k pot at mid point.

 

[johan.b]

The Groove Tubes Trio preamp crunch mode was basically a 2203 pre with different values at that point.. and if I remember correctly, forum member AlvisX, talked about messing with those values and the peaker cap...
J

 

[neikeel]

@johan.b iirc AlvisX likes removing the treble peaked cap.

 

[StingRay85]

Basically it decides if 50% or 100% of the signal is removed, depending on position of the preamp gain pot. It avoids overloading the next gain stage. But in theory you can experiment with these values no problem.

 

[johan.b]

.. I've been meaning to try it for a long time but never seem to get to it... but you could try reducing the 470k paralleled with the 470p cap. It should make the treble slope less steep...or more gentle, if you will,... I'd try a 100k and go from there... same with the 470k/470p before the preamp pot... perhaps someone else already tried it..?

 

[Max Gahne]

Take a look at an AB165 schematic which might be the amp you're working on - AB165 is a 1965 Bassman. The 470K resistor you're describing might be the one connecting the B voltage to the signal path before the coupling cap of the second preamp tube. Something I'm not familiar with. But it's not a typical 470K/470K attenuator like in a 2203/04.

 

[Pete Farrington]

Yes, a schematic / model ref for this amp would be great, I looked at a few but couldn't match any with the description in post #1.

 

[Max Gahne]

here's the schematic I'm referring to. Whether it's the right one or not I don't know but @Pete Farrington take a look at where the 2 channels merge at the 470K resistor going to the plate voltage. What does that do?

 

[johan.b]

If ab165 it could be the local feedback after the mixer. Yes reducing mixer values would change things..

 

[Max Gahne]

It does look like local NFB. If this is the right schematic and resistor @Matthews Guitars did they change it to 22K or 22M?

 

[Pete Farrington]

where the 2 channels merge at the 470K resistor going to the plate voltage. What does that do?

Yes, that stage is configured as an inverting mixer 'virtual earth 'NFB amplifier.
Gain = about 220k/470k so about 2.
Altering the 470k to 22k would reduce the gain to 0.1, ie a 1/10 attenuator.
Aiken outlines the theory https://www.aikenamps.com/index.php/designing-single-stage-inverting-feedback-amplifiers

Whatever, it's not something that's akin / applicable to any regular classic Marshall.

 

[neikeel]

Sorry, from the description I was thinking the mixer resistors on a Blackface 6G6-B ('64) which has a 470k/470k pairing before the grid of the PI.

Changing the either 470k after V2 (bass channel) or V1 Normal for a 22k would really give a very high signal boost for that channel