6G3+’Blackvibe deluxe’ layout

[6G3Victory]

Hey guys,

I’ve ripped apart my 2nd(of 4) build (6G3) to combine the 6G3’s bright channel with a blackface deluxe bright channel. No Trem, no ‘Verb.

Up until now I’ve built from kits. This is my first build drawing the layout myself. I think it seems ok… one thing that concerns me is the long lead that runs from the 10nF cap going into the PI, to the first grid of the PI. It’s kind of a weird set up, having the mixing resistors and 10nF cap stationed in between each preamp channel, but I’m not sure how else I could design it, or if it’s even an issue at all.

Do I need to worry about that long lead, or anything else for that matter? Any feedback would be very much appreciated. Much respect for the knowledge and experience that many of you possess.

[pdf64]

The negative feedback network here is essentially a potential divider. The 820R negative feedback series resistor option doesn't seem a good idea. Combined with a 5k shunt resistance, it allows far too much feedback, 0.86 of the speaker voltage will appear at the LTP 2nd grid, which will greatly reduce the closed loop gain, and may cause instability, possibly oscillation.
Fender use 100R or 47R shunt with a 820R series negative feedback resistors, sampled from a 2, 4 or 8 ohm speaker output, typically 47R with an 8 ohm speaker output, ie 0.054 of the speaker voltage.

[6G3Victory]

Thanks for your reply, PDF64. I just noticed I drew my 820r/56K NFB resistors twice by mistake. I’ve corrected my drawing.
Does an issue still remain?

[6G3Victory]

Ok, I can see there is more to this than just my drawing error. I’ve thrown a 5K pot in there without adjusting anything else. I’ll do some reading and see if I can figure this out.

[B Ingram]

… one thing that concerns me is the long lead that runs from the 10nF cap going into the PI, to the first grid of the PI. It’s kind of a weird set up, having the mixing resistors and 10nF cap stationed in between each preamp channel ...

6G3 Deluxe
AA763 Deluxe

Both of your donor-amps use 220kΩ resistors for channel-mixing.

Look closely at each amp's layout and you'll see Fender put those resistors very close to the Next-Stage-Grid that those channels feed. So if you see "everything before the resistors" as a plate wire, and "everything after the resistors" as a grid wire, the rule followed was "Keep Grids Wires Short, and Make Plate Wires long if Needed." In your amp, this means moving both 220kΩ resistors to be on the board near V3 Pin 2, so that the "grid wire is short." If you look at the AA763 Deluxe layout linked above, Fender made the wiring after coupling caps leading up to the 220kΩ resistors very long, and made the path from 220kΩ --> phase inverter input cap ---> phase inverter grid very short.

Separate/Related Issue:
Your AB763 channel adds a coupling cap between the 2nd Stage plate and the 220kΩ resistor.
Your 6G3 channel needs a coupling cap between the 100kΩ/15kΩ and the 220kΩ channel mix resistor.
It appears you hoped the 10nF phase inverter input cap would do double-duty; it will not. Use the blackface 0.001µF (1nF) for the phase inverter input, then give the 6G3 channel its own coupling cap of at least 10nF before its 220kΩ.

- The high impedance looking in to the long-tail means the 1nF input cap has a lower roll-off frequency that you might expect.
- The coupling cap out of each channel is effectively in-series with the phase inverter input cap for AC, which makes the resulting capacitance smaller.
- Make the channel-coupling-cap 10x bigger than the phase inverter input cap to avoid shrinking the effective value of the latter:
1nF in series with 1nF ---> 500pF effective
1nF in series with 10nF ---> 909pF effective
1nF in series with 100nF ---> 990pF effective

- Looking at the above, Fender's use of 100nF coupling cap into a 220kΩ mixing resistor (for that channel) into a 1nF phase inverter input cap oughta start making a lot of sense.

I’ve ripped apart my 2nd(of 4) build (6G3) to combine the 6G3’s bright channel with a blackface deluxe bright channel.

You have a tagboard in-hand, and the sentence above suggests you have a donor chassis in-hand whose board will be modified/replaced. Your present tagboard layout envisions neatly using each tag-position one-by-one, and seems to have been drawn from right-to-left (from Input Jacks to phase inverter).

I would encourage you to break up each functional set of parts/tags: "phase inverter parts" are there own drawing, "6G3 preamp parts" are their own drawing, "AB763 preamp parts" are their own drawing.
Now lay your drawing out on the graph paper roughly to-physical-scale of the tagboard & parts you're using. Cut these out so you can physically place them between the pots & sockets in your chassis.

Doing the above may help you identify the best physical orientation of parts to minimized grid-wire length, and to avoid the circuit "crossing over itself." It will also highlight where it will be advantageous to leave one or more tags unused, to get the parts closer to where they will be used & reduce wire-length connecting various bits together.

[6G3Victory]

Super helpful replies. Thanks, guys!

Looks to me like the 6G3 channel will need a 180K NFB resistor, and the black channel will require an 82K, to go with the 5K presence pot.

I’ll redo my layout to add the 100nF coupling caps, and to move my mixing resistors closer to the PI.