Best way to combine two dissimilar channels into PI?

[dehughes]

Hey...just thought of something...what if I ran the 6SN7 in a cascode arrangement? Would that solve my phase/mixing issues?

[Andy Le Blanc]

I noticed that there's no bypass cap on the cathode of the 6sn7 second stage,
That in my mind might act as a source of feedback even when the channel
isn't being used.

[Wayne]

Sure would! Not sure what it would sound like (that is, until you try it and let us all know ), but it would be in reverse phase from the other channel.

W

[dehughes]

I noticed that there's no bypass cap on the cathode of the 6sn7 second stage,
That in my mind might act as a source of feedback even when the channel
isn't being used.

Ah! Didn't think of that. Thanks! I'll try that before I dive off into cascode 6SN7 land as a last resort....

[dehughes]

Sure would! Not sure what it would sound like (that is, until you try it and let us all know ), but it would be in reverse phase from the other channel.

W

COOL! I'll try a cap across the 2nd triode of the 6SN7 first and see what that does, and if it doesn't magically make all my problems disappear, then I'll try a cascode arrangement.

Any suggestions as to circuit values for a cascoded 6SN7? I've never tried that with any tube, let alone an octal which I'm relatively unfamiliar with...

[Andy Le Blanc]

Now... by cascode, you intend to use both halves of a 6sn7 or 6sl7 as one
gain stage?
You can wire for two parallel stages, diode bias, cascode, srpp, mu-follower.

I'd go with the parallel stages or diode bias, srpp makes a good bass pre,
but cascode, srpp, and the like, Ive played with and they might not be as
guitar friendly as you hope. I've found diode bias to be a nice surprise,
and would love nothing more than to get another set of ears to listen to it.

[dehughes]

Now... by cascode, you intend to use both halves of a 6sn7 or 6sl7 as one
gain stage?
You can wire for two parallel stages, diode bias, cascode, srpp, mu-follower.

I'd go with the parallel stages or diode bias, srpp makes a good bass pre,
but cascode, srpp, and the like, Ive played with and they might not be as
guitar friendly as you hope. I've found diode bias to be a nice surprise,
and would love nothing more than to get another set of ears to listen to it.

ANDY! Thanks man! FANTASTIC IDEAS! Yes, I intend to use a 6SN7, as for some reason I'm quite enamored with this valve (maybe it's more like infatuation...), and was thinking of using both stages as one stage in a cascode setup (one "on top" of the other). HOWEVER, you're suggestion to run them in parallel is frankly brilliant in its simplicity....

That said, what do you recommend for someone looking for a nice paring (be it through similarity or dissimilarity) with a 6G3 Brown Deluxe channel into a set of cathode biased 6V6s? I'm all ears...even about the diode bias thingy... I am your Padawan...

[Andy Le Blanc]

I was in a traynor recently to do mods, the verb recovery was is used in exactly
the manner your attempting, mixing two dissimilar sources. Its nearly the
same as a fender of coarse. But instead of placing it before an extra gain stage,
they placed it just before the inverter, and used the extra free stage as tremolo
oscillator. CH1 can be your 12ax7 channel and CH2 is the verb or any other channel.
The whole point really is to mix just before the inverter with a reverb recovery
circuit you already know.

[Andy Le Blanc]

Another way to look at it....

[Andy Le Blanc]

Here's the simplest gain stages to try.....

SRPP is finicky, it wants a fixed load, but its simple and easy and appealing to
noodle with, it tends to be overtone-ish with a guitar.

Ive liked the diode form, again its very easy, start with a relatively small
plate resistor, 22k, and work your up to find the better load for your ear
and project reality.

[dehughes]

VERY interesting. A few questions regarding the mixer idea:

1) I wager the 1M pots are the volume controls for the channels?

2) Could they be substituted for resistors (being as I don't have chassis space for more pots)?

3) Are the .02uf caps coming from the plates themselves?

4) Is there an appreciable benefit to either schematic, given my goal?


As for the various gain stages:

1) Which in your mind would be the better way to go given a 6SN7 used for guitar purposes?

2) What type of output/gain does your bias idea kick out? That's intriguing...

As well....would it even be necessary to alter the way I'm running the tube if the "Traynor" mixing idea works? That is, I like how both channels sound (by themselves...not linked....), so I'm only tinkering with stuff in order to get them to not mess with each other in the same amp using the same PI.

THANKS MAN!

[Andy Le Blanc]

1) I wager the 1M pots are the volume controls for the channels?

2) Could they be substituted for resistors (being as I don't have chassis space for more pots)?

Yes but it s just a example keep it as simple as you have too.

3) Are the .02uf caps coming from the plates themselves?

Yes.

4) Is there an appreciable benefit to either schematic, given my goal?

In the fender recovery circuit the tube sees a fixed load, the traynor mixer
will have more tone variation depending on vol.

As for the various gain stages:

They all work, and I haven't posted the cascode sheme yet, they are of course just suggestions,
I like a 6sn7 very well too, it would be nice to
try that mixer and have it be the fix. I did premise the post with the assumption
that the 1M mix pot. would also be ch2 vol., you can put a trim pot. inside
the amp to set the mix without making more holes.

The gain and bias of the diode is really dependent on the load resistor and
plate voltage, 12au7 with a 39k load made around 1.7v bias at around 270v supply.
Two of these made approximately the same gain as a non-verb channel in a fender w/ 12ax7,
which surprised the hell out of me, and is why Id like to see someone else try it.

[dehughes]

1) I wager the 1M pots are the volume controls for the channels?

2) Could they be substituted for resistors (being as I don't have chassis space for more pots)?

Yes but it s just a example keep it as simple as you have too.

3) Are the .02uf caps coming from the plates themselves?

Yes.

4) Is there an appreciable benefit to either schematic, given my goal?

In the fender recovery circuit the tube sees a fixed load, the traynor mixer
will have more tone variation depending on vol.

As for the various gain stages:

They all work, and I haven't posted the cascode sheme yet, they are of course just suggestions,
I like a 6sn7 very well too, it would be nice to
try that mixer and have it be the fix. I did premise the post with the assumption
that the 1M mix pot. would also be ch2 vol., you can put a trim pot. inside
the amp to set the mix without making more holes.

The gain and bias of the diode is really dependent on the load resistor and
plate voltage, 12au7 with a 39k load made around 1.7v bias at around 270v supply.
Two of these made approximately the same gain as a non-verb channel in a fender w/ 12ax7,
which surprised the hell out of me, and is why Id like to see someone else try it.

Okay, I'm getting it. Slowly.

1) So, the "Fender/Traynor" reverb recovery idea would be a way to mix them without as much tonal influence as is found in my original method to mix the channels, yes? I suppose I don't see how this method would be appreciably different than doing what I did. I suppose what I'm saying is that I believe you when you say it's different....I just don't get how it's different. However, this Traynor idea still does nothing to change the phase of the channels, so it's more a method to mitigate tonal vibe than it is a "fix" for the problem of phase and plate loading, yes? Check out the attached schematic...am I close? In this Traynor schematic, the signal from the Reverb channel would be out of phase with the signal from the non-reverb part of the circuit...thus allowing mixing on the same side of the PI without issue, yes? As well, wouldn't a 3M3 resistor really kill a lot of gain from that channel? Could I use a smaller value resistor setup (say 220k/2k2) to mix instead? Lemme know if I'm even close...

2) However, if I inverted the phase on one channel (say, ran the 6SN7 in cascode or paralleled triodes, etc.), then I could mix the channels more simply...even use my original method. Right? In fact, if I got them in opposite phase from each other, I could then put them into opposite sides of the PI quite easily, yes?

3) Also, if the 1M pot in the "Fender/Traynor" mix schematic is to be the volume on one of the channels, how would I be able to retain distinct signal paths, being as each channel has a volume pot in between two triodes?

4) As well, you mentioned that you dig 6SN7 tubes, but that diode bias or paralleled triodes would be your recommendation for guitar amp use (especially without a tone control on the channel). That said, would the diode bias have more or less gain than paralleled triodes?

Thanks so much, man. I appreciate the time and mental effort you're putting forth... Lemme know what you think...

[Andy Le Blanc]

Its just another way to use an old solution for your design needs. You have
two very different channels. Yes you are gonna loose gain, but you can use
it to your advantage, and as you mentioned, you can choose a different ratio.
You have a choice. If you mix into one side of the PI you can take better advantage
of global feedback, or you use the PI to "electronically" mix thru the amp.

I have a distant memory of an old army radar manual that defined different types
of electronic mixing. There's resistive, which your trying, there's also capacitive
and electronic mixing. We're using resistors after the coupling caps of the
stages, you can also use resistors before one common cap, thats capacitive
mixing. Remember the two channel audio mixer in the back of the receiving
tube manual. You can also do it with a tube, say you had two stages sharing
one common resistor as a cathode follower. You had mentioned the idea of
going to a single stage, every cap in the path adds phase lag, the phase
will never be perfectly out of phase, the verb recovery is ment to deal with
unpredictable phase delay from a verb tank , so it seems a usable strategy.
You could have the second stage of both channels sharing a common cathode resistor,
and use resistors before a common coupling cap to balance levels.
But you have dissimilar channels, The gains of the 12ax7
will swamp the 6sn7 so the losses of the 3.3m may well be advantageous.
You can use the verb recovery as the volume control for a single gain stage,
or stick a trim pot inside to better balance. There will also be balance issues
using both sides of the PI, you may well have to incorporate a resistive
network to the same end before the pi with the 12ax7 channel any way.

You should explore the variety on the 6sn7 channel. Thats the fun part.
If you have a 6sl7 handy it would be good, and if the gain difference between
channels is impractical, you have the opportunity to use a 6sj7 with gains
of up to 300, to balance against the 12ax7 channel.

[Andy Le Blanc]

Another thought..... you said 6g3.....

Your going thru an awful lot for two dissimilar channels.
Why not stick closer to the 6g3 scheme.... have half the 12ax7 as one pre
use the 6sn7 in various forms as the other...... and the other half of the 12ax7
as a common stage before the PI..... It will split the difference as for as gain
differences go, and you will still have a 6sn7 pre, the volumes become the
mixer and you will very much simplify the signal path.