Channel Mixing Resistors

[TIMBO]

Hi guys, Just investigating this part of a two channel amp.
I know it's common to channel jump and mix the two flavours together, seems to work well.

What I have in mind is to run two totally different channels from one input and mix them before the phase inverter.
I have a vintage amp that mixes the two channels through each sides of a 12ax7 and is mixed via the anodes.

Extra valve for nothing......

[st]

Not sure what your question is... But yes, you can use mixing resistors for that. Just make sure your 2 channels are in phase.

[tubeswell]

You can use mixing resistors or you can use opposite grids of a differential amplifier (e.g. LTP), or if you have an extra dual triode, you can use a common anode mixer, or if you have one extra triode, you can use it as a virtual earth mixer to mix channels.

[TIMBO]

You can use mixing resistors or you can use opposite grids of a differential amplifier (e.g. LTP), or if you have an extra dual triode, you can use a common anode mixer, or if you have one extra triode, you can use it as a virtual earth mixer to mix channels.

Thanks mate.
Opposite grids - PI will have NFB/presence
Common anode mixer.

Hmmmm, was thinking of using this.
Virtual earth mixer.
Don't know this one?????

[tubeswell]

Here you go. For higher input impedance, increase the 2 x 100k to something like 470k or 1M

[sluckey]

Hmmmm, was thinking of using this.

That's probably your best choice since you need some more gain in front of the PI.

[TIMBO]

Thanks guys.

[silverfox]

Thanks guys.

Not so fast. I've often wondered how two inconguerent signals are mixed to produce a cognizable and pleasing output. Anyone got an explanation? Two different signals superimposed on a resistor doesn't seem to work in my mind or is it the beauty of Carbon?. Somethings gotta give. Shooter??

silverfox.

[shooter]

I mix with resistors all the time, a couple 10k's work good for stereo to mono, 470k's are my tube audio happy R, but 220k gets you there in a pinch, NOW if you want the science behind it, you're smoking with the wrong hick 

[TIMBO]

Thanks guys.

Not so fast. I've often wondered how two inconguerent signals are mixed to produce a cognizable and pleasing output. Anyone got an explanation? Two different signals superimposed on a resistor doesn't seem to work in my mind or is it the beauty of Carbon?. Somethings gotta give. Shooter??

silverfox.

And me TOO!!

Clearly you can see in the background is a Soldano head sitting atop a Fender head, which turns out to be a Dual Showman Reverb.
After a long discussion with a learned mate about some of Gary's LIVE rig, he confirmed Gary did play Soldano's and many others.

The two amps were run in parallel to achieve the many dark and light shades by simply turning the volume control on his guitar.

SO,
Can a similar sound be achieved by paralleling the two preamps and blending (resistors or anode mixing)prior the phase inverter and maintain clarity?



TOO MUCH 

[tubeswell]

So how many tubes is this thing gonna have?

[TIMBO]

10 ........ 11 if a mixer valve is used

TOO much 

[pdf64]

Electrons just do what they gotta do - the signals combine to produce an output that, relatively speaking given the overall attenuation usually ~6dB, is a sum of the two.
Given the tube compliment, I'd tend towards something based on a 5F6a / JTM45; the current arrangement seems a waste of a triode.
Though the plate mixing works well as a mod for 2 channel, post tweed Fenders, to get the fx on both channels.

[Tony Bones]

Not so fast. I've often wondered how two inconguerent signals are mixed to produce a cognizable and pleasing output. Anyone got an explanation? Two different signals superimposed on a resistor doesn't seem to work in my mind or is it the beauty of Carbon?. Somethings gotta give. Shooter??

They'll mix without the resistors, but then each channel would see the other as a (possibly changing) load. The resistors provide some isolation letting each channel work as it would if the other channel wasn't there.

As an example of the wrong way to do it, look at the '68 Custom Deluxe Reverb where they just tied the plates together to get effects on both channels. In an ordinary Fender circuit V1B and V2B each see as a load its own 100k plate resistor in parallel with whatever it's driving. That's a 220k mixing resistor for the normal channel and the reverb for the trem channel. With the CDR, both channels not only see their own plate resistor and the reverb, but they also see the other channel's plate resistor and the resistance looking down into the plate of the other channel.

This works fine if both tubes are making the same music at the same time; it's just like parallel tubes. But if only one channel is operating, then the plate load that it sees is very different when the plates are connected together. That's why Fender originally fed them through a pair of 220k series resistors; to isolate the two plates.

[PRR]

> wondered how two inconguerent signals are mixed to produce a cognizable and pleasing output.

It's simple arithmetic. 2+2=4. (In many cases 2+2 will come out 2, but we can find make-up gain.) As pdf64 says.

For purely resistive sources not working hot, you can just tie them together.

Tubes are non-linear resistors. In Tony's example, the maximum output at low distortion is small, far less than the same tubes working apart. But that plan shows the nominal signal level is far under 1V. You can just tie plates together for signals that small, and the few pennies saved is a blessing.

When you try to tie two volume pot wipers together, if either is at zero, the output is zero. Mix-resistors larger than pot values makes this interaction very small.

There is also Active Mixing, which gives *exact* addition. This eliminates interaction in 16+input studio mixers which have different instruments in each channel. Or broadcast master control rooms with very different programs (bird-watch, football) in each channel. That's over-kill for a single-player guitar amp.

[tubeswell]

Not so fast. I've often wondered how two inconguerent signals are mixed to produce a cognizable and pleasing output. Anyone got an explanation? Two different signals superimposed on a resistor doesn't seem to work in my mind or is it the beauty of Carbon?. Somethings gotta give. ...

Yeah you will still get some crosstalk even with mixing resistors. Active mixing (virtual earth mixer/common anode mixer/differential amplifier) eliminates more of this, for higher cost, but its a guitar amp so... I can think of many other cooler uses for a spare triode/tube


(Edit: i.e. ^What PRR said^)