Mixer for preamps

[swisher]

Hi!
I'm building my whole guitar rig. At this time I'm working on GCX Audio Switcher and Mesa High Gain Amp Switch in one enclosure. I'm looking for solution for signal mixer. Here is the part of chain:



Splitter is build (schematic) with TM018 transformers and TL072s. But I'm looking for perfect solution for mix part. I found GGG mixer but I'm not sure is it good choice for mixer between preamps and poweramp. I don't need any volume pots just straight in-out signal without any noise and constant volume.
Anyone have a correct schematic for something like this? I didn't find any schematic of Mesa switcher, it will be helpful.

[Phoenix]

Are AMP1/2/3/4 preamps? Or full amplifiers? If amplifiers, are they tube, or solid state?

[garcho]

Just use an op amp in summing amplifier mode, easiest to build the inverting type. 180º out of phase but that matters more for sound reinforcement than signal processing, I wouldn't worry about it.

Here's the idea:


Throw some caps in there and be done.
You're not really worried about noise are you?

[swisher]

I didn't receive notification
Yes, Amps 1-4 are preamps for sure (tube preamps, my versions of TriAxis).
I saw schematic like this (with caps) but I wanted to ask about something more 'pro'. And I need clear signal, without any boost and noise.

[PRR]

> something more 'pro'.

The plan garcho posted is THE way professional mixing consoles mix.

If your preamps have any real gain, mix hiss is negligible. (For a 4-input mixer, if preamp gain is much-more than 4, say 10, mix-hiss is much less than preamp hiss. Gain of 10 is "20dB". We normally run mike preamps at gains above 30dB.

[ElectricDruid]

+1 agree with PRR. It doesn't get more pro than a proper virtual-earth op-amp mixer. Pretty much every analog desk since 1970 has been built that way.

T.

[swisher]

Thanks for all answers.
What do you think about something like this?




1. Could phase switches be like this or them should be before transformer or it doesn't matter?

2. What do you think about mixing part? Maybe better option is use op-amp on separate inputs and then second op-amp to mix them?

3. What about mute relay? Do you think it will make some shots during switching?

[PRR]

Mute will have less stray hiss if _not_ shorting to ground. (This is maybe the real advantage of active mixing: you can open inputs without changing gain of other inputs, and the hiss goes _down_ the more you open.)

[swisher]

You talked about this, right?

[swisher]

I read that TM018 are not the best solution for buffer. Which transformers will be better for this project?

[Ben N]

Why do you need isolation transformers at all?

[swisher]

As I showed in first post - I'm gonna connect "guitar signal" to 4 different preamps.

[Ben N]

And again I ask--why do you need transformers?

[swisher]

ok. I suggested Mesa "High Gain Amp Switcher" which has one transformer per output. Isn't it more "safe" and noisless than classic op-amp splitter?

[ElectricDruid]

ok. I suggested Mesa "High Gain Amp Switcher" which has one transformer per output. Isn't it more "safe" and noisless than classic op-amp splitter?

It provides isolation for the earths, which might help prevent ground loops, if that was an issue. It also makes the phase switching simple. I can't see any other reason to include the transformers.

Doing the phase inversion switch with op-amps is probably going to cost you another op-amp per channel, or mess with your input impedance, so perhaps in this situation the transformers aren't a bad solution. Although I hate stuff like that generally, and another TL074 is still a lot smaller and simpler than four transformers.

[swisher]

So do you think something like this will be enough?



And for phase switching I have to use another op-amp similar like this (with switch of course)?

[GibsonGM]

So do you think something like this will be enough?



And for phase switching I have to use another op-amp similar like this (with switch of course)?

Yes

[Kipper4]

Maybe something like this even.

[PRR]

> So do you think something like this will be enough?

TOO much.

You do not need separate opamps for every output.

However you may need a flip-switch for each output. Which comes back to an opamp. But not your over-wrought dual opamp plan.

As usual, R.G. has the hard dope. http://www.geofex.com/article_folders/polarity_reverser/polarity_reverser.htm

Stealing Kipper's input buffer to mask the low Z of the polarity reverser (and trimming redundant bias), we get:

R1-R3 may conveniently be 10k or 22k. (R1 should equal R2 better than 10%; R3 is not fussy.)

EDIT: better plan here.

[ElectricDruid]

*Now* we're talkin'! That's much more like it, PRR!