Mixer instead of input selector

[oon_the_kid]

I was thinking to myself. Is it possible or advisable to use a mixer for input stage instead of selector. If you look at the logic of it. Why bother having input selector when a mixer automatically selects for you.


Oon

 

[as_audio]

I agree ..
Do this since 30 years now.
You may want to short unused inputs.

 

[oon_the_kid]

Good point on the unused input. I was thinking of this design using a common base bipolar transistor amplifier circuit.

Audio Signal Mixing

Any thoughts on how to improve on this circuitry, use of mosfet/jfet instead of bipolar?

Oon

 

[as_audio]

Yes, a summing junction is what you need for the inputs.

The low input impedance should be done without feedback
preferably, because noise is increased otherwise.

So do not as in Fig. 4 of the link above, but like in Fig. 5.
Bipolar is in advantage here due to "natural" low impedance
(gm related).

 

[indianajo]

When my mixer had 4558's on the mag phono input, it was too hissy. But now I have 33078's it is quieter than the pilot light on the stove. I turn down the inputs that have no source, makes it quiet enough. I have a tube PAS2 but walking around the organ to switch the input after I changed from LP to CD or radio, was a nuisance. Also the PAS2 draws 100+ watts 16 hours a day.
Op amp circuit is in this post: Improving a "Disco mixer" to mid-fi performance - diyAudio
There is really no reason to connect all those individual parts IMHO with great op amps these days. If you buy a junk mixer like the Herald RA-88a or its clone the Olson, you get 5 great slide pots and a steel case to keep the radio of the CBer driving by out. I paid $15 for mine. The circuit is okay, but the packaging engineer was an idiot. Corrected as above. The PAS2 has a box inside the box for the high gain stuff keeping the hum away. The RA-88a needed one, but a wall transformer will do. If you build this find a CPO .0033 uf cap for the RIAA.

 

[tmuikku]

I've got long unbalanced run from TV to a mixer, which has some noise if the TV is not powered on = have to mix the TV channel down when listening to music. Until I get the long TV line some transformers someday. No problem for me but the wife sometimes calls me to the work "what the heck, wheres the TV sound, do I have to be a freaking DJ nowadays". Anyway, not much downsides using a mixer unless you've got golden ears.

 

[as_audio]

I think the intention here was to avoid the use of input selector switches
in a preamp, these being among the parts with high failure rate.

 

[PRR]

For your application, I don't see ANY point in the fancy CB mix-amp.

A simple passive bunch-of-resistors and maybe a modest make-up amp will do.

Any such mixer has mix-loss essentially equal to the number of (connected) inputs. 2 or 3 inputs makes loss of 1/2 or 1/3. Most hi-fi rigs have enough reserve gain to cover that; just turn it up.

Beyond 3 or 4 inputs you probably do want make-up gain.

The "advantage" of zero-impedance mixing is that one input will not leak to another, and you can disconnect inputs without change of loss. But you do not care so much if phono leaks into tuner; it's not like Studio or Broadcast where strict separation of multiple active sources is needed. And the whole point is to NOT be disconnecting things while using the system.

For small numbers of sources a 5532 opamp is a wonderful thing. Is used in a lot of small studio and broadcast mixers. Every record you own has been through a bucketload of 5532s, one more won't be noticed.

If you must go active, the mix resistors must be AT the mix amp. Long wire on the inverting input invites instability. That would be the reason I would favor passive mixing and a simple non-inverting gain stage. (Also avoids the "need" to invert the mix-amp's inversion, if you truly think your system is Absolute Phase.)

The common-base plan has been tried over and over. One company based all their products on it; most others quietly go back to standard base-input amplifiers. There is a lot of freedom in amplifier design and you can usually make a CE amp as "good" as a CB amp without the minor troubles of large input current. Rod likes to "play" and it is good for the mind, I don't care if you "play" too, but it is not necessary even for much more elaborate mixing than you propose.

 

[oon_the_kid]

Yes, a summing junction is what you need for the inputs.

The low input impedance should be done without feedback
preferably, because noise is increased otherwise.

So do not as in Fig. 4 of the link above, but like in Fig. 5.
Bipolar is in advantage here due to "natural" low impedance
(gm related).

Hi,

Thanks for the comment. I forgot to mention that I was looking at Fig 5.

Any comments on how to improve on it? I think adding a resistor (10-100K) at the input might reduce pickup of noise. I probably put a buffer stage at the output but beyond that I am open to any suggestions on how to improve on it.

Oon

 

[as_audio]

.. adding a resistor (10-100K) at the input might reduce pickup of noise.

This resistor to ground ? No, it is not useful and not necessary . No unusual noise pick up here.

 

[mt490]

In practice, you want your inputs switched or gain adjustable as you will soon grow tired when a device decides you need to be notified at earblasting volume and have to mute from a thousand different places. Just mixing it all together seems simple but the user experience is disgusting.

 

[DonKu]

This mixer works for me in a lab environment.

Altoids audio mixer by Don Kuenz

Greetings, this is my first post.