[Duran]

Hi. I have a few questions regarding combining audio sources. I'm hoping this is a good place to ask.


I first assumed it would be as simple as wiring two outputs to one input (in fact, I've done this before... so I know it can work fine in some situations). However more recently, a conversation with a friend and some research has revealed to me that it's not a good idea because combining audio signals can be unpredictable, possibly sound poor, and be potentially damaging to the source circuits.

The first thing I want to know is: why exactly?

I thought two waveforms would be combined in the sense that the amplitude of each signal would just be added together at any given point along the waveform. How I understand the problem now is that if one of the outputs is a good amount higher voltage than the other, it can "seep backward" into -and overload- the other output's circuits. (I'd like to better understand how/why this occurs)


The other main thing I want to know is: how do I solve this?

Some examples:

Have my computer, TV, and a CD player (or iPod) all connect to one set of speakers and sub. (not necessarily all sending a signal at once, but perhaps two simultaneously (and maybe all 3 if my computer decides to make a system noise))

Or

Combine the audio from my XBox 360 mic with an audio source (computer/ipod/ect.), and have them both go to the mic input on the controller. Thus playing background music or whatever else over XBox Live chat.

I'd like to make a simple mixer circuit such as one of these.

Though reading that page rose more questions, such as:

Should I know the load impedances of my sources as well as the device accepting the new signal, and how how they compare to the resistor values in the schematic (so to find what optimal values they should have, and not introduce noise or lose some of the signal)?
Does attenuation simply mean the same thing as signal loss (I'm guessing beginning at low amplitudes of the waveform)?
Should I definitely go for the active stages (circuit 2 and higher), or will a passive mixer work just fine (circuit 1)? (I'm not completely sure what is meant by "Suffers with the problem of no buffering and possible missmatches and losses.")
What does "Op-Amp" refer to, and what's meant by "busses" and "sends"?
Should I bother with the 27p and 47p caps?

Help or answers to any of these questions would be very appreciated.

Thank you.

 

[Armaegis]

Just get yourself a cheap pre-amp. Problem solved.

 

[mmd8x28]

Hey, I think he wants to know the specifics, not just to "Get a cheap pre-amp".

Besides, I encourage people making their own stuff.

 

[eucariote]

Um, because you'll be running current from one source to all the others, with your speakers being just one of many current sinks. If you want to know exactly how it will distribute, check the ohmage across all those +/- output sources and refer to ohms law. I'll bet that they suck current even more than your speaker. That will at best divide your signal from the active source to all the others if those are turned off, and at worst, create weird summations that could damage your hot components for running current in places and directions for which they weren't designed.

 

[Armaegis]

Think of it as traffic flow. In a simple souce-to-speakers setup, traffic only flows one way. If you have multiple sources, it's basically like slapping down another road with no idea which way the flow is going and no signs or lights to direct traffic. If you're lucky it might just work out, but chances are something could go wrong. The more roads you put down, the worse it becomes.

You need some sort of preamp to direct traffic. A standard A/V receiver is probably the most common form of preamp. Most receivers also have an amp, so not only does it let you select which input you want, but it will also provide adequate power for your (passive) speakers.

 

[Duran]

So the most basic mixer shown here simply attempts to detur the signel from heading into the other sources by way of the resistors (and upping the impedance into that section of the circuit), and (whether the other sources are on or off, as long as the sources aren't the same strength) a small percentage of the strongest source (or will the weaker attempt it too?) will still attempt to go through to the other sources... but I guess the use of this circuit assumes that a diode or cap in the source circuits will stop this small amount. Right?

Quote:

Originally Posted by eucariote /img/forum/go_quote.gif check the ohmage across all those +/- output sources and refer to ohms law.

I assume you should only do this while the circuits are off... but what if you do it while they're on? (I'm quite sure this is a stupid question. )

 

[Duran]

I'm still looking for answers to the questions in my first post, however, maybe I can also get a quick answer to something else. In the schematics I linked to, there are unmarked transistors... what are good choices of transistors for audio circuits such as these?