Why is there no mixer functionality that allows a track to be aux send/return-ed while preserving the track's pan setting?

Question

(This is ramble-y, please don't ream me in comments.)

For my own purposes, when using a mixer, it would be nice if I could send a left or right panned track out the aux send and then return it to the aux return with the returning signal being panned left or right identical to the original track. When I first started learning to use mixers I assumed aux sends and returns would behave like this.

I now understand the fact that aux sends don't necessarily send to aux returns and aux returns don't necessarily receive from aux sends, and therefore neither has any reason to hold any of the settings of any given track that might be going from or returning to them.

That said, the use case in the first paragraph doesn't seem to me so unreasonable that a mixer designer wouldn't have thought of it, and I don't imagine I'm the only person who either made the false assumption that sends and returns would work like that or would find it convenient if they could work like that.

Given these two presumptions (which I realize could both be false!), I would think that at least some mixer makers would have implemented this functionality. In my experience with mixers though (which I acknowledge is far from exhaustive and therefore potentially misinformed!), there is no mixer maker that does.

Would this be an unreasonable or insufficiently practical feature? If so, what are alternate use cases that make it such? If not, is there some concrete reason why it is never seen?

Answer 1

I think you've misunderstood what an FX send is for. Sends are for bus effects, i.e. effects shared by multiple different tracks. Of course you can use it with just a single track, but in general what comes in on the return will be a mix of signals that originated from multiple tracks. The only way that a send loop can preserve stereo-field information is if the entire thing is in stereo. Now, that is actually perfectly possible, indeed standard in DAWs. But even if, it's kind of missing the point if the return has the same stereo position as the sending track, because the whole point of having an effect in the send loop is that it should merge together the signals.

This is most evident in reverb, the by far most common send effect: reverb should facilitate that the different instruments sound like they're actually in the same room. It is in principle possible to apply mono reverb to each channel individually and then pan them together, but this sounds completely different – the reverbs won't sound roomy anymore but will instead be perceived as each instrument having a “wetter”, but still isolate sound.

If you want to apply an effect to a single channel and have it “follow the position” of that channel, then you should insert the effect into the channel itself, not use any send. In that case, the effect is applied before the pan control and is thus, indeed, panned to exactly the same position as the dry component on this channel.

Answer 2

We're straying beyond the original remit of the question, so to cover that first.

It is physically impossible to send a mono effect to the same position as the original panned mono instrument. There is no correlation between the two. As the return of your effect pedal is mono, then you can place that in the stereo field manually, but all sounds sent through it must, of necessity, be panned to the same location.

You get around this, especially on a modern DAW or large studio desk, by having true stereo sends to true stereo effects, which can then return signal in the same part of the stereo field as the send.
This isn't always the case. Many effects are 'fake' stereo - mono in, stereo out. Others are true stereo right through.

I disagree with some of the other answers here in as much as if I send several instruments to a given room reverb I want the returns to simulate each of those instruments being in their own place in the room & not just generically splashing the whole lot with generic room sound, placed centrally. This way, each sound retains a sense of placement in the final mix, rather than what I would consider a very 80s way of the reverb being a complete stereo wash that ignores that placement.

Answer 3

Would this be an unreasonable or insufficiently practical feature?

I don't think so. I would think of it as a stereo, post-pan aux send. The pan control is effectively where the mono channel gets 'turned into' stereo - a certain amount sent to L, and a certain amount to R. If you had a stereo aux send after this point, you could effectively send the panned signal to the effect.

This could be useful too, musically, in some cases. If the effect you are sending to could deal with the panned signal in a stereo way - e.g. a delay effect that maintained stereo separation (i.e. that was effectively two mono delays with the same settings), it could allow each instrument and its delays to appear in the same place in the stereo image. The same would be true of other linear-response effects, but it wouldn't be true of e.g. distortion based effects, because the way distortion works means that the channels would interfere with each other.

If not, is there some concrete reason why it is never seen?

It costs extra, and there are a limited number of cases where it is useful. As leftroundabout says, reverb is the most common send effect, and maintaining complete separation between L and R doesn't make for a convincing reverb effect - you want to hear the sound bouncing around the soundstage. Often, stereo reverbs sum to mono before the actual reverb processing - they will be stereo in that the dry signal separation will be maintained, and in the characteristics of the wet signal, but not in any way that makes a panned-left input signal give a different wet output than a panned-right one. You could just about imagine modern technology being able to create a reverb that did something sensible with the pan position, but it would have to be very clever and effectively 'unmix' the sent signals before processing (remember that sends are usually used to share the same effect for a number of channels).

People who really want to achieve what you can achieve with a panned send will find another way to do it in the studio - e.g. use multiple insert effects, or the same insert effect applied in turn to different tracks in the recording process.

But your idea isn't a crazy one at all, and it would actually make a lot of sense with certain types of effects.

Answer 4

There are. They are called "inserts" and are per track since you could not otherwise preserve the pans of different channels independently. If you have effects where you are ok with FX-ing a stereo mix, you can use them with two FX channels: you have to establish your pan manually, but then pan controls are not intended for changing all the time like faders may be, and you often have some post-fader (or switchable post-fader) FX buses where the channel volume is being tracked.

Or you just add the effect on the main out: if you have a power mixer, it tends to have separate inserts for that so that you can add, say, extensive equalisation. Without a power mixer, you just add your whole-mix effects at the end anyway.

Answer 5

I'll assume that this comment of yours is the actual question:

To give a concrete example, I would like to pan a track 50% left, send it out to a guitar pedal (mono), return it and have the wet signal be panned 50% left.

You can do this with an insert effect. (Let's not care about the effect's input being designed guitars now.) Does your mixer have channel inserts? That would be mono-in, mono-out though. For real stereo, you'll need a stereo-in, stereo-out, true stereo effect device.

With an effect used in an aux send configuration having the return panned differently for every channel becomes complicated and its usefulness is not clear to me. If you send channel A and channel B to the same mono-input effect, A and B sends are mixed together as A+B and then sent to the effect input, and the effect's output (wet signal) will have both A and B baked in the same signal. If you pan channel A hard left and channel B hard right, that effect return will still be from A+B. If you pan the effect return left for channel A, and right for channel B, then it will be just effect(A+B) on the left and exactly the same effect(A+B) on the right as well. The same thing on the left and on the right! Why would you want that?

If you absolutely want, you could split the effect return to a number separate channels and pan/balance each of them differently. But I still fail to understand what sense it would make to have the exact same thing going to a bunch of channels that are just panned differently.

Edit. Maybe it could make sense to have e.g. reverb appear in the same spot in the stereo field. But trying to simulate that with a mono-input reverb feels very difficult, and it would require lots of components. And even with a complicated multiple-return-channels contraption, the reverb will still have all channels baked in the same signal. What IS often needed however, is having the reverb send post-fader, so that the amount of reverb follows the amount of dry signal.