Send / Return Compression

Combine the best qualities of multiple compressors, and you can make your lead vocal the star of the show.

Typically, engineers use auxiliary sends for effects such as reverbs and delays. If an aux is sent to a dynamics processor, it’s usually as a way of setting up parallel compression. But it doesn’t have to be...

In this article, I will explore the send‑based compression technique renowned mixing engineer Michael Brauer uses for processing lead vocals in his famous ‘Brauerize’ setup. Matt Houghton briefly touched on this in his October 2022 article about multi‑stage compression, but here we’ll go into detail. Brauer himself credits fellow engineer David Kahne with developing and showing him this sophisticated send/return approach to compression. At first glance, it looks like your typical parallel compression setup, and in a way it is — but the key difference is: there’s no dry signal!

The idea is to send the lead vocal to multiple different compressors, using faders to balance their returns to craft the desired lead vocal tone. So, instead of using only one of your favorite vocal compressors, or two in series, you use them all, independently from each other. They are in parallel in the sense that they each process their own ‘mult’ of the vocal signal, but there’s no dry signal.

Float Like A Bus

The best way to set this up, in any DAW, is to set up a group or bus that ‘collects’ all lead vocals and ‘floats’ them out to the different compressors. This is why I like to call it ‘LV float’, but feel free to use whatever name you prefer. What’s important is that the output of this group or bus is not routed anywhere. We want to ensure it does not pass (unprocessed) audio directly to the stereo bus.

What’s important is that the output of this group or bus is not routed anywhere.

This diagram illustrates the basic signal flow involved in the author’s send/return compression setup.

Next, we need to create mixer channels for the different compressors. In his original hybrid setup, Brauer used five. Now, in the box, he uses as many as seven different processing paths. In my setup, which I’ll describe here, I’m using four different processing paths.

But first, let’s take a minute to understand the reasoning behind the concept. At its heart lies the fact that many compressors not only reduce dynamic range, but impart character to the sound in other ways. Some push the midrange, some put a sheen on top, some distort, and so on. Michael Brauer uses terms like “throaty” or “chesty” to describe what the compressors he uses sound like on vocals. I love this idea of attributing emotional and physical terms to gear and sound.

So, when selecting your weapons of choice, try to think along those lines. I’m sure you have some favourite vocal compressors. What is it that you like about them? How would you describe their sound? What do they contribute to the emotional impact of a vocal? The idea is to find a collection of compressors that sound distinctly different from each other, where each brings something to the table that the others don’t.

A good place to start, obviously, is with different compressor topologies. Sure, you could use two different 1176 clones from two different makers or a Rev A and a Rev D original. But the advantages are clearer with different types of compressors. In my studio, I settled on the following:

These four faders give me all I need to craft a great lead vocal tone in any genre or production. The 1176 brings midrange punch and tightness, and brings the voice right up front. The LA‑2A delivers a smooth and big tone that makes the singer shine. The Weight Tank cuts through with a lot of character and fatness, making the singer “spit with muscle”, while the Sebatron is set to really crunch with its air EQ engaged, delivering a very saturated, aggressive tone.

Happy Returns

Let’s look at the setup in detail. As well as the ‘LV float’ bus or group, you’ll need as many extra mixer channels as you want to use compressors. These might be aux, FX or group tracks, depending on the DAW you use. While making sure ‘LV float’ is set to no output, set up four unity gain sends from it to each of the four processing channels. Finally, route the outputs of all four processing channels to a single bus or group called ‘LV master’ or similar, to recombine your compressed sources on a single channel that can be automated, sent to effects and further processed.

Although it can work fine without it, this technique works best after an initial calibration. You want to make sure each of your chosen processors is operating in its sweet spot, and especially, that they all put out the same level. The easiest way to do this is by inserting a sine tone generator on ‘LV float’ set to 1kHz at your preferred operating level (mine is 0VU = ‑18dBFS). With this test tone now sending to all of the processors, you can calibrate them for the desired amount of gain reduction or response and make sure that they maintain unity level at their individual and collective outputs.

The first stage in calibrating the setup is to set all the compressors to deliver the same amount of gain reduction and the same output level in response to a test tone.

I chose to calibrate the three compressors all to deliver 5dB of gain reduction with this test tone, but there’s no rules here. Due to their different topologies and ratios, the actual response in a mix is different, with the Weight Tank compressing the most on average and the 1176 the most on the highest peaks. My ‘LV float’ fader defaults to ‑3, though. That way, I initially start with less compression and leave myself some room to push. Come mix time, the process is simple: I have a quick listen to each of the four processors to get a feel for how they interact with the actual voice in question. Then I decide whether I want to run the Weight Tank in the cleaner ‘round’ or the more saturated ‘drive’ mode, and the Sebatron on my default ‑25dB gain setting or the more aggressive ‑15dB.

The great thing about this approach is that you have different aspects of your vocal sound on different faders, which you can mute and automate.

While listening to the mix, I play around with the faders to achieve the best vocal tone. This depends on the voice and instrumentation, but after using this setup for so long, it has become a very intuitive process. The great thing about this approach is that you have different aspects of your vocal sound on different faders, which you can mute and automate. Sometimes I run all four faders on a similar level all the way through. Sometimes I have them on wildly different fader levels and do lots of automation. Sometimes I use one of them just for the chorus. Sometimes I’ll heavily automate the ‘LV float’ channel to control the ‘push’ into the compression, sometimes I don’t touch it.

Complete the calibration by finding suitable default levels for the faders.

Tone Control

One of the most surprising things about this technique is how it minimises the amount of EQ needed on the lead vocal track. Between the full‑bodied low end of the Weight Tank, the midrange push of the 1176, the sexy sheen of the LA‑2A and the aggressive high‑end crunch of the Sebatron, there’s always a great sound to be found within a minute. Every time, the vocal easily gains authority and importance and will stand its ground in the mix even at a relatively low volume, which is especially important for rock music.

Those of you who never heard about this technique before may have some questions. I’ll try to answer the obvious ones: Yes, I likely do EQ on the lead vocal tracks that feed the ‘LV float’ bus. Yes, I will use compression on those tracks if necessary. Yes, I do send to delays and reverbs from the lead vocal tracks. And yes, I often will have additional processing on the ‘LV float’ and ‘LV master’ channels; typically, it’s a either a de‑esser or an instance of Oeksound’s Soothe 2 on the former, and sometimes a broad‑stroke EQ like a Pultec on the latter.

After you get the hang of it, it’s not hard to think of further options to explore. You can add more or different compressors, or a dry path on its own fader. And of course this send/return compression scheme isn’t limited to lead vocals. It also works very well on bass guitar, especially. Whatever you do, have fun experimenting and getting creative, and enjoy the feeling of manipulating emotion with faders. It’s a rewarding approach.