Parallel compression can add interest and sparkle to a vocal, but almost always you’ll need to automate the lead vocal in the mix. So what can go wrong if you parallel compress?
Here’s the scene…
You have a lead vocal track, and you’ve added parallel compression. One way to do this is to bus the vocal track to an aux track and compress the aux. Like this (different DAWs will look slightly different, but the method is the same) Click the image for a larger version…
In the pic you’ll see the vocal track on the left routed to Bus 1. The send to Bus 1 (the skinny panel next to the compressor) is set to post-fade. The track named PComp takes its input from Bus 1 and there’s a compressor inserted. The level of PComp is set so that it brings up the low-level parts of the vocal, but at higher levels it hardly makes any difference, which is the essence of parallel compression.
There’s a problem already. Can you see it?
The send to Bus 1 is set to post-fade, so the level of the send goes up and down according to the setting of the Vocal fader. This would be fine for reverb because that’s exactly what you want in 99% of cases. But for compression, when you bring the Vocal fader down it pushes the compressor less hard, and because compression is sensitive to input level where reverb is not, it changes the sound. Whoops. Similarly if you bring the Vocal fader up.
So what about using pre-fade so that the level of the send to the compressor never changes?
This is no better because if you bring the Vocal fader down, the compressed signal will become louder in proportion to the Vocal track. If bring the Vocal fader up, the compressed signal will be less in proportion.
This isn’t such a problem however because if you want to change the level of the Vocal track, you can easily adjust the level of PComp to where you want it. It’s a problem, but a tiny one.
Now consider that this is a lead vocal track. In most mixes you will want to automate the level of the vocal as the song plays through. Legend has it that producer Mutt Lange would automate so heavily that the fader (on a physical mixing console) would buzz audibly, it was moving up and down so fast.
So using this setup, the proportion of PComp will vary almost continuously as the automation moves the Vocal fader up and down. And that’s bad, because it’s changing the proportion of parallel compression as it moves.
So what we need is a different signal routing setup
What we need to do is apply the automation to the vocal after it has been parallel compressed. Like this…
What you can see here is that the output settings of both Vocal and PComp have been changed (Look under I/O just above ‘AUTO’). Previously they went directly to the stereo output bus. Now they both route to Bus 2.
There is now a new aux track called Sum. This takes its input from Bus 2 and routes to the stereo output.
So now what you can do is balance the levels of Vocal and PComp to get the sound you want. Following that you can apply any level changes, manual or automated to Sum. You will always get the same balance of Vocal and PComp and your parallel compression will sound the same all the way through the song.
One more thing…
You might not notice until I point this out, but Vocal and PComp have been set to solo-safe. You can see this where the solo button ‘S’ just above the fader is greyed out.
This means that if you press the solo button on Sum, to hear the parallel compressed vocal alone, then Vocal and PComp will not be muted because if they were, you would hear nothing. Solo-safe is an extremely handy feature that (I think but correct me if I’m wrong) all DAWs have.
So there you have it. A potential problem with parallel compression is identified and solved.