I Can't Believe I've Been Making this Mistake for Years

I’ve been using Saturn 2 and other multiband saturators for years now, and honestly, I’m kind of embarrassed that I just now realized this.

So, if you want to avoid making a mistake that I’ve been making for longer than I’d like to admit, stick around.

Basic Saturators Work Perfectly

Multiband saturators or frequency-specific saturators have been very welcomed additions for mixing and mastering engineers, whether we’re talking about exciters or newer plugins.

But for most of us, or at least I can speak for myself, I got pretty used to regular single-band saturators. So much so that when multiband versions came out, I figured they’d work the same way - but I’m realizing now that they really don’t.

Basic or full-spectrum saturators generate harmonics from the fundamental frequency; aside from some soft-knee compression, this is their primary job.

And fortunately, they do it pretty much perfectly.

If I feed an A2 note or 110Hz into a saturator, I know I’m going to get perfect multiples of that fundamental - this is what a harmonic is, a multiple of a fundamental.

Even when I have a more complex signal, say a bass guitar that has overtones and naturally occurring harmonics that make up the instrument’s timbre, I can still count on this saturator to generate harmonics from this fundamental note.

If the saturator generates a 2nd order harmonic, then I know that it’s creating a note an octave above A2, which would be A3. Which is perfect, because I know this will be in-key with my song.

Or if it generates a 3rd order harmonic, I know this will be 3x the frequency of the fundamental - in this case, an E3, which is just about 330Hz.

This may or may not be in key, but, at the very least, it’s a note and not some random frequency.

So, everything’s working as it should. I don’t need to think too hard about what the saturator generates since it’ll use the fundamental frequency of the input to create harmonics.

But here’s something I just learned.

Watch the video to learn more >

These Harmonics do not Carry Over

Let me explain - I always knew saturators used the lowest frequency, highest amplitude aspect of the signal to generate harmonics. This is what a fundamental basically is; it’s a high-amplitude root note.

What I didn’t know until this week is that the harmonics generated from one band do not affect what’s generated in the next band when using a multi-band saturator.

For example, if I saturate the low-frequency band, we can observe the formation of harmonics. If I create a higher frequency band that includes one or more of these harmonics, you’d think that any additional saturation I introduce would take these harmonics into account - but it doesn’t.

It doesn’t register these added frequencies as part of the signal. You’re probably thinking, ‘who cares? Why does this even matter, and why is this internet audio nerd talking about this?’

Well, remember what I said a moment ago, a saturator uses the lowest-frequency, highest-amplitude aspect of the signal to generate harmonics. This is what the saturator reads or measures as the fundamental and how it determines from which frequency harmonics will be generated.

When I assumed the saturator measured harmonics generated from other bands, I assumed that these in-key or at least tied to the fundamental harmonics would help it determine how to add new harmonics, which would also be in key, or again, at least tied to the fundamental.

But, since it doesn’t, whenever we use a MB saturator, the harmonics we create could very well not be harmonics at all.

All the saturator can measure is ‘what is the lowest frequency, highest amplitude aspect of the signal.’

If that frequency is an out-of-key note, or maybe not even a note at all, guess what you’re generating - possibly a whole lot of unmusical nonsense.

With something like a single instrument, we could probably get away without knowing this. The natural overtones of the instrument will likely be loud enough to trigger the saturator.

Watch the video to learn more >

But what about a complex mix?

Or a mix bus? What if the loudest, lowest-frequency note the saturator measures is simply not something from which you want to generate harmonics?For example, say I have a song in the key of D Major, but I set the saturator’s crossover at 523Hz, or right below C5, which happens to be present in the mix. C5 isn’t in the key of D major, so the harmonics it generates won’t be in the right key either.

This is a bit of an extreme example I’m about to show you, but I made it aggressive to show what I’m talking about.

Whenever C5 is prevalent enough, the saturator greatly increases its level as well as unrelated harmonics.

But this problem is even more interesting in the high-frequency range, where we have a lot more space between notes. Say the mix has a resonance at 4.3kHz, which is a frequency in between notes.

Then, without thinking about it, I set the band to right below this frequency - again, if this frequency is what gets picked up by the saturator, we have multiples of it being amplified in the mix.

Watch the video to learn more >

A Couple of Solutions to This Problem

First, setting the saturation crossover or band needs to be done thoughtfully. With single-band saturators, we could just insert it and know it’s going to generate harmonics that relate to the song; with multiband saturators, we can’t be so sure.

This means we should set the band to right below a high amplitude in-key note to get the best chance of introducing related harmonics.

Another solution or one to combine with what I just mentioned is an emphasis, de-emphasis technique. With an EQ before the saturator, we could amplify a note we want the saturator to pick up or measure as the fundamental for that band.

After it generates harmonics, we use another EQ with an equal and opposite curve to balance the spectrum back out.

Watch the video to learn more >