Short Version
On a reverb bus, put an EQ before the reverb. Cut 600Hz off the low end and 10kHz off the high. Use 12dB/oct steepness. Notch cut about 3db around 2kHz for strong vocals.
Long Version
Bobby Owsinski, an audio engineer and producer, is a treasure trove of information. With vast industry experience, he has knowledge in spades, and I can’t recommend his videos enough for learning interesting techniques, tips, and trickery regarding mixing and music production. The Abbey Road Reverb is one such technique.
Put simply, the Abbey Road reverb is a low-pass and high-pass filter placed on the reverb bus. What makes it unique is that the filtering is done before the sound hits it, rather than after. It’s main benefit is for vocals, which can be difficult to reverberate without sounding washed-out and ‘blurry’ (for lack of a better term). The example here will use normalised vocal cuts I made as part of an upcoming track called ‘Delta-v’, with a clip ending the article so it can be heard in a somewhat busy mix. For brevity’s sake, I won’t detail the Abbey Road reverb chamber nor the necessity of this effect in the ’60’s. I’ve included a video from Bobby at the end of the article.
Examples
First, let’s hear the vocal on it’s own:
Place a fully-wet reverb effect on it’s own mixer track, and route the sound to it (i.e., a bus reverb). Here’s the same vocal with a decent amount of reverb and a little volume adjustment:
Right away, we have issues. The vocal doesn’t quite stand out as much, and emphasis and dynamics have been taken away. This would be fine if the sound was destined for the background of the mix, to pad up the sounds in front, but not helpful when we want it up front (foreground/background mix terminology is a tale for another day). There are a few ways to proceed from here. The reverb could be side-chained, reducing it’s volume while the sound plays. Or, the decay time could be dialed right back.
Abbey Road
This is where Abbey Road comes in. Suppose we want a reverb long and drawn out for creative effect. Normally, we would shape the reverb sound following the effect, but this time, we’re shaping the sound itself going in. This makes the reverb not react to the strong low and high frequencies. For vocals, a lot of power and presence comes from the 2-3kHz region. This can be remedied by gently reducing the 2kHz to 3kHz area.
The EQ here illustrates the setting. For my sound, I like a less prominent reverb, so I have removed a little more of the high end, say, down to roughly 8kHz. My vocal cuts benefitted from a gentler notch too, and just above 2kHz was a sweet spot for resonance. You can hear the difference below:
The reverb itself is now less frontal, and doesn’t react so much to the strong cuts. In effect, the reverb appears to sit ‘behind’ the vocal cuts now, and doesn’t wash out the transients (the sharp, pointed bits of each sound). Tonally, the reverb has less thickness. I used this effect on ‘Delta-v’ to ‘push’ the vocal cuts forwards, despite a very busy mix. This way, blurred, heavily delayed vocal cuts could retain their soundscape but not resign to the mix background.
While it can be difficult to hear the minute changes, a comparison back to the reverberated sound without EQ is telling of the kinds of issues badly treated reverb can create. For more detail, see Bobby’s video below for a more detailed explanation.
Final Thoughts
About the only hard set number in this technique is the cut below 600Hz. This is necessary to stop the reverb algorithm reacting to the low end, and grant the reverb a clarity to let the vocal sit forward. However, different values for the low-pass attenuating the highs can be adjusted to flavour, as different reverbs will have different high end character. Play around with the values of the low-pass and the small dip, and hear how it benefits your particular vocal or sound. Some swear by cutting as far down as 6kHz, while others cut only the extreme highs and let the reverb retain an airy and shimmering quality.
The steepness of the filter also matters, and should be left at either 12dB/oct (recommended) or 18dB/oct. Anything under this range will nullify much of the effect as too much bass is still hitting the reverb, and anything above this range will begin to sound artificial and thin. That said, I am a stranger on the internet and no authority; play with different values and see how it sounds.