Vocals · 7 min read
Mixing vocals the Motown way: plates, compression and tape
How the 1960s Detroit vocal sound was built, from close mics and quick compression to plate reverb and tape, and how to get it in any DAW tonight.
By Daniel ·
Put on a mid-60s Detroit single and listen to the lead vocal. It sits right in front of you, bright but never harsh, with a short shimmer of space behind it. The band is loud and busy, yet the voice never gets lost.
None of that came from one magic box. It came from a handful of simple choices stacked together: a close mic, a compressor that never let the singer drop out, a bright reverb, and tape that rounded everything off. Each one is easy to copy today. Here is how I approach it.
Start with the performance and the mic
The singers were recorded close to the microphone, with the band often in the same room or tracked live around them (the setup varied from session to session). Being close does two things: it adds warmth in the low mids (the "proximity effect", where a directional mic gets bassier the nearer you get) and it keeps the room out of the sound.
At home:
- Sing or record a hand's width from the mic, not a metre away.
- Hang a duvet or coat behind the singer to kill reflections. You want the space to come from the reverb later, not your bedroom.
- If the low end gets boomy, use a high-pass filter around 80 to 120 Hz rather than moving further away.
Compression: keep it riding up front
The vocal in those records barely moves in level. Every word, even the quiet ends of lines, comes through. That is compression doing its job: turning down the loud parts so the whole performance can be pushed forward.
The old valve compressors were not surgical. They grabbed fairly quickly, let go smoothly, and added a little thickness on the way through.
Settings to start from:
- Ratio: 4:1, or a fixed "medium" setting if your compressor has one.
- Attack: fast to medium, around 5 to 15 ms. Fast enough to catch the front of a word, slow enough to keep a bit of bite.
- Release: around 100 to 300 ms, or auto. You want it to recover between phrases, not pump between syllables.
- Gain reduction: 4 to 8 dB on the loudest lines. Watch the meter. It should be moving most of the time.
If one compressor working hard sounds squashed, use two working gently: the first catching only the peaks, the second adding 3 or 4 dB of steady levelling.
Brightness without harshness
Those vocals feel bright, yet they never sting. Part of that is tape and old electronics gently rolling off the very top while the upper mids come forward.
- Add a broad lift of 2 to 4 dB somewhere between 3 and 5 kHz for presence.
- Then tame the very top with a gentle low-pass filter or shelf, starting around 12 to 15 kHz.
- Use a de-esser if the "s" sounds start to spit. Brightness plus compression makes sibilance louder.
The plate: short, bright, behind the voice
Many studios of that era used echo chambers (a real room with a speaker and a mic in it) and metal plate reverbs. A plate is a large steel sheet that vibrates when you feed it sound. It gives a dense, bright, smooth tail with no sense of walls, which is why it sits so well behind a voice.
To get there:
- Use a plate reverb setting, not a hall or room.
- Decay: 1.2 to 2 seconds for ballads, shorter for uptempo songs.
- Pre-delay: 20 to 60 ms. This leaves a small gap so the words stay clear before the tail blooms.
- EQ the reverb, not the vocal: cut the reverb's low end below about 300 Hz and its very top above about 8 kHz. Muddy reverb is the most common reason vintage vocal mixes go wrong.
- Put the reverb on a send and blend it in until you can just hear it, then back it off a touch.
A short slapback echo, a single repeat around 80 to 130 ms, can add width and energy on faster songs. Keep it quiet and filtered.
Tape: the glue
Everything on those records went to tape, often more than once as parts were bounced together. Tape adds soft saturation (gentle extra harmonics when you push it), smooths sharp peaks, and slightly lifts the low end while softening the top.
You don't need a tape emulation to get some of this:
- Add a saturation stage after your compressor and push it until you hear the vocal thicken, then pull it back halfway.
- Aim for warmth you can feel rather than distortion you can hear.
- If your saturation has a "tape" mode, try it. If it has a "tube" mode, try both together at low settings.
This is the stage I care most about, and why Ember Press puts compression, tube and tape in one box. Turn Drive up on a vocal and you get the levelling, the thickness and the softened top together, level-matched so you can hear if it's really better.
Make the band make room
The vocal sounded huge partly because everything else was arranged around it. Horns, guitars and keys often sat in a different register to the voice.
In the mix:
- Carve a small dip of 1 to 3 dB around 2 to 4 kHz on busy instruments, so the vocal's presence range is free.
- Keep the backing vocals darker and further back, with more reverb than the lead.
- Check your mix in mono. Many of these records were made for mono radio, and a vocal that holds up in mono will hold up anywhere.
Try this
- Record close, with soft furnishings behind the singer.
- High-pass at around 100 Hz.
- Compress 4 to 8 dB with a medium attack and auto or medium release.
- Add presence at 3 to 5 kHz, roll off above 12 kHz, de-ess.
- Send to a bright plate, 1.2 to 2 s, with 20 to 60 ms pre-delay, EQ'd on the return.
- Add a gentle tape or tube stage for thickness.
- Check it in mono, then turn the vocal up 1 dB more than feels polite.