Recording - July 2016 - 54

By Paul J. Stamler

Part 19: Overdubs
What follows is a work of fiction. We didn't do any overdubs during the
Ottoman Underground session, but I promised the Editor that I'd include an
article on overdubbing in this series. After all, it's an integral part of how
most recordings are made nowadays. [Thank you, Paul-MM] So I'm
describing something we didn't actually do... but if we had, here are the
things I would have done, and some of the issues that would have arisen.
1

Basic setup
After the main session was over, Will Soll decided he would overdub a
mandolin part to complement the tenor banjo that's at the core of the recording. I set him up at the same spot in the studio, in the hope that the room
sounds would match.
I didn't, however, use the same Microtech Gefell M930 mic as I'd used on
the banjo. Will's mandolin, a Collings, has an extremely bright tone, and the
M930, with its upper-end brightness, makes a bad match for it-the sound
of the combination is too forward and brash for me. If I'd used the M930,
the mandolin would have taken over the mix, and both Will and I wanted it
subordinated to the banjo.
Instead, I set Will up with the same model microphone I'd used on Dana
Hotle's clarinet, a Neumann KM 84 small-diaphragm condenser. I've found that
the KM 84's remarkably flat frequency response meshes well with the sound of
most mandolins, and it's my "go-to" mando mic. The mic's flat off-axis response
meant that any room sound I picked up would be relatively free of coloration.
Getting fancy
I've talked before about using controlled leakage between microphones to
create a sense of space. Obviously, since a time machine isn't on the studio's equipment list, Will's overdubbed mandolin wasn't going to leak into
anyone else's mic... or was it?
I took advantage of the fact that most modern DAWs have nearly unlimited
track capabilities, and set up several mics intended solely to pick up leakage.
I put another KM 84 in the place where I'd used it for Dana's clarinet (you
can see the layout in Article 6 of this series, Recording, June 2015); I set the
54

RECORDING JULY 2016

console's preamps to the same settings as I'd
used on Dana. I hadn't used any compression or
EQ on Dana's mic, so I didn't use any on this
one. I created a mono track in the DAW (Pro
Tools), which I called "OD Leak C" (the C stood
for "clarinet"). To avoid comb filtering (see
"Outwitting the Phase Monster", November /
December 2014), I made sure that the signal
from the mandolin was at least 10 dB lower in
the leakage track than it was in the main mandolin track, which I called "OD Mando".
I also set up an ORTF pair of Oktava MC-012
mics in the same place I'd had them for the original session-behind Will, pointing away from
the musicians-and set their preamp gains the
same as they had been before. I called their
tracks "OD Room L" and "OD Room R". I decided to stop there; I didn't simulate leakage into
Will's vocal mic or into the mic on Shlomo's
bass amp. Enough was enough. I was ready to
record, but Will wasn't ready to play, because
he had no one to play with.
Cue mix
Will, of course, needed to hear what had
already been recorded, so he could play along,
and in this fictional session I decided to do the
work on the DAW's virtual mixer rather than the
hardware mixer in the studio. So I took the tracks
that had already been recorded (clarinet, vocal,
tenor banjo and bass) and brought them up in
the DAW's mixing window (I didn't include the
room mics). Now I'd get to use the Aux buses I
wrote about in Article 16 (April 2016).
I had the option of creating a stereo bus and
routing the already-recorded tracks to it, using the
DAW's panpots to create a semblance of a stereo
mix. But I decided to keep it simple, so I used the
DAW's Aux buses to create the monitor mix that
Will would listen to. The Aux buses are mono in
most DAWs, as they are in most chips'n'solder
mixers, so I used both Aux 1 and Aux 2, and set
the same levels on each. I routed the outputs of
the two Aux buses to outputs 3 and 4 on the
(mythical) interface box I was using, and connected those outputs to the Left and Right inputs of the
headphone amp that would drive Will's cans.
A brief headphone digression
It's important to know that the choice of headphones for an overdubbing session can actually
make a big difference in the result. If you use

Table of Contents for the Digital Edition of Recording - July 2016