Recording - May 2016 - 62

gets a boost. Now the whole signal is too loud, and it's overloading the summing amplifier, so you try turning down the
bus master.
Unfortunately, the mix still sounds awful, and here's why. The
bus master control in a chips 'n' solder console is typically a simple attenuator. It can turn down the level of the signal coming
out of the summing amplifier, but if that signal is overloaded; all
you get is a quieter overloaded signal. To get the summing amp
to sound sweet and clear, you need to turn down the levels of
the signals going into it, not what's coming out of it.
That's what can happen in a chips 'n' solder console; I outlined a similar way for overload to jump up and bite you in an
earlier episode of this series. Do the virtual consoles in DAWs
have similar problems? We talked about this in a Talkback column several months ago, and our conclusion was that "it
depends." Some DAWs automatically cut down the level of
signals going into summing "circuits" when there are more of
them to mix, which can help prevent overload. Others use formats for the audio numbers that are essentially overload-proof;
that's what the "32-bit floating point" thing is all about. Some
DAWs, usually older designs, will run into the same summingamplifier overload problems as hardware consoles.
Okay, back on the bus. In Figure 3 I showed a mono mixing
bus, but we usually want the output of the console to be a
stereo mix. Figure 4 shows a (hardware) stereo mixing bus
which will do that; the tracks come in as stereo signals (tracks
2L, 2R, etc.), and the bus master level control at the output is a
dual one. I'll talk in a minute about how the tracks become
stereo left and right.
Sends
How do we send stuff to the various buses? Figure 5 shows
the block diagram of a single channel of a (very rudimentary)
mixing console, set up in mixdown mode-it could be hardware, or it could be a DAW.
Signal flow is from left to right. The first thing that the audio
track (real or virtual) hits is a point where an Auxiliary (Aux)
send branches off the main signal path. This includes a level
control, and it pipes a copy of the signal to the AUX1 bus. In
a chips 'n' solder console, this is probably a mono bus, like
the one shown in Figure 3. We'll talk about where you might
want to send the output of that bus in later articles; the important thing is that it's there, and the send for it is the first thing
the signal encounters.
Next in our imaginary console comes the channel's EQ
module (these days almost all hardware consoles have EQ in
each channel). Then comes a box called "Insert". In a hardware console that's usually a jack or pair of jacks enabling

RECORDING MAY 2016

you to connect a rack-mounted signal processor-say, a compressor, or a fancier EQ, or any of many possibilities. In a
DAW the same possibilities exist, but they're built into the software-one DAW (Audition) actually calls the selection of
effects that go into the signal path at the insert the "rack".
This is probably the right time to mention plug-ins. These are
signal processing devices, in software, that extend the options
already available in the DAW. The range of plug-ins has become
enormous, and I can't begin to cover them all, but they're
reviewed regularly in Recording. One category of plug-ins that's
extremely popular is emulation plug-ins; these mimic the functions
and response of classic devices (e.g., Pultec equalizers or
Universal Audio compressors) that few of us could afford if we
bought them as hardware. One of the nice features of mixing on
a DAW is the possibility of mixing many tracks of audio with,
say, a separate Pultec EQ inserted on every track, which you
couldn't manage in any reasonably-priced hardware setup.
There's no such thing as a free lunch, of course, and these
multiple plug-ins cost something besides money. (They do cost
money, of course, but typically you only have to buy them
once.) Plug-ins use computer power; mostly they take up a
good deal of the computer's CPU capability, but they can also
exert demands on the storage system (usually a hard drive).
One answer that manufacturers have come up with is to offer
an external module that does digital signal processing (DSP).
These modules do the processing that plug-ins demand, taking
the load off the computer's own CPU. Universal Audio, for
example, sells modules that incorporate single or multiple DSP
chips, and come bundled with an assortment of plug-ins,
including some remarkable emulations of the company's classic hardware from years past.
Onward. Next in the chain comes the branch to another
Aux send, labeled AUX2, with its level control. The AUX2 send
is downstream from the Insert-that is, it comes afterward in
the signal chain, so anything you send to the AUX2 bus will
have been processed by whatever's connected to the Insert.
Next comes the Fader; that's the big control that affects the
signal's level in the mix. On most chips 'n' solder consoles this
is a large slider, and on most DAWs it's a visual representation
of a slider.
After the fader comes the send for the AUX3 bus. Unlike the
previous bus sends, this one will be affected by the setting of
the main fader. This means that what you feed to the AUX3
bus will change as the fader's manipulated, and will track the
signal's level in the main mix. That's useful if you're sending
the output of the AUX3 bus to, say, a reverb device; if you turn
the fader down or up, you want the signal feeding the reverb
to go down or up in tandem with it, and having the AUX3
send downstream from the fader makes this happen.
AUX3 is therefore called a post-fader or post-fade send,
while AUX1 and AUX2 in our picture are pre-fader. Some hardware consoles, and many DAWs, allow you to switch some or
all of the Aux sends to be pre-fader or post-fader. The typical
hardware console includes 4 or more Aux buses, although
small desktop models may only have two, one, or none at all.

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