Signal Processing For The Home Studio Owner: Part 2, Gates, Delay, and Reverb

Part 1 of this series focused on Compressors, Limiters, and EQ. Part 2 explores Noise Gates, Delay, and Reverb.

In addition to your microphones, DAW/console, and room, an essential part of any home studio set-up is your signal processing gear. From the dynamics control of compressors and limiters to the effects processing of reverb and delay, these tools are necessary to create a professional-sounding final product. But for the new engineer, these effects can be fairly mysterious, and a tendency to overuse plug-ins and outboard gear is commonplace, especially for someone just learning the nuances of the art of recording.

How can you best use your typical signal processing plug-ins to enhance and optimize your recording? Understanding how the dynamic control processors like compressors, limiters, EQs, and gates function, and knowing how to use multi-effects such as delays and reverbs to perfection will make you a better producer and engineer. It’s also important to remember that signal processing tools are just that – tools. There are no rules stating you can’t use them in different or novel ways to create new sounds. But before doing that, it makes sense to learn about the basic parameters of each and the functions they were invented to serve.

Noise gate
The floor tom was close mic’d, and now, listening back critically, the drum’s nearly five-second release time blurs the tom’s definition. You can live with some of the release but you want to clearly hear the attack of each hit on the floor tom. The best signal processor to help solve this is a noise gate. Noise gates are part of the dynamic processing family of plug-ins. Like compressors and limiters, the noise gate has a user-definable threshold, provides variable gain reduction, and offers attack, hold, and release time parameters.

Gates function by setting a threshold level that determines the amount of amplitude required to open the gate, then letting the audio pass through to the gate’s output. Any amplitude level below the threshold value will not open the gate – so the gated track remains silent. On this particular song, there are a few breaks that Doug leads into with the floor tom, but it rings on too long. Loop the phrase so it plays continuously, then insert the noise gate on the floor tom track and set the threshold level to the point at which the hit on the tom just barely opens the gate. Now adjust the attack, hold, and release parameters to achieve the desired floor tom effect reducing the long decay.

Noise gates are very useful when you need to eliminate any unwanted incidental sounds that may have been recorded. For instance, use one on vocals to eliminate breathing sounds between lyrical phrases, or on a distorted lead guitar to eliminate overdrive noise between lead passages. Noise gates could even be tried on the stereo mix bus output to really tighten the breaks in the song.

Noise gates can also create their own problems, since everything recorded on the track you are gating is eliminated according to the gate’s envelope, including any ambient leakage. This can sometimes cause a perceptible and distracting dropout within the song’s mix. As with all signal processing, use your own ears to decide how much noise gating is useful in your mix.

Delay and Reverb
So far the plug-ins mentioned in this article are generally considered to be dynamic, having to do with varying or controlling amplitude in real-time. The next type of signal processing is time-based processing, and we’ll focus on the bread and butter effects of delay and reverb.

In our example, the bass player recorded her parts directly into the DAW interface via direct box. Her Fender Precision Bass sounded great, and with a little EQ and compression, the track is all set. The guitar player tracked the leads with his guitar processing pedals, but recorded the rhythm guitars direct and dry. Now you are faced with the challenge of giving life to his rhythm guitar parts.

Let’s start with delay. A delay is a time-based processor that generates discrete wave fronts of the input signal according to the delay time. Delay settings of 250 to 500ms will create rhythmic interest while smaller times such as 20 to 80ms can create a sense of depth. You can also create echo effects by increasing the amount of feedback, a parameter that returns the output of the delay circuit back into itself.

Many delays provide rhythmic note values, such as whole, half, quarter, eighths, etc., and offer a sync option that times the delay precisely to the tempo of the original track. The delay also has low and high-cut filter parameters, so you can change the frequency content of the delay generation when feedback is used. You can also modulate the delay time using the depth and rate parameters, and create variable moving rhythmic echoes.

Here’s one practical approach, assuming there are two rhythm guitar tracks. Start by bussing one rhythm guitar to an Aux Track and insert a medium delay. Set the delay time to 40ms and pan the Aux Track to the right, leaving the original rhythm guitar in the left channel, creating a delayed stereo spatial spread.

For the second rhythm guitar track, a long stereo delay provides a good option. For this plug-in, a stereo Aux Track is required, or if inserting on a mono audio track, you can automatically convert it to a stereo track. For the most part, the controls are the same as the first delay used, but now there are separate left and right channel parameters on the delay itself, allowing you to create complex rhythmic and spatial movement in the stereo field.

Finally, you turn to the last solo instrument, a melodica, a three and a half octave reed instrument played by blowing into its mouthpiece and fingering its keys. It’s a warm sounding instrument that requires some compression but generally sounds fine. Here the decision is to add Reverb.

Reverb (short for reverberation) is one of the oldest and most widely-used time-based effects. It can add lush ambient room sound to any instrument. Like delays, reverbs generate multiple wave fronts, but there are a large number of fronts and the time differential between each front is extremely short. It’s easiest to think of these fronts as reflections of the original sound, like the way an instrument sounds when played in a well-designed concert hall. The sound generated by the instrument moves out in all directions. It comes directly toward the listener but it also hits the floor, walls and ceiling. The sound reflections from these surfaces return back to the listener slightly delayed from the original sound, depending on the size and depth of the space. Of course the reflections off the floor, walls and ceiling also continue to bounce off of the surfaces in the space and listeners perceive all those reflections at slightly different times, creating the perception of a spacious concert hall.

Today’s reverbs emulate a wide variety of acoustical spaces. Some of the most common environments include Hall, Room, Church, Club, and Stage. Some reverb plug-ins offer additional emulations taken from the analog reverb days such as Plate, Spring, and Chamber. In all cases there are a few common parameters that can be selected and adjusted to good effect.

Reverb type refers to the room being emulated (hall, room, arena, church). Reverb size refers to how large of a space you can create. You might have a large room, a small church, or a medium hall. Diffusion is a parameter that determines how far apart each reflection spreads out from the instrument, giving a sense of depth of the enclosure. Reverb Decay adjusts how fast the reflections die out after the initial attack of the sound. Pre-delay is an important parameter that determines the time differential between the direct sound and the point at which listeners perceive the reverb reflections. Finally, most reverbs have low and high cut filters that can reduce or increase harmonic partials as a part of the reverb’s reflections. These filters are very useful to create transparency within the reverb process.

It is important to remember, the best sounding reverb is the one that enhances the sound without being too noticeable. For the melodica solo, the large hall setting, a pre-delay of 40ms, a wide diffusion and cutting high frequencies at 8kHz, results in a dreamy-sounding solo for this tune. Everything is now set to begin the final mix, with signal processing tools helping to address the issues that would have made this EP project sound less polished.

Final Thoughts
When using plug-in processing it is critical to keep in mind the style of project on which you are working, the type of instruments you will be recording, how they will be recorded, and what kind of plug-in processing will help when it comes time to mix. As you get more familiar with how signal processors work, listen to some of your favorite recordings and try to reverse engineer what types of processors and settings may have been used.

Addendum: The Lowdown on Impulse Reverbs
The various reverbs known as impulse reverbs offer a different approach to time domain processing. These plug-in effects processors provide reverberation, but their processing method is very different from the traditional reverb plug-in. Impulse reverbs, although they have many of the same parameters mentioned above, access a complete library of sampled sonic spaces known as impulses, taken from various rooms known for their unique acoustical characteristics found all over the world.

Famous concert halls, cathedrals, and classic recording studio tracking rooms are just some of the options available when shopping for impulse reverbs. Instead of algorithms that emulate or calculate the dimensions of a hall, church, room, the impulse reverb actually loads the acoustic signature of a given space with all the actual time variables included. This results in a totally convincing audio reverberant spatial environment.

Addendum: Using Auxiliary Tracks to Preserve Your Computer’s CPU Power
It should be noted that reverb and delay plug-in processing demands more CPU power and larger amounts of RAM in order to accommodate the time differential between the input of the unprocessed signal and the output of the plug-in processor. Because of this time differential, something known as latency (the time difference between the audio signal written to the drive, passing through the CPU, getting processed, and then returning to the audio output) can result in slight phase anomalies. Most professional audio recording applications compensate for latency by imperceptibly delaying the returning audio to the track output. The user is never aware of the delay and for the most part operates the application as usual. However, it is important to keep in mind the total number of time-based processors you insert on any given audio track.

The more time-based plug-ins inserted on a track, the greater the amount of compensation required. For this reason, when using time-based processing, you may wish to create an Auxiliary track (Aux Track) as you are setting up your mix and insert the time-based processors on the Aux Track and then bus the audio via the individual channel sends to the Aux Track. You can set up one for reverb, another for delay and so forth. In this way, the original audio track with the recorded instrument information is separate from all the time-based plug-in processors, minimizing the need for latency compensation of the audio track. Doing so will result in a sharper and more defined audio image throughout your mix.

Read more
Recording with Reverb and Echo – Tips and Lessons from Six Classic Tracks (November 2010)

Home Studio Posts: advice on how to record, music gear, guides, and pro insights (March 2012)

Keith Hatschek is a regular contributor to Echoes, author of two books on the music industry and directs the Music Management program at University of the Pacific. Jeff Crawford is a recording engineer and producer with more than 30 years industry experience. He also teaches music technology at Pacific.

–>

Tagged as:

audio compressor,

compressor limiter,

delay and reverb,

disc makers,

dynamics control,

effect processing,

eq,

home recording,

home studio,

Keith Hatschek,

limiter,

noise gate,

reverb,

signal processing

Advertisements

Signal Processing For The Home Studio Owner: Part 1, Compressors, Limiters, and EQ

In addition to your microphones, DAW/console, and room, an essential part of any home studio set-up is your signal processing gear. From the dynamics control of compressors and limiters to the effects processing of reverb and delay, these tools are necessary to create a professional-sounding final product. But for the new engineer, these effects can be fairly mysterious, and a tendency to overuse plug-ins and outboard gear is commonplace, especially for someone just learning the nuances of the art of recording.

We’ll explore the functions of these units by way of a fictitious example. You spent the last two weekends tracking instruments and vocals for a friend’s rock band using your new digital audio interface and music production software. Because the tracking took place in the same room as your interface and computer, everything was monitored on headphones. Now you’re faced with mixing the project in a spare bedroom using near-field monitors. In all, the first part of the process went well: the tracks sound pretty good, the band’s time and tuning were on, and the performances had the right energy. Inevitably, there are a few things that caught your attention that will need to be addressed with your signal processing software.

First, there is a very wide dynamic range in the lead vocals and various solo parts. Second, the direct-recorded rhythm guitars sound flat and dry. Finally, the drums were so loud in the room while the group’s drummer recorded that it was difficult to do any critical listening during recording. All you are sure of is that each mic was working and that none of the drum levels seemed to overload, but you’re pretty sure the drum tracks will need some fine tuning to sit properly in the mixes.

How can you best use your typical signal processing plug-ins to enhance and optimize your recording? Understanding how the dynamic control processors like compressors, limiters, EQs, and gates function, and knowing how to use multi-effects such as delays and reverbs to perfection will make you a better producer and engineer. It’s also important to remember that signal processing tools are just that – tools. There are no rules stating you can’t use them in different or novel ways to create new sounds. But before doing that, it makes sense to learn about the basic parameters of each and the functions they were invented to serve.

Compressors and Limiters
While it’s common to use compressors and gates when going to tape, let’s assume that little to no plug-in processing was used in the tracking of the instruments. (One reason not to use plug-in processing to tape is that it is destructive, which is to say, you can’t go back and remove it once the audio has been printed to the drive.) In our scenario, the tracks were recorded clean, with only guitar amp tone control settings, stomp boxes, and microphone placement used to establish the overall sound of the track.

After setting the instrument levels, you begin the mixing process by working on the lead vocal of track one. The singer was varying his voice tone and volume levels throughout the song, toggling between low whispers, soft and loud vocalizing, and ultimately to a primal scream and growl in the outro. How will you treat these very different vocal dynamics so they fit into the overall mix?

Two of the most useful plug-in tools for such a scenario are compressors and limiters, which are used to manage the wide fluctuations caused by an overly dynamic delivery, vocal or otherwise. The difference between compressors and limiters is basically that compressors have a variable output level while limiters have a fixed output level.

A compressor reduces the amount of output signal level in relation to the input signal level according to a given ratio, beginning at your user-defined threshold. In other words, it brings the loudest sounds down, and brings the softest sounds up. This is ultimately determined by the ratio setting. For example, you set a threshold for the output signal. A compression ratio of 2:1 means that for every 2 dB the input level is above the threshold you’ve set, the output level is reduced by 1 dB. A ratio of 4:1 means that if the input signal is 4 dB over the threshold, it will be brought down to 1 dB over the threshold, or reduced by 3 dB. It’s as if you were riding the gain on a console fader. When the input signal gets too loud, you pull the fader down, lowering the gain. When the signal gets too soft, you push the fader up, raising the gain.

A limiter allows you to set a maximum output level that will not be exceeded, regardless of the amount of input signal level. It’s often described as a 60:1 ratio, or ∞:1 ratio. Anything that exceeds the threshold is brought down to the output level you’ve set.

In our example, the sections where the vocalist sang the verses were dynamically consistent, but in a few spots, the vocal level dropped down to an intimate, whispery style, and the signal is getting lost in the mix. Here, compression will do the trick.

Depending on how great of a dB variation, start by setting the ratio to approximately half the difference between the highest and lowest vocal level on this track. For example, if there is a 10 dB difference between the vocal’s dynamic high and low point, you can set the compressor’s ratio to 5:1. Now reduce the threshold setting to the point at which you want the gain reduction of the vocal to start. You begin to notice the sound of the gain reduction as it kicks in and it may sound a little unnatural. Try adjusting the attack time of the compressor. The attack and release time parameters control how fast the compressor will respond as the signal crosses the threshold. Now you may find that the overall vocal level has been reduced considerably. Even though you like the evenness of the vocals, the level is now too low in the mix. When using gain reduction, it is often necessary to raise the output gain of the compressor to bring the vocal back to a usable listening level.

On the phrases where the singer screamed and growled, a better choice might be a limiter. If the verses and the screaming phrases are on the same track, separate the sections by copying the screaming and growling vocals to a new track. Let’s say that the screaming is consistently much louder than the growl and that the growl is near the level you want the verse vocal to be. Insert the compressor plug-in on the new track and set the limiter’s ratio to the maximum value. Now adjust the threshold of the compress/limiter to a point at which both the scream/growl vocals produce the same output signal level. With a limiter you can easily knock down the louder scream so it’s equal to the growl’s amplitude.

As you learn about compressors and limiters, try experimenting with different ratio and threshold values. Tweak the attack and release parameters and experience their effect. Try some of the presets found in the plug-in preset menu in your software. If you are working with the Compressor/Limiter Dyn3 in Pro Tools, a helpful visual indicator for setting compressor parameter values is the real-time dB level indicator. This small square in the input/output graphic reflects input-to-output levels as compression or limiting begins to affect the signal.

Try playing with the factory presets. The presets that come with your music production software were designed by recording studio professionals and are offered to you as a starting point. You can also create, name, and edit your own presets for future use.

EQ
The drums offer a completely different challenge. Although you may use some dynamic processing on the drums, such as a limiter placed on a stereo group channel output, if recorded individually, each drum needs to be listened to and treated on its own. During recording, it’s often difficult to distinguish between the direct sound of the drums in the room and what was being recorded. Now at mix, you might hear things you didn’t notice during the tracking session. The kick drum sounds a bit “tubby,” there’s an overtone ring somewhere in the snare and the deeply-tuned floor tom has a five-second tonal decay time. None of these drum issues are insurmountable when using the right kind of plug-in processing.

First, it is important to keep in mind the type and style of drum sound you want to produce. For rock, the kick and snare are primary. They need to be tight, clearly defined and in your face. Let’s start with the kick. After placing a limiter on the kick to even out the level, you found the overall kick drum timbre did not cut through. The large diaphragm dynamic mic you placed on the kick delivered a deep fat bottom but the midrange frequencies are over emphasized and the top end frequencies are weak. For making adjustments relating to frequency, the right plug-in tool is the equalizer (aka EQ).

The EQ is a frequency-specific amplifier, and it comes in two basic flavors: graphic or parametric. Both essentially make tonal adjustments by increasing or decreasing amplitude at specific frequencies, but in the case of the graphic EQ, the bands are set at fixed center frequencies across the 20-20kHz bandwidth. The number of bands may vary from five to 30.

To fine tune the kick drum, we’ll grab a 7-band parametric EQ. It provides a smaller number of bands, yet gives the user more precise control over each band than a graphic EQ. Each frequency band has a dB control, usually +/-15 dB, a sweepable frequency range control, and a “Q” control that sets the width of the frequency band to be adjusted. The higher the Q, the narrower the frequency band that will be affected. Conversely, lower Q values result in a wider bandwidth range being boosted or cut.

In the case of the kick drum, the low-mid to mid-range frequencies (500Hz to 2.5kHz) are causing the “tubby” kick sound. Tune the EQ’s frequency band to emphasize the tubbiness in both amplitude and bandwidth. Don’t be afraid to be extreme with the frequency’s amplitude control. You want to really hear the influence of the EQ on the kick drum’s sound. Once you have found the frequency at which the tubby sound is most extreme, drag the frequency point into negative values. This should greatly reduce the proper frequency range to minimize the kick’s unwanted tone.

The technique of emphasizing and then subtracting unwanted frequencies is one way to eliminate annoying hums, rings, and any other frequency zones that need to be equalized. This technique will also be very effective on the ringing snare drum overtone. Finally, to give the kick drum a bit more definition, you can use the same method of experimenting to find the right frequency to boost to emphasize the kick drum’s attack. It turns out that it is at the high-frequency EQ band at 7kHz. By boosting the EQ to brighten the transient attacks, the kick sounds fat, but now has the attack to punch through the mix without overpowering the other tracks.

Continue reading about Noise Gates, Reverb, and Delay in Signal Processing For The Home Studio Owner: Part 2.

Keith Hatschek is a regular contributor to Echoes, author of two books on the music industry and directs the Music Management program at University of the Pacific. Jeff Crawford is a recording engineer and producer with more than 30 years industry experience. He also teaches music technology at Pacific.

–>

Tagged as:

audio compressor,

compressor limiter,

delay and reverb,

disc makers,

dynamics control,

effect processing,

eq,

home recording,

home studio,

Keith Hatschek,

limiter,

noise gate,

reverb,

signal processing