|Home > Electric Guitar > Guitar Effects > What They Do|
|I was thinking about how people seem to read the Bible a whole lot more as they get older, then it dawned on me . . . they're cramming for their final exam.|
Before looking at the individual effects, most guitar effects can be placed into these broad categories:
Here are the effects described on this page:
The wah effect moves a peak in the frequency response up and down the frequency spectrum. This movement is usually controlled by rocking a foot pedal, but there are also stomp-box effects which allow the peak to be triggered up or down by your playing intensity.
The resonant (peak) frequency is usually be moved from around 400Hz to 2Khz. One factor that makes different pedals sound special is how the resonance changes as the frequency is moved. Typical wah pedals have increasing resonance as the frequency is lowered.
Some other controls you might see are:
The original Vox and Cry-Baby pedals are considered something of a benchmark. These use the same circuit that has been extensively copied by other manufacturers over the years. They use a coil & capacitor combination to provide the peak, and much fuss has been made of their special coils which apparently contain magical properties.
There are some likely reasons for this, besides plain nostalgic sentiment.
Here is my schematic showing some popular wah pedal modifications.
Phasers use an internal low frequency oscillator to automatically sweep notches in the frequency response up and down the frequency spectrum. An important difference between phasing and flanging is that phasers space these notches evenly across the frequency spectrum, while the notches in flanging and chorus are harmonically (musically) related.
You don't hear the notches as such (because they are the frequencies that are removed); what you hear is the resulting frequency peaks between these notches. Early phasers did not provide any feedback, so the original effect was quite subtle; ideal for textural rhythm playing.
Phasing works by mixing the original signal with one that is phase shifted over the frequency spectrum. For example, a four stage phaser signal could be from 0 degrees at 100Hz, shifted to 720 degrees at 5Khz (these extremes are not quite possible practically, but are near enough to explain the effect). This is how the term phase shifter comes about.
Where the signal is in phase (at 0 degrees, 360 degrees and 720 degrees) the signals reinforce, providing normal output. Where the signals are out of phase (180 degrees and 540 degrees), they cancel each other, giving no output at these frequencies. Constantly varying the frequencies where these cancellations occur, gives the movement associated with phasing.
Adding resonance enhances the frequency peaks where the signals are in phase. A 4 stage phaser has 2 notches with bass response, a central peak, and treble response. By using resonance to enhance the central peak, you can get a sound similar to an automatic wah.
Each phaser stage shifts the phase by 180 degrees, so a 6 stage phaser gives a shift of 1080 degrees, providing 3 out-of-phase frequency notches along the way. Designs with 4, 6, 8 and 10 stages were common, although each stage adds noise to the final output.
Using a phaser with lots of stages and setting the resonance high can give a sound similar to flanging, although they are really quite different.
The controls common on a phaser are:
The Univibe (made famous by Hendrix) is an early implementation of a phaser. A phaser uses matched FETs to control the changing frequency response, while the Univibe used incandescent light bulbs and light dependent resistors, giving a more erratic, somewhat pulsating phaser sound.
Compressors are commonly used in recording to control the level, by making loud passages quieter, and quiet passages louder. This is useful in allowing a vocalist to sing quiet and loud for different emphasis, and always be heard clearly in the mix.
Compression is generally applied to guitar to give clean sustain, where the start of a note is "squashed" with the gain automatically increased as the note fades away. Compressors take a short time to react to a picked note, and it can be difficult to find settings that react quickly enough to the volume change without killing the natural attack sound of your guitar. It works like someone adjusting your volume control while you play - turning volume down when you pick a note, then turning the volume up as the note fades out.
This diagram shows all dynamic effects:
The compressor drawn here shows that gain is reduced above a threshold level. In practice, overall gain is increased to make-up for the lower maximum level, so this has the effect of boosting lower levels which is perceived as longer sustain. Typical stompbox pedals implement this effect simply by applying boost that increases as signal decreases.
Common controls are:
I have a lot more background on these effects on my Amplifier Overdrive page. I will summarise here by saying that these effects are intended to produce the sound of an overdriven amplifier, pushed well into its clipping region.
There is an enormous array of pedals available today, tailored for different markets. The first commercial designs were fuzz boxes and produced a thin (lots of bass-cut) buzzing tone. However, later designs were aimed at a natural overdrive sound, and these are still popular, whether used for their overdrive tone, or as a relatively clean booster to push the amplifier into overdrive. Later pedals have been tailored to heavy rock, metal, blues, grunge, retro, and so on.
Smooth overdrive and distortion effects were born from the many fuzz-circuit designs of the 60's. A wide variety of methods that contorted a guitar signal were marketed under the generic description of Fuzz. One of the most popular was the Fuzz Face as used by Hendrix, while the most useless was probably a Schmidt-trigger design that only worked monophonically (one note at a time) producing a synth-like squarewave.
Towards the end of this era, the back-to-back diode pair became popular as a technique to provide soft clipping (with germanium diodes) and hard clipping (with silicon diodes).
Today, overdrive effects usually means soft clipping, where gain is reduced beyond the clipping point, while distortion usually means hard clipping, where the level is fixed beyond the clipping point. Distortion is a little harder sound, good for rock, while overdrive gives a more natural sound.
A common variation is called asymmetrical clipping, where one side of the wave is clipped more than the other. This just gives the final waveform a slightly different sound, but regardless of the method used, the more overdrive, the more they sound alike. Of course, real guitar signals are not pure sine waves - I've just used those to demonstrate how clipping works.
Usual controls are:
Some companies label their controls with terms they think their customers will relate to, such as Grunt, Guts, In-Yer-Face, and so on. I'm not sure whether I'm amused or insulted. I'm definitely confused!
Some classic overdrive pedals are:
These effects are designed to give more tone control than is possible with the basic amplifier bass, middle and treble controls. There are 2 common varieties; graphic and parametric.
Graphic equalisers use sliders to control the level at fixed frequencies, called bands. These provide a graphic representation of the overall frequency response. The bands are usually logarithmically related, meaning that each frequency is always a fixed multiple of the next lowest frequency. This corresponds to the way our ears perceive frequencies, including notes in the scales we use.
This diagaram shows 10 bands, each of which can be boost or cut between the extremes shown:
The total frequency range can be limited to suit particular instruments, such as bass or guitar, or it can cover the entire audible range from 20Hz to 20khz. Additional bands give you finer control, but require more adjustments to make broad changes.
Parametric equalisers generally provide a bass and treble control that work as normal tone controls to allow broad shaping. They have one or more middle controls, each offering:
This diagram shows the same type of boost and cut as a graphic equaliser, however, a parametric allows you to select the frequency you want. You can also increase the Q to affect a narrower band of frequencies. Q is sometimes labelled resonance (again higher resonance means a narrower frequency band). Q can also be labelled as bandwidth, in which case a higher setting affects a broader band of frequencies, which is probably more logical for us musicians.
Both equalisers often include a level control to allow you to compensate for any overall loudness changes made by the tone changes.
The graphic is probably the easiest and most intuitive to use, but if you need to fine tune problem frequencies for feedback, or acoustic guitars, a parametric is more useful.
These add one or more notes to what you are already playing, and come in several variations.
The first harmonisers were octave dividers, which added a distorted signal one or more octaves below your playing. These only worked on a single note at a time, and are still interesting as a vintage effect, but I think it's fair to say that they are not going to change the world.
Modern harmonisers use digital storage and retrieval techniques that preserve the tone and timbre (character) of your playing. It is still easier to provide monophonic (single note) harmonies, so several models also offer this as an option with improved accuracy and/or quality. Monophonic mode is readily applicable to vocal and solo instrument harmonies as well. For guitar, you will sometimes want polyphonic harmonies to allow things such as pitch shift and 12 string emulation on chords.
You can set the harmonies to be fixed interval, such as up 5 semitones, or down 7 semitones. Many harmonisers now offer "intelligent" chord based harmonies, so the interval is determined by a key you set, and the note you play. You could set harmonies to be a 3rd and 5th, in the key of C major, and the harmony intervals will change to always play in C major.
Advanced options allow you to set your own chord intervals, and even apply random pitch variations or corrections to add extra realism to vocal harmonies.
This is a relatively new digital effect, designed to emulate a whammy bar. You can set how far and how fast pitch is bent, and how long it takes to return to normal.
Often these effects are combined with other pitch effects such as vibrato and some basic harmoniser options.
Vibrato varies the pitch smoothly between slightly flat and sharp, similar to the fingerboard technique of string bending, or wiggling the whammy bar. Of course, you can't bend a string flat!
Fender amps have an effect labelled vibrato which is actually volume modulation, or tremolo (see below). I have read that Fender originally did provide pitch modulation (true vibrato), but later changed to volume modulation to suit the "surf sound". When they changed the effect, the amp labelling remained as vibrato. I don't know if this story is true; I've never seen a Fender amp or even a Fender schematic with true vibrato.
Common controls are:
Flangers mix a varying delayed signal (usually from less than 1 millisecond to a few milliseconds) with the original to produce a series of notches in the frequency response. The important difference between flanging and phasing is that a flanger produces a large number of notches, and the peaks between those notches are harmonically (musically) related. A phaser produces a small number of notches that are evenly spread across the frequency spectrum. The short delay used for flanging is usually set too short for the extra signal to be perceived as an echo.
Flangers, Phasers and Choruses each produce a series of notches in the frequency response that are modulated across the frequency spectrum. The notches correspond to no sound, so except for a little tremolo (pulsating volume), we don't really hear the notches; we hear what's left which is a series of peaks.
This diagram is an actual calculated response of a 1 millisecond delay with an equal mix of dry (the chart is 20Hz to 20KHz plotted at quarter-tone intervals). Even at this resolution, the chart doesn't completely show the detail at higher frequencies, however, you can see that the notches occur at harmonic multiples instead of being evenly spread across the frequency response like a phaser.
Most flangers provide a resonance control to use internal feedback to enhance the peaks in the frequency response. Flanging got its name from a trick used in recording studios where the same track was played on 2 reel to reel tape machines, and recording engineers gently touched the flange of one tape reel to produce a small delay between the machines. Then, by touching the flange of the other reel, they would bring the machines back into synchronisation again, removing the delay.
With low resonance, the effect is similar to the original studio trick With high resonance, you get the "jet plane" effect.
Common controls are:
True vintage chorus works the same way as flanging. It mixes a varying delayed signal with the original to produce a large number of harmonically related notches in the frequency response. Chorus uses a longer delay than flanging, so there is a perception of "spaciousness". The delay is usually on the verge of a perceived echo: with some settings echo won't be noticeable while longer delay settings can make give a distinct slap-echo effect as well. There is also little or no feedback, so the effect is more subtle.
There are several variations of stereo chorus that are effective in providing a powerful "surround-sound" effect through a stereo system. The most common arrangement is to have a separate delay for each channel, and while the delay is increased in one channel, it decreases in the other, and vice-versa. These delayed signals are mixed with the original in each channel, and sometimes a small amount of delayed signal is applied in the opposite channel with bass cut.
Common controls are:
The original chorus effects used BBD (bucket-brigade device) technology which moves analogue charges (representing your audio signal) though a series of steps causing the delay, when tit it then rebuilt into an analogue signal. Although these were ground breaking devices in their day, their audio quality was poor and many tricks were used to get the best from them, such compressing the signal before the device and expanding it afterwards. Also, treble response was limited, sometimes dynamically with delay times. Some players feel these measures contribute to the overall effect, but they were really just measures to get the best from the available technology.
Early digital processors produced chorus in a different way, which provides a stronger chorus effect, but also adds a small out-of-tune effect. It is produced by mixing the original signal with one that is modulated slightly flat then sharp. Personally I don't like them at all, but they've been so commonly recorded now that many people have forgotten what vintage chorus actually sounds like.
Noise gates are used to electronically turn the volume down when you're not playing, so you don't hear the noise produced by other effects. Those with high gain, such as overdrive and compression can be especially noisy. So to work at all, noise gates MUST be placed after the effects producing the noise.
They work by detecting the signal level, and then slowly fading down the volume while your playing level fades away. This prevents notes that are fading naturally being cut off dead. All noise gates need to respond as quickly as possible to a new note after they have turned down, so there is rarely a control to set how fast you want the turn-on time to be.
With very noisy effects, it can be hard for the unit to separate the signal from the noise. It is usually better for the level detector to have its own input, which you would feed direct from the start of the effects chain. This feature is more common on rack multi-effects units.
There are more sophisticated noise gate units that offer additional noise reduction techniques, such as treating the bass and treble components of the signal separately, offering minimum volume and tone settings, etc.
Common controls are:
These devices are used to limit the maximum volume. They have no effect on signals below the threshold level sets, but hold signals above that level at a fixed level. This effect is similar to a compressor reducing high volumes, but a limiter does not boost low level signals.
These are commonly used in PA systems to prevent overloading the power amps and/or speakers. They can be useful in guitar systems for simulating valve power amp dynamics in a solid state system, but they really are not as good as "the real thing".
Common controls are:
Not much to explain here. The pedal controls the volume. Some pedals allow you to set a minimum volume, so you can always be assured of something, even with you heel fully down.
Something to watch for is whether you can walk away from the pedal with it set at some specific volume, without it falling on its own to maximum volume.
If you always use the pedal after some other effect that uses electronic switching (or in the send/return loop from your amplifier), you will probably be best served by a medium impedance pedal (say, 50K). On the other hand, if you need to use the pedal straight after your guitar, you will need to use a high impedance pedal (at least 500K).
This modulates the guitar volume, like rapidly turning the volume control up and down. When used with reverb, you can hear the surf guitar sounds of old. Fender incorrectly label this sound as Vibrato on their amps. Different effects have different wave-forms to modulate the volume level. The originals pretty much used sine waves, which gives a smooth effect. Other offer choices, such as saw wave (slightly less of a pulsating sound), square wave (which just turns the sound off and on very quickly), and other interesting variations.
Common controls are:
Panning is just like 2 tremolo effects, one for the left and one for the right channels. They are linked so that when volume is high in one channel it is low in the other, and vice-versa. When connected to a stereo system, the sound "moves" from one side to the other.
Common controls are:
A typical guitar speaker box is not designed to faithfully reproduce the sound presented by the amplifier. Unlike hi-fi systems and front of house systems that strive for a wide bandwidth and uncoloured sound, guitar speaker boxes are an important part of the sound creation process.
Without a speaker simulator, you are likely to get the best guitar sound through front of house by using one or more microphones around your guitar amp. The quality of speaker simulators varies enormously to my ears. All simulators apply a general guitar speaker response where lows are rolled off gradually while highs are cut dramatically above about 6KHz. Good speaker simulators emulate other cabinet frequency response characteristics such as general low and mid biases as well as detailed peaks and notches usually above about 1KHz.
Common options are choice of cabinet type and speakers, closed or open back, microphone types and positions, and a mix of direct vs simulator. Digital emulators offer choices between specific types of guitar cabinets and possibly specific models.
Delay is an echo effect that replays what you have played one or more times after a period of time. It's something like the echoes you might hear shouting against a canyon wall.
The original delays, like the legendary Watkins Copy Cat, were tape machines running a loop of tape that recorded your playing. The sound was replayed through one or more replay heads positioned further around the loop, then ultimately erased, ready for the next recording. By varying the mix from different replay heads and the speed of the tape, you could get a wide variety of delay effects. You could even set up different rhythm patterns in the delays! These units suffered some problems, mechanical ones with broken tapes, head alignment was important, and they were quite noisy as well.
Modern delays are digital, where your playing is stored in memory, and retrieved at some later time. Common controls are:
Other popular controls are tone, often used to cut treble response of the delay so it does not distract too much from the main playing. More sophisticated units offer multiple taps, like the multiple replay heads on older tape units, with options to position taps anywhere between left and right output channels for interesting stereo effects. One neat stereo effect is ping-pong delay where the repeat sounds as if it "bounces" from left to right as it fades out.
Using a single delay set to a short delay (say 50mS) at nearly the same level as the original gives you the doubling effect, because it sounds like two players playing the same thing in near-perfect unison. By increasing the delay a little more (say 100ms) you get a slap-back echo effect.
Reverb is the sound you hear in a room with hard surfaces (such as your bathroom) where sound bounces around the room for a while after the initial sound stops. This effect takes a lot of computing power to reproduce well. Reverb is actually made up of a very large number of repeats, with varying levels and tones over time. Reverbs usually offer you a choice of different algorithm to simulate different environments such as different sized rooms and halls, studio effects such as plate, chamber and reverse* reverbs, and sometimes emulations of guitar spring reverbs.
These algorithms serve as a good starting point for the more basic controls:
Sophisticated reverbs give you control over a large number of reverb parameters, such as:
* Reverse reverbs were initially used by Phil Collins for his legendary gated snare sounds, where the reverb actually builds in intensity before cutting off abruptly. Almost all reverbs offer this effect, and yet none is allowed to give Phil any credit for it (otherwise they'd have to pay royalties, I guess). I don't sell reverbs, so I don't mind saying "Thanks, Phil!".