HOW GUITAR EFFECTS WORK

� 2004 H. Davis

 

How we love things that make chords sing and notes take wing! The electric guitar revolutionized popular music a half-century ago, and it continues to be its greatest tool of expression. Through electronic technology our electrified beat enthralls the world with harmony and joy!

 

When I began designing guitar pedals I discovered much that is alien to engineers in the related field of high-fidelity audio. Their goal is "transparent" sound reproduction, as unaffected by the reproduction system as is possible. Seeking only to reproduce, not originate music, they have no concept of the amplifier and speakers as being parts of a musical instrument itself. The more I hung out with musicians and the more I learned to play guitar, the more obvious this concept became and the more I came to appreciate its importance. Your sound, all that you say musically, merely begins with your hands on the strings. Between those strings and your listeners' ears lies a vast universe of complex technology, few parts of which do not affect what is finally heard. Here is where art and science lose their distinctive identities, and where success lies in working with both as symbiotic media of expression. As with a truly great band, the product of the whole becomes greater than the sum of its parts.

 

I hope to convey here an understanding of how guitar effects work, without being overly technical. Musicians with a little technical familiarity should have no problem with this. If you find you often draw a blank, a book on basic electronics might be a good starting point.

 

The electric guitar creates its signal through its steel strings vibrating in a magnetic field. The resulting fluctuation of the field generates a small voltage in the pickup coils, and this is passed on to an amplifier, directly or through one or more effects pedals.

 

The simplest audio signal - a pure pitch, or fundamental - is called a sine wave, and depicting this signal helps to understand such concepts as clipping and phase shift

 

Let's say you are playing through a small amplifier. Not loud enough, you crank it up - but you soon reach a point where distortion sets in. The reason is that both the positive and negative voltage peaks of the signal are trying to go higher than the limited range the amplifier allows. At these limits the signal flattens out, or clips. So the first distortion "effect" was probably just an overcranked amp, or perhaps a cheap radio or record player that someone hooked their guitar into.

 

Fuzz is what this distortion is called when the clipping is hard - not gradually increasing as the signal level increases, but appearing all of a sudden at the maximum level the circuitry allows. When the distortion comes on more gradually with increasing level, rounding the signal peaks rather than chopping them flat, it is termed overdrive distortion. Compared to harsh, buzzy fuzz, this has the softer, mushier sound typical of a good overdriven tube amp. A properly designed solid-state overdrive pedal like the Pigtronix Polysaturator can emulate the sound of tube distortion without the problems of reliability, fragility, inefficiency, size, weight, and cost that are the drawbacks of tube equipment.

 

Limiting in its simplest form is the result of the limited voltage range that leads to clipping. Above a certain point the sound can get no louder because the signal voltage can go no higher. Amps produce limiting with very audible distortion, but limiters are electronic devices designed to limit the maximum signal level without perceptible distortion.

 

Compression is similar to limiting, but rather than cutting in at a certain level and preventing it from getting any higher, compression operates over the entire dynamic range, from no signal to the maximum level possible. With compression you still have volume changes in the output when the input level changes, but they are reduced in magnitude. A change of say 6db, which would be very audible, will come through as a change of only 3db if the compression ratio is two to one. Thus compressor pedals are useful for increasing sustain, or if you have a tendency to play some notes or chords too loud or too soft. These pedals usually have high compression ratios that make everything you play come through at about the same level. They often have a threshold control that allows you adjust the signal level at which the compression begins to take effect. The opposite of compression is dynamic range expansion, but there are no pedals that do this other than internally for noise reduction purposes. Both compressors and expanders operate by sensing the signal level and adjusting their internal gain (amount of amplification) to achieve the desired effect.

 

Actually there are pedals that do expansion, but not in a recognizable way. These are the noise gates. They operate by greatly decreasing their gain when the signal falls below a certain level. This level usually can be adjusted with a threshold control, as with a compressor. When you stop playing and damp the strings the signal drops below this threshold, and the gain cuts off or goes way down so that the noise or hum from your pickups, cables, and preceding pedals is suppressed. If you set the threshold too high you will lose some sustain, so proper adjustment is important.

 

Distortion, compression, limiting, and noise gating are all effects that operate on or with gain. More sophisticated effects are obtained by controlling the gain throughout an entire note or chord with ADSR circuitry. ADSR is an acronym for attack, decay, sustain, release - the four defined and electronically controllable periods over which an individual sound can be heard. A normally picked guitar note has a fast attack, slow initial decay (same as a long sustain), and a release that is just a continuation of the slow decay. The sound starts with a sharp attack and then slowly dies away, which is typical of a plucked string or percussion instrument. But it can be made quite different electronically, different to the point of not even being recognizable as a guitar. The vintage Electro-Harmonix Attack Decay pedal and the recently designed Pigtronix Attack Sustain allow independent control of attack and decay times, making possible a realistic tape-reverse effect, toy piano sound, bowed string effects, percussive and horn-like sounds, and weird slow-attack synth effects - all from your electric guitar or bass! The technical term for these effects pedals is envelope synthesizers.

 

The oldest gain modification effect was around before the first guitar pedals ever came on the market, and was a feature built into the amp itself - tremolo. A low frequency oscillator (LFO) in the amp generates a sine wave at an adjustable subsonic frequency, typically from one to ten cycles per second (usually termed Hertz, or Hz). This is then used to amplitude modulate (vary the level of) the guitar signal. Some pedals allow modulation with square or triangle waves as well to obtain different modulation effects.

 

There are pedals that synthesize a second harmonic (octave up, or frequency-doubled sound) such as the Octavia or the Pigtronix Disnortion, or a subharmonic (one or two octaves down) such as the EH Deluxe Octave Multiplexer. These first generate a clean signal with the fundamental frequency of the note being played. They then process it to produce a signal rich in the second harmonic, or divide it down with digital circuits called flip-flops to generate the first and possibly second subharmonics, which are 1/2 and 1/4 the fundamental frequency. Mixed with the dry (unmodified) signal, subharmonics can make a guitar sound like a bass guitar or even a thundering pipe organ. As these devices must extract the fundamental pitch of the note, the clean playing of single notes is necessary for proper operation. They do not work with most chords.

 

Filters in general are circuits that alter the frequency content of signals passing through them. Lowpass filters allow frequencies below their designed cutoff frequency to come through, while reducing the level of higher frequencies. The higher a signal's frequency is above the lowpass cutoff point, the more it is reduced, or attenuated. Likewise there are highpass filters, and bandpass and bandstop or notch filters that only operate on a limited part of the frequency spectrum. There are also allpass filters that do not affect the frequency response at all but cause a frequency-dependent phase shift. More about this later.

 

You must be thinking that filters of all sorts are useful for guitar effects. Damn right they are!

 

The wah-wah is among the earliest of filter-based guitar pedals, and probably gave the generic name pedal to stompboxes in general, including those not footpedal operated. In its simplest form, the wah-wah is a lowpass or bandpass filter whose cutoff frequency is controlled by a footpedal-operated pot.

 

That footpedal is a pain-in-the-kiester to operate while playing onstage. So why have it at all? Maybe someone can come up with a little electronic musician-in-the-box that will work it for you while you play, leaving you free to dance around, kick your drummer for missing a beat, and dodge anything the audience may throw. That clever musician-in-the-box is called the envelope follower.

 

The envelope follower is basically a rectifier circuit, one that changes the AC voltage of the guitar signal to a varying DC voltage whose level follows that of the signal. Instead of having that wah-wah filter operated mechanically, it can be designed to be voltage controllable, and controlled by the envelope follower's output. Thus was born the auto-wah and a wide variety of filter-based effects that funkily sweep themselves up or down as you play. The Electro-Harmonix Q-TRON is one example of this.

 

The phase shifter or phaser is a unique filter-based guitar effect. If we add two sine waves of the same frequency that are in phase - both reaching their positive and negative peaks at the same time - the sum of the two is another sine wave of an amplitude (level) equal to the sum of the two. If however we add two sine waves that are 180 degrees out of phase, with the positive peak of one corresponding to the negative peak of the other, they subtract in level, and if equal in amplitude they completely cancel each other out, or produce a null. An allpass filter does nothing to signal amplitude, but it causes a phase shift from zero to 180 degrees as the signal frequency varies from well below to well above the filter's designed center frequency. At this center the shift is half the maximum, or 90 degrees. Now suppose we put two allpass filters in series, and mix their output with the unmodified input signal. At frequencies well below the center frequency the dry and filtered signals are in or close to in phase, so their sum is about double the voltage of the input, assuming unity gain (no gain or loss in level) through the filters. The same is true at frequencies well above center, as there the total phase shift is 360 degrees, so the signals are again in phase, a 360-degree total shift bringing the phase relationship full-circle around to where it started. But what about at the center frequency, where the dry and filtered signals are 180 degrees out of phase? Here we get a null, a cancellation - nothing! A notch is produced in the overall frequency response, and this is the basis of the phaser effect - alternating peaks and notches in the frequency spectrum.

 

How does the sweep effect of the phaser come about? A little guy inside the box pushes the notches - - -

No. Actually, as with other filters, allpass filters can be made electronically controllable. An LFO with an adjustable sweep rate controls the filters, and the effect sweeps up and down as the filters do. Most phasers have two or three response notches, corresponding to four or six stages of allpass filtering. It is the peaks you actually hear, not the notches, but without the hole, would you have a donut? The feedback ("color") feature many phasers have is an enhancement of the peak and null response brought about through positive feedback of the processed signal.

 

An entire family of guitar effects is based on time delay - echo, chorus, vibrato, reverb, flanging. The first delay devices were loops of tape or other recording media that first passed over a recording head, then one or more playback heads. The delay time is determined by the distance between the heads and the tape speed. Multiple repeats are produced by feedback from the playback head to the recording head. Tape echo units are large and heavy, and due to tape breakage are not very reliable, but as with tube equipment some people like them and still use them.

 

The 1970s brought the innovation of analog delay chips, and revolutionary new echo stompboxes like the Memory Man hit a market that was begging for them. Rather then being recorded on tape, the signal enters a long series of charge storage cells in the chip. These pass the charge packets, which are actually samples of the signal voltage, at a rate determined by a clock oscillator in the same way that a "bucket brigade" passes buckets of water down the line to put out a fire. The delay time is determined by the number of charge storage cells in the chip and the clock frequency. Increasing the clock frequency moves them along faster and thus decreases the delay time. Multiple repeats are produced by feedback, just as with a tape delay.

 

As the clock oscillator controls the delay time, it allows various effects to be produced by modulating its frequency. When a source of sound is moving towards you, the Doppler effect causes a perceived increase in pitch, and the pitch likewise drops as it moves away from you. In the same way, the pitch of the delayed signal increases during the time in which the clock frequency is increasing, and likewise the pitch drops as it is decreasing. Modulate the clock with an LFO, and lo and behold, we have vibrato - a pitch (frequency) modulation of the signal. At low modulation levels, vibrato and tremolo sound the same.

 

Want chorus? Use a short delay, modulate it at a very slow rate to produce a little pitch shifting, and mix it back with the dry signal. Now you have a simulation of the slight timing and pitch variations between the several players or singers in a group. The effect is full and rich, and pedals with stereo chorus outputs can sound awesome when used with two separate amps. The double tracking or slap echo effect is the same as chorus, but without the modulation of the delayed signal.

 

Reverb is the result of many echoes with differing delay times that feed back upon each other just as sound does in an acoustically reflective hall. The earliest reverb device, aside from an actual live room echo chamber, is the spring reverb still used today in amps. A transducer converts the signal to vibrations that reverberate in the springs and are converted back to an electrical signal. Springs and other mechanical reverb simulators do a good job, and with proper equalization (frequency response correction), a very good job. Their chief drawback is they are microphonic - sensitive to mechanical impact and vibration. Kick that amp and hear them scream! Electronic reverb pedals exist, but opinions differ on their quality and advantages over spring reverbs. Reverb is about the only area in which digital technology is more cost-effective than analog for stompbox guitar effects, and here also an old electromechanical device - the spring reverb - still holds its own.

 

Ever hear a phaser that sounded more like a flanger than other phasers do? It probably had three rather than the more common two response notches. A flanged signal has a frequency spectrum pattern similar to that of a phaser, but with many more peaks and nulls that are spaced more closely together at the higher frequencies. Rather than using phase shift, the flanger introduces a short time delay and mixes the delayed signal with the dry signal. The amount of delay determines the number and location of the peaks and nulls. Modulate the delay, and the flanging sweeps up and down just like phasing does. The first flangers were two synchronized tape recorders playing the same program material at the same time. By slightly slowing one of them with one's hand on the flange of the tape reel, a small and varying time delay was introduced, and thus was born the flanging effect. Do any old purists still flange with tape recorders? Vacuum tube-based iron-wire recorders, perhaps?

 

Ring modulation produces a weird, sometimes chime-like effect that you recognize immediately if you've ever heard it before. The music signal undergoes a suppressed carrier modulation with the signal from an adjustable local oscillator in the pedal. This process is like the amplitude modulation of a radio signal, with the carrier frequency (the local oscillator signal) being filtered out, leaving only the sidebands in the output. Sidebands have frequencies that are the sum and difference between the frequencies of the oscillator and the incoming signal. These are not harmonically related to the note played, but with proper adjustment of the local oscillator frequency they can be made to have a musical relationship to some of the notes. The Pigtronix MOTHERSHIP GUITAR SYNTHESIZER is a breakthrough in modulation technology I am proud to have invented. This has the first tracking ring modulator, an "intelligent" ring mod that follows the notes you play to always maintain the same musical relationship between the note and the sidebands.

 

Very deep bass can be produced by a ring modulator. If the local oscillator is set at 20 Hz above or below the note played, a 20 Hz note is heard (or felt), as this is the difference between the two frequencies. This pitch is two octaves below the low E of a guitar, one octave below the E of a bass. Good bass speakers driven by solid-state amps are essential to take advantage of this.

 

Actually, good speakers are essential, period! They are often the weakest link in the audio chain, especially if you like clean and heavy bass, understandable vocals, crisp and clear highs, or effects that rely on extreme low or high frequency production. Most amp cabs have no tweeters, and open backs waste low bass energy. Sheer volume is no substitute for tonal quality. Just as audiophile product designers can learn from us music guys, it's time for vice-versa.

 

Home