Delay: A Guide to Making Things Happen Later
The varieties of delay and how to use them.
By Ken Pearsall
Thirty milliseconds, give or take. That's the measurable cut-off for when a human can distinguish one sound from another.
Though we don’t hear delay effects clocking in at less than 30 milliseconds as distinct and separate sounds, we do hear them, and they can be used to thicken a signal or add ambience via effects like doubling, phasing and flanging.
Here, we're going to focus on what happens when a signal is delayed 30 milliseconds or more.
What is Delay?
Simply put, delay is making a sound repeat at a later time. The methods of doing this vary, but the three most common are acoustic delay, mechanical delay and electronic delay.
Acoustic delay is the speed that sound waves travel through the air. This is what allows us to yell bad words into the Grand Canyon and hear them come back a few seconds later. You may know this kind of delay as “echo". You will also notice this playing your guitar through an amp with no effects in a large room.
Without getting into complicated math, we can use the formula that speed travels about one foot per millisecond. So if you play a note and mute it quickly, you can hear a short slap echo; this is the time equivalent to the distance from your amp to the far wall and back to you. In a lot of rooms, these echoes will be so close together it can sound more like reverb—a good example would be Mark Hollis’ guitar on “Runeii” by Talk Talk from the 1991 album Laughing Stock, which was recorded in a giant studio room that used to be a church.
Mechanical delays are far more common. These range from delays that record the sound to a metal disc, oil cans, and most famously, analog tape. While we think of tape machines as electronic with magnetic storage, it is the mechanical aspect that can be used to create a delay.
Tape machines usually have at least three tape heads. One head is capable of erasing, another can record and playback, and the third is for playback only. The tape is played back at certain speeds measured in inches per second (or IPS). The delay time is equivalent to the time it takes one point of the tape to travel from the record head to the playback head. One unique use of this technique was employed by pioneering guitarist Robert Fripp (King Crimson, David Bowie) who expanded the idea into what he called “Frippertronics”. This tape delay was actually two tape machines. Sound was recorded on the first machine, and the tape was spooled onto the second machine for playback. The delay time in this case was changed by how far apart the machines were from each other.
Don’t have your own tape machine? No problem. There are many stomp boxes that simulate this effect quite well. You’ll also find it in most multi-effect pedals and even built right into a variety of guitar amps.
You can think of these types of delay as both the “analog bucket brigade” delays-refering to the vintage-style analog delay pedals-and the “DSP” digital delay types.
The so-called “analog bucket brigade” delays work very similar to tape delays. Instead of recording to magnetic tape, however, the signal is recorded to memory chips. The first chip passes its signal to the second, which passes it to the third, and so on. Instead of relying on physical placement of tape heads, the delay time is the rate at which the chips pass the signal from one to the next and how many total chips they go through.
Digital delays work in a similar fashion, but allow us to set very precise delay times. This means we can easily set the delay time to the tempo of a song and it will stay perfectly in time. There are few guitarists who do this as well as the Edge from U2. Check out “Where The Streets Have No Name”. The Edge is using a delay time that makes it sound as though he’s playing double the amount of notes and creating a huge sound that seems like many guitars at once.
Feedback and Modulation
If you listen to guitarists who use delay, you've heard them play notes that repeat multiple times before fading away. This is called feedback. Conceptually, it's the same ear-splitting idea you may be familiar with. On most delay pedals it will be labeled "feedback" or “repeats”. This circles the output of the effect unit back to the input, making the signal delay again. The more aggressively the repeats (or feedback) control is set, the more times it will get sent back through the effect to repeat before fading out. We can use this to create long trails of notes, make a guitar solo sound bigger or, with enough feedback, make a note last forever.
Modulation is another control we may see on delay effects. If you play your guitar through a delay with a lot of repeats, you may notice that the sound builds in volume, louder and louder. Modulation means “change”, and in this case, we use modulation to alter the pitch of the delay repeats or delay time slightly so our delay sound can blend better with our original guitar signal. This can also add a nice thickness to the tone or can be exploited at the extremes for more psychedelic expressions.
There are no rules about what you should or shouldn’t do with delay. Want to turn the repeats knob all the way up until it’s a wall of noise? Go for it. Want to crank the modulation knob until your guitar is a warbly sci-fi raygun? We salute you. Pink Floyd’s David Gilmour often used two delays set to different time signatures. Throw on a copy of “Run Like Hell” from The Wall (1979) and listen to two different delay times creating their own groove within the riff. There’s no right or wrong way to use delay, and you’re really only limited by your imagination.