Last month, we talked about the nuts and bolts (parameters) that you’d find in a typical delay, whether it be the plug-in or the hardware variety. This month I have some tips and techniques that you are free to borrow if you promise not to tell anyone about them. The editors of FRONT of HOUSE are already sworn to secrecy. But before we get into specifics, we need to mention one last characteristic of delay.
Tap Tempo
Tap tempo lets the user set delay time by tapping along with the beat on either a hardware or a virtual button. Older hardware delays calculate the time between your taps and set the delay to that interval. More versatile delays let you choose a note value that the delay “interprets” from your tapping. For example, if you set the note value to eighth notes, the delay analyzes the tempo of your tapping (it assumes the taps are quarter notes) and then calculates a delay time to generate eighth-note repeats. A lot of digital consoles with onboard effects provide the ability to assign a user key to tap tempo, making it easy to adjust delay time — simply by tapping on the button (with no need to access an effects menu). It’s nice to have a delay that rhythmically agrees with the song. However, echoes that fall precisely on the beat can sometimes be masked by other instruments, such as kick and snare.
And Now… The Secret Sauce
Here are a few things that I do in prep for a live show. First I create a few presets to use for the main vocal delays. Typically I’ll do four presets to cover a range of song tempos: 150 milliseconds, 205 ms, 265 ms and 305 ms. I use the shorter times for up-tempo songs, and 265 ms or 305 ms as the main delay for the lead vocal. The two shorter delays get little to no feedback so that they stop short; the two longer delays have 3 to 5 percent feedback, giving me about two or two-and-a-half repeats. If the delay is stereo, I’ll offset one side by 10 to 15 ms to add a bit of stereo width.
Here’s the secret when adding delay to lead vocal: You want to create a sense of space and add a few repeats, but you don’t want to turn it into a mess by letting it get in the way of the dry lead vocal, especially when the mix becomes dense. For that reason, I EQ the delay fairly aggressively, using either an FX return, an extra channel or (given the ability) inside the delay itself. Specifically, I often dump the low-end below 300 Hz or so and chop off the high-end above 3 or 4 kHz. Sometimes I’ll add a broad, mild boost around 1 kHz (see Figs. 1 and 2). What this does is band-limit the delay so that it (a) makes the delay sound distinct from the dry lead vocal and (b) takes up less space in the mix. I use less delay when it’s EQ’d like this because it’s easier to hear — as opposed to using a full-bandwidth delay that steps on the dry vocal as well as the guitars and keyboards.
Quack
Next, if the delay has the capability I might apply a bit of ducking to regulate its volume. Ducking automatically lowers the level of the delay while the dry signal is playing and allows the delay to get louder when the dry signal stops. Some delays have a parameter for this: you simply set a threshold and when the input rises above the threshold, the delay level is dropped at the output. Let’s say that you want a long rolling delay on a lead guitar. If that delay rolls around while the guitar is playing, the articulation of the guitar is difficult to hear, so set the delay to duck a few dB while the guitar is playing. When the guitar stops, the delay becomes louder, making it easier to hear the repeats.
If the delay does not have built-in ducking you may need to “build” one by patching a compressor after the output of the delay. Route a bus or an aux send from the guitar channel to the compressor’s key input. Set the threshold and ratio of the compressor so that every time the guitar plays, the delay is compressed by 4 to 6 dB. As each guitar phrase ends, the compressor lets go, creating the illusion of a huge delay. If you are doing this via plug-ins, make sure that the compressor is inserted in a slot below that of the delay (signal flow follows insert slots from top to bottom).
If you want a longer, multi-repeat delay for what I call a gag or special effect, crank up the feedback to 35 percent. If there is a parameter for number of repeats, try five. Once in a blue moon, I’ll crank the feedback up to 60 or even 65 percent to get tons of repeats — but beware that older hardware units like the Yamaha SPX990 may “take off” at such settings and feed back even when there is no signal present. Some engineers turn up the aux send on the source channel and leave it there so that the vocal (or whatever) is always feeding the “special” delay. Then they raise the effect return fader when they want to hear the delay. Others (myself included) leave the return at “0” (nominal) and throw the vocal channel’s aux send knob up to two or three o’clock when they need the effect. If the aux send has an on/off switch, you can preset the position of the aux send knob during sound check, and simply turn it on or off as needed.
Does This Delay Make Me Look Wide?
One trick that you might like is using a stereo delay to fatten up a lead vocal, not so much as a special effect but as an integral part of the vocal sound. Feed the lead vocal channel to a stereo delay. Set one side of the delay to 40 ms and pan it hard left. Set the other side of the delay to 60 ms and pan it hard right. The dry lead vocal remains panned to the center. On some vocals, this adds size and presence to the vocal without sounding like an effect is being used. Start with the feedback at 0 and slowly raise it. You’ll probably find that a few percent is enough; more than that and the delay may start to ring. In this case you can EQ the effect return less aggressively, maybe rolling off the low-end below 100 Hz and the high-end above 10 kHz (see Fig. 3). You’ll definitely want to sound check with this one because adding the effect will make the lead vocal significantly louder and it may not work equally well with all voices.
You can take the previous idea into interesting territory by elongating the delay times and increasing the feedback until you get a “ping-pong” delay. This effect makes a sound appear to bounce between the left and right channels and can be pretty cool when used tastefully (or even distastefully). Let’s say you set the left side to 250 ms and the right side to 500 ms. You’ll hear the 250 ms delay first, then hear the 500 ms delay on the other side. When feedback is increased the process repeats. Many algorithms will feed the output of the first delay to the input of the second, perpetuating the effect.
Delay may also be used to create a variety of modulation effects such as flange and chorus. For a complete look at those effects, check out T&P in the Oct. 2011 issue of FRONT of HOUSE. See you on the road!
Steve “Woody” La Cerra is the tour manager and FOH engineer for Blue Öyster Cult.
