Calculate delay and set sound propagation time correctly

Set audio transit time: Everyone is late at some point, but it rarely goes down well and quickly puts everyone involved and colleagues in a bad mood. Sound engineers have it easier there, because sometimes the sound has to wait a bit. Of course, this workshop is all about decentralized audio amplification, delay lines and time adjustment. We explain how you can adjust your signal operation times in a meaningful way.

What are sound propagation times?

Time flies when you get out of the armchair and go to the fridge for a TV night. The sound must also have time to travel from the sound source (speakers) to the receiver (audience). The speed of sound is 343 meters per second at a temperature of 20 degrees Celsius. If we divide 343 meters by 1000, we know that sound travels 34.3 cm per second. millisecond. In practice, the rule of thumb is: 1 meter = 3 milliseconds. That’s enough to get most runtime problems under control.

A laser range finder provides accurate readings quickly.
A laser range finder provides accurate readings quickly.

Sound transit times in a box

There is hardly a box that is a perfect point sound source, where all drivers radiate at the same time and are therefore on the same time axis. In some boxes, the woofer radiates out in a long horn, while the tweeter radiates out in a short. This means that high frequencies leave the box before the bass. To compensate for this runtime offset, the drivers are matched to each other via the controller in the amplifier rack or the DSP on an active box via a delay, so that they sound as simultaneous as possible. This is not only necessary in a box, but often also in a complete sound system if it consists of several boxes.

Driving times between top parts and subwoofers

What are the most important criteria when buying a house? Right: location, location, location. The same applies to the sound positioning of a PA, especially with regard to the position of the individual boxes in relation to each other. If you place two peaks to the left and right of the stage for a little speech amplification, you don’t have to worry about run times. The same applies to a conventional satellite PA (subwoofer, short distance bar, top section)

However, it becomes problematic when the bass and top parts are placed at a greater distance from each other. An example: the tops are on stands to the left and right of the stage, and the subwoofers are in a mono cluster in the center of the stage. The audience in front of the stage therefore perceives the subwoofer more than the upper parts. You should therefore delay the subs to the peaks, otherwise unwanted audio artifacts will occur, especially in the crossover frequency range.

measuring systems against Loss of sound due to cancellations

Let’s assume that bass and treble are separated by 100 Hertz, then this frequency range is reproduced by both the bass and treble. The different driving times can lead to sound loss due to cancellations. If you delay the subwoofers to the top, the sound will be much tighter. But how to measure the distance and what value should I enter? For the ambitious lecturer, a measurement using a measuring tape or a cheap laser distance meter (around 30 euros) is usually sufficient.

Professionals use measurement systems with software such as Smart or Easra Systune to determine run times, correct them and ensure correct phase response. Such a setup is quite expensive and requires a lot of experience. A simple runtime correction using distance measurement, on the other hand, is still audibly better than nothing.

So we measure the distance between the basses (left bass to left peak, right bass to right peak). Let’s say the tops and subs are five meters apart. We remember the rule of thumb: 1 meter = 3 milliseconds and choose a delay of 15 milliseconds for the subwoofer output of the speaker controller or active module.

Almost all current digital mixers allow output buses to be delayed via line delay

Running time between PA and delay speakers

Sometimes a normal PA on the left and right side of the stage is not enough to cover the audience area evenly. Balconies, galleries or utility areas then require additional, decentralized sound reinforcement, also known as delay lines. But how do you recognize that additional boxes are required? A clear indication is the lack of line of sight. Acoustics behaves in the same way as optics: angle of incidence = angle of reflection. This means that if I can’t see the PA, I won’t hear it well either, except perhaps for the ubiquitous bass frequencies. In very reverberant rooms far away from the main system, I might be able to see the PA, but loud reflections prevent good speech intelligibility.
In closed venues, a delay line becomes necessary no later than when the direct sound from the main system and the room reflections are in a 50:50 ratio. Outdoors, there are hardly any problems caused by reflections. Here, the main system’s range and audience volume are the decisive factors. At the latest when the howl factor approaches the level of PA, it’s time for delay lines.

Sound propagation time 3 ms

However, it is easier said than done to set up extra boxes. As we know, sound needs 3 milliseconds to cover a distance of one meter. The audience at the back of the hall first hears the sound through the additional speakers and only then through the main system. This creates unsightly comb filter effects or – depending on the distance – significant delays, which in turn sabotage speech intelligibility. The solution: The additional boxes are delayed on the main system. The delay corresponds to the time it takes the sound to travel between the main PA and the delay line.

Propagation delays are an integral part of any PA controller

Specific product recommendations for timing adjustment and delay lines

Place and set up delay speakers

To create a homogeneous, discrete sound pattern, delay lines should not be higher than the main system in their installation location: set up a delay speaker and walk back a few meters, the delay line is still turned off. Now we send pink noise to the front PA and look at an SPL meter like the Digital Sound 8928 (product link on thomann.de) set to “dB a slow”.

We turn up the volume to 84 dBa and then turn off the main system. The delay box is now activated and the volume increased until the SPL meter also shows 84 dBa. The delay line is only intended to support the main system and not to replace it. However, if necessary, we can run a well-established delay line up to 10 decibels higher than the main system without jeopardizing the placement of the main PA on stage.

A simple SPL meter helps set the volume of a delay line

As a rule, delay lines are set up without a subwoofer. But why is that? Subwoofers have the unpleasant property of radiating omnidirectionally, i.e. also in the direction of the main system. This in turn can lead to phase cancellations. If you absolutely want to use delay lines with a subwoofer (open air at a rock’n’roll arrangement), then these should at least be fixed, e.g. with a cardioid or endfire setup. If we are dealing with a speech event (lecture, comedy, etc.), we do without subwoofers and provide the delay boxes with an extra low cut.

Regarding the placement of a delay line, the following applies: You can almost never set up the delay lines where they would make the most sense. Emergency exits, walkways, immovable rows of chairs or blocked views – delay lines are almost always in the way. So compromises are required.

The crux of the matter: The precedent effect

Contrary to expectations, “The Law of the First Wavefront” or “The Precedence Effect” are not RTL thrillers. Rather, these terms describe a psychoacoustic effect relevant to sound engineers: if an identical sound signal arrives at a listener from different directions with a time delay, the listener perceives only the direction of the signal that arrives first. The delayed signal is assigned the direction of the first signal (the precedence). Helmut Haas found that with delays between 10 and 30 milliseconds, the sound that arrives first is crucial for localization, regardless of the direction from which the delayed sound comes.

The Haas effect has a practical use for our delay line: we add just ten more milliseconds to the set delay value, ensuring that the listener is guaranteed to hear the sound from the main PA first. The sound appears to be in one piece. If the level of the delay line needs to be raised for a short time because, for example, a lavalier microphone is already working at the coupling limit due to a quiet speaker, we also do not lose the precedent. The ear is deceived into thinking it is only hearing the faster main system. This Haas addition can be kept variable depending on whether the sound of the delay line or the localization is more important. My last authority is the ear: I always check the overall sound in several places in the audience.

Transit times between main PA and backline

As soon as the audience sound consists of stage sound and PA, we again have to contend with runtime differences. The sound from the PA usually reaches our ears faster than the crosstalk from the back row. Here, too, the law of the first wavefront applies, which sometimes confuses us as listeners: we see and hear the band on stage, but perceive the signals from the PA earlier, which directs our acoustic localization to the PA. The visual impression does not match the acoustic one. Solution: We delay the PA to the highest part of the backline.

Usually this is the drum kit or drum head. We measure the distance from the PA to the drums and again use the well-known rule of thumb. This approach is always a compromise. The back line is not always in line with the drums, and loud monitoring on the edge of the stage causes further chaos. Nevertheless, a PA delayed on the backline often provides a clear sound advantage.

Digital mixers and PA controllers also cause propagation delays.  The values ​​can be determined using a reference measurement

When I set the delay time, I usually proceed as follows: I adjust the level of the PA to the stage volume, so that both sources are about the same volume. Then I activate the runtime delay. It’s not uncommon for the overall sound to become more defined and fatter as a result. Sometimes the effect is minor or barely noticeable. It’s worth a try. If there are other digital devices in the signal chain (digital mixer, PA controller, etc.), then these already create a small delay (usually between 2 – 4 milliseconds), which we should also take into account. So take your time and give it a try.

More interesting content

Leave a Comment