Delay Time Chart

Path difference is the distance from the main loudspeaker to the listener, minus the distance from the fill loudspeaker to that same listener. Measure to a seat in the middle of the area the fill covers, not to the nearest or furthest one.

Setting the delay to the exact path difference is a common and correctable mistake. At exact alignment, the two arrivals are simultaneous and the ear splits the difference, placing the source somewhere between the platform and the fill. Adding a small extra delay hands localisation entirely to the first arrival, which is the main system.

Five common positions with the measurements taken as they would be on site. Main distance and fill distance are both measured to the same representative seat.

Sound travels faster in warm air. For an indoor sanctuary this is a refinement, but for an outdoor service where a delay tower is aligned at dawn and used at midday it is a real shift, and for an unheated building in winter it is worth checking.

They fix coverage, not level. A delay loudspeaker exists to reach seats the main system geometrically cannot, typically under a balcony where the overhang blocks the direct path, or beyond the distance at which the mains have run out of level. If every seat already has a clear line to a main loudspeaker at an adequate level, a delay speaker adds nothing but complexity.

They do not fix intelligibility in a reverberant room. This is the expensive mistake. If the words are unclear because the room has a long reverberation time, adding more loudspeakers adds more energy into the same reverberant field and makes the problem worse. Under a balcony this is often the reverse, because the overhang is acoustically dead and a local speaker genuinely helps. In an open reverberant nave, absorption is the answer, not another box. See reverberation time for churches.

Level matters as much as timing. A correctly delayed fill that is too loud still draws attention, because the precedence effect breaks down when the later arrival is substantially louder than the first. Aim to have the fill contribute just enough to restore clarity, typically sitting a few decibels below the main arrival at the listener. Set the delay first, then turn the fill down until it disappears, then stop.

Two delay rings need two delay settings. A long nave with fills at 60 feet and at 120 feet needs each ring timed to the mains independently, not the second timed to the first. Timing ring two to ring one compounds the Haas offset and pushes the rear of the room past the echo threshold.

It assumes the main system is the reference. Every delay on this page is measured against the arrival from the main loudspeakers. In a room where the platform source is largely acoustic, such as an unamplified choir or organ, the reference is the physical source rather than a loudspeaker, and the measurement starts from the singers.

It does not cover subwoofer alignment. Aligning a subwoofer to the main loudspeakers is a phase problem across a crossover region, not a path difference problem, and it is solved with measurement software rather than a tape measure. The figures here will not help.

It assumes a single coherent main arrival. In a room with widely spaced left and right loudspeakers, a listener under a balcony may be much closer to one than the other, so there is no single main arrival to align to. Time the fill to the nearer main, and accept that the far side will not be perfect.

Plan view distance is not acoustic distance. A loudspeaker flown 20 feet above the listening plane and 40 feet back is 44.7 feet away. Churches routinely measure along the floor and set a delay several milliseconds short. Measure the straight line path from the loudspeaker to the listener’s ears.

It says nothing about the processing you need. Delay has to live somewhere, usually in a loudspeaker processor or in the output section of a digital console. Analogue systems generally need a dedicated processor, and that is a purchase most delay projects forget.