Skip to content
Church Sound Calc
Menu

System tuning

How to Tune a Church Sound System

Last updated

Quick answer

Tune in this order: aim, gain structure, high-pass, system equalisation, then fill delay and level. A blended service sits at 88 dBA average at the mix position with 12 dB of peak headroom, and the house curve should be flat to about 2 kHz and roughly 6 dB down by 16 kHz.

Most church systems are tuned in exactly the wrong order. Somebody hears a problem, reaches for the graphic equaliser, and pulls down whatever band seems closest. Two years later the equaliser looks like a mountain range, the system has 12 dB less headroom than it started with, and the original problem is still there, because the original problem was a loudspeaker aimed at a wall.

Tuning is a sequence, and the equaliser is step four of five. Each step removes a cause that would otherwise be papered over by the next one. Work through it in order and most rooms need only three or four gentle filters at the end. Before you start, confirm the level you are aiming for with the SPL loss by distance calculator and the targets in the dBA targets by service style chart.

On this page
  1. The five steps and what each one fixes
  2. The order of operations, step by step
  3. What the house curve should look like
  4. Gain structure is where the headroom lives
  5. What equipment you actually need
  6. Tune with the room in the state you use it
  7. When tuning cannot save you

The five steps and what each one fixes

Read the table top to bottom and do not skip. Roughly 70 percent of the audible improvement in a typical church comes from the first two rows, and both are free.

Four of the five tuning steps happen before the equaliser is touched, and a blended service should end up at 88 dBA average at the mix position.

The tuning sequence, the tool for each step and the target
StepToolTargetWhat it fixes
1. AimA ladder and an inclinometerWithin 6 dB front to backFront rows too loud, back row thin
2. Gain structureConsole meters and preamp trimAverage near 0 on the meter, peaks 12 dB below clipNoise, distortion, no headroom
3. High-passChannel filters80 to 120 Hz on voice, 120 to 160 Hz on fillsMud, stage rumble, wasted amplifier power
4. System EQProcessor or graphicFlat to 2 kHz, about 6 dB down at 16 kHzRoom resonances and horn response
5. Fill delay and levelProcessor delayPath difference in ms plus 10 to 15 ms, 6 to 10 dB downImage pulling to the ceiling, echo

Levels are A-weighted averages at the mix position over a full song or a full reading, not instantaneous peaks. A phone meter is accurate enough to compare seats, but not accurate enough to claim a calibrated number.

The order of operations, step by step

  1. Aim the boxes before touching anything electronic. Put the on-axis point of each main on the back row and check that no box is behind the microphone line. A system that measures within about 6 dB from the front row to the back row is aimed. The full method is in how to aim and place church speakers.
  2. Set gain structure from the source outward. With the loudest expected source, set each channel preamp so the average sits near the top of the green and the peaks stay about 12 dB below clip. Then set the output and the amplifier so the system reaches its design level with the console faders at unity, not at the top of their travel.
  3. High-pass everything that is not a kick drum or a bass. Roll off below 80 to 100 Hz on speech microphones, 100 to 120 Hz on most vocal microphones and 120 to 160 Hz on fills and small boxes. Nothing below those frequencies on those channels is signal, and removing it returns headroom to the amplifier.
  4. Play pink noise and listen, then measure. Play pink noise through the mains at a moderate level and walk the room. Ringing at one pitch is a room mode, dullness is a coverage problem, harshness around 2 to 4 kHz is usually the horn. Free measurement software with a measurement microphone at the mix position turns those impressions into a curve.
  5. Cut narrow, boost never. Apply cuts of 3 to 6 dB with a narrow bandwidth at the two or three frequencies that misbehave. Do not boost: a boost costs headroom directly and a room resonance cannot be filled in from the console. If the curve needs more than about six filters, go back to step one.
  6. Set the fills last, then save the scene. Delay and level the under-balcony and front fills after the mains are right. Then save the processor settings, write them down on paper, and lock the processor so a volunteer cannot move them. An undocumented tuning is lost the moment the person who did it leaves.

What the house curve should look like

A system measured with pink noise should be roughly flat from about 100 Hz to 2 kHz and then tilt gently downward, ending about 6 dB down at 16 kHz. That downward tilt is not a fault: a flat measured response in a real room sounds harsh, because the direct sound and the reverberant field sum differently with frequency. Anything steeper starts to sound dull and starts to cost intelligibility.

Below 100 Hz the room dominates completely. Room modes in a sanctuary of typical proportions land in the 30 to 80 Hz range and can create 10 dB peaks that move as you walk. Do not chase those with narrow cuts across the whole system: fix the worst one, and solve the rest with subwoofer placement or with treatment, which is covered in acoustic treatment for a sanctuary.

The single most useful filter in most churches is a gentle cut of 2 to 4 dB somewhere between 200 and 400 Hz, where a hard-surfaced room builds up energy and speech turns boxy. The second most useful is a cut around 2.5 to 4 kHz where cheap horns get harsh. Everything past those two is refinement.

Gain structure is where the headroom lives

A system tuned for level but not for gain structure runs out of clean output long before it runs out of watts. The principle is that every stage in the chain should hit its own nominal level at the same time, so the preamp is not straining while the amplifier idles, or the reverse. The detail is in gain structure for church sound, and the reason it matters here is headroom: program material peaks 10 to 12 dB above its average, so a system set to just reach 88 dBA average clips on every transient unless it can actually produce 100 dB.

That is why the amplifier sizing on this site always includes a headroom term. A 250 seat room at a blended 88 dBA with a 45 foot throw calls for roughly 473 watts per box, and about 1,000 watts of amplifier per side once you round to a size that is sold. Work your own room through the amplifier power calculator before you conclude that the system is under-powered: usually it is under-tuned.

What equipment you actually need

A loudspeaker processor such as dbx DriveRack PA2 2x6 PA Complete Professional Loudspeaker Management System, Rack Mountable with Android and iOS Control. Black at $426.00 gives you crossover, high-pass, parametric equalisation, delay and limiting on every output, which is everything in this list in one box. Behringer Ultradrive Pro Dcx2496 Ultra High-Precision Digital 24-Bit/96 Khz Loudspeaker Management System covers the same ground with more outputs and a steeper learning curve. A dual 31 band graphic such as dbx 231s Dual Channel 31-Band Equalizer is the older approach and still works, but a graphic tempts operators into the mountain-range mistake because every band is visible and adjustable.

You also need a measurement microphone and a laptop, a reliable reference recording that you know well, a sound level meter or a meter app, and a set of closed headphones such as Sony MDR7506 Professional Large Diaphragm Headphone for checking the console output independently of the room. Comparisons between the processor options are in best loudspeaker processors.

A feedback suppressor such as dbx AFS2 Advanced Feedback Suppression Processor with Full LCD Display belongs at the end of this process, not the start. It buys you a few decibels of gain before feedback on a system that is already aimed and equalised. On a system with a loudspeaker behind the pulpit it buys you almost nothing, which is the honest position argued in how to eliminate feedback in a church.

Tune with the room in the state you use it

An empty sanctuary is measurably more reverberant than a full one. A congregation of 250 people adds several hundred sabins of absorption, mostly above 500 Hz, which shortens the decay and softens the top end. Tune in an empty room and the system will be bright and thin on a Sunday morning.

The practical compromise is to do the technical work empty, then verify and make final adjustments during a service with the room occupied. Keep those service-time adjustments small and write down what you changed. If you also run a livestream, check the broadcast mix separately, because the room is not part of it at all and a house tune has no bearing on what the stream hears. That is the point of how to mix for livestream.

When tuning cannot save you

Some problems are not electronic and no equaliser will touch them. A room with a 2.5 second reverberation time and a worship band is a building problem, and the fix is absorption, not filters. A loudspeaker with a 90 degree pattern in a 35 foot wide room is a coverage problem, and the fix is a different box. A stage running at 95 dBA acoustically before the system is turned on is a stage volume problem, and the fix is in reducing stage volume in church.

Being honest about which category a complaint falls into is the most valuable thing a volunteer tech director does. Spending a Saturday on the equaliser when the answer is 78 absorption panels wastes the Saturday and delays the fix by a year.

Sources

  • SPL targets, headroom figures and amplifier sizing as implemented in src/lib/avmath.mjs
  • Loudspeaker processor filter and delay ranges from manufacturer published specifications

Frequently asked questions

Do I need a measurement microphone to tune a church system?

Not to get 80 percent of the way there. Aiming, gain structure and high-pass filters are done with a ladder, the console meters and your ears. A measurement microphone and free software make the last step faster and stop you cutting frequencies that were never a problem. If you buy one thing, buy a calibrated measurement microphone rather than another processor.

How flat should the system measure?

Flat from about 100 Hz to 2 kHz and then tilting down to roughly 6 dB below that level at 16 kHz. A genuinely flat measured curve sounds harsh in a real room because the reverberant field adds energy differently at each frequency. Below 100 Hz the room dominates, so treat the low end as a placement and treatment question rather than an equaliser question.

Should I boost or cut when equalising?

Cut. A boost costs headroom directly: 6 dB of boost means the amplifier reaches clipping 6 dB earlier at that frequency. A room resonance is a peak caused by the room, and cutting it removes the excess, while boosting a dip caused by a cancellation just pours energy into the cancellation. Cut narrow, by 3 to 6 dB, at two or three frequencies.

How often should a system be retuned?

A permanent install that has not changed does not drift. Retune when something physical changes: new carpet, new seating, a removed wall, a different loudspeaker, or a new processor. Otherwise verify once a year against your written settings and confirm nobody has moved anything. Documenting the tuning on paper matters more than repeating it, because volunteers rotate and memories do not transfer.

Why does the system sound different with a full congregation?

People absorb sound. A congregation of 250 adds several hundred sabins of absorption, mostly above 500 Hz, which shortens the reverberation time and takes some brightness out. A system tuned in an empty room will be bright and thin when full. Do the technical work empty, then verify with people in the seats and keep those final adjustments small and documented.

Can I tune the system from the sound booth alone?

No. The booth is one position, often against a back wall where low frequency builds up by 3 to 6 dB, which makes the room sound bassier there than anywhere else. Walk the room during every stage of tuning: front row, centre, back row, both outside aisles and under any balcony. What you set from a single chair is a tuning for that chair.

Researched, not professional advice. This page is compiled from published manufacturer specifications, operator manuals, FCC rules, published standards and owner-review consensus, not hands-on testing. Sound system design, rigging loudspeakers overhead, and any electrical work are jobs for a qualified professional: have flown loudspeakers and their attachment points signed off by a structural engineer or a certified rigger, and have all wiring done by a licensed electrician to your local code. Wireless microphone rules change, so confirm the current FCC position before buying. As an Amazon Associate we earn from qualifying purchases.