Multi-Zone Audio Gain Structure: Setting Levels End to End

Good gain structure means every stage in the audio chain runs at a healthy level, with the same headroom, so that the noise floor stays low and no stage clips before the next one does. Set levels from the source towards the loudspeakers, one stage at a time, and trim zones only at the point where they split.

This note is written by Singapore-based AV and IPTV integrator Prestige Solutions for technical managers and AV consultants who run AV control and signal routing in multi-purpose venues. It covers nominal levels, headroom budgets, where zone trims belong, how to size amplifier gain, how to build the remote feed for hybrid events and the order we use when commissioning. Reference levels vary by device, so confirm them against the data sheet of each unit in the chain.

What are the reference levels in a typical audio chain?

Professional gear treats +4 dBu as nominal line level, which is about 1.23 V RMS, while consumer gear uses -10 dBV, about 0.316 V RMS. The difference is roughly 11.8 dB, which is why a laptop plugged straight into a pro mixer input often sounds weak and noisy.

A reference of -18 dBFS leaves 18 dB of headroom before digital clipping. That is enough for speech peaks and most music, and it is the figure we plan around unless the data sheets say otherwise.

Where should gain be added, and where should it be left alone?

Add gain once, early, at the input stage, then pass the signal through the rest of the chain at roughly unity gain. Every extra boost later amplifies the noise that earlier stages already added, and every cut later wastes headroom.

• Input stage: set the preamp or input trim so a normal speaker reaches about -18 dBFS on average, with peaks no higher than about -6 dBFS.
• Processing stage: EQ, filters and dynamics should add or remove no more than a few dB. If an EQ band needs +10 dB, fix the cause upstream instead.
• Matrix and zone mixing: sum several sources into a zone and the level rises. Two uncorrelated sources of equal level add about 3 dB, four add about 6 dB. Leave room for that sum.
• Output stage: set the DSP output so the amplifier input sits close to its rated sensitivity at the loudest wanted level, with no further analogue attenuator in between.

This is the same discipline we describe for network microphones in the Dante beamforming microphone reference: get the level right at the source, and the downstream stages stay simple.

How should zone trims be arranged in a multi-zone venue?

Put one master level per zone and one trim per speaker group, and keep source levels out of the zone controls. Users should adjust "foyer" or "ballroom section B", not an individual source in an individual zone.

A divisible ballroom needs one more rule: when the partition opens, the combined zone should take its level from a preset, not from the sum of two operators' faders. Store a preset for each partition state and have the touch panel recall it when the sensor reports the wall position.

How do you size amplifier gain from the numbers?

Work out the voltage needed for full power, compare it with the amplifier's rated input sensitivity, and the difference is the gain the amplifier must supply. The sums are short enough to do by hand for every zone.

Example for a direct low-impedance zone: a 150 W amplifier driving an 8 ohm load needs V = square root of (P x R) = square root of (150 x 8), about 34.6 V. If the amplifier reaches that at 1.4 V input, its gain is 20 x log10(34.6 / 1.4), about 27.9 dB. A DSP output at +4 dBu is 1.23 V, so it is just under the 1.4 V that gives full power. That is the intended arrangement: normal programme at nominal level uses most of the amplifier's range, and the limiter catches the rest.

Example for a 70 V distributed line: ten ceiling speakers each tapped at 6 W present 60 W. Plan the amplifier at no less than 75 W, which gives 20% spare capacity. A 6 W tap on 70 V corresponds to roughly 70 x 70 / 6, about 817 ohm per speaker, which shows why a long line with many taps can be tuned for coverage rather than raw loudness. Speaker spacing and tap choice are covered in the ceiling speaker spacing table.

Level also drops with distance. In free field the sound pressure level falls about 6 dB per doubling of distance, so a zone with a deep seating area may need more than one speaker row, delayed in time, instead of one louder row.

How is the remote feed built for hybrid events?

Send the remote participants a mix-minus: everything the room hears, except the remote participants' own voices. Without it, the far end hears an echo of itself and the operator starts turning things down until the room cannot be heard.

Set the remote feed with its own limiter and a level that matches speech at the codec input, not the room mix. Check that the room speakers do not feed back into the presenter microphones at the level you set; echo cancellation on the conferencing path is a safety net, not a substitute for correct gain.

In what order should levels be set during commissioning?

Work from source to speaker and lock each stage before touching the next. Changing an early stage after later stages are set invalidates every measurement downstream.

1. Confirm every device's input and output reference levels from its data sheet, and write them in the commissioning sheet.
2. Feed a 1 kHz tone at nominal level into each source input and set the input trim so the DSP meter reads the planned reference, for example -18 dBFS.
3. With amplifiers off or the speaker lines disconnected, set the zone matrix and master faders at 0 dB and confirm each zone output meter reads the same as the input.
4. Set the amplifier input attenuators to a documented starting position, normally fully clockwise for rated sensitivity, and leave them there.
5. Connect the speakers, play pink noise at a low level, and raise the zone output trim until a calibrated meter at the listening position reads the target SPL.
6. Set the zone limiter threshold about 3 dB below the point where the amplifier begins to clip, tested with a tone.
7. Run speech and programme material from each source and confirm no stage clips on peaks and none sits below about -40 dBFS during normal use.
8. Build the mix-minus feed, call the remote end and confirm there is no echo and no level jump when sources change.
9. Save presets for each room state, lock the installer controls and record final trim values in the handover pack.

FAQ

Why does the system hiss only when all zones are on?

Usually because some stage carries too little signal and is being amplified later to compensate. Each active zone also adds its own amplifier noise. Re-check input levels first, then see whether an analogue attenuator sits between the DSP and the amplifier; moving the attenuation into the DSP output and running the amplifier at full sensitivity often lowers the audible hiss.

Should the touch panel control individual sources inside a zone?

Rarely. Give the operator source selection and one zone level, and keep the source levels fixed after commissioning. Exceptions are a lectern microphone or a guest performer's input, where a single level control with a limited range is reasonable.

How often should levels be rechecked?

Recheck after any amplifier, speaker or DSP firmware change, and as part of the periodic maintenance visit. Store the original trim values and test-tone results, so a later technician can compare them quickly instead of starting from zero.

Visit Prestige Solutions for the wider AV and IPTV service range.

Recommended next step

Contact Prestige Solutions to review your site drawings and current operating pattern. Call +65 8010 2337 or +65 9825 8218, message us on WhatsApp (+65 8010 2337), or email sales@prestigesolutions.com.sg. You can also browse the full product range before the site walk.