House of Worship: Multi-Zone System with Balcony Delays

The Challenge: Three Zones, Three Problems

Grace Community Church, a 1,200-seat modern worship facility, had grown from a single-level auditorium to a three-zone venue after a recent expansion. The main floor seated 800, a new balcony added 350 seats, and a glass-enclosed cry room at the rear accommodated 50 parents with young children. The main floor PA system worked well, but the expansion created three distinct acoustic problems that the original system was never designed to handle.

Congregants in the balcony reported a noticeable echo effect — they could hear the main PA system from below and the balcony speakers from above, but the two were not synchronised. The cry room was even worse: an ambient microphone hanging from the ceiling captured a muddy, reverberant version of the main room sound and played it through a small wall speaker, resulting in unintelligible audio that parents simply ignored.

Measurement: IR, SPL, and STI Across All Zones

Using three SonaVyx tools in sequence, the consultant mapped the acoustic conditions in all three zones.

Impulse response measurement at the balcony front row:

• Main PA arrival: 0 ms reference (direct sound from main clusters below)
• Balcony speaker arrival: 85 ms late — the balcony speakers were 29 metres further from the stage than the main listening plane, but had zero delay applied
• The 85 ms offset was well outside the Haas fusion zone (30 to 50 ms for speech), creating a distinct echo rather than reinforcement

SPL measurement across zones:

• Main floor (row 15 centre): 88 dBA
• Balcony (front row centre): 80 dBA from main PA, plus 84 dBA from balcony speakers
• Net balcony level: approximately 85 dBA (combination), but with the echo artefact
• Cry room: 68 dBA with poor spectral quality

STIPA measurement:

• Main floor: 0.65 (good)
• Balcony: 0.38 (poor — the echo degraded the modulation transfer function)
• Cry room: 0.28 (bad — the ambient mic captured reverberation rather than direct speech)

Diagnosis: Timing, Level, and Signal Path

The AI diagnostic engine identified three independent problems:

1. Balcony delay: 85 ms uncompensated. The balcony speakers had been installed with direct feed from the main amplifier rack, with no DSP delay processing. The 29-metre path difference created an 85 ms arrival gap that was perceived as a distinct echo.
2. Balcony level: -8 dB relative to mains. The direct sound from the main clusters was 8 dB lower at the balcony than at the main floor reference point. The balcony speakers partially compensated but their timing mismatch degraded the intelligibility benefit.
3. Cry room signal path: The ambient microphone captured a combination of direct sound, reverberation, and HVAC noise. By the time this signal reached the cry room speaker, the direct-to-reverberant ratio was approximately -6 dB — meaning the reverberant energy exceeded the direct speech energy.

Solution: Delay Alignment, Level Matching, Direct Feed

Three targeted corrections were implemented:

1. Balcony delay speakers set to 85 ms: A DSP delay module was inserted in the balcony speaker signal path. The delay was set to 85 ms to match the acoustic arrival time of the main PA at the balcony. SonaVyx IR measurement verified that the two arrivals were now within ±2 ms at the balcony front row.
2. Level matching: The balcony speaker level was adjusted so the local reinforcement was 2 dB below the main PA arrival at the listening position. This maintained the Haas precedence effect, ensuring the congregation perceived sound as coming from the stage.
3. Cry room direct feed: The ambient microphone was disconnected. A direct line-level feed from the mixing console's matrix output was routed to the cry room speaker via a dedicated amplifier. Volume control was provided via a wall-mounted attenuator.

Results: Consistent Experience Across All Zones

Before/after measurements:

Zone	Metric	Before	After
Balcony	Delay offset	85 ms (echo)	±2 ms (fused)
Balcony	SPL	80 dBA (main only)	86 dBA (coherent)
Balcony	STI	0.38 (poor)	0.62 (good)
Cry room	STI	0.28 (bad)	0.65 (good)
Main floor	STI	0.65 (good)	0.65 (good, unchanged)
All zones	Minimum STI	0.28	0.62

The balcony echo was completely eliminated. Congregants reported that they could no longer distinguish any timing difference between the main PA and the local speakers. The cry room transformation was even more dramatic — parents reported that they could finally follow the service clearly, and the cry room began seeing regular use again after months of being avoided.

Lessons Learned

• Delay alignment is non-negotiable for multi-zone systems: Even a few milliseconds of offset degrades intelligibility. An 85 ms offset creates a clearly audible echo that devastates STI. SonaVyx IR measurement provides the exact delay value.
• Level matching preserves the Haas effect: Local speakers should be 2 to 6 dB below the main system arrival to maintain the illusion that sound originates from the stage. If local speakers are louder, the congregation perceives sound as coming from above rather than from the platform.
• Ambient microphones are never acceptable for overflow rooms: A direct console feed provides clean, intelligible audio at a fraction of the cost of trying to make an ambient pickup work. The STI improvement from 0.28 to 0.65 demonstrates the magnitude of this difference.
• Measure all zones, not just the best seats: The main floor STI of 0.65 masked severe problems in the balcony and cry room. A comprehensive multi-zone survey with SonaVyx ensures no congregant is left with a substandard experience.