Feedback Elimination for Live Sound: A Practical Guide

Feedback is not random. It follows the laws of physics precisely. When sound from a loudspeaker reaches a microphone, it is amplified and sent back to the loudspeaker. If the total gain around this loop exceeds 0 dB at any frequency, that frequency sustains and grows into audible feedback. The frequencies that feed back first are determined by room modes (resonant frequencies of the enclosed space), speaker and microphone polar patterns, and the distance and angle between them.

The gain-before-feedback (GBF) is the maximum system gain achievable before feedback occurs. Every 3 dB of GBF improvement doubles the effective operating range of the microphone. A well-optimized system provides 6-12 dB more GBF than an unoptimized one, which is the difference between a system that barely works and one that performs confidently.

Before touching any EQ, open SonaVyx RTA and measure the system response with all microphones open at normal operating gain. The frequency response reveals existing resonances and the overall spectral balance. Note any frequencies where the response is already elevated, as these are likely feedback candidates.

Switch to the problem detector and slowly increase gain. SonaVyx identifies emerging feedback frequencies automatically, showing the frequency, severity level, and recommended notch filter parameters. This removes the guesswork from the ring-out process.

With all microphones open and pointing toward their intended sources, slowly raise the system gain until the first feedback frequency appears. Do not let it develop into a full howl. As soon as you hear the ring beginning, note the frequency on the analyzer and reduce gain immediately.

Apply a parametric notch filter at the identified frequency. Start with a 3 dB cut at a Q of 15-20. Raise the gain again. If the same frequency returns, increase the cut depth to 6 dB. If a new frequency appears, apply a second notch filter. Repeat for 4-6 frequencies.

Stop after addressing the 6 most troublesome frequencies. Each notch filter you add removes a small amount of useful audio content. Excessive notching creates a thin, unnatural sound that is worse than moderate feedback risk.

EQ is the last resort for feedback control. Better strategies address the root cause. Use directional microphones (cardioid or hypercardioid) with the null aimed at the loudspeaker. Maintain maximum practical distance between microphones and loudspeakers. Mute unused microphones, as each open microphone reduces GBF by 3 dB per doubling of open mic count.

Loudspeaker aim is critical. Point speakers away from microphone positions. In churches and theaters, aim speakers at the audience, not the stage. Use speakers with controlled directivity to minimize energy sent toward the stage area. In-ear monitors eliminate monitor wedge feedback entirely.

Monitor wedge feedback is the most common feedback problem in live sound. Stage monitors are close to microphones, pointed roughly toward them, and operating at high volumes requested by performers. Solving monitor feedback requires working with the performers to find acceptable monitor levels, using in-ear monitors where possible, and ring-out each monitor mix independently.

Main speaker feedback is less common but more disruptive when it occurs. It typically happens when a microphone is carried into the coverage area of the main speakers, such as a pastor walking off the platform with a handheld microphone. Prevention requires clear communication about coverage zones and, where possible, automatic gain control that reduces level when the microphone enters a high-risk zone.