How to Find and Eliminate Feedback
Quick Answer
Feedback occurs when amplified sound from a speaker re-enters a microphone, creating a self-reinforcing loop at specific frequencies. Finding feedback frequencies means identifying these resonant peaks using spectral analysis, then applying precise narrow notch filters or adjusting mic and speaker positions to break the feedback loop and maximize gain before feedback.
Try It Now
Open RTA Mode — free, no download required.
Equipment Needed
- ✓SonaVyx with RTA mode and spectrograph
- ✓Parametric EQ (console channel EQ or system processor)
- ✓Measurement microphone for initial analysis
- ✓Performance microphones (cardioid or supercardioid)
- ✓Monitor speakers or in-ear monitor system
Step-by-Step Guide
Understand the Feedback Mechanism
Acoustic feedback happens when the loop gain (microphone to preamp to EQ to amplifier to speaker to acoustic path back to microphone) exceeds unity (0 dB) at any frequency where the total phase shift is a multiple of 360 degrees. The frequency where this condition is first met as gain increases is the most feedback-prone frequency. Multiple frequencies may ring simultaneously in severe cases. Understanding this helps you address the root cause rather than just the symptom.
Open SonaVyx RTA Mode
Launch SonaVyx and switch to RTA (Real-Time Analyzer) mode. Set the FFT size to 4096 or 8192 for adequate frequency resolution. Use 1/12 octave smoothing or no smoothing to see narrow feedback peaks clearly. The RTA displays the microphone input spectrum in real time, allowing you to watch for emerging peaks as you increase system gain. Enable the spectrograph view for a time-frequency display that shows feedback building over time as persistent bright lines.
Slowly Increase System Gain
With the room quiet and all microphones open, slowly increase the master fader or aux send level. Watch the RTA spectrum for any frequency that begins to rise above the surrounding level. A frequency that grows 10 dB or more above its neighbors is a feedback candidate. Stop increasing gain immediately when you hear the first ring. Note the frequency displayed by SonaVyx's cursor readout. Typical first-ring frequencies in small rooms are 800 Hz to 4 kHz for vocal microphones.
Apply Narrow Notch Filters
Apply a parametric EQ notch at the identified feedback frequency. Start with a narrow bandwidth (Q of 10 to 30) and cut 3 to 6 dB. A Q of 16 at 2 kHz affects only a 125 Hz wide band, minimally impacting overall sound quality. After applying the notch, continue increasing gain until the next feedback frequency appears. Repeat the process for up to 6 to 8 frequencies. Beyond this, excessive notch filtering degrades audio quality and you should address the problem physically instead.
Optimize Physical Setup
Before relying entirely on notch filters, address physical causes. Move microphones closer to the sound source and farther from speakers. Angle speakers so their coverage pattern directs sound toward the audience and away from microphone positions. Use directional microphones (cardioid, supercardioid) with their rejection null aimed at the nearest speaker. Reduce monitor volume and consider switching to in-ear monitors for maximum feedback rejection.
Verify Gain Before Feedback
After optimization, measure the gain before feedback (GBF) by slowly increasing system level until the onset of ringing. The GBF should be at least 6 dB above the maximum operating level needed for the performance. If GBF is insufficient, consider using feedback suppressors (automatic notch filters) or redesigning the physical layout. Document the final GBF and notch filter settings for future reference. SonaVyx's before/after comparison tool can quantify the improvement.
Technical Background on Acoustic Feedback
The Nyquist stability criterion states that a feedback loop becomes unstable when the open-loop gain exceeds unity (0 dB) at a frequency where the phase shift equals a multiple of 360 degrees. In acoustic systems, the path from speaker to microphone introduces frequency-dependent gain and phase shift determined by room acoustics, speaker directivity, and microphone pickup pattern. The room's impulse response creates multiple paths with different delays, each contributing constructively or destructively at different frequencies.
Feedback Margin and Headroom
Professional system design aims for a feedback margin of at least 6 dB, meaning the system operates at least 6 dB below the threshold of instability. This margin provides headroom for acoustic variations during performance, such as a singer moving closer to a wedge monitor or audience absorption changing between soundcheck and show. Each notch filter typically provides 2 to 3 dB of additional GBF.
Automatic vs Manual Feedback Suppression
Automatic feedback suppressors like the Sabine FBX or dbx AFS2 detect and notch feedback frequencies in real time. While convenient, they can sometimes misidentify musical tones as feedback. Manual ring-out using SonaVyx's RTA provides more control and avoids false triggers. A hybrid approach works best: manual notches for the primary feedback frequencies found during soundcheck, plus one or two automatic filters for protection during the performance.
Common Mistakes to Avoid
Using too-wide notch filters (Q below 5) which remove musical content along with the feedback frequency
Boosting EQ at other frequencies to compensate for notch filter cuts, which raises the overall loop gain
Applying notch filters without first optimizing the physical microphone and speaker arrangement
Ringing out monitors at excessive volume, risking hearing damage and speaker damage
Ignoring the contribution of reflective surfaces near microphones that create secondary feedback paths
Applicable Standards
| Standard | Clause | Relevance |
|---|---|---|
| IEC 60268-16 | Annex B | STI measurement in the presence of feedback and equalization |
| AES-2id:2023 | Clause 5 | Transfer function measurement for feedback analysis |
Try It Now
Open RTA Mode — free in your browser.
Related Guides
How to Ring Out Stage Monitors
Step-by-step guide to ringing out stage monitors for maximum gain before feedback. Identify feedback frequencies and apply precise notch filters.
How to Tune a PA System
Learn how to tune a PA system with transfer function measurement, system EQ, and delay alignment. Free browser-based tools for live sound engineers.
How to Time-Align Speakers
Learn speaker time alignment using impulse response and delay measurement. Align mains, fills, and subwoofers for coherent summation and coverage.
How to Measure Speaker Frequency Response
Learn how to measure speaker frequency response using swept sine, pink noise, or transfer function. Step-by-step guide with free browser-based tools.
Design-Time Companion
Need to design room acoustics before measuring? AcousPlan handles room acoustic design, while SonaVyx handles deploy-time measurement and tuning.