Understanding Feedback and Its Causes

Acoustic feedback occurs when a sound reinforcement system creates a loop between a microphone and a loudspeaker. The microphone picks up sound from the speaker, amplifies it, and sends it back out – each cycle increasing in volume until the system reaches maximum gain and produces that characteristic high-pitched squeal or low-frequency hum. Without proper management, feedback can ruin a live performance, distract an audience, and potentially damage speakers or hearing.

The Physics of Acoustic Feedback

Feedback happens when the gain in a system exceeds the acoustic attenuation between the speaker and the microphone. Every room has natural resonances that reinforce certain frequencies. When a microphone is placed within the coverage pattern of a speaker, those frequencies are amplified disproportionately. Feedback elimination tools are designed to break that loop by detecting the offending frequencies and applying a narrow notch filter to suppress them. Understanding both the acoustics of your venue and the polar patterns of your microphones is essential for effective feedback control.

Common Scenarios for Feedback

  • Vocal microphones placed too close to stage monitors or front-of-house speakers.
  • Omnidirectional microphones used in room corners where reflections accumulate.
  • High‑gain settings required for quiet instruments or distant speakers.
  • Undertrained operators who boost problematic frequencies on graphic equalizers.
  • Reverberant rooms with hard surfaces (glass, concrete, wood floors).
  • Multiple open microphones on stage simultaneously, which increases the probability of a feedback loop forming through phase summation.

Recognizing these scenarios early during setup allows you to combine positioning best practices with feedback eliminators for the cleanest possible result. A proactive approach to placement and gain structure will always yield better results than relying solely on processing to fix problems after they occur.

What Are Feedback Eliminators?

Feedback eliminators are specialized processors – either hardware units or software plugins – that automatically detect and suppress feedback frequencies. They differ from standard graphic equalizers in that they apply extremely narrow, dynamic filters rather than broad fixed bands. By targeting only the ringing frequency, they preserve the overall tonal balance of the mix while preventing howling. A typical feedback eliminator can deploy anywhere from six to twenty independent notch filters, each with adjustable depth and width, to surgically remove problem frequencies without affecting adjacent musical content.

Hardware vs. Software Solutions

Hardware feedback eliminators (such as the dbx AFS2, Behringer FBQ2496, or Shure DFR11EQ) are common in live sound racks. They feature physical controls for filter depth, number of filters, and lock modes. Software plugins – for example, Waves F6, FabFilter Pro-Q 3, or iZotope Neutron’s Feedback Assistant – are used in recording studios, broadcast, and virtual live streams. Both types rely on similar detection algorithms, but hardware units often have faster real‑time response suitable for dynamic stage environments.

Key differences:

  • Latency: Hardware units typically process with zero or near‑zero latency; software plugins depend on your audio interface buffer size. For live monitoring applications, any added latency can disorient performers, making hardware the safer choice for wedge monitors.
  • Permanence: Hardware filters can be locked to prevent accidental changes; software settings can be recalled with a session file. Lockable filters are especially valuable when multiple operators share a console during a festival or multi-act event.
  • Integration: Software eliminators can be inserted on individual channels or bus groups, whereas hardware units are usually inserted on the main output or monitor mix. This means software-based solutions offer more granular control – you can apply different suppression settings to a vocal channel versus a guitar channel.
  • Cost: Hardware units can be expensive but are a one-time purchase; software plugins often require annual subscriptions or license renewals, though many offer free trial periods for evaluation.

How Feedback Elimination Algorithms Work

Most feedback eliminators use a combination of real‑time spectral analysis and adaptive filtering. When a sustained frequency exceeds a programmable threshold, the processor places a notch filter at that exact frequency. The width (Q) of the filter is automatically adjusted to remove only the feedback while affecting as little program material as possible. Advanced algorithms also learn the room’s resonances during a calibration mode and apply fixed filters before the performance begins. This dual approach – fixed filters for predictable resonances and dynamic filters for live feedback – provides the most reliable suppression.

Some high-end units employ a technique called "frequency shifting" in conjunction with notching. Frequency shifting moves the entire audio spectrum by a few hertz, which disrupts the feedback loop without removing any frequencies. While effective, this method can introduce a subtle pitch-shifting artifact that may be noticeable on sustained tones. Most engineers reserve frequency shifting for spoken-word applications and rely on notch filtering for musical performances.

Real-Time Analysis vs. Learned Mode

Real-time analysis mode continuously monitors the spectrum and applies filters on the fly as feedback begins to build. This is essential for unpredictable performances where microphone positions or stage volume changes frequently. Learned mode, on the other hand, runs a calibration sweep before the show and commits filters to memory. Learned mode offers the advantage of zero latency during the performance and ensures that filters are placed precisely before any audio passes through the system. Many professionals run calibration in learned mode and then enable a small number of dynamic filters as a safety net for unexpected resonances.

Step‑by‑Step Guide to Using Feedback Eliminators Effectively

1. Strategic Microphone and Speaker Placement

Before activating any feedback eliminator, optimize your physical setup. Keep main speakers at least 1.5 meters in front of the front row of microphones. Angle stage monitors so their coverage stops at the back of the vocalist’s head – not at the microphone. Use directional microphones with tight polar patterns (cardioid, supercardioid) wherever possible. This reduces the amount of speaker energy reaching the microphone capsule, lowering the risk of feedback and reducing the workload on your eliminator. Place monitors on the floor rather than on stands whenever feasible, as floor coupling reduces low-frequency spill into vocal microphones.

2. Set Optimal Gain Structure Before Calibration

Proper gain staging is the foundation of any feedback-free mix. Set your microphone preamp gain so that the average signal level sits around -18 dBFS on a digital console or 0 dBVU on an analog desk. If your input gain is too low, you will compensate by pushing the channel fader or master output, which raises the noise floor and increases feedback potential. If your input gain is too high, the preamp itself may clip, causing distortion that resembles feedback. Once gain is set correctly at every stage of the signal chain, the feedback eliminator has a clean signal to analyze and can operate with greater precision.

3. Run a Full System Calibration

Most hardware eliminators have a "setup" or "learn" mode. During soundcheck, mute the house mix (if possible) and walk the stage while a band member or assistant produces sound at typical performance levels. The feedback eliminator will automatically ring the system back from a safe gain level and capture the room’s resonant frequencies as fixed filters. Do not skip this step. Calibration ensures that the eliminator commits its sharpest filters to the most likely feedback points before dynamics occur. For software-based eliminators, some plugins require you to play pink noise through the system and let the algorithm analyze the room response. Follow the manufacturer's instructions carefully, as incorrect calibration can result in filters placed at wrong frequencies.

4. Adjust Threshold, Filter Count, and Attenuation

Once calibration is complete, set the detection threshold carefully. A low threshold will cause the eliminator to react to transient peaks (like snare hits), unnecessarily notching out useful frequencies. A high threshold may miss feedback until it becomes audible. Start at a moderate level and increase gain on problematic channels until you hear the beginning of a ring – then lower the threshold slightly. Most units let you choose between 6 and 20 dynamic filters. Use the minimum number that suppresses feedback, because each filter removes a tiny slice of your mix.

Attenuation depth is the amount of gain reduction per filter. –6 dB is typical for live sound, but you may need –12 dB for extremely resonant rooms. Avoid exceeding –18 dB, as wide or deep filters begin to sound unnatural and can hollow out vocal or instrument tone. As a rule of thumb, start with the shallowest attenuation that stops the feedback, and only increase depth if the feedback persists after the filter is applied.

5. Lock Filters After Show Start

Once the performance begins and the system is stable, lock the filters on your hardware unit or disable further automatic filter allocation in your software. This prevents the eliminator from reacting to intentional musical content – such as a guitar feedback solo or a sustained vocal note – and notching out frequencies that should be present. Some consoles allow you to switch from "automatic" to "manual" mode after the first three songs of a set, preserving the filters that were established during soundcheck while preventing new filters from being added.

Advanced Techniques for Feedback Suppression

Using Multiple Filters and Dynamic EQ

Many modern feedback eliminators allow you to assign filters with different Q factors. Use narrow Q (high selectivity) for sustained feedback, and wider Q for frequencies that tend to ring only during specific parts of a song. Some software plugins like FabFilter Pro‑Q 3 support dynamic EQ mode, where the filter only activates when the frequency exceeds the threshold and releases afterward. This preserves the natural sound for 90% of the performance and only intervenes when necessary. For example, a dynamic filter on the 2.5 kHz range can clamp down on a ringing vocal only when the singer leans into the monitor, then release when they step back.

Combining Feedback Eliminators with System EQ and Compression

A feedback eliminator should not be your only defense. Use a graphic or parametric EQ to shape the system’s overall response before the eliminator. Cut problematic low‑mid frequencies (around 200–400 Hz) that commonly cause feedback in vocal mics. Apply gentle compression on vocal channels to reduce sudden gain jumps that trigger feedback. In digital consoles, use high‑pass filters on every microphone except kick drum and bass DI – this rolls off rumble and low‑end resonance that contribute to subsonic feedback.

Workflow tip: Insert the feedback eliminator after the system EQ and after dynamics processing, but before the main output limiter. This order ensures the eliminator sees the final frequency content and can react accurately. If you place the eliminator before compression, the compressor may later boost the notched frequencies, partially undoing the filter's effect.

Using Feedback Eliminators in Monitor Mixes Separately from Front of House

Feedback on stage monitors typically occurs at different frequencies than feedback through the main PA, because the acoustic relationship between microphones and monitor speakers is different from that between microphones and front-of-house speakers. If your console allows, insert a dedicated feedback eliminator on the monitor bus. Configure it with a lower threshold and more aggressive attenuation, since monitor feedback is more disruptive to performers. On the main mix, use a more conservative approach that prioritizes tonal quality. This separation of processing allows you to tailor suppression strategies to each zone of the venue.

Best Practices for Live Sound and Recording

Protecting Equipment and Maintaining Sound Quality

Overly aggressive feedback suppression can harm your mix. Always listen to the effect of each filter. If a vocalist’s tone becomes dull or thin, check whether the feedback eliminator is notching out fundamental frequencies. Many units have a "listen" function that lets you hear what the filter is removing. Use this feature periodically during rehearsals to verify that the filters are still appropriate. If you notice that a filter has been active for several songs without any feedback occurring, consider releasing that filter – the room conditions may have changed, making the filter unnecessary.

For recording studios, avoid using feedback eliminators during tracking unless absolutely necessary. Instead, employ room treatment, proper mic placement, and soundproofing. If you must use one (e.g., for a live broadcast mix), engage it only on the broadcast output, not on individual tracks that will be processed later. Recording raw tracks without feedback suppression gives you maximum flexibility during the mixing stage, where you can use surgical EQ and volume automation to address any residual feedback.

Workflow Integration and Automation

In digital mixing consoles (Yamaha CL/QL, Allen & Heath SQ/dLive, Behringer Wing), feedback eliminators can be assigned to specific mixes – for instance, only on monitor sends. This allows you to apply different levels of suppression to the stage wedge mix versus the house PA. Some consoles support snapshot automation that recalls feedback eliminator settings per song. Use this to reset filters between acts or when the room occupancy changes (e.g., empty rehearsal vs. full concert hall). Automating filter recall ensures that the same room is treated consistently across multiple shows in the same venue.

Managing Feedback Across Different Venue Types

Small clubs with low ceilings and reflective surfaces require aggressive feedback suppression, as the acoustic environment is inherently unstable. In such venues, use a higher number of dynamic filters (12-16) and a lower threshold. Outdoor stages, on the other hand, have minimal reflections and more predictable acoustics. Here you can reduce the filter count to 4-6 and raise the threshold, relying primarily on fixed filters from calibration. For large concert halls with complex reverberation patterns, run calibration at multiple locations on stage to capture the room's full acoustic signature. Some advanced eliminators allow you to store multiple calibration profiles and switch between them during a show as conditions change.

Troubleshooting Common Issues

  • Feedback still occurs after calibration: The threshold may be too high, or the microphone position has changed. Rerun calibration after any significant move. Also verify that the calibration signal was played at the same volume level as the actual performance – a quieter calibration sweep may not excite the room's resonant frequencies fully.
  • Sound becomes thin or lifeless: Too many dynamic filters have been activated. Increase the threshold to allow more gain before notching, or reduce the number of allocated dynamic filters in the hardware menu. Check for overlapping filters – two filters placed at nearly the same frequency may create a wide, unnatural notch.
  • Low‑end rumble feedback: Enable high‑pass filters on all vocal and instrument mics. Low frequencies (below 100 Hz) are rarely produced by vocalists and often cause subsonic feedback loops. In rooms with wooden floors, low frequencies can couple structurally and cause feedback that feels like vibration rather than a clear tone.
  • Different frequencies ring at different times: This indicates a change in room acoustics (opening doors, crowd moving). Use a feedback eliminator that offers "continuous" or "adaptive" mode rather than fixed filters only. Adaptive mode will track the changing resonance and adjust filters accordingly, though it requires more CPU power in software implementations.
  • Channel bleeding between microphones: Tighten polar patterns and use gates with short release times to keep microphones closed when not in use. In extreme cases, automate microphone muting for channels that are not actively being used during a performance. This is especially effective in theater productions where actors speak in sequence.
  • Feedback eliminator introduces audible clicks or pops: This occurs when filters engage or disengage too quickly. Adjust the attack and release times of the filter deployment. Most units let you set a "filter ramp time" – a value of 100-200 milliseconds is usually enough to smooth the transition without delaying suppression.

Feedback elimination technology continues to evolve. Machine learning algorithms are being integrated into both hardware and software solutions, allowing the processor to learn the acoustic signature of a room from a single soundcheck and predict feedback before it occurs. These systems can analyze hundreds of frequency bands simultaneously and generate filter sets that are optimized for the specific combination of microphones, speakers, and room geometry. Some manufacturers are experimenting with adaptive filter Q values that automatically widen or narrow based on the frequency content of the program material, ensuring that the notch is only as wide as necessary.

Another emerging trend is network-based feedback management, where multiple feedback eliminators across a Dante or AVB network communicate with each other. When one device detects a feedback frequency, it shares that information with all other devices on the network, allowing the entire sound system to respond coherently. This shared intelligence prevents a situation where the front-of-house eliminator notches a frequency while the monitor eliminator boosts it, effectively canceling out both processors' efforts.

Practical Advice for Adopting New Technology

When evaluating new feedback elimination gear, always test it in the venues you work in most frequently. A unit that performs well in a dry rehearsal space may struggle in a reverberant hall, and vice versa. Read user reviews from engineers who work in similar settings, and take advantage of demo periods for software plugins. Most importantly, never rely on feedback eliminators as a substitute for good system design and operator skill. The best feedback elimination strategy combines thoughtful physical setup, precise gain staging, and intelligent processing – with the eliminator acting as a safety net, not the primary line of defense.

Conclusion

Feedback eliminators are powerful allies in audio mixing, but they work best when integrated into a thoughtful system design. By understanding the physics of feedback, choosing the right hardware or software, calibrating carefully, and balancing automation with human oversight, you can achieve clean, feedback‑free sound without sacrificing tonal quality. Whether you are mixing a church service, a national tour, or a studio broadcast, applying these techniques will improve your mix’s clarity, protect your gear, and give your audience a reason to listen.

For deeper reading, refer to Shure’s guide on feedback and Sound On Sound’s technical overview on the subject. Additional resources include the Audio-Technica feedback control guide and the Behringer feedback eliminator tips wiki. For a deeper dive into acoustic measurement and room analysis, the Room EQ Wizard (REW) software is a free tool that can help you identify problematic resonances before you even turn on a feedback eliminator.