Understanding Feedback in Live Sound

Feedback is one of the most common and disruptive issues in live sound reinforcement. It occurs when a sound system creates a positive acoustic loop: a microphone picks up sound from a loudspeaker, that sound is amplified and sent back through the system, and the loop continues until the system reaches its maximum gain. The result is a piercing, often high-pitched squeal or low-frequency howl that can interrupt performances and damage equipment. To prevent this, engineers must employ a combination of physical placement, equalization, and dynamic processing. Compression plays a vital role in controlling the signal dynamics that often trigger feedback, but it must be used carefully to avoid further complications.

The Feedback Loop

A feedback loop begins when the gain of a system exceeds a critical threshold at a specific frequency. Every microphone, speaker, and room has resonant frequencies that are more prone to feedback. When a microphone is placed too close to a speaker, or when the system is pushed too hard, those resonant frequencies amplify exponentially. The sound becomes a sustained oscillation. Feedback can occur at any frequency, but it is most common in the midrange and high frequencies, where microphones are most sensitive. Understanding this loop is essential for applying corrective measures.

The Role of Compression in Live Mixes

Compression reduces the dynamic range of an audio signal by attenuating peaks above a set threshold. In live sound, this helps maintain consistent volume levels, prevents distortion, and protects speakers. But compression also influences how a system behaves acoustically. By smoothing out abrupt loud passages, the compressor reduces the instantaneous energy that can provoke feedback. However, improper compressor settings—such as too much gain reduction or a very fast release—can actually make feedback more likely by bringing up the level of quiet frequencies that are prone to ringing. Thorough understanding of compressor parameters is therefore critical.

Key Compressor Parameters Explained

  • Threshold: The level above which compression begins. A lower threshold engages compression on softer signals, which can raise the overall average level and potentially increase feedback risk. Set the threshold so that it only activates on peaks that could cause system instability.
  • Ratio: Determines how much compression is applied once the threshold is exceeded. A 4:1 ratio means that for every 4 dB above threshold, only 1 dB passes. Higher ratios (8:1, 10:1) are effective for limiting sudden transients that could start feedback, but they can also make the mix sound lifeless. Moderate ratios (3:1 to 6:1) are usually best for feedback management.
  • Attack: Controls how quickly the compressor responds after the signal exceeds the threshold. Fast attack times (1–5 ms) catch and clamp down on sharp peaks that may trigger feedback loops. Slower attack times allow initial transients to pass through, preserving punch but increasing feedback risk on percussive sounds.
  • Release: Determines how long it takes for the compressor to stop reducing gain once the signal falls below threshold. If the release is too fast, the compressor quickly stops reducing gain, which can cause the signal to jump back up and create a pumping effect that may excite feedback frequencies. A moderate release (50–200 ms) helps maintain natural dynamics while keeping feedback in check.

Why Compression Helps Prevent Feedback

Feedback is most often triggered by strong, sustained tones at specific frequencies. A compressor that acts quickly on these peaks reduces their duration and intensity. It also prevents the system from entering the runaway gain loop that defines feedback. Moreover, compression can allow the engineer to push the overall volume higher without exceeding the feedback threshold. That said, compression alone is not a cure-all. It works best in conjunction with other feedback reduction techniques.

Feedback Reduction Techniques

Before reaching for the compressor, live sound engineers should master the foundational techniques of feedback control. These methods address the root causes of feedback rather than just suppressing the symptom.

Microphone and Speaker Placement

The single most effective feedback reduction technique is proper placement. Keep microphones as far as possible from loudspeakers, especially when using directional mics. Cardioid and supercardioid patterns reject sound from the rear and sides, but they still pick up sound from the front. Point the null area of the microphone (the rear) toward the nearest speaker. For monitors, place them so that the microphone capsule is in the acoustic shadow of the performer's head. Even a few inches of movement can dramatically reduce feedback potential. Always check the polar pattern of the microphone before installation.

System Equalization

Equalization (EQ) allows the engineer to identify and cut frequencies that are prone to feedback. This is typically done during sound check using a graphic equalizer or a parametric EQ. A systematic approach: ring out the system by slowly raising the master volume until feedback begins, then cut that frequency by 3–6 dB. Repeat for each problematic frequency. Many engineers use a real-time analyzer (RTA) to see which frequencies are peaking. Subtractive EQ (cutting) is safer than boosting, as boosting can introduce new feedback nodes. For monitor mixes, apply EQ to the monitor feed independently from the front-of-house mix to avoid affecting the main sound.

Feedback Suppressors and Notch Filters

Automatic feedback suppressors are digital processors that detect sustained, narrowband signals and insert notch filters to eliminate them. These devices work in real time and can be a lifesaver in fast-paced environments, but they can also reduce audio quality if too many filters are engaged. Use them as a safety net, not a primary tool. Manual notch filters—where you identify the exact feedback frequency and cut it with a very narrow bandwidth—offer more precise control. A parametric EQ with adjustable Q allows you to cut only the offending tone without affecting adjacent frequencies. For example, if feedback occurs at 2.5 kHz, set a notch with a Q of 10–30 to attenuate just that narrow band.

Dynamic Equalization

Dynamic EQ is an advanced technique that only applies EQ cuts when a specified threshold is exceeded. This is especially useful for managing feedback that only occurs during certain notes or phrases. A dynamic EQ on a vocal channel can automatically attenuate a resonant peak (say at 1.2 kHz) only when the singer hits a note that triggers feedback. This preserves the natural tone during quieter passages. Many modern digital mixing consoles offer built-in dynamic EQ, making it easier to implement in live settings.

Integrating Compression with Feedback Control

Combining compression with the above techniques requires a systematic workflow. The goal is to create a stable, gain-before-feedback margin that allows the mix to be loud without howling. Here is a practical approach to integrating these tools on a typical vocal or instrument channel.

Setting Up the Compressor for Feedback Prevention

  • Start by achieving a clean mix without compression. Place mics, set EQ cuts for feedback reduction, and establish a stable gain structure with suitable headroom.
  • Insert a compressor on the channel. Set a moderate ratio (4:1) and a threshold that engages only on the loudest peaks (around 6–10 dB of gain reduction on peaks).
  • Use a fast attack (2–5 ms) to catch brief transient peaks that could excite feedback loops. A fast attack also helps prevent the compressor from “listening” to sustained tones that are already triggering feedback.
  • Set a release time that is slow enough to avoid pumping but fast enough to recover between musical phrases. A medium release (100–200 ms) works for most sources.
  • Adjust makeup gain to restore the overall level. Keep in mind that excessive makeup gain can negate the feedback prevention benefit by boosting the signal back into the feedback threshold.
  • Check for feedback by slowly raising the fader. If feedback occurs, identify the frequency and cut it with a notch filter rather than further adjusting the compressor. The compressor should only control dynamic peaks, not fix frequency-specific resonance.

Using Multiband Compression

Multiband compression splits the audio signal into several frequency bands (e.g., low, mid, high) and compresses each band independently. This is extremely powerful for feedback management because it allows you to compress only the problematic range. For example, if a vocal mic is prone to feedback in the 1–3 kHz range, you can apply 4:1 compression only to that band with a fast attack, while leaving the low and high bands uncompressed or lightly compressed. This prevents the compressor from over-reacting to other frequencies and retains the natural character of the voice. Many digital mixers offer multiband compression as a built-in processor; external units like the dbx 1066 or DBX 276XL are also options.

Sidechain Compression for Feedback

Sidechain compression allows you to use an external signal to trigger the compressor. In live sound, you can route a copy of the monitor feed (or a key input microphone) into the sidechain of the compressor on the vocal channel. This way, when the monitor mix gets loud enough to risk feedback, the compressor reduces the vocal channel’s gain before the loop begins. This technique works best when the sidechain signal is filtered to isolate the feedback-prone frequencies. For instance, you can route the monitor mix through a bandpass filter (e.g., 1–4 kHz) and use that filtered signal to trigger compression. This creates a feedback “detection” system that acts proactively rather than reactively.

Practical Example: Vocal Channel in a Small Club

Imagine a small club with a vocalist using a Shure SM58 and two side-fill monitors. During sound check, you hear a resonant howl at 2 kHz. You cut 3 dB with a narrow notch at 2 kHz on the monitor EQ. The vocal channel has a compressor with a 4:1 ratio, 3 ms attack, 150 ms release, and threshold set to −10 dBFS. You engage a multiband compressor on the same channel, compressing the 1–3 kHz band by 3 dB with a 2:1 ratio. The vocal now sits more consistently in the mix, and you can push the overall monitor level 3 dB higher before any feedback starts. During the show, if the singer leans into the monitor, the fast compressor catches the sudden level spike and prevents the system from howling. This integrated approach works reliably throughout the performance.

Advanced Strategies for Live Engineers

Beyond the basics, experienced engineers develop a set of strategies to further reduce feedback while maintaining dynamic expression. These include monitoring monitor mixes independently, using real-time analysis, automating changes, and training the ear.

Monitor Mixing and Feedback

Monitor mixes are the most common source of feedback because speakers are placed close to microphones. Use separate EQ on each monitor mix, and apply compression to monitor sends rather than only to the channel input. A compressor on a monitor send can be set with a fast attack and a high ratio (8:1) to act as a limiter. This prevents sudden bursts from reaching the monitor speakers at dangerous levels. Also, consider using “monitor wedge” equalization—cutting the low-mid frequencies that are particularly problematic in small wedge cabinets.

Using Real-Time Analysis (RTA)

A handheld RTA microphone or a software app on a tablet provides a live spectrum view of the room. While ringing out the system, watch the RTA display for the feedback peak. This confirms which frequency to cut. RTAs are also valuable during the show: if you see a narrow peak building up, you can preemptively cut that frequency with a notch before it becomes audible feedback. Many digital consoles include an integrated RTA; use it to correlate your hearing with the visual data.

Automating Changes During Performance

In larger venues with digital consoles, automation can be used to recall specific feedback-reducing settings for different sections of the show. For example, if a certain song has a loud guitar solo that pushes the monitors to the brink, you can automate a slight reduction in monitor gain or a more aggressive compressor ratio on the vocal channel during that song. This prevents the need for manual adjustments in the heat of the show. Use scenes or snapshots to store these variations.

Training Your Ear

No amount of technology replaces a trained ear. Experienced engineers develop the ability to identify the approximate frequency of feedback immediately. Practice by using test tones and a graphic EQ: sweep through frequencies while listening for the room’s resonant peaks. Over time, you’ll learn to distinguish between a 2 kHz squeal and a 3.5 kHz sizzle. This skill allows you to make fast EQ cuts on the fly, often before the audience even notices the feedback.

Conclusion and Best Practices

Combining feedback reduction techniques with compression is a cornerstone of professional live sound. The best results come from a layered approach: start with proper microphone placement and system equalization, then apply compression to control dynamics, and finally use advanced tools like multiband compression, sidechain techniques, and RTA analysis for fine-tuning. Key takeaways for live engineers include always ringing out the system before the show, setting compressor attack times fast enough to catch feedback-triggering transients, and using notch filters for specific frequencies while letting compression handle broad dynamic control. For further reading, consult resources from Shure’s guide to feedback elimination, Sound on Sound’s article on avoiding feedback, and this practical video from Pro Sound Guy.

Remember that every venue and every performance is different. The feedback frequencies shift with the room, the performer’s position, and the microphone choice. Stay flexible, listen critically, and trust your tools—but never rely on them blindly. With practice, the integration of feedback reduction and compression will become a natural part of your mixing workflow, delivering clear, powerful sound that keeps the audience engaged and the performers confident.