If sound engineers are Luke Skywalker then feedback is most certainly Darth Vader, or various other strained analogies that you can probably think of. That high pitched scream coming out of a microphone or low hum reverberating out an acoustic guitar is a sure way to make your audience wince and automatically blame you for not doing your job properly. Feedback is a constant source of pain for most sound engineers and to make matters worse every single situation will bring its own difficulties. In this article we look at what feedback actually is, how it can be prevented and how best to avoid it all together.
So what is feedback? Audio feedback (or the Larsen effect as it’s sometimes referred to) is a sound loop between an audio input device and an audio output device. The signal received by a microphone for example is amplified then passed out through a loudspeaker, the audio output device. The signal, or at least parts of it, is then picked up by the microphone, amplified again and then passed through the speaker again.
This cycle continues until it gets to the point where a certain frequency becomes so dominant it results in a shrill or low pitched screech depending on the offending frequency or frequencies. All this can happen in a matter of milliseconds the resultant noise can depend room size, room dimensions, speaker type, amplifier type, the diaphragm on a microphone and a whole host of other factors.
That’s the technical explanation, practical reasons why this occurs are that microphones or acoustic instruments are placed too close to a speaker or monitor. If the audio input signal is too close to the audio output method the input is bound to pick up some of what is emanating from the output source. Too a high a gain structure can also be reason. If gain input is too high then it makes the microphone extremely sensitive to any audio that gets near it. Other reasons can actually depend on the room you are in.
If the room naturally excites a particular frequency, by design or otherwise, it is highly likely that frequency will be prone to feedback. Obviously you can’t start knocking walls down in a venue or remove the roof so we have to look at other methods to eliminate this issue.
So how do you get rid of it? Each situation will require a different approach. Obviously moving the input device away from the output device is an obvious one. If you find speakers are pointing directly into the pickup pattern of a microphone then its best to move the mic. The same can be said for monitors, just move the mic a bit further away. If you are in completely fixed environment you can look at mics with a narrower pickup pattern to avoid the excess acoustic energy within the environment. Gain structures also play an important role. If all gain inputs on a mixing desk are pushed too high there will naturally be feedback. Try setting the main fader level to 0dB then adjusting the gain input to point where you receive feedback. At this point back off the gain slightly and then adjust level to where you need it. Setting the gain structures is key to avoiding feedback right from the start.
When setup and gain structures have been set correctly and feedback still occurs it’s time to break out the EQ. Ideally a graphic EQ would be used to pull out offending frequencies. Rather than using your channel strip EQ which is best suited for tonal correction a graphic EQ will pin point a specific frequency and reduce it if needed. Your channel strip EQ is a lot less precise when it comes to this type of correction and trying to take out a specific frequency will affect the others around it to the detriment of your mix. If you find the feedback is a high pitched shrill try reducing some of the higher frequencies, if it’s more of a hum it may be the lower frequencies. Experienced sound engineers will be able to detect the exact frequency of feedback just by listening. This is all down to experience and is part of the live sound learning curve.
The diagram on the right shows how a graphic EQ is pin pointing the 4.5 kHz frequency whereas the parametric EQ is adversely affecting all frequencies between 3.15 kHz and 6.30 kHz. 31 band graphic EQ devices can be an incredibly useful piece of equipment but are not always available. Fortunately a lot of digital desks have graphic EQs built into most outputs. Simply by selecting the Graphic EQ of the offending output you can eliminate the frequencies causing a problem without losing the tone of your mix. However digital desks can be expensive, so are there any other ways to combat this without paying for extra outboard gear or costly digital desks?
Mackie’s new Pro FXv2 range of desks actually has a small graphic section built in. This is a great addition as it allows you target the most common areas of feedback within one piece of equipment. An increasing trend recently has been to put a feedback eliminator in the speakers themselves. This clever little device will automatically detect where a loop is occurring and reduce the frequency by the required amount of dB. When looking for a set of speakers it may be worth checking if they have an auto feedback eliminator, while it gives you less control than a graphic it’s as simple to use as pressing a button.
So to summarize, to treat feedback in the most effective way possible;
- Make sure your mics are in optimal positions
- If required and possible move monitors and speakers to better positions
- Establish proper gain structure
- Avoid where possible using channels strips to eliminate feedback
- Develop an ear for feedback to get to the correct frequency quickly
- Use a 31 band graphic EQ or digital desk with a graphic option if possible.
- Where this is not look for equipment with a smaller built in Graphic eq.
- When selecting speakers think if it would be useful to have a built in feedback eliminator.