Hi all.
I've seen a lot of info regarding
feedback, and I would like to share with you some of the interesting things about this.
I'ld also appreciate if you could add more stuff, from your experience regarding on how to prevent
feedback.
Here is some of my findings on how to prevent
feedback:
- Place your loudspeakers ahead of your mics
- Reverse the polarity of your loudspeaker or mic (this is really good stuff)
- If you have monitors for the musicians, try to remove, or lower the level, of their own sound in the monitor speakers (e.g. for vocals - send instruments, for violins - send metals, and so on.) (I also like this one)
- Lower the volume on stage, increase on the FOH
There is still something that I really don't know how to do, is using the
equalizer to eliminate
feedback. I understand that
feedback occurs at some frequency, and the EQ can lower this frequency. But I don't know how to find this frequency from the spectrum of frequencies.
Technically,
feedback is when the electroacoustic loop
gain exceeds unity, in other words when a reproduced sound gets picked up by a
microphone at a
level greater than the original
level, which then continues that loop building in
level each time. Also technical, the general issue is wanting a Potential Acoustic
Gain (PAG), which is how much
gain the
system can provide before
feedback, that is greater the the Needed Acoustic
Gain (NAG), which is how much
gain you are trying to get from the
system.
Placing the speakers in front of the microphones is really addressing a more general issue of how the patterns of the speakers and microphones relate. Speakers and microphones generally have some directionality, which can often vary at different frequencies. The goal is to use that to your advantage in order to try to get the sound from the speakers directed at the listeners rather than at the microphones and to have the microphones
pick up the desired source(s) while tending to reject, or
pick up at a lower
level, other sources such as sound from the speakers.
Reversing the
polarity of a
microphone or
speaker has its place but not usually in preventing
feedback. The concept is that inverting the
polarity of either the original signal or signal from the
speaker will cause them to cancel but the reality is that since they typically originate at two different points in space they are not
phase coherent and thus such simplistic cancellation does not occur. So I won't say it can't help in some cases but it is not as simple as some sources present it and it can have other implications.
There is a lot you can do with monitors to
address gain before
feedback including the
speaker pattern, the
microphone pattern, the relative location of the
microphone, EQ,
etc. But probably the one that often gets overlooked is that monitors are intended to give the performers what they need to perform and not to sound like it does out in the audience. Try to always think in terms of less is more and what can be taken out of a
monitor mix rather than what needs to be added.
Since the original sounds, the directional characteristics of the speakers and microphones, the relative
phase of the direct and indirect signals and so forth are all frequency dependent,
feedback is very frequency dependent. There are different ways to determine a
feedback frequency, or preferably ringing as it starts and before it turns into
feedback. One is by ear training and there are plenty of resources out on the Internet to help there. The other is with an
RTA, FFT analyzer, spectrograph,
etc. One trick I've seen with functionalities such as the spectrograph mode in
SMAART is to set the
threshold high enough that you don't or rarely see anything during normal use but then see the signal as
feedback starts to
build and exceeds that
threshold level.
