Practical Tips for Taming the 'howl'. Minimising feedback and maximising the music.
In part one we took a look at what feedback actually is, what causes it, and how it takes place at different frequencies.
In part two we took a look at physical measures we could employ to minimise feedback and maximise gain. Loudspeaker and microphone positioning, etc.
In this part, we are going to take a close look at how equalisers (or EQ's) can be employed to fight the dreaded howl. Graphic and parametric EQs are often employed to EQ sound systems although they are by no means the only devices that can be used. Graphic EQs are the most popular and important devices for couple of reasons. Firstly using practical examples of implementing graphic EQ's we learn about the nature of feedback and it's occurrence at particular frequencies/wavelengths, and secondly digital versions of graphic EQ's are now found in abundance in the latest digital mixing consoles and loudspeaker processors. So what we learn by using a traditional analog graphic EQ can be applied to the latest digital consoles and processors. Parametric EQ's can also be used and, once again, most digital devices offer parametric EQ options.
The first item we'll take a look at is the traditional analog graphic EQ. Although this type of device may be considered slightly 'old fashioned', digital versions of graphic equalizers are often to be found in the latest digital mixing consoles and loudspeaker processors. I think the analog graphic EQ is a great place to start, because it illustrates the fundamentals. What we can learn from an analog graphic EQ can be applied to the other devices that employ digital versions.
|2 x 30 Band Graphc EQ|
'Ringing Out' a Front-of-House PA Using a Graphic EQ.
The process used to EQ a sound system for minimal feedback is called 'ringing out'. Since this is a 'practical' guide to minimising feedback, we'll use this is a process as not only is it extremely effective, but it also teaches us about the nature of feedback. More comprehensive processes are available. Using pink and white noise in conjunction with real time analysers to measure room response and then applying EQ curves is a more 'scientific' and effective way of carrying out room EQ, which is great for venues and system installs where the time and equipment is available. But from a practical musicians perspective, ringing a system out using the lead vocal mic channel can deliver excellent results, fairly quickly, and with quite basic equipment.
'Ringing out' involves repeatedly advancing the gain of a channel (usually the lead vocal mic channel) which induces feedback at a particular frequency, and then using the graphic EQ to remove the 'ringing', then repeating the process.
Of course, we need to have our sound system setup properly in the first place in terms of gain structure and phase. You can check out my post on gain structure here
I would generally ring the system out between completing the setup and the soundcheck, so I would suggest getting your system setup with a graphic EQ in line with the master outputs of your mixing console, or via the master insert points. At this point, you really only need the lead vocal mic, so if your engineering a band that hasn't arrived yet, just setup a lead vocal mic in it's usual position. If the band arrives with you, then you can carry out the ringing out between setting everything up and the soundcheck. You may want to play some music to check everything is working.
On the mixing console, I would suggest starting with a flat channel EQ, set the gain control at the top of the channel to it's usual position (this will depend on the mic and how loud the singer is) so it delivers a healthy signal without clipping. A few 'one two's' down the mic channel might be useful for this. The channel fader at the bottom, Set the master fader(s) at unity, or '0' on their scale (usually about two thirds of the way up). Set the graphic EQ to unity input/output gain with all the faders in the middle detented position. If you have a choice of 6db, or 12db cut/boost, I would suggest setting it to 12db.
Since most live sound systems tend to be stereo, and therefore a stereo graphic EQ will be employed, you can 'ring out' each side separately. Since rooms are rarely completely symmetrical this will give you better results and ringing out both sides simultaneously may get rather confusing in terms of which side is actually feeding back, etc. So I would pan the channel left, carry out the ringing out procedure, reset the fader to the bottom, then pan right and carry out the ringing out procedure for the right side.
So with the channel panned left, open the lead vocal channel and slowly bring up the fader. As you advance the fader at some point you will reach your first 'ringing' point. This will be at either a low, mid, or high frequency. Identify the frequency that is ringing and bring the corresponding fader on the graphic EQ down 3db. It will take some practice to identify the frequncy of the ringing. If you bring down the fader on the EQ and nothing happens, then you've identified the wrong frequency. Replace that frequency back to the centre position and try adjacent faders. Once you've selected the correct frequency centre the ringing will stop. Then we repeat the process. Advance the channel fader on the mixing console some more until you hear more ringing. Identify the frequency and bring down the corresponding fader on the graphic EQ by three db. Keep repeating the process until you can't squeeze anymore gain out of the system. Then repeat the process for the right side.
Notes On Graphic EQ Types:
Graphic EQ's and Digital Mixing Consoles:
|Stereo Parametric EQ|