Ringing out a room..

Si_

I've sort of become the bands unofficial sound man, although I know very little about it other than what I've read on websites and youtube videos I've watched. 

I think I'm getting there, but we still struggle with monitor feedback. I've just picked up a 31-band EQ with the hope of using it in the monitor signal to help with that. 

So I guess I need to learn how to ring out a room. I've bean reading about it and I just had 1 query.. when doing it, is it best to just use the mic in the main singers position on stage? and just set up as he would use it? I've read contradicting websites with some saying it should be on the floor, some on the stand, etc..

Just to check I've got the procedure correct:

• set monitor volume to  unity
  
• set monitor send of channels to full
  
• set mic gain to 0
  
• slowly raise mic gain until feedback occurs
• use 31-band EQ to reduce offending frequency
• repeat until no more feedback

Assuming that's correct how high do I go with the mic gain before it's done?

John_P

Si_ said:

Assuming that's correct how high do I go with the mic gain before it's done?

That would work -  just a slight tweak I use -  set the channel gain for the mic to a reasonable level for your singer, to optimise the gain structure in the desk,  then bring up the output level til you hear it edging into feedback and drop the offending frequency - keep the adjustments as small as you can.

Having plenty of headroom, careful positioning of the speakers and mics (check you know the pattern for that mic) should minimise the amount of adjustment needed. 

ESBlonde

You are on the right track, the channel gain should always be set so there are no red lights indicating clipping at the loudest point, remember adding ANY tone eq will take it back over the clip point, it's usually better to subtract eq.

Next the gain to the monitors should be below clipping point, always set the power amp gain about 60% up (the amp works at full power all the time the 'gain knob' is not a volume control).

Before you try to ring out the room you should know the natural peaks that your mic and monitor will resonate at (no monitors are perfectly flat). Try setting it up outside one time to find the hot spot.

With the channel gain set and the fader up somewhere near normal front of house volume (all volume is additive) start to bring up the aux send for the first mic until it starts to ring, with the graphic find the offending frequency and reduce it 3-6 dB (always subtract never add on a graphic). once the ring is gone increase the channel aux send until the next freq rings, pull down a freq on the graphic to stop it. Repeat once more and the stop. Now turn down that aux send and bring up another if applicable. Then turn the master aux send down a little to preserve headroom and prevent a squeal mid set. Do not attempt to ring out more than three nodes as you will hack the spectrum and make it too hard to hear whats left, also on graphics they create further peaks in other octaves to those you pull down. 

All mics on the same mix should be the same brand/model to reduce the natural peaks you have to fight.

Study the effects of the mics polarity and position the wedge in the NULL area (directly behind on a cardoid but at 120 deg for a typical hyper cardoid - find out and make sure).

The quality of your wedges matters A Lot! Some cheap 12" cones and piezos are not adequate. Bi amped boxes are usually the best in budget boxes.

Graphic Equalisers are old fashioned and a bit of a blunt instrument, you can buy digital versions that do the graphic feature and have parametric Equalisers and feedback destroyers all in a single rack. These powerful tools can enable you to tune the mic/wedge before you even get to a gig and memorise it. At many gigs it then becomes a simple thing to turn them on often without any feedback at all.

If you have more questions ask away, you do get a knack for spotting the ring quickly with experience.

maltingsaudio

Your just about on the money with these replies. With regards to your last question the aim is to get the mic as loud as possible in the monitors befor it feeds back so if you can get the slider all the way up with out a squeel you've done it. Also do bear in mind this is a monitor mix so it doesn't have to sound as nice in the monitors as it does out front, you can probably knock out every thing on the graphic below 160 hz and above 5k befor you start

You may find this helpful http://sft.sourceforge.net/

Si_

When doing this should I have the FOH up at gig levels? or just the monitors and no FOH sound?

Sporky

I've always done it (but for installations) with a pink noise generator and a multi-band parametric EQ.

Turn the pink noise on through the monitors, set one band to a high Q and a boost and sweep that frequency. At the room's "live" frequencies it'll resonate and you can then knock that frequency down - I normally start at -3dB, usually it needs a bit more cut than that. Rinse and repeat...

ESBlonde

Them there feedback destroyers are clever/usefull and a pain in the harris. They sweep and detect the offending frequency then narrow the width and lesson the depth of cut to suit. The frequency they cut is exact rather than the nearest frequency on a graphic. Once the FB device has cut three freqs, copy or transfer these to a parametric and turn off the FB Destroyer otherwise it continues to hunt and cut to the detriment of the sound. As MS said use a band pass filter on the Parametric to remove all below 160 Hz it wastes amplifier energy and makes the stage sound muddy.

Whilst you can learn how to do it by hand/brain, it is an acquired skill requiring a fair bit of practice and tolerance by others in the room at soundcheck.

Cheap option but requires some book reading and is self contained.

http://www.studiospares.com/effects+amps/behrigner-fbq1000-feedback-supressor/invt/383000?source=215_74&gclid=CODD5cfT4bkCFTLMtAodqjYAAA#

The better option that also requires some operator knowledge but is well worth it.

http://www.gak.co.uk/en/behringer-ultra-curve-pro-deq-2496/1139?gclid=CJLW5OHT4bkCFU_MtAodpmsAfw

They kick the pants of any graphic three times their cost in terms of performance and none colouration. They (and pretty much all digital devices)have a downside if you push the system hard you can't grab a fader and pull a freq on the fly. If used as intended they out perform a graphic.

Si_

Lots of people put me off the automatic feedback destroyers which is why I opted for the 31 band EQ instead (although the Behringer one has Feedback detection which as said above, lights the lights of the offending frequencies to make life a little easier).

The Ulragraph has an adjustable low pass filter so I can use that to cut the low frequencies out of the monitor mix. 

Sporky

I'd avoid a feedback eliminator (especially the crap-tastic Behringer ones) unless you're going to have roving microphones. Better to properly EQ the system.

Jalapeno

The advice up there is accurate ^^^^^^ 

Most people who advised you meant the Behringer DSP110 - really clunky

Use the FB destroyer to find the most likely culprits, then remove it.

Leaving it inline will eventually kill all the freq bands in turn

ESBlonde

Jalapeno said:

The advice up there is accurate ^^^^^^ 

Most people who advised you meant the Behringer DSP110 - really clunky

Use the FB destroyer to find the most likely culprits, then remove it.

Leaving it inline will eventually kill all the freq bands in turn

Once the freqs and Q's are found, convert them to the parametric or lock them and disable further search. They are effective but you must use them correctly.

maltingsaudio

Si_ said:

When doing this should I have the FOH up at gig levels? or just the monitors and no FOH sound?

Initially do it with FOH off then put it in. Quite handy during sound checks to alternate between monitors only and FOH in, also if you have a loud monitor system be aware that it can intrude on the sound in the room, your call but from an audiences point of view they prefer to hear front of house.

Si_

Jalapeno said:

The advice up there is accurate ^^^^^^ 

Most people who advised you meant the Behringer DSP110 - really clunky

Use the FB destroyer to find the most likely culprits, then remove it.

Leaving it inline will eventually kill all the freq bands in turn

It was the DSP1124P i was looking at.

I've bought the EQ now so will have to live with that.. I'm sure it will do the job as required. (or I hope it will anyway)

Sporky

It's doable with a 31 band, it's just that you have to be better at it than me. Parametric is easy.

Si_

Well, that didn't go as planned... hooked everything up on Friday night, new EQ in place in the monitor mix, just couldn't get it to work. all cables were Ok but I wasn't getting anything on the input level son the EQ, although sound was coming form the monitors (although a lot quieter than usual).

I assume I did something wrong but I'm not sure what.. so will try again at this weeks gig.

maltingsaudio

What did you get in the end? If it was connected and sound was coming through the wedges then it was doing something. Two things spring to mind, Not enough level from the desk to the EQ do you use jack or xlr to connect it, there is often a big gap in the input level depending on which one you use. Second is was the input level meter switched to monitor the input?