Gain staging for vocal mics in "lively" room

[Bullitt5135]

In a small/lively practice room, is it advisable to roll back the input gain on the vocal mics a little bit (to make them less sensitive) and increase channel volume in order to minimize potential feedback? Or will increasing the channel volume to the desired level result in the same potential for feedback. I try to shoot for -18db RMS on the input gain meter on my Soundcraft ui24r.

This is what my pea brain is trying to understand... If I set the gain:volume ratio to 3(gain):7(volume), is the potential for feedback less than a ratio of, say, 7(gain):3(volume)?

Background: I am having some difficulty with feedback in our small rehearsal room. Even though I've rung out the monitor with a graphic EQ, I'm still always teetering on the edge of feedback. IMO, we are not overly loud (my ear is sensitive to moderately loud noise, and I am able to play without earplugs without any discomfort). I think I already know the answer, which is turning down guitars and bass to the absolute bare minimum.

 

[byacey]

You want a certain amount of headroom on the trim control, otherwise there is a tendency to clip the input preamp on high peaks. You should never be approaching clipping on your preamps; 18dB isn't really that much headroom. Open up your master to unity gain, set your channel faders to unity gain (0db) and bring your trims up to get the appropriate volume in your mix. This won't give you the best s/n ratio, but it will give you the best headroom before clipping on your inputs.

Feedback can be caused by many things, speaker placement, the room dimensions and reflections, microphone choice, speaker quality, etc. Sometimes EQ just doesn't have enough control over the situation if some of the other factors are creating excessive instability in the sound system.

I know this doesn't provide any definite answer to your problem, but you have to closely examine all these factors.

 

[s0c9]

You want a certain amount of headroom on the trim control, otherwise there is a tendency to clip the input preamp on high peaks. You should never be approaching clipping on your preamps; 18dB isn't really that much headroom. Open up your master to unity gain, set your channel faders to unity gain (0db) and bring your trims up to get the appropriate volume in your mix. This won't give you the best s/n ratio, but it will give you the best headroom before clipping on your inputs.

ugh.. the dreaded "mix by trim" method. If one is doing this, then the channel fader may as well not be there - it's just a unity passthru to the master!
The trim/gain should be set - with channel fader set to infinity(off). Use the PFL meter to set the gain/trim to the top of the "green" or lower "orange" levels. Where that level is Db-wise varies by vendor. Makes sure the input is not clipping at it's loudest level. Put a high-pass/low-cut filter on vox (cuts off < 100 Hz) to remove the "rumble".
Sure, set master to unity, then bring up the channel fader to the volume you need. Same applies to every other input.
If there's clipping, you may need to reduce the gain, but other than that, volume adjustments are via the channel fader.. that's what they are designed for!
It may be that there's too much noise in the room and you won't get enough vox volume to be hear over the top of the instruments! Plan B required!

Feedback can be caused by many things, speaker placement, the room dimensions and reflections, microphone choice, speaker quality, etc. Sometimes EQ just doesn't have enough control over the situation if some of the other factors are creating excessive instability in the sound system.

+100% on this one!

 

[filmtex]

ugh.. the dreaded "mix by trim" method. If one is doing this, then the channel fader may as well not be there - it's just a unity passthru to the master!
The trim/gain should be set - with channel fader set to infinity(off). Use the PFL meter to set the gain/trim to the top of the "green" or lower "orange" levels. Where that level is Db-wise varies by vendor. Makes sure the input is not clipping at it's loudest level. Put a high-pass/low-cut filter on vox (cuts off < 100 Hz) to remove the "rumble".
Sure, set master to unity, then bring up the channel fader to the volume you need. Same applies to every other input.
If there's clipping, you may need to reduce the gain, but other than that, volume adjustments are via the channel fader.. that's what they are designed for!
It may be that there's too much noise in the room and you won't get enough vox volume to be hear over the top of the instruments! Plan B required!

+100% on this one!

Yup.

 

[byacey]

ugh.. the dreaded "mix by trim" method. If one is doing this, then the channel fader may as well not be there - it's just a unity passthru to the master!
The trim/gain should be set - with channel fader set to infinity(off). Use the PFL meter to set the gain/trim to the top of the "green" or lower "orange" levels. Where that level is Db-wise varies by vendor. Makes sure the input is not clipping at it's loudest level. Put a high-pass/low-cut filter on vox (cuts off < 100 Hz) to remove the "rumble".
Sure, set master to unity, then bring up the channel fader to the volume you need. Same applies to every other input.
If there's clipping, you may need to reduce the gain, but other than that, volume adjustments are via the channel fader.. that's what they are designed for!
It may be that there's too much noise in the room and you won't get enough vox volume to be hear over the top of the instruments! Plan B required!

+100% on this one!

I didn't say this was the best way to mix, but if you want to maintain maximum headroom on the input preamps, this works.

Also the strip faders are still used to do your fine mixing; they don't have to reside at zero, which is just a baseline starting point. Either way, I would still check levels with pfl.

 

[s0c9]

I didn't say this was the best way to mix, but if you want to maintain maximum headroom on the input preamps, this works.

Also the strip faders are still used to do your fine mixing; they don't have to reside at zero, which is just a baseline starting point. Either way, I would still check levels with pfl.

Point taken!!
It's just not the method I was taught how to use.. my mentor (ex-FOH/mons for Clapton, RHCP, Buffett and others) cringes when he sees folks mix by trim!

 

[byacey]

Point taken!!
It's just not the method I was taught how to use.. my mentor (ex-FOH/mons for Clapton, RHCP, Buffett and others) cringes when he sees folks mix by trim!

There's a couple of schools of thought on mixing; you can run your trims as close to clip as possible which will give you maximum signal to noise ratio, and then attenuate the signal back down to the desired level with the fader. This is best for recording and live production when you want the hottest, low noise signal possible.

The other method only uses as much gain as necessary, but sacrifices / trades off signal to noise for maximum headroom before clip at the input stage. This method lends itself better towards live mixing at festivals, etc; especially when you are running multiple acts one after another, without the luxury of any sound checks.

 

[s0c9]

There's a couple of schools of thought on mixing; you can run your trims as close to clip as possible which will give you maximum signal to noise ratio, and then attenuate the signal back down to the desired level with the fader. This is best for recording and live production when you want the hottest, low noise signal possible.

The other method only uses as much gain as necessary, but sacrifices / trades off signal to noise for maximum headroom before clip at the input stage. This method lends itself better towards live mixing at festivals, etc; especially when you are running multiple acts one after another, without the luxury of any sound checks.

I always do/mandate line checks when mixing festivals and still never mix with trims.
Easily done with a little prep, practice and console layout!

 

[TomB]

Are you using compression on the mics? It can create feedback havoc in small spaces and stage monitors if not carefully applied.

 

[Wasnex]

In my experience, if you push any part of the signal path too close to distortion your chances of incurring feedback increase dramatically. Before a signal starts to distort, it will start to compress. Compression essentially means quiet signals pass without attenuation, and hotter signal are attenuated a bit. The result is the quiet signals have a bit higher gain. The extra gain will tend to pull more feedback out of the "ether."

I am an advocate of mixing off the trims to balance out the board for live sound, and I have a good reasons for it (IMHO and YMMV). I typically run both FOH and monitors from one console. If I set up the main mix to be balanced with all the faders in the same place, that means the monitors will be pretty darn close if I set them up the same way.

So I balance out the main mix with the trims, and rely on a bit of a formula to set up the monitor mixes. The formula is to give each person a full monitor mix, and then add 6-10dB to the channels that relate to each person. For example if the guitar player sings, the related guitar and vocal channels get 6-10dB bumps on the appropriate monitor sends.

I use this formula to preset each of the monitor mixes before sound check even begins. The last thing we do during sound check is give every one a chance to fine tune their monitor mix. In my experience, people still need to ask for tweaks to their monitor mix, but frequently the tweaks are very minor.

A second benefit IMHO: Inevitably the stage will break out into sort of a volume war. One person turns up and this sets up and series of chain reactions, where everyone eventually turns up. This can go round and round the band, causing the stage volume to get louder and louder over the course of the concert.

Instruments where the Master Volume impacts the level of the signal sent to FOH compound this problem. If someone bumps the signal they are sending to FOH, it impacts every monitor mix that includes the effected channel. For example the guitar player turns up a mic'ed amp and the bass player hears more guitar in his/her monitor mix. Then the bass player has to turn up to compensate. If this bass is using a Post EQ DI, the bass may get louder in the guitar player's monitor. So then the guitar player has to turn up again.

When I notice someone bump the volume on their amp, I typically correct for the extra SPL by turning the trim down on the affected channel. The idea is to rebalance all of the mixes, and not just the main FOH mix. It's not perfect, but I find this does help.

In order to use this trick, it's helpful if the board is setup in a way where you have visual references. In other words, all of the faders have sort of the same home position that you can use as a reference.

During most of the show, I actually mix off the faders.

If I were mixing a recording session, I would adjust the trimmers in the normal way (typically for the best S/N ratio). The S/N ratio tends to be a bit more important during recording, but I don't typically find it to be much of a problem for live sound.

 

[byacey]

In my experience, if you push any part of the signal path too close to distortion your chances of incurring feedback increase dramatically. Before a signal starts to distort, it will start to compress.

Clipping also creates harmonics in addition to the original signal which can touch off feedback in a system that is balancing on the edge.

 

[Ryan in PDX]

In a small/lively practice room, is it advisable to roll back the input gain on the vocal mics a little bit (to make them less sensitive) and increase channel volume in order to minimize potential feedback? Or will increasing the channel volume to the desired level result in the same potential for feedback. I try to shoot for -18db RMS on the input gain meter on my Soundcraft ui24r.

This is what my pea brain is trying to understand... If I set the gain:volume ratio to 3(gain):7(volume), is the potential for feedback less than a ratio of, say, 7(gain):3(volume)?

Background: I am having some difficulty with feedback in our small rehearsal room. Even though I've rung out the monitor with a graphic EQ, I'm still always teetering on the edge of feedback. IMO, we are not overly loud (my ear is sensitive to moderately loud noise, and I am able to play without earplugs without any discomfort). I think I already know the answer, which is turning down guitars and bass to the absolute bare minimum.

Ui24r user here. I've been unable to make the thing feed back (with AFS active). I've logged dozens of gigs and many more rehearsals with it.

My first thought would be to check the acoustics/physical layout. Is a monitor bouncing off a wall into a vocal mic, or is someone pointing the rear of a super cardioid mic at the monitor? If your amps are so loud that they're a problem, then yes, that's a problem. I sincerely hope your guitarist is open minded enough to turn down for the sake of the mix. Again, with the layout: where is the amp pointing?

On quiet gigs and practices, I just turn AFS on (Live mode) for the mains and monitors. I also use the HPF/LPF if I think there might be an issue, and send most things to the monitors pre-processing if they're heavily compressed. For louder gigs, I will ring out the aux send (Fixed/measurement mode), simulate those cuts using the graphic EQ, clear the fixed list, ring it out again to use up the Fixed slots, then switch it to Live. Probably overkill, but it only takes a few seconds per send, and it's worked for me. Danny Olesh has some great videos on YouTube (mix-record-connect is the channel name) about setting up these features.

As far as gain staging, I set the gain so that I'm confident I won't clip when the big chorus comes. I am watching peaks, not RMS level. As long as you're not clipping, use your faders however you like. After soundcheck, I prefer to deal with the mix screen rather than the gain screen.

And here's a quick and dirty trick that sometimes works when all else fails: flip the polarity on the aux channel. Takes half a second and you'll know instantly if it helped.

Feel free to DM me about the ui24r. It's a great little unit, but not that many people are using them. I am more than happy to share what I know!

 

[Medicine Man]

To a certain extent, mic placement and room/stage volume just needs to be controlled, imho. The board can only do so much.

 

[Bullitt5135]

The trim/gain should be set - with channel fader set to infinity(off). Use the PFL meter to set the gain/trim to the top of the "green" or lower "orange" levels. Where that level is Db-wise varies by vendor. Makes sure the input is not clipping at its loudest level. Put a high-pass/low-cut filter on vox (cuts off < 100 Hz) to remove the "rumble".

This is how I normally set up my inputs, but I'm thinking I probably have the gain set too hot – but not clipping at all.

Sure, set master to unity, then bring up the channel fader to the volume you need.

This is a new concept for me. Right now, most of my channel and master/aux faders reside around -35dB on the meters. I think I will have to rethink my approach...reduce input gain to safer levels, and compensate on the channel and master faders.

If I were mixing a recording session, I would adjust the trimmers in the normal way (typically for the best S/N ratio). The S/N ratio tends to be a bit more important during recording, but I don't typically find it to be much of a problem for live sound.

Yeah, this is how I'm set up right now. I'm starting to realize that this is not the best approach for live sound.

My first thought would be to check the acoustics/physical layout. Is a monitor bouncing off a wall into a vocal mic, or is someone pointing the rear of a super-cardioid mic at the monitor? If your amps are so loud that they're a problem, then yes, that's a problem. I sincerely hope your guitarist is open-minded enough to turn down for the sake of the mix.

The monitor is most certainly bouncing off the walls and into the vocal mics. The monitor is on a stand at head height in the corner of the room. The room also serves as a home office, so I'm not at liberty to start moving stuff around. Right now, the lead singer is furthest from the monitor (and closest to the drums). I'm going to suggest moving him away from the drums and closer to the monitor. The guitar player is not the problem. The instruments are never loud, they are just too loud for the room even at low volume – probably due to the flaws in my gain structure.

 

[Bass4Brkfast]

I subscribe to the "don't mix by trim" school of thought, although I see the merits explained by Wasnex.
I also subscribe to the K.I.S.S method, whereby before jacking up the gain staging, I would try to reposition the speakers and/or reposition the vocal mics to minimize the bleed that sets off the feedback loop.
I know the OP has a small space, but speaker and/or mic orientation can go a long way to reducing feedback.

 

[Bullitt5135]

Visual reference for anyone who is interested. (Slow day at work!) This is how the room is currently set up. It's tight – maybe 12x16 at the most, but with office furniture adding to the congestion. Drywall and an area rug over a tile floor. Not much room at all to spread out or move around.



My inclination is to move the LEAD VOX to the RHYTHM GUITAR position, closest to the monitor and furthest from the drum kit. BASSIST moves to the former LEAD VOX spot next to the drum kit, leaving the bass amp in the same spot. RYHTHM GUITAR moves to former bassist spot. LEAD GUITAR stays put – bonus point if I can get him to move his amp across the room and facing him (and angled toward his head).

 

[beatmachine pro]

This is what my pea brain is trying to understand... If I set the gain:volume ratio to 3(gain):7(volume), is the potential for feedback less than a ratio of, say, 7(gain):3(volume)?

no your incorrect,..you will have the same volume in the end
the reason to have the most gain avaliable at the board is that you can turn the faders down quick!
if you have hot mics they will feedback when the singer steps closer to them
you dont want all your gain at your amp either

it sounds like the mic volume chain is what you need to work on
try moving your mics and monitors around. you could even try getting them facing the band off the floor about 5 ft back. ]

*mics with less pick up at 180*polarity would also clean up the mess

 

[Bass4Brkfast]

Visual reference for anyone who is interested. (Slow day at work!) This is how the room is currently set up. It's tight – maybe 12x16 at the most, but with office furniture adding to the congestion. Drywall and an area rug over a tile floor. Not much room at all to spread out or move around.

View attachment 4368007

My inclination is to move the LEAD VOX to the RHYTHM GUITAR position, closest to the monitor and furthest from the drum kit. BASSIST moves to the former LEAD VOX spot next to the drum kit, leaving the bass amp in the same spot. RYHTHM GUITAR moves to former bassist spot. LEAD GUITAR stays put – bonus point if I can get him to move his amp across the room and facing him (and angled toward his head).

Sounds like a good plan to start with.
It looks like the feedback comes from reflections off the back wall and into your lead vox mic.
Keep that in mind when deciding on the orienration of the other mics.

 

[DanDBassist]

Since the monitor speaker is at head height and positioned as shown, and going by the placement of the mics for the rhythm, lead gtr, and lead vox, the sound is probably bouncing from the walls causing the feedback. A simple experiment would be to lower the height of the speaker a little at time to hear what happens. Changing one thing at a time is the best way to troubleshoot.

 

[SoCal80s]

In a small room, the sound from the speaker is going to bounce around and get into the mics. Experiment with positioning of the monitor. Consider having two monitors at lower volumes instead of 1 at higher volume. Consider setting the monitors on the floor and angled back at you. Also, what mics are you using? All should be quality cardioid mics (Samson Q7 or Shure SM-58 at minimum).