Feedback prevention trick:Myth or Fact (live sound question)

[chconnor]

Hi - (sorry to post a live sound question on the recording forum... if inappropriate, please do remove the post)

I've heard a theory from a couple people now, and it sounds crazy to me, and I always want to tell them so, but i figure that before running my mouth off i should check with the board. :-)

The theory is this: when facing a feedback-prone situation, you can turn the input gain down and compensate by turning the fader up. This will somehow make the mic "less sensitive" and cause it to feed back less easily. My expectation would be that the overall system gain is the only relevant factor, regardless of where that gain is coming from, and that this "trick" would be useless, and probably evolved just because gain knobs typically have a much wider range than faders.

However, I know that mic pre-amp circuits do occasionally have unexpected (to me) effects on mics -- though I can't give an example off-hand, i faintly recall reading about subtle impacts that can be exerted on the mic by the pre-amp circuit (as in, on the physical mic and its responsive properties and/or resilience, not on the mic signal per se), so I have thus far bitten my lip in reaction to this "trick" in case there is something subtle about which i am ignorant.

What say ye? Persistent myth? Kernel of truth to it? The gospel?

Anyone else heard this theory?

Thanks!
-c

[Scodiddly]

Myth.

But you'll even find people out there who believe it's better to line up all the faders at 0dB and literally mix using the preamp gain knobs.

[thethingwiththestuff]

never heard that, but no. here is what you should worry about...

spend a lot of time learning to recognize ringing frequencies, and dont eq out more than you need. a parametric on a digital board will get you closer than a 31-band graphic.

keep stage volume at a decent level.

MAKE SURE YOU ARE NOT SENDING POST COMPRESSION. could require some trickery on certain boards.

"a little bit of everything, i guess..." is never the right answer to "do you need anything in your wedge?" don't send things to monitors by default.

and finally, get the hell up on stage yourself and listen to the monitors. talk into the mic. take it off the stand and walk around a little. slowly surround the windscreen with your hands. make an O shape with your mouth and carefully approach the capsule. hear a tone? hum it to yourself all the way back to the console.

[chuckfurok]

Myth.

But you'll even find people out there who believe it's better to line up all the faders at 0dB and literally mix using the preamp gain knobs.

I have never understand this! I've even seen people do this when they are recording!

[MoreSpaceEcho]

i try and track everything so the mix sounds really good with the faders at zero. this approach doesn't make sense when recording to tape, but why wouldn't you do this if you're recording to digital?

[Snarl 12/8]

Myth.

But you'll even find people out there who believe it's better to line up all the faders at 0dB and literally mix using the preamp gain knobs.

Hey wait a minute? I resemble that remark. I always thought this was proper gain staging in a lot of situations. Because, you're essentially removing a gain stage, or two. The way I always start on a mixer, live or other wise, is with the master fader at zero and then all the channel faders at zero. Then I bring up the pres until I've got the right level. Often, if the source is too quiet or too loud I need to compensate some at a fader, either up or down. And sometimes I want to drive the pre more, or less, so I'll set the fader accordingly. But all faders at unity, and things sounding really good, tells me something is going right, tracking or mixing. I don't, generally, make mixing moves up at the gain knobs after these initial settings though. When I'm fading out a song, or bringing down a guitar for a verse or something, I always do that at the fader.

[chconnor]

Thanks all. Armed with your wisdom, I'll be sure to be appropriately scornful the next time someone presents this idea to me. :-)

[inasilentway]

WAIT A MINUTE. FACT. The nonsense about leaving the faders at zero and mixing with your preamp gains is indeed ridiculous in a live setting, but it is also a completely separate issue.

Gain structure is important in the recording studio but in sound reinforcement it's the MOST important thing. Most of the people responding in this thread come from a recording background, where feedback is not an issue. Preamp gain does make a mic more sensitive to what other stuff is happening on stage. Try this: Listen to a source (preferably a singer while a drummer is playing behind them, so you can see what I'm taking about) with a 58 and a mixing board. Turn up the mic pre 10db, then turn it back down and push the fader up 10db. They have different effects on the sound of the microphone. Now picture that instead of a drummer it's an already-cranked vocal wedge.

A very common rookie move is to have the preamp gain too high, and then compensate by chopping out frequencies from the graph, which just cuts your volume more and you have to crank the send more to compensate and then you chop more and so on and so on. Literally a feedback loop. When I first started it took me about 6 months to understand the concept that a graphic EQ is a selective volume control and if you're cutting too many frequencies or cutting an area by more than 6-8db, you might as well just turn the whole thing down.

Many (not by any means all) analog live sound consoles have their knobs laid out such that noon on the preamp knob is about where a vocal mic needs to be, so I'll use that for my example. I generally start somewhere between 11 o'clock and noon and adjust based on the voice in question. I rarely if ever take it past noon; I use the monitor send/fader to determine the level going to the PA or wedges. In a perfect world, if the sources are all good and already balanced it does indeed look like all faders are at zero, but this is rarely the case in club land and often to get the vocals as loud as the band you have to pump the faders (or, as I often do, put a compressor on the vocal bus and use the makeup gain to goose it a little. important that it's on the bus if you're running monitors from FOH because, as mentioned earlier, compression in the monitors can be deadly. With a separate monitor engineer you have a little more leeway with all of this). Cut frequencies for sonic quality and feedback reduction, if you're getting feedback from several different frequencies at once you might just be running it too hot (or the singer is cupping the mic, or the wedge is pointed in the wrong place, or the hi amp on your bi-amped wedge is too high, or a number of other factors).

"But it's not loud enough and the fader/send is cranking!" Then you set the preamp gain too low, and you do need to turn it up. But if the gains are set correctly at soundcheck/line check, you shouldn't have to touch them to mix except to compensate for differences in stage volume (i.e. guitar player switched to the backup guitar which isn't as hot, singer's adrenaline is pumping and he's projecting way more than his nervous soundchecking self, the drummer's shots of Jack are kicking in and he's slamming his snare drum way harder, etc).

And sometimes the band is too loud for their singer's voice or the PA/wedge is underpowered. Acceptance that not every band will sound good despite your best efforts is a key step towards not losing your mind. Some singers are just too quiet, some bands are just way too loud, and some wedges, PAs, and rooms just suck. The first stage of system gain is the acoustic volume of the band playing in the room, and if that step isn't managed properly then nothing else will be.

[chconnor]

Preamp gain does make a mic more sensitive to what is happening on stage. Try this: Listen to a source (preferably a singer while a drummer is playing behind them, so you can see what I'm taking about) with a 58 and a mixing board. Turn up the mic pre 10db, then turn it back down and push the fader up 10db. They have different effects on the sound of the microphone.

Aha, the plot thickens! So, my understanding is that the relevant parameters for inducing feedback are the SPL at the mic and the gain of the system; when the gain of the (entire) system produces SPL from the speakers such that the resultant SPL at the mic is roughly the same or greater than the inciting sound SPL, ringing or feedback results. Ignoring the effects of multiple mics, etc, this would imply that gain at the preamp stage or the volume fader are basically equivalent as far as feedback is concerned. In other words, that increasing the preamp gain does not increase the sensitivity of the mic, but rather the "sensitivity" of the system as a whole (via the preamp), as does increasing the fader (via the volume part of the gainstaging).

Is your argument that the sonic changes that happen with a preamp as its gain is raised or lowered explain the greater or lesser propensity for feedback? It's clear that preamps sound "different" at various gain levels (especially prosumer gear, i take it; that being an argument against mixing with gain knobs), so it's not absurd to me to suggest that this would change the feedback equation; what's not clear to me is that a reasonable increase in gain would, for example, cause a bulge at, say, 5k so significant that the feedback equation would change significantly.

At least in terms of feedback-relevant amounts, the frequency response of most pre-amps are relatively non-changing in the range of normal gain, are they not? When i test this stuff (turn the gain down, turn the fader up) i hear virtually no overall change (again, within reasonable ranges) and notice no reduction in feedback.

A corollary theory we see bouncing around is that the gain affects the pickup pattern (more/less proximity, more/less off-axis response, etc). This is usually "debunked", but I defer to the experts; it would obviously bear on this discussion.

I wish i could remember the "subtle effects" i referred to in the OP about the preamps effect on the mic... i seem to remember posts challenging the notion that the preamp circuit (being "downstream") has no relevant effect on the mic's performance... something about the circuit changing the electrical properties of the physical mic, changing the max SPL it can handle, or something to that effect, and i was wondering if one of those dynamics had something to do with all this.

-c

[JohnDavisNYC]

Myth.

John

[Scodiddly]

Gain is gain. Unless you've got stone-age era gear, the sound and response of the mic will not vary depending on what stage it gets gain from. As long as you're not getting distortion from overdriving a stage or excess noise from some other gain stage screwup, the mic will sound the same. Different gain staging will NOT change the polar pattern of the mic, nor the frequency response.

But anyway, why would I line up all the faders at 0dB to mix? Well, I'm not really using them, and they're probably the most expensive variable resistors in the console with the best resolution and repeatability. Unless I'm doing some kind of silly mix where everything is the same level, I don't see the point. Yes, my hi-hat fader ends up lower than my vocal mic fader. That's the way it should be, because I'm not trying to make the hi-hat as loud as the vocals.

If I set the channel gains to get the optimum blend of headroom vs. noise floor, then I can construct my mix on the faders and of course instruments that need to be quieter in the mix will end up with a different fader position. That's why the console designer specified faders, and why I bought a console with faders.

I mean seriously - if you mix with the faders lined up all pretty, why not just buy a bunch of mic preamps and a summing box?

[Andy Peters]

What say ye? Persistent myth? Kernel of truth to it? The gospel?

Despite all protestation to the contrary, it's a myth.

-=a

[Scodiddly]

When i test this stuff (turn the gain down, turn the fader up) i hear virtually no overall change (again, within reasonable ranges) and notice no reduction in feedback.

Trust your ears, not something you read on the Internet.

[Andy Peters]

Aha, the plot thickens! So, my understanding is that the relevant parameters for inducing feedback are the SPL at the mic and the gain of the system; when the gain of the (entire) system produces SPL from the speakers such that the resultant SPL at the mic is roughly the same or greater than the inciting sound SPL, ringing or feedback results. Ignoring the effects of multiple mics, etc, this would imply that gain at the preamp stage or the volume fader are basically equivalent as far as feedback is concerned. In other words, that increasing the preamp gain does not increase the sensitivity of the mic, but rather the "sensitivity" of the system as a whole (via the preamp), as does increasing the fader (via the volume part of the gainstaging).

Yes, the mic sensitivity is fixed, and whether you get the desired gain at the input trim or the fader or both is irrelevant, except in the sense that the input gain control sets system noise figure.

Feedback is caused when the gain (in the Bode plot sense) exceeds unity. Typically this happens in a frequency-dependent manner, which is why feedback tends to happen at one frequency first.

While people can argue for days about "preamp tonality" and such, the two elements of a sound system that don't have flat frequency response are the microphone and the speaker cabinet. A hot spot in the monitor wedge response means it's more likely to feed back at that frequency. And the commonly-used microphones have a presence peak which is why your SM58 will like to feed back at 4 kHz.

Put another way, if your console channel's frequency response is dependent on preamp gain and/or fader position, then something's very very very wrong with that channel.

-a

[Andy Peters]

Myth.

But you'll even find people out there who believe it's better to line up all the faders at 0dB and literally mix using the preamp gain knobs.

Hey wait a minute? I resemble that remark. I always thought this was proper gain staging in a lot of situations. Because, you're essentially removing a gain stage, or two.

Nope, those gain stages are always there (the fader is a voltage divider in front of a fixed-gain amplifier).

When you trim cold and push the fader up to unity, you also push the circuit noise post-preamp up as well.

-a