horrible feedback no matter what i do. help!

[revstate]

so, my wife and my project will be playing our 5th or 6th show (i can't remember right now) in a little while. problem is, no matter what i do i can't seem to be rid of feedback, really bad feedback. we have tried a sabine, and a behringer feedback eliminator with no luck. we purchased bet57a mics for both of us because they were supposed to have the highest gain before feedback of any mic on the market (at least that is what we were told). we have tried a variety of different fx processors itb and otb, still with no luck. we never have monitors on stage because it makes it much worse. and i am just stumped.

our fx consist of the following:

noise gate (useless because the pa is always too loud to gate out unless we scream our vocals)
compression
distortion or overdrive (digital and analog)
chorus
delay
and reverb

this time i am using my ensemble with it's pres and mainstage 2 for vocals and playback, then going to DIs to go to the snake and then to the foh. same problems, feedback, feedback, and more feedback. i have even purchased a cb mic (sound really cool btw) for the temporary on switch. this does not fix the problem as i sing too long of phrases before i can let go of the switch. three words in, and the feedback is there.

what is causing all of this feedback? i know distortion causes feedback, but i have been to shows where the band was using distortion and they were not having feedback problems. i know mic positioning is important, but i have watched shows were the lead singer gets off stage with a wireless mic and runs around the crowd singing to them (not very pleasant for me, but you get what i am saying). delay in itself is feedback (bbd), but plenty of bands have delay on there vocals. and chorus and verb are commonplace with tons of bands vocals.

where am i going wrong, is there alternatives to what i am doing to create the same effect? my anxiety for the next show is growing in me like a monster because i can't seem to fix this problem.

PLEASE, PLEASE help

-ben

[ghellquist]

As always, a bit of method in the madness might help. My guess is that the combination of compressor and distortion is the main culprit. ( Distortion and chorus on voice by the way, odd in my world? )

Start by using no fx at all. You should be able to reach a decent level without feedback problems singing in the mics. If not, there might be something else wrong with the PA.

Then add fx one at a time and try again. Compression can create all kinds of feedback problems unless you set it just right. Can you live without the compression or could you set it up differently?

// Gunnar

[John Hedger]

I use a very straightforward non-technical PA setup for my blues band and jazz trio, with 200W main/200W monitor amps, 8 channels, using Sure SM58's mainly. I have a 10-band EQ on both my monitors and mains. I know EQ settings can affect the feedback threshold quite a bit. I use either a smile or a sine wave eq shape usually and tweak it more on the gig. I try and go for as natural sounding vocal EQ as I can get with no feedback. You might try and note what frequency range your feedback is howling at...Is it around A-440? (very common) or in the low to midrange? Set EQ accordingly. As I said, I have 10-band EQ, but it would be much better to have a higher band count for subtle tweaking. Good luck!

[Enginearing]

can you provide a diagram of how you're using these effects? (how/where in the chain etc.) i suspect the culprit is your Dist/OD and/or comp but knowing things like the console/speakers you're using is also going to help.

Gain Staging - pretty much this is what its all about. (also make sure you're not doing any crazy boosting with EQ...)







oh and John... a sine wave eq shape? is this something you developed after looking at a guitar centre catalog? start flat, go from there...

[NorseHorse]

Also, where are you in relation to your speakers? Position is just as important (if not more) than gain staging in reducing feedback.

Are you using monitors? Where are those located?

Last time I met a band with horrible feedback, it was because they were running both their MAINS and MONITORS from the "master" outs on their PA head. Meaning every time they tried to turn up the mains, they also were turning up the monitors. And they couldn't EQ them differently.

[John Hedger]

Originally Posted by Enginearing

oh and John... a sine wave eq shape? is this something you developed after looking at a guitar centre catalog? start flat, go from there...

Guitar Center!? That was a cheap shot. I know my advice was vague, but no more vague than starting flat and going from there... just that any EQ settings generally sound more musical when sliders are set in some kind of a curved fashion, as opposed to having everything flat and then moving only the 500 hz slider down 40 db. For vocal mics, there should be a delicate balance of EQ setting vs channel EQing. I always go by ear. If the vocals sound nice and natural, with no feedback probs... then I don't fool with it. I have come close to firing some sidemen for messing with my EQ! Usually my EQ does end up in some kind of wavy pattern. Hard to give any real advice about this in a forum post. I have no real clue yet what kind of power or PA this fellow is running...and of course there is the factor of what kind of rooms he is playing.

I usually prefer using my own PA system over house systems that cost five times what I've got, mainly because often their settings suck, or their speakers are half blown or their mics have been dropped numerous times, or their PA tech has no clue how to get good monitors without feeding back. Many house systems sound dull to me. I would rather take the extra time to set up my own PA which I know sounds great...I know I'll have monitors, etc..etc..

[Enginearing]

wasn't a crack at you mate... GC if anything... you obviously know your boxes and how to get your PA to sound the way you need to... i could tell that from your 1st post. the octave EQ is probably wide enough that you would end up with a similar curve each time, to get the results you want.

I do however shudder when i see catalogs with 31 bands of EQ with 3 or 4 waves up the graph... i just know that there's people that take that as an audio lesson and use it when they setup.

[revstate]

so if i use compression, it can and most likely will cause feedback? does a limiter on my main outs cause similar problems?

as far as gain staging, i will be only sending the final "mix" of everything plugged into my ensemble, so the only thing i can do is control the mic input on the ensemble and the volume that Mainstage.

as for the monitors... i have never been able to use stage monitors live because of the feedback.

and then, the problem with speaker placement... doom, utter and horrible doom. every show we play the speaker placement is different, and BAD. one of our shows, the speakers were behind the front and on the stage, and this meant we had to stand behind the stack, leaving us with a few feet between the back of the mains and the wall of the venue. another show, the stage was in the corner of the building and the mains were lined up at a 90 degree angle with the back left of the stage (feedback even if we turned off all fx and ran dry vocals). and with this next show, we have no idea what the venue will be like, and the promoters never listen or read our tech rider, so you can tell how much they care. one time, the front of house engineer<---- was sitting behind the left speaker and wouldn't listen to us when we tried to explain that he needed to be in front of the stage, not behind it.

all of the fx are in line and inserted logic plugins (3rd party eats my little macbook's cpu like crazy). here is how they flow for my vocal chain:

eq (just highpass filter, and a little lowpass filter) then compression then overdrive then delay (all inserts), and a little send to some verb

my wife's vocal chain is: eq, compression, chorus, reverse ir reverb, delay and a little bit of a send to verb. then her second vocal chain for some different songs is: eq, ensemble, distortion, chorus and then delay with some send to verb

thanks!

[63hz]

Originally Posted by John Hedger

I use either a smile or a sine wave eq shape usually and tweak it more on the gig. As I said, I have 10-band EQ, but it would be much better to have a higher band count for subtle tweaking. Good luck!

sorry i broke up your post abit, john you would be better off with a few channels of parametric eq instead of a 10-band, you basically pulling alot more than you need to get the sound your after and will be loosing headroom in your system, on a graphic eq you will be overlapping frequencies your cutting and the result if done with an rta and pink noise you will find is a very jagged eq.

[GZsound]

It seems that something else is wrong here.

I would start at the beginning and see what is causing the feedback.

First, just hook up the mics and run to the FOH system. No effects, no EQ, etc. and see what happens. If you use monitors, bring them in after the FOH mix is working properly without feedback.

Then start adding all the additional effects. Add one at a time and see which one is causing the feedback.

Something in the signal chain is causing the problem.

So, my suggestion would be to start with the most simple setup and make that work properly then add any effects, processing (distortion AND compression to voice?) and see what is happening.

Get a good non feedback sound in the FOH system before bringing in monitors.

[Remoteness]

I believe it’s more about the relationship between your microphones, amplifiers and speakers and not so much the effects you’re using. The potential for feedback occurs when the sound from your speakers loop back into your mics.

Proper gain structure between all gadgets involved is paramount.
Do you have enough gain before feedback?
The more power you have the less chance for feedback.
Do you have enough power to drive the system correctly?
Maybe it is as simple as that; your system is not large enough for the dates you're doing.

Perhaps JvB can chime in on this one, since he's also a (real) live sound engineer.
I just play one on TV.

Check this reference out: Audio feedback - Wikipedia, the free encyclopedia

[mixedupsteve]

If the speakers are behind the mics you will have feedback. Same with corners and near the back wall. The FX will make it worse but it does sound like you have a very basic setup problem. Small rooms are very difficult and even with a good setup you may have to go with a lower SPL than you want.

[bishopthomas]

I don't think there's anything more to be added here. You have so much processing in line that you really have to be careful about your gain structure. Process of ellimination is your best bet. Go forth and report back...

[fifthcircle]

I find that with a well tuned system, it really doesn't matter where the mics are, you should be able to keep it from feeding back. I've mixed live vocals in front of the speakers at 95dB SPL without issue.

That said, so many systems out there are not tuned properly. Add to that sub-quality speakers and things get even more difficult.

Most of the important stuff has been said here- a 30band EQ will get you a long ways to notch out feedback points. Choosing the right mic will also help. Some mics are more likely to feed back than others. Also effects can remove gain before feedback as well. If you have too much reverb in the mix, you can have big issues.

--Ben

[Remoteness]

Well said, Ben! thumbsup

[unitymusic]

I believe the solution to your problem has already been stated in this thread... but anyway... if you get a graphic EQ (31 band) you should be able to eliminate all of your feedback problems with that.

If that doesn't work, I would take a look at the quality of the effects processors you're using. I'm not sure if I missed something but I don't believe you mentioned specific products, and I would think using that many effects if they were not high quality would produce some less than desirable results. Most of the bands that use the effects you're referring to on vocals have FOH engineers controlling the effects with high quality outboard gear using balanced lines, which is quite different than trying to rig everything to be controlled on stage, if that is what you are doing.

[fallforward]

Have you tried lowering the output level from your rig that you are sending to the house? If that doesn't work try turning everything the else down.

You have a lot of gain stages in front of the console, and any one of those could muck up the whole signal.

I played sax in a rock band where I used a whole pedal board of efx, I had to add a line console before the di to pad the output because it would feed back constantly. So my flow was:

Mic>Mic Pre>Pedal Board>Line Mixer>DI>FOH

I had to be very diligent in my gainstagin to make sure I wasn't sending too hot of a signal.
Robby

[silverking]

The effective solution to the OP's original problem, as stated above, is a 31 band (1/3 octave) EQ.

Every time you output sound through a speaker in a closed room, you create a single system comprised of the speaker and the room.
That single system has a pre-defined amount of gain it can accept before the system enters into feedback.

Oversimplified, when you notch individual problem frequencies out with the 1/3 octave EQ, you create additional gain for the remainder of the frequencies you haven't attenuated.

You will often find that you're attenuating frequencies which are related to each other (octaves), and you should exercise caution that you're not reducing the overall gain of the system, something that's very simple (even for experienced live engineers) to do if you're not paying attention.

It takes a knack, but the 1/3 octave graphic EQ is the standard tool for live audio feedback control. Although a parametric EQ would work for feedback attenuation, the 1/3 octave graphic is far simpler to use for those who may not have extensive live mixing experience.

No offense to those who own them, but the "feedback suppressor" type devices are really trying to calculate rough approximations of those frequencies that are prone to feedback (through internal processing), and applying very steep parametric EQ's to those frequencies. Although sometimes effective, they're far more often quite a ways off the mark, and only succeed in reducing your overall gain, or making your system sound like crap.

You're basically doing the same thing with the 1/3 octave graphic, but you're using your own ears to hit each and every frequency needing attenuation, and leaving those frequencies that aren't problematic alone, which increase your overall system gain without any feedback.

BTW, compressors either on the system, or on the mics causing feedback only add to your problems if you're not effectively EQ'ing first. More often than not, they'll apply compression as per your settings, and in doing so reduce feedback to a point where the feedback will come on like suddenly, and like gangbusters, rather than giving you a quiet little ring first to let you know it's on its way.

Hope that might help.

[GZsound]

I retired last year from playing in a classic rock and roll show band with a female singer.

Our singer never got on stage and did all her songs out in the audience with a wireless mic. In other words..in front of the FOH system.

We never had feedback problems unless she got way too close to the FOH speakers. Proper gain setting is critical.

I never had a 31 band EQ in my system either, just dual 15 band EQ's that rarely got used.

We did outdoor concerts for 4,000 folks with contractor provided PA systems that could handle that many people and never had any feedback problems.

And I have been running sound for bluegrass festivals for 12 years and use a feedback destroyer with excellent results. The bands typically insist on a single large diaphragm condenser mic and I have found that "tuning" the venue with the feedback destroyer gives me substantially more headroom with the LD mic than without. I save the settings as a preset and just dial it in whenever a band wants a single LD mic.

Also, two of my sound clients are "track singers" who sing to backing tracks. Both of them frequently sing in front of the FOH speakers and as long as they stay past a certain distance, there is no feedback.

It can be done. But setting up the system needs to be done right.

[fifthcircle]

I'll add to my previous post that a crappy 31band EQ won't do you much good either. The phase shift from a massive cut can really screw with the sound.

One venue where I work has the DBX low end EQs (231?). I've found with those EQs that the sound changes even when bypassed. Add to that, the +/- 6dB sounds MUCH better than the +/- 12dB setting.

If you make a system sound really hi-fi, chances are you won't have major feedback issues. Remember, if you over-EQ, you just make the system quieter and you really aren't helping yourself.

--Ben

[silverking]

It may be worth pointing out that folks who are posting and stating that they don't use a 1/3 octave EQ in their systems, and yet have excellent sound are not operating their system past the point of maximum gain before feedback.

This is obviously the most desirable way to operate a sound system.
It's not always possible though, due to a variety of factors, so the 1/3 octave graphic is simply a tool for those times when it's required to operate your sound system past the maximum point of gain before feedback.

[fifthcircle]

All systems should be run lower than their maximum gain before feedback. If you are running higher than that, you have feedback which should always be avoided. Given a choice, I usually tune my systems with parametric EQs. Then, if there is a problem point or two, I may hit a band or two with the 30band, but I even EQ my monitor rigs this way. More often than not, I don't even hit the 30 band EQs. The parametric does what I need.

Tuning a system is about more than just knocking out feedback.

A well tuned system gets more than you can imagine.

I'll say it again. A well tuned system gets you more than you can imagine.

When mixing outside in the city, I have to deal with max SPL issues all the time. For the big rock acts, we run into that all the time. However, I can mix a well-ballanced and well EQ'd show at 93-96 dB SPL and I'll have folks think that I'm mixing over 100. The imact from a tonally balanced system and mix is MUCH more than you get without. The same goes for monitors. If the monitor mix is muddy, it doesn't matter how loud it is, the performer won't be able to hear. Make the mix clear and you can run your monitors much quieter.

-Ben

[DarbyOhara]

If I missed somoething in the first post, then excuse me. But I wouldn't even begin to try to debug a feedback system without knowing products and placement. That is just silly. You can write until your blue in the fingers, but very little is gonna help this cat. The best advice would be to start from scratch (without any effects) and find the cause. But my first quest would be... what and where is the mixer/power amp(s) from the speakers. I've had some large Behringer PA speakers that I could never get to use in small venues. Something about they way they are built. Now with the same venue, pop down a couple of Peavy PA speakers.. and man! You can crank it it. Same thing about a mixer. I had nice Behringer mixers... but pop down a Yamaha mixer... and hear the difference. I am not saying all Behringer are bad, I am trying to make a point about products in general. You honestly get what you pay for. Lot's of studio folks don't like to hear it... but they know it is true.

But I would agree with a few others... start from scratch. Turn all effects off... even devices and slowly turn them on until you find the culprit.

[revstate]

Originally Posted by DarbyOhara

I am not saying all Behringer are bad, I am trying to make a point about products in general. .

I am using the ultra di pro (di4000) to go from the ensemble to the house snake. Can this cause feedback problems, you know, if it is a not so good product?

[revstate]

I won't have a 31band eq for this show, so it doesn't matter if it would work or not. I found the venues website, and this is their gear list (not very fancy).

STAGE

The hi-dive stage is approximately 16 feet wide by 16 feet deep and is elevated 3 feet from the ground. We have a Sony DVD/VHS projector with a retractable 7′ × 9′ screen.

SOUND
Mains: stereo system
2 x tops: 1 per side
each cabinet: Production Services 3 way 2×15, 1×10, 1×2
loaded with DAS components
powered by QSC
2 x subs: 1 per side
each cabinet: 2×18
loaded with McCauley components
powered by Crest
processor DBX Driverack PA

Monitors:
4 mixes from FOH
4 x Community 2 way boxes (1×15, 1×1)
1 x drum monitor: B52 (2×15, 1×1)
powered by QSC

Front of House (FOH):
Console: Mackie SR-24 VLZ
20 mono channels with inserts
2 stereo channels
4 monitor sends
2 effects sends
3 band EQ (mid sweepable)
4 submixes with inserts
19 channels from the stage
DBX EQ on mains
Yamaha EQ on monitors
8 channel Presonus gate/comp
2 channel DBX 266 gate/comp
Yamaha Rev.7 Reverberator Digital Reverb

Mics/DI:
4 x Shure SM58
3 x Shure Beta58
4 x Shure SM57
1 x Audix D6
1 x Audix D1
2 x Audix D2
1 x Audix D4
1 x Audio Technica Pro25
4 x DI boxes

[revstate]

Originally Posted by fifthcircle

I find that with a well tuned system, it really doesn't matter where the mics are, you should be able to keep it from feeding back. I've mixed live vocals in front of the speakers at 95dB SPL without issue.

That said, so many systems out there are not tuned properly. Add to that sub-quality speakers and things get even more difficult.

Most of the important stuff has been said here- a 30band EQ will get you a long ways to notch out feedback points. Choosing the right mic will also help. Some mics are more likely to feed back than others. Also effects can remove gain before feedback as well. If you have too much reverb in the mix, you can have big issues.

--Ben

this is a one off show for us, because i cannot play many live gigs due to my health. and as i have stated, all of previous shows and this one have poorly tuned systems, bad rooms, bad engineers, well... bad everything. because we are an "industrial band" we get treated lower than dirt ("you don't have a guitar, what the f*ck is wrong with you?" attitudes everywhere). i wish i would have had more time to talk to trent's (nin) engineer. he said that the red rocks show had horrible problems, but i didn't notice a thing. real life live engineers are just amazing, dealing with a different monster venue every night, and just the shear number of mics for that show... jesus christ. that was the best large venue show i had ever seen, and when he said that he had so many problems... i just couldn't believe it. our shows have such a minimal setup to prevent problems, and here he is with gear piled around him, saying he had a bad show, and it sounded fantastic.

thanks for everyones responses here, you have all been great. wish us good luck and take care!

-ben

[LX3]

I'll admit now that I haven't read the entire thread... so someone may have already mentioned this, but one suggestion for anyone suffering from feedback issues:

(And you may already be doing this, but just in case)

Make sure you sing tight on the mic. No more than an inch away. That way the gain on the mic can be lower for the same output level, and lower gain means less (hopefully no) feedback.

You'll also get less spill in the vocal, and therefore a better live mix.

Just today I finished mixing a gig by a new-ish but very successful US band, and there's more drums and bass in the lead vocal mic than there is vocal. Everything sounds fabulous until you put the vocal in, at which point the whole thing turns to mush. Not unsalvageable, but disappointing.

Then I saw the footage from the show. Came as no surprise to find that the singer is singing a good foot away from the mic. This band is also renowned for playing very loud on stage (even though they're not what you'd consider a "loud" band). Not a good combination. Makes you wonder why, after two years solid touring with some very "particular" live engineers, no-one's mentioned this issue to them...?

[mixedupsteve]

Originally Posted by revstate

I won't have a 31band eq for this show, so it doesn't matter if it would work or not. I found the venues website, and this is their gear list (not very fancy).



DBX EQ on mains
Yamaha EQ on monitors

I'll bet the DBX is a 31 band and the Yamaha might be too.
Here's a down and dirty way to help with feedback:
http://www.techietalk.co.uk/articles/eq-pa/

To help with identifying frequencies you can make a CD from here:
Michael Knowles: Extras

If you get to know the freq then you can cut the eq there.

[Enginearing]

that spec may not be fancy but surely enough to get the job done well.... I've had plenty of successful shows using gear like that at local venues. you may even want to split your signals to give the house engineer a few more channel options instead of just LR...

[WDG]

I would recommend that you

A. check to make sure that the internal mic on the laptop is off.
B. I am not familiar with the FX that you are using but make sure that you aren't creating a feedback loop with in the the FX on the laptop. Sometimes digital will let you do goofy things
C. Make sure that all mics are behind the house speakers.
D. If A,B,C fail to stop feedback. Try to make your set up as simple as possible to eliminate problems then add on till you get feedback.

Warren