Vocal Mic Feedback: Cardioid vs Super/hyper Cardioid

[Craig Vecchione]

 

With some pickup patterns and some mics the proximity effect can be quite strong when you get right up on the mic, my solution was to always get up on the mic and only back off for dynamic effect if needed. Then listen to how the volume and proximity effect differ and make it part of my sound, use it to my advantage. It takes some work but it can do wonders.

 

 

Eating the mic helps because you are increasing the ratio of desirable source (your voice) to the total signal (your voice, plus stage volume, plus noise) at the mic. If you sing softly or are far from the mic, more gain has to be used to get your voice to a given level. The higher gain also increases the gak level behind your voice, and the total sound volume is much higher than if you sing loudly or eat the mic, increasing the change of feedback.

[Bobby1Note]

Bobby (and stratguy), I'll repeat what Don and Andy stated: gain structure has NOTHING to do with feedback. It affects signal/noise ratio. Overall gain is what matters. A given system in a given state can accept a given signal power before it becomes unstable. How you arrive at that power is irrelevent. If your mixer is changing the EQ as gain and fader settings changes, that's probably why you perceive greater volume in one configuration than the other.

 

 

Yeah, I understand that Craig. As I said in my reply to Sam, "gain is gain".

 

I also believe that this entire subject has to be explained a little more in-depth to noobs. One sentence observations/answers/comments, do little to enlighten or demystify the subject for them.

 

In my particular situation (jam I described earlier), our success was due to sensible levels overall. Regarding "noise floor", this was not an issue in our case. The mixer was very quiet (undetectable noise) with channel gain set to 10 o'clock.

[Craig Vecchione]

 

In my particular situation (jam I described earlier), our success was due to sensible levels overall. Regarding "noise floor", this was not an issue in our case. The mixer was very quiet (undetectable noise) with channel gain set to 10 o'clock.

 

 

I believe I stated that gain structure affects S/N ratio, and does not affect feedback.

[Dogoth]

Here's the two sentence answer.

Gain is Gain (changing where that gain comes from, only changes the S/N ratio & distortion specs).

Other than being able to lessen electronic gain due to a louder initial source (I.E. eating the mic), it has NOTHING to do with feedback.

[Bobby1Note]

Geez these threads can get frustrating FAST. How many times now, have I said that I already understand the premise of "Gain is Gain"?????? Twice?? Three times???

OK. let's look at this again. Let's assume that you're in a small room, with a PA speaker, a mixer, and a mic,,,, no external EQ. Now you've picked the perfect placement for your monitor relative to the mic. Monitor is perfectly flat (no spikes). Mic gain is properly set, faders at unity. and the powered speakers' output level is perfectly set. Theoretical Nirvana.

Suddenly, you're just over the threshold of feedback. Could be any reason,,, someone bumped a fader, twisted a knob, or moved the mic, or the monitor, etc. There is a "balance" to that feedback, and it keeps propagating itself until corrective action has been taken (reduce gain somewhere in the chain) Knee-jerk reaction says, you can twist any knob in that "loop", or yank a fader, and achieve the desired effect of eliminating feedback.(channel gain knob, channel fader, master fader, or powered speakers output level knob).

Now, as I see it, each of those choices changes the "balance" that supported the feedback loop, but each of those choices carries with it a specific penalty. Question is, which "penalty" are you most prepared to live with?

Turn down the speaker output volume knob, and your sound is too low Yeah, you have a TERRIFIC signal-to-noise ratio, but who gives a rats'-a$$,,,, it's too low. Yanking back the master fader has the same result. It's a no-go.

Yank the channel fader, and you have virtually the same scenario (lower volume), with perhaps a bit of degradation in your signal-to-noise ratio.

Each of these solutions cured the feedback issue, but resulted in unacceptably lower room volume. That leaves the "crank back the channel gain" option. What exactly is the impact of that choice?

Yes, you reduce overall room volume as well, and, you have a "less-than-optimal" signal-to-noise ratio",BUT if the noise-floor is inaudible, who gives a rats' a$$. You can now turn up the room volume elsewhere. You've changed the "balance" that supported the feedback loop. You've simply reduced the mics' ability to amplify the speakers' output............ No?

[agedhorse]

Bobby... you are missing the gain thing entirely.

Removing gain anywhere in the signal chain will resolve the feedback problem, and the volume will decrease exactly the same. The only change will be in the signal to noise ratio, but lowering the gain by 3dB isn't going to cause any meaningful noise problems anyway.

[Shaster]

Now, as I see it, each of those choices changes the "balance" that supported the feedback loop, but each of those choices carries with it a specific penalty. Question is, which "penalty" are you most prepared to live with?

I found a reference to what I was talking about before

Feedback will occur at unity loop gain - that is, when the loudspeaker produces the same level at the microphone as does the (singer's) voice.

Yamaha SoundReinforcement p. 48.

If you're on the verge of feedback, and turn something up, and then get feedback.... just turn it back down.

As AH has stated, the amount of reduction would be small so it's unlikely you'll wreck your sound by destroying your signal to noise ratio..

I normally start with the channel fader, seems like the logical place - others may differ. As long as you turn it down, or reduce the offending frequeny(ies) you're good to go.

[Bobby1Note]

I found a reference to what I was talking about before Yamaha SoundReinforcement p. 48. If you're on the verge of feedback, and turn something up, and then get feedback.... just turn it back down. As AH has stated, the amount of reduction would be small so it's unlikely you'll wreck your sound by destroying your signal to noise ratio.. I normally start with the channel fader, seems like the logical place - others may differ. As long as you turn it down, or reduce the offending frequeny(ies) you're good to go.

Well geez, That's EXACTLY what I've been saying all along, inn't. Reduce the channel gain slightly to where there is little-to-no negative impacts on signal-to-noise ratio(inaudible),,,, and you can turn up the room volume. If the singer has to sing a little louder into the mic, it's not contributing to a feedback loop, since the singer, is the "source" of the sound.

[Bobby1Note]

My Brain hurts. I can see I'm really gonna sleep well tonight.

Don't worry though, I'm workin' on another (possibly/probably) scatter-brained explanation of how I'm understanding this (or not). :poke:

[lifeloverwg]

Well geez, That's EXACTLY what I've been saying all along, inn't. Reduce the channel gain slightly to where there is little-to-no negative impacts on signal-to-noise ratio(inaudible),,,, and you can turn up the room volume. If the singer has to sing a little louder into the mic, it's not contributing to a feedback loop, since the singer, is the "source" of the sound.

That sums up what you aren't getting.:poke: If you turn it down at the channel gain and then turn it back up the same amount on the "room volume" you are right back where you started as far as the system is concerned. As far as feedback is concerned, total gain is total gain no matter where in the signal chain it comes from. When we ring out monitors with EQ we reduce system gain by the amount you cut the EQ.

I have experienced situations where the channel strip gain setting along with other factors could lead you to think that way though.

We played a festival once where the providers approach was to max out the channel gain in order to "get a good strong signal to work with". I'm talking input gain of +10 - +12 dB before anything else in the chain happens. Monitors were Mackie 450's with the gain set wide open and no EQ. This was also one of those small town affairs where band BE's weren't allowed to mix because "they had had too many problems with feed back in the past"....

FOH wasn't a problem feedback wise all day, but anytime somebody asked for just a little bit more of anything in the monitors they would howl like crazy. (Monitors were plenty loud!) With both sides of the signal chain set the way it was, the sensitivity in the monitor send was magnified to the point that any small adjustment up in level was too much. But the problem was still all about the total gain............ What the muso's should have asked for was for something to be turned down so that they could instead hear better what they needed/wanted to hear.

Like that would happen, Winston.

Edit: for more and different.

[AcousticRock]

My drummer son and I have just started a duet ala the Black Keys or White Stripes. Our rehearsal setup for the moment is: - Acoustic drums - Guitar through my cranked 17 watt Goodsell - Me on vocals into Allen & Heath MixWiz 12 and out to two QSC K12 floor monitors. I'm having trouble getting enough gain before feedback with my supercardioid AKG D870 dynamic mic. It's a great little mic for the money and has served well on acoustic gigs. But it doesn't seem to be cutting it here. In fact, I have to put my amp across the room facing us for practice or it really howls. Wondered if a cardioid pattern would be better since it supposedly wouldn't pickup anything behind the mic (i.e., monitors). And I'd like to move the amp back over near me like it would be at a gig. In particular, I'm interested in the Shure 55 SH mic which is cardioid. http://www.shure.com/americas/products/microphones/classic/55SH-Series-II-iconic-unidyne-vocal-microphone I'd appreciate experienced opinions and guidance.

 

 

I frequently play in small spaces and at times with loud guitars and drums. In these situations I have found the following to work well (in the order I prefer):

 

-put monitor in the correct position based on your mic's polar pattern

 

-reduce stage/monitor volume (get drummer to play softer and turn guitars down)

 

-use a mic w/ a tight polar pattern (my favorite for small spaces w/ loud volume is Audix OM7 or Beyerdynamic TGX60)

 

-add a 31 band EQ on your monitor (for gigs where I won't have time to ring my monitor out I have a Shure DFR22 that can grab offending frequencies if feedback occurs while I'm playing and I can't get to the 31 band eq myself)

 

-last but not least, switch to in-ear monitors. If you're using a decent mic, a 31 band eq and your monitor is still feeding back,you're at levels that can permanently damage your hearing. Consider switching to IEM's, adjust volume to taste and say goodbye to feedback and hearing damage. (Of course IEM's can also damage hearing if turned up too loud)

[Bobby1Note]

Well, I went to bed, tossed and turned for a few hours, and just may have finally figured this thing out.

We're not talking solely about "speaker volume", but rather, speaker volume measured at the mic aren't we.

When speaker volume measured at the mic meets and exceeds the source volume (singers' voice measured at the mic), feedback occurs. Is that it?

That's where the confusion was for me. I was thinking that a mic works like a speaker, but "in reverse", and by turning down the channel gain, you decreased the mics' ability to "hear' the speaker if there was distance between the mic and the speaker. In that instance, if the singer is singing louder into the mic, you could turn up the speaker volume, as long as it doesn't meet or exceed the same level as the singers' voice when measured at the mic.

Well,,,, Do I "pass" ????

In actuality, I'm thinking when any frequency from the loudspeaker, measured at the mic, exceeds any frequency from the singers' voice measured at the mic, that frequency will cause feedback.

[Pater Familias]

I frequently play in small spaces and at times with loud guitars and drums. In these situations I have found the following to work well (in the order I prefer): -put monitor in the correct position based on your mic's polar pattern -reduce stage/monitor volume (get drummer to play softer and turn guitars down) -use a mic w/ a tight polar pattern (my favorite for small spaces w/ loud volume is Audix OM7 or Beyerdynamic TGX60) -add a 31 band EQ on your monitor (for gigs where I won't have time to ring my monitor out I have a Shure DFR22 that can grab offending frequencies if feedback occurs while I'm playing and I can't get to the 31 band eq myself) -last but not least, switch to in-ear monitors. If you're using a decent mic, a 31 band eq and your monitor is still feeding back,you're at levels that can permanently damage your hearing. Consider switching to IEM's, adjust volume to taste and say goodbye to feedback and hearing damage. (Of course IEM's can also damage hearing if turned up too loud)

Read my mind! Started a thread yesterday on how to go about setting up an IEM.

Will definitely try to rearrange the wedges also. Can't believe it didn't register with me before that that's why I've seen wedges angled on stage for years. Started paying attention to the monitors yesterday while watching an Old Grey Whistle Test greatest videos DVD. Needless to say I had a Homer Simpson moment. Doh!

Funny thing is that I played bluegrass/acoustic music for a few years with a bassist that has 25 yrs of gigging experience. He would not only just have the monitors set straight towards the mics but also put his bass amp shooting right up my ass with me being on acoustic guitar. Then he would bitch about my guitar was what caused all the feedback problems we had. He would keep eq'ing me out of the mixer and everyone would wonder why no one could hear my guitar. I tried to explain and demonstrate to him repeatedly that hey it's your bass amp hitting my guitar and vocal mike that's feeding back, but even with me showing him that it fedback even while holding my strings clamped on the neck he refused to believe it was anything but me.

[lifeloverwg]

Well, I went to bed, tossed and turned for a few hours, and just may have finally figured this thing out. We're not talking solely about "speaker volume", but rather, speaker volume measured at the mic aren't we. When speaker volume measured at the mic meets and exceeds the source volume (singers' voice measured at the mic) , feedback occurs. Is that it? That's where the confusion was for me. I was thinking that a mic works like a speaker, but "in reverse", and by turning down the channel gain, you decreased the mics' ability to "hear' the speaker if there was distance between the mic and the speaker. In that instance, if the singer is singing louder into the mic, you could turn up the speaker volume, as long as it doesn't meet or exceed the same level as the singers' voice when measured at the mic . Well,,,, Do I "pass" ???? In actuality, I'm thinking when any frequency from the loudspeaker, measured at the mic , exceeds any frequency from the singers' voice measured at the mic , that frequency will cause feedback.

Your getting closer. But remember, you can make a system feedback without anyone singing or playing a note. Especially in a small space. People often ring out monitors without any body singing into the mic. If I want to do that, I just keep raising the gain on the monitors until the first one or two rings occur and then cut them out.

This is where good mics and quality monitors make a big difference. If you happen to have crappy monitors with a very peaky response that lines up with say the presence peak on a vocal mic, then that crappy monitor is going to feedback sooner than a better monitor with a smoother frequency response.

Winston

[lifeloverwg]

Read my mind! Started a thread yesterday on how to go about setting up an IEM. Will definitely try to rearrange the wedges also. Can't believe it didn't register with me before that that's why I've seen wedges angled on stage for years. Started paying attention to the monitors yesterday while watching an Old Grey Whistle Test greatest videos DVD. Needless to say I had a Homer Simpson moment. Doh! Funny thing is that I played bluegrass/acoustic music for a few years with a bassist that has 25 yrs of gigging experience. He would not only just have the monitors set straight towards the mics but also put his bass amp shooting right up my ass with me being on acoustic guitar. Then he would bitch about my guitar was what caused all the feedback problems we had. He would keep eq'ing me out of the mixer and everyone would wonder why no one could hear my guitar. I tried to explain and demonstrate to him repeatedly that hey it's your bass amp hitting my guitar and vocal mike that's feeding back, but even with me showing him that it fedback even while holding my strings clamped on the neck he refused to believe it was anything but me.

Some people forget or never knew, that a mics pattern is a "polar response", meaning that it goes all the way around the mic. On a cardioid mic like the SM58, for best rejection you want the back of the mic where the cord attaches pointed at the monitor wedge or whatever other sound source you don't want getting into the mic. With a super cardioid like the Beta 58 there is a small pic up mode pattern directly behind the mic and the null is at an angle of something like 28 degrees off of that.
For the best rejection of the monitor sound, you could leave your mic setup like with the SM58 and move the monitor off to the side, or you could leave the monitor in the same place as with the SM 58 and instead change the angle of the mic so that the back end with the cord attachment is now pointing straight out at the audience. Either method achieves the same goal of placing the monitor in the mics null point.

Winston

[fishmanrod]

Lots of good thoughts, here.

Feedback is weird stuff.

Often in bars you end up with much less than optimal speaker placement, and that causes issues. I mixed a country band a month or so ago with a very inexperienced singer, who would not approach the mic, or sing loud. She could sing loud, she just chose not too. One speaker tower ended up a bit behind her (on the side of the "stage" but behind) and I was messing with it all night to keep her loud enough in the mix and keep the feedback down. The monitors were fine, and actually I mixed her down in the monitors and she started to eat the mic more, and that helped. I EQ'd like crazy but sometimes it is all speaker position. I could have done better if I had had a sound check, but it was all done on the fly. Sometimes it's hard to even figure out where the feedback is coming from when you have the whole band going.

I also had an issue one time when the drummer kicked the kick drum I would get feedback. I eq'd the kick mic out, gated it, compressed it, could not get a good sound or enough volume without this howl or resonance that made the kick sound awful. It turned out to be several things. The feedback was NOT on the kick channel, it was coming from a guitar amp mic next to the kick, that was on a short stand sitting on the boxy boomy stage, and had high gain because the guitar player did not like to run his amp very loud (the only one I ever met). I turned the guitar players amp up a bit, put a pc of foam under the mic stand, and reduced the gain on that mic and the feedback went away.

The funny thing was I saw a very similar problem occur at an out door show on a big stage with a huge PA. The bass amp was right next to the kick drum, and every time the bass player would play they would get this huge low frequency resonance. The sound man was going ape, because he could not figure out what was going on, because the bass was DI'd, how could it feed back? Well, it could through the mains if cranked enough, but the bass itself was not that loud (and at one point totally shut off in the FOH, but there was still bass in the FOH, he's going WTF???), but the feed back was huge! In frustation, someone turned the bass amp off, and the feedback went away. So, some smart guy changed the angle of the bass cabinet, and the problem got better, then another smart guy re-equ'd the kick mic, and the problem was fixed. The bass amp was cranked, and the kick mic was picking it up, this sound was louder than what was coming from the kick, so when the kick mic picked up the bass, it fed back due to how loud this was in the mains. Weird, eh?

My advice to anyone doing sound is:
a)make sure you have some sound check time to ring out your monitors, insure you will not get feedback in any situation, and you have some room to boost without feedback. Sometimes you have to ring out the room, too! Feedback can occur from the FOH, too.
b) have good speaker placement. This saves you more times than anything else, but sometimes in tight quarters it's hard to get what you want.
c) look for other stuff, ie, a secondary mic that is causing problems. Use your mute buttons to kill channels till you find the where the feedback is.
d) spend the dough on an eq dedicated to each monitor channel.

[Dogoth]

Some people forget or never knew, that a mics pattern is a "polar response", meaning that it goes all the way around the mic. On a cardioid mic like the SM58, for best rejection you want the back of the mic where the cord attaches pointed at the monitor wedge or whatever other sound source you don't want getting into the mic. With a super cardioid like the Beta 58 there is a small pic up mode pattern directly behind the mic and the null is at an angle of something like 28 degrees off of that. For the best rejection of the monitor sound, you could leave your mic setup like with the SM58 and move the monitor off to the side, or you could leave the monitor in the same place as with the SM 58 and instead change the angle of the mic so that the back end with the cord attachment is now pointing straight out at the audience. Either method achieves the same goal of placing the monitor in the mics null point. Winston

 

 

Basicly you are correct but even an SM58 has a small pickup node exactly 180degrees off axis from the address side. The way a mic get's it's directionality is through the porting of the sound to the back of the element (or at least into the resonant cavity that drives the digaphram). This ported sound is (to the digaphram) 180 degrees out of phase with the front, therefore cancelling out the sound from the rear. There are all flavors of directionality ranging from barely directional (the best place for the monitor is quite close to the back of the mic (where the cord plugs in)) to super directional (a figure 8 pattern where the lobe in the back is as big as the one in the front and the best place for the monitor is 90 degrees off axis (not usualy recomended for SR)). Also the more directional a mic is the more proximity effect (that bass boost you get when near the mic). The best angle for the monitor (as you pointed out) depends on the mic, and the offending frequencies, but it is rarely exactly 180 degrees off axis (although sometimes it's close).

 

The afore mentioned technique is only a rule of thumb though. With differing acoustics, this all can change because you're trying to get the best null point of the mic's pattern pointed at the loudest SR source (which can sometimes, at certain frequencies, be a reflection rather than the speaker it's self). Life can get interesting in a live environment :-)

 

Cheers

[Shaster]

Or a singer can just cup the mic, "cancelling" the null, and thus causing feedback... had that happen once too often.

EDIT: Okay cancelling might be the wrong word, rendering the null's feedback rejection null and void? Oh boy, going to get coffee.

[Dogoth]

Or a singer can just cup the mic, "cancelling" the null, and thus causing feedback... had that happen once too often. EDIT: Okay cancelling might be the wrong word, rendering the null's feedback rejection null and void? Oh boy, going to get coffee.

Never happened