Feedback in PA systems

Sorry, don't have much more time for today, but tomorrow morning should be good. Regarding cone resonance, basically we can think of it as a spring-mass system. An often easy way for electronics guys to understand that involves an analogy with an RLC resonant circuit. For a mechanical system model, the equivalent lossy element in electronics, a resistor, is replaced with a mechanically lossy element called a dashpot. Probably possible go into more detail later, but in the absence of too much in the way of external complicating factors (at least to begin with), maybe you already know how to calculate the resonant frequency of an RLC circuit, say, series or parallel, to keep it reasonably uncomplicated. Anyway, the resonant characteristics don't change with signal passing through (again, sticking to idealized models to start with). Same thing with mechanical resonance. The resonant frequency stays pretty much what is natural for the spring-mass system.
 
Sorry, don't have much more time for today, but tomorrow morning should be good. Regarding cone resonance, basically we can think of it as a spring-mass system. An often easy way for electronics guys to understand that involves an analogy with an RLC resonant circuit. For a mechanical system model, the equivalent lossy element in electronics, a resistor, is replaced with a mechanically lossy element called a dashpot. Probably possible go into more detail later, but in the absence of too much in the way of external complicating factors (at least to begin with), maybe you already know how to calculate the resonant frequency of an RLC circuit, say, series or parallel, to keep it reasonably uncomplicated. Anyway, the resonant characteristics don't change with signal passing through (again, sticking to idealized models to start with). Same thing with mechanical resonance. The resonant frequency stays pretty much what is natural for the spring-mass system.
Ok, all noted.

Do you have experience of making PA systems including monitors in a live environment?

I would politely point out that this is essential for any empirical corroboration to what you say. Particularly where you say it’s unlikely to be possible that any resonances play a significant role in feedback.
 
Last edited:
Speaker cone resonance can be viewed as a 'memory' effect. It can be reduced by connecting the speaker to close-by amplifier with a high damping factor. In any case, cone resonance is by nature exponentially decreasing. An amp with a high damping factor speeds up that exponential decay. Maybe in some case exponential growth of acoustic feedback in the presence of exponentially decaying cone resonance could interact in some way, but in most cases its probably not going to happen at the same frequency for the speaker cone natural resonance and the acoustical feedback path phase shift and gain. It would be a special case if were. In that case the most practical solution might be to notch out the cone resonance a little, and or otherwise increase mechanical damping of the speaker a little.
I would also point out that there is a slight, but important error in the definition of ‘feedback’ here.

Feedback, in a live environment (which is my usual environment) can be and (as long as is controlled to the best of the engineer’s ability) be exponentially decreasing.

In fact, whilst the gig is happening (and as was alluded in previous comments) there is a background of low level feedback in the FOH PA. you can hear it if you go up close.

So I think a thorough definition of feedback is any communication between transducers that is doubling the original signal. This may have been an unfortunate misunderstanding, but again, should not have provoked insult.

Feedback is not just run-away feedback, but any unwanted or inevitable communication between the transducers.

Although somewhat separate, there is a control on many reverb plugins called ‘feedback’. You can control this for a little tail, all the way up to runaway feedback.

Actually, now I thought of this, there is a pertinent example which should demonstrate my point/ideas: high reverb levels cause feedback. Well known. I would point out, this is a totally separate system (as separate a system as an amplifier or woofer is- the reverb comes from software handling of the hardware within a digital mixer, for example.), and yet they interact.

I’m afraid I’m writing this as if I know what I’m talking about, because I do.
 
Do you have experience of making PA systems including monitors in a live environment?
Not recently. Did it professionally for about 7 years right out of college. Eventually, it included Michael Jackson at what was then called the Oakland Coliseum Arena (10,000 very reverberant seats). Did a lot of West Coast sound. Among other things a fair amount of concert gigs at UC Berkeley Greek Theater (where I once got a standing ovation from the audience for house sound, as reported in the SF Chronicle much to the surprise of my boss who had never seen such a thing), Concord Pavilion, etc. Crazy long hours. Also did corporate sound, smaller theater productions 2,000 - 3,000 seat type things), a George Carlin tour, churches, pretty much the whole gamut of live sound at the time..Part of what I did involved design of a type of standard small-gig PA speaker. After that decided it was time to get a real 8-5 job. Over the next 30 or so years an entirely different career passed by. Now here we are.
 
Last edited:
Not recently. Did it professionally for about 7 years right out of college. Eventually, it included Michael Jackson at what was then called the Oakland Coliseum Arena (10,000 very reverberant seats). Did a lot of West Coast sound. Among other things a fair amount of concert gigs at UC Berkeley Greek Theater, Concord Pavilion, etc. Crazy long hours. Also did corporate sound, smaller theater productions 2,000 - 3,000 seat type things), a George Carlin tour, churches, pretty much the whole gamut of live sound at the time.. After that decided it was time to get a real 8-5 job. Over the next 30 or so years an entirely different career passed by. Now here we are.
As relates to my other post, which I hope you will puruse, are you aware that high levels of reverb (as isolated a system as an amplifier) can cause feedback?
 
Hi there, I’m looking for any info on reducing microphone feedback in PA speakers.
If I may add a few comments to this thread...

First, the frequency response charts and polar plots we typically see on "spec sheets" are almost always "smoothed", Whether 1/3 octave, 1/12 octave, or whatever, and whether it's loudspeakers or microphones, we're typically not seeing the nitty-gritty soft white underbelly of the unit's performance. There are often wild swings which don't show up on these charts. In particular, off-axis response can be pretty nasty. This is one reason why there are inexpensive ones, and very expensive ones. Second, high quality condenser microphones will typically have smoother actual response curves than dynamic microphones, but they tend to be less rugged and more expensive than dynamic microphones.

Also, I've watched this thread morph from a request "for any info on reducing microphone feedback in PA speakers", to you becoming an acolyte for an unproven hypothesis, and arguing with people who have tried to help you. I just thought I'd mention that.
 
  • Like
Reactions: Markw4 and JMFahey
If I may add a few comments to this thread...

First, the frequency response charts and polar plots we typically see on "spec sheets" are almost always "smoothed", Whether 1/3 octave, 1/12 octave, or whatever, and whether it's loudspeakers or microphones, we're typically not seeing the nitty-gritty soft white underbelly of the unit's performance. There are often wild swings which don't show up on these charts. In particular, off-axis response can be pretty nasty. This is one reason why there are inexpensive ones, and very expensive ones. Second, high quality condenser microphones will typically have smoother actual response curves than dynamic microphones, but they tend to be less rugged and more expensive than dynamic microphones.

Also, I've watched this thread morph from a request "for any info on reducing microphone feedback in PA speakers", to you becoming an acolyte for an unproven hypothesis, and arguing with people who have tried to help you. I just thought I'd mention that.
I think you represent the majority of comments as help and to ignore anything to the opposite I don’t think this is reasonable.

I expect, though I may be wrong that my idea will prove significant.

Thanks

PS I would add, even if this does not prove true, I don’t think the tone of some of the previous comments, particularly the one I pointed out is justified.
 
Ok, all noted.

Do you have experience of making PA systems including monitors in a live environment?

I would politely point out that this is essential for any empirical corroboration to what you say. Particularly where you say it’s unlikely to be possible that any resonances play a significant role in feedback.

Yes.

I've also been published a few times in Live Sound International Magazine, have a degree in Physics, and have designed, built and deployed numerous PA systems and stage monitor systems, some of which have been used by Artists You've Heard Of. I've also restored around 50x AKG D12s (previously considered un-fixable), among other vintage mics, and have taken lots of measurements of microphones themselves.


Please have another look at my previous post (#32) in this thread. I believe it covers most of your questions.

Chris
 
  • Like
Reactions: EarlK
"This thread morphed into an argument"
Guess why.

You started asking about improving acoustic feedback in Live situations, then started jumping all over the place.

For your information,Acoustic Feedback as mentioned happens in the acoustic domain (duh) ,and is influenced by Acoustic factors: distance, directivity, polar patterns, resonances, nearby surfaces, boundaries, phase differences, etc., which happen at the speed of Sound.

Electric signals inside amplifiers, which to boot are close to Perfection (incredibly flat response, infinitesimal distortion, etc.) compared to Acoustic signals, transducers, speakers, cabinets, etc. happen at the speed of Light (or comparable to it).

No Fairy Dust Magic Amplifier is going to influence Acoustic problems.... which is what you are trying to push.
 
  • Like
Reactions: simon7000 and rayma
Acoustical feedback is due to the signal time delay.

Time delay in the air (due to the distance of several feet) is of the order of milliseconds.
Time delay in the amplifier (due to inverse of the electrical rad/sec bandwidth) is of the order of microseconds.

Which is significant, and which is not?
 
  • Like
Reactions: JMFahey
"This thread morphed into an argument"
Guess why.

You started asking about improving acoustic feedback in Live situations, then started jumping all over the place.

For your information,Acoustic Feedback as mentioned happens in the acoustic domain (duh) ,and is influenced by Acoustic factors: distance, directivity, polar patterns, resonances, nearby surfaces, boundaries, phase differences, etc., which happen at the speed of Sound.

Electric signals inside amplifiers, which to boot are close to Perfection (incredibly flat response, infinitesimal distortion, etc.) compared to Acoustic signals, transducers, speakers, cabinets, etc. happen at the speed of Light (or comparable to it).

No Fairy Dust Magic Amplifier is going to influence Acoustic problems.... which is what you are trying to push.
I haven’t jumped all over the place.

I used the term ‘feedback’ I wasn’t any more specific than that. I did not preclude either form, and in my comments have implied different forms can interact.
 
Last edited:
So I have an explanation from KV2 about why their amplifiers, among other components, affect feedback.

I have noticed with KV2, they make too much of certain things in their advertising, for example digital sampling rate needing to be 20 megahertz or similar.

George, however is not like their salespeople, and wouldn’t in my opinion outright lie, though I realise he is critical of line arrays- and line arrays can work well.


If you would like to see the explanation, I can send it.
 
For anyone who would like to take a look, the advertising claims of the gentleman speaking in the videos linked to in post #11 are in written form at: https://www.kv2audio.com/technology.html
As long as you don’t tell me I wasn’t aware of the claims about huge sampling rates, I’m happy.

And also I was aware that KV2 fudges quite a lot in their advertising material. But interviews with the designer, George Krampera are different.
 
20MHz sampling, from the visuals on their webpage, refers to DSD256 and its impulse response. We already understand that pretty well. At least we understand it in much greater depth than they talked about there.
They also present themselves as ‘ushering in a new era of live sound’, which is quite a claim.

But it doesn’t mean to say they aren’t a good manufacturer- and that underneath the sales talk they actually know what they are doing. In my experience, their products speak for themselves.
 
To sell stuff in a capitalistic world there has to be a story, and the story often includes a fair amount of advertising puffery. I find George Krampera videos to include some exaggerated puffery. Not that I have a problem with him trying to compete in the marketplace system that we have, but I don't take it as pure fact. I would feel rather gullible if I did.
 
To sell stuff in a capitalistic world there has to be a story, and the story often includes a fair amount of advertising puffery. I find George Krampera videos to include some exaggerated puffery. Not that I have a problem with him trying to compete in the marketplace system that we have, but I don't take it as pure fact. I would feel rather gullible if I did.
Neither do I.

Although, I don’t think it’s really possible to say something totally water-tight in the time available for these interviews, because language (or at least language that is to be understood by a relative layman) is naturally vague.

However, reading between the lines, I think certain things can be ascertained.