…and attach up to one MBytes per post. This is a forum not a hosting service… 
Andreas,
I’m sorry if I seemed to confuse the issue with my extended response. I started out saying that feedback is a complicated issue. It is perhaps impossible to solve a specific feedback problem without being in the room. I just tried to address most of the factors involved.
Unless one understands the issues, it is partly luck when one solves a feedback problem, depending upon the specific circumstances. Of course, every feedback problem is a mix of mic/gain/speaker adjustments, but unless you say how to handle each of these factors, you haven’t said anything helpful. If you’ll read my instructions and suggestions carefully, you’ll find that I did address the question asked, and I did state that the vocalist was the last contributing element to be considered.
Consider this: You can stop feedback by turning down the system gain. That says that the feedback at any frequency is dependent on the level of that frequency in the room. If the room tends to feedback at, say 10 kHz, a female vocalist is more likely to generate sound at that frequency than a baritone.
If you’ll review the other suggestions in this thread, you’ll see that they are all “guesses” because they are not able to know the exact situation of the person having the feedback problem. Until the operator is able to get a feel for all the factors involved and then determine how each applies to his situation, he will struggle with the problem. He may fix the problem for this one situation, but what happens in the next situation?
I’ve even heard listeners, even operators, say “I heard feedback,” when what they heard was reverb!
No, we have not purchased our A&H yet, but we’re getting close. I realize that I did not confine my remarks to the limitations of the A&H QU. There’s a saying that goes something like this: “If you give a person a piece of bread, you may have solved his immediate hunger. But if you teach him how to make his own bread, you will help him feed himself and others, and lessen the chance that he will be hungry again.” I was just trying to help the operator “feed himself.”
Again, I’m sorry if I only created confusion. However, the more one learns, the less confusion one will have. You’ll learn a lot more if you truly listen to what one has to say. Sometimes, what one says may seem to disagree with what you think you know, but listening will give you a broader perspective and sometimes correct erroneous or limited conceptions.
avaen1,
Hey, I like that hat! I also like your attitude - consider what someone has to say and see how it jibes with your experience.
Just consider the conversation between DR and myself a discussion between two old codgers. We really are not enemies, just old guys with somewhat different types of experience and observations. If you see a point of disagreement between us, just perform an experiment that will tell you which is right. You will have learned something!
For example, put an SM81 and an SM58 beside each other in a somewhat reverberant room. Connect them to your mixer and EQ them for maximum volume without feedback. Which mic tends to feed back at low frequencies the most? Which one tends to have more high-frequency feedback? Or is there no difference between them?
What did you learn? If you choose to do this experiment, tell us what you learned.
Dick,
We certainly have different experiences in the audio field and seem to have formed different opinions! I became interested in electronics at the age of 10, when the typical circuit constructed by young boys was a crystal radio - remember those? As I worked toward my doctorate at Oak Ridge, I delved into electronics somewhat deeper, building my first hi-fi system from Heathkits, along with a tube-type color TV. I didn’t just follow the instructions, I also looked at the circuit diagram to better understand what I was building.
I studied speaker design extensively and built a number of speakers over the years. In the 70’s my church needed a new mixer, so I built an 8-channel powered solid-state mixer from circuit boards that were available at the time. We were one of the first churches to have stereo sound.
I’ve been singing since I was 10, mostly in church. Later, I sang in a couple of praise bands and designed sound systems for churches.
As a scientist, I learned how to make critical observations and perform experiments to learn what I needed to know about a situation. When confronted with a problem, I tried to examine it from as many perspectives as possible. That experience carried over into my work as an audio technologist. Generally, I don’t tend to make statements that I have not somehow verified as accurate.
Nevertheless, it is conceivable that I missed something along the way, and you may have also, that has resulted in us having differing opinions on some audio technology subjects. All I can suggest is that we try to see our disagreements from the other’s perspective and see if we need to do some experiments to decide where the truth lies. I hope that those who read our submissions will do the same, not just take sides in an argument that doesn’t really exist.
Hi
PM it me please and I’ll host it on here for everyone to access
No can do. I cannot do attachments in PM’s just like I can’t do it here.
The PDF is subject to periodic revisions and customization for specific individuals at various skill/experience levels, so I do it on a case by case basis. I do not care to have my work posted on the nintershnet, thank you, but if it will help you then you are welcome.
All I need is a valid email address.
DR
@DocG…
I’m not going to get into an endless refutation of your misconceptions regarding audio. Those of us who do this for a living certainly see through your misinformation. I did a point by point reply to clarify the misleading conceptions in your first reply not to cavil with you but to point out for the benefit of the OP and other beginners the fallacies inherent in what you wrote.
Either your wording was errant or you operate with only a partial knowledge of the basics physics of audio. Your characterizations of GEQ /PEQ are demonstrably fallacious and misleading to those trying to learn. Additionally, your statements about the causes and characteristics of feedback fall way short of the mark.
I’m not going to continue this as a debate. I come here to help those who are starting out, not to disabuse folks who insist on clinging to what I’ve heard called " faith-based physics."
To the OP and anyone else interested: GEQ, though often used for the task, is not the best tool for dealing with standing waves and/or room resonances resulting from room/system interactions. PEQ is the better tool and should be applied to the mains bus, not the individual channels.
The bulk of problems in this area lie below 300hz and 3 or 4 notch filters commonly suffice. GCumbee spoke of this in his post above. If you happen to have a DSP unit or DSP built into your power amps or powered speakers, you should take the time and trouble to optimize your system in free space (outdoors) so you’re not dealing with both system AND room anomalies at the same time. It gets tricky fast.
DR
Consider this: You can stop feedback by turning down the system gain. That says that the feedback at any frequency is dependent on the level of that frequency in the room. If the room tends to feedback at, say 10 kHz, a female vocalist is more likely to generate sound at that frequency than a baritone.
The first sentence is true.
The second sentence does not follow from the first as implied and is misleading in that it is the sound at the microphone, not the sound in the room that is the determinant. For a good part of the sound spectrum the propensity for feedback at a specific frequency can be reduced simply by moving the mic a few inches one way or the other toward an anti-node.
The third sentence is just a poor example for a variety of reasons, primarily that 10kHz is far above the point at which room resonances occur and any sonic anomalies in this region are due to sound reflected directly into the microphone from a mains or monitor speaker or an extremely unfortunate microphone placement.
DR,
I too am tiring of this. You simply can’t follow logic. You say “anomalies in this region are due to sound reflected directly into the microphone from a mains or monitor speaker…” My question is, what generates that sound?
Resonance, as already stated. Yes it needs stimulus, but the more unstable the system, the less of a stimulus it needs. This can include thermal noise.
Not sure if reducing the overall volume is an option for your band, but in smaller rooms you should think about that as well.
I always try to keep the volume of the gig as low as possible (Playing to the volume of the drummer who isnt a massive basher)
My question is, what generates that sound?
Anything at all. There is no voice that I know of that can produce the 10kHz in your example as anything other than a harmonic. You also cite room resonance:
." If the room tends to feedback at, say 10 kHz, a female vocalist is more likely to generate sound at that frequency than a baritone"
Mervakas post above gets to thhe crux of the matter. First of all, “the room” does not feed back. The sound system feeds back. The only way the room can be excited and resonate (not feed back) occurs as a standing wave directly related to room dimensions. This does not happen at wavelengths as short as 10kHz. There will be many, many other (lower) frequencies which become problematic before 10kHz becomes a problem.
In fact, you can reduce the propensity of 10kHz to be a part of the problem not just by notching the frequency itself, but by cuts at 1/2, 1/4, 1/8 the value: 5kHz, 2.5kHz, 1.25kHz and so on. Yes, these are octaves…
For those interested in understanding these phenomena, here’ a link to a calculator:
https://www.mcsquared.com/metricmodes.htm
I think you’ve thrown out a random figure in choosing your 10kHz example and linking it to room behavior only applies if the sound is directly reflected into the mic. The wavelength @ 10kHz is 1.35 inches. Such short wavelengths tend to get cancelled out to a great degree by the complex interactions of the reverberative sound field of the room.
Again, for those who wish to understand the physics involved, here’s another calculator link:
As I stated before, you are either mistaken in your basic understanding of the physics involved or you are having difficulty expressing yourself in clear terminology which does not render discussion moot. I refer once again to your blanket, unqualified statement of GEQ filters being “narrower” and your confusing of room resonances and feedback. We know what you SHOULD mean, but your choice of terms is so inappropriate as to confuse logical discussion. The end result is semantic quibbling rather than discussion of the topic. I endeavor to provide hard information for those interested in learning the realities of the physics of audio. That way SOMEONE reading this will understand what you seem not to.
This is getting to be pretty Socratic…if you know what I mean.
Here’s a link to a better room mode calculator than that linked above:
https://www.marktaw.com/recording/Acoustics/RoomModeStandingWaveCalcu.html
10kHz #3: I also assume that the 10kHz is a badly picked example and would be difficult create with digital desks at all. Not because they’re digital but because they introduce some sort of latency, about a mSec on the Qu. Maximum feedback frequency (at the fundamental) is slightly below 1kHz, getting feedback on harmonics first is pretty unlikely…
Thanks, Andreas.
For those of you interested in the different types of GEQ filtering, here’s another good link:
I’m not a big fan of computer screens (eye problems make them hard for me to read), especially if they draw the attention of a sound person to the visual realm instead of the aural. But the EQ trace shown on the Qu screen can be very informative…should you actually own or use a Qu console.
The common Q value of GEQ filters has been 1.4. Filters with this Q value act over a one octave range, the 1/3 octave nomenclature denoting the center frequencies of the filters, not the width. If you wish a visual representation of such a GEQ filter, set one of your parametric filters to have a Q of 1.4 and have at it. I believe this is 1/3 on the Qu, but I’ll have to check when I get the console out of the vehicle in a couple of days.
Here’s another link, this one dealing with filter Q and bandwidth:
DR
DR and Andreas,
Now that I have you focused on the correct issue - the frequency of the sound feeding back – maybe we can communicate.
Part of the “room” equation are the speakers and the mic. For feedback to occur at a given frequency, all that has to happen is for that frequency to be generated from the mic, transmitted through the mixer, come out of the speaker and hit the mic again at a level sufficient to start a feedback loop. The room itself doesn’t have to contribute to that loop, only the transducers and electronics. You stated that in other terms.
I said that I was assuming that the operator knew how to take into account the factors having to do with proper speaker placement vs microphone position and the directionality characteristics of each device. I wasn’t addressing those issues; my focus was on the frequency response of the system.
My choice of 10 kHz and the female voice as my example was perhaps unfortunate but also deliberate; I was just trying to make a point, get you on the subject and at the same time give you something to argue about. A high-pitched female voice can reach 3000 Hz and the 3rd harmonic would be at 9000 Hz. If the microphone can reproduce that frequency strongly, and/or if the mic or speaker has a response peak in that range, feedback can occur at that frequency without room resonance being involved. At any rate, a female voice would be more likely to generate high frequencies than a male. That’s all I was trying to say. Pick another frequency - the argument remains the same.
If one then reduces the system gain at that frequency with a GEQ or PEQ below the feedback threshold, the feedback can be eliminated. Now that you are talking about the Q values of GEQ and PEQ, you are on the subject I was trying to address; I just did not want to completely ignore all of the factors that contribute to feedback overall, but I wasn’t trying to address them in detail.
Once all the other factors have been cared for, you are left with the frequency response of the system. If you have smoothed room/speaker response with the appropriate GEQ, eliminated the primary feedback frequencies with HPF. LPF, and the PEQ on the channel in question, and some unwanted feedback is still being generated at the volume level you wish to use, you must resort to the PEQ of the group and main controls, then finally the GEQs on those outputs. With it’s fairly narrow Q values, a GEQ can be used to remove particular frequencies without adversely affecting the overall sound, if used judiciously. If the EQ devices are not sufficient to eliminate the feedback, one must then reduce overall system volume. The Qu has these capabilities.
I spoke about using an RTA to determine where feedback is occurring; the Qu screen can display this - it is very informative.
Lord! I hope we have arrived at the answer the operator was asking originally!
Hey Doc…
Coming back to parrot the corrections to your spurious posts as if you had stated them correctly in the first place or hadn’t gotten into all the ramifications in depth because of this or that is classic trolldom. Once and for all, you lack a basic understanding of the physics of audio…or the practical application thereof.
Basically, you’re nuts. The PM’s I’m getting are 100% in agreement. To paraphrase W. C. Fields:
Some of the participants here are habitues. Some are just sons of habitues…
Go away and stop posting nonsense.
It hurts when you’ve been shown to be a fool, doesn’t it Dick? I was wondering when you would get to the vitriolic level. Now we all know just the kind of person you are.
Indeed, Doc, we’ve learned more about one particular person, but that’s not Dick…
I hope that’s a good thing, Andreas, but did you learn anything more about audio technology, or do you feel there is nothing more you can learn?
It hurts when you’ve been shown to be a fool, doesn’t it Dick? I was wondering when you would get to the vitriolic level. Now we all know just the kind of person you are.
Sorry you’re in such pain, Doc. It appears pretty close to terminal. It also appears your primary intent is to start arguments in this and the other threads in which you’ve posted.
As to the person I am, I am an audio professional with enough verifiable experience and credibility to earn my living at it. I also am not afraid to post under my real name rather than hide behind a pseudonym.
You wanted an argument, you got an argument. You should be happy now although you’re still wrong on 90% of the statements you’ve made here.
You appear to be a classic example of the “majority of one”…