You can't equalize away standing waves (1 Viewer)

[Richard Greene]

I use a 30-band equalizer with my DIY subwoofer and recommend narrow-band equalization to other audiophiles
with subwoofers.
If you want to set an equalizer, everyone "knows" you use pink noise or warble tones.
What everyone knows is completely wrong for finding and "fighting" room standing waves.
A standing wave needs time to develop.
Pink noise and warble tones don't remain at the same frequency long enough to fully excite a room standing wave.
The right test tones to use for your battle against standing waves are sine waves: slow sine wave sweeps to estimate
frequency response variations (+/- xdB) including all room resonances ... and discrete sine wave tones to discover the specific frequencies that make your listening room "boom" and cause the sound to decay at a much slower than normal rate.
Using pink noise or warble tones with standing waves is sort of like looking at the world through rose-colored glasses -- pink noise and warble tones make the bass frequency response measure a lot better than it sounds!
There seems to be a misunderstanding about what
equalization does to standing waves in a home listening room.
Equalization does absolutely nothing to standing waves.
The standing waves were there before equalization ...
and they will be there after equalization.
The equalizer does absolutely nothing to absorb the energy of surface-to-surface room reflections that cause standing waves.
Using an equalizer to restrict the output of your stereo at certain frequencies is creating deliberate frequency response errors -- reducing output of certain frequencies from your speakers. This makes the sound less accurate --
but the problem you've deliberately created with the equalizer is likely to be a good trade-off:
When you have a listening room standing wave that does bad things to certain sound frequencies, you can improve the sound quality by restricting output of those frequencies from your speakers (NOW) so you don't have to suffer as much with the room effects (LATER). The standing wave will not be as loud if you deliberately restrict ouput at certain frequencies. But the standing wave will still be there.

 

[Jones_Rush]

Richard,
Do you imply that even if I'll (theoretically) succeed to EQ my speakers for flat response in my room, it's still won't be as accurate as the sound in a room without room modes ?. In other words, do you imply that a tone frequency of 40hz being EQ'ed for flat response won't be as accurate as a 40hz siganl that do not have to be EQ'ed in order to be flat ? is this right ?.

 

[Jack Gilvey]

I believe that the practical implications of what Richard is saying is that , while it's all well and good to talk about theoretically "flattening" response "perfectly" in your room with unlimited bands of eq , this "perfection" will only apply to a limited (very) area. Move your head over a foot (literally, in my room), and you're in a different world. In fact, the sound in other locations can possibly be much worse than before eq, because of what you've done to the signal to "optimize" one spot. This is something most seem willing to do, though.
[conjecture mode]Even in the "sweet spot", the sound would probably be different from an un-eq'd sub in a theoretically "perfect" room. EQ is causing a lot less power to be delivered to the sub at certain frequencies than others, and it may have a different sonic signature, especially at high power levels, where the unattenuated frequencies are reaching the system's limits. Not that it would be necessarily noticeable. After all, when will we be able to compare it to our sub in the perfect, mode-less room?[/conjecture mode]
Probably not until DSP is more powerful and accessible will we be able to attack this issue effectively.

 

[Timmy]

If you want to set an equalizer, everyone "knows" you use pink noise or warble tones

Everyone does?
Richard, this has already been discussed on HTF many different times, where room waves need a few hundred milliseconds to develop; more time that what is given with pinknoise or warble tones.
I think its great you have discovered this as well; but HTF'ers have already "been there; done that"

I do agree it is much more an eye opener, and more is learned when one discovers this on their own as you did.
Welcome to HTF!

 

[MichelF]

Richard is absolutely right.
Trying to EQ (boosting in this case) a frequency because there is a dip at the listening position at this frequency due to a room mode is absolutely useless: all the energy you will put in this frequency will anyway be "eaten" by this dip.
Imagine the result of the room mode as a multiplication factor that you apply to your signal; this multplication factor depends on the position in the room and the frequency of the signal.
Standing waves introduces in some places and for some frequencies multiplication factor equal to 0... (this could happen at your listening position for certain frequencies). Trying to EQ the signal at this frequency would mean you try to boost the signal that is anyway multplied by 0... and you then know the net result in these locations: 0!
The only thing achieved is to put a lot of stress on the amplifier and loudspeakers that are requested to work harder for nothing except probably more distortions.
Michel

 

[Ricardo E Garcia]

Please forgive a newbie for posting, but I have a question. Would the information on this site http://www.guidetohometheater.com/shownews.cgi?388 help with avoiding alot of what you're talking about in this thread?
From what I understand on several of their articles, the lower frequencies are more difficult to eliminate problems with nulls (I'm not sure if thats what you mean by standing waves) than higher frequencies. So room dimensions are important for, say a starting point, then going from there you can use acoustical treatments, or use equilization for other problem frequencies.
If I got this all wrong, please be easy on the flames. I'm just trying to learn.
Thanks,
------------------
Rick

 

[Mike Boniferro]

Ricardo, I don't think you have to worry about too many people flaming you here about an observation
From what I understand, you are correct that the lower frequencies are much more susceptible to larger areas of nulls in the room.
On another note, what is it that you people mean by modes? When I think back to Physics class, I remember learning about standing waves and (n)odes, would nodes be the same as modes?? Is it a typo or a common mistake people make?
thanks!

 

[Wayne A. Pflughaupt]

When you have a standing wave that does bad things to certain sound frequencies, you can improve the sound quality by restricting output of those frequencies from your speakers.

And this is exactly what an equalizer accomplishes.
I’m not sure what Richard is trying to tell us here—-narrow-band EQing is a waste if time and money? After 1/3 octave EQing my system was much improved, especially the subs. Every sub I ever used in my room exhibited a large boost at 40Hz. 1/3 octave EQ has made my sub response very linear and infinitely more listenable.
Regards,
Wayne A. Pflughaupt

 

[Tom Vodhanel]

this site has some good stuff on modes/nodes...
http://www.headphone.com/EditorialHe.../RoomTubes.asp
TV

 

[Brian_J]

Good read, but i dont think i'll be doing that in my house anytime soon!
Brian
------------------
Zed's Dead Baby...

 

[Richard Greene]

The main point of my post was to get people to realize that an equalizer has no direct effect on room acoustics problems ... and a bass frequency response that measures "flat" may not sound as ideal as it looks on a chart.
If you greatly restrict the energy launched into the listening room at certain frequencies to "fight" standing waves using an equalizer, you have changed the timbre of the music leaving your speakers. There's no "free lunch" here.
We are all using equalizers to "fix" a problem that they are not capable of fixing ... and they can even make the problem worse at some listening positions. At best, equalizer can make a room mode less offensive. Sometimes moving your
listening seat a few feet (away from a mode maxima) can help just as much as equalization.
The typical home listening room is too small to produce the relatively smooth bass response you'd hear outdoors or in a large auditorium where standing waves are much less audible than in a typical home listening room.
To Jones Rush:
If you are using equalization to "fight" standing waves, the sound meter may read flatter than before but the frequency response launched from the speakers will be less flat than it was before equalization.
Reason: you are deliberately putting "notches" in the frequency response of your speakers at frequencies that "act up" in your listening room.
You're correct that the equalized sound will not be
as accurate as sound produced in a room without modes ... but I have never heard of a room without modes except
for an anechoic chamber. So that's a moot point.
Assume your subwoofer has a relatively flat frequency response in an anechoic chamber, your room has a 40Hz.
standing wave that is being excited by your subwoofer,
and is easily audible at your listening position:
(1) The launch of the sound wave (SPL) is correct.
(2) After a brief delay, the reflected sound will cause
constructive reinforcement that makes the tone (SPL) too loud and causes a slower decay than other frequencies.
Both (1) and (2) combined make the 40Hz. tone (SPL) seem much too loud.
If you equalize the sound to make the sound meter read flat
at your listening position, the result will be:
(1) The launch of the sound wave (SPL) is now too weak
(2) After a brief delay, the reflected sound will still cause constructive reinforcement that makes the tone (SPL) sound louder and causes a slower decay than at other frequencies
Both (1) and (2) combined make the average SPL over time seem okay ... but only because you have averaged a too-weak attack from the speaker and a too-strong reflection/decay from the room. These combine when using a sound meter as if they were offsetting errors ... but in fact these two errors do not occur at the same point in time, so the sound meter is only being fooled into reading flat.
To Jack Gilvey:
Great point. Standing waves are all about the uneven
distribution of sound energy in the listening room.
If your listening seat is near the maxima for a 40Hz.
mode, for an example, and another listening seat is near
a minima for the 40Hz. mode, for which listening seat
do you set the equalizer?
The answer is easy for audio - you optimize for the "sweet spot" listening chair. The answer is difficult for video where the sweet spot is quite wide due to the use of a center speaker. If you averaged two microphone readings, for one listening chair near a mode maxima and another listening chair near a mode minima, the result might be "flat" due to offsetting frequency response errors at the two positions!
To Timmy:
I discovered sine-wave signal generators and room modes in 1971 during my first year in engineering school. If you're interested, the resonant frequency of the human chest cavity
(diaphragm) falls in the 100 to 150Hz. range (depends on the size of the chest cavity - higher resonant frequencies for smaller chest cavities) -- don't think I wasted my time in engineering school on unimportant subjects such as differential equations! I also checked the wear on my stylus with a $10,000 electron microscope!
I only mentioned that pink noise and warble tones don't excite room modes because quite a few people on this website (certainly not the majority) don't seem to realize that.
To MichelF:
Trying to boost a standing wave minima (aka "null") is just a waste of amplifier power, as you've detailed.
To Mike B.
Since no one can fix room standing waves, they just confuse the subject with different terms meaning roughly the same thing: peaks/troughs, maxima's/minima's, nodes/anti-nodes, standing waves, modes, and there are more!
To Wayne P:
Pink noise won't excite room bass resonances if you leave the test tone on all day!
I'm trying to say an equalizer is not the proper tool
to eliminate room modes although it can work better than doing nothing. 1/3 octave controls are too wide for most room modes and equalizers do nothing about the time-based effect the room has on certain bass tones.

 

[Jones_Rush]

If you equalize the sound to make the sound meter read flat at your listening position, the result will be:
(1) The launch of the sound wave (SPL) is now too weak
(2) After a brief delay, the reflected sound will still cause constructive reinforcement that makes the tone (SPL) sound louder and causes a slower decay than at other frequencies
Both (1) and (2) combined make the average SPL over time seem okay ... but only because you have averaged a too-weak attack from the speaker and a too-strong reflection/decay from the room. These combine when using a sound meter as if they were offsetting errors ... but in fact these two errors do not occur at the same point in time, so the sound meter is only being fooled into reading flat.

Richard, what you actually say is, that when EQ attenuated frequencies are played, I will hear them at a slight delay with comparison to other frequencies which are untouched by the EQ device. (the delay is the actual time that it takes for the standing-wave to develop). This is a very annoying fact, nevertheless, I will tade this small delay ANYDAY with the 10-15db peaks I get at some bass frequencies.
After reading your post, one can think that with a parametric EQ, all you're doing is to trade one problem with another. While OBJECTIVELY (according to what you say) this is correct, subjectively, I think it's VERY wrong. I heard my room with EQ and without, and I can honestly say that the bass with EQ is (subjectively) 100 times better than without it. I never even noticed the delay you are talking about (though it will be a hard thing to notice If I don't have the option for an A/B comparison).
I'm sure that what you say is correct (it has much logic in it), but if this delay was *subjectively* a big problem (like the modes themselves), I'm sure that today's EQ devices would have been able to deal with it. It shoudn't be that hard, all you have to do is insert this small delay into to the EQ chip, so it will only attenuate frequencies after a couple of milliseconds they are played. How will you know to set the delay correctly ? simple, by trial and error. If the delay is subjectively very annoying, than there will be no problem to set it by ear, maybe not with 100% success, but even 60% is not bad. An even better way to calculate the delay is to use a RTA software which can show a 3D graph of time/spl/frequency ("waterfall"), this way you can very accurately see the time it takes for the standing-wave to develop. True, you'll have to set a different delay for each attenuated frequency, but it is sure worth it if the subjective results will show a similar gain to what some of us have been able to get with only "correcting" the standing-wave itself (in our listening position, of course).
The fact is, that the delay issue has been completely ignored in all EQ devices known to me, which leads one to believe that it might not be that much of a big deal.
I haven't even seen such feature in the pricey Tact TCS room correction device (http://www.tactaudio.com/web-files/TCS/TCS.htm).

 

[Thomas Ward]

The Headroom site is more than a "good read"...Those "giant toilet paper tubes" really work!
My room is 12' x 12' and has(HAD)a 48Hz peak that would tear your head off. I used four 1/4 wave tube traps and now have a significantly flatter room. I don't even use my parametric EQ anymore...really!
Those 16" diameter tubes are "butt ugly", but they improve the response so much that my wife wants me to keep them...really! They will look a lot better when they are each covered with a black polar fleece sleeve stretched over a couple layers of polyester quilt batting (a little bonus, high frequency absorption).
When you wave those tubes around (in and out of the nodes) you get BIG swings on the Radio Shack SPL! Those tubes are radiating "negative sound" when excited into resonance.
Give it a try...Sonotube isn't that expensive!
All the best,
Tom

 

[Wayne A. Pflughaupt]

The typical home listening room is too small to produce the relatively smooth bass response you'd hear outdoors or in a large auditorium where standing waves are much less audible than in a typical home listening room.

Outdoors, yes. In a large auditorium, absolutely not! Standing waves might not be a problem in a large auditorium, but reverberation and delay definitely is! The larger a room is, the more difficult it is to achieve smooth, intelligible bass. Usually it is dependent of the construction of the room (i.e., shoebox, square, round) coupled with acoustical and diffusing treatments (or more typically, the lack thereof). Personally, I’ve never been in an auditorium large enough to seat several thousand where the bass was not absolutely horrible. And as noted above, perceived bass response changes drastically depending where in the room you are sitting.
My hat’s off to Richard for this thought-provoking thread. He definitely has his theory down about the limitations of equalizers.
Fortunately the ear seems to be very forgiving of these time-domain problems. Even though it is theoretically impossible to deal with room modes with 1/3 octave equalization, I was able to (perceptively) achieve very linear response in my system. The difference was night and day. Time-domain issues or not, I’m not about to chunk out my equalizers!
Regards,
Wayne A. Pflughaupt

 

[Jones_Rush]

The delay situation is time-domain based, and has nothing to do with the use of an equalizer. An equalizer will not in and of itself cause delay.

Wayne, what I meant was that when a certain frequency which is EQ attenuated is being played, at the first couple of milliseconds it will be -Xdb (Xdb refers to the amount of attenuation you set for the frequency), after a couple of milliseconds, when the standing-wave will develop, the -Xdb should turn into 0db variation (assuming you did a good job). Again, the dealy is the actual time it takes for the standing-wave to develop, it has nothing to do with the EQ device itself, the EQ device is only responsible for what you'll hear (ie spl variations) during the delay and after it. If you are not using an EQ to attenuate the standing-wave, you will still get delay, only this time, during the delay you'll get 0db variation (assuming you speaker is
unechoicly flat), and after the delay you'll get +Xdb (Xdb refer to the amount of room gain your standing-wave was able to produce).
Of course, during the delay itself, the db variation is constantly changing, it's not static, it is gradually taking the spl-shape of the standing-wave which is developing (actually, it IS the standing-wave developing).
I've also said that since this time domain problem is so easy to fix, the only reason that there is no product (that I know of) which fix it, maybe prove that this issue is not subjectively a problem . As a matter of fact, the solution is so simple that I might ask my friend who works with computer fiir filters to try and program a corrective solution for this delay. I am not expecting a night and day difference, but it will be nice to do some A/B comparisons with delay and without it.

 

[Richard Greene]

To Tom Ward:
Great post. You are dealing directly with your
room acoustics problem using tuned tube bass traps.
They are ugly but they work is you use enough of them.
They do work best in room corners and other room surface
intersections where the sound pressure is high .
To Jones Rush:
Let's say you have a room with a 40Hz. standing wave,
so you greatly restrict output at and near 40Hz. using
your equalizer.
When a long sustained organ note plays at 41Hz.,
for one example, there is still a 40Hz. mode but
not as loud as before.
That's usually a big improvement and the equalizer
may not even be noticed.
But what happens when you play the Jimi Hendrix song
"Purple Haze" with its 42Hz. kick drum?
Well now the sharp attack of the kick drum is not as
loud as a result of the 40 Hz. equalization.
You've changed the timbre of the music launched into your room by your speakers.
Could it be that the kick drum transient is too quick to fully excite your 40Hz. room mode?
Could it be the kick drum would have sounded better with less 40Hz. equalization or no equalization at all?
My point is that equalization to reduce the SPL of a room
mode works pretty well for sustained bass notes like those from a pipe organ or bass guitar ... but not so well for kick drums and other bass sounds that have a quicker decay.
Fortunately, most bass musicians prefer slow resonating bass tones that are not harmed much by equalizers. But there are some kick drums, synthesizers set for a fast attack and fast decay, and sometimes a Fender bass played with a pick that don't sound quite right when played through an equalizer being used to "fight" standing waves.
Using acoustic tools such as bass traps to fight room modes
can actually solve the problem without any negative side effects (except they are not much to look at).
An equalizer is not the right tool to "fight" a time-based listening room acoustics problem ... but until digital sound processing tools become much less expensive, an equalizer is an inexpensive tool that can make some very serious room acoustics problems a lot easier to tolerate. Especially
in a small listening room.

 

[Timmy]

Could it be that the kick drum transient is too quick to fully excite your 40Hz. room mode?

No; if you filtered all freq content from pink noise except 40hz; you probably wouldn't hear anything except very faint popping. But it would easily show up on a RTA screen.
Even with worse case scenario; it only takes milliseconds for standing waves to develope in a room; more time, though, than pink noise gives at any one time at any one frequency.
The reason for pink noise being used for speaker calibration originally was to negate any room interference from doctoring the results.

 

[Sundar Prasad]

But what happens when you play the Jimi Hendrix song
"Purple Haze" with its 42Hz. kick drum?
Well now the sharp attack of the kick drum is not as
loud as a result of the 40 Hz. equalization.
You've changed the timbre of the music launched into your room by your speakers.
Could it be that the kick drum transient is too quick to fully excite your 40Hz. room mode?

Actually a transient in a kick drum note will have frequencies higher than 42 Hz. If you are familiar with Fourier transforms, you will know that a signal with a rapid attack and decay comprises of many high frequency components. Consequently, EQ'ing a room mode of 40 Hz by using a parametric equalizer to tame a band that is 5 or 6 Hz wide is not going to do great harm to the overall nature of the trasient in the drum signal. The note coming from the speaker will be changed, but only in a small band of frequencies. The poor phase response of the equalizer may change the character of the note somewhat (whether that is audible is a debatable issue), but it again will be restricted around the band of frequencies being attenuated.
I use a BFD to tame room modes by only processing 4 narrowish bands below 60 Hz. The mains are unprocessed and any phase nasties occur only in the subwoofer. I'll take the defined sound of this equalised setup any day in comparison to the amorphous mush that resembled bass when the room modes were doing their thing.
Sundar