10-25 Hz, is it necessary for HT or Music?

bentoronto said:
"This means that if the 40 Hz 2nd harmonic of a 20 Hz tone is at a 24 dB lower level, then it will sound equally as loud as the fundamental. This corresponds to 6% 2nd harmonic distortion. The 3rd harmonic distortion would have to be below 1%, or over 38 dB down, in order that it is less loud than the 20 Hz fundamental. It all leads to very low distortion requirements. The fundamental frequency sound pressure level needs to be above 70 dB to even become audible and it should not be masked by higher frequency distortion products.
For a detailed investigation of requirements see: Louis D. Fielder & Eric M. Benjamin, "Subwoofer performance for accurate reproduction of music", JAES, Vol. 36, Number 6, pp. 443 (1988)."

Sooo, some folks who think they are hearing 20 Hz are really hearing the distortion... and you may not be able to tell the difference by ear, no kidding.
Hi there,

I'm late to the party....juz my 2 cents here. :D

Yes its true that its an uphill task trying to produce high-level 20Hz/infrasonics with low distortion.

But luckily there are quite a number of solutions nowadays.

General good info for anyone, check out the equal loudness contours (with new ISO curve).
http://en.wikipedia.org/wiki/Equal-loudness_contour

It will be a relief to know that DIY solutions of 15/18" ported or PR with 500-1kW, or commercial units like SVS PB13U, Epik Conquest, DSL DTS-20/TH-50, or even sealed designs like Mark Seaton's Submersive....easily surpass those distortion goals at very good levels.

And don't forget there is ample room gain below the lowest modal freq, which gains the low freqs and leaves the H2/H3 untouched.

You could even implement a 2-way sub system, 1 for VLF and 1 for mid/upper bass (and you could use your high quality mid/upperbass driver with your fav Altec to pro-audio stuff or the ever popular DIY Lambda).

All depends on the need for bass. :)
 
bentoronto [/i]I "hear" a big drum but not feel it: yes said:
You only need the frequency range to understand what is being said and what happens, for music maybe a little more. But the bandwith of a telephone is sufficient.


But your comments about distortion gave me some food for thought.

I havn't come to a conclusion, but for the time being I decided not to care.

I find the music to be richer and more enjoyable with the added infra sub. If I sense the infras itself or it is harmonics I hear (and my brain adding the missing fundamental. ), it adds to the size of things ( Big Drums) or gives the sensation of danger or doom ( Dark Ambient, e.g. Steve Roach ), that is lacking without it. Distortion or not.... thinking about the missing fundamental: Maybe without distortion I wouldn't sense a thing!
 
there are ways to achieve 10hz-25hz easily. Ok it might not be that easy but it's worth it. One way is to take your current subs and put them in a sealed box with a "Variable Mass Passive Radiator" For those not familiar with the concept of passive radiators, it is basically a sub with no magnet and instead of a coil there's a tube. you install it in the box with the cone facing inward, play low frequencies thru the subs, and load the radiator's tube with supplied washers to add mass to the moving assembly which tunes the radiator. When you get the sound you're looking for you epoxy the washers in place and there u have it, an accurate 15hz. As far as I know they are only available in 10" - 18" If you need an 8" you can make one yourself with a sub & a cardboard tube & some epoxy.

The other way to achieve such low frequencies is to build a "Transmission line enclosure" Which is basically a ported box with a port so long it is tuned to the END of the wavelength. With a properly designed cabinet you might even be able to get these frequencies with as little as a 12", only problem is a trans. line for a 12" would probably be close to around 18cuft to 22cuft.

Then there is the easy way... Get a 32" sub (kenwood, clarion...etc) and build a portedbox blueprinted to TS specs.

Idon't think these frequencies would show in music, but may be found in movies/tv explosions & such
 
The quote from Linkwitz is correct regarding the equal loudness curves, but they are what their name implies, curves that tell you how loud a given frequency needs to be to sound as loud as a different frequency.

The ELC have little to do with real world sounds and recorded source.

Consulting firms make lots of money designing car doors that, when closed, emphasize the extreme low frequencies and attenuate the higher frequencies through mechanical means. This is because the lower frequencies translate to the perception of quality in the case of a car door closing.

This certainly doesn't mean that the ELF of the event has to bee in excess of 120dB when a car door closes, as common sense should lead anyone to conclude.

The fact is that all transients have content to DC. Arbitrarily selecting a cut off point for playback of a recording of these events detracts from the perception of realism, as will any other distortion.

Movie soundtracks work within the format for the LFE channel to produce 115dB peaks at the listening position at 0dBFS playback level, from 3Hz to 120Hz, a bandwidth of 5 1/2 octaves.

There is never a Low Frequency Effect that is comprised of a single frequency. They are instead, complex combinations of various recordings of real events mixed with synthesized sounds. Because the mix engineers realize that the 115dB peaks are the combined total SPL of the effect and that the effect is a spread of simultaneous frequencies over 5 1/2 octaves, the fundamental frequency of the effect is normally encoded at -10 to -15dBFS.

This drops the requirement of playback level of a single frequency in the effect to 100 to 105dB at the listening position with playback level at 0dBFS.

Since a 10Hz wave length is 110 feet (33.55 M) and 5Hz is 220 feet (67.07 M), most every room will provide a considerable amount of pressure vessel gain without the possibility of room mode problems (because there is no single room dimension long enough to set up a mode).

The bottom line is that the average actual requirement for a subwoofer at these frequencies to satisfy the most demanding source is 90dB at one meter. Since the room will amplify the sound at an increasing rate as you move down in frequency, the harmonics generated by the subwoofer are decreased as you move up (2HD, 3HD, 4HD, etc), which gives the effect of lowering total harmonic distortion.

Search for spectrograms of transients as well as spectrograms of movie soundtracks for verification by measurement that transients have content to DC and that many movie soundtracks have content to below 3Hz at -10 to -15dBFS.

Are they necessary? Do they contribute to the perception of realism? That's one for each individual to decide, but the fact is that the Dolby specification for the LFE channel is 3-120Hz and scores of movies use the entire range available. Also, the fact is that, from a kick drum to the stuccatto pluck of a stringed instrument, there is content to DC that may or may not have been filtered in the recording process.

Bosso
 
Issues need to be kept separated.

The folks who create those scary post-Jaws rumble sounds are entitled to screw around with the bass any way they please. And listeners can do likewise at home - indeed, I think there is no defencible way of using physical measurements - in practice and in theory - to achieve "the right" pressure spectrum at your chair (just as there is no way to put a rumbling rail boxcar in my living room or even to adjudicate what is the right level to play a string quartet... surely never as quietly as fourth row in even a small hall or did Beethoven really mean my livingroom... which brings us smack up against the equal-loudness spectrum question).

The Fletcher- Munson-type curves of "equal loudness" seem to my mind an unreliable kind of lab exercise... and I just happen to be a perception psychologist (as one of my professional hats). But they do contain a certain rough truth. Maybe most meaningful is the threshold of hearing (with earphones) curve because that is based on a fairly solid perceptual judgment (present versus absent) and a good sense of what the physical input to your ears is.

These curves are very important for the home builder and the person trying to decide what LF range to try to achieve. You need to look at the curves upside down. If I put equal voltage into the speaker (which is the kind of test many people are talking about here), what do I hear? You hear the equal loudness curve appropriate to the level you are producing. At any level short of deafening, you don't hear much at 20 Hz. as the upsidedown curve indicates.

But, if you are hearing much sound at very low frequencies, what you are hearing (as contrasted with feeling), as per Linkwitz, is likely to be the harmonics (unless you have a truly exotic system with 18 inch driver, servo motor, etc). And that's that. If you are happy with the sound, that's just fine. But don't imagine (or post) that you are making loads of 10 Hz sound.

When you are exposed to real-world sounds... not sure if that includes stuff in disaster movies or not - not sure earthquakes make any rumbling noises, they always have their harmonics present. Your ear-brain-system hears a 16 Hz organ pipe (pretty rare) perfectly well even when you are not hearing anything at 16 Hz at those volumes. Music and speech would be ghastly if you couldn't fill in the harmonic blanks downwards that way, which is the plain way your hearing happily works.

My Klipschorn with little sound below 37 Hz sure shakes me when the rail boxcars move through or the Queen's Birthday Salute guns fire. And without dwelling on it again, oddly somewhat OK and wholly musical to listen with a 24 dB/8ave cut-off of 140 Hz. Everybody should try it briefly.
 
Dear oh dear.

Just for a laugh I turned off my labhorns and set my midbass horns highpass to 140Hz / 24dB/oct.

Just listening to the radio with male vocal / guitar folk type stuff all weight and body went from the acoustic guitar leaving a sense of listening through a single speaker transistor radio type stereo.(tinny)

If bentoronto really can't hear much difference then I'd suggest either his hearing needs checking, or more likely his high pass isn't doing what it should, or his speakers are not doing what they should.

I do agreee that you should not really be hearing sound below ~ 32Hz (I hear test tones down to ~32Hz on my system, and below that I just get a feeling of pressure / fluttering in my ears,oh and rattles. ) However the 30Hz - 140Hz band contains a lot of information and it should be obvious that its missing.

I'd also suggest that here on this forum as opposed to a 'normal' hi fi forum he'll find a lot more people with sub bass systems that don't just kick out harmonics. ie multiple drivers / IB's / TL's / big horns / tapped horns etc so shouldn't assume that anyone posting about the benefits of having response down to ~10Hz is only running a 'rubbish' sub that is kicking out distortion.

Rob.
 
My experience with klipsch horns is that they have plenty of bass below 37hz in room, but they also aren't subwoofers. The plenty of shaking experienced is the energy created in the 50-60hz range rather than the 10-20hz range that most people think the "boxcar" shaking is coming from. It's generally as important to ensure you have plenty of energy there as you do down lower, more important really. A lot of people have theater systems capable of over 110db's from 50hz down to say 20hz, but yet have main speakers that can barely do 90db's above 50hz. I think the biggest thing separating the average home surround system and a really good professional theater system is the ability of the main speakers to play loudly and cleanly at those levels. Obviously you don't want and should not listen at those levels for any sustained periods, but the clean headroom is needed during those crazy dynamic peaks our precious action movies have.

A lot of speaker systems actually roll off quite a bit below 100hz, and I would agree that they have a surprisingly full sound. None the less, speakers capable of lower than that do sound more real, more complete. I think the notes below that point tend to help add scale to the sound of instruments. A tympani drum just doesn't sound realistic if its missing its lowest octaves. Additionally, there is a visceral effect you get with a live instrument like that which can be lost in home reproduction if the home system doesn't have the same basic range or output capability.

As for the argument about content below 30hz in movies, the answer is a resounding yes, there are measurements of the spectrum of bass content in movies over on AVS forum showing tons of very low frequency information going into the 10hz and below range, with plenty below 30hz. While you may not need this to enjoy the movie, its there, and if you want to hear the movie as it was intended to be heard, you need a system that can reproduce that.

As for music, well we know that live music has content that low, and again, there is value to our perception of the music in having it, but unfortunately the CD and worse yet MP3 lovers have killed any chance of us actually hearing those frequencies. While most cd's can do 20hz, and many mp3's at least to 30hz, spectral analysis of the content reveals that either its not being recorded onto the content or something else is up the the formats (i.e. there is rarely much content below 40hz on most standard and low def digital sources. Records have it, but they also have rumble and large amounts of distortion at those low frequencies. Real to real tapes also have it, with out the issues of records, but are very rare and pretty impractical. Various HD audio formats will have to be our saving grace I suppose, assuming the engineers don't decide to remove the bass for increased extra's space.

Hmm, I apparently missing the comment about hearing below 30hz, or something like that. I have my hearing checked pretty regularly and I'm quite positive I test to below 30hz. I'll admit to not hearing 20hz, but I normally hear the 25hz tones just fine. I think the cut-off point of people's hearing is going to be person dependent, with age, gender, and genetic variation responsible for this.
 
Here's an internet hearing test.

http://www.phys.unsw.edu.au/jw/hearing.html

Did you spot the bunch of fallacies and silly assumptions that render this test all but worthless? Not to mention many tests by "dispensing audiologists" (who, in many countries are not well trained in their profession) and of course all DIY testing. A lot of "fantastic" testing going on out there.

Can't say as I have much to cavil about with pjpoes and thanks for a clear presentation. I hope that ends a lot this sub-thread debate that is largely theological. BTW, I don't think routine audiometric testing goes below 50 Hz and not sure it would be reliable/meaningful if it did (and certainly outside the realm of interest for any diagnostic purpose), research settings excepted. Anybody have any idea what an isolation booth certified to even 30 Hz would look like?

I sure wish other people would stop saying I can't hear the difference when the woofer is turned off. I only meant to say, as pjpoes said, there's diminished richness rather than an aural awfulness added. But if after cutting off below say 140 Hz, your system then sounds like the proverbial car radio or terrible as some said, trust me, you've got other problems that need attention right away.

Footnote: I don't mean to be mean to audiologists. They do important work and each of us should get a work-up now and then (esp from a Hearing Society or high class testing operation). They are concerned about losses in the useful 2k-8kHz range and their tests (including spoken word hearing) are satisfactory for that purpose. The smart HiFi person knows to get repeated tests to see trends because that's the most reliable test strategy.
 
pjpoes said:
..., but unfortunately the CD and worse yet MP3 lovers have killed any chance of us actually hearing those frequencies. While most cd's can do 20hz, and many mp3's at least to 30hz, spectral analysis of the content reveals that either its not being recorded onto the content or something else is up the the formats.

While the sampling frequency of the CD is limiting the upper range, the lower range of MP3s and CDs goes down into the single digits Hz (probably more limited by the eqipment than the format). I got a lot of CDs and MP3s with content below 20 Hz.

I don't know much about mastering and mixing, but I assume that a the bass below the capabilities of the monitoring system is sometimes cut to make sure there is no low bass rumble and rubbish on the CD, or simply delete it because it is out of their control.

And sometimes there just ain't no infra to record in the first place...
 
Yes, sub-25Hz is mandatory

I would say 10-25Hz is definitely mandatory, as there are several sources such as pipe organs, explosions, cannon fire, etc, that create frequencies in this region. Back when I had an audio system and some audiophile friends, we got together to try to solve this dilemma. We settled on a TL type design, and the only low-fs woofer we knew about back then was a Radio Shack 18” with Fs of 16Hz. The first attempt yielded a cabinet resonance of 10Hz, the second, a bit smaller, ended up at 12Hz. So we actually missed the 16Hz target, but in a beneficial way. Here is a picture of the larger of the two subs.

It was interesting using a signal generator to find the Fs of the combination; at around 13Hz the windows and garage door were also resonant and rattled noisily even though we couldn’t hear the sub working, though it could be felt some. 👍
 

Attachments

  • EB90B4F0-3924-43B9-8469-84EE35ACE49B.jpeg
    EB90B4F0-3924-43B9-8469-84EE35ACE49B.jpeg
    54.8 KB · Views: 263
I'm sorry Daveis but you wont be able to hear sub 20hz. And even at 20hz it will take 120db for you to even start hearing it as that's the threshold of human hearing.

For normal music there is practically nothing lower the low E string on an upright bass which is 41hz. Very few mainly classical works will have organs or even live recording of canon fire going lower than that.

The need for subbass in home audio has grown out of the mistake of trying to recreate earthquake-like tremors in home theaters.

There's a couple of things people seem to forget though, it takes enormous sound pressures for frequencies below 40hz to be even audible and that human hearing is adaptive in the sense that if the harmonic overtones off the base frequency is reproduced correctly then we will hear and feel the base frequency even if it's not there.

So in order to answer your question, no, 10-25hz is absolute not necessary for HT or music. If you absolute want the tremors then buy a bass shaker instead.

this isnt true. the lowest note on the piano is 28 cycles and you can hear 28 hz at far lower sound levels than 100db. much electrical music especially has resoundent bass tones in the 30-40hz range. many drum kicks although mainly higher up do have a broad wave and much can still be felt and heard. as the frequency drops below 30 it does transition more to feeling. but it is still perceiving. and yes you Can hear 20hz. and 30 hz you can hear very well. with a decent sub you should hear the dominant fundamental node louder than the overtones produced.
 
On the Enya track “Longships” there is a drum fundamental at 12Hz that is completely lost without a true subwoofer.
No, you can’t hear these frequencies, but you can feel them as a “shudder” in the air around you. No, they are technically not audio, but I consider them to be a desirable part of music.

Amd no, they don’t have to be “loud” to be appreciated.
 
Last edited: