Subwoofers that go below 30 hz for music

[Hayabusa]

Shows the max useful SPL (that fall within the "acceptable" distortion range)

Nice graph, not clear how ' acceptable' was defined but at least the SB2000 has 94dB usable output at 20Hz, still nicely audible, and the pro version has some extra 3dB output to work with. So we can conclude it is use full and not as you said; "Not without massive distortion which is not really useful. "

 

[sigbergaudio]

Quite so but I suspect like in many engineering efforts near extremes what people want and what is technically feasible diverge

An old bloke like me who enjoys "The Dream of Gerontius" would rather have the bass than the treble, I probably hear nothing above 10kHz, probably less, but I still find the low organ pedal notes in the final "softly and gently" are a key aspect of this very moving piece but have only ever heard them on one system (mine) using a very old REL Studio, bought from Richard Lloyd in Wales when he still owned the company.
On most HiFi one wouldn't know they exist.
Personally I would MUCH rather have "listenable" the lowest octave from 20-40 Hz than the highest octave from 10-20kHz which I can no longer hear even though it is only on a few pieces of music.

Totally agree, but 20-40hz is much easier than 10-20hz.

I also suspect quite a few people in these types of discussions have never tried to listen to a 20hz sine wave. It's so low it's felt more than heard, and is not audible at all unless played very loud.

It looks good when you measure your system with REW and see that you can reproduce 20hz, but then when you play music at a moderately high level, say 80dB, 20hz will be 100% inaudible.

 

[Frank Dernie]

Totally agree, but 20-40hz is much easier than 10-20hz.

I also suspect quite a few people in these types of discussions have never tried to listen to a 20hz sine wave. It's so low it's felt more than heard, and is not audible at all unless played very loud.

It looks good when you measure your system with REW and see that you can reproduce 20hz, but then when you play music at a moderately high level, say 80dB, 20hz will be 100% inaudible.

Quite so.
I have absolutely no interest in 10-20Hz, that is sub sonic film effect territory and the 16Hz from the 64' organ pipe is rarely used so easier sensed in a cathedral than reproduced at home but there is organ repertoire in the 20 to 40Hz band and most speakers "ignore" it and many subwoofers don't go loud enough.

 

[Andysu]

below 30Hz good enough for my THX cinema

 

[olieb]

If you can find another study great , please post it, I certainly don't think this is definative but at least they tried, which is more than you can say about the CEA standard where they apparently just made the numbers up or got industry input (which is my guess but I obviously can't prove it).

The problem here is that these results are somehow up in the air (at least it seems to me). What was the setup and how did they measure this? Without the text this is just a sheet with numbers.
I will have to read the documents @sigbergaudio provides (Thanks!) to check about your claims of the "industry".
But I found another "study" on a German webpage. https://www.hifi-selbstbau.de/index...erschiedenes/klirrfaktor-wie-viel-ist-zu-viel

They did not measure themselves, but used results of Zwicker (in ISO532B) about masking of sounds to derive the audibility of harmonics in the presence of the fundamental. (Masking has been studied in the context of MP3 and such, too).
This procedure takes Fletcher-Munson into account.
These are their results for the audibility of H2 ("K2") at different levels of SPL. (Other harmonics are hidden behind a subscription paywall.)
In the context here the most left part of the curves is relevant.

Here is a diagram with the results from your sheet to compare.
Blue is for H2, Red for H3. (Thicker line means higher SPL.) The dashed line is the result for 80dB from the above graph.
The trend is somewhat similar but there is a difference of at least 15dB.
The values for audibility in this result are obviously much higher than in your sheet and more in line with "common wisdom".
Now, who is right? Again, it would be interesting to know how the results in your sheet were measured.

 

[Hayabusa]

below 30Hz good enough for my THX cinema
View attachment 473511

nice! -3dB 5.5Hz !! max spl at 10 and 20 hz? at what distortion?

 

[Andysu]

nice! -3dB 5.5Hz !! max spl at 10 and 20 hz? at what distortion?

i like the JBL cinema professional subs

 

[levimax]

The problem here is that these results are somehow up in the air (at least it seems to me). What was the setup and how did they measure this? Without the text this is just a sheet with numbers.
I will have to read the documents @sigbergaudio provides (Thanks!) to check about your claims of the "industry".
But I found another "study" on a German webpage. https://www.hifi-selbstbau.de/index...erschiedenes/klirrfaktor-wie-viel-ist-zu-viel

They did not measure themselves, but used results of Zwicker (in ISO532B) about masking of sounds to derive the audibility of harmonics in the presence of the fundamental. (Masking has been studied in the context of MP3 and such, too).
This procedure takes Fletcher-Munson into account.
These are their results for the audibility of H2 ("K2") at different levels of SPL. (Other harmonics are hidden behind a subscription paywall.)
In the context here the most left part of the curves is relevant.
View attachment 473497
Here is a diagram with the results from your sheet to compare.
Blue is for H2, Red for H3. (Thicker line means higher SPL.) The dashed line is the result for 80dB from the above graph.
The trend is somewhat similar but there is a difference of at least 15dB.
The values for audibility in this result are obviously much higher than in your sheet and more in line with "common wisdom".
Now, who is right? Again, it would be interesting to know how the results in your sheet were measured.

View attachment 473512

The study you quoted is interesting but very far removed from an actual study of people listening to music or even test tones.

The "Table 2" I posted is all over the internet but the complete study is behind a paywall or maybe not available as I tried to see how much it was and could not add it into the cart. Anyway below is the complete title of the study, if you happen to come across the entire study please let me know how to get it as I am interested in reading it. There seems to be more than just distortion mentioned.

 

[levimax]

Where did that come from? The numbers appear to be garbage.

 

[Andysu]

tos remix 5.1 has some infersonic , it could have been there or maybe not been there but instrumentation reads it was there , " balance of terror " has some neat infersoincs
to some infersonic maybe nonexistence often there maybe only noise down there or well depends on point of view

i have design idea , oh don't we all , this is into the known of unknown deeper sub bass below 30Hz way below 10Hz way beyond the known physical side of sub universe

even my room does infersonic down to very single low digits , shhh not saying how

 

[Anton S]

I've never had a subwoofer that doesn't roll off steeply below 30 hz, and I understand that very little music content is below 30 hz. I've read in a few places that when a music recording does have content below 30 hz, it's usually filtered out during mastering in order to make it play better on most listeners' systems. However, I know that there is music with synths, pipe organs, and even some pianos (the lowest notes on acoustic pianos) that has sound components below 30 hz. I just turned 60 and I'm thinking maybe it's time I spring for a sub that goes really low just to hear what I am missing.

Has anyone had a sub that sounds good below 30 hz and have you ever listened to music (not talking about movies) where it makes a difference?

My SVS SB16 reaches down well below 20 Hz without any problem. And I have a dbx subharmonic synthesizer in the system to restore the super low notes that are generally left on the cutting room floor during mastering. Love it!

 

[Andysu]

i knew there was something below beyond the known sub universe

 

[olieb]

The study you quoted is interesting but very far removed from an actual study of people listening to music or even test tones.

There is nothing wrong with using data from (extensive) research, and this research includes vast amounts of "people listening" to all sorts of sounds. Masking has been studied a lot!

Thanks for the Fielder paper reference, but this only shows how much is lost if the context is not there. This is named a "cornerstone paper" but without the text it remains obscure how CTA-2010 can come up with their "high" requirements if the "true" thresholds of audibility would be so much lower. In my view it is much too simple to blame the "industry to make themselves look better than they are and claim BS on the standard". After all the paper comes out of the Dolby Lab! (One could claim BS on it as well.)

From the abstract I would not assume that Fielder and Benjamin actually measured something/much (about harmonic distortion).
"Psychoacoustic data are used to establish to was degree errors are perceptible" seems to me that they were doing similar stuff as in the German webpage (but with different assumptions?).

And what Linkwitz refers to about the Fielder-Benjamin paper hints in the same direction of "assessing data".
From https://www.linkwitzlab.com/thor-measmt.htm

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).

After all this is the same Louis Fielder who came up with the need for ≈115dB dynamic in sound reproduction by taking the range from the loudest peak he measured with live music to the threshold of hearing room noise in a very quiet environment (perfect silence with digital zero). [Dynamic-Range Issues in the Modern Digital Audio Environment]
One can argue for that, but it is quite an extreme position and it needs proper context, as otherwise Red Book would seem insufficient for music.

Long story short, without the full text I would take the numbers from the 1988 sheet with a rock of salt.

 

[Andysu]

My SVS SB16 reaches down well below 20 Hz without any problem. And I have a dbx subharmonic synthesizer in the system to restore the super low notes that are generally left on the cutting room floor during mastering. Love it!

View attachment 473584

show the dbx as a yt video only show it in action , there bundles of these now cheaply , show me the dbx with king kong

 

[bmc0]

What's with the 110% for 10Hz at 80dB, then for 100 and 110 dB the levels are way lower?

Well, the 2nd harmonic of 10Hz is 20Hz and the threshold of hearing at 20Hz is about 75dB SPL. In fact, the 10Hz numbers are all the same absolute level for each harmonic: H2 is 81dB SPL, H3 is 67dB SPL, H4 is 56dB SPL, and H5 is 49dB SPL. This approximately follows the ISO 226:2003 threshold curve, except all values are about 6dB higher. In other words, the data appear to show that masking is not significant for the 10Hz fundamental and that the limiting factor is simply the threshold of hearing. The 20Hz and higher fundamentals do not follow the same pattern.

 

[levimax]

There is nothing wrong with using data from (extensive) research, and this research includes vast amounts of "people listening" to all sorts of sounds. Masking has been studied a lot!

Thanks for the Fielder paper reference, but this only shows how much is lost if the context is not there. This is named a "cornerstone paper" but without the text it remains obscure how CTA-2010 can come up with their "high" requirements if the "true" thresholds of audibility would be so much lower. In my view it is much too simple to blame the "industry to make themselves look better than they are and claim BS on the standard". After all the paper comes out of the Dolby Lab! (One could claim BS on it as well.)

From the abstract I would not assume that Fielder and Benjamin actually measured something/much (about harmonic distortion).
"Psychoacoustic data are used to establish to was degree errors are perceptible" seems to me that they were doing similar stuff as in the German webpage (but with different assumptions?).

And what Linkwitz refers to about the Fielder-Benjamin paper hints in the same direction of "assessing data".
From https://www.linkwitzlab.com/thor-measmt.htm

After all this is the same Louis Fielder that came up with the need for ≈115dB dynamic in sound reproduction by taking the range from the loudest peak he measured with live music to the threshold of hearing room noise in a very quiet environment (perfect silence with digital zero). [Dynamic-Range Issues in the Modern Digital Audio Environment]
One can argue for that, but it is quite an extreme position and it needs proper context, as otherwise Red Book would seem insufficient for music.

Long story short, without the full text I would take the numbers from the 1988 sheet with a rock of salt.

I wonder why there are only 2 papers on this subject neither of which actually use listening tests? In any case I think there is an obligation for the CEA-2010x authors to explain the methods they used to determine their thresholds as their tests are used extensively for marketing purposes. If there isn't any hard data then that should be disclosed. In the absence of hard data to me it seems like an arbitrary and conservative distortion figure for all frequencies should be used. Yes this might be "too high" a threshold for some frequencies but at least it would be consistent and easily understood by people that wanted to look a little deeper. The current system and the discrepancies between 2010-A and 2010-B with some apparently arbitrary threshold level jumps, is very confusing even for relatively technically oriented readers.

 

[Andysu]

tested real world bus ride , no years ago , spectrum lab ( too many colours ) plenty below 20Hz infersonic wow best sub bass mechanical vibration period and a lot , lot , lot more besides that , a project idea i have in mind

pictures while recording

screen capture while multi tasking on the bus

 

[Andysu]

i get decent fair lows infersonics in the THX cinema

 

[Andysu]

well heard felt the subs 30Hz and below low , below low , low now for something different
jedi's blaster lightsabers JBL with beam splitters few lasers fitted on microphone booms and angled aligned to a tin can with a balloon stretched over near the JBL subs

 

[Palmspar]

Quite so.
I have absolutely no interest in 10-20Hz, that is sub sonic film effect territory and the 16Hz from the 64' organ pipe is rarely used so easier sensed in a cathedral than reproduced at home but there is organ repertoire in the 20 to 40Hz band and most speakers "ignore" it and many subwoofers don't go loud enough.

A 32ft organ pipe has 16 hz if i remember correct, 64ft can play 8hz.
Use 3 diy sealed subs with 5x12'" drivers they can play 105db @10hz inroom.
I can hear a 16hz sine wave, but above 12000hz not so much, invest in good base is a better choice for me.