Opinions Please Frequency Response and SPL capability

[oneoldude]

Looking for the modern thinking re: ideal speaker system freq. response and SPL capability (within Xmax) at 1 m for home use.

The thinking I am used to would be about:

30 Hz to 16K Hz with 110 dB SPL across the band for both music and home theater purposes.


Are these still proper targets or does modern media actually have frequencies below 30 Hz for either music or HT?

Perhaps the better question is: What is the actual current frequency bandwidth (actually recorded on the media, not its capability) on modern music and HT sources?

Can you point to useful URL's on this? I could not find any.

Thanks

[Face]

Yes, some HT and Music does have material below 30Hz, and in some cases even below 20Hz.

http://www.avsforum.com/t/1347751/purpose-of-flat-response-below-20hz-why-does-it-matter

http://www.avsforum.com/t/1439484/music-whit-20hz-bass

[oneoldude]

Thanks for the URL's.

There was some synthetic music with 7 Hz but most of the rest seemed to be residual noise rather than actual signal. Most real bass seemed to be around 30 Hz.

Thanks

[AdelaaR]

Interesting, Face, thanks.

[cjd]

Thanks for the URL's.

There was some synthetic music with 7 Hz but most of the rest seemed to be residual noise rather than actual signal. Most real bass seemed to be around 30 Hz.

Thanks

As long as you ignore classical. Organ and bassoon both dig a good bit deeper than 30Hz.

That said, you may not be able to hear anything much below 30Hz, and it's more of a subtlety that it will add. You'll end up with 1 octave higher on most stuff being rather audible and that's what you're most likely to care about.

On the other hand, something is missing for me without solid sub-bass. There are some movies that make good use of it (rather than just "oooh" momemts) - LOTR, for example, when you've got all the horses running, it's not just noise at that point. Bass shakers can get you pretty close to this without nearly the expense or complexity.

Is it worth it? That's up to you. It's there. It's less common. And I have no idea if you notice it.

110dB is painful for too long.

C

[BobEllis]

To me, 30 Hz seems to be a good target for a system crossed to a sub at 50-60 Hz - try to keep driver response flat an octave either side of a crossover. I'm with Chris on there is something missing if the total system doesn't extend well below 30 Hz. Sure, there is a lot of gratuitous noise down in the 20s especially in movies, but there is something worth having on a lot of music and movies. Tympani or even kick drums don't sound as solid and realistic if the system doesn't dig deep. A piano can go to 27Hz. http://en.wikipedia.org/wiki/Piano_key_frequencies

I recently changed from a ported sub with a quasi fifth order rolloff theoretically 3 dB down at 24Hz to an infinite baffle system that has a theoretical second order rolloff and 3 dB down point of 23 Hz. The specs don't sound like much of a change, but the difference is amazing. I cross to the sub at 60Hz, and even electric bass is more realistic. Movies are more interesting when the soundtracks are good. Funny that Chris mentioned the horses in LOTR. That was an "Aha!" moment for me testing out my IB sub.

110 dB capability seems reasonable for dynamic peaks, but as Chris suggests don't stay there long. Tweeters will likely be the limiting factor. Clean to 105 dB isn't bad.

A related question: Are some/many recordings equalized so the deep bass is very hot with the idea that most consumers have systems that roll off "early"? Flat system response sounds very bass heavy to me on many pop recordings and special effects heavy movies. Any mastering engineers out there care to comment?

[---k---]

I agree. I enjoy my sub tuned to 15.5 hz. I wouldn't go back to a sub that couldn't produce sub 25 hz.

[oneoldude]

Oh, I do not leave out classical. Among my favorites are the Telarc 1812 and and the Toccata and fugue in D minor. Fanfare for the Common Man, Also Sprach Zarathustra and many other bass heavy pieces are in the group as well. See the graphs below. I was really surprised at the bass in organ pieces. I thought it was much lower than was measured.

I usually do not save old magazines. But I have saved the July 1989 issue of Audio because of the following article. On page 43 there is an article titled "How Hot Are CDs?" The author measured the peak levels of the recorded sound for 40 CDs. He went from Mozart's Quintet for Piano and Winds to Telarc's 1812 and included rock and pop music as well. It is the best representation of what was actually on CDs that I have ever seen.

I was wondering if the peak level relationship today is any different than it was in 1989 and I have been unable to find an equivalent contemporary study. That is why I started this thread.

For those who are interested, I have attached images of the graphs illustrating the author's findings. Most of the big bass peaks are at around 40 Hz. If that is still the case, and given the Fletcher Munson curves showing our poor sensitivity to low freq sounds, I wonder what is really happening when people find their system more satisfying if it extends to 15 Hz. It might be that by extending the system capability to 15 Hz, the system is better able to handle bass where it really is, about an octave+ above 15 Hz.

[BobEllis]

I think that's part of it. If the sub can handle decent output <20 Hz then 30 Hz excursion is minimal and distortion is lower. But I believe that being able to reproduce the fundamentals adds a lot even if it is at a low level. Somewhat like the cheap systems that use harmonics to simulate bass extension are acceptable to some but most of us here recognize the need to truly dig deep. For me reaching into the teens adds realism to a lot of music and is worth the expense. As always in this hobby, YMMV.

Fletcher Munson probably tested using headphones. As Chris' bass shaker recommendation suggests we may not hear the low frequencies but we can sense them in other ways.

[Face]

Yes, tactile sensation. Having a speaker being able to produce below 20hz really brings everything together.

[---k---]

I don't think you'll find any disagreement that the majority of bass lives in the 40 - 60 hz range. And, I'm betting most people who have very capable subs down to say 30 - 35 hz are very satisfied and will recommend their systems to friends. Bass below 20 hz seems to be one of those things that people who have it say that it adds a magical, indescribable something. Those without it, say that they've read articles that say we can't hear it, so they don't believe it.

I think you got most of the answer above. I also think there is some voodoo magical interaction between the frequency and the harmonics that is missing when you can't reproduce them. This seems to be one of areas that isn't researched yet and have definitive answers.

I should point out that I'm coming at this mostly from a HT perspective. CJD and Bob can probably provide more of the music perspective.

[madmac]

When I spun a bass test EQ disc with my setup for my subs, when it approached 30-20hz, my windows started to 'rattle'!!. It was kinda' cool!!. The only time I've heard bass that low after that was when the martians came out of the ground in the modern version of "War of the Worlds". My daughter was asleep at the time upstairs and I was SURE her bed was 'skirting' across the floor as that scene happened!!. By the way.....she DID NOT wake up!!. Kids will sleep through anything when they are trained to hear music and movie's whilst asleep. It's a audio/video snob dad that plays loud stuff kid kinda' phenomenon!.

[BobEllis]

That scene sent the cat running to her safe spot in my closet.

[jim1961]

(measured at the listening position)

When I went from the black curve to the red one, their certainly was some "magic" added as some are calling it. I am a bit skeptical in gaining more by extending things below 20hz though. I think if one does movies, there is more to gain than in dedicated 2 ch audio. Of course, there is that occasional 16hz pipe organ to consider

[jim1961]

Oh, I do not leave out classical. Among my favorites are the Telarc 1812 and and the Toccata and fugue in D minor. Fanfare for the Common Man, Also Sprach Zarathustra and many other bass heavy pieces are in the group as well. See the graphs below. I was really surprised at the bass in organ pieces. I thought it was much lower than was measured.

I usually do not save old magazines. But I have saved the July 1989 issue of Audio because of the following article. On page 43 there is an article titled "How Hot Are CDs?" The author measured the peak levels of the recorded sound for 40 CDs. He went from Mozart's Quintet for Piano and Winds to Telarc's 1812 and included rock and pop music as well. It is the best representation of what was actually on CDs that I have ever seen.

I was wondering if the peak level relationship today is any different than it was in 1989 and I have been unable to find an equivalent contemporary study. That is why I started this thread.

For those who are interested, I have attached images of the graphs illustrating the author's findings. Most of the big bass peaks are at around 40 Hz. If that is still the case, and given the Fletcher Munson curves showing our poor sensitivity to low freq sounds, I wonder what is really happening when people find their system more satisfying if it extends to 15 Hz. It might be that by extending the system capability to 15 Hz, the system is better able to handle bass where it really is, about an octave+ above 15 Hz.

I am afraid those graphs are not legible.

[BobEllis]

[ATTACH=CONFIG]22046[/ATTACH]

(measured at the listening position)

When I went from the black curve to the red one, their certainly was some "magic" added as some are calling it. I am a bit skeptical in gaining more by extending things below 20hz though. I think if one does movies, there is more to gain than in dedicated 2 ch audio. Of course, there is that occasional 16hz pipe organ to consider

Because you can extend it down there. There are a number of movies that will challenge a ported sub without a high pass filter. An interesting thread with lots of graphics showing low frequencies here: http://data-bass.ipbhost.com/index.p...ic-etc/page-17 Google got me to page 17, but the whole thread is interesting for bass head movie lovers like me. Yes, I have some 16 Hz pipe organ music, 1812 Overture and others that dig deep.

[JonMarsh]

One point not mentioned here, which sometimes weighs in favor of a low cut off frequency, is avoiding interacting with the behavior near the actual cut off frequency.

Starting from the point that the best bass definition I've heard or measured (in terms of transient response) is from IB's or sealed systems with a Q of 0.6 or less, if you do choose to go with a ported system with an Fb of 15-18 Hz, in the normal music range, you're well above the range of port output, and operating more like a typical sealed box system. This sounds better to my ears than having a port with significant output right in the music range (say, tuned to 35 or 30 Hz, where the port output is substantial for up to one octave above Fb.)

For HT extension and definition, nothing touches a IB manifold array with a bunch of drivers with Fs of 20 Hz or so, eq'd to be flat to Fs in room.

[cjd]

For HT extension and definition, nothing touches a IB manifold array with a bunch of drivers with Fs of 20 Hz or so, eq'd to be flat to Fs in room.

I admit, I'm looking forward to attempting a comparison between my IB (4 Tempest-X) and my some-day dipole subs (4 Maelstrom-X 18's)...

C

[oneoldude]

Sorry Guys,

Here are the peak levels in legible size.

I would love to see a similar comparison for contemporary recordings including HT audio. It would also be nice to extend the measurement down to as low as current signals exist. But LF noise that is not intended must be not be part of the act.

If it is too small to read try to click on the image to expand. I have done my best.

[oneoldude]

[ATTACH=CONFIG]22046[/ATTACH]

(measured at the listening position)

This is what I was talking about. You gained improvement to about flat in the real bass range being almost flat from 25 to 45 Hz. If you have good transient response and can hit sufficient SPL (within Xmax) down there, it should be a beautiful thing to hear.

[oneoldude]

Hello jonmarsh,

Where should an IB or other low Qt sub be crossed to the main speakers? Are there guidelines or rules of thumb for that sort of thing other than when the mains are running on the edge of Xmax and the xover should be low enough to not affect directionality??

[Face]

Before the dipole peak, where the woofer's distortion is as low as the dipole's, and preferably before 120hz as localization may occur.

[oneoldude]

Before the dipole peak, where the woofer's distortion is as low as the dipole's, and preferably before 120hz as localization may occur.

Hi Face,

You must be a mind reader. Yes, I am a dipole sort of guy, but have not discussed it in this thread. Interesting. Should I be wearing an aluminum foil helmet on my head? 8O

I am unclear about your post. As I understand it, Dipole woofers should be used below their dipole peak so as to be in full dipole mode throughout their limited bandwidth and not have to deal with the irregularities that occur at or above the dipole peak. Ideally, they should run from around Fequal or a bit more (well below the dipole peak) down to where they start to run out of Xmax and distortion begins to rise.

Hopefully the dipole woofer is able to get down to reasonably low frequencies. By that I mean re: amplitude localization, below 200 Hz and Re: phase localization, below 80 Hz. The subwoofer should then pick it up from there. Hmmm, I guess I just answered the question I posed to jonmarsh.

So, the rule of thumb should be that a sub should be crossed to the main speakers at or below 80 Hz and the design of the main speaker should be such that at Xover its woofer does not exceed Xmax for a system design level of 110 bB SPL or better.

Did I get it right?

[benchtester]

[QUOTE=oneoldude;576730]Sorry Guys,
...If it is too small to read try to click on the image to expand. I have done my best. ..QUOTE]


Clicking didn't help me. But "control +" (to expand) and "control 0" (to return to normal) are working fine.

Nice charts by the way.

[oneoldude]

Nice charts by the way.

I think so too. I go back to them from time to time to remember what is really going on. Gives me a warm and fuzzy feeling of stability. I was hoping that there has been some contemporary published work that would show whether things have changed much since 1989. Guess not.

[Face]

Hi Face,

You must be a mind reader. Yes, I am a dipole sort of guy, but have not discussed it in this thread. Interesting. Should I be wearing an aluminum foil helmet on my head? 8O

I am unclear about your post. As I understand it, Dipole woofers should be used below their dipole peak so as to be in full dipole mode throughout their limited bandwidth and not have to deal with the irregularities that occur at or above the dipole peak. Ideally, they should run from around Fequal or a bit more (well below the dipole peak) down to where they start to run out of Xmax and distortion begins to rise.

Hopefully the dipole woofer is able to get down to reasonably low frequencies. By that I mean re: amplitude localization, below 200 Hz and Re: phase localization, below 80 Hz. The subwoofer should then pick it up from there. Hmmm, I guess I just answered the question I posed to jonmarsh.

So, the rule of thumb should be that a sub should be crossed to the main speakers at or below 80 Hz and the design of the main speaker should be such that at Xover its woofer does not exceed Xmax for a system design level of 110 bB SPL or better.

Did I get it right?

My only dipole sub experience is with ripoles. LF response and 110db would be very difficult. For a standard OB, you would need multiple, large drivers, and response at 110db near 20hz would still be tough.

[oneoldude]

Face, you really are a mind reader! I am in the process of building a folded dipole as we speak.

Basically it is 1/2 of the Linkwitz "W" dipole he used for the Phoenix project. I guess the Ripole is a folded dipole with the openings squeezed down a bit to increase acoustic impedance on the driver, but in principle it is still a folded dipole. Sure, a folded dipole will not go super low, but that is what subwoofers are for.

In using the Linkwitz spread sheet it appears that a hefty 12" with Xmax of 14+ mm in a folded dipole configuration should be able to produce around 100 dB SPL at around 50 Hz in free space. Put on the floor makes it half space for an additional 6 dB. Add another 6 dB from the sub through the crossover and another 6dB from the other side, you end up with about 118 db SPL. Even if the lift from half space only gets you 3 dB and if the other side only gives you 3 dB that is still 112 dB SPL. And that is at 50 Hz. If the crossover to the sub is a bit higher,the numbers will be even better.

Now I know this is all theoretical and must be proven in practice, but it seems like a pretty good bet.

Since you have experience with Ripoles, will you share that experience? I'd like to hear the good the bad and the ugly, especially re: the quality of bass produced by a Ripole and crossing over from a Ripole to a sub.

Now if your experience has been with trying to make a Ripole a sub. Well then I feel for you. That would be real tough sledding.

Thanks

Now where is that tinfoil?

[Face]

Here is a Ripole using two inexpensive JBL 15" woofers, wired at 8ohms, measured at 1 meter. http://www.jbl.com/estore/jbl/us/pro.../GT5-15_JBL_US





At this SPL, distortion started to rise, which had me concerned that I may run out of headroom at 4 meters away(listening position). I had given the thought of trying a pair of AE dipole woofers, but have too many other irons in the fire right now.

[oneoldude]

Face, nice driver choice. Nice response curve.

I did a Linkwitz spread sheet analysis of your system and you made some fine choices. The numbers speak for themselves.



Your work is very exciting. Will you please share some info? ;x(

Will you review the sonic results of your work? You left me hanging. How does it sound?

Your SPL graph looks ideal. What were the measurement conditions?

What were Fs and Qt with the drivers in the Ripole?

Was efficiency significantly reduced by the Ripole?

Did you notice any energy storage problems with the Ripole?

How does the Ripole distortion compare against other low frequency designs you have measured?

I know I am asking a lot. But the results of your excellent work should be disseminated.

Thanks

[lhwidget]

Hey oneoldude,
You and John, have me wondering about something I've seen a couple of times. An old JBL write-up about a subwoofer they introduced back around 1983 (http://www.lansingheritage.org/html/.../1983-subs.htm), and Lynn Olsen's comments about good sounding bass (buried somewhere in the epic 900 page thread about his Ariels (http://www.diyaudio.com/forums/multi...ariel-748.html Sorry, I have no idea where now...).

You take an efficient (read Pro style, look for a low Vas value and high efficiency) woofer and put it the biggest box you can live with, tune it very low, then apply a simple first order filter at the box tuning frequency to flatten out the response. When the box is big enough, the rise in response will hit 6 dB/octave.

This approch puts the port contribution basically in the sub-sonic region, and you rely on the driver's high efficiency to generate a usable response curve after applying the filter.

Using Jeff Bagby's Woofer Box Model and Circuit designer spreadsheet, I modeled an 18" B&C pro woofer just playing around with the factory TS parameters and I arrived at the following:
Vbox = 6.7 ft^3
Fb = 20 Hz
F3 for the 1st order low pass filter is 20 Hz
I did include an 8th order high pass filter @ 15 Hz to keep the displacement in check, it's perfectly willing to self destruct without one...
A 100 watt input will drive it to about 105 dB @ 1 meter, the graph below is with 300 watts input.

[lhwidget]

By the way, I'm all for strong response down to 20 Hz, I just don't have much real experience with it

Right now I'm using 2.1 systems with smaller subs that are only getting down to about 30-35 Hz.

What I really want to hear is Hans Zimmer's Mombassa (from the Inception sound track) on a system that can actually reach 20 Hz without much effort. I watched the line level signal on SoundEasy's RTA and there's this rolling bass line that stays strong and slides down into really deep (subsonic) bass regions. I'd love to experience that the way he intended. http://stuckincustoms.smugmug.com/Po...29GQa&lb=1&s=A

Now I'm curious about the bass in other pieces after hearing your opinions, thanks for posting guys.

[BobEllis]

I was happy with a pro woofer in a big box until I went IB. I had a pair of JBL2245h in 8.5 cubic feet each tuned to 24 Hz with a Q=2 high pass filter at 24 Hz to flatten it out. I was happy with the sound but having 2 of these behemoths in any room made it too much of a bachelor pad. Then I went IB, which just sounds more realistic even in the 30-40 Hz range. Maybe it's the group delay of the ported box with EQ on top of it. It's also pretty nice to have that floor space back.

Not to discourage making a decent conventional sub, just acting like any religious convert and going a little overboard touting the benefits of IB. Given a choice between a 30 Hz sub and a huge box digging into the low 20s, I'd take the huge box(es) any day.

[lhwidget]

Thanks Bob, that's good to know.

[AdelaaR]

A related question: Are some/many recordings equalized so the deep bass is very hot with the idea that most consumers have systems that roll off "early"? Flat system response sounds very bass heavy to me on many pop recordings and special effects heavy movies. Any mastering engineers out there care to comment?

I have been asking myself this very same question for a long time now.
Especially music intended for younger audiences is likely recorded with exaggerated bass because those youngsters a) just lOOOve the bass and b) mostly listen with small speakers that don't dig deep.

[Face]

A related question: Are some/many recordings equalized so the deep bass is very hot with the idea that most consumers have systems that roll off "early"? Flat system response sounds very bass heavy to me on many pop recordings and special effects heavy movies. Any mastering engineers out there care to comment?

This heavily depends on the room. Tuned flat to 20hz is only a little bass heavy here.

[jim1961]

This heavily depends on the room. Tuned flat to 20hz is only a little bass heavy here.

+1

Ive seen rooms add 10db or subtract 20db in the low bass region.

[BobEllis]

I've noticed the issue with my system(s) in a number of rooms, as well as in my car (standard issue, sounds flat-ish on most music). It does rear its head as "one note bass" so you may be on to my answer, Face.

[Face]

Playing around with this will help determine what LF response will best match your room: http://audio.claub.net/software/jbabgy/BDBS.html

[oneoldude]

My experience in real life has shown me that except for extreme HE explosives which sound like a sharp crack, mortars, bombs, cannons and the like have a very tight bass sound unless they are very far away from you or you are so close that your eardrums clip. The same is true for deep musical bass.

Unfortunately, in order to get proper WAF and to save living room space, most base units, including sub woofers are placed in a smallish box with a Q of at least .7 and often much more. Also, vented boxes have their own resonances. Most in home reproduced music and movie sound effects end up with a rubber quality that makes the bass notes booooom rather than have that natural tight sound. On top of that, with loud low sounds, there is often severe clipping of both the amp and speaker that breaks up the sound in a very dirty way. Lots of folks get used to those problems and end up thinking they sound realistic. I guess they never heard the real thing.

My preference is for a very low Q woofer/sub woofer (Q .5 to .6) that makes use of the the room modes and then is equalized as necessary. Its much tighter and more realistic to me. This can be achieved with large sealed boxes, smaller boxes with a Linkwitz transform or other equalization to lower effective Q or IB subs in the attic, basement, etc. I guess it could be done with a dipole as well but I have never tried it. And size matters. Even in moderate to smaller rooms it takes at least two long throw 12s and serious amplifiers to really get the job done.

Grab the Linkwitz spreadsheet spl_max from his web site and see what it takes. Remember to add 6 dB for half-space and possibly more for quarter-space if applicable and another 6 dB for each doubling of woofers/subwoofers (2,4,8 etc.) in the very low frequencies.

When you use that spreadsheet properly it will tell you how close you are to getting it as good as it can be. Linkwitz has a bunch of very good spreadsheets on his site. Pull them down and check them out.