Bass Quality in Sealed Speakers

[dualazmak]

Hello OP @mike7877,

Just for your reference, I have intensively "measured" transient behavior of my 30 cm woofer YAMAHA JA-3058 in sealed NS-1000 cabinet and my L&R sub-woofer YAMAHA YST-SW1000; if you would be interested in my method and results, please visit my posts on my project thread;
- Measurement of transient characteristics of Yamaha 30 cm woofer JA-3058 in sealed cabinet and Yamaha active sub-woofer YST-SW1000: #495, #497, #503, #507
The objectively measured data (especially the 3D Fq-Gain-Time sound-energy-distribution color-spectrum given by ADOBE Audition 3.0.1) are/were very much useful for determination of optimal XO frequency and the LP/HP filters (and slopes) for sub-woofers and woofers.

I assume these related posts would be also of your interest and reference;
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-1_ Precision pulse wave matching method: #493
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-2_ Energy peak matching method: #494
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-3_ Precision single sine wave matching method in 0.1 msec accuracy: #504, #507
- Perfect (0.1 msec precision) time alignment of all the SP drivers greatly contributes to amazing disappearance of SPs, tightness and cleanliness of the sound, and superior 3D sound stage: #520

- Not only the precision (0.1 msec level) time alignment over all the SP drivers but also SP facing directions and sound-deadening space behind the SPs plus behind our listening position would be critically important for effective (perfect?) disappearance of speakers: #687

- Reproduction and listening/hearing/feeling sensations to 16 Hz (organ) sound with my DSP-based multichannel multi-SP-driver multi-amplifier fully active stereo audio system having big-heavy active L&R sub-woofers: #782

Furthermore, if you would be curious, you can find here #931 the details of the latest setup of my multichannel multi-SP-driver multi-amplifier fully active audio system.

Edit:
If you would be seriously interested in using the test tone signal tracks I prepared and used in above measurements and tunings, and if you would be also interested in the contents and PDF booklet of "Sony Super Audio Check CD" (ref. #651, #750, #760), please simply PM me writing your wish.

Furthermore, many of the "music tracks" within my "Audio Reference/Sampler Music Playlist" consists of 60 tracks (summary ref. here as independent thread) would be very much suitable for your subjective listening confirmation after your objective measurements and tunings, I believe.
Again, if you would be seriously interested in using/listening such intact tracks of my "Audio Reference Playlist", please simply PM me writing your wish.

 

[mike7877]

If you check the post that I linked above you will see that the main drawback of using LP filters is the elevated pre-ringing
Now if that is audible that is a different question... the amount of pre-ringing depends on the filters you use (Q value mainly) and the implementation of the linear phase plugin

Another drawback is the time delay but if you are just listening to music that is a non-issue (given that you apply the same LP plugin on all the channels ensuring equal time delay for all). And even if you watch movies there are players (like Jriver) that will automatically delay the video to sync to the audio - so again, no issue

Basically you can boost the lows of a sealed speaker until you reach either the xmax of the driver and/or the power output limit of the amp
I ensure no clipping in the digital domain by using 64-bit digital volume control upstream (=volume control is the first DSP item in the chain) in Jriver

Yes, I did see that after I replied to you.

Very good!

 

[mike7877]

Hello OP @mike7877,

Just for your reference, I have intensively "measured" transient behavior of my 30 cm woofer YAMAHA JA-3058 in sealed NS-1000 cabinet and my L&R sub-woofer YAMAHA YST-SW1000; if you would be interested in my method and results, please visit my posts on my project thread;
- Measurement of transient characteristics of Yamaha 30 cm woofer JA-3058 in sealed cabinet and Yamaha active sub-woofer YST-SW1000: #495, #497, #503, #507
The objectively measured data (especially the 3D Fq-Gain-Time sound-energy-distribution color-spectrum given by ADOBE Audition 3.0.1) are/were very much useful for determination of optimal XO frequency and the LP/HP filters (and slopes) for sub-woofers and woofers.

I assume these related posts would be also of your interest and reference;
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-1_ Precision pulse wave matching method: #493
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-2_ Energy peak matching method: #494
- Precision measurement and adjustment of time alignment for speaker (SP) units: Part-3_ Precision single sine wave matching method in 0.1 msec accuracy: #504, #507
- Perfect (0.1 msec precision) time alignment of all the SP drivers greatly contributes to amazing disappearance of SPs, tightness and cleanliness of the sound, and superior 3D sound stage: #520

- Not only the precision (0.1 msec level) time alignment over all the SP drivers but also SP facing directions and sound-deadening space behind the SPs plus behind our listening position would be critically important for effective (perfect?) disappearance of speakers: #687

- Reproduction and listening/hearing/feeling sensations to 16 Hz (organ) sound with my DSP-based multichannel multi-SP-driver multi-amplifier fully active stereo audio system having big-heavy active L&R sub-woofers: #782

Furthermore, if you would be curious, you can find here #931 the details of the latest setup of my multichannel multi-SP-driver multi-amplifier fully active audio system.

Edit:
If you would be seriously interested in using the test tone signal tracks I prepared and used in above measurements and tunings, and if you would be also interested in the contents and PDF booklet of "Sony Super Audio Check CD" (ref. #651, #750, #760), please simply PM me writing your wish.

Furthermore, many of the "music tracks" within my "Audio Reference/Sampler Music Playlist" consists of 60 tracks (summary ref. here as independent thread) would be very much suitable for your subjective listening confirmation after your objective measurements and tunings, I believe.
Again, if you would be seriously interested in using/listening such intact tracks of my "Audio Reference Playlist", please simply PM me writing your wish.

Definitely going to check all of this out, thank you!
My next step is going fully active

 

[mike7877]

Just one more thing I would add which is totally subjective:
For some systems I do prefer to use normal min-phase low-shelf filters to purposefully add that group delay - it gives that 'meaty' bass sound that sounds pleasing
(whereas I do not prefer that with some other systems)

I can see how this would be better with some systems or genres of music. To me, so far, it only sounds sloppier. 90% of the music I listen to is only with real instruments, though

 

[mike7877]

+30dB is 1000 times the power!!!!

You are clipping your amplifier and hearing bass harmonics/distortion. If your amplifier had enough power you'd burn-out the little woofer.

Every 3dB represents a doubling of power (approximately):
+3dB = X2 power
+6dB = X4
+9dB = X8
+12dB = X16
+15dB = X32
+18dB = X64
+21dB = X128
+24dB = X256
+27dB = X512
+30dB = X1024

...And because power is equal to the square of the voltage, +6dB represents a doubling of voltage. 30dB is sort-of a "magic number" where 30dB of gain is a voltage gain of about 30X.

But if the +30dB is way down at 20Hz, there's not much down there anyway. I've found, while giving a +12dB shelf centered on 56Hz (ie. 56Hz is the +6dB point), that real headroom is only reduced by 3-4dB with most music. If there's content down at 20-35Hz that's just as high as the bass around 80Hz, you'll lose all 12dB, but that's rarely the case. For example, Daft Punk Random Access Memories album (bass heavy, including low bass), with my +12dB shelf, reduces headroom by around 8dB, and that's about the worst case I've run into.

With Sokel's small speakers, if he had a 100W amplifier and he had +30dB down at 30Hz, and bass heavy content around 30Hz, it's extremely likely his woofers would be moving too far, causing harmonic distortion as the cone moved out of the linear range, before his amplifier clipped

 

[mike7877]

Buchardt A10 which I own are decent with bass, but every time you half the frequency, like 40hz vs 20hz, you need to move 4 times as much air. a 6.5 inch mid-bass can only do so much at 40hz and below. So the A10 is better for listening at 80db or so full range.

The KH310 (sealed 8" woofer) can probably output more than the A10. KLH model 5 is a sealed 10" woofer. I have a sealed 10" DIY 3-way and it can get to 30hz with good punch, but it uses big cabinets.

Dear SX as in Pioneer SX
Pioneer SX-950, my first receiver SX?

Just a random guess

 

[mike7877]

Quoting Linkwitz from that fine link, thx ...
"This allows to extend the response of a closed box woofer to lower frequencies, in the above circuit example from 55 Hz to 19 Hz, provided the driver has adequate volume displacement capability and power handling."

Adequate volume displacement capability has been the wall I've run into whenever applying a Linkwitz transform.
I've had to either put in a high-pass to protect from over excursion, or limit the sub's overall SPL by the amount of the LT boost.
The high-pass makes the sealed LT sub behave/sound more like a ported, ime.

Whereas with no high pass, I'm OK with overall limiting, other than needing to double up on subs to get back to starting SPL.

With no-high pass and no limiting ....sooner or later = bye bye sub. (for me, being a SPL and bass junkie)

What I do to offset too much cone movement is use a peaking filter at ~25-35Hz with a Q of around 1.5-2.5, bring it down 3-6dB, depending. Usually I find that when there's too much low frequency content down there, before cone movement becomes a problem, bass balance in each ear does. The lower it gets, the more likely it is that one ear will be receiving all the bass, it's even possible the bass will be out of phase, causing a dizzying effect. In rooms smaller than like 14x20 feet, I don't really like stuff much under 40Hz for this reason

 

[Petrushka]

I still have the Bose 901s I bought in 1969. I’m not using them at the moment, but they were the poster child for sealed speakers and equalization. The supposedly had a lot of distortion at 40 Hz, at least compared to AR3as.

I don’t have any reliable way of testing for room effects, but I find myself turning the bass down on ported speakers. The port resonance sounds really bad to me.

 

[mike7877]

As @ppataki says, that's not surprising. You changed the frequency response, therefore the Q and consequently group-delay of the system. They will inevitably be closer to a traditional ported system.

Given that frequency response is king when it comes to SQ, the best thing you can do is make sure you have a flat in-room response and have it extend as low as possible. That will deliver the best group delay, and shift the group delay bump as far down in frequency as possible. Usually, it's more efficient to do this with a ported system, and it elevates the need for a linear phase filter, which for bass, @ppataki already mentioned the drawbacks for.

But with a ported system, once you're below the port frequency it's like the speaker is in free air... I think, at least for my drivers (woofer Qts 0.85, fs 46Hz), sealed is the best option.

Shouldn't Q be not any higher than it was to start with after equalization with the same Q? I think the ringing from the Q might be more audible because the frequency is lower, making the time of ringing longer. I could be completely wrong though, idk, it could be compound

 

[mike7877]

I still have the Bose 901s I bought in 1969. I’m not using them at the moment, but they were the poster child for sealed speakers and equalization. The supposedly had a lot of distortion at 40 Hz, at least compared to AR3as.

I don’t have any reliable way of testing for room effects, but I find myself turning the bass down on ported speakers. The port resonance sounds really bad to me.

I remember reading about those! They had a bunch of drivers, right?
Have you restored them, or are they still holding up?

 

[mike7877]

12dB in the most power consuming point can be quite a task for an amp even at low levels,and that adds to the rest of the EQ applied.

Did you applied equal amounts of negative gain?How that affected gain structure,SNR,etc?Did you measure the chain before-after?

I do apply negative gain on the preamp, and I adjust according to the peak meter. A lot of music I listen to isn't normalized to -1 or 0dB (it's older) so a lot of times I don't need to reduce gain at all. When I do, it's usually 3-4dB. With modern, bass heavy music, it can be -8 to -10dB down.

Since usually I only have to adjust 0 to -4dB, I haven't done any measurements to see if SNR is affected much. My DAC is the D90 III, preamp A70 Pro, amps are two LA90 Discretes bridged, so if I were to measure, my interface wouldn't have the resolution. Subjectively, I think it's around -25dB (in the digital domain) where I can begin to tell the difference (with some content, exceptional content)

 

[Sylvanus]

For the best treatise of sealed-box woofer design and implementation, you need not look any further than Linkwitz' page on the Thor system:

THOR - subwoofer

Design of a sealed box subwoofer with electronic equalization

www.linkwitzlab.com

https://www.linkwitzlab.com/thor_splmax.htm (take note of the last statement on this page.)

It's important to remember that once you put Woofer A in Box B, many of the system parameters are set in stone.
You can't equalize your way into some magical transformation. The trade-offs are always present. Always.

 

[Petrushka]

I remember reading about those! They had a bunch of drivers, right?
Have you restored them, or are they still holding up?

Still work; pre-foam. One driver replaced, out of 18. I found two more sets at garage sales and became an unexpected collector.

I have two minidsps, pre-programmed, to replace the equalizers.

 

[DearSX]

Dear SX as in Pioneer SX
Pioneer SX-950, my first receiver SX?

Just a random guess

Good guess, I love old Pioneer and Sansui receivers. It's for my Nissan 240SX drift car and love of anime (DearS) and general love of audio sound.

 

[bmc0]

Regarding Linkwitz transform usage: One must keep in mind that it's just a special-purpose linear equalization filter. As such, it can only correct linear characteristics (i.e. frequency and phase [time] response). It cannot correct nonlinear characteristics, which will become a significant problem if you try to push things too far. As @Sylvanus said, it isn't magic. You can't reasonably expect good results trying to extend the LF response of a small speaker by 2.5 octaves.

In my system, I use Linkwitz transforms (implemented as digital biquad filters) in two places:

1. To correct and extend the response of my two 15-inch sealed subs. Unequalized, the Fs and Qts are 37Hz and 0.87 respectively, which I transform to 20Hz and 0.7. The in-room response rolls off at about 7Hz (not a typo) thanks to a convenient room mode. Distortion is respectable (<1% below 25Hz; <0.3% above 25Hz) at normal listening levels.
2. To achieve a (theoretically) exact LR4 highpass characteristic with my main speakers. In this case, I'm actually moving the corner frequency up (71Hz to 85Hz) rather than down.

 

[Sokel]

With Sokel's small speakers, if he had a 100W amplifier and he had +30dB down at 30Hz, and bass heavy content around 30Hz, it's extremely likely his woofers would be moving too far, causing harmonic distortion as the cone moved out of the linear range, before his amplifier clipped

???
What small speaker?

I do apply negative gain on the preamp, and I adjust according to the peak meter. A lot of music I listen to isn't normalized to -1 or 0dB (it's older) so a lot of times I don't need to reduce gain at all. When I do, it's usually 3-4dB. With modern, bass heavy music, it can be -8 to -10dB down.

Since usually I only have to adjust 0 to -4dB, I haven't done any measurements to see if SNR is affected much. My DAC is the D90 III, preamp A70 Pro, amps are two LA90 Discretes bridged, so if I were to measure, my interface wouldn't have the resolution. Subjectively, I think it's around -25dB (in the digital domain) where I can begin to tell the difference (with some content, exceptional content)

Yes,it's absolutely music dependable.
If for example you play the deadly LFE from Interstellar or more sanely Amir's reference Fading Sun demands can skyrocket.

 

[MAB]

Has anyone else extended the frequency response of their sealed speakers? How did you like the results?

It's done very often. Linkwitz has a page on it, aka Linkwitz transform:

Active Filters

It can be done with analog filter, like this:

Linkwitz Transform Subwoofer Equaliser

ESP Project Pages - Linkwitz Transform Subwoofer Equaliser. If you have tried to get a sub to sound right (and go low enough), you already know that it is not easy. This simple circuit makes a huge difference.

sound-au.com

Spreadsheets and Biquad calculators available Elliott Sound, as well as DSP implementations from MiniDSP, RePhase, etc...

Sealed + EQ is really a great way to go. I enjoy building cabinets without ports. My last project has almost exactly the same bass extension as the review of the same studio monitor:

I give up ~6dB of headroom, more on bass heavy tracks. I can't listen at max SPL these are capable of anyway, so it's a bit of a health benefit. I much prefer multiple sealed subs, and deal with the room's cancelations and peaks and phase issues. Ported speakers have other issues, like the port resonance. And if I need the headroom, or want to reduce cone motion, I tend to add extra woofers.

 

[bmc0]

Just in case it's useful to someone, here are discrete-time formulae for the Linkwitz transform (discretized using the bilinear transform):

If you're familiar with Robert Bristow-Johnson's Audio EQ Cookbook, you may notice the similarity with the high pass filter. The derivation from the given high pass coefficients is straightforward, but so far I haven't actually seen the equations written out anywhere.

 

[Waxx]

There are subwoofer drivers that can go very low without or with very minimal eq, but they tend to demand bigger cabinets than most want and are not so common as (drivers for) ported speakers. They also tend to be quiet more expensive and not so common.

I got a pair of Scanspeak Discovery 26W8534G00 in 77L sealed cabinets, and without any bass boost they get to 32Hz F6. The F3 of a sealed speaker does not say that much, as the natural cutt off slope of the box is very shallow (typical 6dB/octave in a 0.707 QTC config). In room this gives me without extreme boosting a response to below 30Hz if I want (i tune it with dsp to 30Hz F3 altough). The cavecat is that this driver only got limited xmax. But as it's a 10" it still moves enough air to got a very loud subbass in my small room. This is one of the cheaper drivers that can do this in good quality that i know about. There are others off course also but i got personal experience with this one in diy builds.

There are specialised subwoofers who can even do better than this, Dayton Audio and CSS make (or made, i'm not so up to date) drivers who could go lower in a much smaller cabinet. But those are strictly subwoofers (not useable above 100Hz) and big. In my diy build from above the driver needs to cover the 30-300Hz range, so those did not fit.

From a 4 or 5" you will never have solid bass on decent volume. Even 50Hz is often already a challenge for those. The cone surface to move air is just to small to do that. From a 6" it start to be big enough (with a lot of xmax) for decent bass, but real subs start for me at 8". That combined with enough xmax can go low enough on decent volume. And the bigger the driver, the easier to sub frequencies can be played in general. But you loose good top end in the frequency range of the driver as the driver becomes larger.

 

[Holmz]

…
Given that frequency response is king when it comes to SQ,..

For a subwoofer one my say the low group delay… is king.

… And if you must roll off, make it gentle. It’s way less audible and can make quite a difference.

The part we are in agreement on.