Multiple Enclosure Loudspeaker Advantage, why does it matter

Multi Enclosure Speaker Design, Why does it matter?

May 2019

After an online conversation with a client, I realized that this subject deserved a page.

If you’re familiar with my website, you may have realized that I seldom do traditional box designs. The “all in one” box so familiar in HiFi circles, doesn’t appeal to me at all for a variety of reasons.

Yet, at the very top of the food chain, few “Stereophile Recommended” audiophile companies offer an “all in one box”. The appeal for simplicity and the associated “one size fits all” never gets old.

Drink the cool aid they say. Just look at the five figure Magico Q7, Vivid Giya G1 or the Sonus Faber Aida, Kef Muon. Those are flagship models, so their head designer, must know what they’re building. Or do they? Well, I agree to strongly disagree.

Magico Q7 Kef Muon Vivid Giya G1 Sonus Faber Aida

I’ll list my top three reasons for why I think it’s too compromised to be used by any serious builder.

First reason: The all-important bass, knowing that 99% of commercial speakers miserably fail here. No wonder it’s my top reason.

In the movie reality, THX research in acoustic perception found that below 80 Hz, average humans can’t locate a sound origin. Most mass marketed receivers use THX 80 Hz recommendation, and cut low-bass from the main speakers, sending everything below 80 Hz to the subwoofer channel.

-Dolby Pro-Logic and the later updated Atmos system also use 80 Hz as the subwoofer crossover point.

I personally tested myself, in an unscientific manner, and without too much effort, I located 80 Hz origin but not 70 Hz and below. And at ~70 Hz, I’m not really sure I can locate sound source with 100% accuracy. I just feel a sense of broad direction more than a pinpoint location, and my own tests weren’t double blind. I may have had a bias.

The point being, that at ~70 Hz and below, no audiophile can locate a sound origin. Knowing that, literally means that loudspeakers and subwoofers can be together in one box or separated in distinct enclosure, with no harm to sound perception. Speakers could be designed only to reproduce 80 Hz and up (or 70 Hz and up in my case) with the subwoofer picking up the rest.

And now, that we have separate subwoofer around, a new dilemma appears. Given that 80 Hz and below is essentially mono (remember, the average human brain can’t deduct subwoofer location below 80 Hz), it doesn’t matter if the signal is stereo or mono source. As far as the subwoofer frequency is concerned, the brain perceives it as mono. If one mono subwoofer would fit the need. How many should we use?

A single subwoofer can’t create a uniform sound pressure into a room. Room acoustics dominate and interfere too much, and at such low frequencies, response variations are often in the 30 dB range from one spot to another. Corners often reach even bigger peaks. DSP correction can lead to improvements at one location, but often at the expense of worse results in another location. All of the DSP wonders are overwhelmed in such cases.

Two Subwoofers (hopefully far apart from each other) lead to more uniform frequency across the room, but are still severely limited. (refer to my links section for article on multiple subwoofer benefit)

Four subwoofers (best location being at all mid-wall in a rectangular room) get much more acceptable results, which I consider the bare minimum for accurate sound performance. A sound system without four or more subwoofers, usually doesn’t get my interest.

Six and more subwoofers scattered inside the room, is really the best one can get. It evenly spreads the peaks and valleys, and counteracts room acoustics to great lengths. By using more subwoofers, room treatment requirement is diminished (in the hard to treat 10 Hz-80 Hz band). It also puts less strain on a single unit, as more units can work less together. As a group, they require much less DSP compensation to obtain fantastic results. More subwoofers REALLY is the way to go. It’s not about the ability to play stupid loud. Rather, multiple subwoofers are the only effective weapon to fight back nasty room acoustics we all encounter.

Back to our “one box for all”.

If one pair of speakers have to reproduce the whole frequency range, bass will come only from two origins: the left and right speakers. Those two source will be relatively close to each other, compared to one behind you for a separate subwoofer system. In order to achieve frequency balance and good soundstage, both left and right loudspeakers should be placed in a symmetric manner. Perfect symmetry is the very worst case to excite more room mode in bass frequency! It’s unavoidable. By focusing on midrange accuracy, depth and wide soundstage, one sacrifices the very low frequency in-room performance. Can’t beat it. You’re stuck with a compromised bass, and have to live with significant room mode.

This reality can’t be understated. Separate enclosures enable better subwoofer placement to work better in actual listening room. Multiple subwoofers have and always will, provide best result in real use (i.e.: in room).

Please Do not buy this----------->

Second reason: Cabinet Vibration Management.

Imagine one classical three-way speaker design: (see image of a Harbeth speaker above, very bad design BTW)

This one uses a 12” bass unit, a 5” mid-range unit and a 1” dome tweeter.

All those drivers, traditionally will be forward facing, sharing the same front baffle. (inducing time delay error too!)

Let’s now assume the speaker reproduces music and is playing @ 100 dB white noise.

The woofer, in order to reach 100 dB output @ 1 meter, will have to move ~5.7 millimeters to reproduce 40 Hz frequency. Only ~1.1mm to reproduce 90 Hz @ 1m. At 32 Hz, the woofer is required to move a healthy 8.9mm.

The vibration transmitted onto the front baffle is much more, if the speaker cone has to move 8.9mm compared to only 1.1mm (8 folds reduction). This vibration will travel through the front baffle and be transferred to the mid-range and the tweeter, because they are mounted on the same non infinitely rigid structure. The whole speaker enclosure vibrates. All loudspeakers do. Some more than others, but they always do.

The 1” tweeter, to reproduce the same theoretical 100 dB output @ 6khz at 1meter only has to move ~0.03mm (30microns). At 14 kHz, it only has to move ~0.005mm (5 microns). The reality is that often, the front baffle of a cabinet vibrates at a bigger amplitude than the tweeter diaphragm does. The tweeter then has no chance of being accurate, if it’s being mounted on a surface that vibrates more than its dome does.

If the enclosure vibrate more than 0.1mm (10 microns), the tweeter output is seriously compromised.

Of course, our 5” mid-range will add its grain of salt, and vibrate the enclosure a tad more. Whilst much less than the woofer, it will add some extra energy into the mix. Not ideal at all.

I repeat: This is really bad. A tweeter has no chance to be accurate if it’s being vibrated more than its diaphragm does. Of course, making heavier, better braced enclosures reduce this effect, but never eliminates it entirely. It only attenuates the problem. To completely get rid of this, the best solution is to break up the physical connection and build separate enclosures. By detaching the subwoofer, our main speaker will only reproduce 80 Hz and above. The vibration induced to the cabinet will be reduced, and as such is a tremendous advantage.

Aerial 20T loudspeaker and Verity Audio Sarastro, are both high end designs using a two chassis speaker system.

1 bottom cabinet for bass duty

1 layer of isolation platform

1 midrange + tweeter enclosure on top.

Aerial 20T

Verity Audio Sarastro

This concept yields solid improvement over the single enclosure above.

Bass speaker vibration is separated from the rest of the midrange loudspeaker enclosure, and they are mostly absorbed through the middle layer (mostly, not totally, as it never absorbs 100% of the energy). The mid-range vibration produced is much smaller and only induces a limited detrimental effect on the tweeter. While still an effect, it’s much less than a single enclosure attached to a bass driver. Huge progress can be achieved at the expense of additional carpentry work.

Back to our three-way example. Pushing the concept further and having three separate enclosures would be the best. The 12” enclosure won’t interact with the mid-range enclosure and won’t disturb the tweeter output. All drivers are then operating independently.

The Wilson Audio Alexia does exactly that!

One box for the dual bass driver, one box for the single mid-range driver and one box for the tweeter.

And they also apply vibration isolation mechanism between enclosures. Life is great.

Pushing the concept to its limit, the flagship Wilson Audio WAMM model use six separate enclosures!

One for the tweeter
Two for each high mid-range
Two for each low mid-range
One for the bass module (10” and 12” working in tandem as per usual Wilson Audio fashion)

Wilson Audio WAMM

Wilson Audio Alexia

All of those boxes vibrate independently, and thanks to a semi-rigid assembly method (decoupling), each enclosure shouldn’t interfere with each other too much. On that aspect, this is the way to go. This is a technical home run on vibration management.

Those Wilson Audio WAMM’s, as per my first reason for separate enclosures and multiple subwoofer advantages...

I have to be honest. It’s a 750,000USD epic fail. Rooms dominate and always will below 80 Hz. Wilson can't change that.

Third reason: Ease of Integration and Tweaking.

It’s no secret that building loudspeakers is time consuming. Doing a “piano finish” isn’t a walk in the park, and requires considerable time. There are $$ signs associated with that glossy finish.

If you build a classic three-way as per the Harbeth example, you have to build a certain outside dimension in order to get an internal air volume to work with the mid and bass drivers selected. The volume needed for both the mid-range and bass drivers is unique to that driver you carefully selected. Once you cut your speaker hole, you’re deeply committed. From one bass driver to another, the hole cut-out size for the basket will often be unique to that driver as well.

If you build superbly finished three-way loudspeakers, all Scan Speak based drivers and somehow succeeded in getting a good sound, you’re now stuck with it. Let say you used a 5” paper mid-range, and then suddenly, you read a fantastic review and now want to try a ceramic cone, like the Accuton 5.5” mid-range. The change will never be "bolt-on". It will require some woodworking, and if you need to cut a hole bigger for the new driver, reverting back to the previous one will be an impossible. A risky, permanent alteration required for a test, which may or may not work out.

For bigger bass drivers, the tuning frequency of a bass reflex cabinet will be screwed if you change a driver, as it’s required internal volume won’t match the loudspeaker you previously designed. Hoping for the best, is all you can do.

By building separate enclosures, changing a single driver to test a new product becomes a brisk task. Simply build a new test enclosure for the new driver and swap'em out. If you don’t like the result, put the old one back in. No harm, and no alteration of the original design and it’s ten times easier/faster to do. The new box will always be the perfect size, as it’s built for this specific driver and nothing else. None of the other boxes require alteration. If it’s a fail, to the fire pit it goes, with the rest of speaker remaining intact.

Furthermore, you can time align the driver only by moving the enclosure back and forth relative to each other, to obtain the precise required offset. A five-minute task at the most. With a solid single “one size fits all” enclosure, you just can’t, and are stuck with the original distance as building a shim won’t look right. The way it is, is the way it has to stay. The only way to keep time alignment intact with single enclosure, is to use active crossovers and DSP for time compensation, which is not something most DIYers do.

Proper time alignment is easy to detect in subjective testing. And once you listen to dialed-in, time aligned speaker, there’s no going back.

Multiple boxes give the builder that option without much drawback.

-And time alignment is possible without the use of DSP, with passive crossover!

Conclusion.

I think engineers that design an “all in one” speaker seen on popular magazines, understand what I wrote here. Its not exactly a secret to anyone who has a bit of experience in speaker design.

My theory is that they willingly compromise on sonic quality against aesthetic marketability of their product.

They deliberately choose traditional, greatly inferior designs that are aimed to visually please potential buyers and their “significant other” (used to be known as WAF : Wife Acceptance Factor).

Ultimately, they design something that look nice… just that.

Sales departments found out that the better their product looks, the more it sells, and they are in business to sell and make money. That is why they build products the way they do, with impeccable finishes, archaic designs, to sell them to the broader masses. The masses want shiny, so shiny they get.

“Sounding nice” is just a bonus, and most people never know what they’re missing anyway. Sounding “REAL” is something one has to build themselves, IMHO.

The industry isn’t providing it, and likely never will.

DIY’ers have no fear! As long as you know what you’re doing, you can humiliate most five figure or even six figure flagship speakers…at a fraction of the cost