"Monkey Coffins" (box speakers) and Resonances...

[dlaloum]

So, I became an audiophile in my late teens, and at 20 was working in a "HiFi" store (also the first home theatre store in our region!) ... circa 1985.

This retailer sold speakers that have affected my audio journey ever since...

We had Quad ESL-63 electrostatics, Revox Agora B active, Boston Acoustics (A40 through to A400), also Pioneer, Marantz, and Magnat.

I also spent some time getting to know our competitors, visiting them chatting with them listening to their speakers....

The speakers I had a huge preference for, were the Electrostatics ... Quad ESL-63, and its competitor down the road... Martin Logan CLS, I also loved the Boston A400's. - The Revox Agora B was nice.... but the others were better.... the Pioneer, Marantz speakers all were middle of the road "boxes", and I disliked the Magnats, finding their Metal dome tweeters harsh.

I heard many speakers then, and in subsequent years (when my career took me into IT rather than audio) continued listening to a wide range of speakers....

Typically I can tell a "Monkey Coffin" (AKA Box Speaker) - almost instantly - there is a distinctive "sound" to them - I believe this is caused by resonance of the speaker walls (some peope have reported a common resonance around 1.5khz for many speakers.... this may be what I have been noticing).

This was (and is!) totally absent on Electrostatics, and other panel speakers - I have also found it to be absent on "Orb" speakers - and some other "funky shape" designs... including my current Gallo Nucleus Reference 3.2's.

Another observation from long term listening, is that smaller bookshelf speakers, tend to suffer markedly less from this "box" sound than do large floorstanders... perhaps a simple property of the larger surface area.

Also of note, that certain floorstanders don't seem to have this identifying "box" sound - the Boston A400's didn't have it - they were a very wide, but relatively thin rectangular box....

(also noteable for having relatively small 8" woofers but two of them - they had great bass!)

I remember the annoyed look on a high end audio retailers face, when after less than 60 seconds listening to a very expensive set of speakers - I said "they sound like boxes" - In the next room he had a pair of Magnepan 3.7's set up.... listened to those for 60 seconds..... ahhhh that's more like it... settled in for a bit of a listening session.... I don't recall the "boxes" but they were priced up around $10k at the time.

This would have been in the late 90's - when I spent 4 years in NYC...

All of which is to open up the topic - what is it exactly that makes a box sound "boxy"

And what is it about those (rare) "boxes" that don't have that distinctive sound?

During my skint student years - I had small bookshelves... a set of AudioPro's and a set of Tandy/Radio Shack - cheap and cheerful - no serious vices, and no "boxy" sound! - Possibly because small boxes are more rigid, or their smaller radiating surface, makes the resonance less obvious?

That might lead to - how does one eliminate the boxy sound if a speaker has it?

(or is it so fundamental to the design, that it is best to move to a different speaker... as I did)

 

[jeffbook]

Any Siegfried Linkwitz design eliminates the boxy sound of Monkey Coffins.

 

[dlaloum]

Not really, those speakers are another boxless design!

 

[jeffbook]

Not really, those speakers are another boxless design!

Well, that's my point. Only with massive internal bracing that raises the natural frequency of flat surfaces that make up enclosed box speakers can the boxy sound begin to be tamed. Small enclosures can more easily tame the boxy sound as the flat panel natural frequency is raised simply be design for those enclosures, but this comes with the disadvantage of reduced lower end response. Just an engineering design trade off. Before you begin to discuss ported speakers, these come with other trade offs that impact sound quality.

 

[dlaloum]

Yep, it is worth noting that my speaker and subwoofer preferences have always been towards sealed designs... (or open baffle) - possibly because none of the ported sub's I tried ( or the ported speakers) could match the nimbleness in the bass of the sealed designs and the electrostatics... ported has always seemed to me to tend towards more rather than better. (hence my response to a search for "more" would be to add more subs of the nimbler sealed variety)

But yes, my observations are that smaller boxes suffer less (or not at all) from the issue - it is usually the larger floorstanders where it becomes an issue.

 

[dlaloum]

Interesting... - But most of my audiophile life, I have had dipoles, (Quad Electrostatics) - so there was an even stronger backwards radiation, bouncing back of the wall and reflecting into the room....

So it wouldn't be the reflected sound from the back....

The Gallo's I use now, are Stainless Steel spheres stuffed with a specialised absorbent material.... - they don't have panels to radiate... but they also have an absorbent interior like your pillow example.

It is interesting that the concrete box didn't work - but if all it was, was 6 concrete sides, without bracing - what you have done is only increase the mass of the panels.... so increased inertia - but once they start to move/radiate - they will keep doing so for longer.... it might exacerbate the problem rather than solve it.

Possibly a layered design, where the box walls are made like an isolating platform, with layers of damping materials as well as layers of different density materials... so as to reflect vibrations back into the damping layer - and a sealed, "suspension" design box, with further sound absorbing materials on the inside....

Does that sound like a familiar set of design principles to anyone?

(and yeah - this whole thread could be boring... for those not interested in the topic!)

 

[howard416]

Are your experiences with monkey coffins, having given you the impression of "boxiness", coming from sighted or double-blind testing?

 

[Frgirard]

Are your experiences with monkey coffins, having given you the impression of "boxiness", coming from sighted or double-blind testing?

Double test bind with speakers, How you proceed and can we see your installation?

 

[dlaloum]

Are your experiences with monkey coffins, having given you the impression of "boxiness", coming from sighted or double-blind testing?

Sighted.

 

[MarkLawrence]

I've had magnepans for 43 years. I also owned Dave hermeyers original direct drive Electrostats and 100 pound 1200 volt tube amp. (Dave taught Roger Sanders and wrote the original audio amateur article). I agree completely on monkey coffins. And I agree it's the box. I'm not a big fan of Walsh, but they don't do this thing . And the original amts mostly didn't. The amt driver was mounted open air. Also, somehow the Linn isobarics were ok.

 

[dlaloum]

I've experimented with resonance control on my TT stand - including measurements, experiments with different materials, working on spring maths etc.... - resonance is a sneaky / tricky foe - and some of the solutions can be counter-intuitive... and solutions that work for one frequency may well do nothing for another or (much worse) create a problem in another frequency range...

But yes lead sheets, sandloading, all of these can be used... also differing materials, as sound, like light can reflect and refract... so if you know the frequency you are focusing on, layered materials can be used to refract the soundwaves (change their frequency) - reflect them back (so they radiate in rather than out - and back through absorbent materials) - and use absorbing materials focusing on the specific frequency... which effectively turn the resonance into heat.

 

[dfuller]

This is sort of a quandary of speakers with reflex designs. Absorbent materials inside help with reducing internal reflections, but reduce the effectiveness of the port or passive radiator - they depend on the rear side of the speaker cone to work! Then there's the pipe resonances of the port itself - imagine blowing across the top of a beer bottle. A lot of time and effort has gone into suppressing those resonances as much as possible because they can make pretty awful peaks in the midrange to lower presence region.

Sealed alignments on the other hand are almost always stuffed full of some kind of damping material because they don't need the in-cabinet resonance to do anything, they just need an air spring - I took the back panel off one of my KH310s a while back (I was trying to diagnose an intermittent buzz so I was checking the power supply caps) and it's stuffed full of the same kind of stuffing you'd find in a pillow.

Of course you need to make the cabinet as rigid and nonresonant as possible, too - but absorptive material to reduce the internal reflections is very important.

I think that what he was hearing was the reflected sound wave of the upper bass and lower midrange reflecting off the interior walls and exiting through the cone material. Although it would be reduced in level, it would still be a significant secondary (delayed) signal. I could be wrong, though.

Could be, yeah.

 

[dlaloum]

This is sort of a quandary of speakers with reflex designs. Absorbent materials inside help with reducing internal reflections, but reduce the effectiveness of the port or passive radiator - they depend on the rear side of the speaker cone to work! Then there's the pipe resonances of the port itself - imagine blowing across the top of a beer bottle. A lot of time and effort has gone into suppressing those resonances as much as possible because they can make pretty awful peaks in the midrange to lower presence region.

Sealed alignments on the other hand are almost always stuffed full of some kind of damping material because they don't need the in-cabinet resonance to do anything, they just need an air spring - I took the back panel off one of my KH310s a while back (I was trying to diagnose an intermittent buzz so I was checking the power supply caps) and it's stuffed full of the same kind of stuffing you'd find in a pillow.

Of course you need to make the cabinet as rigid and nonresonant as possible, too - but absorptive material to reduce the internal reflections is very important.

Could be, yeah.

Explains my long term dislike (maybe too strong.... perhaps "wariness" ) for ported designs...

 

[polmuaddib]

Sighted.

Can you rule out visual bias?
The more speaker looks like a box, sounds like a box?
Because that's how your OP reads if you take that aspect into account.
Mind you, I am not claiming anything.

 

[dlaloum]

Can you rule out visual bias?
The more speaker looks like a box, sounds like a box?
Because that's how your OP reads if you take that aspect into account.
Mind you, I am not claiming anything.

Except I have heard quite a few boxes, that did not sound that way... the A400 remains prominent in my memory, but the Revox Agora B also did not have the issue.
I have heard excellent box speakers at audio shows ... but heard even more that did have what I characterise as the " box " sound.

So I have good reason to believe that it exists and is no figment of my imagination.

I have never identified that sound from a panel or open baffle speaker, nor from orb designs.

It seems to be a flaw inherent in the six sided box shape, but which can be overcome with good design / engineering, and which is far more often present in ported designs. ( perhaps the required mitigations are more difficult to implement in a ported design?)

 

[mcdn]

The Boston A400 and the Revox Agora B are both wide baffle loudspeakers. They'll have very well controlled (i.e. relatively narrow) directivity down to a much lower frequency than typical narrower box loudspeakers. The more exotic "no box" speakers also have atypical directivity in their various ways. I'd suggest this is a more likely source of the sound signature you associate with "boxes" than any panel resonances. Panel resonances would show up as distortion.

 

[dlaloum]

I think the trade off's related to using ported speakers, come from two things:

1) Increasing the efficiency (SPL's) of the speaker - this was particularly important in the days when the average home amp would have between 15W and 50W.... and a 50W amp was considered "powerful" - so designers sought to eke out every bit of SPL from their designs.
This particular trade off has decreased in importance, as many mass market amps today sit in the 75W to 100W range - power is more readily available, so trade offs for higher SPL/wm are less valuable!

2) Low end extension

This primarily is about floor standers... ie: big speakers, not "bookshelves"
Floor standing speakers have always tried to be "full range" - and therefore to approach 20Hz as their lower end limit
Again this comes down to a tradeoff - does the designed fit a Subwoofer into the design? Limit the bottom end to 25Hz or 35Hz ? or use woofers with porting to extend their bottom end as far as possible... more economical.... - also low end can chew up a lot of amp power - so the power equation SPL/wm appears again - getting serious low end output without porting, can easily require 150W++ amplifiers (not uncommon for subs today to have 300W amps)

 

[dlaloum]

The Boston A400 and the Revox Agora B are both wide baffle loudspeakers. They'll have very well controlled (i.e. relatively narrow) directivity down to a much lower frequency than typical narrower box loudspeakers. The more exotic "no box" speakers also have atypical directivity in their various ways. I'd suggest this is a more likely source of the sound signature you associate with "boxes" than any panel resonances. Panel resonances would show up as distortion.

I would agree with your characterisation of the A400's as a wide baffle design...

But would you really consider the Revox Agora B design as wide baffle... - It's horizontal proportions are square like most speakers. (and they are a ported design)

They aren't particularly tall either... at 72cm - my recollection of them was circa thigh height

They were very heavy - very solidly built, and with substantial on board amps - a fully active speaker in the early 80's...

 

[mcdn]

But would you really consider the Revox Agora B design as wide baffle... - It's horizontal proportions are square like most speakers. (and they are a ported design)

42cm wide apparently - that is much wider than most mainstream speakers. The Grimm LS1 is 52cm wide and that's a design explicitly based around using a wide baffle for directivity control.

 

[suttondesign]

Linkwitz.