Speaker enclosure vibrations - a few measurements with accelerometer

[Tangband]

Here also some knock test audio recordings on different board combinations

DUO-DXT

hifi-selbstbau.de

Very interesting .

My own listening experiments with bitumen on the speakerwalls with two speakers ( jbl 530 and monitor audio rx6 ) is that the sound got worse with bitumen on all the walls, and better than without bitumen If I only used it on one of the opposite walls inside the speakers.
I have no measurements to show, but it seems that it ( might ) be better to spread the resonanses at different frequencys inside the speaker walls than lowering all the resonanses , damping all the walls.

It might also be good to use different thickness of the walls , ie the baffle maybe 2,2 cm and the backwall 1,6 cm for a smaller speaker ?
The structural resonances would then be different depending on the thickness of the wall. If this makes sense, it might be a good thing also to mix the material in the walls , maybe using plywood for some of the walls, and mdf for others ?

There are, in my opinion, rather big differences to be heard with music using bitumen , at least with the two speakers I tested it with.

 

[thewas]

I have no measurements to show, but it seems that it ( might ) be better to spread the resonanses at different frequencys inside the speaker walls than lowering all the resonanses , damping all the walls.

It might also be good to use different thickness of the walls , ie the baffle maybe 2,2 cm and the backwall 1,6 cm for a smaller speaker ?

Practically with the different size cut outs and mounted hardware on the front and rear baffle (drivers, ports, termines) the resonant frequencies will be different without using different thicknesses, if you are unlucky you could even get them closer again by using different thicknesses, so without computations or measurements such empiric approaches can be counterproductive. Also like in the examples above it is rather better to reduce them with a constrained layer sandwich and calculated bracing.

 

[Tangband]

Practically with the different size cut outs and mounted hardware on the front and rear baffle (drivers, ports, termines) the resonant frequencies will be different without using different thicknesses, if you are unlucky you could even get them closer again by using different thicknesses, so without computations or measurements such empiric approaches can be counterproductive. Also like in the examples above it is rather better to reduce them with a constrained layer sandwich and calculated bracing.

Yes, I agree that its probably only good luck to get the resonances were they disturb the music less , only by ear.

 

[Alexium]

Why would damped enclosure sound worse than undamped? Could it be that you simply like the coloration from those resonances, as long as it's mild? Perhaps, it complements the dips in the speaker's natural response?

 

[DanielT]

Why would damped enclosure sound worse than undamped? Could it be that you simply like the coloration from those resonances, as long as it's mild? Perhaps, it complements the dips in the speaker's natural response?

Try to overfill, really slam in, cram the speaker completely full of damping material, press in really hard. Then squeeze a little more stuffing/moff into the boxes. Screw in the speaker elements and listen. What do you think is going on with the speaker's sensitivity? Poof there it disappeared.

 

[Alexium]

Try to overfill, really slam in, cram the speaker completely full of damping material, press in really hard. Then squeeze a little more stuffing/moff into the boxes. Screw in the speaker elements and listen. What do you think is going on with the speaker's sensitivity? Poof there it disappeared.

Bitumen only damps the enclosure walls, not the interior volume of air, doesn't it? Shouldn't only the drivers be playing in a speaker, rather than the enclosure walls?

 

[DanielT]

Bitumen only damps the enclosure walls, not the interior volume of air, doesn't it? Shouldn't only the drivers be playing in a speaker, rather than the enclosure walls?

That's how it is. I showed the extreme case if you fill the entire box with some type of moff, filling material (really press in with a lot of force). I've tried it for fun, the results were exactly as you'd expect, not so fun.

Edit:
Incidentally, there is talk of forces on the baffle where the speaker elements are mounted, but the back wall of the speaker box should really be the one that receives a lot of force, or?

You can try building speakers from cast pipes in cardboard. It works, even though the tubes are thin, surprisingly well as tube subwoofers, see picture. Double 12 inch sub bass drivers mounted in the bottom. Maybe not super hifi (okay, I can always EQ it if I want), but with that bass support the floor gives damn what bass it will be. Party subwoofer, currently not in use, but if there is a party in the future, I will connect it.

No filling at all in that subwoofer.

Speaking of just that:

Filling/stuffing in Subwoofer?

Hi I am grateful if anyone could sort this out; Is there any point in having filling in a Subwoofer? The question based on the assumption of an otherwise well-constructed Subwoffer, with , for example,cross braces in the box and with steep LP filter, for example 24 dB, crossover frequency 80 Hz...

audiosciencereview.com

 

[witwald]

Yes as long as it is isotropic. Question is what happens when adding a driver with its own resonant behaviour. I can’t see that the combination is isotropic but rather anisotropic.

Many of the materials commonly used in loudspeaker cabinet panels are isotropic and have quite uniform properties throughout the panel. I can't think of how adding another mechanical system (e.g., a driver and its basket) and connecting it to a panel of the loudspeaker wouldn't make the overall system anisotropic. In general, it would still exhibit linear elastic behavior, unless of course some distinct nonlinearities were introduced by design.

The vibration modes of panels would of course be affected by the stiffness of the loudspeaker basket and the way in which it is attached to the panel. For example, a relatively rigid bolted/screwed assembly could be used, with adhesive bonding at the contact surfaces. Alternatively, the basket could be resiliently mounted in some manner, which would affect the modal frequencies as well as the transfer of vibrations to and from the panel. As we already know, the loudspeaker enclosure, stiffeners, damping methods, and driver basket will combine in a complex manner.

This complete vibratory system is far from simple, with undesirable resonances making a nuisance of themselves on many an occasion. Some of those could be avoided by a bit of careful finite element analysis early in the design stage, but that is no guarantee of success as the modeling process can require quite a few approximations along the way. Still, it's better than having to build a prototype and finding it has problems that might easily have been avoided with a bit of thoughtful modeling taking place.

 

[witwald]

What about the speaker cone tap test? What affects the sound you hear when gently tapping the center of a base or midbase driver?

When you tap the driver, one result is that you excite some of the resonance frequencies of the cone. The resultant vibrations produce some acoustic output, which you hear. More heavily damped cone materials will produce a duller sound, as the cone vibrations will decay more swiftly.

If the enclosure is a vented box, the movement of the cone will excite the port resonance at its tuning frequency. Higher-order pipe modes might also be excited, depending on the design of the port. That's probably the dull low-frequency sound that you hear upon tapping the cone.

I have a pair of floorstanders and a pair of more or less matching bookshelves. On the bookshelf the tapping sound is what I expect it to be, but on the floorstanders it's colored. Both 6.5", but the drivers are different models.

The floor standers may have a lower tuning frequency, hence the resultant sound could be a bit duller.

Is it affected by the speaker parameters, by the enclosure type (sealed, BR, TL etc.)?

All of those, probably. The sealed system will generally have a higher tone, as the driver in the enclosure is tuned to a higher frequency.

Is this "test" useful for anything?

I'm not sure. It's not really a measurement as such in a strict sense. There may be some loudspeakers that display abnormalities that could be picked up using such a simple test.

Others may be able to chime in with their experiences.

 

[witwald]

Incidentally, there is talk of forces on the baffle where the speaker elements are mounted, but the back wall of the speaker box should really be the one that receives a lot of force, or?

A force applied at a given location can lead to large vibration amplitudes at another location as a result of the modal response of the complete system being excited. Hence, resonances are best avoided. However, sometimes the best that can be done, in a cost-effective manner, is to dampen down the unwanted response or alter the frequency at which it occurs to make it less audible. In some cases, both approaches can work well (e.g., the BBC thin-walled loudspeaker cabinets with lossy structures).

 

[DanielT]

Speaking of dampening movements. One should not forget the amplifier's ability to dampen the movement of the speaker element. Here it is tested, 4:25 in the video, by setting a speaker element in motion without any damping, result:

6:20 in the video with an 8 Ohm resistor:

 

[witwald]

MDF, like most composite materials, is anisotropic. E1, E2 (elastic moduli parallel to the sheet, say, in x, y directions) are very different from E3 (perpendicular to the sheet, z direction).

https://www.researchgate.net/publication/285495335_Mechanical_properties_of_MDF_as_function_of_density_and_moisture_content

Thanks for that link. I didn't realize E3 was so different.

I still think that the resonance frequencies of the panels made from MDF material are going to be largely controlled by the E1 and E2 elastic moduli, due to their plate bending behavior, aren't they?

Also, glued joints behave very differently in compression (when being pressed together, compressing the MDF) than in tension (when being pulled apart, stretching the glue). In fact, screwed joints can be more linear if the screw preloads aren't exceeded, i.e. the joints are under compression at all time (may not be easy for MDF/wood as they tend to relax with time).

The compressive modulus (E3) is much smaller, so it would need to be taken into account in the design of the mounting system for the driver basket on the panel. In any case, the system is quite complex!

"Constrained layer damping" is also inherently nonlinear as it depends on the shear modulus of the viscoelastic layer being a strong nonlinear function of frequency (loss/damping is strongest in the transition region).

It depends a bit on the definition of nonlinearity. At a given frequency (and temperature), the shear modulus is constant. This allows a linear elastic analysis to be performed, one frequency at a time, of the constrained layer damping system. That's what the handbooks do, and finite element analysis uses a similar approach. One of the goals with constrained layer damping is to choose a viscoelastic material whose loss factor is close to its maximum value at the frequencies of interest. 3M and other manufacturers have tried to design such materials for aerospace applications, whose frequency range often covers that which is typically found in many loudspeaker panel resonances.

 

[Thomas_A]

Does anyone know iif there are other comparisons of two enclosures of the same size, one with constrained layer damping, and a loudspeaker driver (with BL in the 6-7 range)? Mechanical resonance and acoustic measurements included.

 

[fineMen]

MDF, like ...
To summarize, the structural behaviour of loudspeaker cabinets should not be assumed to be linear.

Any proof, I'm asking for any? Me thinks not.

Some reply to the link to so called knocking tests at "hifi selbstbau" was:

Very interesting . ...
My own listening experiments ...

Any correlation of knocking against a freely vibrating plate with the driving forces and subsequent motion when the ready-made enclosure is in actual operation? Me thinks not, just none.

So why is it, that You all refuse to investigate the made-up case any deeper than just citing irrelevant data from third parties, or reiterating wild speculations? And then You praise the "constraint layer" as a solution for a problem that isn't understood to the least; neither in regard to auditory relevance nor in regard to its possible origin.

This is exactly how the otherwise disregarded audiophile shmockery works. Invent an utterly irrelevant problem, propose a singular magical pseudo-science solution of Yours and make some money from missing education on the customer's side.

Congrats!

 

[Tangband]

Any proof, I'm asking for any? Me thinks not.

Some reply to the link to so called knocking tests at "hifi selbstbau" was:


Any correlation of knocking against a freely vibrating plate with the driving forces and subsequent motion when the ready-made enclosure is in actual operation? Me thinks not, just none.

So why is it, that You all refuse to investigate the made-up case any deeper than just citing irrelevant data from third parties, or reiterating wild speculations? And then You praise the "constraint layer" as a solution for a problem that isn't understood to the least; neither in regard to auditory relevance nor in regard to its possible origin.

This is exactly how the otherwise disregarded audiophile shmockery works. Invent an utterly irrelevant problem, propose a singular magical pseudo-science solution of Yours and make some money from missing education on the customer's side.

Congrats!

No , the reason why this is interesting to us former DIY:er , is that we KNOW that the sound from cabinet walls might color the sound , and in some cases, one can hear it.
Just look at these two pictures ( Elac B6 and below Genelec G3 ) cabinet resonanses measured from Stereophile, and read John Atkinsons comments :

”A prominent discontinuity in the traces just below 200Hz suggests that the enclosure has a serious resonance at that frequency. While setting up the speaker on my turntable stand to measure it, I noted that the cabinet seemed very lively—investigating the panels' vibrational behavior with a plastic-tape accelerometer, I found a very high-level, high-Q resonant mode at 188Hz that was present on all surfaces but highest in level on the side panels (fig.2). As HR wrote of the sound through the B6es of a recording of male voice and double bass, "I heard that vibrating box—not a little, but a lot. . . . [The speakers] were vibrating like sex toys!" There are also some high-level resonant modes present in the midrange.”

…and Genelec G3 :

”I investigated the enclosure's vibrational behavior with a plastic-tape accelerometer. It was extremely inert. The only resonant mode I found was on the sidewall, at 637Hz (fig.1), but this is vanishingly low in level, even at SPLs >90dB.”

————-

…So as everyone can see - this is in certain cases a problem for sound quality .

 

[fineMen]

No , the reason why this is interesting to us former DIY:er , is that we KNOW that the sound from cabinet walls might color the sound . ...

Third party impressions, third party data, no documentation, no investigative model, nothing to work with. Just there is something. Please DIY--the examination, and do it right, as I suggested.

 

[Tangband]

Third party impressions, third party data, no documentation, no investigative model, nothing to work with. Just there is something. Please DIY--the examination, and do it right, as I suggested.

But this is exactly what Thomas A has done - dont you think his investigations are interesting to discuss ?
And where are your DIY experiments ?

 

[Thomas_A]

Third party impressions, third party data, no documentation, no investigative model, nothing to work with. Just there is something. Please DIY--the examination, and do it right, as I suggested.

As mentioned, you and others are free to replicate the experiment and present the results. Then we can discuss.

 

[fineMen]

But this is exactly what Thomas A has done - dont you think his investigations are interesting to discuss ?
And where are your DIY experiments ?

Who am I to serve You righteously faithless fellow with precious data from my lab, eventually explaining and discussing everything for love? I'm only suggesting that You DIY the stuff at Yours.

 

[fineMen]

As mentioned, you and others are free to replicate the experiment and present the results. Then we can discuss.

Can't! As I personally took it, Your's holding an accelerometer against some cabinet walls doesn't provide enough clues to replicate the case. I told You and others already, what the problems are. I felt that my argumentation wasn't taken seriously. It was disregarded directly. So, it is up to You.