Subwoofer Isolation

[MartinC]

mosfet - Just to add with reference to your swimming pool example that in underwater acoustics (sonar etc) reflections from the surface water/air interface are strong and very obvious. I frequently make measurements in the MHz range and see this no problem. I don't believe there is a significant frequency dependence on the reflection coefficient but maybe I ought to look into this...

 

[mosfet]

Well granite/air would be bigger...

And granite / vacuum larger still. I was trying to think of an example that was illustrative so it could be visualised.

Good example. Surely you'd admit the sound level is reduced though, which is all I was suggesting above.

Yes certainly Martin. However I suspect this reduction is primarily due to absorption and less so a result of reflection.

The significant mismatch is between the granite and carpet under it, or the wooden floor.

I'd be cautious of this ââ'¬â€œ I think we've established the significant mismatches occur between solid/gas , solid/liquid , gas/liquid etc rather than solid/solid. Which you acknowledge in your following point:

Yes you'd get a bigger mismatch going granite/air, so an air platform as I use specifically for this reason under my speakers takes things a step further.

The air platform is a good choice Martin ââ'¬â€œ I think air suspension has much to offer. However the interface is still solid/solid (unless of course the platform is literally air in which case I'd be very interested!). The principle at work here is to lower the resonant frequency of the structure, ideally to just a few Hz, rather than shift it higher by adding mass.

I believe Symposium Acoustics are quite keen on the acoustic mismatch idea. Although I don't give much credence to the efficacy of this, their products do look well thought out if you want to have a look.

http://www.symposiumusa.com/main.shtml

 

[MartinC]

Yes certainly Martin. However I suspect this reduction is primarily due to absorption and less so a result of reflection.

I need to think about your specific example, but it absolutely isn't an absorption issue. Sound absorption in water, especially at audio frequencies and over the distance scale your talking about, is completely insignificant.

I'd be cautious of this ââ'¬â€œ I think we've established the significant mismatches occur between solid/gas , solid/liquid , gas/liquid etc rather than solid/solid. Which you acknowledge in your following point:

I need more time to look at this properly (I'm intrigued now) but doing some rough sums the energy reflection coefficients for the following interfaces are about right:

Air/Water ~ 1
Granite/Wood ~0.5

The mismatch for the granite/wood interface is not insignificant I believe.

Martin

 

[Matt F]

You could always place the sub atop several levels of Mana.

:mana:

Matt.

 

[mosfet]

I need to think about your specific example, but it absolutely isn't an absorption issue. Sound absorption in water, especially at audio frequencies and over the distance scale your talking about, is completely insignificant.

Perhaps another visualised example.

Returning to the lake on a particularly hot day. Let's say I'm skinny dipping with a nubile young female ââ'¬â€œ that's not too outrageous of course! And I decide to do a bit of ahem.. diving . The ahem.. sounds I would hear from the surface above would be dependent on frequency or wavelength; given low frequencies travel through water more easily than high frequencies. The question is, is this decided by reflection at the surface or absorption thereafter. On reflection (no pun intended) I'd say it probably is more likely the former.

How this might relate to a sub on a granite platform in considering what proportion of energy is reflected by virtue of the interface, given the above example, I'm now a little less focused to say!

Either way, if isolation is the objective, then a lossy material that causes the energy to take a convoluted path (like open cell acoustic foam) and hence dissipate through heat losses will be more effective than the isolation offered through an 'acoustic impedance mismatch'.

 

[MartinC]

Returning to the lake on a particularly hot day. Let's say I'm skinny dipping with a nubile young female ââ'¬â€œ that's not too outrageous of course! And I decide to do a bit of ahem.. diving . The ahem.. sounds I would hear from the surface above would be dependent on frequency or wavelength; given low frequencies travel through water more easily than high frequencies. The question is, is this decided by reflection at the surface or absorption thereafter. On reflection (no pun intended) I'd say it probably is more likely the former.

I'm going to have to come back to this when I've had time to check this out properly, specifically any frequency dependent issues, but I will.

My current best guess as to still hearing things in a swimming pool is that we're just underestimating quite what the loss is due to the non-linear way we percieve loudness. A 12dB loss would correspond to 94% of the energy being reflected for instance. The calculation I did above corresponds to a 40dB attenuation.

Either way, if isolation is the objective, then a lossy material that causes the energy to take a convoluted path (like open cell acoustic foam) and hence dissipate through heat losses will be more effective than the isolation offered through an 'acoustic impedance mismatch'.

Well I think you're right this would be better, and the effect of using such foam or a rubber based material would be a combination of absorption and the reflection issue I've raised. The impedence mismatch going to such a material will be significantly greater than that of granite/wood discussed above. I was discussing the granite/floor case since that was what was being proposed in this thread, presumably as most audiophiles are wedded to the stable speaker idea, and for this case the impedence mismatch is significant as losses will be minimal.

 

[mosfet]

My current best guess as to still hearing things in a swimming pool is that we're just underestimating quite what the loss is due to the non-linear way we percieve loudness. A 12dB loss would correspond to 94% of the energy being reflected for instance. The calculation I did above corresponds to a 40dB attenuation.

If you're looking to try to figure out an accurate objective model for the above Martin (as opposed to the more simplified model I gave) then indeed there are other factors that would need to be accounted for. The freq response of human hearing (as you say) , the salinity of the water, ambient temperature and quite probably half a dozen or more other factors I haven't mentioned. But don't let this stop you from trying!

ELF and VLF surface to submarine transmissions may yield some information.

Well I think you're right this would be better, and the effect of using such foam or a rubber based material would be a combination of absorption and the reflection issue I've raised.

Agreed. But by what percentage do you suppose the effect of isolation is attributable to reflection where the support is lossy ââ'¬â€œ say neoprene or acoustic foam? I'd say (as a guess) no more than a few percent.

 

[MartinC]

If you're looking to try to figure out an accurate objective model for the above Martin (as opposed to the more simplified model I gave) then indeed there are other factors that would need to be accounted for. The freq response of human hearing (as you say) , the salinity of the water, ambient temperature and quite probably half a dozen or more other factors I haven't mentioned. But don't let this stop you from trying!

I was just thinking of the basic issue of transmission and any frequency dependence of that. The other factors you mention are I suspect of relatively little importance, and are certainly not worth bothering about given this was just an example.

Agreed. But by what percentage do you suppose the effect of isolation is attributable to reflection where the support is lossy ââ'¬â€œ say neoprene or acoustic foam? I'd say (as a guess) no more than a few percent.

The honest answer is I've no idea! I've never looked into energy dissipation in such aborbers.

 

[maddog 2]

I use two large limestone tiles from Fired Earth, bonded together, sitting on three alu cones under my sub.

Seems to work well on my floating oak floor

 

[Cloth-Ears]

Dare I say the likes of this back in the days of LP12/Naim/Mana mania and all things spiked but, I'm not convinced rigidity and spiking is always best. I'm not saying its not, just that it could be seen as an "unchallengable orthodoxy" and you should be prepared to try other solutions. Oh and by the way, at the same time as all those spikey things you had to use, there was Townsend's Seismic Sink(of cash) , somewhat diametrically opposed ! As some have pointed out, what WORSE thing you could do to isolate a turnatable than to couple it to a moving floor by rigid little spikes? Duh-oh ! Yet, all those racks, and the likes of Mana..... I can't tell you what's right. Its an industry remember and its full of lots of little naked emporers. Right, subwoofer isolation... It is said that you dont want the box moving, only the drive unit, ok yes thats logical. But, do you also want the floor vibrating with it, really, wouldn't you want the air moving, driving the room more purely. How about an opposite approach, borrowed from the music industry. Consider soft isolation to produce a better tone of bass and, very importantly, to limit the effect of vibrations on your playing equipment ( if your floor is bouncing along to the explosions don't expect a fancy rack to isolate your DVD player and amp !!!) . There's a product called Gramma by Auralex. They are used under subwoofers a lot in the U.S and for guitar amps and such like in live music. The makers claim significantly better results with it. Whether that's true or not the Grammas make a huge difference to directly transmitted sound going through the floor and along to your walls to your neighbours below or to the side. I use Grammas under my main speakers for that reason alone and they do make a huge difference there. Lets be realistic, if you have neighbours and you are into AV there is a clash, particularly if you desire a subwoofer. Perhaps the use of Grammas could enable use of a sub when otherwise you might baulk at it. Sonically, I reckon they work. I've tried both. Also, if your floor is not quite even they solve that problem too. No longer speakers to sit on 3 spikes, never quite being able to get the adjustment right. The Grammas consist of specialised thick wedges of foam with a covered layer of wood on top. www.auralex.com/c_sound_isolation/c_sound_isolation.asp , I believe they could also be used for great big valve power amps, main floorstanding speakers etc ( as I do, I use spike plinths under my speakers and they sit firmly on the even surface of the Grammas ) . They are cheaper to buy mail order from the States than over here. Try E-bay.

 

[Cloth-Ears]

OOPS, I see Matt F above has posted an image of the Grammas already !. Oh well, you still have my windbag post anyway.

 

[7_V]

It's already been pointed out by Cloth-Ears and others that you can't possibly come up with an optimum solution to the speaker isolation/coupling question without paying regard to the room itself, particularly the type of floor (concrete or suspended wooden).

I would also add that the design of the speaker, its bass response and cabinet vibration is also a factor, as is the method of supporting the rest of the equipment.