stone and concrete ring like a brass bell. Ever liked the sound you get from drilling in concrete?
But its perfect material for constrained layer type of box.
This guy makes wonders with stone material. http://www.faktiskt.se/modules.php?name=Forums&file=viewtopic&t=41863
He uses two kinds of stone wich he glues together with a special damping glue called DG-A2
Akustik ljudisolering bullerdämpning ljuddämpning vibrationsisolering
But its perfect material for constrained layer type of box.
This guy makes wonders with stone material. http://www.faktiskt.se/modules.php?name=Forums&file=viewtopic&t=41863
He uses two kinds of stone wich he glues together with a special damping glue called DG-A2
Akustik ljudisolering bullerdämpning ljuddämpning vibrationsisolering
There still seems to be confusion about structural resonances versus internal reflections and resonances. These are 2 different issues and both should be effectively dealt with.
The speed of sound in a material and its inherent damping mean that each cabinet wall will have various resonance frequencies. At those frequencies a cabinet wall will ring on and be highly transparent. Internal spl will get out. The BBC Harwood paper (linked to by the German site) shows that he developed a criterion of keeping all cabinet contributions 20 to 30 dB below the woofer output for inaudibility.
Note that the 30 dB threshold was at about 500 Hz. There are cases where you might have resonances at lower frequencies (where you can tolerate a little more) and, through increasing stiffness. push the typical resonances up to 500 where they would be more audible, not less! Stiffness alone, or mass alone, aren't enough. You need to consider an effective way to add damping.
Now, internal to the cabinet the hard surfaces will reflect sound. The reflected sound will come out the woofer cone, the port, or possibly through the cabinet walls. The dimensions mean that there will be standing wave frequencies for all 3 dimensions of a typical 6 sided box. These are not structural resonances but are dimension related and can be treated by acoustical absorbtion (i.e. fiberglass or rockwool). It works out that a thickness of 1/4 to 1/6th each dimension (of fiberglass) will effectively damp the lowest standing wave of each dimension.
Comparing the 2 issues to your listening rom, the first is what sound gets through the walls to the neighbors, the second is what are the room's internal acoustics like?
David S.
p.s. the Harwood paper tells you everything you need to know on the subject.
http://www.bbc.co.uk/rd/pubs/reports/1977-03.pdf
The speed of sound in a material and its inherent damping mean that each cabinet wall will have various resonance frequencies. At those frequencies a cabinet wall will ring on and be highly transparent. Internal spl will get out. The BBC Harwood paper (linked to by the German site) shows that he developed a criterion of keeping all cabinet contributions 20 to 30 dB below the woofer output for inaudibility.
Note that the 30 dB threshold was at about 500 Hz. There are cases where you might have resonances at lower frequencies (where you can tolerate a little more) and, through increasing stiffness. push the typical resonances up to 500 where they would be more audible, not less! Stiffness alone, or mass alone, aren't enough. You need to consider an effective way to add damping.
Now, internal to the cabinet the hard surfaces will reflect sound. The reflected sound will come out the woofer cone, the port, or possibly through the cabinet walls. The dimensions mean that there will be standing wave frequencies for all 3 dimensions of a typical 6 sided box. These are not structural resonances but are dimension related and can be treated by acoustical absorbtion (i.e. fiberglass or rockwool). It works out that a thickness of 1/4 to 1/6th each dimension (of fiberglass) will effectively damp the lowest standing wave of each dimension.
Comparing the 2 issues to your listening rom, the first is what sound gets through the walls to the neighbors, the second is what are the room's internal acoustics like?
David S.
p.s. the Harwood paper tells you everything you need to know on the subject.
http://www.bbc.co.uk/rd/pubs/reports/1977-03.pdf
That's an informative post, though to save all this understanding malarkey you could just build open baffles and have very few issues with cabinet colouration. You'd also hear much less of the room.
Back to the boxes and internal resonances. I found through experimentation that carpet underlay of the rubbery foam variety was most effective at dealing with the internal reflections without introducing colouration. It's surprising given it's not absorptive. In the same small ported enclosures, no damping caused terrible audible resonances and polyester wadding caused an unacceptable loss of natural dynamics. This in a 2-way speaker. With that underlay the bass and mids were both superb, and I still love those speakers when I hear them (at a friend's house). Oh I also capped it off with some teased out lamb's wool, which did improve matters without the side-effects of polyester.
Back to the boxes and internal resonances. I found through experimentation that carpet underlay of the rubbery foam variety was most effective at dealing with the internal reflections without introducing colouration. It's surprising given it's not absorptive. In the same small ported enclosures, no damping caused terrible audible resonances and polyester wadding caused an unacceptable loss of natural dynamics. This in a 2-way speaker. With that underlay the bass and mids were both superb, and I still love those speakers when I hear them (at a friend's house). Oh I also capped it off with some teased out lamb's wool, which did improve matters without the side-effects of polyester.
I'm about to get my hands dirty with carbon fibre/kevlar altho I'm sure it must vibrate at some point.
By curving the cabinet design as much as possible adds plenty of strength and minimises vibration - to me this is just as important as the material, I can't see the point of aerospace aluminium if you keep the panels flat, even with bracing.
By curving the cabinet design as much as possible adds plenty of strength and minimises vibration - to me this is just as important as the material, I can't see the point of aerospace aluminium if you keep the panels flat, even with bracing.
for 2 reasons:
To help attenuate reflection, and to damp the very resonant stone 'ring'. I figured to dissimilar materials would work the best, a la ally perf sheet and glass fibre, or polyfoam and lead loaded rubber.
To achieve very good downward dynamic range (DDR) you need better then 20-30 dB.
To get there i have looked at the problem in another way. For a structural resonance to get excited it has to be feed sufficient energy within its bandwidth to get excited. This graph shows the relative level of <proportional to the inverse of the square of frequency> and <... of the 4th power...>
More explanation: http://www.diyaudio.com/forums/cons...erials-build-speakers-out-77.html#post2562940
Somewhere between these curves * some konstant, is the likelihood of exciting a resonance when the stimulus is music.
If a resonance is not excited it is as if it does not exist.
Pursuing builds based on this has produced boxes with VERY low resonant signatures (and they can be lifted without breaking yoir back).
Relative to the recent discussion of granite, it also shows why, thou it may ring like a bell when externally stimulated with a broad band signal, with music as the stimulus, they could be made quite non-resonant.
dave
I use memory foam for this purpose. Very wideband absorber. (See the pic above)
yes open cell polyurthane foam seems to be a very good damping material. in the bass its still good to use ordinary white "fluff". It has very good isothermic prorperties even though its guite useless for damping (wich can be good in the bass where you dont have any resonances in the working band you need to dampen, and can benefit from the less restrictive material).
The ordinary glassfibre or stone fibre wool is very good at damping (some of the best sound damping properties) but have some disadvantages.
Sheeps wool has similar damping but is "healthier" if you like.
Foam has an additional property that can be good in that its easier to get it to stay staionary. The movement of the damping in the box can actually affect the driver and cause delayed motion (shown as group delay). This is tricky though since you get the best absorption in the middle of the box.
For instance, the designer of the stone speakers above uses a specially made foam in his speakers, shaped blocks that fit snugly between the walls.
But Im sure all this have been covered on this forum before?
All absorption coeffecients I have seen tell me that foam is pretty much useless. While Fibreglass being one of the better absorption materials (depending on its density).
Not following you. You are talking about the probability of music exciting a resonance but not explaining how you get your music to miss the resonances. Even if we assume western music sticks pretty well to a 12 tone scale we still have a lot of noise in musical instruments and also plenty of impulsive content to spread the spectrum around. No real chance to "hide the resonances between the notes".
David S.
I think he's saying that although the resonant frequencies will be played in most recordings, they won't have enough energy to excite the resonances. And the graph shows us that the probability of those resonances being excited above ~600hz is quite low. So if you construct your box to resist resonance of frequencies below 600hz (ie panels no larger than 340/600 m), you'll be relatively safe.
Do I understand you correctly Dave?
I think I agree and that is the most practical way of constructing boxes. I had that somewhat in mind for my next project. However, how are you sure that the energy must be "high" to excite a resonance. This is where I struggle to understand your position. Also that, your graph is formed from what evidence? Experimentation I suppose?
Do I understand you correctly Dave?
I think I agree and that is the most practical way of constructing boxes. I had that somewhat in mind for my next project. However, how are you sure that the energy must be "high" to excite a resonance. This is where I struggle to understand your position. Also that, your graph is formed from what evidence? Experimentation I suppose?
Thanx, yes, that should answer speakerdave's question.
You need to feed energy into a resonance to get it to move.
Striking a tuning fork feeds a wideband impulse of energy into the fork. The energy within the bandwidth of its tuning excites its resonance and the rest goes away. If you feed the tuning fork with energy not within its bandwidth it does not resonate.
The 1/f^2 curve (red) comes from the physics. Somewhere i have the page dogeared in one of my text books. The extra 2 powers are unproven but not unwarranted suppositions.
Actual test data that i have accumulated so far limited to actual builds based on this approach and listening to the box walls with a mechanics stethescope (and very useful tool)
dave
I don't try to miss them, i just use "brute" force to push the resonances up to the point where the music does not have enuff energy to excite them.
dave
The topic has been about damping for a speaker box (eliminating structural vibration). Absorption of air borne sound (inside the box) is something different. A number of folks are not understanding this distinction (damping vs absorption).
Take a look at the Harwood paper. Based on his listening tests there was a constant threshold of audibility for resonances that occured above 500 Hz. Resonances below that could be stronger for equal audibility. Also, the energy level of resonances he measured on a viriety of cabinets was fairly flat from 100Hz to 2000Hz (with some scaling due to cabinet size). Probability of excitation isn't falling as your graph suggests. Both his graphs and his comments show that raising cabinet resonances by making walls stiffer can make resonances more audible.
Hi Q resonances are interesting. Tests show that with LF standing waves in a room, once excitation near but not on a resonance is turned off, the tone shifts to the nearby peak and continues.
David S.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.