I am trying to get enclosure damping straight, particularly fiber stuffing vs lining the internal walls with mat/foam sheets.
Here's what I think that I know so far:
Fiber stuffing performs the function of reflection damping and/or increasing the apparent size of the enclosure by up to 20%. This most often how sealed enclosures are treated.
Lining the internal walls with foam or mat fibers dampens internal reflections. This is the way many/most ported enclosures are treated.
Do I have this right? When, if ever would you use both?
Thanks,
George
Here's what I think that I know so far:
Fiber stuffing performs the function of reflection damping and/or increasing the apparent size of the enclosure by up to 20%. This most often how sealed enclosures are treated.
Lining the internal walls with foam or mat fibers dampens internal reflections. This is the way many/most ported enclosures are treated.
Do I have this right? When, if ever would you use both?
Thanks,
George
Hi George, have a read.
Volume filling a reflex box
All you need is stuff the cabinet to the extent when you don't see resonances anymore. You don't need lining at all. In the vented cabinets, you usually leave some space between the woofer and a vent so the air can freely move.
Volume filling a reflex box
All you need is stuff the cabinet to the extent when you don't see resonances anymore. You don't need lining at all. In the vented cabinets, you usually leave some space between the woofer and a vent so the air can freely move.
I'd certainly use both in a closed box, and even a ported enclosure is allowed to have a degree of stuffing (keep the port clear though).
Use something like thick felt for the internal walls, where pressures are high and velocities low.
Towards the centre of the enclosure, pressures are low and velocities are high, so a relatively lower density of fibrous material is appropriate. To absorb sound energy, the idea is for stuffing to convert air movement to heat, so a fibre which exhibits a high degree of friction is best. In practice, fibreglass seems to work well, and polyester not very well at all (in my limited experience).
Use something like thick felt for the internal walls, where pressures are high and velocities low.
Towards the centre of the enclosure, pressures are low and velocities are high, so a relatively lower density of fibrous material is appropriate. To absorb sound energy, the idea is for stuffing to convert air movement to heat, so a fibre which exhibits a high degree of friction is best. In practice, fibreglass seems to work well, and polyester not very well at all (in my limited experience).
George,
I tested a few different materials for use in the cabinet of my sealed line array. I have posted pictures and impedances measurements to show the effects of the different materials. You might find it helpful.
http://www.diyaudio.com/forums/full-range/303417-full-range-tc9-line-array-cnc-cabinet-20.html#post5053920 From Post 195 onwards.
All the other information you have been given is good, if you want to damp or absorb low frequencies then you want fibreglass and felt. The acrylic stuffing does not do much at all unless you use large quantities of it as volume fill to try and increase the apparent volume of the enclosure.
The thing is fibreglass and felt do the same thing and they do it much better and with less of it.
Fibreglass can be an issue in vented cabinets as some fibreglass sheds and it can escape through the port and out into your room.
Vented cabinets really only need resonance damping and not volume fill in most cases.
I tested a few different materials for use in the cabinet of my sealed line array. I have posted pictures and impedances measurements to show the effects of the different materials. You might find it helpful.
http://www.diyaudio.com/forums/full-range/303417-full-range-tc9-line-array-cnc-cabinet-20.html#post5053920 From Post 195 onwards.
All the other information you have been given is good, if you want to damp or absorb low frequencies then you want fibreglass and felt. The acrylic stuffing does not do much at all unless you use large quantities of it as volume fill to try and increase the apparent volume of the enclosure.
The thing is fibreglass and felt do the same thing and they do it much better and with less of it.
Fibreglass can be an issue in vented cabinets as some fibreglass sheds and it can escape through the port and out into your room.
Vented cabinets really only need resonance damping and not volume fill in most cases.
So then damping material can have up to three purposes?:
1) Reduce internal reflections. This can be lining the walls with a continuous material and/or a loose fill. I should think that this is what will "clean up the midrange".
2) Increasing the apparent volume of the cabinet in a sealed enclosure. This is with loose fill. This changes the bass tuning of the system.
3) Further reduce cabinet box/wall resonance by adding mass to the walls with a stiffer/heavier material. This reduces output peaks at the resonant frequency of the box.
Do I have this right?
In the case of 3), is this any better than thicker or more rigid box material and/or more internal bracing?
Thanks,
George
1) Reduce internal reflections. This can be lining the walls with a continuous material and/or a loose fill. I should think that this is what will "clean up the midrange".
2) Increasing the apparent volume of the cabinet in a sealed enclosure. This is with loose fill. This changes the bass tuning of the system.
3) Further reduce cabinet box/wall resonance by adding mass to the walls with a stiffer/heavier material. This reduces output peaks at the resonant frequency of the box.
Do I have this right?
In the case of 3), is this any better than thicker or more rigid box material and/or more internal bracing?
Thanks,
George
Just to sum up, if you want to increase effective box volume, then filling with almost anything will add up to 20%, converting low frequency compression/expansion from adiabatic to isothermal (look those terms up).
To suppress internal acoustic energy at middle frequencies, use dense material on the internal walls and less dense towards the middle of the enclosure - and use something that actually works.
To reduce enclosure wall vibration, just make it very rigid (using bracing as necessary) and heavy. You can use constrained layer damping in larger boxes if they're thin-walled, but there'll still be significant output at low frequencies.
To suppress internal acoustic energy at middle frequencies, use dense material on the internal walls and less dense towards the middle of the enclosure - and use something that actually works.
To reduce enclosure wall vibration, just make it very rigid (using bracing as necessary) and heavy. You can use constrained layer damping in larger boxes if they're thin-walled, but there'll still be significant output at low frequencies.
So enclosure resonance is best taken care of when building the enclosure (thick/dense wall material, good bracing) and if done properly will not require any damping material for the purpose of taming resonance?
If you do not need to increase the apparent volume of the enclosure then how do you determine the mix of dense on the walls vs looser in the middle - same question for both vented and sealed?
Thanks,
George
complicted subject - even hard to find agreement on basic principles
few have the education and expeirence to properly weight the effects, proposed 'solutions' - many go to extremes in (different) directions that aren't justified by the numbers, even fewer by actual measurements
walls vibrating at their mass/stiffness panel resonance(s) is a different issue than acoustic resonance inside the box where the sound wavelengths small integer rational fractions fit inner box dimensions
how to treat, how to know when you've achieved a good result is difficult with box construction
Geddes (posts here as gedlee) uses urethane modeling board - neither dense nor stiff in comparison to often used 'better' cabinet materials, and uses damping bracing for his OS waveguide family of 2 way speakers with ~ 1 kHz XO
I think constrained layer damping of walls might be useful too but Geddes seems happy with his approach
Constrained layer damping could be applied to many other box construction materials like M/HDF, Baltic Birch... but again it needs some engineering chops to design, test
for acoustic resonances inside the box any fill or unconstrained soft 'damping' material will be mostly useless at frequencies where the sound wavelength is less than 1/2 wave in the damping material
and there should be very little sound to absorb/damp above the XO frequency
few have the education and expeirence to properly weight the effects, proposed 'solutions' - many go to extremes in (different) directions that aren't justified by the numbers, even fewer by actual measurements
walls vibrating at their mass/stiffness panel resonance(s) is a different issue than acoustic resonance inside the box where the sound wavelengths small integer rational fractions fit inner box dimensions
how to treat, how to know when you've achieved a good result is difficult with box construction
Geddes (posts here as gedlee) uses urethane modeling board - neither dense nor stiff in comparison to often used 'better' cabinet materials, and uses damping bracing for his OS waveguide family of 2 way speakers with ~ 1 kHz XO
I think constrained layer damping of walls might be useful too but Geddes seems happy with his approach
Constrained layer damping could be applied to many other box construction materials like M/HDF, Baltic Birch... but again it needs some engineering chops to design, test
for acoustic resonances inside the box any fill or unconstrained soft 'damping' material will be mostly useless at frequencies where the sound wavelength is less than 1/2 wave in the damping material
and there should be very little sound to absorb/damp above the XO frequency
Enclosure construction considerations
The previous two post are excellent, in that, they raise valid considerations.
What are the sources of box colourations ?
Vibrations from the woofer to the enclosure walls; It has already been established that since the box walls consist of a much greater amount of possible acoustical radiation as compared to the driver unit, that these vibrations shall be cancelled. Newton's law of physics is called upon here. This would be applied to the very lowest of frequencies, say only the first few octaves, because, by the time we get up past the 80-100Hz range, the required cone excursions diminish to the point where baffle board damping can accomplish what we want, but then we must consider the resonances in the box itself, due to the standing waves. Addressing this issue can be accomplished in two ways; the use of adequate acoustical damping AND reducing the possibility of standing waves in the first place, by the use of non parallel walls. It is obviously a wise choice to have two, separate and devoted sections. One to handle deep bass and then again one to handle the lower mid-range/mid range section. This is because, in my opinion, there is exists no such [simple]cure-all for handling both together. If design constraints are such that you are FORCED to do so, then I would follow the likes of some of the finest British builds (B&W), Magico, or Von Schweikert.
The previous two post are excellent, in that, they raise valid considerations.
What are the sources of box colourations ?
Vibrations from the woofer to the enclosure walls; It has already been established that since the box walls consist of a much greater amount of possible acoustical radiation as compared to the driver unit, that these vibrations shall be cancelled. Newton's law of physics is called upon here. This would be applied to the very lowest of frequencies, say only the first few octaves, because, by the time we get up past the 80-100Hz range, the required cone excursions diminish to the point where baffle board damping can accomplish what we want, but then we must consider the resonances in the box itself, due to the standing waves. Addressing this issue can be accomplished in two ways; the use of adequate acoustical damping AND reducing the possibility of standing waves in the first place, by the use of non parallel walls. It is obviously a wise choice to have two, separate and devoted sections. One to handle deep bass and then again one to handle the lower mid-range/mid range section. This is because, in my opinion, there is exists no such [simple]cure-all for handling both together. If design constraints are such that you are FORCED to do so, then I would follow the likes of some of the finest British builds (B&W), Magico, or Von Schweikert.
Roger White's paper on volume filling is technically very accurate, but is applicable to one class of filling material. Roger used the equations and fiber parameters from Les Bradbury (Reference 1). Les Bradbury is a well respected Fluid Dynamist who applied his knowledge to a speaker topic on the paper. You can see in Roger's graphs that the sound attenuation of fiber filler is a strong function of frequency and that Fiber glass is only half as effective as the (long fiber) wool. When you apply the Bradbury equations to polyester, you will find polyester fiber is more less useless in sound attenuation at lower frequencies.
Also note that the Les Bradbury formulation is specifically geared toward the Arthur Bailey transmission line speakers with an uniform filling distribution. You can get Les Bradbury's paper from the Audio Engineering Society for $25.
In the formulation of similar cabinet by Martin King and George Augspurger, they applied the electrical filter theory which does not account for the frequency dependency of the fiber filling. They treat the speaker as a resonator and came to a completely different conclusion.
The fiber filler in a completely closed cabinet is operating on different principle and may be mathematically treated as a thermodynamics problem. You can go to Ken Kantor Aural Fixation website for a good collection of references.
Excellent point.
Last edited:
No measurements.
I've found that once a panel is glued up (both sides), it can be stuck on anything and with anything (as a separate panel) with the result being that it is completely dead.
Harder to come to definite conclusions when the box is made up, but it doesn't seem to react any differently.
BBC only mass damps one side; I think it really needs both sides to be treated.
I've found that once a panel is glued up (both sides), it can be stuck on anything and with anything (as a separate panel) with the result being that it is completely dead.
Harder to come to definite conclusions when the box is made up, but it doesn't seem to react any differently.
BBC only mass damps one side; I think it really needs both sides to be treated.
It seems like the discussion getting a bit beyond the original intent. I think we might want to separate out these different types of phenomenal into something more basic.
1. Filling a seal box to increase volumn - no really a problem but more of a effect.
2. Vibration/resonant caused by enclosure walls.
3. Internal air resonant caused by enclosure geometry - a more difficult to solve.
#1 is something that is not really a problem but more of a implementation to your advantage.
#2 the best way is using strong material and using well braced cabinets. There is really no magic bullet.
#3 is somewhat tricky. There are two or maybe three ways to solve this problem. Resonant can make its way to our ears in two ways - by excite the enclosure walls, and by reflecting back to the drivers which then move the external air. You can build a strong cabinet, but there is nothing you can do about the drivers.
1. Cabinet geometry: most resonant occurs at certain peak frequencies of certain integer/or integers. Human ears can be very sensitive to those peaks. One way is to "smear" out or distribute those peaks more even across multiple more continuous frequencies. Cabinet geometries can influence how the peaks are distributed. The worse type of enclosures are the long rectangular where one dimension just dominates the others, and to make thing worse, it happens to be the easiest to build so most speakers are shaped this way. The square towers are usually the worse. Since it's the easiest to build, most DIY projects end up this way. A lot of high end speakers have irregular shapes that are curved, or tapering that looks like a triangle, or something fancy that hopefully will break up the resonant peaks into smaller more continuous peaks. But I don't think they completely eliminate peaks just maybe reduce them to a certain extend.
2. The other technique I've seen is similar to what "Humble HiFi". He built a regular cabinet, but internally, he placed braces with a certain geometries that supposedly break up the internal air resonant. I don't know how effective this is but it makes some sense.
3. The only technique that I think is almost effective is transmission type of enclosure. I've read an article from Vandersteen and he said his enclosure even for the mid range uses transmission type. You can use a real transmission type or a pseudo one. For the bass you probably need a real one since it has a lot of energy. For the mid range, Vandersteen said he makes his mid range tapering at the end that will eventually absorb the energy before it bounces back to the driver. Interestingly transmission line is not that difficult to build, but it is bulky so that is the trade off.
1. Filling a seal box to increase volumn - no really a problem but more of a effect.
2. Vibration/resonant caused by enclosure walls.
3. Internal air resonant caused by enclosure geometry - a more difficult to solve.
#1 is something that is not really a problem but more of a implementation to your advantage.
#2 the best way is using strong material and using well braced cabinets. There is really no magic bullet.
#3 is somewhat tricky. There are two or maybe three ways to solve this problem. Resonant can make its way to our ears in two ways - by excite the enclosure walls, and by reflecting back to the drivers which then move the external air. You can build a strong cabinet, but there is nothing you can do about the drivers.
1. Cabinet geometry: most resonant occurs at certain peak frequencies of certain integer/or integers. Human ears can be very sensitive to those peaks. One way is to "smear" out or distribute those peaks more even across multiple more continuous frequencies. Cabinet geometries can influence how the peaks are distributed. The worse type of enclosures are the long rectangular where one dimension just dominates the others, and to make thing worse, it happens to be the easiest to build so most speakers are shaped this way. The square towers are usually the worse. Since it's the easiest to build, most DIY projects end up this way. A lot of high end speakers have irregular shapes that are curved, or tapering that looks like a triangle, or something fancy that hopefully will break up the resonant peaks into smaller more continuous peaks. But I don't think they completely eliminate peaks just maybe reduce them to a certain extend.
2. The other technique I've seen is similar to what "Humble HiFi". He built a regular cabinet, but internally, he placed braces with a certain geometries that supposedly break up the internal air resonant. I don't know how effective this is but it makes some sense.
3. The only technique that I think is almost effective is transmission type of enclosure. I've read an article from Vandersteen and he said his enclosure even for the mid range uses transmission type. You can use a real transmission type or a pseudo one. For the bass you probably need a real one since it has a lot of energy. For the mid range, Vandersteen said he makes his mid range tapering at the end that will eventually absorb the energy before it bounces back to the driver. Interestingly transmission line is not that difficult to build, but it is bulky so that is the trade off.
Is the best way to find cabinet resonance to sweep the cabinet with a sine wave and either listen or measure and graph?
Same question for internal stuffing to "tame" mid frequency reflections.
Since most of us don't have an anechoic chamber is the best way to listen/measure outside?
Thanks,
George
Same question for internal stuffing to "tame" mid frequency reflections.
Since most of us don't have an anechoic chamber is the best way to listen/measure outside?
Thanks,
George