When I built my sealed subwoofers, I filled them with polyfill, and it was quite difficult for me to calculate exactly the virtual volume increase, I should have asked you to do me a favor to calculate it. Maybe I'm still on time, because it would seem that you are an expert in cabinet measurements, what do you say ? 😉
Just run impedance checks while filling the enclosure with polyfill, You should see the impedance peak moving down and to the left.
(you see the virtual volume increase visually, while doing the impedance measurement check and filling some more between measurements)
If the peak starts moving to the right again, your fill is too dense and you'd lose (virtual) volume. 🙂
Not that hard to do with a simple impedance rig and REW... Why gamble if it can be this easy.
See an example here:
https://www.vandermill-audio.nl/?p=950
Different materials results in different damping characteristics though. As was said before, real wool does a very good job.
So does the itchy stuff I used in combination (worked better than one or the other).
You don't have to do it like that, but it doesn't hurt to try it once... 😉
(you see the virtual volume increase visually, while doing the impedance measurement check and filling some more between measurements)
If the peak starts moving to the right again, your fill is too dense and you'd lose (virtual) volume. 🙂
Not that hard to do with a simple impedance rig and REW... Why gamble if it can be this easy.
See an example here:
https://www.vandermill-audio.nl/?p=950
Different materials results in different damping characteristics though. As was said before, real wool does a very good job.
So does the itchy stuff I used in combination (worked better than one or the other).
You don't have to do it like that, but it doesn't hurt to try it once... 😉
Last edited:
I'm just a hobbyist and I don't want to spend money on gear I won't be using anymore... thanks anyway ! 😉
I will tell you that I am also very satisfied with the sound obtained, but I will read the article, it seems interesting. Although I would be more interested in learning to calibrate REW in my living room, I have resisted it, despite having DSP......
REW is free, and the resistor I used for my impedance rig was a free gift too, from the electronics shop around the corner.
The wires used were just old jack plugs I had laying around. I use an old Windows Vista Laptop nobody here used anymore...
Total investment: some time to put it all together, really, there isn't much to it.
I'm a hobbyist too, no more, no less... but I like to learn and to experiment. The site I linked is mine, done for fun and certainly incomplete.
The wires used were just old jack plugs I had laying around. I use an old Windows Vista Laptop nobody here used anymore...
Total investment: some time to put it all together, really, there isn't much to it.
I'm a hobbyist too, no more, no less... but I like to learn and to experiment. The site I linked is mine, done for fun and certainly incomplete.
I perfectly understand what you are saying, but there is no longer so much enthusiasm on my part to embark on new inponderables, I would not want you to think that I am lazy, only that I think it is not worth the loss of time for me (at this moment and stage of my life, I hope you understand, if not, just wait till you're 72 and have a spine like an F1 circuit!)
But I can assure you that I always carefully read your contributions here.
But I can assure you that I always carefully read your contributions here.
Thank you, everyone, for sharing your experience 👍 An OB design is not something I’m considering atm, my question was intended to help me understand why I’m getting a boxy/cupped colouration from my implementation when the colouration isn’t present when listening to to the unit in free air.
Reading your replies it’s clear that I may need to pay more attention to reducing the back wave generated by the drive unit from re-emerging through the cone and/or more effective box stuffing.
One interesting point of discussion was about the audibility of the cones back wave. In order to test this I propose to place a portable radio inside the enclosure and listen for any sound escaping. Hopefully this’ll give some idea of its audibility and the effect of enclosure stuffing.
I like the idea of measuring the speakers impedance whilst adjusting the fill density. For those who have tried this technique, what effect does tuning for minimum impedance have on the sound of the speaker?
Reading your replies it’s clear that I may need to pay more attention to reducing the back wave generated by the drive unit from re-emerging through the cone and/or more effective box stuffing.
One interesting point of discussion was about the audibility of the cones back wave. In order to test this I propose to place a portable radio inside the enclosure and listen for any sound escaping. Hopefully this’ll give some idea of its audibility and the effect of enclosure stuffing.
I like the idea of measuring the speakers impedance whilst adjusting the fill density. For those who have tried this technique, what effect does tuning for minimum impedance have on the sound of the speaker?
The stuffing (up to a point) makes the box look bigger. That means a lower frequency resonance with higher damping.
Is there a difference in performance between stuffing to make the box look bigger and building a bigger box?
A closed box design often relies on the stuffing to moderate the cabinet volume to produce a specific acoustic response. The cabinet volume may effectively be increased by some 5-10 %.
Building too big a closed box can be detrimental to the bass performance.
Building too big a closed box can be detrimental to the bass performance.
Again, would increasing the enclosure volume achieve the same thing?
In what way would it be detrimental? My understanding is that increasing the volume changes the bass slope which reduces bass at say 80hz, whilst also improving the transient response. Am I right in thinking that some people would feel that less bass is detrimental and some people would feel that an improved transient response would be desirable?
Stuffing allows micro-adjustment of acoustic response while choosing enclosure volume is a macro design decision made to suit the parameters of the bass driver.
Too large an enclosure can result in a weak bass response due to overdamping i.e. too much reduction of the system Q.
Increasing the enclosure volume reduces the enclosure resonant frequency, but also reduces the system Q.
Low Q gives better transient response, but lacks the fuller bass response higher up that some listeners prefer.
Attachments
Most designers aim for a closed box system Q that is between 0.7, the value that gives the flattest response curve, and about 1.
Some rock fans like a higher Q because it accentuates the bass, even going as high as 2.
So, for a 'full' bass go over Q = 1 and for 'ultra-clean' bass stay under Q = 1.
Some rock fans like a higher Q because it accentuates the bass, even going as high as 2.
So, for a 'full' bass go over Q = 1 and for 'ultra-clean' bass stay under Q = 1.
I’m still wondering if increasing the apparent volume by 10% is the equivalent of increasing the actual volume by 10%. I can imagine that increasing the actual volume may not be quite as good due to increased panel radiating area and maybe additional mass, but does increasing the stuffing have its downsides too?
Also, I’m still wondering about how tuning for minimum impedance affects sound quality?
Also, I’m still wondering about how tuning for minimum impedance affects sound quality?
The answer is Yes - it is. And getting the correct density of stuffing gives you about 20% increase in effective volume, apparently.
Since the mechanism is down to heat transfer between the volume of air in the enclosure and the stuffing, the thermal properties and surface area of the fibres will have some effect too...
(Didn't one of the major speaker manufacturers fairly recently develop a special stuffing material to optimise this, or was I dreaming?)
There's some good, practical info in this recently resurrected thread here (which you've probably already spotted):
https://www.diyaudio.com/community/...bsorption-material.340878/page-2#post-7028463
https://www.diyaudio.com/community/...bsorption-material.340878/page-2#post-7028463
When the air in a sealed enclosure is compressed, its pressure increases. At the same time, the mechanical work done in compressing the air is transformed into heat energy which causes the temperature of the air to increase.
If the heat is allowed to escape then the process is said to be isothermal, meaning the temperature remains the same.
However, rapidly changing the pressure at audio frequencies in a sealed loudspeaker enclosure gives the heat no time to escape. Adiabatic is the name used when heat does not leave a system, so unstuffed sealed cabinets are classed as adiabatic.
Filling a sealed enclosure with absorbent material allows heat to transfer from the air to the absorbent. This converts the loudspeaker enclosure to isothermal action by stabilising the temperature of the air.
The advantage of isothermal action compared to adiabatic action is that the compression of the air is more linear, which, in a loudspeaker context, means less distortion.
The air in a sealed enclosure obeys the Gas Laws. Only in an isothermal setting does the product of pressure and volume remain constant, allowing the compression and expansion of the air to remain linear. In an adiabatic setting, the compression and expansion is non-linear.
The air in the enclosure is a good insulator, but the absorbent is a less good insulator, allowing heat to conduct into the absorbent.
An interesting experiment may be to incorporate fine copper strands into the absorbent material!
Why address that to me? I said that earlier in the thread, and it's something I've been saying for years.
By the way, as a spring, the volume of air in a closed box is very non-linear. In an isothermal system PV is constant. So P is proportional to 1/V - describing a hyperbola rather than a straight line. Distortion is intrinsic, even if stuffing is used to make the behaviour approximately isothermal (adiabatic - unstuffed - would be slightly worse).
Distortion due to the above would be minimised by minimising the changes in volume caused by driver excursion by making the enclosure larger (for a given SPL). But, since the ear isn't sensitive to distortion at low frequencies, I doubt it's a problem anyway. And even if the same driver handles higher frequencies, the cone will be mass-controlled above resonance, so largely unaffected by compliance non-linearity.
Last edited: