Reading some more, I am wondering how acoustic impedance and possibly other damping characteristics will vary with linear density of fibers and volume density of application. It seems that lots of lower priced products use fibers of larger diameter. Once I asked for a finer fibers, it had to be special ordered. Even on special order, the fiber diameters seemed much larger than the sample I provided which came from a pillow.
It could be.
But the only example of that are the measurements of R. A. Robinson in his Ph. D. Thesis (with Marshall Leach).
RAR uses the same technique as MJK, but with a TL filled of fiberglass (brrrrrr !).
RAR makes his own model (purely mechanical, translated into electrical: there you see the Leach "hand"). This model too is completely "out of the mainstream". It's made for fiberglass and it is hardly adaptable to other fibers.
It seems working better than standard, "rigid", models.
With those I get:
An externally hosted image should be here but it was not working when we last tested it.
(2.6 kg/m3)
An externally hosted image should be here but it was not working when we last tested it.
(5.2 kg/m3)
An externally hosted image should be here but it was not working when we last tested it.
(7.9 kg/m3)
"Rigid" models loose completely the resonance that is quite evident at 40 / 50 / 70 Hz, and the fit is not so satisfactory.
Are those measurements correct ? I don't know.
Is the fit "good enough" to predict the SPL (what we hear ...) from the line ? I don't know.
soongsc
BAF and AcoustaStuf are interesting material, but I don't have any measurement on then (apart the usual generic statements).
Again on fiberglass: how can I be sure that the fiberglass I buy in the store near me has the same fiber diameter as the one used by RAR ?
Dacron Hollofil II is quite easy to find in the States (Amazon.com: Dupont ® Dacron Hollofil II Standard Pillows (2 Standard Pillows): Kitchen & Dining). You pay (a lot !) for pillows that then you destroy, but, at the end, you have something whose characteristics are well known (and "good enough" for a TL).
BAF and AcoustaStuf are interesting material, but I don't have any measurement on then (apart the usual generic statements).
Again on fiberglass: how can I be sure that the fiberglass I buy in the store near me has the same fiber diameter as the one used by RAR ?
Last edited:
Heh, my enclosure supplier suggested I use US$3 pillows from which my original samples were from. 
My question was what kind of pain he would be willing to go through to measure within 10% for each speaker. But I will review the models to gain more understanding. Thank you!
I also noticed that the more course fibers were stiffer. Of course fiberglass is also quite stiff, but it does not seem to expand to fill, so this might also effect measurement data. In once set of speaker, I stuffed fibers in trying to distribute it as good as I could, but the sound was compressed. Then I just tore the fiber sheet into small pieces so that they could be spread out more evenly, this then made the sound more relaxed.
My question was what kind of pain he would be willing to go through to measure within 10% for each speaker. But I will review the models to gain more understanding. Thank you!I also noticed that the more course fibers were stiffer. Of course fiberglass is also quite stiff, but it does not seem to expand to fill, so this might also effect measurement data. In once set of speaker, I stuffed fibers in trying to distribute it as good as I could, but the sound was compressed. Then I just tore the fiber sheet into small pieces so that they could be spread out more evenly, this then made the sound more relaxed.
Last edited:
The introduction of the "Wilson model" page still says "Allard-Champoux model".
I am not sure whether I presented the above clearly or not.
By linear density, I mean the actual material density of the fiber, not the suffing density term normally used. By volume density, I mean how much physical volume is accupied by the fiber. When the fiber has low linear density, it will occupy more volume for the same stuffing density.
Another aspect that would be very important is the finess of fiber constructed of the same material. This will also effect the acoustic impedance at different frequencies.
All the models seem to address stuffing density.
The fiber models I'm using rely on the flow resistivity. There is (commonly) another model, the Johnson-Champoux-Allard that relies on more parameters which require specialized measurement tools.
The JCA model is (best) suitable when you have some specific material, of which you don't change the (apparent) density when you use it (look, e.g., at fiberform, just another PET composite, that come in well defined densities).
I don't use it because I want to do just the opposite: control the density.
The other model I'm using is Bies-Hansen that gives you the flow resistivity as a function of the fiber diameter and the (apparent) density.
This is why I don't use porosity, which in turn would require the bulk density of the material.
The JCA model is (best) suitable when you have some specific material, of which you don't change the (apparent) density when you use it (look, e.g., at fiberform, just another PET composite, that come in well defined densities).
I don't use it because I want to do just the opposite: control the density.
The other model I'm using is Bies-Hansen that gives you the flow resistivity as a function of the fiber diameter and the (apparent) density.
This is why I don't use porosity, which in turn would require the bulk density of the material.
I wish I had the math skills, but I can only rely on inuition and test articles. My gut feeling is that the fibers used should to have some desireable properties:
1. They need to be able to compress and expand well to allow better control of volume density. Volume density would be related with resistive nature of the stuffing.
2. The fibers should be as fine as possible. More fibers per cross section will create more resistance.
3. They need to have low linear density. This will make the fibers more easy to be moved by soundwaves/airflow such that irregular fiber movement can create additional resistance for less suffing density.
Do you have any links to information on the JCA model? I have just searched...
"the Johnson-Champoux-Allard model which supposes a motionless frame, they may yield wrong parameters when this condition is not satisfied."
This site is also interesting:
http://apmr.matelys.com/index.html
1. They need to be able to compress and expand well to allow better control of volume density. Volume density would be related with resistive nature of the stuffing.
2. The fibers should be as fine as possible. More fibers per cross section will create more resistance.
3. They need to have low linear density. This will make the fibers more easy to be moved by soundwaves/airflow such that irregular fiber movement can create additional resistance for less suffing density.
Do you have any links to information on the JCA model? I have just searched...
"the Johnson-Champoux-Allard model which supposes a motionless frame, they may yield wrong parameters when this condition is not satisfied."
This site is also interesting:
http://apmr.matelys.com/index.html
Last edited:
Hi teodorom,
*
I guess you’ve already read DR Putlands thesis?
http://grputland.com/files/thes.pdf
*
If not I,m looking forward for your eventual comment(s) on Putland's take on 'fiber absolute velocity' in : Chapter 7.1.1.
Do you agree with with Putlands approximation to neglect any:
...for...
b
*
I guess you’ve already read DR Putlands thesis?
http://grputland.com/files/thes.pdf
*
If not I,m looking forward for your eventual comment(s) on Putland's take on 'fiber absolute velocity' in : Chapter 7.1.1.
Do you agree with with Putlands approximation to neglect any:
...for...
?
b
At the bottom of the page is a history chart related with research of various material.
http://apmr.matelys.com/pages/AcousticalPorousMediaHistory.jpg
Continuing some search...
Hmm, I am wondering whether periodical "Applied Acoustics" is a good source of such information or not.
Such as http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V1S-3WCSYKP-2&_user=10&_coverDate=06%2F30%2F1996&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1444463620&_rerunOrigin=scholar.google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=85d6b10a3abbcfb5376990d68ee5d2de&searchtype=a
Last edited:
Thnks,
I alrady knew that chart. It shows clearly what I mean by "mainstream". You clearly see that the names known by "the people of loudspeaker builders" are not mentioned.
Secondo point: I'm full of papers that "could be interesting if they were free". Ooops ! I forgot to say: A FREE PAPER .
I alrady knew that chart. It shows clearly what I mean by "mainstream". You clearly see that the names known by "the people of loudspeaker builders" are not mentioned.
Secondo point: I'm full of papers that "could be interesting if they were free". Ooops ! I forgot to say: A FREE PAPER .
This is the paper I was mentioning: http://orbit.dtu.dk/getResource?recordId=173922&objectId=1&versionId=1
"For a wave with frequency 100 Hz in glass wool of density 30 kg/m3, the attenuation of a layer of thickness 0.20 m is 4 dB if the fibers move, and 12 dB if they do not move."
This density seems higher than what I have used in the past. Just a quick browse through the paper, it seems the material is placed against a wall rather than in a duct where the waves propogate through. I think this will be different because porous material work more effeciently where the velocity is highest, whereas when placed against the wall, the velocity is lowest.
This density seems higher than what I have used in the past. Just a quick browse through the paper, it seems the material is placed against a wall rather than in a duct where the waves propogate through. I think this will be different because porous material work more effeciently where the velocity is highest, whereas when placed against the wall, the velocity is lowest.
This is true, but the low frequency attenuation improves when most damping material is at a certain distance from the wall. Normally these are measured for wall treatment products. I guess I need to look into the paper a bit more to see if there is anything applicable to speakers.
Thank you for the information.
I've always wished there were some software that will allow users to define the suffing distribution and simulate it.
Thank you for the information.
I've always wished there were some software that will allow users to define the suffing distribution and simulate it.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.