I think that I said that I had no intention of searching it out.
I also have a poor knowledge of the mating habits of North African Elephants, but, somehow I'm not all that concerned about it.
The "bulk" properties of fibers is sufficient for my uses since I only use closed boxes. The details of the fiber density is simply not that big a thing for me. I did a lot of work on this kind of thing decades ago, so I thought that I could help, but these days there are other more presssing problems to deal with.
I am not quite so steeped in the theoretical aspects of fibrous stuffing as I am in the practical applications thereof and thus am not qualified to add to the theoretical discussion.
Closed box stuffing is not rocket science. I think Mr. Gedlee has pointed this out using different words. The key objective is to make the box appear larger in volume than it actually is and thus optimize (minimize box Fc) to its lowest possible level and thus increase bass extension.
When stuffing closed boxes there is a nexis where the benefits of adding more stuffing reverses it's downward trend. In the case of the AR3a about 20 oz. of FG is optimum yielding an Fc of 42 hz on avg. OTOH, I have found (emperically) that the optimum level of PET results in a few Hz higher Fc AT A LOWER STUFFING DENSITY on the order of 16-18 oz. I believe this is simply due to the inherent differences in the density of PET vs FG (i.e. it takes more PET fibers to = the same number of FG fibers at a fixed weight of each). However, this will depend on the denier of PET being considered. 1.5 dpf PET is about 12 microns. So it will take less weight of 12 micron fibers to = the number of 33 micron PET fibers.
There is one other aspect of modeling stuffing with fibrous materials and that is their inherent bending stiffness. I believe the fibrous mass in an enclosed volume (and maybe open as well) acts like a spring against the back wave from a woofer. Is this parameter a part of the models under discussion? I don't know.
It's my guess that modelling stuffing in closed boxes vs open tubes (TL's) is like comparing apples to oranges perhaps? In conclusioin, I should apologize for taking this thread OT.
Closed box stuffing is not rocket science. I think Mr. Gedlee has pointed this out using different words. The key objective is to make the box appear larger in volume than it actually is and thus optimize (minimize box Fc) to its lowest possible level and thus increase bass extension.
When stuffing closed boxes there is a nexis where the benefits of adding more stuffing reverses it's downward trend. In the case of the AR3a about 20 oz. of FG is optimum yielding an Fc of 42 hz on avg. OTOH, I have found (emperically) that the optimum level of PET results in a few Hz higher Fc AT A LOWER STUFFING DENSITY on the order of 16-18 oz. I believe this is simply due to the inherent differences in the density of PET vs FG (i.e. it takes more PET fibers to = the same number of FG fibers at a fixed weight of each). However, this will depend on the denier of PET being considered. 1.5 dpf PET is about 12 microns. So it will take less weight of 12 micron fibers to = the number of 33 micron PET fibers.
There is one other aspect of modeling stuffing with fibrous materials and that is their inherent bending stiffness. I believe the fibrous mass in an enclosed volume (and maybe open as well) acts like a spring against the back wave from a woofer. Is this parameter a part of the models under discussion? I don't know.
It's my guess that modelling stuffing in closed boxes vs open tubes (TL's) is like comparing apples to oranges perhaps? In conclusioin, I should apologize for taking this thread OT.
Anyway: these are the diagrams of the O'Hanlon work:
http://www.teodorom.altervista.org/Temp/Specific%20Acoustic%20Impedance.html
So I repeat my (uninteresting for someone) question: it seems that there is something wrong on the assumptions (I'm still investigating on them) that the properties of a model depend only on flow resistivity, and that flow resistivity depends only on fiber density (and diameter).
So it seems that to reproduce the fiberglass behaviour the polyester needs to have 10 times more density than fiberglass. It seems to me unrealistic that at such densities the models for the polyester fiber stay valid.
It seems (Tarnow, Robinson) that at certain frequencies fiberglass resonnate, even if the resonance peak is not so pronounced, so the "rigid" models are still good.
http://www.teodorom.altervista.org/Temp/Specific%20Acoustic%20Impedance.html
So I repeat my (uninteresting for someone) question: it seems that there is something wrong on the assumptions (I'm still investigating on them) that the properties of a model depend only on flow resistivity, and that flow resistivity depends only on fiber density (and diameter).
So it seems that to reproduce the fiberglass behaviour the polyester needs to have 10 times more density than fiberglass. It seems to me unrealistic that at such densities the models for the polyester fiber stay valid.
Agree. O'Hanlon would have been less confusing if he had published the Absorption Coefficient instead of Acoustical Impedance.
All the models I'm dealing with are "rigid" models. Some are purely phenomenological (Delany-Bazley), semi-phenomenological and based on fluid theory (Champoux-Allard), others (Wilson, I don't have a derivation) based on thermodynamics.
It seems (Tarnow, Robinson) that at certain frequencies fiberglass resonnate, even if the resonance peak is not so pronounced, so the "rigid" models are still good.
Thanks for the additional references.
I hope your modelling efforts are going to be beneficial to you in the future. Perhaps a Phd thesis?
I hope your modelling efforts are going to be beneficial to you in the future. Perhaps a Phd thesis?
I'm sorry, I did not mean to insult you. But you have to understand that not everyone iis going to be as interested in this obscure topic as you and you have no right to expect them to hunt out papers and do research on the topic because it interests you.
From my experince the "rigid" assumption is very limiting, and seriously questionable, because, lets face it, none of these fibers are "rigid". This fact alone could completely invalidate everything that you are doing. Lets not "miss the forest for the trees".
Sincely I expected much more from a person with your reputation.
If you are not interested in some topic, simply go to next one.
If you are interested but you don't understand, simply ask for clarifications.
If you understand but you disagree, simply point out your opinion.
If you understand and you agree, I'm happy !
If you understand and you have some further point that can make an advancement to the discussion, all the people interested in that topic are happy !
This is the point. You and many others are happy with those generic statements.
What do you mean by "your experience" ? did you make any measurement ? if you hade them I would be happy If I could work on them and then test the various models.
If a "rigid" model works and predicts the correct acoustical behaviour of a fiber, that fiber doesn't move (or at least, doesn't move noticeably).
"Hypotheses non fingo", as someone much, much, much better than me, once upon a time said.
I have a great future behind me.
I have a Ph.D. in Theoretical Particle Physics (1981).
I worked in I.T. for 30 years.
Now I'm retired (but I still work part-time in IT), so I have the time to write in "good copy" the material I have developed since 5 years. I shall have also the time to put in good shape (a nicer box) the TL loudspeakers I made since then.
Congratulations then on your unbridled enthusiasm for perfecting your TL project via the modelling process rather than using simple trial and error most of use when stuffing speakers.
Situation is even worse than that.
I tried to normalize the Acoustical Impedances, that, as everybody knows, should go to the Characteristic Impedance (ZC) as the frequency goes to infinity.
It happens with the Wilson model (and the other I mentioned):
An externally hosted image should be here but it was not working when we last tested it.
but it doesn' happen with the O'Hanlon model:
An externally hosted image should be here but it was not working when we last tested it.
Hi speakerdoctor
You can check the MJK site (Quarter Wavelength Loudspeaker Design) or the R.A. Robinson Ph.D. Thesis (http://etd.gatech.edu/theses/available/etd-04092007-102240/unrestricted/robinson_robert_a_200705_phd.pdf).
There is much more material there than I could never reproduce (and I don't will to).
we are talking abaout two different things: electrical impendance and acoustical impedance. Not to be confused.
You can check the MJK site (Quarter Wavelength Loudspeaker Design) or the R.A. Robinson Ph.D. Thesis (http://etd.gatech.edu/theses/available/etd-04092007-102240/unrestricted/robinson_robert_a_200705_phd.pdf).
There is much more material there than I could never reproduce (and I don't will to).
I would assume the rigid model would be sufficient for stiffer fibers, most of which are seldom used these days, but was quite popular when I started out DIYing.
Interesting collection of fiber information.
Animal fiber - Wikipedia, the free encyclopedia
Animal fiber - Wikipedia, the free encyclopedia
Theodoro
I would almost rather talk about Quantum Physics! You studied Quantum Mechanics with a hobby in acoustics and I studied acoustics with a hobby learning quantum mechanics! What a pair.
As I have said, I studied this problem a long time ago - much more than 20 years - and I am not so interested to go back to it. So all I can offer is what I remember about what I studied, but I don't have any data and I am not going to read any papers.
But clearly the rigid model fails as the frequency drops. The flow past a fiber causes a force and the fiber, not being perfectly rigid, has to deflect. This much is clear. At HF it is also clear that the fiber is rigid because there is not enough time for the fiber to deflect and hence it will remain stationary. So the reality is somewhere in the middle - but clearly changes with frequency.
I would almost rather talk about Quantum Physics! You studied Quantum Mechanics with a hobby in acoustics and I studied acoustics with a hobby learning quantum mechanics! What a pair.
As I have said, I studied this problem a long time ago - much more than 20 years - and I am not so interested to go back to it. So all I can offer is what I remember about what I studied, but I don't have any data and I am not going to read any papers.
But clearly the rigid model fails as the frequency drops. The flow past a fiber causes a force and the fiber, not being perfectly rigid, has to deflect. This much is clear. At HF it is also clear that the fiber is rigid because there is not enough time for the fiber to deflect and hence it will remain stationary. So the reality is somewhere in the middle - but clearly changes with frequency.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.