Transmission Line - Sound Speed

Status
Not open for further replies.
Jump to Latest
#2
What is the purpose of the line, is it to absorb the back wave or is it a resonator? In the first case, the line length is not critical. In the second case, use the normal speed of sound and only put damping material near the closed end of the line. If there is damping material near the open end, the line won't resonate any more at the fundamental frequency and bass output is reduced.

I would recommend to use Hornresp or similar software to simulate and design the system.
 
#5
chevchevchelios said:
Hello. I'm projecting a TL loudspeaker and my question is about the sound speed.
Should I calculate the line lenght using normal sound speed or sound speed on damped material? I heard that the sound speed changes with damped material inside cabinet.
Click to expand...
Yes the speed of sound is changed when travelling through a mixture of air and other materials compared to just air. The amount for typical foam or fibrous damping materials will vary with the geometry and material properties of the fibres, how they are fixed, density, temperature and pressure, frequency,... It isn't easy to calculate with accuracy for particular materials and so performing experiments is often the most practical approach. However, most quarter wave designs use light stuffing density leading to fairly modest changes and so an estimate after a bit of reading about the values determined by others should be sufficient for a ballpark estimate.
 
#6
I remember, from AR Bailey's seminal 1965 article in Wireless World “A Non-resonant Loudspeaker Enclosure Design”, that he described the filling as critical. In his design it performs the dual function of slowing down the back radiation (thereby reducing the physical line length necessary for correct operation) and absorbing upper bass frequencies, such that only the lowest bass frequencies appear at and are augmented by the vent.

Unfortunately, I don't remember the exact details, other than the fact that he found that the only really viable material was long-fibre wool which had been teased out just the right amount, and that density was greatest immediately behind the driver.

Hopefully, a little bit of research will yield further information. Good luck!
 
#7
OK, some facts.

There is relatively little reduction in the SoS (there is some) due to damping, and this is caused by viscous losses and non-adiabatic process:

http://www.quarter-wave.com/TLs/Test_Line_Results.pdf
http://www.quarter-wave.com/TLs/Damping_Coefficient.pdf

There is nothing particularly special in a damping sense about long-hair wool -it works very well, but plenty of other materials also work well, with subtle differences in absobtion efficiency at different frequencies depending on their physical properties: see Augspurger's AES paper for some comparative details http://diyaudioprojects.com/Technical/Papers/Loudspeakers-on-Damped-Pipes.pdf

In practical terms, Bailey's enclosure was akin to a Stromberg-Carlson acoustical labyrinth (a well-known and oft-used box type) with the lagging replaced by stuffing and with slightly different objectives for testing. The effects of the taper & end correction do not seem to have been accounted for though. Bailey's second article notes an apparent reduction in the SoS of ~0.7 - 0.8 -this was likely inferred from the lowered pipe Fp. However, a good proportion of this will actually have resulted from the modest taper of the line, with the damping providing the rest.

Short version: the SoS in a damped pipe is reduced by damping, but not by anything like as much as is often implied. What tends to be more significant in altering Fp (pipe fundamental) is its geometry: an expanding pipe has a higher Fp for a given physical axial length than a straight, a contracting pipe a lower, how much so depending upon the degree of taper applied.
 
Last edited:
#9
I almost always use a stuffing density of 0.75 lb/ft3 with stuffing (polyester fiber) usually located in the first 2/3 of a tapered line (contracting area). Comparing the line's resulting f3 in the same non-stuffed line to that with the line stuffed, the modeled results show that f3 is typically lowered by 1-3 Hz by the stuffing. Stuffing is essentially a low-pass acoustical filter and while different stuffing materials can be used (polyester fiber, fiberglass and wool) they all have the same effect but to different degrees and, therefore, require different stuffing densities, but none is inherently superior to the others.
Paul
 
#11
Scottmoose said:
OK, some facts.
There is relatively little reduction in the SoS (there is some) due to damping, and this is caused by viscous losses and non-adiabatic process:
Click to expand...

This is incomplete. There can be substantial changes in the speed of sound although usually not with the typical materials and light stuffing densities typically used in speakers with pipes that seek to let through low frequency sound relatively unimpeded and absorb high frequency sound. The motion of the solid material is an additional and relevant physical mechanism in some cases despite the author you linked to getting upset by it for reasons that aren't fully apparent but may be due to historical disputes where it was claimed to be the sole or perhaps dominant mechanism.

Despite claims of novelty and odd/incomplete physics within the DIY speaker world the behaviour of sound in porous media has, not surprisingly, been extensively studied, modelled and measured in the scientific literature. Here is a brief review from the web but you would likely need to go to a library to get anything much done due to the age and closed journal publication of most of the relevant work.
 
#13
andy19191 said:
This is incomplete. There can be substantial changes in the speed of sound although usually not with the typical materials and light stuffing densities typically used in speakers with pipes that seek to let through low frequency sound relatively unimpeded and absorb high frequency sound. The motion of the solid material is an additional and relevant physical mechanism in some cases despite the author you linked to getting upset by it for reasons that aren't fully apparent but may be due to historical disputes where it was claimed to be the sole or perhaps dominant mechanism
Click to expand...

To date (as far as I am aware in the scientific literature) Bradbury et al's hypothesis of coupled motion at low frequencies with an aerodynamic drag coefficient has yet to provide a particularly accurate corrolation with measured results at low frequencies. This was why MJK abandoned it, as its predictions failed to accurately match his own measurements. And it is worth noting that Bradbury pointed out in his original paper that the accuracy of his model fell away at lower frequencies.

Since I happen to work in the academic field, I have access to most journals through my institution and do try to keep abrest of developments; that being said, I have not kept a close eye on this rather narrow matter of the coupled motion of porous media (or fibrous tangles) in the wider literature since it is in essence irrelevant for audio purposes as a perfectly ordinary model based on viscous damping has proved consistently effective for such uses, with excellent corrolation to measured results. Both MJK and George Augspurger, working independently, with different modelling approaches, found this and literally thousands of loudspeakers have been developed on this or equivalent basis, which work as intended. Perhaps you could specify some of the papers you are referring to (with particular relevence to loudspeaker design)? I would very much like to read them, and compare to other work and my own test line setups.
 
Last edited:
#16
Hi. Been lurking for a little bit now, first time post. Very interested in DIY projects and have been reading all the normal literature. TLs are what interests me the most, and of course they are the most elusive.

Excited to also be receiving my first TL speakers next week... a Dennis Murphy and @pkitt collaboration. 🙂

I've been to the two sites, also. Trying to digest King, as well.

Left with some questions though, with my very remedial understanding.

Apologies now if this is not the best place to post these questions.

What is the relationship between Sd and the Area of the closed end of a tapered pipe. Is there an equation I've missed for this somewhere?

Also, Vas and the volume of the line: I see many saying their is a relationship but feel like I'm missing that somewhere?

Appreciate any help pointing me in the right direction!!!

Cheers,
R
 
#18
As indicated above, there is no direct functional relationship between Sd and line CSA. An indirect [several stages removed] only as a constituent part of Vas. Martin for example used multiples of Sd as a convenient means of expressing CSA in some of his earlier work, but I think he later regretted doing so as it perpetuated the belief that there was some close relationship between them; his later material simply stated WxD dimensions to express CSA to avoid this.

Broadly speaking, Fs, Qt and Vas are the dominant driver parameters as far as Vb and tuning are concerned.
 
Last edited:
#19
If your new speaker is the BMR monitor, it's not a TL, although I did use TL modeling techniques to find an optimal location for the port relative to the woofer's location and internal cabinet height. As already mentioned by Scottmoose, there's no specific formula that dictates a line's area based on the Sd of the driver. Once the optimum tuning frequency, based on the driver's fS and Qts, is determined from a line's length and taper ratio (that creates its 1/4-wavelength resonant frequency) the volume in the line, relative to Vas, determines the bass response, like f3. IOW if you want a lower f3 you increase the line's volume while keeping the line's length and taper ratio the same.

Paul
 
#20
Fortunately, Paul, I know the difference between the BMR and the Phil3. 🙂 Interesting tidbit about the BMR cabinet... Dennis did mention that you helped on that too. Great speakers! Talking to Dennis was a major inspiration in wanting to follow the DIY path. Besides, in 10 years, when I want to upgrade beyond the Phil 3s and BMRs, I likely won't be able to spend 30K to do it! And so here I am.
To all: thank you for your replies. Yet again, it seems I am still missing something important at a very basic level. Back to the beginning again.

I will figure this out!
Best,
R
 
Status
Not open for further replies.
Top Bottom