"AN ELECTROACOUSTIC ANALYSIS OF
TRANSMISSION LINE LOUDSPEAKERS"
Robert Allen Robinson, Jr.
School of Electrical and Computer Engineering
Georgia Institute of Technology
May 2007
This document, from an associated of Dr. Marshal Leach, at the Georgia Institute of Technology, seems to contradict some of the Augspurger's findings, specially around the behaviour of stuffing material...
For the sake of the this discussion interest, I'm not taking party for one or the other..
http://www.google.es/url?sa=t&source=web&cd=4&ved=0CDgQFjAD&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.115.9825%26rep%3Drep1%26type%3Dpdf&rct=j&q=Olney%20labyrinth%20loudspeaker&ei=DY0XTdjXNMut8QPsoMSABw&usg=AFQjCNHmbg4D5YV5En4tPcoToyutKPYORg&sig2=POsLP-J2QyKyCTv6eyhrEg&cad=rja
TRANSMISSION LINE LOUDSPEAKERS"
Robert Allen Robinson, Jr.
School of Electrical and Computer Engineering
Georgia Institute of Technology
May 2007
This document, from an associated of Dr. Marshal Leach, at the Georgia Institute of Technology, seems to contradict some of the Augspurger's findings, specially around the behaviour of stuffing material...
For the sake of the this discussion interest, I'm not taking party for one or the other..
http://www.google.es/url?sa=t&source=web&cd=4&ved=0CDgQFjAD&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.115.9825%26rep%3Drep1%26type%3Dpdf&rct=j&q=Olney%20labyrinth%20loudspeaker&ei=DY0XTdjXNMut8QPsoMSABw&usg=AFQjCNHmbg4D5YV5En4tPcoToyutKPYORg&sig2=POsLP-J2QyKyCTv6eyhrEg&cad=rja
Last edited:
Thanks for a very interesting and involved analysis. It will take me a while to wade through it.
The nonlinear effect of output due to stuffing is interesting. His more complex model for line stuffing makes it look as if there isn't a clear cut ideal for stuffing density. Effects around the corner frequency seem everpresent.
His simulations don't make a good case for the TL vs. either the sealed box or vented box: less extension and much rougher response.
David S.
David Weems in his Book DESIGNING, BUILDING & TESTING.... makes a distinction between the labyrinth and transmission line systems, where the transmission line-type is a stuffed labyrinth. Would it be correct to say that there is virtually no acoustic suspension or excursion limiting similar to the BR in a transmission line? If so, it would seem to follow that (1) a suitable driver for a transmission line sys. must have an uncommonly stiff suspension, and/ or (2) displacement limited acoustic power output Par of the transmission line will be generally much less than that of a comparable closed-box and BR systems.
interesting read. i too have Weems' book, along with many people. the definition of a labyrinth in most books ive read is that of a single or multipath(since not all use a single path at all, and a series of staggered perforated baffles instead) line, with the walls of the 'line' lined rather than stuffed.
Also i may be wrong, but, 'acoustic suspension' woofers are pretty much extinct. what you say in pts 1 and 2 are essentially correct, IMHO, when applied to an 'acoustic suspension' woofer in a vented alignment, and apply to a lesser extent to a heavily stuffed TL.
a minimally stuffed TL, will see the reflected pressure wave in a similar way to reflex's. the effect in a minimally stuffed 1/2 wave sealed TL, I can only guess at. without medium-heavy stuffing in both cases, i would bet both 1/2 wave closed TL or 1/4 open TL would see this acoustic damping of cone displacement.
Nowadays it seems that most drivers are designed to satisfy vented alignments, rather than closed box (a pity IMO). strictly speaking a TL is introducing another resonant system into the mix. we already have 3 dimensional, then we add the pipe.....BUT even using GR for box dims, we still have 3 dimensional resonances, and in a typical closed box system, arent these likely to be closer together than in a tapered, adequately stuffed QWTL?
if both the closed box and TL are equally stuffed (re density) then i think id rather take the 4 well spaced resonances, than the 3 closely spaced ones.
however, if i am wrong, i really welcome some other opinions, or smth else to read to sway me otherwise 😀
Last edited:
I'd take those distinctions with a grain of salt. Its a bit semantic but I though the terms were interchangable. I actually prefer labyrinth since the reality is so far removed from an actual transmission line.
Regarding power output, you always have to put things into the context of how much damping you intend to apply. If damping is fairly low then you will get significant bass gain and a useful null in woofer excursion. At the same time the upper modes will likely be stronger than you want. Damp enough to remove the upper modes and you will likely have no excursion null but you may still have a couple of dB of bass gain. This will put you ahead of sealed boxes when comparing output vs. excursion, but behind well designed vented boxes.
For the same voltage input and high damping then, as you state, the excursion will be higher than the same woofer in a sealed enclosure, where the sealed box would raise resonance.
Uncommonly stiff? Both the vented box and TL will "unload" at very low frequencies so moderate stiffness will help. Beyond that, generalizations will get you into trouble.
David S.
Mondogenerator,
The distinction that you make between labyrinth and transmission line systems agrees exactly with that of Weems. Sorry, I guess that I did a lousy job of paraphrasing Weems's explanation.
Is there some way to know from the specs of a driver whether or not it must be acoustically suspended? From Weems again, acoustically suspending is required (or used to be required) for woofers of an advertised diameter greater than 6 inches.
Speaker Dave,
In the equation for Par for closed-box systems, Kp is very low for say Qtc = 0.5. My vision of a well-stuffed TL is that of the volume of enclosed air of the TL having a negligible affect on Q of the system. What I should have said I suppose is that Qts of the driver of the TL system should be approaching 1.0 in order to avoid an early onset of over-excursion. Is this what you mean by taking damping into consideration?
The distinction that you make between labyrinth and transmission line systems agrees exactly with that of Weems. Sorry, I guess that I did a lousy job of paraphrasing Weems's explanation.
Is there some way to know from the specs of a driver whether or not it must be acoustically suspended? From Weems again, acoustically suspending is required (or used to be required) for woofers of an advertised diameter greater than 6 inches.
Speaker Dave,
In the equation for Par for closed-box systems, Kp is very low for say Qtc = 0.5. My vision of a well-stuffed TL is that of the volume of enclosed air of the TL having a negligible affect on Q of the system. What I should have said I suppose is that Qts of the driver of the TL system should be approaching 1.0 in order to avoid an early onset of over-excursion. Is this what you mean by taking damping into consideration?
MCpete:
i have the 3rd edition of Weems' book, which doesnt go into detail RE transmission line/labyrinth design, so im afraid i have no idea 😀
i have the 3rd edition of Weems' book, which doesnt go into detail RE transmission line/labyrinth design, so im afraid i have no idea 😀
Sorry if I wasn't clear but for damping I was referring to line damping, the damping of line resonances, rather than changing woofer Q.
Lets differentiate between damping and varying woofer Q with different rear volumes. That isn't damping but a compliance change shifting resonance and Qt naturally shifting with it. (fs/Qt is approx. constant).
Smaller boxes will raise resonance (sealed box) and increase power handling for a constant voltage input. But when looking at maximum output there is no change. Vented boxes and, hopefully, well designed lines will give more output vs. excursion hence greater total output.
Back to your original comment. I'd agree that a stiffer woofer (higher Q) will be better with subsonic signals.
David S.
Hi all
Reading through this excellent thread, I realised that we try through stuffing to attenuate the harmonics of the tube's foundamental as though they are definitely a detrimental byproduct of this enclosure type.
But are they really detrimental?
They are angly looking when viewing a measurement FR or a SW simulation file, but it may be benign -or even beneficial- when our acoustic perception is concerned.
It seems that the frequency spacing of harmonics, is the clue that our brain uses to "synthesize" the initial harmonic (This was Aristotle's wonder "how is it that the high tone contains the low tone?").
Missing fundamental - Wikipedia, the free encyclopedia
Instead of trying to eliminate them, it might be better to study the perception effect of the gradual and controlled attenuation of the harmonics.
It remains open the treatment of the "one tone" effect , but this has to do with the geometric aspects of the tube and not with the stuffing scenarios PS*
As to the wise question of what effect does stuffing has to the sound that escapes back through the speaker's cone, I think that there too is important -perseption wise- to retain the whole spectrum of this escaping sound unaltered albait attenuated if possible, instead of attenuating only part of this spectrum (easier said than done of course).
Regards
George
PS* Perhaps we can apply some semi-terminations spread unevenly over the whole length of the tube, to artificially produce more than one "one tone" effects.
Reading through this excellent thread, I realised that we try through stuffing to attenuate the harmonics of the tube's foundamental as though they are definitely a detrimental byproduct of this enclosure type.
But are they really detrimental?
They are angly looking when viewing a measurement FR or a SW simulation file, but it may be benign -or even beneficial- when our acoustic perception is concerned.
It seems that the frequency spacing of harmonics, is the clue that our brain uses to "synthesize" the initial harmonic (This was Aristotle's wonder "how is it that the high tone contains the low tone?").
Missing fundamental - Wikipedia, the free encyclopedia
Instead of trying to eliminate them, it might be better to study the perception effect of the gradual and controlled attenuation of the harmonics.
It remains open the treatment of the "one tone" effect , but this has to do with the geometric aspects of the tube and not with the stuffing scenarios PS*
As to the wise question of what effect does stuffing has to the sound that escapes back through the speaker's cone, I think that there too is important -perseption wise- to retain the whole spectrum of this escaping sound unaltered albait attenuated if possible, instead of attenuating only part of this spectrum (easier said than done of course).
Regards
George
PS* Perhaps we can apply some semi-terminations spread unevenly over the whole length of the tube, to artificially produce more than one "one tone" effects.
Last edited:
I'm not sure if the resonances are beneficial but I certainly could believe they are more benign than we think.
The Augspurger paper seemed to optimize to 1dB of resonance ripple. Since there is a strong tradeoff between ripple and bass gain, I would be tempted to go for +-2dB ripple and get a little more bass output. When you measure a finished system in a real room I think you can get by with a little more ripple and never see it amongst all the other room effects.
David S.
The Augspurger paper seemed to optimize to 1dB of resonance ripple. Since there is a strong tradeoff between ripple and bass gain, I would be tempted to go for +-2dB ripple and get a little more bass output. When you measure a finished system in a real room I think you can get by with a little more ripple and never see it amongst all the other room effects.
David S.
id have to agree here, the room effect is almost as important as the speaker, crossover and everything else put together.
- Status
- Not open for further replies.