Hi,
I'm currently using my "full" range electrostatic loudspeakers combined with TL woofers. I use a dsp to switch from esl full range mode to esl + TL woofer mode, depentent on the music I listen to.
The TL woofers are actually a 2-way exiting loudspeaker DIY model:
As I only use the mid / bass driver, I would like to add another fold to the TL design in order to make the design less tall. I'm wondering if this would be possible without changing acoustical properties / sound?
I'm currently using my "full" range electrostatic loudspeakers combined with TL woofers. I use a dsp to switch from esl full range mode to esl + TL woofer mode, depentent on the music I listen to.
The TL woofers are actually a 2-way exiting loudspeaker DIY model:
As I only use the mid / bass driver, I would like to add another fold to the TL design in order to make the design less tall. I'm wondering if this would be possible without changing acoustical properties / sound?
Me too.
But why are you switching depending on the music? There's nothing I ever play that demands any change of my system although I can imagine garbage pop recordings that might sound boomy on a good system.
A TL behaves a lot like a tuned BR box (which it functionally resembles) but with lots of stuffing. There's no magic to any particular length (unlike BR theory) and the longer the better until it behaves just like a long labyrinth.
Specifics about bends in your TL will arouse theological arguments from the faithful.
The following links might help:
17 foot pipe sub 12-230 Hz ±5dB
Long pipe to sequester rear wave
B.
The response will change a little, because of the different path length between the "port" and the driver. I do not agree with the longer the length the better the TL, the lengt is critical, like in a ported speaker.
@bentoronto: I’m switching depending on music as on some recordings I find midrange seems a bit “veiled”: a bit duller sound. I don’t know why because I filter at 80 Hz @ 48 dB/oct on both subwoofer and esl panel. (dsp) Even when I crossover at 60 Hz I perceive a veiled mid range effect. I don’t get that. Any idea what could be the cause of this perception?
Thanks for sharing your TL project ideas.
Thanks for sharing your TL project ideas.
Wrong.
Read carefully what G. Augspurger and M. J. King wrote about the transmission line loudspeakers.
First you have to determine what is the cause for the veiled midrange. Try this experiment: set the crossover at 60 Hz and 48 dB/oct., disconnect the subwoofer and listen to the ESL panel only. Than remove the miniDSP from the audio chain and listen the ESL directly (sound level must be the same). If the veiling is gone, than the problem is DSP (which is unlikely, but worth a try).
Than try this: set the crossover at 80 Hz and 48 dB/oct., on both subwoofer and ESL. Listen, than try 24 db/oct at the same 80 Hz. Which one is better?
Last edited:
Sonce:
It is not altogether correct to say that Ben is wrong.
Ben has described (and evidently built) an enclosure design that owes much -- except, perhaps, on the issues of folding and duct cross section -- to Arthur Bailey whose design (and design intent) originated the moniker "transmission line".
Bailey was concerned that "the design of loudspeaker cabinets has not kept pace with (driver design)... and there is little doubt that many enclosures now introduce more coloration than that produced by good loudspeakers."
His objective:
"The loudspeaker enclosure to be described [Bailey's article was entitled, "A Non-Resonant Loudspeaker Enclosure Design"] was developed to give as little coloration as possible."
One form of cabinet effect that he was trying to reduce was the "abrupt change in slope of the amplitude/frequency characteristic [the "ringing"]...which is the necessary price to pay for the extension of bandwidth by the use of resonance effects."
His premise:
"The only safe way of removing the rear cone sound energy is by transmitting it down an infinite transmission line. This is obviously impractical so the nearest approximation was examined."
and in his subsequent article:
"In general it is very difficult to obtain absorption if the path length is less than one-quarter wavelength of the sound in free space"
Clearly, quarter wavelength lines are marginal lines with respect to the original design intent, one with which Ben seems aligned. King and Augsburger et al are after something else... something different... at or outside the margins of Bailey's notion... but neither are they "wrong". But perhaps what they describe are not "transmission lines" in the original sense. They do seem to have co-opted the name.
On the other hand Stu Hegeman developed an enclosure design of several "lines" of varying lengths and used the dispersed resonances to good effect.
It is not altogether correct to say that Ben is wrong.
Ben has described (and evidently built) an enclosure design that owes much -- except, perhaps, on the issues of folding and duct cross section -- to Arthur Bailey whose design (and design intent) originated the moniker "transmission line".
Bailey was concerned that "the design of loudspeaker cabinets has not kept pace with (driver design)... and there is little doubt that many enclosures now introduce more coloration than that produced by good loudspeakers."
His objective:
"The loudspeaker enclosure to be described [Bailey's article was entitled, "A Non-Resonant Loudspeaker Enclosure Design"] was developed to give as little coloration as possible."
One form of cabinet effect that he was trying to reduce was the "abrupt change in slope of the amplitude/frequency characteristic [the "ringing"]...which is the necessary price to pay for the extension of bandwidth by the use of resonance effects."
His premise:
"The only safe way of removing the rear cone sound energy is by transmitting it down an infinite transmission line. This is obviously impractical so the nearest approximation was examined."
and in his subsequent article:
"In general it is very difficult to obtain absorption if the path length is less than one-quarter wavelength of the sound in free space"
Clearly, quarter wavelength lines are marginal lines with respect to the original design intent, one with which Ben seems aligned. King and Augsburger et al are after something else... something different... at or outside the margins of Bailey's notion... but neither are they "wrong". But perhaps what they describe are not "transmission lines" in the original sense. They do seem to have co-opted the name.
On the other hand Stu Hegeman developed an enclosure design of several "lines" of varying lengths and used the dispersed resonances to good effect.
Last edited:
King’s modeler is capable of a wide range of design simulations, depending on the designers input. He has his own take on what he prefers but it is not a limitation of the software.
Bailey’s work is often misunderstood, and the cherry picked quotes, including the misleading title. Our understanding has dramatically improved since those days.
dave
Bailey’s work is often misunderstood, and the cherry picked quotes, including the misleading title. Our understanding has dramatically improved since those days.
dave
Bailey's core statement of intent is clear. Citing it is not "cherry picking" but getting at the heart of the matter.
And the journal seemed happy enough that the article's title was representative of the content or they would have fixed it or issued a correction.
This is not to take away from the work of King et al. Only to say they have been plowing a different -- and fruitful -- furrow.
And the journal seemed happy enough that the article's title was representative of the content or they would have fixed it or issued a correction.
This is not to take away from the work of King et al. Only to say they have been plowing a different -- and fruitful -- furrow.
The intent (if that is given in the title) is not what his example is. The enclosure would need to be damped until aperiodic for that to happen, ie no extra bass.
All TLs have a series of resonances where we try to add a LP such that the ones we do not want are supressed.
It is an important article, but dated.
dave
All TLs have a series of resonances where we try to add a LP such that the ones we do not want are supressed.
It is an important article, but dated.
dave
His statement of intent was made in the body of the article... "The loudspeaker enclosure to be described was developed to give as little coloration as possible".
Bailey may have failed to accomplish his aim fully but that was what he aspired to do.
And we know this because it was also the focus of his second article (and amplified in the letters). And there may be better ways to achieve those goals. Bailey simply proposed his.
Certainly by setting aside the topic of wringing out more bass enhancement by exploiting lingering resonances he opened up a fertile field for others to explore. I guess that was important. Again, I'm not denigrating the work of the modellers in their pursuit of their objectives.
But Bailey evidently thought that getting rid of coloration was a priority. It was his stated intent and objective. And in that effort, longer is better. Ben wasn't wrong if that was his intent too.
Can't we just leave it there?
Bailey may have failed to accomplish his aim fully but that was what he aspired to do.
And we know this because it was also the focus of his second article (and amplified in the letters). And there may be better ways to achieve those goals. Bailey simply proposed his.
Certainly by setting aside the topic of wringing out more bass enhancement by exploiting lingering resonances he opened up a fertile field for others to explore. I guess that was important. Again, I'm not denigrating the work of the modellers in their pursuit of their objectives.
But Bailey evidently thought that getting rid of coloration was a priority. It was his stated intent and objective. And in that effort, longer is better. Ben wasn't wrong if that was his intent too.
Can't we just leave it there?
That has been the goal of many designers, before & after Bailey. He stood on the shoulders of those before him (Onley is credited with one of the 1st TL papers, some decades before Bailey’s, April 1937, Karlson too), and those after him stood on his shoulders.
http://www.t-linespeakers.org/download/Onley-acoustic-labyrinth.pdf
Designing a TL the way he did is a crap shoot. Say maybe 1 in 10 reasonable lines. These days with a modeler in the hands of someone who knows what they are doing are better than 9 out of 10. And the modelers have shown us that the “TL space” is much, much larger than Bailey ever imagined.
I have been a big TL fan since i 1st heard the Radford S90 (the 1st commercial development of what Bailey built) and have read everything i could get my hands on. There was a huge game changer in late 1999 when the 1st modelers hit the streets.
dave
http://www.t-linespeakers.org/download/Onley-acoustic-labyrinth.pdf
Designing a TL the way he did is a crap shoot. Say maybe 1 in 10 reasonable lines. These days with a modeler in the hands of someone who knows what they are doing are better than 9 out of 10. And the modelers have shown us that the “TL space” is much, much larger than Bailey ever imagined.
I have been a big TL fan since i 1st heard the Radford S90 (the 1st commercial development of what Bailey built) and have read everything i could get my hands on. There was a huge game changer in late 1999 when the 1st modelers hit the streets.
dave
I have the Onley article but not the one by Karlson (Carlson... as in Stromberg Carlson, the company Onley worked for?). I've seen it referenced but I haven't seen a link or even a title or publication info. Got one you can share?
I did a lot of my reading in TAA/SB a long time ago. That includes the Augsperger series and others. I don't remember much discussion about how the modeling helped remove cabinet coloration (not that I could undrrstand). Was that because these objectives were readily solved without resorting to reducing bends (making the line less labyrinth-like), increasing line length, stuffing the line fully and with an appropriate (particular) material and employing particular wall configurations?
Was Bailey swatting flies with a sledge hammer? Or was he on to something?
By the way, what drew you to TLs in the first place and kept you interested? It wasn't the lack of coloration was it?
I did a lot of my reading in TAA/SB a long time ago. That includes the Augsperger series and others. I don't remember much discussion about how the modeling helped remove cabinet coloration (not that I could undrrstand). Was that because these objectives were readily solved without resorting to reducing bends (making the line less labyrinth-like), increasing line length, stuffing the line fully and with an appropriate (particular) material and employing particular wall configurations?
Was Bailey swatting flies with a sledge hammer? Or was he on to something?
By the way, what drew you to TLs in the first place and kept you interested? It wasn't the lack of coloration was it?
I am waiting for freddi to notice and post the paper (he is very likely to have it).
I was at the 1999 AES in NYNY when he 1st presented his work. At the same time i was working to help Martin get his paper up.
2 models with pretty much the same results, but each used different paradigms. It is unfortunate that George did not take his work any further, Martin is still improving his.
Colouration, other than that inherent to the particualr driver cabinets affect colouration by reflections back thru the driver cone, standing waves, and cabinet resonances.
A TL (with volume fill damping) kills most, if not all of the 1st, has a strong 1/4 wave resonance with a trail of harmonics, and if folded the internal panels act to brace the box. The goal, if one is attempting to get extra bass reinforcement, then one wants a low pass that takes out (most of) the harmonics which are not wanted. Historically this is been done significant volume fill damping, but inevitably one kills some of th efundemental. Modern modeling have shown us tricks to do this with less damping so that less of the fundemantal resonance is not also reduced. Restricted terminus (mass-loading), a pre-camber and a driver offset are 2 prime ones, allowing for less damping and more of the fundemental coming thru.
Also, modeling has shown us that the claim of decreased speed of sound is actually that the taper affects the natire of the resonance train. Tapering from large to small at the terminus decrease the length of the line for a given fundamental resonance.
An aperiodic line, most useful in a midTL, has no bass extension and if done well can suppress most of everything coming off the back of the driver, with the bonus of damping down the resonance peak.
At the frequencies involved the back wave does not see the folds. If ne leaves out the deflectors shown in illustrations above, the LF is not affected, but the expansion of the line acts as an additional LP filter. The deflectors do add a tiny bit of bracing.
Bailey was important in that it brought earlier work to attention, and goosed a lot of empirical work on TLs. The methodology (called Classic Design) was really quite bad, and those who either kept plugging away with box after box until things got decent, and those OK speakers gained a following. The many bad ones had many saying WHY?
You do see hints of the restricted terminus, and in things like the Radford & IMF the large pre-cavity. But this was from trial & error.
The excellent performance (in its day) of the Radford S90. And meeting the really clever & smart guy who was building them.
dave
An interesting discussion, maybe even a bit less cult-following than usual when TLs are discussed here. Thanks to mhenschel for clarifying matters so nicely.
But the OP asked about lengthening the pipe. As I said in post #2, "A TL behaves a lot like a tuned BR box (which it functionally resembles) but with lots of stuffing. There's no magic to any particular length (unlike BR theory) and the longer the better until it behaves just like a long labyrinth."
I don't think anybody has disagreed with that except to mention that "Martin" offers ways of finding the shortest path and hence the smallest box that - to the ears of his adherents - works OK.
Apropos silvershadelynx's post #4, I wish I better understood the description of how the system was running and what the perception was. Almost all systems benefit when you stop harassing them with signals outside what they comfortably play, esp full-range ESLs and their hardworking amps. But seems odd that one configuration is better on some music and the other configuration on other music.
B.
But the OP asked about lengthening the pipe. As I said in post #2, "A TL behaves a lot like a tuned BR box (which it functionally resembles) but with lots of stuffing. There's no magic to any particular length (unlike BR theory) and the longer the better until it behaves just like a long labyrinth."
I don't think anybody has disagreed with that except to mention that "Martin" offers ways of finding the shortest path and hence the smallest box that - to the ears of his adherents - works OK.
Apropos silvershadelynx's post #4, I wish I better understood the description of how the system was running and what the perception was. Almost all systems benefit when you stop harassing them with signals outside what they comfortably play, esp full-range ESLs and their hardworking amps. But seems odd that one configuration is better on some music and the other configuration on other music.
B.
That is not true at all.
A BR only works if the ratios of the dimensions do not become too large
As one dimension becomes considerably larger than the others a BR transitions to a Mass Loaded TL.
Compilation of MJK ANSYS simulation of BR vrs ML-Voigt (an ML-TL variant).
As length increases the line frequency goes down, and the frequencies that exit the terminus is lower.
Now if you damp the ine until aperiodic, nothing or near nothing is coming out the line, them a longer line does not hurt. I use that scheme in midTLs, but it is rare in a woofer, and if it is, it is more often called a labyrinth (after Onley) not a TL.
dave
Yes, it is.
He is wrong.
You are wrong.
Augspurger and King gave a scientific proof (theory + measurements) that Bailey's TL is a marginal case of a general Transmission Line loudspeaker, and a not optimal one! Just pay attention to the 2 dB depression in the measured frequency from 45 through 110 Hz in the Fig. 9, from the original Bailey paper. That is far from optimal! In fact, it is a bad execution of a TL . But in spite of that, it was better than the majority of bass-reflex speakers of that era - it was back in year 1965 (bear in mind that Small paper on vented box was published in Journal of AES in 1971)!
No, he did not:
He asked about different folding, keeping the same total pipe length ("without changing acoustical properties"). If you add another fold to the present box, the box will be just as tall as before, but with completely different acoustical properties (as a result of the longer pipe).
Answer to the OP question: Yes, it is possible, just keep the total pipe length the same (center line of the pipe).
Last edited:
@Ben
"Almost all systems benefit when you stop harassing them with signals outside what they comfortably play"
Peter Moncrief subscribed to that theory and I think it has merit.
"Almost all systems benefit when you stop harassing them with signals outside what they comfortably play"
Peter Moncrief subscribed to that theory and I think it has merit.
Thanks for your replies. A little offtopic: I discovered another TL design at https://www.quadesl.com/speaker/diyesl/tl_cad.pdf
It uses a Seas ExcelW22 8" woofer. This woofer is much more expensive than my Scanspeak 18W8535-01 woofer. I'm wondering if one could expect an improvement in sound quality when utalizing a more expensive woofer? I'm aware more expensive doesn't neccesarry mean better.
It uses a Seas ExcelW22 8" woofer. This woofer is much more expensive than my Scanspeak 18W8535-01 woofer. I'm wondering if one could expect an improvement in sound quality when utalizing a more expensive woofer? I'm aware more expensive doesn't neccesarry mean better.