Instead of berating me for being to stupid to understand points that have "been clearly shown to you", why don't you simply try to name those advantages for mid- and upper-range drivers and how that beats a simple box in those passbands. And wouldn't that be helpful for those poor "lurkers and others" you are so sympathetic to?
Just two or three point-form comments would help your case, eh. Dave, time to stop berating and simply post the points, even if they apply to the TLs you sell.
BTW, Dave, regarding your post 141, are you saying I don't know the difference between a sealed box and a TL? C'mon now, not nice of you.
B.
Last edited:
Not easy to tell the difference between a TL and just a box with an opening acting like an OB without knowing the numbers.
For sure, millions of quality tweeters (esp domes) have been made with sealed rear chambers barely bigger than the envelope of the driver. So hard to imagine any advantages to building a labyrinth when just a box will do.
Also, as we see with the Nautilus, even respected manufacturers can go "populist" and use gimmicks that appeal to the eye or unexamined mind.
In olden tymes, when you might want to squeeze some mid-range out of your woofer (not smart), the shape of a TL box sort of helped nicely putting the mid-range at ear level. But there no advantages to the mid-range sound by forcing it to play into that TL. And that TL then had to fight hard with lots of stuffing calibrated "to taste" to suppress the awful spurious mid-range sounds a TL creates.
But as a sub, a rear pipe or normative TL can be great.
B.
Last edited:
I think we have some confusion between the open ended (usually folded) TL used to reinforce output near the lower cut-off and a closed exponentially tapered tubes used to absorb sound emitted at the rear of the driver.
The former is designed is designed for bass duties and is a resonator around its lower cut-off. Usually it has a parabolic taper. The latter is designed to only allow a plane wave travelling toward the narrow end and have it absorbed on its journey. It has an exponential taper.
The former is designed is designed for bass duties and is a resonator around its lower cut-off. Usually it has a parabolic taper. The latter is designed to only allow a plane wave travelling toward the narrow end and have it absorbed on its journey. It has an exponential taper.
Last edited:
Thanks! This makes sense. Although I use the "closed" loosely as I believe in the Nautilus designs the ends are open, but stuffed with absorbent materials. There is no low-end reinforcement.
Then there's the Focal / Sopra tweeter which appears to be a rear, open ended horn.
A closed TL is what it sounds like: a pipe closed at both ends (technically a 1/2 wave). An open TL is open at one end & is a 1/4 wave, nominally resonanting at 1/4 wave of its acoustical (as distinct from its physical) length.
You can use either for midrange lines. An open TL, e.g. what Dave uses, but many others too, can be stuffed to ~aperiodic levels, thereby flattening the overall impedance load. Very effective they are.
Can we please cease this twaddle (and it is twaddle) about TLs being only useful for subwoofers. No they aren't.
Caveat: 'Transmission Line' has become a catch-all phrase to the point that people use it to describe a vast range of enclosures, many of which are in loading intention the reverse of each other.
-An aperiodic TL is useful if you want a max-flat impedance, either for midrange or bass duties, which are not exclusive, last I checked, to 'sub-bass'.
-A mildly resonant quarter-wave line, mass-loaded either via a vent or by tapering (which is inherently mass-loaded) can be very useful in allowing a lower practical tuning frequency than a regular T/S vented box. This is demonstrated by the thousands of examples of both types in existance, created by DIYers and professional designers. Any attempt to claim otherwise is going up against this unassailable fact. The operating physics are well understood. Excellent information may be found on Martin King's website & George Augspurger's AES paper (these used different methodologies & had somewhat different objectives, but corrolate extremely well). From there, it is possible to branch out into other areas if desired, but they contain all the data many are ever likely to require.
-A TL / QW does not have to be a 'labyrinth' as has been pointed out repeatedly (which makes it curious the statement is being repeated, as though this minor fact had never been mentioned...) and said QW / TL may in fact be created in an enclosure with a single internal panel, or none at all using a tapered or untapered box no more physically complex than a conventional reflex. This level of complexity is unlikely to cause grown men to swoon, women to tear their hair, and have urchins weeping in the streets.
-Personally I'd be intrigued to know where this business of 'TLs being an attempt to get more midrange out of a woofer' came from, in 'olden tymes [sic]' or otherwise. That's news to me. Perhaps some documentary evidence showing this was the specific design intention in all such cases would be in order. To the best of my knowledge, to use Bailey's original article as an example, his intention was in fact to produce a cabinet with reduced ringing than a large contemporary reflex box. I'm not entirely convinced that that can be used as a synonym for 'the only objective for all early TLs was to try to squeeze more midrange out of a woofer'. Because they weren't. The numerous 3-way examples of TLs from the late 1960s - early 1970s rather obviously underline that fact. First, Rogers, then Radford & IMF lines all immediately leap to mind of multiway designs where the woofer was not run to an especially high frequency, and that was far from being the primary objective behind using a line.
-As I have shown above, blanket assertions that all pipes have 'awful spurious resonances' are completely incorrect. For example, I posted the plots of two closed pipes above, both of which were completely undamped for the sake of clarity, which clearly demonstrate only moderate issues with harmonic resonances that are very easily suppressed with minimal lagging of the pipe walls (also shown). I requested at the time that somebody explain how these easily damped harmonic modes were in fact catastrophically bad, as had been previously claimed, but answer came there none. The same question must, I fear, be asked regarding the thousands of quality MLTLs and tapered pipe designs in existance, viz. where, exactly, are the hideous resonances that have been attributed to them in a blanket assertion that QW / TL designs are somehow only suited to subwoofers. This must be further asked of George Augspurger's work, which shows the reverse, that of Martin King (likewise), and the many others who work with QW / TL designs of all kinds on a daily basis.
You can use either for midrange lines. An open TL, e.g. what Dave uses, but many others too, can be stuffed to ~aperiodic levels, thereby flattening the overall impedance load. Very effective they are.
Can we please cease this twaddle (and it is twaddle) about TLs being only useful for subwoofers. No they aren't.
Caveat: 'Transmission Line' has become a catch-all phrase to the point that people use it to describe a vast range of enclosures, many of which are in loading intention the reverse of each other.
-An aperiodic TL is useful if you want a max-flat impedance, either for midrange or bass duties, which are not exclusive, last I checked, to 'sub-bass'.
-A mildly resonant quarter-wave line, mass-loaded either via a vent or by tapering (which is inherently mass-loaded) can be very useful in allowing a lower practical tuning frequency than a regular T/S vented box. This is demonstrated by the thousands of examples of both types in existance, created by DIYers and professional designers. Any attempt to claim otherwise is going up against this unassailable fact. The operating physics are well understood. Excellent information may be found on Martin King's website & George Augspurger's AES paper (these used different methodologies & had somewhat different objectives, but corrolate extremely well). From there, it is possible to branch out into other areas if desired, but they contain all the data many are ever likely to require.
-A TL / QW does not have to be a 'labyrinth' as has been pointed out repeatedly (which makes it curious the statement is being repeated, as though this minor fact had never been mentioned...) and said QW / TL may in fact be created in an enclosure with a single internal panel, or none at all using a tapered or untapered box no more physically complex than a conventional reflex. This level of complexity is unlikely to cause grown men to swoon, women to tear their hair, and have urchins weeping in the streets.
-Personally I'd be intrigued to know where this business of 'TLs being an attempt to get more midrange out of a woofer' came from, in 'olden tymes [sic]' or otherwise. That's news to me. Perhaps some documentary evidence showing this was the specific design intention in all such cases would be in order. To the best of my knowledge, to use Bailey's original article as an example, his intention was in fact to produce a cabinet with reduced ringing than a large contemporary reflex box. I'm not entirely convinced that that can be used as a synonym for 'the only objective for all early TLs was to try to squeeze more midrange out of a woofer'. Because they weren't. The numerous 3-way examples of TLs from the late 1960s - early 1970s rather obviously underline that fact. First, Rogers, then Radford & IMF lines all immediately leap to mind of multiway designs where the woofer was not run to an especially high frequency, and that was far from being the primary objective behind using a line.
-As I have shown above, blanket assertions that all pipes have 'awful spurious resonances' are completely incorrect. For example, I posted the plots of two closed pipes above, both of which were completely undamped for the sake of clarity, which clearly demonstrate only moderate issues with harmonic resonances that are very easily suppressed with minimal lagging of the pipe walls (also shown). I requested at the time that somebody explain how these easily damped harmonic modes were in fact catastrophically bad, as had been previously claimed, but answer came there none. The same question must, I fear, be asked regarding the thousands of quality MLTLs and tapered pipe designs in existance, viz. where, exactly, are the hideous resonances that have been attributed to them in a blanket assertion that QW / TL designs are somehow only suited to subwoofers. This must be further asked of George Augspurger's work, which shows the reverse, that of Martin King (likewise), and the many others who work with QW / TL designs of all kinds on a daily basis.
Last edited:
@planet10 I do not think bentoronto's "sub" should be taken literally but what he actually meant was a setup where Fs is within the passband. In your full-range driver scenarios, none of which are "subs", but all of them have Fs within the passband and would have fitted my updated definition. Therefore it makes perfect sense to put FR's in TLs.
Now, take a tweeter with a Fs of 1kHz, and quarter-wave TL it at the same, open-ended. See how it doesn't make any sense?
(Disclaimer: I am mixing up Fs and Fp assuming they are always the same for simplicity. Of course that does not need to be the case.)
Now contemplate what TBTL told us: put that same tweeter in a close-ended TL, preferably with an exponential taper, and now it makes perfect sense again.
Now, take a tweeter with a Fs of 1kHz, and quarter-wave TL it at the same, open-ended. See how it doesn't make any sense?
(Disclaimer: I am mixing up Fs and Fp assuming they are always the same for simplicity. Of course that does not need to be the case.)
Now contemplate what TBTL told us: put that same tweeter in a close-ended TL, preferably with an exponential taper, and now it makes perfect sense again.
And attached below is what Augspurger has to say about "moderate issues" in a pipe with light damping*. Not what i'd call moderate, even without seeing the additional noises that would be higher in freq.
Yes, you can add enough stuffing to kill the higher noises but it seems to have killed the lower bass too. Yes, you can make very good TLs for woofers as Augspurper shows later in his article; done it myself. But you can't make a good one without stuffing.
We seem to have a case of people who truly love pipes - both exponential and parabolic. What we aren't seeing is any reason to favour them over simple damped boxes for higher range use where the driver's resonance is not within the passband. Maybe some vendors like to fix problems that are imaginary.
B.
*ummm, Augspurger's group of plots on the left sort of resembles what I posted.
Attachments
Last edited:
TBTL's explanation was perfect. Would you put a midrange in a vented enclosure but then high-pass it at an octave above vent tuning? You won't.
Now put that same midrange in a TL and it will be high-passed an octave above Fp. But then you would close-end it, or leave it open-ended but stuff it so much that it makes practically little if any difference.
Now put that same midrange in a TL and it will be high-passed an octave above Fp. But then you would close-end it, or leave it open-ended but stuff it so much that it makes practically little if any difference.
So what you are saying is that you have taken a single example of an open pipe not optimised for such use (no taper, no mass-loading, no driver offset, no terminus offset, nothing: just an end-loaded pipe providing maximum excitation to the fundamental and all harmonic modes, fully open at the opposite end) and are claiming universal applicability.
In other words you claiming that driver offset, taper, mass loading, terminus offset &c. have no effect or relevance. Since that flies in the face of thousands of examples in daily use and the laws of physics, I'm afraid that is a somewhat questionable assertion. You have also inadvertently neglected to mention
-The pipes I referred to as requiring minimal damping were closed (as in 1/2 wave) pipes, not QW which is what Augspurger is referring to
-Augspurger had a fixed set of alignment goals for the TLs in his paper, and
-Refers to some of the effects of these characteristics later on in it also.
Augspurger's work was very good for near aperiodic QW TLs approaching flat impedance designs, and is required reading on the subject. He did not extend it though into higher taper ratios, detailed examination of driver offsets for these or other pipes, and / or in association with other types of line, the effects of mass loading &c. Nor did he refer to sealed lines. Since you make very strong statements on the subject of all lines irrespective of type, it is a trifle surprising you have yet to mention these rather salient matters and their effects. Perhaps you would care to do so, and explain why you appear to be claiming they have no influence on behaviour, and how, to use a simple example, a well designed MLTL suffers from all these horrific issues you describe. Especially the ones that don't.
I'm sorry, but you need to stop making these nonsensical statements that are directly against well known physics and engineering reality. There are numerous designers on this forum, notably Paul Kittinger, GM, and even little old me, who produce numerous QW designs for our fellow DIYers that do not exhibit such issues as those you describe. There are literally hundreds here, just do a search. And that's before you consider the ones elsewhere. You now need to state, very clearly, how and why not one of them works as designed.
In other words you claiming that driver offset, taper, mass loading, terminus offset &c. have no effect or relevance. Since that flies in the face of thousands of examples in daily use and the laws of physics, I'm afraid that is a somewhat questionable assertion. You have also inadvertently neglected to mention
-The pipes I referred to as requiring minimal damping were closed (as in 1/2 wave) pipes, not QW which is what Augspurger is referring to
-Augspurger had a fixed set of alignment goals for the TLs in his paper, and
-Refers to some of the effects of these characteristics later on in it also.
Augspurger's work was very good for near aperiodic QW TLs approaching flat impedance designs, and is required reading on the subject. He did not extend it though into higher taper ratios, detailed examination of driver offsets for these or other pipes, and / or in association with other types of line, the effects of mass loading &c. Nor did he refer to sealed lines. Since you make very strong statements on the subject of all lines irrespective of type, it is a trifle surprising you have yet to mention these rather salient matters and their effects. Perhaps you would care to do so, and explain why you appear to be claiming they have no influence on behaviour, and how, to use a simple example, a well designed MLTL suffers from all these horrific issues you describe. Especially the ones that don't.
I'm sorry, but you need to stop making these nonsensical statements that are directly against well known physics and engineering reality. There are numerous designers on this forum, notably Paul Kittinger, GM, and even little old me, who produce numerous QW designs for our fellow DIYers that do not exhibit such issues as those you describe. There are literally hundreds here, just do a search. And that's before you consider the ones elsewhere. You now need to state, very clearly, how and why not one of them works as designed.
Last edited:
I’m not exactly sure what you mean.
In the ML-TLs we use for bass, Fs is typically > Fp. In the midTweeters Fs is at least an octave below the start of the passband.
dave
From their website
http://www.bowers-wilkins.com/Downloads/Product/Manual/ENG_FP10313_Nautilus_manual.pdf
http://www.bowers-wilkins.com/Downloads/Product/Brochure/ENG_FP10313_Nautilus_brochure.pdf
The woofer section definitely appears to be sealed,
Whereas from the description and graphics that the other 3 pass bands are highly resistive, but not sealed.
And yes, as originally marketed, it was a fully active system with dedicated crossover - page 9 of brochure above. The choice of amplification was left to the customer.
http://www.bowers-wilkins.com/Downloads/Product/Manual/ENG_FP10313_Nautilus_manual.pdf
http://www.bowers-wilkins.com/Downloads/Product/Brochure/ENG_FP10313_Nautilus_brochure.pdf
The woofer section definitely appears to be sealed,
An externally hosted image should be here but it was not working when we last tested it.
Whereas from the description and graphics that the other 3 pass bands are highly resistive, but not sealed.
And yes, as originally marketed, it was a fully active system with dedicated crossover - page 9 of brochure above. The choice of amplification was left to the customer.
As Dave says, with the LF leg they were aiming for an alignment ~equivalent to a sealed load but with more effective suppression of the back wave. Or at least something that they could make a case for providing the latter. 
Whether it was or not, we don't and can't know unless somebody provides full data on the drivers and line so an equivalent sealed box alignment could be created and a comparison made.
Whether it was or not, we don't and can't know unless somebody provides full data on the drivers and line so an equivalent sealed box alignment could be created and a comparison made.
Last edited:
Doesn't the end cavity of the proto work as a resonator chamber? The final version definitely is sealed without a chamber.
Extracted from the manual linked previously,
"Research showed that the exponentially tapered pipe was an even better absorber than the cylinder. So complete was its absorbing action that the pipe could be left open or closed. This was the breakthrough."
Extracted from the manual linked previously,
"Research showed that the exponentially tapered pipe was an even better absorber than the cylinder. So complete was its absorbing action that the pipe could be left open or closed. This was the breakthrough."
Last edited:
Dave,
I did say Fs was not necessarily equal to Fp and I did request you temporarily ignore the difference - the rationale being that they are within an octave of each other and that difference was not enough to sway the debate one way or another.
My interpretation of bentoronto, hopefully not distorting what he said, was, "an open-end quarter-wave transmission line (straight, tapered, expanding, ML-TL) is useful for a subwoofer, woofer or full-range driver only when Fp is within the passband. It confers no advantage over a sealed enclosure for a midrange or a tweeter where the high-pass filter is higher than Fp." (I am including ML-TL because this statement is applicable to ML-TL although it is debatable if it should be counted as open or closed.)
Many less-than-cordial exchanges were the result of semantic differences which dissipates away when the preconditions are properly specified.
The Nautilus was the exemplar of how to use close-end exponential transmission lines to sequester the back wave. The tweeter enclosure certainly was superior to a plain sealed box because the exponential profile has natural standing wave suppression built-in. You may get some (or even most) of the benefit with a straight taper, but to go from a great sound to a legendary sound requires an exponential increase in effort. It is clear that 3 out of the 4-way had Fp outside of the passband, but this is closed end. Opening the end is not useful except for the lowest driver because you only want the first harmonic out of the terminus - all else is midrange leakage. For the upper 3 drivers the first harmonic was filtered out.
Transmission line is such a broad concept that encompasses so much that it is very easy to get into misunderstanding.
One more addendum: if you designed a midrange "transmission-line", open-end, and filter away Fp, I do not even doubt for a second it would sound excellent - if you had stuffed it sufficiently there is no output from the terminus, so whether it is open or close becomes irrelevant. If you had only stuffed it loosely, you could have designed an excellent-sounding midrange dipole. That which we call a rose smells as sweet by any other name.
Last edited: