• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Transmission-line speakers

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
889
Likes
1,676
Location
Norway
I did a short text on transmission-line speakers, perhaps someone finds it entertaining enough to read here as well:

---------
The classic transmission-line speaker is a long acoustic channel with decreasing area, filled with damping material to absorb all sound. The claimed benefit is removal of all resonances and reflected sound from inside the cabinet, so the driver operates in a cabinet that is removed acoustically - like an acoustic black hole.

It was supposed to be driven by a driver with heavy moving mass, and low motor force. The length of the channel determines the low frequency cut-off, where this length equals 1/4 wavelength. This gives a speaker with quite low cut-off, and low efficiency.

Then someone came up with the idea of letting some of the sound radiate out from the end of the pipe, to increase efficiency. Since this sound is delayed from the sound from the driver diaphragm, the port output and driver output combine in-phase and the total sound output is then increased.

Of commercial designs utilizing this concept, the IMF were famous for its reproduction of the very lowest frequencies - in a time where subwoofers did not exist, this was among very few speakers that could reproduce sound down to and below 20Hz.

Transmission lines are complicated to design and build. And there are some technical challenges that can be difficult to solve. The channel itself can have resonances, especially upwards in frequency, above its intended frequency range. The port output is delayed and only sums correctly with the sound from the front of the driver across a limited frequency range - there will be cancellation above this range, and also problems with delayed sound output.

The ported variations will have the benefit of acoustic loading and increased efficiency due to the port output, similar to an ordinary ported bass-reflex box.

The closed-end variant does not have those issues with delayed sound and problematic summation, and it is possible to achieve resonance-free response, and this black-hole cabinet behavior. But the ordinary sealed cabinet can perform just as good, when filled appropriately with damping material.

Great care must be taken when designing a transmission line with acoustic port at the end of the channel, to ensure there are no resonances left, and attenuate the output from the port at higher frequencies. If done properly, it will perform quite similar to a ported box.

I do not have a transmission line to show, but I can use the F2 as an example of a ported horn design that works in a similar way, and presents the same challenges. The driver is mounted off-set into the channel of a short horn, and this horn is then ported at the bottom of the cabinet. Like this:
F2wf10242.jpg

For this to work properly, the horn channel must be damped very accurately, in the right places with the right type and amount of damping material. Here is the frequency response of the port output - yellow is with no damping, red is damped and how the speakers is when finished:
Port output med uten kabinett demping.png

We see that the undamped response is quite horrible and useless, with severe resonances. The damped is smooth and falls off towards higher frequencies.

Then the question is - why bother with such a complicated design, with all its potential issues? For the F2, the answer is simply that this bass driver needs a cabinet with acoustic reinforcement from around 200Hz and down, to provide any bass at all. The F2 now achieves reasonably flat response down to a roll-off at 60Hz, which gives typical in-room response down to around 40Hz. (Now, what is the point of this speaker - it does not play full-range bass - still needs a subwoofer, and then you really do not need the extension down to 40-60hz..)

Whether you call it transmission-line or something else, what we are dealing with here are cabinets with some sort of acoustic loading using a channel that is long compared to the wavelength. And they can have benefits like increased output capacity, extended low-frequency range and better acoustic coupling.

------

My opinion is that transmission-line - or similar acoustic loading using long channels - is problematic for full-range speakers covering both bass and midrange frequencies. Today we can use simulation to optimize the design, but physical laws still apply.
 
Transmission lines work fine, when they're real transmission lines, i.e. infinitely long, or effectively so. The problem is that the loudspeaker then has to be BIG, (and expensive) and that limits the market, so to make them smaller such that they'd sell, they're rarely full transmission lines, but some 1/4 wave line, semi reflex or other fudge that ruins the LF response.

IMF made some more or less true TLs, and PMC still do in their larger Pro models, but as far as I'm aware, nobody else. (B&W's Nautilus is one, but don't know if they're still made). It's a very valid technique for DIY, although a lot harder to get right as success depends on getting the line stuffing correct such that all the back wave is absorbed, without creating enough back pressure to colour the sound from the driver. It has to be graduated getting increasingly absorbent away from the driver, such that it doesn't matter if the end of the line is open or closed, as there's nothing left to reflect back.

S
 
Whether you call it transmission-line or something else, what we are dealing with here are cabinets with some sort of acoustic loading using a channel that is long compared to the wavelength. And they can have benefits like increased output capacity, extended low-frequency range and better acoustic coupling.
Thank you for this but as you correctly said at the very end, the channel must be long enough. We expect good bass response from a large speaker. The wavelength of 30 Hz is 11 metres or 37 feet. Try fold that long a channel in a speaker that fits in a living room.
 
Thank you for this but as you correctly said at the very end, the channel must be long enough. We expect good bass response from a large speaker. The wavelength of 30 Hz is 11 metres or 37 feet. Try fold that long a channel in a speaker that fits in a living room.
That PMC Twenty.21 measures poorly, could it be because the channel is too short? Or maybe it's due to other factors? Do you have any idea about that?

image (1).jpeg


 
That PMC Twenty.21 measures poorly, could it be because the channel is too short? Or maybe it's due to other factors? Do you have any idea about that?

ATC specs say:

Effective ATL™ Length: 1.72m / 5.6ft
Frequency Response: 50Hz – 25kHz

The wavelength of 50Hz is 7m, which is four times longer than the transmission line used in that speaker. If a manufacturer that is a big advocate of transmission line flunks their own theory that much, and openly, I have no respect for them nor I think should anyone else.
 
ATC specs say:



The wavelength of 50Hz is 7m, which is four times longer than the transmission line used in that speaker. If a manufacturer that is a big advocate of transmission line flunks their own theory that much, and openly, I have no respect for them nor I think should anyone else.
Thanks for the answer Sarumbear.

Wondering why PMC has embarked on such an impossible projekt? They should have known that it was doomed to fail.
 
Wondering why PMC has embarked on such an impossible projekt? They should have known that it was doomed to fail.
Companies crave for a USP, even if it is not realistically achievable.
 
Thanks for the post, learned something today but still not understand fully. For ported version I can see why the low end gets boosted, but for closed version do we get any boost at the low end? Some people in this thread is talking about transmission line going infinitely long is the best, why would that be?
 
Some people in this thread is talking about transmission line going infinitely long is the best, why would that be?
Transmission Line is a way to block the rear sound of a driver while not pressure loading the box as in a closed box or accentuating some frequencies like in a vented enclosure. The minimum length is the wavelength of the minimum frequency it is expected to work at. There is no maximum for the TL length.
 
Thank you for this but as you correctly said at the very end, the channel must be long enough. We expect good bass response from a large speaker. The wavelength of 30 Hz is 11 metres or 37 feet. Try fold that long a channel in a speaker that fits in a living room.
The wavelength of 50Hz is 7m, which is four times longer than the transmission line used in that speaker. If a manufacturer that is a big advocate of transmission line flunks their own theory that much, and openly, I have no respect for them nor I think should anyone else.
AFAIK, the lines were designed to approximate 1/4 wavelengths.
 
AFAIK, the lines were designed to approximate 1/4 wavelengths.
That is when the port output is out of phase and interacts with the piston’s front output. A TL by definition is to achieve an infinite baffle. However, TL is such a messy design that there are infinite variations to it.
 
I was given to understand that the PMC cabs actually act as bass reflex designs, not transmission lines. Does anyone know whether that is true or false? Does anyone have measurements of the bass response and accompanying impedance charts for one (or more) of the PMC speakers? I'm curious. Jim Taylor
The company insists that it is a TL, and not a ported design.
 
Thank you for this but as you correctly said at the very end, the channel must be long enough. We expect good bass response from a large speaker. The wavelength of 30 Hz is 11 metres or 37 feet. Try fold that long a channel in a speaker that fits in a living room.
It does not need to be 11m, cut-off is typically at 1/4 wavelength. And how it is designed also affects the acoustic length - porting at the end with a reduced port area is one alternative.
 
  • Like
Reactions: 617
Thanks for the post, learned something today but still not understand fully. For ported version I can see why the low end gets boosted, but for closed version do we get any boost at the low end? Some people in this thread is talking about transmission line going infinitely long is the best, why would that be?
The idea is to make the box kind of invisible to the driver, so that it acts like there was no cabinet behind it. So the TL should then absorb all sound radiated from the back of the cone.
 
That is when the port output is out of phase and interacts with the piston’s front output. A TL by definition is to achieve an infinite baffle. However, TL is such a messy design that there are infinite variations to it.
Exactly. So the F2 I show in my example is really not a transmission-line, since there is sound radiated from the end.
 
Exactly. So the F2 I show in my example is really not a transmission-line, since there is sound radiated from the end.
Tell that to other posters who argue :)
 
I could insist that I'm young and handsome, too. That wouldn't make it true. Maybe they are a TL, maybe they aren't. I looked into Stereophile magazine:


quote: "Although PMC describes the fact.8's woofers as being loaded with an "Advanced Transmission Line," the solid trace in fig.1 doesn't look significantly different from the impedance of a conventional reflex loading, with a low-frequency saddle at what would be the tuning frequency of the port." and "As with a reflex design, the output of the port is in opposite polarity to that of the woofers, though the length of the line delays the vent output by several milliseconds."


quote: "In a true transmission line, the vent output is in-phase with that of the line-terminated woofer; by contrast, the output of the PMC DB1i's vent is in opposite polarity to that of the woofer, again typical of reflex behavior."

And a page from someone named Kalman Rubinson. I never met him, but he sounds like a bright kid. :)

I am not disputing what you say. I was being sarcastic as TL alignment is the company’s USP even though, as you correctly said, it is not a TL, nor an advanced one (ATL). It is simply a vented enclosure with a lossy port.
 
I could insist that I'm young and handsome, too. That wouldn't make it true. Maybe they are a TL, maybe they aren't. I looked into Stereophile magazine:



quote: "In a true transmission line, the vent output is in-phase with that of the line-terminated woofer; by contrast, the output of the PMC DB1i's vent is in opposite polarity to that of the woofer, again typical of reflex behavior."
In a true transmission line, there's no output from the vent. It doesn't matter whether the vent is there or not. That so-called TLs are vented, indicates they're probably not true TLs.
The only example I can think of, of a true TL without a vent is the B&W Nautilus, which has a line of decreasing cross section, ending in a dead end.


S
Nautilus.jpg
 
Back
Top Bottom