The Labyrinth Speaker DIY Project

[Deleted]

From: Webring.org  (click 'Labyrinth Speaker link on left of 1st page)

By Masaaki Takenaka, 2011

The Labyrinth Speaker







I have built a unique speaker named "labyrinth" .
This is an extension of the "angular spiral" concept devised by myself.

Anticipated advantages of the labyrinth are :

(1) Elongated sound passages.
(2) Prevention of standing wave with slanted fins.
(3) Decrease of high frequency by repeated cycles of expansion and reduction of cross-section area of sound passages.
(4) Suppression of unwanted vibration by reinforcing the cabinet with fins.

The sound is very good with a clear and strong low frequency region.
I think the labyrinth speaker may not belong to TLS because it dose not use damping material and does not resemble one in impedance characteristics.
My previous home pages are discarded because this labyrinth speaker is too good.


8 cm full range: SA/F80AMG

Building


1) All 22 fins have the same shape with 30 degree slanted cut end. Only 15 mm MDF is used for whole cabinet.



2) At the start, divider plate is bonded at the center of the side plate.



3) Next, fins are bonded on the two side plates.



4) Bottom view.



5) front view.



6) upper end.

[hemster]

The link above doesn't work for me.

I'd like to learn how he can get that performance out of a single 8cm full-range blower.

[Deleted]

Apr 20, 2014 23:42:36 GMT -5 hemster said:

The link above doesn't work for me.

I'd like to learn how he can get that performance out of a single 8cm full-range blower.

Try that link now...

Yes, so would I, but then that's why I've posted the design....maybe a better DIY person than myself can test it all out...

[hemster]

Link works now.

The design seems to compromise his premise of "Elongated sound passages". Based on the design, I really can't see how this speaker would sound good across the normal audio frequencies. I'll admit it is braced very well but that's not the only tenet of good loudspeaker design. Seems overkill on the bracing. Normally one just needs to stop the cabinet from flexing and doesn't need concrete-pillar solidity.

A single 8cm driver won't cut it IMHO. For a start, there'll be nothing in the mid-upper frequencies. Just my 2 cents.

[hemster]

Just noticed their mantra is:

The Ultra-Fi webring is dedicated to unique, deeply personal, extreme, and even bizarre expressions of audio reproduction. Special emphasis is placed on DIY and homebrew vacuum tube amplifiers, speakers, and other equipment...

I guess the design doesn't appeal to me as 'deeply personal' but sure is a bizarre expression of carpentry!
Not that there's anything wrong with that.    

 

[KeithL]

Generally, in a labyrinth design, the idea is that the overall length of the labyrinth (tunnel) works out to be half the wavelength of the lowest frequency you expect it to produce. This way the sound coming from the back of the speaker (which is 180 degrees out of phase from the front wave) is shifted another 180 degrees by the time it gets to the port - and so it  adds to the front wave and reinforces it. This doubles the output from the front of the driver at that frequency, and the cabinet should be designed such that it provides damping for the driver as well. (Usually transmission lines include stuffing in the tunnel - which increases its "effective length" and adds damping. There is also an intent that higher frequencies shouldn't "make it out of the tunnel". Even without stuffing, the author says his does pretty well in that regard, which makes sense since, while bass travels through the tunnel as a pressure wave, higher frequencies would have to ricochet back and forth until they find their way out.

You do get about 3 dB of "boost" at the tuning frequency by doing this, but the biggest advantage is supposed to be cleaner, more well damped sound (very sharp; very little ringing). As for maximum capacity, you will get about 3 dB more than the driver alone can do, which is not going to be much low bass with a 4" driver (think "punchy"; not "thunderous"; putting the vent opening in a corner or on the floor may help there - this author did not.).

You're also going to get some high-frequency roll off - because the driver isn't a tweeter. (And, even though some tiny drivers can be EQed to go pretty high, you usually get all sorts of distortion and beaming effects when you try to make a midrange work as a tweeter. You've also got the fact that the cone is moving quite a bit to make low frequencies - and that moving cone is trying to make high frequencies at the same time - we call that MODULATION - as in "intermodulation distortion"; you're basically moving the tweeter back and forth with the bass. )

For anyone looking to try a DIY labyrinth or TL design, but dubious of their woodworking skills, you should note that several (supposedly successful) designs have been made using 4" (or larger) PVC drain pipe for the "cabinet". It's strong, rather inert, easy to work with, and you can make any shape you want by combining elbows and corners. I've seen ones that zig-zag, ones that spiral, and long straight ones..... (you can even make a long one and snip off - or add - length to tune it)

Apr 21, 2014 5:11:32 GMT -5 hemster said:

Link works now.

The design seems to compromise his premise of "Elongated sound passages". Based on the design, I really can't see how this speaker would sound good across the normal audio frequencies. I'll admit it is braced very well but that's not the only tenet of good loudspeaker design. Seems overkill on the bracing. Normally one just needs to stop the cabinet from flexing and doesn't need concrete-pillar solidity.

A single 8cm driver won't cut it IMHO. For a start, there'll be nothing in the mid-upper frequencies. Just my 2 cents.

[DYohn]

And just FYI, a tuned labyrinth transmission line is the same theory Bose uses for their speakers.

[stiehl11]

Apr 21, 2014 10:12:05 GMT -5 DYohn said:

And just FYI, a tuned labyrinth transmission line is the same theory Bose uses for their speakers.

I was thinking the same thing (and we know about how everyone raves about Bose speakers on this forum). I might be all wet, but the graphs see to show a significant roll off below 100Hz and after ~12k Hz.

[Boomzilla]

The chance that the single 4" driver can produce the extended treble frequency response shown is highly unlikely. The chance that this speaker can play at anything above "background level" is equally unlikely. This is a prime example of measurements taken at an amplitude where the design measures best rather than how it will have to perform in the real world, In short, the design, the "frequency response," and the other measurements are fraudulent.

There is no free lunch.

[DYohn]

Apr 21, 2014 11:27:36 GMT -5 stiehl11 said:

Apr 21, 2014 10:12:05 GMT -5 DYohn said:

And just FYI, a tuned labyrinth transmission line is the same theory Bose uses for their speakers.

I was thinking the same thing (and we know about how everyone raves about Bose speakers on this forum). I might be all wet, but the graphs see to show a significant roll off below 100Hz and after ~12k Hz.

That's due to the inadequacies of the driver.

[DYohn]

FYI, here's an image from Bose's latest patent filing showing how they use the same sort of labyrinth design, with their tuned chambers, as a waveguide. Note that they are using six small drivers in an isobaric alignment to achieve sufficient dbSPL.

[stiehl11]

Apr 21, 2014 11:53:44 GMT -5 DYohn said:

Apr 21, 2014 11:27:36 GMT -5 stiehl11 said:

I was thinking the same thing (and we know about how everyone raves about Bose speakers on this forum). I might be all wet, but the graphs see to show a significant roll off below 100Hz and after ~12k Hz.

That's due to the inadequacies of the driver.

Maybe if the builder went to AutoZone and got a 2-way Kraco speaker it would help?

[lsdeep]

beside the driver size, the fin design advantages compared to a 'normal' transmissionline are rather not clear to me. i own transmissionlines and i am happy with them (link to the diy version of my speakers). the fin design seemed to be not worth consideration bei newtronics.

cheers, L

[kazoo]

The best way to see how they sound is to build them. It looks like a fun and easy build. Don't under estimate smaller drivers. I heard a little 3" in a back loaded horn that amazed me on how much bass it produced, to the point I was looking for a sub. It was a speaker my uncle made and got me started on DIY speakers. A lot of great designs out there. If it sucks, sell the drivers and use the boxes as fire wood.

[KeithL]

A small driver in a horn can indeed deliver more bass than you would expect - but there's a reason why that works - and why that DOESN'T happen with a transmission line.

One of the reasons that typical "piston" loudspeakers aren't very efficient (and that includes cones, domes, and even planars) is that they are a poor impedance match for the air they are intended to move. In other words, if you compare the weight of even a very light 8" cone to the weight of the air it is moving, you will find that the cone weighs much more, and so most of the energy you pump into that cone is "wasted" waving the cone around. Compare trying to "push" air with the palm of your hand to pushing water in a bathtub; the water is a much better impedance match. (The technical term is that "the energy isn't very well coupled to the air".) Even electrostatic speakers, where the moving parts are very light, aren't perfect in this respect - because, compared to what air weighs, nothing mechanical is "very light". This "effect" is most significant at lower frequencies, where a lot of air has to be moved to make high SPL's.

A horn offsets this effect by acting as a "transformer" for air pressure. The tiny driver is housed such that it is "working" against higher air pressure, which improves its coupling to that air. The horn then "transforms" a small amount of air moving at high pressure into a much larger volume of air moving at standard "room pressure". Without going into too much detail, a 3" driver mounted in the throat of a horn with a 24" diameter mouth is roughly equivalent to a 24" diameter driver - but one whose "moving mass" is still that of the 3" driver that powers it. (If you look inside a huge metal "PA horn", you'll usually find a tiny little 1" or 1-1/2" dome, designed out of very strong material, and driven by a relatively powerful voice coil -designed to operate against lots of air pressure.)  This is why horns are usually so efficient. (The technical way to say it is that the horn acts as an impedance matching transformer between the driver and the air, and so improves the coupling between the driver and the air - a lot.)

My point here, though, is that this ONLY works for horns because they are tapered. Even though the reality always falls a bit short of the theory, a 3" driver in an enormous well designed horn enclosure really CAN make as much bass as a large woofer. (Since a normal piston driver only couples a few percent of its energy to the air at low bass frequencies, there's a lot of room for improvement by putting it in a horn enclosure). The drawback to horns, other than the difficulty in getting them to deliver clean, flat frequency response, is that the size you need to build the horn is directly related to the lowest frequency you want it to deliver. The size of the mouth is related to the wavelength of the lowest frequency you expect it to produce, and there are mathematical requirements for how it tapers along the way.

For example, a horn with a massively powerful 3" driver could deliver loud bass down to 50 Hz - if it was well designed - but it would require a cabinet the size of a refrigerator. If you wanted to make it a sub, and extend that response down to 20 Hz, then that cabinet would have to be the size of an SUV. Of course, in the real world, there are all sorts of possible trade-offs here. Use an 8" driver, and one that can move a bit of air, and you don't need quite so much "gain" from the horn - which lets you do it in a slightly smaller cabinet... etc, etc, etc.

In contrast, a TL or tuned reflex cabinet, by "flipping the back wave and using it" can only get about a 2:1 benefit. (So a 3" driver in a TL cabinet isn't going to make any more bass than a 5" driver in a regular cabinet - at best.) Of course, you can get a bit more by intelligently using room boundary gain (for example, by putting the port near the floor, or in a corner), but a TL with a tiny driver just isn't going to make lots of bass. The real benefit of a TL design is that it controls the driver very well, which results in tight, clean bass (at least hopefully)... and you get a little extra bass because of the port... but it's way different than a horn. A typical TL design features a "labyrinth" that isn't tapered, or that tapers to a smaller opening (rather than larger; like a horn), and the length required is related to half the wavelength of the lowest note you expect it to produce. (So, for that design with the 3" driver, you need a tunnel that's several square inches in cross section, and quite a few feet long. To make an equivalent speaker that goes down to the same lowest frequency, but uses an 8" driver, you would need a tunnel the same length, but with a cross section about 5 TIMES as large. (So, no matter how you arrange the zigs and zags, the box would have to be 5x as big as well.) This is why TLs that go low are always very large. (And why they're a lot more practical if you're willing to limit their low frequency response and use them with a sub.)

Apr 24, 2014 7:36:58 GMT -5 kazoo said:

The best way to see how they sound is to build them. It looks like a fun and easy build. Don't under estimate smaller drivers. I heard a little 3" in a back loaded horn that amazed me on how much bass it produced, to the point I was looking for a sub. It was a speaker my uncle made and got me started on DIY speakers. A lot of great designs out there. If it sucks, sell the drivers and use the boxes as fire wood.

[KeithL]

Incidentally, one of the fun things about a TL speaker is that you can change the way it sounds by how you stuff the labyrinth. (The design you showed had an unstuffed labyrinth, but most designs stuff it to some degree, with the intent of making very sure that nothing much above the tuned frequency "makes it out the end". The real purpose of the tunnel is to act more or less like an infinite baffle at higher frequencies, while still allowing some bass at near the tuning frequency out to reinforce the response. The tunnel can be stuffed with very dense material, or none at all; the sides can be lined with felt, or not; and many designs even "taper" the stuffing density - by stuffing it tighter towards one end, and more loosely towards the other. This is one reason why you want to design your cabinet so it can be opened up - so you can fiddle with the stuffing )

[kazoo]

Keith,
Thanks for that explanation, learn something new every day. I made a few BLH but not really a TL, I don't think. I was just saying without hearing it you will only guess on how it sounds. Also don't know if I missed it but not sure what driver that is. They are quite different in specs between manufacturers. I have heard a few designs that don't follow the rules and sound better than they should. The fun part of this hobby is building it and seeing what happens.

[ocezam]

Apr 21, 2014 11:30:14 GMT -5 Boomzilla said:

The chance that the single 4" driver can produce the extended treble frequency response shown is highly unlikely. The chance that this speaker can play at anything above "background level" is equally unlikely. This is a prime example of measurements taken at an amplitude where the design measures best rather than how it will have to perform in the real world, In short, the design, the "frequency response," and the other measurements are fraudulent.

There is no free lunch.

Perfectly stated.

[Gary Cook]

Tip........it's not a 4 inch driver.
1 inch = 2.54 centimetres.
8 centimetres = 3.1 inches.

The speaker is a SA/F80AMG which has a stated frequency Range of 90Hz~22KHz it has a magnesium alloy cone, is rated at 8 ohms, 84.3 db efficiency, with 1.96 Kgs MMS. Specs can be found here www3.ocn.ne.jp/~hanbei/eng-f80spec.html

Cheers
Gary

[meldhache]

I beleive way back kef use to provide kits for transmission Line enclousure that were huge . It used a 9 by 15 inch flat piston woofer, a matched 6 inch mid and 1 inch tweeter.
I built such a monster about 40 years ago and my neveux still has it in his home.
They went down to almost 20 Hz at the time. They had a huge front opening 9X15 inch bass port and they were about 2 x 2 x 6 feet high and of about 150 + pounds.

Mel