All New Metronomes.

[Scottmoose]

Looks like you were in this game long before me ! ( although my dad and I built a pair of Heathkit speakers in 1978 ) . Have you got a link to what an acoustic labrynth design looks like then ? - is it a way of making a resistive port ?

Yes, you can look at them that way. Plenty of similar analogies in audio.

[MJK]

I did find speakers very difficult to understand at first, perhaps this was caused by the various uses of the terminology by different 'camps' in speaker design ( bass reflex / horns/ ML TLs ).

There are way to many names floating around. Some are pure marketing and others are just people trying to stake a claim to a particular enclosure or just apply a name due to a lack of understanding. The air in the box does not care about what it is called, it follows the laws of physics. In my opinion, all enclosures reside on a spectrum of box designs that gradually transition in behavior. There is no solid black line that is crossed going from one style of enclosure or naming convention to another. The way I look at the enclosure family is as follows.

1. Acoustically Small Boxes - A sealed or open volume enclosure added to the back (or front I guess) of a driver that acts like an air spring or a mass respectively. The dimensions of the box are small enough so that the air spring rate or mass are only functions of volume. The acoustically small box does not add any additional resonances to the speaker system.

2. Acoustically Large Boxes - These boxes can be sealed, ported, or open ended but the common factor is that standing acoustic waves exist inside the box and these resonances interact with the driver and modify the system's SPL response. You need to account for the standing waves in the design if you want to accurately simulate/predict the speaker's electrical or SPL response. I refer to both closed or open ended boxes as TLs, quarter wave modeling must be used to accurately design the geometry (two quarter waves form a half wave in a closed box). The boxes can have tapered, straight. or expanding cross-sectional areas as a function of the long direction. In reality, to be very accurate you need to model the standing waves in three orthogonal directions but at low frequencies just considering the longest dimension is usually adequate. A mass load can be imposed on the standing wave using an acoustically small port. The key here is resonances inside the volume impact the electrical and SPL response. Many different named enclosures fall under this category but the physics of them are all exactly the same and the transition from one terminology to another is gradual. Hence I consider bass reflex, classic TL's, ML TL's, TQWT's, Voigt pipes, almost all back loaded horns, tapped horns, and anything else relying on standing wave resonances to fall under the general category of TL's. Understanding and accounting for the physics is more important than a long string of buzz words used for naming.

3. Horns - A horn is an open ended length where the opening is large enough to damp standing waves and increase the efficiency of the driver output coupling into the room. These are non-resonant enclosures, one peak on the electrical impedance curve, producing a broad frequency band of efficient SPL output. The regime that they operate is frequency dependent based on the mouth size. An enclosure can act as a TL at low frequencies (mouth too small) and gradually transition to horn behavior at higher frequencies (mouth now big enough). Sizing these enclosures is tricky and if using rules of thumb or standard design equations you have to know the assumptions or conditions for which they are applicable (very common mistake).

So my path through enclosure definitions has simplified over the years to be as follows.

non resonant enclosure (small boxes) ----> lots of resonances impacting SPL response (TL's) ----> non resonant enclosure (horns)

The coining of all of these ever expanding new naming conventions is overly complex, confusing, and to me demonstrates a lack of understanding of the big picture of what is happening acoustically in the air volume inside the box. It tends to be a marketing ploy in the DIY world to carve out an intellectual advantage over another "typically less experienced" speaker builder, an elitist mode of operation in the forums. These definitions cannot be rigidly applied since there is a large gray area between names and a lot of mental masturbation is taking place classifying, defending, and debating what exactly falls under each name. For me, when I see a discussion using these exotic names I almost immediately conclude that the participants do not really understand the physics, they have not done the math themselves, and are blindly applying observations as fact.

My opinions, logic, and explanation.

[Scottmoose]

As far as pipes go, GM and I have a somewhat different classification, which came from his visits to Altec et al and which their engineers used as convenient practical handles:

Expanding pipe = horn
Untapered pipe = labyrinth or TL
Contracting pipe = Tapered pipe / tapered labyrinth

Speaking for myself, I generally take the trouble to establish how somebody is using a term rather than simply assuming them to be ignorant. To each their own of course. GM and I use the above quite happily in the same way, make no apologies for doing so, and will not stop.

[MJK]

As far as pipes go, GM and I have a somewhat different classification, which came from his visits to Altec et al and which their engineers used as convenient practical handles:

Expanding pipe = horn
Untapered pipe = labyrinth or TL
Contracting pipe = Tapered pipe / tapered labyrinth

Speaking for myself, I generally take the trouble to establish how somebody is using a term rather than simply assuming them to be ignorant. To each their own of course. GM and I use the above quite happily in the same way, make no apologies for doing so, and will not stop.

So a straight untapered pipe is a Labyrinth or TL. But if it contracts uniformly by a total of 1% along the length it is now a tapered labyrinth. If that pipe expands uniformly by a total of 1% along the length it is no longer a Labyrinth or TL but is now a horn. Why is it not an expanding TL or Labyrinth?

The acoustic behavior of all three pipes (measured and/or simulated) will be for all practical purposes identical. Your jargon makes no sense, it is not based on technical performance differences. Do you really believe there are significant enough differences in the three pipes I propose to assign them different names/labels which in effect indicates different performance? I don't, these definitions demonstrates ignorance of the physics.

[IslandPink]

I think I'm with Martin on this - the posting from 3pm goes a long way to clearing the confusion in my mind about the operation and significance of the difference sorts of loading. Logically I'd have to accept that if a Fostex BLH is not operating as a horn below eg. 150Hz due to the small mouth, then Danley's tapped horns are better defined as dual-input transmission lines, at least over most of their range. I hadn't thought of it this way before, but following my experience of building the longer folded horn and hearing the effects of having the length ( 1/2 wave at 100Hz ) and mouth size adequate to be horn-loaded to 100Hz , I can understand that one has to be carefully about claiming horn-loaded behaviour from anything shorter or smaller.

[MJK]

... but following my experience of building the longer folded horn and hearing the effects of having the length ( 1/2 wave at 100Hz ) and mouth size adequate to be horn-loaded to 100Hz , I can understand that one has to be carefully about claiming horn-loaded behaviour from anything shorter or smaller.

One point, there is no half wave action in a horn. One end closed and the other open enclosures (TL's and horns) are all based on quarter wavelength solutions. But because of the area expansion the length of the quarter wave is much longer than the simple L = c /(4 x f) equation would predict, that equation is only valid for straight geometries.

[rowuk]

Yes, we are living in a world where are words are stolen, complex situations are simplified so that parakeets can repeat them and the mark of quality is the amount of hits that a post or opinion gets.

After using the MathCAD worksheets, it became apparent to me that there were many common denominators in the "boxes" that we build. The most common was most designers quoting Thiele-Small but only to define system behaviour at the LF cutoff. Those that design horns and TLs seem to pay more attention to what happens above the cutoff/lower knee. Many other box types would benefit from a more detailed look and listen.

The idea of non resonant small boxes and resonant big boxes has an additional appeal if we know how much resonance actually passes through the speaker cone. My experience is that many drivers are very "leaky" in this respect.

[Cressy Snr]

I designed my Metronomes in the first instance as an expanding quarter-wave thing using Voigt, to which mass-loading was later added.
The calculations made with the design that is/was the subject of this thread, were all done with pencil and paper, using established theory, on the basis that the Met is a quadratic expanding quarter wave-transmission line. The sound with the drivers in the test cabinets my son has, which were an earlier version of the same hand-worked-out design is very, very good. So I'm confident that a QQWTL is what the design is.

I'm a big fan of simple, in that if you are going to design/build a speaker, you have to be fairly sure that you are actually designing what you think you are designing, otherwise you are stuffed before you start. For me I thought I had designed a QQWTL, it gave the predicted results when tested in a cabinet that was "close enough" so yes, I have designed what I thought I was designing.

The big question for me is now, has to be; am I parakeet, macaw or African grey? (squawk)

I'll get me coat.

[MJK]

The idea of non resonant small boxes and resonant big boxes has an additional appeal if we know how much resonance actually passes through the speaker cone. My experience is that many drivers are very "leaky" in this respect.

I do not believe arbitrary sound waves pass through a speaker's cone.

I do believe internal standing wave resonances exert pressure on the back of the driver's cone causing peaks and dips in a speaker's SPL response at discrete frequencies. This is predictable and can be factored into design simulations. You can see it in near field measurements.

[cressy]

Who cares about the semantics?
Just build the bloody things and see what they sound like.
Or send me some wood and I'll build them and you lot can argue about the relative merits of the multiple possible definitions of what they might or might not potentially be classified as as defined by the potential expansion of the line as defined by the calculations based on the correct or incorrect assumption that they might behave in a way not necessarily dependent on the definition of what classification they fit into 3 days ago and wether it was a Monday or a thursday when pen hit paper for the first time they were drawn up.

It doesnt really matter

[Cressy Snr]

It does matter Anthony. I like to see these discussions.
I came out of left field with this design in 2006 and I know what it is
Martin et al are arguing about what is a horn and what isnt, not really about the Mets themselves.
I think Mark is getting into a way of thinking about his own design, that he wouldn't have got into but for this discussion.
I think the original subject of the thread is irrelevant at the moment, but I'm not bothered.

We'll build them, get them veneered and I'll enjoy a great sounding, great looking pair of speakers.
No-one who has built a set has ever complained about the way they sound. That's good enough for me.

The build pics will follow when we get started. It'll be a Christmas project. Better get the inductors/caps and the veneer ordered pretty sharpish.

[Nick]

Who cares about the semantics?

Anyone who hopes to have a conversation that has more meaning that just hearing their own voice. Unless there is prior agreement on the meaning of terms, then at best a conversation is a happy way of spending some tine, the chance of exchanging meaningful information is zero, or at least the other party getting the same meaning from the words is zero. Thats why mathematics matters, as it provides a way of describing reality that can be communicated in such a way that is hopefully devoid of muddle.

It all depends on the goal. Some people get pleasure from building stuff and hearing the result, others want to have a deeper understanding of the processes involved. Both goals have merit, but the two methods are not equivalent or at least exchangeable.

[cressy]

You're exactly right nick
but if we presume that things may not be right first time, and that at this stage the description is largely irrelevant, we need a prototype to measure.
If it is supposed to behave in one manner and behaves in another, then the description becomes relevant as the design can be altered based on the measurements and wether it behaves in a manner predicted by calling it an abc or an xyz.
Until there is a physical 'thing', to me it is all theory and semantics as what the speaker has been described as, is predicted to operate in a very similar way by whichever terminology is used
It seems pointless to debate this until there is an actual thing to debate about.

[IslandPink]

Well no, not at all, because speakers are very messy, time-consuming things to build. I wouldn't want to launch into any speaker construction without having some plan of what I wanted out of it and how the next build might achieve that. Hence I'm going to go to some effort to design and/or simulate that first, and discussions may form part of the plan. Then we need to know what the words mean, so we are not talking at cross-purposes.

[cressy]

But if person a says a term means x and person b says it means y, who is right?
If there isn't a consensus, but there are compelling arguments for either, where do you go from there?