Sealed mini speaker, 3-way

[Heinrich]

There is a little box that nobody can believe in.

It's supposed to be a real shelf box, and therefore no deeper than 22 cm and no higher than 30 cm. At least that's how plus/minus European shelves are built. This means that a size of 30x30*22 cm^3 would still be pretty reasonable, resulting in a maximum internal volume of around 15 liters.

Furthermore, the frequency response should reach well down, 40Hz is the minimum. After a lot of back and forth, it is not possible to set up a bass reflex in this small volume. The port is always either too long or too narrow - resonances or wind noise. Passive membranes don't help either, because where to put them on the shelf, and they would also be much too heavy and, last but not least, expensive.

What remains is a sealed design with the largest possible cone and plenty of excursion. A two-way speaker won't work, because the excursion would lead to distortion of the midrange in any case, no matter how fancy the bass is.

So three ways with the lowest possible crossover frequency. This can practically only be done (digitally) electronically. Furthermore, the necessary bass correction can again only be achieved electronically.

This then leads almost inevitably to the design as shown. Three ways, bass on a separate amplifier with boost, bass / mid digitally separated, mid / high passively separated on a second amplifier. A quick calculation shows that 50 watts each will exploit the limits of the speaker fully. The digital separation and smoothing of the frequency response can be done via the PC / Equalizer APO / Peace and a 7.1 USB interface; 2 channels stereo to four channels speakers. With 24-bit digitization depth, volume control on the digital side is completely sufficient. (This ensures synchronization).

Since this is an experiment, fairly inexpensive drivers are sufficient. The bass allows +/-6mm travel. And the result is actually really good.

The key data: Cabinet made of 9mm poplar plywood, 21 x 30 x 21 cm^3, stiffened inside, approx. 8 liters, midrange 0.5 liters, drivers from Peerless and ??, x/overs at 300Hz and 2kHz. The bass is digitally tuned to a cut-off frequency of about 30Hz in the room.

I have a question: how can I prove that the choice of a three-way system is advantageous? I am not aware of any examples of a similar design.

 

[terryforsythe]

I have a question: how can I prove that the choice of a three-way system is advantageous? I am not aware of any examples of a similar design.

Harmonic distortion and compression measurements.

You can peruse speaker reviews herein and/or Erin's Audio Corner and see those measurements on many 2-way and 3-way speakers. EDIT: I don't recall whether all of Amir's measurements include the driver compression data, but I do know that most of Erin's do. But almost all of both of their reviews include harmonic distortion measurements.

 

[ZolaIII]

JBL Professional Control HST

Wide-Coverage Speaker with 5-1/4" LF, Dual Tweeters and HST Technology™

jblpro.com

5.25" with all the reinforcement he could get and 5 orders of magnitude more engineering.

 

[thewas]

I have a question: how can I prove that the choice of a three-way system is advantageous?

The superiority of a 3-way designs can be shown with multitone distortion measurements and also with smoother direcitivity measurements, although second can be compensated with the use of a appropriate tweeter waveguides. With good drivers like some Purifi things can get hard though if the 3-way competitor drivers are not of a high engineering level.

I am not aware of any examples of a similar design.

Hundreds of compact 3-ways in the 70s from almost all German loudspeaker brands and not only, for example original and new Wharfedale Super Denton

to the Neumann KH310 and its predecessors

I have several such in my small vintage loudspeaker collection, for example from 1979 with also a 7" woofer

 

[Salt]

https://www.audiosciencereview.com/...sive-drivers-and-components-mechano325.60789/

 

[Salt]

For passive closed with highpassed woofer might be a way...

 

[ZolaIII]

Worst part is that I actually favor close buffle design. When you properly relax and isolate (separate enclosure) mid range driver then it does it's part with extreme low distortion and no impact from refractions or direct impacts from those under it doing heavy lifting and 99% of true peeks. Now if you want my help to go 4 way properly provide documentation.

 

[DSJR]

Damn - I was sucked here by the thread title concerning those tizzy-toned screech boxes that 'audiophiles' still wet themselves over - another 'monitor' designed to do a particular job rather well, pleasurable reproduction of music at home not being it!


The UK view in the 70s was generally that it was bad enough to integrate two drivers well (the drive units weren't as good then), let alone three. I do appreciate things have come on a long way and of course, I believe that active operation with decent crossover slopes, better phase integration and dsp, also a greater understanding of integrating dispersion patterns properly, means a good three way should 'sound' better one hopes.

Back in 1974, I remember the smaller Wharfedale three ways were extremely coloured sounding to be honest and easily beaten by the then new two way plastic coned models coming along (the KEF Chorale being one of the best shelf mounted models I remember)..

 

[ZolaIII]

I can do it eyes closed without any documentation but I want you to choose 10" subwoofer's to a extreme Fs will be to 23 Hz but we are not here for that so something with nice response and still low Fs of 30~32. 7" will be stressed out to do main bass peaks and we would see what you choose for midrange one (mex I do need documentation).

 

[Salt]

Give us some curves ...

 

[Heinrich]

https://www.audiosciencereview.com/...sive-drivers-and-components-mechano325.60789/

How could I miss that?!

Damn - I was sucked here by the thread title concerning those tizzy-toned screech boxes... the smaller Wharfedale three ways were extremely coloured sounding to be honest and easily beaten by the then new two way plastic coned models coming along (the KEF Chorale being one of the best shelf mounted models I remember)..

Seems I can't change the title anymore. Otherwise I would, sorry. But you make a valuable point.

 

[ZolaIII]

If this whose a derail then it sure worked for me.

 

[Heinrich]

Give us some curves ...

Sure:

Low frequency demand, killer tracks for ported designs, except for Revel etc that tune to sub 40Hz (recorded in room at some arbitrary position, the little one's output):

Frequency response and distortion at same arbitrary position, stereo, 76dB at about 2m distance, watch room interference:

Intermodulation, see captions:

I didn't tell, the design was aimed at 86dB in-room, stereo peak level. And coincidently, with 86dB single tone at 33Hz, the 10% mark in harmonic distortion is reached. So, with pink (brown) noise the user (having very little interest in "perfect" stereo) will squeeze out like 96dB of full spectrum SOUND. That is well beyond any need, actually.

ps: I've got a little bit of rattling in the stands, which shows randomly in the graphs above.

 

[Heinrich]

An addition, a more speaker friendly mix, see how the spectrum is limited to what decent ported design could do, 40Hz namely:

A piece from the good old days, frankly, the sound is fuller, nearly "fat" despite 40Hz is 20dB down:

So, would you say that a ported design is superior to a closed, equalized speaker box?

What we didn't discuss is the subjective relevance of intermodulation. I personally think that two-way has fallen out of time. KEF mostly, and others did a good job in designing speaker by the book. They wrote the book. But today, well, with digital even the smallest decent speakers could be three-way, because the complication of a lower x/over has vanished.

 

[ZolaIII]

Nope mate, when you cross average 6.5" high at 120 Hz with 10" one cone excursion on it will be so low that separate midrange would only marginally improve things even to the main bass peaks which would still be kept on it. It's still 3 way when you look at it. To where you will have concentration of peeks and ELC boost especially where it's significant on lower SPL you use crossovers to limit transistance of their own fundamentals and keep ringing of it low as if it leeks it will ruin mids. Ported design can again done good as reinforcement and less to the extension and you get a bit higher max SPL (tho it still cost both box size, amp power and THD). And you get F box from port. Under port tuning things explode regarding THD increase and pretty much fall of a cliff.
When I asked data I meant direct drivers reference preferably in database with extended data and emu not in room response.

 

[Heinrich]

Nope mate, ...

I'm not quite sure I got your caveat right. How would you decide?

- ported as to decrease intermodulation (I personally don't think ported helps that much in that regard) and THD; but limit frequency response electronically so that no "explosion" occurs below the tuning frequency, which has to be a bit higher due to the very limited size of the speaker

- let the ported speaker run down into "explosion" (I like that wording, it is utterly correct)

- don't bother with small speakers, become the man you are

When I asked data I meant direct drivers reference preferably in database with extended data and emu not in room response.

O/k, you could hire me ...

 

[ZolaIII]

I already stayed that in the opening of 7th post but I can't be exclusive to all ported designs. For me the port is not worth with all of it's benefits to troubles it brings. However main advantage of separate subwoofer's from mains is physical placement different to mains. Basically push the sub's to the walls and mains into the room still keeping them on tight leesh so that you get reinforcement where you need it and better ratio of back to front refractions, that is if you are lucky enough to have such large and long room in the first place. If you are not it's close to the wall anyway and you can still improve it tuning the bandwidth dispersion (not really widening it more than a little bit but certainly narrowing it) to listening distance. In small room getting in generally good RT's is hard task but far from unachievable (well not dandy ones but still good).

 

[Heinrich]

...advantage of separate subwoofer's from mains is physical placement different to mains. ...

Admittedly, the room modes are a problem. This can also be seen in my measurements, a broad dip around 200Hz. A random position, on purpose, guys, reality! Because who wants to sit in a vice? It's about music, funny lala on the side, or important “state of the union”-type podcasts. Of course that makes you run.

But then the user also wonders about the share of the rent for the stuff. The “real estate” factor: no subs! The boxes as shown are pretty much the maximum for normal people. Otherwise it's bluetooth mono, practicability.

And still, I've got 30Hz

 

[a4eaudio]

...The boxes as shown are pretty much the maximum for normal people...

Can you tell us more about these "normal" people?

 

[Matt_Holland]

Back in 1974, I remember the smaller Wharfedale three ways were extremely coloured sounding to be honest and easily beaten by the then new two way plastic coned models coming along (the KEF Chorale being one of the best shelf mounted models I remember)..

I’ve got my dad’s KEF Chorale he bought in 1978 and are still in excellent condition, as I offered to help him sell them. They are best placed on low stands to get some floor gain and avoid floor bounce cancellation, but pulled away from the front wall they image terrifically and are very clean and quite linear sounding. Or they can be placed right up against the front wall if you need to position them higher up.

I temporarily replaced my Revel M16 with them and they are less refined, and lack bass in treble extension in comparison, but in isolation are great fun and look cool if you like retro.

[Star Wars balloons were for my son’s birthday]