A new approach to point source?

[PristineSound]

This is an open discussion.

In recent years, there are many manufacturers who are using multiple tweeter sized drivers that surrounds a tweeter proper for midrange.

See examples below.

Unfortunately there are no polar maps for any of these speakers, with the exception of Perlisten which only uses two tweeter sized drivers for midrange, one above and one below the tweeter proper. The polar maps on the Perlisten appears very much point source like, see below.

Is this a new approach to point source? We know coaxial drivers can potentially suffer from intermodulation distortion and the few MEH's spin that I've seen aren't all that great.

So is this design a modern day alternative to point source?

Andrew Jones once commented on Facebook that using tweeter sized drivers for midrange does not provide sufficient surface area as compared to a proper midrange sized driver. But the Perlisten's spin and polar maps smashes this theory.


The Elac M807

The Monitor Audio Hyphn

Perlisten

Edit: adding Genelec 8381A

And of course Tekton, but their intentions aren't to emulate a point source. And there is enough evidence based on Eric Alexander's own actions and words that he doesn't even measure his speakers. So we can discount Tekton.

 

[Esprit]

8381A SAM™ Adaptive Point Source Main Monitor - Genelec.com

Our 8381A empowers engineers and creators to perceive every sonic detail, offering true-to-life sound level and dynamics – with the full flexibility of free-standing monitor placement and advanced room adaption.

www.genelec.com

 

[TNT]

A true ideal point source would be the site and form of a ping pong ball transducer able to do < 0,01% THD 10-50kHz at 120 dB 1m with a perfectly uniform spherical sound field - ie. measure the same in all direction.

The directivity would be like:

Many say "point source"... this is what it would be... any takers? No, I didn't think so.

But I would take one every day of the week...

//

 

[PristineSound]

8381A SAM™ Adaptive Point Source Main Monitor - Genelec.com

Our 8381A empowers engineers and creators to perceive every sonic detail, offering true-to-life sound level and dynamics – with the full flexibility of free-standing monitor placement and advanced room adaption.

www.genelec.com

Ah yes, I forgot about this Genelec as another example.

 

[PristineSound]

A true ideal point source would be the site and form of a ping pong ball transducer able to do < 0,01% THD 10-50kHz at 120 dB 1m with a perfectly uniform spherical sound field - ie. measure the same in all direction.

The directivity would be like:

View attachment 453190

Many say "point source"... this is what it would be... any takers? No, I didn't think so.

But I would take one every day of the week...

//

That seems like and omnidirectional speaker with perfect sound power.

But I get your point, it is physical not possible to have a text book definition of a point source, particularly bass is omnidirectional.

 

[Timcognito]

That is why infinite baffle has so many benefits, 2Pi amplitude, no baffle step, no front wall reflections. If kept away from corners, floors and ceilings, can be very convincing sound.

 

[thewas]

It should be said though that the Genelec 8381 Quad Midrange System has 127 mm(!) drivers and is used between 150 and 500 Hz so not really tweeters used as bass but rather midbass drivers used for lower mids.

 

[thewas]

One of the best such applications in my opinion is this DIY(!) one where an asymmetrical tweeter sized driver array is being used to form a wide horizontal and and narrow vertical radiation:

Source and more: https://www.diy-hifi-forum.eu/forum/showthread.php?17618-Quasikoax-1

 

[TNT]

That seems like and omnidirectional speaker with perfect sound power.

That's a point source, yes.

But I get your point, it is physical not possible to have a text book definition of a point source, particularly bass is omnidirectional.

Not sure what you mean/say... My statement above was a text book definition of a Point Source. Do you wanted to say that is not physically possible to implement one - that I would agree to. Bass it is the easiest one to make to a point source - almost tall are.

Or did I misunderstand you... sorry in that case...

Many call their design point source or just use the word... but there are very few, if any...

//

 

[JPA]

Andrew Jones once commented on Facebook that using tweeter sized drivers for midrange does not provide sufficient surface area as compared to a proper midrange sized driver. But the Perlisten's spin and polar maps smashes this theory.

You've misunderstood Jones' surface area comment. He wasn't referring to directivity, but to the inability of tweeter-sized drivers to play as loudly without distortion as larger midrange drivers.

 

[PristineSound]

You've misunderstood Jones' surface area comment. He wasn't referring to directivity, but to the inability of tweeter-sized drivers to play as loudly without distortion as larger midrange drivers.

Thanks for the clarification.

Although, the Perlisten proves to be very low in distortion, also qualifying for the THX Dominus certification.

 

[Timcognito]

There are BMR drivers with wide dispersion patterns especially in the mid and upper mid ranges.

 

[PristineSound]

That's a point source, yes.


Not sure what you mean/say... My statement above was a text book definition of a Point Source. Do you wanted to say that is not physically possible to implement one - that I would agree to. Bass it is the easiest one to make to a point source - almost tall are.

Or did I misunderstand you... sorry in that case...

Many call their design point source or just use the word... but there are very few, if any...

//

The way how I read the polar maps you shared, is that no matter where you stand encircling the speaker and no matter what frequency, the SPL is always constant. It doesn't seem like a point source.

Below polar map seems more consistent with a wide and controlled directivity point source, textbook and unachievable in the real world.

 

[PristineSound]

One of the best such applications in my opinion is this DIY(!) one where an asymmetrical tweeter sized driver array is being used to form a wide horizontal and and narrow vertical radiation:

View attachment 453194

View attachment 453195
View attachment 453196

Source and more: https://www.diy-hifi-forum.eu/forum/showthread.php?17618-Quasikoax-1

This guy needs to start a business selling these speakers.

 

[thewas]

This guy needs to start a business selling these speakers.

Yes, he has made more superb designs which most commercial companies would like to call their own:

http://loudspeaker-research.de/

 

[PristineSound]

Yes, he has made more superb designs which most commercial companies would like to call their own:

http://loudspeaker-research.de/

Very very interesting.

Especially this waveguide design, if you can even call this a waveguide.

 

[Arindal]

Is this a new approach to point source?

It is a very interesting approach, indeed. The primary goal seemingly is not only to get a point source (there are easier methods to do this), but to solve the problem of controlled directivity in the transitional band midrange/tweeter and in the freq band below that. Speaker designers run into a dilemma here as there are not many ways to get an identical dispersion pattern of the midrange´s highest octave and the tweeter´s lowest one. You either have a step up or down in directivity, which is a problem, or you have drivers/waveguides so big that the distance between them poses other problems: vertical localization, interference or other phenomena.

With a tweeter plus a midrange array, you can save this problem in the transitional area and neighboring bands, in theory.

We know coaxial drivers can potentially suffer from intermodulation distortion and the few MEH's spin that I've seen aren't all that great.

MEH?

For home application, I would not be overly worried of intermodulation distortion. The SPL needed is usually not as high (compared to P.A. applications) and there are measures to reduce the excursion of the midrange driver, like a 3-way concept with a x-over freq of 250...500Hz between woofer and midrange.

Combining a tweeter and an array of small midrange drivers surrounding it, does not really solve the x-over problem, as these midranges have to be small and you end up in a similar range of necessary x-over freq. That is the problem Andrew Jones was touching in his comment.

Most of conventional coaxial designs have a different problem as they shift the interference problem to a higher band than the transition. Depending on the design, you have a band in which the tweeter positioned in the center of a midrange would on axis produce soundwaves which meet reflected waves of +180deg phaseshift, hence a narrow-banded cancellation. There are measures to decouple tweeter and its central position (KEF, Beyma, Fyne), but they usually come at the cost of further increasing the tweeter´s directivity index.

Many say "point source"... this is what it would be... any takers?

There are different concepts of a point source in hi-fi, and commonly it is referred to a concept having the drivers producing localizable frequencies on the same axis or the frequencies originating from the same point. This is not the same as a pointsource referred to in acoustic experiments which has a diaphragm area of almost zero hence producing a perfectly spherical wavefront.

There are BMR drivers with wide dispersion patterns especially in the mid and upper mid ranges.

Certainly true, but BMRs are kind of the opposite concept to a midrange array. Their diaphragm breaks up in a controlled way (usually into two rings coupled to each other with something flexible) so in theory they can keep a very broad dispersion pattern while reaching very high in the frequency spectrum (that is why some people like them as quasi-fullrange drivers).

One of the best such applications in my opinion is this DIY(!) one where an asymmetrical tweeter sized driver array is being used to form a wide horizontal and and narrow vertical radiation:

That is a very promising concept, and in theory from directivity point of view close to an ideal solution. Reason being the dispersion pattern of the tweeter is matching the potential dispersion pattern of the midrange array, in theory.

In practice it remains to be listening to the result. Seemingly one of the main obstacles has been solved, the x-over freq and how the array behaves in this band and slightly below. In a listening test, I would focus on the other one which is the dynamic reserves and subjective sound quality of the tiny midrange drivers. I would be afraid they cannot go low enough w/o signs of compression. BTW Perlisten has seemingly solved this problem by employing two additional tweeter domes in a very narrow band. Just enough to keep directivity in the same ballpark for the transitional band between central tweeter and midwoofer array.

This direction is nevertheless a promising one. I am working on a completely different solution to bringing a large tweeter and a controlled dispersion midrange array together. The idea is to achieve a perfect pointsource plus constant directivity in the transitional band, but without the need of additional midwoofers, but either fullrange or to combine with a subwoofer so this whole thing should reach 90Hz. It is still in an early conceptual stage.

 

[Arindal]

Especially this waveguide design, if you can even call this a waveguide.

For P.A. applications, I know such concepts referred to as ´slot drivers´ or ´line drivers´ with a planar wave former, which are modestly popular for big line arrays. In this case, coming with a transition element between compression driver and slot providing identical distance between any point of the slot and the diaphragm.

They are capable of producing a perfectly even cylindrical wavefront which is a very interesting and promising concept. In theory, they can go as low as 800Hz or lower with a 1.75" compression driver. Which is making the combination with a midwoofer array much much easier.

 

[xanalog]

...Andrew Jones once commented on Facebook that using tweeter sized drivers for midrange does not provide sufficient surface area as compared to a proper midrange sized driver.

Andrew Jones is the creator of the three MoFi "SourcePoint" series.

...Unfortunately there are no polar maps for any of these speakers,

Erin review of SourcePoint10s provides 'contour' and 'globe' plots:

Are these the types of 'polar maps' you reference?
I wish more speaker measurements also provide for those waterfall plots, that I think are telling.

 

[Arindal]

Andrew Jones is the creator of the three MoFi "SourcePoint" series.

...and kind of a pioneer of coaxial speaker design. Before MoFi, he did really interesting pointsource designs for KEF, TAD Labs and Elac.