That was in a response to my question about why the panels were mounted horizontally instead of vertically.
I assumed it was the adhesive that connects the paper skin to the honeycomb core.
But that's a guess.
Eric
I looked at both.
Nothing obviously great about the proportions. Not bad, but nothing special to me in terms of the distributions of modes.
The most preferred exciter location (1.62) appears to be extremely close to an NXT position, i.e 5/13-3/7.
Generally though it looks like a good design to me, but nothing obviously inventive, as far as I can see. The only thing I've never seen before is the special position near the corner for an HF exciter.
Light, stiff panel, using honeycomb core, and stiff light skins. Good for efficiency.
Stiff frame with foam around the entire perimeter. Good for full range, and smooth response (if enough damping).
Spine to support the exciter. Good to prevent sag.
Reasonable aspect ratio.
Reasonable exciter position.
While the description is pretty detailed in terms of construction, for me it lacks any technical explanation of "why". They refer to empirical results, but don't even show any of them. Too bad.
My favorite part (of the translation) is where they translate exciter as "pathogen".
Eric
Last edited:
Thanks for the English copy of that patent. I searched by the inventor's name, Danilo Herger, and maybe that's why I couldn't find it. The US patent is filed by Sotis AG, a patent holding and selling company in Switzerland. Both US patent state is pending, while priority claimed from the Russian patent.
Best not to trust the Google automatic translations. 🙂
------
The sound-emitting membrane (2) is made of a light and rigid material - a sandwich structure, including a honeycomb core, a surface layer glued to the honeycomb structure from both sides and a stabilizing impregnating composition based on polyurethane primers and varnishes covering the surface layers.
Such a membrane (2) begins to conduct traveling wave structures over the surface, formed by the drive of acoustic vibrations (3) applied to the membrane surface. Surface traveling waves, which have a finite propagation velocity in the membrane material, are repeatedly reflected from the edges of the membrane itself, forming resonance-conditioned, frequency-dependent modulations, zonal localized over the panel area. These modulations have one distinctive feature: they arise in the form of strictly opposite balanced oscillations within one indivisible sound-emitting membrane (2).
For ease of understanding, these opposite bending oscillations can be represented as a set of incoherent point acoustic emitters (speakers) strictly out of phase at the level of 180 degrees, see Fig. 3. This mode of operation of the proposed acoustic emitter is the main and necessary, since in modes that go beyond the limits of the resonant balanced formation Opposite modulations stop the process of effective generation of the audio signal as such, and the conditions necessary for the formation of the transverse component of the wave do not arise.
The advantage of our proposed technical solution in the form of a special line within the membrane area, involving the attachment of excitation sources within it, is to ensure the optimal distribution of resonant modulations within the membrane area, which in turn has a positive effect on the uniformity of the amplitude-frequency characteristic, as well as ensuring such parameter, as the naturalness of sound, closely related to the total amount of distortion introduced by the operation of the speaker system, the reduction of phase shifts, as well as ensuring the maximum frequency range in the operation of such a system.
In our acoustic device, no special measures are required to maintain the condition for the existence of a transverse sound wave. The resonant mode of operation of such a device itself assumes the constant presence of suitable conditions for the emergence and maintenance of a transverse wave. Moreover, these conditions exist as a continuous readiness for the radiation of a transverse wave in a gas at almost any frequency of the acoustic range, including wider limits in the region of low and high frequencies, if necessary. the only source of excitation fed by a single-channel power amplifier and apply the appropriate signal (for example, a sinusoidal, a certain frequency, or wideband ("pink noise", music content, etc.))
-------------------------
As in all patents, the inventors don't say everything, but these Russian registered patents are quite detailed. Maybe, the inventor, just can't claim, but has to prove the claim with a product or to a certain level of scientific evaluation. The Mescalito speakers had been produced since 5 years or so, but the patent is applied in June 2020, next in July on the transducer.
Last edited:
This thread is impressive and I congratulate every one !
After working hard on this type of loudspeaker, my final findings where that first of all it’s very difficult to derive a loudspeaker with a “neutral” sound, every material will give a special coloration and thus be very dependant of the material properties of the plate, such as density, rigidity and thickness.
Moreover the radiation pattern isn’t not usual like with a direct radiator and obviously very far from a wave-guided loudspeaker. It’s very difficult to obtain a detailled image and most often the soundstage is not very precise, as a consequence there isn't any sweet spot ! The listening area is larger with quasi the same amplitude everywhere…
And last but not least, the low and the high frequencies aren’t easily reproduced by only one panel, thus the medium range is the predilection working zone of this type of loudspeaker. The add of a sub and sometimes a tweeter results in a 3-way system with the known difficulties to tweak the whole thing !
I gave up with this transducer for all these reasons…
Anyway, I’ll be very happy if somebody could overcome the main issues I faced with DMLs !
After working hard on this type of loudspeaker, my final findings where that first of all it’s very difficult to derive a loudspeaker with a “neutral” sound, every material will give a special coloration and thus be very dependant of the material properties of the plate, such as density, rigidity and thickness.
Moreover the radiation pattern isn’t not usual like with a direct radiator and obviously very far from a wave-guided loudspeaker. It’s very difficult to obtain a detailled image and most often the soundstage is not very precise, as a consequence there isn't any sweet spot ! The listening area is larger with quasi the same amplitude everywhere…
And last but not least, the low and the high frequencies aren’t easily reproduced by only one panel, thus the medium range is the predilection working zone of this type of loudspeaker. The add of a sub and sometimes a tweeter results in a 3-way system with the known difficulties to tweak the whole thing !
I gave up with this transducer for all these reasons…
Anyway, I’ll be very happy if somebody could overcome the main issues I faced with DMLs !
Have you shown pictures and recordings or plots of your panels?
There are many reasons , for how and why colourations can creep into panel construction .
With a good panel you should have the sweet spot all over you room.
Very large panels in a small room might not have the same pin point accuracy as a small panel,because you are too close to a large panel, but the large panel will have a larger scale performance ,which I like very much.
To be honest i like listening to both large and small, the sound is very similar it is just the soundstage that changes, in my room .
I use a sub for both, except for the proplex ,unless I am playing something with heavy low frequencies below 40hz.
Steve.
There are multiple reasons for different FR's (Frequency Response "colouration") in DML Panels. But in general they are far more forgiving of poor FR than point-and-spit technology.
Even with a basic design, I cannot concur that imaging is not precise. But with a properly-designed panel and a good source recording, the reproduction is so convincing that one is almost tempted to reach out and touch the instruments. Transient response, dynamics, clarity and definition of instruments are far superior to pistonic systems.
Bass is dependent upon panel surface area.
HF is dependent upon panel stiffness and density. And, especially if you're not too concerned with high efficiency, then good HF is not impossible to achieve.
The problems with pistonic point-and-spit systems are multiple, and ruler-flat frequency responses are an unavoidable pre-requisite in order to hear different instruments especially in complex recordings. In box speakers, one is almost forced to have either the same FR as the master recording engineer used (mostly ruler-flat) or better, in order to define all of the instruments upon play-back. And if the box speaker FR is not ruler-flat, then colouration or masked instruments are the inevitable result. This is like looking at Rembrandt painting through different-coloured lenses and trying to decide which makes the painting look "nicer."
Add to this the problems with cross-overs, phase responses, complex amplifier loads, driver beaming, wall reflections, tiny sweet spots etc...
DML panels seem to side-step these issues: No cross-overs required. No impedance phase issues. No beaming. Smooth coverage. Reduced reflection problems. Reduced standing wave problems. Massive sweet spot...
I would say that pistonic drivers will go the way of the Dodo in a few years' time.
Since I never reached a good sounding result I never publish any od the DMLs I built. This is why I gave up several years ago… The reason I say that there isn’t any "sweet spot” and stereo reproduction is the first the issues I faced, is related to the physics of the DML radiation pattern and wave reproduction. It’s simply impossible to get a coherent phase from this technology.
I did recordings and mixing several years : the use of these panels isn’t an option for mastering purposes, at least again for me.
But they have other advantages as a consequence and one very interesting way to use them might be in WFS as demonstrated elsewhere (cf. studies from Delft University).
As much as I love (personal preference) these DML panels I will have to disagree that they are far superior to pistonic (conventional cone) systems. I do believe DML technology has the potential to compete with high end pistonic/conventional cone speakers. At this point in time though there are only a few DML designs that can actually compete with high end conventional cone systems. Conventional cone systems in terms of advancement in technology is way ahead. DML designs are way behind and need to catch up. Even if DML's eventually do catch up with modern conventional cone systems the conventional cone will never become obsolete as technology keeps advancing.
One can also have the best of both worlds and combine DML tech with Conventional cone driver tech in which most of us do if using a conventional cone powered subwoofer.
Like with everything there are pros and cons, DML's have there pros but also cons as they are far from perfect.
Not to brag but my panels sound pretty darn good though. lol
Last edited:
I’m very surprised that you obtained good stereo imaging, since this is a constant issue with the DMLs. Anyway, I’m happy that you finally overcome this issue and I wonder to know how you did it considering the radiation of these panels.
The radiation efficiency is less than half that of a direct-radiator (0.039% vs. 0.089% in controlled experiments) and this poor radiation efficiency below coincidence frequency is a physical constraint of flexible panels.
Since DMLs relies on resonant modes of the panel, non-linearities arises and this results in higher harmonics distorsion.
Several authors suggested DSPs as a good option to improve the audio quality of DMLs.
I'll keep watching the present thread to see if some good solution emerges from your work guys.
I agree with you that one cannot get good results using commercially available exciters, most of which are sort of party gadgets. But those, who are actually making the flat panels and marketing them are making their own 'exciters' to match what they try to achieve. If a company is selling flat panels, the cheapest being ~1200$/pair, they'd have to deliver what they promise. Some first produce speakers, get the market approval and then apply for patents. They usually are engineers and scientists, or employ such.
Could you please give a link those studies from Delft Uni? The Ammos was produced at the Delft Uni by a student group that won the James Dyson Award.
On post 390 LineArray shows an early bending transducer.
It is sad that after all this time we are still talking about the same problems trying to build good sounding panels.
I have built many panels, of different sizes, shapes and materials.
Most of them I would regard as being equal to cone, electrostatic, and planar and probably better.
Definitely better when it comes to soundstage presentation and dynamics.
I think NXT thought they were going to take the world by storm, it is a pity they made such a mess of it.
Steve.
https://www.diyaudio.com/community/threads/piezo-nxt-type-panel.133711/page-20
It is sad that after all this time we are still talking about the same problems trying to build good sounding panels.
I have built many panels, of different sizes, shapes and materials.
Most of them I would regard as being equal to cone, electrostatic, and planar and probably better.
Definitely better when it comes to soundstage presentation and dynamics.
I think NXT thought they were going to take the world by storm, it is a pity they made such a mess of it.
Steve.
https://www.diyaudio.com/community/threads/piezo-nxt-type-panel.133711/page-20
Sticking a commercially available, mostly low grade, exciters on to a randomly found sheet is not building panels, sorry.