Sorry to disappoint you lekha, but no, I have not heard what you are trying to say. Much too vague for me.
Simply not true at all. You obviously have not read through the majority of this thread.
Good for them! We all would like to see more commercial DML speakers. But you seem to forget that this is a diy forum. Most members here have limited means of developing new stuff like exciters or BMR drivers or even composite panels.
Discover what Dr. Graham Banks has written and spoken about, and you will gain insight into what you are trying to achieve. For those living in England, they could visit Cambridge and search the libraries for written materials. Reading what Henry Azima has partially said won't suffice, as it was not his work. His brother Farad, the millionaire, simply placed him in that position, as Farad was financing the research at the time. Azima had just come out of the Iranian navy. As I said, good luck!
People like Graham Banks often take their knowledge to the grave, leaving little for future generations, so you may not find many texts or even videos from him. However, here is a video featuring one of the first physicists, Timothy Whitwell, who worked on the team when Verity Group was commissioned to turn this (DM) into a commercial product. He was one of the leading figures at Tectonics later. It’s a lengthy video, but it’s worth watching.
Christian,
Yes, if you are looking for constructive interference from two signals that are out of phase you would need a half wavelength delay between them. But for a peak to happen from the 1,3 mode, or other odd,odd mode, you don't need such a delay. The pressures oscillations caused by the shaded regions cancel each other out, leaving the two unshaded regions to radiate in-phase, and constructively interfere with each other (i.e. sum together) with no delay. Or is there something I am missing?
Eric
.
My reflection was oriented by a model I used just before for a dipole made from 2 distant point sources. So I was thinking in words of small sources at different locations. In the sketch above, I think there is an implicit condition which is the listening position is far away the plate so compare to the wavelength (in the air), there is not enough path differences between the different areas to create additional phase shift or level change. In the listening position is closer to the plate, some path differences occur but not enough in my opinion to have an effect in the 500Hz range.Or is there something I am missing?
Eric
.
Christian
Now that you've listened to Timothy Whitwell, one of the former development and research team members at NXT, you may have noticed that Tectonic had shifted away from creating DMLs and had progressed even further by adopting a distributed mode, and this was well before 2020. You might have heard how the entire saga began and how it developed thereafter. What was not mentioned is how the technology relocated to the US and how many of the former team members accompanied it. It was also not noted that when the technology moved to the US, the British Defence Ministry had to grant permission for this. Therefore, it can be safely assumed that some of that technology did not make the transition. The British undoubtedly had to sign some non-disclosure agreements before the move to the US.
(Henry Azima was not mentioned at all. Once he was ousted, he was not permitted to continue with the technology or to start another company. He subsequently moved to Canada.)
No further patents were released later, around the year 2000. There was also not much progress with BMRs. The BMRs can be designed to operate safely while achieving the intended sound quality at a smaller size, roughly around a radius of 4.2 to 4.5 inches. Attempts were made to create the standard 6.5-inch model, but these were unsuccessful.
It appears that Tectonic handed over their renowned DML500 and other similar projects to Denny Mayer, who is now the CEO of Flat Panel Audio, while they shifted their focus to BMRs. As mentioned earlier, the issues with the Balanced Modal method arise when the radius exceeds 4.2 inches. In any case, any developments in this area have not been published, and no further patents have been released—most likely, some of the findings are classified in some manner.
The available BMR sizes can be quite easily utilised in today's consumer products, such as mobile Bluetooth speakers, computer speakers, earphones, and even high-end speakers (like those from Mission and Cambridge Audio), particularly by eliminating the tweeter in some manner.
On the other side of the world, some (still DIY) researchers are attempting to implement BMR technology in somewhat larger panels—small speakers are difficult to sell at exorbitant prices there—and are experiencing a degree of success, despite being overlooked by audio bloggers. They are encouraged by the "import substitution programme" declared by their government. Whether they will achieve their goals remains to be seen. It’s akin to saying, "Buy our panels, and also purchase our amplifiers to achieve the best sound from the flat panel speakers"—Linkwitz's miniDSP comes to mind. The only caveat is that you have to manufacture the speaker drivers (exciters ?) yourself; they do.
Now, if you listen carefully to Timothy Whitwell, Vice President of Engineering at Tectonic Audio Labs, in that video, you'll notice that it revolves around how to use the "standard" voice coil to transfer vibrations to the suspended flat panel. This system is easy to produce and predictable, for them at least. You can utilise two voice coils instead of one (in your DML), with each receiving a different signal.
And for those who are particularly keen on where to position the exciters ...
@lekha
Hello Lekha,
Thank you for the links to those videos I found of very good technical level (ok, the level of today or even yesterday from a company point of view but very good for the DIYer I am). I recommend to every DML builder here to watch them.
The first one speaks shortly about DML and more in details about the principle of the BMR. I appriciate to see the directivity plots. We have started to do some here for DML. I have noticed the remark about the dip because of the baffle used (48' 3" speaker) which is part of my last directivity measurements.
The second is a discussion with B Zenker. His papers are known here but it is probably better to have him explaining the content. The slides are good. He has simulated different size and ratio of panels. The funny point is what he calls a small panel is 0.1m², with a ratio of 2, it 22x44cm which is not that small in my vocabulary.
Beyond that, I found the history aspect in the post of the first video very interresting... unfortunatly without understanding where you want to come.
Christian
PS : according to the slide showing the results of similuations (see around 51") 0.2m² is a minimum to enter in the range of the lowest deviations this with a ratio of 1.25 (40x50cm) or even 0.3m² with a ratio 2 (about 40x80cm). Not that small!
Hello Lekha,
Thank you for the links to those videos I found of very good technical level (ok, the level of today or even yesterday from a company point of view but very good for the DIYer I am). I recommend to every DML builder here to watch them.
The first one speaks shortly about DML and more in details about the principle of the BMR. I appriciate to see the directivity plots. We have started to do some here for DML. I have noticed the remark about the dip because of the baffle used (48' 3" speaker) which is part of my last directivity measurements.
The second is a discussion with B Zenker. His papers are known here but it is probably better to have him explaining the content. The slides are good. He has simulated different size and ratio of panels. The funny point is what he calls a small panel is 0.1m², with a ratio of 2, it 22x44cm which is not that small in my vocabulary.
Beyond that, I found the history aspect in the post of the first video very interresting... unfortunatly without understanding where you want to come.
Christian
PS : according to the slide showing the results of similuations (see around 51") 0.2m² is a minimum to enter in the range of the lowest deviations this with a ratio of 1.25 (40x50cm) or even 0.3m² with a ratio 2 (about 40x80cm). Not that small!
Last edited:
Putting this table in a graph, it comes the curve below which is more a less a straight line that can be approximated by BestRatio = 7. PanelSurface (m²).
And going deeper leads to a constant width of 38cm
Really?
If Benjamin reads this thread, maybe we will get more input?
Christian
And going deeper leads to a constant width of 38cm
Really?
If Benjamin reads this thread, maybe we will get more input?
Christian
At 0.1m², the ratio is 1, which is quite small! The ideal sizes would be 0.2m², 0.3m², and 0.4m², where not much material would be required. However, on the other hand, one might need to purchase Xcite exciters.
At around the one-hour mark, Zenker was discussing a woofer in a panel that pushes air to excite the panel; what he meant was a woofer housed in a box with a panel in front. I believe I've come across a very old American patent on that, either in this thread or elsewhere. I just can't remember where.
In the video there is a first slide with a panel 0.1m² ratio of 2 (45") and in synthesis of the simulation the best ratio seems to be 1.25 for 0.1m².
Anyway you are right, best results are 0.2 to 0.4m²
The dependence to the exciter is not in the video.
The initial paper : Low Deviation and High Sensitivity -Optimized Exciter Positioning for Flat Panel Loudspeakers by Considering Averaged Sound Pressure Equalization
says DAEX30HESF at least for the measurements. I don't know if there is a paper about the simulations.
Christian
Maybe useful : from the paper, details about the panel for the measurements
In the video, he says no difference for the placement with a closed box
Lekha, You sure have a lot of knowledge on the history of DML's/BMR's and the patents! I'm curious to what your background was in the audio industry and your professional occupation, if you don't mind sharing.
Thanks,
Bruce
Maybe I misunderstood, but despite their enthusiasm, it seems to me that they are admitting that the DML system essentially has an inherent problem that makes it impossible to reach certain levels. Of course, the convenience of having a flat system remains, but it is extremely limited to certain applications. Please correct me if I'm wrong
Hello,
I don't remember that from the video probably because it is not at the top of my "specifications"... but it is my understanding of the DML too. This is opinion as I have no experience in that domain. Due to their wide dispersion up to high in frequency, they are interesting on a very wide frequency range (this is their main characteristic with maybe the IR which is said not coherent out of axis, not to be flat). In the 2 videos, the question of the open back, the front and rear waves cancellation is asked but I don't have in mind a clear and detailed answer. To reach very high level it might be a problem because of the needed displacement induced in the material. The levels we can reach for a home application is enough for me. I suspect also the exciters to be a limitation. I don't think they are design for low distortion (even for home applications), and for sure not for high displacement/high power. They are most probably first design to be cost effective.
Christian
Indeed, but more precisely, two. The commercial one for general public is well explained in the last videos that lekha post (thanks again for those): the shift in the way most people hear music content.
The scientific one derives from the transverse sound wave knowledge (as exposed previously) and…. Metamaterials.
It`s a pleasure to read all of the efforts that some members developed to understand the DML sound wave propagation but, for me, the real breakthrough will be in the metamaterial side.
For the record I use plastic 1m/50cm hanging on fishing strings 150Hz up, 12Db crossover with the daxe25fhe-4, 4,5m apart. And as Homeswinghome wrote “The levels we can reach for a home application is enough for me”, when spl is not an issue.
Full range DML? Sure.
Tried with the previous panels described plus several panels excited by the xcite´s 32, enough for +- 86-90dB listening, but… not for ´real´ dynamics, not to mention power compression due to thermal heat. So not a full speaker for me.
Which brings to other commercial sight, that is not a paradigm but a necessity.
In the pro audio the DML wave propagation characteristics are a bless for several difficult environments, mainly speech intelligibility, as Veleric wrote “how DML's provide better sound intelligibility in places that are notoriously reverberant, like subway stations and the like. If I recall correctly, they attribute this feature to the "incoherent" nature of the sound radiation from a DML”, but they fail in several pro criteria, so… metamaterials once again…
That “incoherent” nature of sound lead me to the same conclusion of Sandasnickaren, open back always.
For whom is interested:
https://www.sciencedirect.com/science/article/pii/S0020768323005371
https://pubs.aip.org/aip/jap/articl...astic-transverse-wave?redirectedFrom=fulltext