A different kind of panel speaker

[Blumlein 88]

Being a lover of electrostats I like panel speakers.

This is a different idea for a panel speaker.

This 31 minute video is wordy and not well organized so feel free to jump around some. And note just past the 20 minute mark some annoyingly loud tones in the video so be prepared to lower volume (a bunch).

Here is a related video of how to make your own contact vibrators.

The 1st video's resonance reduction looks to sort of work. Maybe if worked on enough this could make a good sounding panel, and they are claiming to be quite efficient.

 

[Wombat]

A

Being a lover of electrostats I like panel speakers.

This is a different idea for a panel speaker.

This 31 minute video is wordy and not well organized so feel free to jump around some. And note just past the 20 minute mark some annoyingly loud tones in the video so be prepared to lower volume (a bunch).

Here is a related video of how to make your own contact vibrators.

The 1st video's resonance reduction looks to sort of work. Maybe if worked on enough this could make a good sounding panel, and they are claiming to be quite efficient.

Similar to 'motors' used to add LF vibration to car seats and gamer's chairs back-when. Some guidance on panel size/mass needed for application. Fun for non-HiFi sound, probably. Add many for earthquake vibes to home cinema using panel walls, complete with cracking.

 

[RayDunzl]

Seems to me that this panel is resonating not just at the fundamental, but, at a harmonic multiples.

---

Fundamental - 821Hz, -5.2dB
2nd - 1640Hz, -34.1dB
3rd - 2460Hz -38.2db
4th - 3280Hz, -59.0dB

Could be worse...

 

[Blumlein 88]

Yes, I thought he was clear about that in the video. His other panel was constructed differently to have less resonance and multiple frequencies. Still sounded like it had lots of harmonics to me in his improved version. I don't know if these can be made high fidelity or not. Just thought it a neat panel speaker idea. My guess is should they get resonance distributed well enough for low distortion you'll lose efficiency and they'll become unworkable for that reason, but maybe someone clever will do it. Also occurs to me such a panel needs to be further from the wall though this guy was pushing the idea it fits out of the way like a flat screen TV.

The other thought to me was the idea of having a curtain of these panels with multiple actuators for Atmos type surround systems. Not as obnoxious as 13 or 40 channels of cone speakers.

 

[RayDunzl]

Yes, I thought he was clear about that in the video.

Maybe, but I didn't give him 31 minutes to tell me what I expected to see and soon saw.

Sorry.

---

Just watched the end, after wondering if he tickled them with something other than a tone, and he did.

 

[Wombat]

Does anybody remember Poly-planar speaker from the sixties/seventies?

http://www.audioimprov.com/AudioImprov/Speakers/Entries/2012/6/15_Bertagni_SM-275_restoration.html

There were rectangular baffles/panels of various sizes for DIYers. Build a box and fit the musical baffle - done, no crossovers. They sounded OK for non-discerning listening

 

[Cosmik]

I can't get excited by it. You take an ordinary moving coil speaker with high output level, good linearity and point source qualities and turn it into a very low output, poor linearity speaker which beams in weird ways.

In any case, as far as I can tell, the only advantage of the electrostatic speaker was because it was 'full range without crossovers'. Now that we can control multiple moving coil drivers with ease, it seems to me that they are the perfect speaker. But unfortunately they are boring in their simplicity and perfection!

 

[Blumlein 88]

I can't get excited by it. You take an ordinary moving coil speaker with high output level, good linearity and point source qualities and turn it into a very low output, poor linearity speaker which beams in weird ways.

In any case, as far as I can tell, the only advantage of the electrostatic speaker was because it was 'full range without crossovers'. Now that we can control multiple moving coil drivers with ease, it seems to me that they are the perfect speaker. But unfortunately they are boring in their simplicity and perfection!

Actually all the ESL's I've heard without something like a crossover were by far the least satisfying and measurably the least performing. Though yes, the full range crossoverless ESL was the idea marketed. The real idea maybe was the panel itself radiated all frequencies for the all the good and bad of that situation.

By this I mean that Acoustat used a pair of transformers with 50:1 and 200:1 step up ratios and had crossovers that blended those together on the panel. Soundlab uses the same idea. You can actually bi-amp both speakers. Quad used two different panels crossed over in the original, and has the delay line EQ thing in the later Quads. Pure panels without that like early ML and other designs just have no low end, and are horribly inefficient. They have a certain beguiling midrange purity like nothing else, but no bass. They also lack the finest treble which in my opinion was found in some ribbon tweeters. The best speaker undoable in the past, but not now, might be a box/cone woofer, ESL midrange and ribbon tweeter. The midrange sound spread over a large ESL panel is still very, very low in distortion vs other speakers. Then there are horns which might have lowest possible distortion of all if various issues can be fixed with extreme modern design profiles and/or DSP.

Good time to be an extreme audiophile if you can avoid getting sucked into the audiophile bull shite!

 

[Wombat]

Actually all the ESL's I've heard without something like a crossover were by far the least satisfying and measurably the least performing. Though yes, the full range crossoverless ESL was the idea marketed. The real idea maybe was the panel itself radiated all frequencies for the all the good and bad of that situation.

By this I mean that Acoustat used a pair of transformers with 50:1 and 200:1 step up ratios and had crossovers that blended those together on the panel. Soundlab uses the same idea. You can actually bi-amp both speakers. Quad used two different panels crossed over in the original, and has the delay line EQ thing in the later Quads. Pure panels without that like early ML and other designs just have no low end, and are horribly inefficient. They have a certain beguiling midrange purity like nothing else, but no bass. They also lack the finest treble which in my opinion was found in some ribbon tweeters. The best speaker undoable in the past, but not now, might be a box/cone woofer, ESL midrange and ribbon tweeter. The midrange sound spread over a large ESL panel is still very, very low in distortion vs other speakers. Then there are horns which might have lowest possible distortion of all if various issues can be fixed with extreme modern design profiles and/or DSP.

Good time to be an extreme audiophile if you can avoid getting sucked into the audiophile bull shite!

But individuals who know SFA about speaker design venturing into using existing wall panels and underlying construction, all different everywhere, sounds like fantasy to me. Maybe in limited, designed architecture it has a use.

 

[Cosmik]

The midrange sound spread over a large ESL panel is still very, very low in distortion vs other speakers.

One of my contentions is that active crossovers and DSP allow us to use more 'ways' than we could previously because crossovers are now transparent. Given a three or four-way cone speaker with perfect step response, do we still need the fragility, 'beaminess' and high manintenance of the ESL panel? It sometimes seems to me that people are more focused on the romance of the technology than on the mundane issue of duplicating the recorded signal in the form of air pressure variations.

 

[Frank Dernie]

20 years ago I was Chief Engineer of Lola Cars in Huntingdon. I didn't live locally so when at the factory I stayed in B&B accommodation nearby. There are lots of hifi businesses in this area and one of the people I got friendly with at the B&B was a consultant who specialised in loudspeakers, specifically computer modelling their vibrational characteristics and building prototypes.
He was a consultant to NXT and was computer modelling their theories and testing his model using prototypes.

https://en.wikipedia.org/wiki/Distributed_mode_loudspeaker

I heard some prototypes and ended up being given some cardboard pyramid speakers using the technique.
The obvious considerations. The heavier the panel the exciter is connected to the less the efficiency, and for a good even frequency response the shape of the panel and where the exciter was attached was crucial. I found it very interesting but it really came to nothing.
Thie new stuff leads me to suspect NXT's patents may be coming to their end.
The development which they came up with which looks, or looked, fantastic in principle is the BMR driver. This is used in quite a few speakers and the prototypes sounded fantastic to me.
There is a production engineering limitation though controlling consistency of drivers where the highest frequencies are influenced by small changes in the surround which make no difference to a normal mid driver, if I understand it correctly, leading mass production driver manufacturers to struggle with consistency (I think I have got that right...)
Anyway my free cardboard speakers sound surprisingly good and make a talking point.

http://www.nbcnews.com/id/5404920/n...ts/t/speakers-madeout-cardboard/#.Wrt4xWaZONY

 

[Jakob1863]

<snip>
The best speaker undoable in the past, but not now, might be a box/cone woofer, ESL midrange and ribbon tweeter. The midrange sound spread over a large ESL panel is still very, very low in distortion vs other speakers. Then there are horns which might have lowest possible distortion of all if various issues can be fixed with extreme modern design profiles and/or DSP.

And then there was this beast (afair around 1982, i had to dig in the AES archive):

Sardec CH 2; used 3 woofers (12 or 15 inch) an ESL 63 as midrange and "some" isodynamic ribbon tweeters (presumably Technics/Matsushita TH 800) per channel.

It never made it to full production so just two or three prototypes exist.....

Good time to be an extreme audiophile if you can avoid getting sucked into the audiophile bull shite!

Even better to avoid all sorts of bull shite, be it "audiophile" , "pseudo-objectionistic" or of any other kind.

 

[Fitzcaraldo215]

The best speaker undoable in the past, but not now, might be a box/cone woofer, ESL midrange and ribbon tweeter. The midrange sound spread over a large ESL panel is still very, very low in distortion vs other speakers.

There was Mark Levinson's famous HQD system in the late 70's. Two Quad 57s over/under a Decca ribbon in a wooden mount plus a Hartley 24" woofer per side.

https://www.stereophile.com/content/mark-levinson-hqd-loudspeaker-system

Like JGH, I heard them at Chestnut Hill audio in Philly, but with Bryston, not Levinson amps. I was not all that impressed. The room was much too small for them, I thought, plus other issues.

And, speaking of large 'stat panels, I continue to like them mated to cone woofers at a low xover frequency. My guess is that is so because there is no crossover in the sensitive 2-3k Hz region, like there is for the majority of 2- and 3-way cone speakers. That also means there is no shift in directivity there between a larger mid and smaller tweeter, usually cone to dome.

 

[DonH56]

I heard the HQD system in a large room and was still not terribly impressed.

 

[Don Hills]

... He was a consultant to NXT and was computer modelling their theories and testing his model using prototypes.

https://en.wikipedia.org/wiki/Distributed_mode_loudspeaker

...

I'm surprised you were the only one to mention this. As you point out, computer modeling and lots of experimenting is required for good results. Random hacking such as in the videos above are unlikely to get there.
TDK licensed the NXT technology for several of their computer speakers and micro systems. They don't sound any worse than the prize and size equivalent conventional systems.

 

[Blumlein 88]

I'm surprised you were the only one to mention this. As you point out, computer modeling and lots of experimenting is required for good results. Random hacking such as in the videos above are unlikely to get there.
TDK licensed the NXT technology for several of their computer speakers and micro systems. They don't sound any worse than the prize and size equivalent conventional systems.

Were they also the company that was experimenting with using such panels to send out ultrasonic sound that would via IMD demodulate into audible sound below 20 khz?

 

[Rene]

A recent visit to Tectonic Audio Labs was very enlightening:

http://www.aes-media.org/sections/pnw/pnwrecaps/2017/tectonics_dec2017/

Their panels have tremendous bass impact and extremely wide dispersion. They can cut through the clutter in very noisey and reverberant environments; but they are definitely not high fidelity. I cannot describe how they sound because their sound does not resemble any other acoustic experience in my memory.

If you are interested in hearing their speakers, Marcelo Vercelli is usually quite willing to demonstrate them.

 

[bigx5murf]

I had a pair of vintage speakers that I believe were built like this. They were BES, stood for Bertagni Electroacoustic Systems, they used transducers to vibrate a sheet of Styrofoam. They sounded terrible, hot treble, and not much bass.

I had a TEAC mini component system with small flat speakers, the backs are labeled as panels from NXT. Pretty much the same sound signature, hot treble. This system had a subwoofer that had to be hooked up to work.

 

[JAPN]

Yes, I thought he was clear about that in the video. His other panel was constructed differently to have less resonance and multiple frequencies. Still sounded like it had lots of harmonics to me in his improved version. I don't know if these can be made high fidelity or not. Just thought it a neat panel speaker idea. My guess is should they get resonance distributed well enough for low distortion you'll lose efficiency and they'll become unworkable for that reason, but maybe someone clever will do it. Also occurs to me such a panel needs to be further from the wall though this guy was pushing the idea it fits out of the way like a flat screen TV.

The other thought to me was the idea of having a curtain of these panels with multiple actuators for Atmos type surround systems. Not as obnoxious as 13 or 40 channels of cone speakers.

In my case, I tried to make the panels, it is really very difficult to make them sound like speakers, each material that is used, brings different distortions, you have to mix different materials and use several transducers to achieve an "acceptable" result, if you want to make a Hi-Fi project, you need a lot of investment.

Already analyze most types of flat speakers, and few get to an acceptable sound.

The best systems that I have seen and are more efficient are Bertagni Electracoustic Systems, Fisher PL-6-W "Zebra" Label of this there is a publication that even shows how they are embedded on the wall (this is very difficult to achieve, due to the rear resonance) and Technics SB-AFP1000, and currently exist, Tectonic DML500, J

Each of these projects are very interesting, however I don't think they are Hi-Fi, or at least they play the music faithfully without coloring.

In general, when you are looking to make a project of these dimensions, you encounter many difficulties, more than if conventional speakers were used.

Saaludos.

 

[JAPN]

I had a pair of vintage speakers that I believe were built like this. They were BES, stood for Bertagni Electroacoustic Systems, they used transducers to vibrate a sheet of Styrofoam. They sounded terrible, hot treble, and not much bass.

I had a TEAC mini component system with small flat speakers, the backs are labeled as panels from NXT. Pretty much the same sound signature, hot treble. This system had a subwoofer that had to be hooked up to work.

This system of transducers is very squeaky in treble and difficult to achieve bass, you must have a large surface area to achieve it.