Electrostatic design - is a round design practical?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Put a hex on 'em

Hello Sashua,

Thanks for the clarifying posts - I see now that you've already got the tweeter end of things well under control.

I believe that multiple small conventional woofers is the way to go.

The comb filtering you see in a same-plane driver array is largely eliminated if you use a splayed array. This from Earl Geddes' "Audio Transducers". Very uniform results are obtained from splaying two drivers out by 30 degrees to either side of the centerline, or in other words with an angle of 60 degrees between the drivers.

This happens to work out very nicely - if you like hexagons. The angle between each face of a hexagon is 60 degrees. You could build a columnar hexagonal enclosure and mount a small woofer on each vertical face.

Now because of the way the drivers' responses will sum (remember that each driver's radiation pattern is widening as you go down in frequency), you want to use drivers that have a smoothly and gently tipped-up response across their passband. Morel's smaller Neodymium drivers come to mind. They have very low power compression and very good low-level articulation, which you want if you'll be using an ionic tweeter. The physically compact frame and magnet assemblies of the Morels makes a fairly compact hexagon possible.

A hexagonal ring of small woofers will be smoothly omnidirectional in the horizontal plane. You will still get some beaming in the vertical plane, but probably not enough to be of consequence with a 1 kHz ballpark crossover.
 
Hybrid fourdoor said:
Do a search on Plasma Tweeters. They are probably the truest iteration of what you are thinking about.

However they have drawbacks, such as frequency extension, very high voltage, very little resources available, and they produce ozone.


But they basically use a plasma flame that modulates at the frequency...in <practically> true 360* nature.
http://www.studioerosbarone.it/diego/Plasma2eng.htm
 
Ultrasonics

Ok. You said you wanted something outside of the box. I don't know how DIY this is, but there is a company in San Diego that has been developing systems based on ultrasonic sound generators. As I understand it, the ultrasonic waves are aimed at a point from two radiators. The sound is generated where the ultrasonic waves interfere. So-- figure out how to aim waves from multiple sources, so that the iterference zone is essentially spherical. No box at all!

Sheldon

http://www.atcsd.com/
 
Mangers in a spherical enclosure

A friend of mine has Mangers built into spherical enclosures at opposing poles. When you walk around them there is no "break" whatever in the sound. In the current version the speakers are dipoles but he used an unipolar setting earlier. I didn`t listen to that, but it should do the job.
 
You say that there is no break in the sound but even if the Manger driver had a perfect 180 degree dispersion pattern there would still be a seam in the transition and/or overlap with drivers placed on opposing ends of a sphere. You must also consider the vertical dispersion of sound. Get above or underneath the speaker and there will be a frequency rolloff for sure.

Good idea, probably sounds great....but it still doesn't make the grade for a "perfect" omni design.

Russ
 
I see lots of talk about omnidirectional drivers and have to ask, what does omnidirectionality have to do with quality sound reproduction? What is the idea behind using omnidirectional drivers?

The microphones used to record music usually have some directional pattern. The instruments and voices recorded usually have some directionality as well. If a person were in the room talking, little of their voice energy would be directed toward the back of the room (unless they were facing there). Why would designing a speaker to send the voice equally in all directions be considered realistic?

When I first started making ESLs I was immediately impressed by the improvement in sound quality they provided due to the ability to control some of the reflections in the room.

My ESL-63s have some complex electronics and segmented stators that are supposed to simulate a pulsating sphere. I actually prefer the sound of the a plain-jane flat panel like the ones I make. Sure, you need to be sitting in "the spot", but for optimal performance with ANY speakers, you need to be in "the spot".

What am I missing?

I_F
 
It has been said that the ideal speaker is a point source with infinite bandwidth and dynamic range.

Among the many benefits to this ideal speaker are that issues like polar response, baffles and diffraction, time alignment and phase changes all go away.

In all probability the main difference that you heard when you started to listen to ESLs had little to do with the control of room reflections and more to do with an overall improvement in quality via a reduction in distortion, especially at higher freqs. At least that's my view.

The 63s always sound to me like they are "down the hall" somehow, which is probably due to the way they are set up with the delay lines and all... so it makes some sense that you might prefer a straight ESL over them...

_-_-bear :Pawprint:
 
IF, I don't think it's entirely correct to draw an analogy between mike patterns and speaker patterns. Once you've made the giant approximation of condensing the sonic drive to the room to 2 (or 5.1) points, the analogies start crumbling.

I've heard a few omni systems. Most were pretty poor, but not because they were omni; I was blown away by the imaging excellence of those German watermelons (MBL?).
 
Sy, I was about to say the same thing about the MBL 101. Spectacular sounding. One of the top two or three speakers I’ve ever heard.

For me, an interesting thought experiment when considering polar patterns of speakers is to remember the reflected image theory. Imagine that all walls, floor and ceiling are mirrors. Wherever you "see" reflections of your speakers there are phantom sources of sound placed exactly at that mirror image. All these add up to produce the sound field at your ears, with all the constructive and destructive interference. Using this approach can help you consider certain speaker patterns and placements. Beveridge was on to something…
 
bear said:

In all probability the main difference that you heard when you started to listen to ESLs had little to do with the control of room reflections and more to do with an overall improvement in quality via a reduction in distortion, especially at higher freqs. At least that's my view.

I experimented with them a bit when I first put them together including messing around with tilting their angles to send the back wave either to my ears or elsewhere, and found there was a pretty distinct difference between the sounds (as one would expect due to simple interference between the two signals). I also messed around with putting absorbant materials behind the speakers (along the wall, not close to the speaker) to limit the reflections of the rear wave. I ultimately concluded that the old live-end/dead-end worked best with the speakers being at the dead end of the room, and listener's head pretty far from the live rear wall. The result is very stable and localized imaging with recordings that are mixed for that sort of thing.

I was also quite impressed with the low distortion. The first time one hears such wide band/low distortion audio coming from a single driver (especially a home made one) there's no going back.

I_F
 
sashua said:
You say that there is no break in the sound but even if the Manger driver had a perfect 180 degree dispersion pattern there would still be a seam in the transition and/or overlap with drivers placed on opposing ends of a sphere. You must also consider the vertical dispersion of sound. Get above or underneath the speaker and there will be a frequency rolloff for sure.

Good idea, probably sounds great....but it still doesn't make the grade for a "perfect" omni design.

Russ


Of course this doesn`t work when you come too close, but my friend`s listening room is huge.
 
SY said:
IF,
I've heard a few omni systems. Most were pretty poor, but not because they were omni; I was blown away by the imaging excellence of those German watermelons (MBL?).


Maybe the secret of omnis like the MBL 101 is that they have aequivalent wavefront curving over the whole frequency range. Also aequivalent dispersion, but I believe wavefront curving is more important.
 
Mangers on opposite poles of a sphere? (no sides, eh?)

In phase?

Size of sphere?

This is/was a stereo sphere, or were there two spheres, 4 Mangers?

I had Mangers here for a while, interesting sound. Not particularly omnidirectional as they got above say 5kHz. Start to get raggedy and directional as they go up in freq.

So, not "omni" in my book.

Still, if I needed a quiet speaker for a quiet place, they'd be a great choice...

MBLs, suck power for lunch. No SPLs available as far as I can see. Saw them in MBL rooms several times, run with large "monkey coffin" amplifiers to a deafening 85dB SPL or so... nice though.

Anyone ever hear the Ohm F or Ohm A to good advantage? Fairly well omni there. (doughnut shaped pattern)

Magnat tweeters, Hill Plasmastatic? :D

I tried to make a true cylindrical experimental ESL, but that particular one didn't fly. Think I might be able to pull it off now though, but somewhat differently, and really only an approximation of a cylinder...

_-_-bear :Pawprint:
 
My friend has a five-channel setting. Three spheres with about 50cm diameter made of concrete. the center-sphere has a single Manger. The side spheres are dipoles and are turned relatively far inwards. Even a bit more than directly facing the listener. The Mangers on the back side of these spheres have slightly less amplitude. The rear Mangers house in cubic MDF-enclosues and are looking down onto diffusor cones. The woofers are two Klipschorns. I don`t know the crossover topology he uses now, he has changed it very often. The nine monoamps are Black Devils from Experience Electronics. They are AB, but have a harmonic spectrum with almost only K2. They don`t have anything that prevents overpower, except melting fuses. The power supplies are speeded up with mica capacitors.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.