pixel1,
I aim for a compression of about 0.5 mm. Of course it is impossible to be exact, but about 0.5 mm is my target. That seems to work fine for me. Ideally, probably the best is no push and no pull. But I use only the adhesive tape on the exciter (no epoxy or other adhsive) to attach the exciter to the panel. I expect with no push at all, it could eventually "creep" loose. Having just a tiny push seems to prevent that (at least for me so far).
Eric
Actually, you can.
The speed of sound is limited, so the cone does not 'see' all the room at once. The sound pressure will build up, and it will spread.
Here's a question I've been thinking about lately: If we could actually design a panel that truly reacted like the "pebble in the pond" analogy, would it still have the wide dispersion character of a typical modal bending wave speaker (i.e. dml)? Or does the wide dispersion of a flat panel speaker require a modal character? What other characteristics of the speaker would benefit from (or be harmed by) the the reduction of modal behavior? I'm pretty sure that reduction of modal behavior would improve the impulse response and reduce "ringing" (perhaps these are mostly the same thing). But what else would be the effects?
Eric
Eric
Hi Eric,
The Goebel design which uses the damped slots at the perimeter claims to act like an infinite plane and as a result that the signal disperses like a ripple on a pond. I have not heard one but the reports are of a wide dispersion not unlike a DML.
From a review
“Where things get funky is these drivers and radiators hand off to a rectangular Carbon Excellence bending wave driver capable of a remarkably wide, continuous bandwidth of 170Hz to 31kHz. Moreover, this same bending wave driver has a dispersion of nearly 180 degrees both frontwards and backwards. The rear-firing bandwidth is limited to 170Hz to 4000Hz across all Epoque speakers, but it is said to produce a completely immersive sound field similar to that of an omnidirectional speaker.”
I don’t understand how or why the rear bandwidth is limited. Might be an error from the reporter?
Burnt
The Goebel design which uses the damped slots at the perimeter claims to act like an infinite plane and as a result that the signal disperses like a ripple on a pond. I have not heard one but the reports are of a wide dispersion not unlike a DML.
From a review
“Where things get funky is these drivers and radiators hand off to a rectangular Carbon Excellence bending wave driver capable of a remarkably wide, continuous bandwidth of 170Hz to 31kHz. Moreover, this same bending wave driver has a dispersion of nearly 180 degrees both frontwards and backwards. The rear-firing bandwidth is limited to 170Hz to 4000Hz across all Epoque speakers, but it is said to produce a completely immersive sound field similar to that of an omnidirectional speaker.”
I don’t understand how or why the rear bandwidth is limited. Might be an error from the reporter?
Burnt
Last edited:
The Gobel website confirms the 180 dispersion front and back but also, oddly, the limitation on the rear bandwidth
http://www.goebel-highend.de/technology.html
Burnt
http://www.goebel-highend.de/technology.html
Burnt
Burntcoil.
The original goebel panels were the same, they had a restricted frequency from the rear.
I always presumed that this was from some sort of damping behind the panel ?
The slits at the sides were also filled with ,I think,silicone?
I missed my chance to hear them at the last audio show, I'm sure they would have sounded good.
But the price would probably be way out of my budget.
Would I have preferred the sound to my panels, I doubt it.
At the end of the day it is a heavily damped panel, that would be a problem for me, but some I'm sure will like that.
Steve.
The original goebel panels were the same, they had a restricted frequency from the rear.
I always presumed that this was from some sort of damping behind the panel ?
The slits at the sides were also filled with ,I think,silicone?
I missed my chance to hear them at the last audio show, I'm sure they would have sounded good.
But the price would probably be way out of my budget.
Would I have preferred the sound to my panels, I doubt it.
At the end of the day it is a heavily damped panel, that would be a problem for me, but some I'm sure will like that.
Steve.
Morning Steve,
Yes you are right the slots were filled with silicon. Absorbing the high frequencies from behind does seem a design goal. They appear to have transitioned away from their special laminate to carbon fibre for the panel as well. The main area of the panel appears to be undamped from what I can see
To Eric's question, in this case an 'infinite' panel with no modal action does appear to retain the omni-like sound field. If it truly is mode free that is.
Burnt
Yes you are right the slots were filled with silicon. Absorbing the high frequencies from behind does seem a design goal. They appear to have transitioned away from their special laminate to carbon fibre for the panel as well. The main area of the panel appears to be undamped from what I can see
To Eric's question, in this case an 'infinite' panel with no modal action does appear to retain the omni-like sound field. If it truly is mode free that is.
Burnt
pixel1, your translator is not fuctioning properly. Your post has been deleted as it was not in English.
TRANSLATION said:
Thanks Burnt, and Steve,
Yes, of course I know about the Goebel design and their claims. But as you may also know I suspect the Goebel plate is closer to a DML than an infinite plate than they would like to admit.
As far as I can see, the plate itself is not particularly special. The original one had fiberglass skins with a balsa (end grain) core. The new one has carbon skins and (I presume) the same or similar balsa core. I have made many panels similar to these and the panels themselves are highly elastic (very low damping). I'm sure my panels are not identical to theirs but I doubt they are much different with respect to damping.
They do make a big deal about their slits and perimeter damping. But my reading about acoustical black holes recently suggests that elimination of reflections is a pretty hard thing to do effectively, and if angled slits with silicone filler were really effective, we would see mention of that in the literature (I have not seen it).
I've tried myself to maximize perimeter damping with various materials of various widths, thicknesses, etc. These definitely have a damping effect, but nothing I have dried on the perimeter comes even close to eliminating the modal behavior. Butyl rubber window glazing seems to be one of the best for this, but even butyl only slightly spreads the peaks (in my tap test), but all the same modes are apparent.
I'd love to do my tap test on a Goebel panel! It'd take me about 5 seconds to see how modal it really is. That is, if they'd let me whack it with my mallet!
Given that, I'm not sure that the performance of the Goebel speaker really answers my question about how a truly well damped panel would perform. What I'm thinking about is a panel with significant internal damping (not just perimeter damping) such that the ripples are significantly damped before they even reach the edge of the panel. What would be the speaker's characteristics then?
Eric
Understood. Yes that is new territory and intriguing. I built a nomex honeycomb ply composite panel this year. It would be possible to ‘dope’ some of the cells before applying the final ply which may be a way of getting what you want.
On testing Gobel, I can distract the sales guy while you whack it?
Burnt
On testing Gobel, I can distract the sales guy while you whack it?
Burnt
Last edited:
It is not the matter of the cone, but the matter of the space. Air cannot be pressured in an unlimited space. We and everything around us are in that air. Air is unseen, so we don't consider that we live in a medium, just like fish in water. Sound travels through any medium, in solids it has a very high speed. By the time, a solid passes the sound that is passing through it to a low speed medium (air), the rest of the sound had gone away. There's always some lack of sound coming from solids.
The idea here is to search for means, reasons why sound comes out of surfaces, flat or otherwise, once a sound source is attached to it. Maybe, how the sound travels from that surface to our ears could be important, but is secondary. Why the same flat surface gives out more full sound when curved is more interesting (Telefunken Arcophon 3 speaker of 1926). It'd be pretty hard to send waves on a curved surface. Inner surface is much stiffer/tougher than the front. The fully closed curved surface, a cylinder is the most stiffest, still it gives fuller, clearer sound, than a flat surface of the same material.
The diameter at the bottom of this glass vase is 12cm, open top diameter 18cm, the glass thickness is ~3mm, the glass bottom is 1 cm thick, height is 70cm. We have two of them. They have been tested for sound amplification for a brief moment, once with the 'exciter' at the centre under the thick glass bottom, once with the exciter touching the open side holding the vase upside down. Not an easy thing to do, though. It was much easier to place the 'exciter' underneath, fix the vase in a wooden cylinder, so it won't topple. The resulting sound amplification was very nice. That experiment was done for a few minutes. They have been bought for X'mas decorations. Might be used later for other kind of decorations.
But the preliminary testing for sound was done. Once the exciters are hidden in a decorative wooden cylinder/box, there'd be no way to tell where the sound is coming from. Decorations can stay in. I suppose the X'mas decorations won't come off until February, so got to wait for further testing. Someone might say this vase looks like a horn, but there's no opening at the bottom, only a solid block of 1 cm thick glass. Interestingly, sound doesn't come from the open end, but from the surface of the vase, a curved enclosed panel. If the sound come from the surface, then that surface is DM surface.
Last edited:
It’s a good question. On the one hand, you would have a mostly spherically expanding wavefront, and no edge diffraction effects bc the whole wave is theoretically absorbed before it reaches the edge (which I also doubt is possible to do in a practical setup). On the other hand, you would still have differences in travel time because of propagation delay across the panel is still there, so dispersion I guess would not be as good. Also, wall reflections would be well correlated so the room would have more of a deleterious effect. Also, I would guess the LF performance would be inferior too, because acoustic short-circuit cancellation will still occur below coincidence frequency, and yet you presumably won’t have any uncancelled sound at the edges - the so-called corner and edge modes, which is the source of the LF from normal panels
BTW, I don’t believe the sound from normal panels is random or uncorrelated, since at any one frequency the source of sound is the same. It’s just that with musical content the frequency is always changing, and the locations on the panel are different at every frequency, so it’s ‘effectively’ more random.
It’s 3am here and I’m awake with indigestion from too much Greek food, and too many ideas floating around. I’ve taken delivery of some high density eps panels so I’ll try some tap testing too.
I have a question too. Has any speaker technology attempted to do something like simulate coincidence behaviour? By which I mean this: at coincidence, sound is propagated bc it travels across the panel at the speed of sound, so energy is efficiently transferred to air. Presumably if you had a linear array of exciters on small panels in a pipe, you could delay each subsequent signal so that it is aligned with the passing wavefront from the previous exciters, giving a much boosted output. You could even do it from both ends of the pipe!
Another thing I’ve been thinking about is exciters. Seems to me that exciters derived from conventional speakers are an inefficient way to excite a bending wave. Voice coil exciters are designed to create a velocity, which in turn shakes the panel and creates bending waves, which ultimately gives the velocity of the panel moving at any particular point. But to create a bending wave directly, presumably what you want to do is apply a bending stress to the panel. So you need at least two points on the panel moving in opposition. I would guess this simple arrangement would be more efficient than the flexing cause by the inertia of the exciter and panel. Two points let you change the distance between the points to excite different frequencies optimally. It would also leave open the possibility of directed waves, and more potential for adjusting the frequency response by changing the direction of one exciter, or using multiple exciters in different directions, possibly EQed separately. Or maybe a three-point or four-point exciter where the bending direction is rapidly rotated, giving a more random excitation. Then again, I may be just delirious from lack of sleep. Goodnight.
Last edited:
Sorry about your indigestion problem. Hope, it'd be over by tomorrow.
When you wake up tomorrow, let's know what you think about the "pebble in the pond" type waves on a much more solid surface than water. Any wave would slow down and die in/on water. Waves on water might hit an obstruction and turn back, and obstruct the oncoming waves. When there's no movement, the water surface would settle and stay still. I've seen this still water in/on a private lake. After all water, is a liquid.
Now, if on a EPS sheet surface, which is very poor in terms of rigidity, waves starts moving, the ups and downs of the waves should stay on the surface. It is not a liquid to settle down. After a while, the EPS sheet should have wrinkles. But it doesn't. A very thin paper might flutter, but as paper is also a solid, so too much flutter would deform it. The "pebble in the pond" doesn't cut it, does it?
If you were to use the holes on the frame of the first exciter image below to screw the exciter to the panel, wouldn't that create a bending moment?
I think the inner and outer rings of the second exciter would also create such a bending moment.
Neither is as sophisticated as what you were imagining, but at least they should create a bending moment directly.
Eric
I think
Thanks Paul, I'll have to give these issues some thought.
Other than the Goebel, the other speaker that supposedly tries to create a wave that expands from the center out (i.e. the "pebble in the pond" type), is the later version of the Quad ESL.-63.
What are the characteristics of the ESL-63 and it's later variants (ESL-988 and 989)? And would a highly damped panel as I have described have similar characteristics to it?
Eric
The space is not put under pressure, only part of it.
The speed of sound acts as a barrier, which makes the space small.