Print

[captainjack115]

For sure - dipoles behave different in coupling/exciting the room modes as monopole speakers (boxed speakers) do - move them away from the back wall - this will increase LF output due to better exitation of the lower room modes.

I respectfully submit that bending wave transducers are not dipoles, they are bipoles.

The whole cancellation effect of dipole operation at lower frequencies is out the window, so to speak with dipoles. On the other hand with bending wave bipoles, Bending waves on the rear surface will be in phase with the front surface. It's just a matter of careful dampening at the the edges of the board to retard reflections back to the epicenter without impeding edge motion too much. The edge is where the most travel of board takes place, thus providing the highest amplitude of lower frequencies. Mounting a bending wave transducer can be a tricky situation. Thats why many of us have reported that lifting the board from the floor has a dramatic effect on bass. The "Q" can be very high and narrow, in some cases undetectable by a spectrum analyzer, but quite noticeable by ear.

[captainjack115]

[quote author=Russell Dawkins link=topic=70541.msg688131#msg688131
"I'm afraid this is not correct.
In a bipole system, two drivers fire in phase with one another, but in opposite directions.

This is not possible with one panel unless it's expanding and contracting in thickness according to the input signal.
Well in fact the panel does contract and expand to some extend - but not in a way so that bipolar behavior is the result.

... anyone with a good understanding of the basics of acoustics would realize that mkstat is absolutely correct on this point.

It is when the discussion became argumentative on this fundamental notion that I bailed on this thread months ago. To think a panel could exhibit bipolar or omnipolar radiation in bass frequencies is to demonstrate that science is not supporting proceedings, to put it politely.
To be less polite - this is fantasy, and not helping with any potential understanding."

The photo shown in the original post shows two pack to back conventional pistonic drivers.

The problem in that illustration is one of misdirected phyisics. the distance separating the the cones has not been taken into consideration. You cannot piece together a sine wave that has been pulled apart by considerable distance with any semblance of accuracy.

As a published author of electronic projects and instructional courses, I have learned that text book learning will only take you so far. There's a lot to be said for empirical design. Kudos to Ziggy for using the finest audio equipment for his speaker design, his ears and good sense.

I've designed and built electrostatic loudspeakers for over 25 years and electrodynamic speakers for much longer. Electrostatics are dipoles, very fast and accurate, but they need to be large to get bass. NXT type panals need to be large too, but mostly to move anough air. Comparing the two technologies, I'd say that NXT bending wave has a lot going for it and empirical work needs to be encouraged.

"You can't listen to a graph or a spec sheet!"

[bobloblob]

While I can appreciate the desire to have an undivided single driver, I am going to be happy if I can dial in a very wide range driver, which these panels seem ideal for.  If I can bring in a pair of woofers fairly high, say  100 to150 Hz and then bring in a super tweeter at the top, I will be content, since that would seem to let the panels do what they do best, and in the area where most music lives.  It also means that the Gatorfoam may do perfectly well for my needs.  It would be nice to cut the super tweeter in above  10K, but, judging from what others here have posted, it looks like 8K may be a better point.  For the subs, I like the 18 inch woofer in an H-frame dipole that MJK uses (with the panels stacked on top, maybe?). This should also relieve the exciters from the strain of reproducing heavy bass passages, even though they seem able to do some.

My attitude on subs changed after hearing Bruce Edgar's horn sub at his shop.  That was the first time I had ever heard a string bass reproduced in a way that seemed realistic.  Every other subwoofer I have heard, though it may reproduce the fundamental, compresses it greatly when  the music gets dynamic, even on a piece that doesn't seem that demanding.  I also later realized that I had never heard an audio system reproduce a cello well.  Since about a third of my listening is to classical music, and another third is to jazz with some bass in it, I am now more critical of bass reproduction than I used to be.  I think the dipole bass may mate well with these panels, though they may not put out as much as horn subs.  As good as these panels may be, and given that they seem to get pretty low down,  I don't think they could put out the type of bass that "blooms" with the rest of the music.  What's more, I'm fine with that.  Any thoughts or suggestions?

[el`Ol]

Here is a compact horn sub that will do the range you need.
http://www.hm-moreart.de/13.htm
hm charges a bit for his plans, but should be worth.

[Rudolf]

Electrostatics are dipoles, very fast and accurate, but they need to be large to get bass. NXT type panals need to be large too, but mostly to move anough air.

Why are NXT type panels called distributed mode speakers? Because they are neither pure dipole neither pure bipole. They have opposite phases 'distributed' on every side of the panel. They don't pressurize a room like monopoles and bipoles would do at very low frequencies. But they don't have a clear dipole separation either. And their ability to produce bass is very well defined by their size in the first place. A small DMS will NOT produce bass, regardless how much air you are able to move with it.

[bobloblob]

el' Ol, thanks, I've seen that and have liked some of HM's other designs too.  If I were to go with a horn sub, the one I would probably go with would be Bill Fitzmaurice's, here:

http://www.billfitzmaurice.com/TT.html

What I am more interested in is something that I think may be as open-sounding as this panel, and that would be MJK's H-frame dipole woofer, here:

http://www.quarter-wave.com/Project08/Jordan.pdf

Used in an active system, this should work well with a lot of "fullrangers", and it is short enough that I think the panels can be stacked on top.  I would still like to make the panels fairly large, not so much for the bass as to keep a big dynamic midrange.  I am unsure what sort of bass will mate best with a dml but, since this is most likely a bending wave down lower anyway, I think dipole bass may be the ticket.

I've heard that with age, bass becomes more important to us, and that seems to be so in my case.  Besides, I'm tired of having limited bass in my systems.  I don't need to crack the walls or rattle the windows, but I would like to hear the bass expand out with the rest of the music.  It did this in the old humongous systems of the fifties, and I think that's why corner horns appeal to so many folks.  Get in the same room with a cello sometime, and you will realize that you have probably never heard one before unless you actually go to a lot of live music events, and cello doesn't even go all that deep.

I am looking forward to when I get moved and can start to work on this.  It seems to be almost exactly what was on my wish list.  I am grateful to Ziggy for pulling this project out of the mist, and at such an affordable cost, too.

[captainjack115]

Electrostatics are dipoles, very fast and accurate, but they need to be large to get bass. NXT type panels need to be large too, but mostly to move anough air.

Why are NXT type panels called distributed mode speakers? Because they are neither pure dipole neither pure bipole. They have opposite phases 'distributed' on every side of the panel. They don't pressurize a room like monopoles and bipoles would do at very low frequencies. But they don't have a clear dipole separation either. And their ability to produce bass is very well defined by their size in the first place. A small DMS will NOT produce bass, regardless how much air you are able to move with it.

[/b]

The late Paul Klipsch and I had some interesting chats by phone and snail mail.

One of the things that most concerned him about bass was to keep diaphragm movement to a minimum. his main interest was of course, horns. There's only a few ways to move lots of air, a horn captures air molecules in it's mouth and small diaphragm movement moves lots of air. That leaves us with various other conventional ways to move air, I.E. bass reflex, tuned port ( a variation on bass reflex), infinite baffles, etc.

Bending waves are very interesting in that they are bipolar in a range of frequencies where we get the most appreciation. To some degree they don't seem to exhibit much difference in acoustical phase as compared to pistonic dipoles. I noticed this effect by using a pair of single exciters, one exciter per side mounted on identical surfaces. Running a monaural signal (pink noise) to them, I then proceeded to flip one of the boards over. Had these been true dipoles, the sound would have been diffuse. The results I obtained were not diffuse, not identical either. It wasn't until I flipped the board back and electrically reversed the phase to that board that the sound became diffuse. This is indeed interesting, bending wave transducers definitely reverse phase electrically, but not necessarily acoustically. This would seem in part due to their ability to produce a wide and deep sound stage.

Frankly, I've never heard a pair of transducers deliver the spaciousness of bending wave technology. The freedom of being able to move about a listening area, standing up or sitting down and still be able to perceive the position of one particular instrument.

Simply amazing!!!

So!! It would seem that at some point this technique changes from true bending waves to something more along the lines of pistonic behavior. We may need to think about something Bob Carver did with his woofer system on the original Amazing Loudspeaker by having a high "Q" at low frequencies to make up for cancellation due to baffle width. It's hard to say at this point, we're not dealing with true bipole or dipole operation, but a chaotic transition.

[scorpion]

Sirs,

I certainly appreciate your posts here.

I have heard zygadr's panels and they sound very good. I think they may be lacking in the low bass and in the extreme highs. But as a compromise they are quite good also when compared to common fullrange speaker designs. And also they are cheap. I certainly advocate all steps to try new materials. But one should not be questioning diverse filtering approaches. They can give clues not yet realized.

They do sound omnidirectional, compared to Stig Carlsson's designs, familiar here in Sweden, I think that the panel's response are very much more omnidirectional than Stig Carlsson ever accuired with his dynamic speaker designs. But even those old designs did show remarkable 'no shows'. 

/Erling

[Rudolf]

It would seem that at some point this technique changes from true bending waves to something more along the lines of pistonic behavior. We may need to think about something Bob Carver did with his woofer system on the original Amazing Loudspeaker by having a high "Q" at low frequencies to make up for cancellation due to baffle width. It's hard to say at this point, we're not dealing with true bipole or dipole operation, but a chaotic transition.

I would not call it a chaotic transition, but it gets indistinguishable for the ear (and the eye in this case too).
At the lowest frequencies there is a clear difference between a panel bending in phase (left) and with 180deg phase difference (right):


One octave above the difference already gets a bit blurred:


At high frequencies we can hardly distinguish between this pattern and the one with opposite phase (not shown):


I understand how one could describe the last pattern as 'more along the lines of pistonic behavior'. But isn't this only a mental mistake while in fact it still is a perfectly bending wave?

Rudolf

[bobloblob]

A couple of interesting posts by Moray James about bending wave transducers.  He was one of the principals in developing one a couple of decades ago.  Scroll down for his posts.

http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-47.html

While I think it is interesting to find how and why these panels of Ziggy's work, I think the major point is that they do work, unless we do not believe the reports.  And, if they sound as good as stated, I am happy.  The rest goes more to satisfy my curiosity as to how.

I hope that the frequency response to the listener is accurate.  I had a friend who bought the first iteration of the Vandersteen, and was absolutely in love with them - for just over an hour.  The guy had played in a symphony when he was young, and said that something didn't seem quite right with the speakers after listening to them for a while.  He returned them.  I heard some Vandersteens later in a shop, and compared them to about a half-dozen other systems I knew to be flat.  On one of the records making the "audiophile" rounds at the time, Tarentella Tarentule, on the Vandersteens it sounded like the piccolo player was up at the front of the stage and everyone else was in the background accompanying him.  On all the others, the piccolo was just part of the orchestra.  After that, I reasoned that if a speaker didn't measure relatively flat, I wasn't hearing the music as intended, whether or not I could tell.  Most people who talk about how they have more "life" in their music without bsc can get the same effect by turning up the upper midrange/lower treble, which many shops in the 70's did on the systems they were pushing, telling the customers that the speaker was delivering more "detail".  I doubt that these would need any bsc.  So, while I am not concerned about the measured response, I am concerned about how accurate a speaker is, and usually the frequency response gives a clue about this.  There are also many ways to deal with dips and rises other than eq, so that would be my last recourse if there is a problem here.

What I would like to hear from someone is not so much what the measured response is, since that sounds like it may be more difficult to get with these, but rather how the sound compares with a speaker known to be flat, comparisons like whether the instruments all seem to be in the same location front to back, and have the same "weight".  In other words, in a symphonic piece by the New York Philharmonic, does it still sound like the NY Phil, and the same performance, on both speakers, given the fact that the presentation by the two systems will be different?

Forgive the rambling please.  Just thinking "out loud" in print.

[captainjack115]

It would seem that at some point this technique changes from true bending waves to something more along the lines of pistonic behavior. We may need to think about something Bob Carver did with his woofer system on the original Amazing Loudspeaker by having a high "Q" at low frequencies to make up for cancellation due to baffle width. It's hard to say at this point, we're not dealing with true bipole or dipole operation, but a chaotic transition.

I would not call it a chaotic transition, but it gets indistinguishable for the ear (and the eye in this case too).
At the lowest frequencies there is a clear difference between a panel bending in phase (left) and with 180deg phase difference (right):


One octave above the difference already gets a bit blurred:


At high frequencies we can hardly distinguish between this pattern and the one with opposite phase (not shown):


I understand how one could describe the last pattern as 'more along the lines of pistonic behavior'. But isn't this only a mental mistake while in fact it still is a perfectly bending wave?

Rudolf
[/quote][/b]

The graphics from your original post are interesting.

I'm curious how you obtained the bending wave graphics. Are they computer simulations, or real world tests with actual exciters? If so, have you personally conducted these tests. I ask only because I tend to be skeptical about third party testing methods, such as something found on a web site. If I see a "Wet Paint" sign you know I have to find out for myself.

I use the term "chaotic transition" because there's no proof in my mind to the contrary.
The graphs are indicating full bending wave mode, but they represent an unknown source to me.

Anyway, in keeping with Paul Klipschs' theory of small diaphragm movement keeping distortion low. Vibrating panels, dipole or bipole need to have large areas to move enough air and keep distortion low.

Personally, I love the idea of listening to a full range panel. Dividing music into multiple frequencies and drivers always end up with integration problems when everything hits the ears.
My own everyday system is a two way. I use B&G RD-75's and 4 12" woofers built into an infinite baffle. The bass is more than anyone could ever hope to have, deep, powerful and low distortion.......... However!!!!! The line source B&G monopoles can't come close to an in room DBL. I'm willing to strive for acceptable bass if I can keep the magical 3D.

Jack

[bobloblob]

"Personally, I love the idea of listening to a full range panel. Dividing music into multiple frequencies and drivers always end up with integration problems when everything hits the ears.... Jack'"

I understand.  I've gone down the full-range path a bit over the years, and I love it for what it can do.  I've also made that compromise over what I called "acceptable" bass.  I've had all sorts of what I thought were great systems, and always found integrating a sub into them to be difficult and, in some cases, impossible.  However, it was an "Ahah!" moment when I heard the horn subs, and then live cellos.  It made me extremely aware of how much musical content I had written off as not really necessary for my satisfaction.  I no longer want to toss off that low end as nearly extraneous.  I want it the way the composer intended it, and the way the musicians played it.  This may be partly because I don't hear much beyond 12K, if that, but a lot of it is that now I want to hear the whole piece, and to hear the bass notes rise along with the midrange and treble.

From the posts, it looks like there may be less treble with these panels, at least for some responders.  However, that seems to be out of the range of most instruments anyway, and up in the frequencies that we think of as "airiness", "openness", or whatever.  I don't think that should be too hard to deal with, if necessary.

A compromise I really wish to avoid is to have the lower registers squished into the background.  If it calls for a sub cut in somewhere at the bottom, I think it is worth the work to try to integrate one.  The panels seem close to fullrange, with some work maybe not even needing a supertweeter.  At this point I am far too aware of the bass that was missing or compromised to give it away, even though it seems like these will go pretty low.  I will have to wait and see what my feelings are when I get around to putting mine together.  The major fact is that these are wide-range enough that whatever dividing is done will be out of the range of most critical hearing.  If it affects the rest of the frequencies too much, well, we'll see.  I understand what you are saying though.  That describes most of my audio life.  Compromises, always compromises.  You pick what you can live with.  Then, in a few years, you want more out of your system.

Anyway, enough of my rambling.  Back to what the guys who are building and developing these are coming up with.  Far more interesting than my thoughts.

Take care

[BowerR64]

I just ordered 5 pair of these today to play around with. Im wanting to build one like the guy on youtube. Does anyone know where to get those cool little tweeters he used?

The idea of a flat board style speaker is cool but i dont have a problem mounting a little tweeter to the top to give it the slight brightness it needs nor the low end he used with that seperate sub.

THe sub i can hide anywhere and the tweeter can be positioned to give a nice direction.

I like his whole setup, the chrome stand, the black board then the neat little matching crome dust cap tweeter.

[el`Ol]

They do sound omnidirectional, compared to Stig Carlsson's designs, familiar here in Sweden, I think that the panel's response are very much more omnidirectional than Stig Carlsson ever accuired with his dynamic speaker designs. But even those old designs did show remarkable 'no shows'. 

My DIY Carlsson OA-50 inspired three-way speakers are in fact the reason why I stopped my DML experiments. They have exactly the amount of indirect sound I need.

[sedge]

Zygadr
I am glad you are back on form and ready for action. In all my tests I have been very excited with the outcome so far. The problem is just trying to understand how they function. When is an nxt not an next? Depending on panel size, weight, stiffness and amounts of damping in the panel, my plywood panel seems to go from 20 k to 250 htz in pretty much nxt mode. Below this I am sure it is a piston. As a contrast I tried Foamcore again (as the sound was dull and boomy). I measured the frequency response across the front and back panels, moving the microphone from the back of the driver to the edge of the panel you see a reduction in hf as you move along the surface. On the front surface as you move the microphone from the one inch foot of the exciter, the hf drops dramatically almost immediately down to 10 k within an inch or so. I then cut a hole in the Foamcore just big enough for the exciter to fit. What I did then was cut a disc of this ali. about two and a half inches in diameter and stuck the disc to the front panel and the exciter to the back of the disc so that the disc (drives) the front surface. The sound was much improved but Foamcore is very lossy. It still drops the hf quite fast as you move the microphone across the surface. This seems to match the omni-directional graph seen on these pages, very well indeed. You could make the disc into a strip, say 8 inches wide by the length of the panel. Could this strip of material be bolsa wood? Any ideas. This panel does not act like nxt above say, 10 k, or below 250 htz approx. The panel has to be very rigid for nxt for frequencies to travel along the full length and width of the panel.

[BowerR64]

Can you guys recomend a cheap tweeter i can add to these transducers to add the highs these lack when mounted to typical sufaces?

I want to bild a set like smokinjoe from youtube but im not sure what tweeter he used. There is a cheap 6X9 frame tweeter on parts express i could use maybe or i found some on e-bay that are paper cone type.

somthing like a 2" or somthing compairable to smokinjoe i wanted to order them so everything comes in around the same time.

[FULLRANGEMAN]

Hi Boys,
As a new panel material is hard to find or make or expensive(Nomex, cardboard hive) our standard material is GatorFoam.
I think there is a alternative way to this project we are missing:
we need to do our own big, very big exciters ourselves, I think 4 or 6 inches exciters, in 16 ohms only to do a parallel connection with 4 exciters for a used final impedance of 4 ohms.
We can not afford to use Series connection cause we want Hi-End performance and not frequencies anomalies(serie link), big exciters have a higher sensitivity and we need this to use only 4 exciters.    If anyone know as do exciters I think the exit to success is use 4 big exciters of 16 ohms.

[BowerR64]

I thought i read that these are just speakers with no cone? Why couldnt you take another driver and just cut the basket off and remove the cone then attach it the same way we do these?

[FULLRANGEMAN]

I thought i read that these are just speakers with no cone? Why couldnt you take another driver and just cut the basket off and remove the cone then attach it the same way we do these?

Maybe the suspension of a exciter is different than a cone driver, Iam not sure.

[sedge]

a few weeks ago I was listening to some music on a test disc when my friend phoned ,he said he was interested in the panels and wanted to come over and have a listen .Anyway we had a little chat and I said goodbye ,I then started to look out of the upstairs window wondering who the hell was making all the fog horn noises out in the street below,when the penny dropped [thud]I flew down the staires and shut the power off to the audio system  45 mins of test tones at near full volume [ouch].The smell of burning,and the heat of the exciters ,it was amazing .later when they cooled off  I checked them and they seemed ok but a few days later one of the plastic feet fell off .I had to cut the foot off the  experimental panel I was using , and in doing so found that the heat from the exciter had melted all the foam in the panel around it , if I had not cut the exciter off I  would not have known the panel was hollow!.  apart from that the exciters seem fine , any idea which glue is best to stick the foot back on with? If not Im down to my last 39 exciters