| Motor-based planar speakers-can it be done? - Click HERE for Original Thread |
| Spasticteapot |
I've been looking through speakers, and it appears that some rather decent speakers (which more-or-less fit the definition of "planars") use a combination of a standard electromagnetic motor and another non-electrically active material to actually make the sound. Examples include the Soundpads and "Personal Planars".
Soundpads:
http://www.thinkgeek.com/electronics/audio/6bd8/
Odd wire-magnet-and-diaphragm design I don't quite understand:
http://www.diyaudio.com/forums/show...0821&highlight=
"Personal Planars"
http://www.decware.com/panels.htm
I personally like the idea of using a standard electromagnet assembly to drive a flat panel; that way, one can use a standard amplifier without transformers or high voltage, while keeping a large degree of the excellent sound associated with planars.
Also, I was thinking about using a method of driving these speakers used in "featherweight" R/C airplanes: Rotational actuators.
You can see one here:
http://www.ekmpowershop2.com/ekmps/...CT&productid=77
By placing a magnet in the center of a coil, you can make it spin in alternating directions by putting A/C current through the coil. Once the vibrations are produced, you then just apply 'em to the speaker membrane. However, due to inertia, I doubt these would work well at high frequencies.
Alternately, one could put a coil on the membrane and have it around a centrally placed magnet, somewhat like conventional speakers.
Finally, one could also try a piston-style design, with a small magnet with a coil on each side, running push-pull. You then thread a pushrod out to the membrane; using a high-strength magnet, you'd likely get decent efficiency, too.
So, any tips for doing this?
Or am I out of my gourd? |
|
|
| dshortt9 |
| It must be something in the Bratwurst. |
|
|
| thadman |
| by using a voicecoil you raise the moving mass of the driver greatly, which defeats the main purpose of the planar/ribbon design (excellent transient response). |
|
|
| theAnonymous1 |
I have no personal experience with much of what you posted other than the "soundpads", and I don't recommend them for anything other than pure novelty.
"Distributed Mode Loudspeaker" is the technical term for them, developed by a company called NXT. Its takes a lot of research, development, and access to the right materials to even come close to making a "DML" worthy of listening to. Even "DML" products made by big companies fall extremely short of anything close to HiFi.
I have spent many hours and quite a few dollars messing with this technology, and I don't consider any of it well spent. |
|
|
| Spasticteapot |
I guess you've got a point.
One idea (shown in the second link) is to put an extrelely long, thin coil of ordinary copper magnet wire along the length of a mylar ribbon. Although not as thin or as light as a real ribbon driver, you would'nt need transformers to drive it, and it could be made relatively cheaply if a source for neodiyum magnets could be found. Efficiency likely would'nt be too bad, either.
EDIT:
Now with picture! Can you tell I made it in MS-paint? |
|
|
| el`Ol |
| Search for neodymium at Ebay yields 632 hits. |
|
|
| Ron E |
Museatex did the voice coil on planar diaphragm thing many years ago.
I know they cost a lot for what they were and they never went over well. Supposedly they sounded good when dialed in, but I don't believe what I don't hear. 90%+ of high end audio is just a con game.
Manger is essentially what you describe, but with damping elements and variable stiffness. I haven't seen a DIY Manger, why not try your own? |
|
|
| Spasticteapot |
Hmm...interesting.
I was wondering about the viability of speakers using a standard coil mounted in the center of a taut-streched piece of plastic. (This would be a lot like an electrostatic speaker.) Using a magnet on each side of the coil, you could drive it in both directions equally. As an added bonus, the field remains somewhat linear througout; as the coil moves farther from one, it moves closer to the other.
Also, could someone explain how the "personal planars" speakers work? I can't figure it out.
I wish I was'nt going on a !#$!$!@#$ vacation in three days. |
|
|
| moray james |
Highwood Audio designed and manufactured the Sumo "Aria" loudspeaker which is exactly what you discribed a voice coil mounted to a tensioned mylar diaphragm. This is the speaker which later became the Museatex loudspeaker (when Highwood Audio and Museatex merged). The speaker design worked very well. You need to use a very low inductance single layer voice coil as well as a copper sleeve on the motor pole piece and shorting rings at both the top and bottom of the pole piece. Aside from this you need to deal with diaphragm damping especially edge termination to effectively deal with reflections. The design can be made to work very well and present a very easy load to the amplifier. You will need a cone for the coil and if you can find some the type used for piezo tweeters are excellent (paper) I have some of these.
Quote " by using a voicecoil you raise the moving mass of the driver greatly, which defeats the main purpose of the planar/ribbon design (excellent transient response)." sorry but you are very wrong! Have you ever measured the actual mass of a maggie? You will discover it is far mor than most heavy cone speakers. The moving mass of our speaker with coil, dusy cap (cone) and tinsil leads is 2.2 grams. That's about as good as any tweeter out there. Our voice coil is a 1.25 inch custom made unit but if I were you and wanted to experiment I would go with a one inch tweeter motor assembly (very modified) and note that you will also need ferro fluid to act as a liquid bearing/spider replacement. Any attempt to use a spider will generate far too much noise otherwise. I hope that this helps you good luck and have fun. Very best regards Moray James cofounder coinventor Highwood Audio. |
|
|
| Spasticteapot |
| quote: | Originally posted by moray james
Highwood Audio designed and manufactured the Sumo "Aria" loudspeaker which is exactly what you discribed a voice coil mounted to a tensioned mylar diaphragm. This is the speaker which later became the Museatex loudspeaker (when Highwood Audio and Museatex merged). The speaker design worked very well. You need to use a very low inductance single layer voice coil as well as a copper sleeve on the motor pole piece and shorting rings at both the top and bottom of the pole piece. Aside from this you need to deal with diaphragm damping especially edge termination to effectively deal with reflections. The design can be made to work very well and present a very easy load to the amplifier. You will need a cone for the coil and if you can find some the type used for piezo tweeters are excellent (paper) I have some of these.
Quote " by using a voicecoil you raise the moving mass of the driver greatly, which defeats the main purpose of the planar/ribbon design (excellent transient response)." sorry but you are very wrong! Have you ever measured the actual mass of a maggie? You will discover it is far mor than most heavy cone speakers. The moving mass of our speaker with coil, dusy cap (cone) and tinsil leads is 2.2 grams. That's about as good as any tweeter out there. Our voice coil is a 1.25 inch custom made unit but if I were you and wanted to experiment I would go with a one inch tweeter motor assembly (very modified) and note that you will also need ferro fluid to act as a liquid bearing/spider replacement. Any attempt to use a spider will generate far too much noise otherwise. I hope that this helps you good luck and have fun. Very best regards Moray James cofounder coinventor Highwood Audio. |
Mind explaining that to me in more simple terms?
(Where'd I put my copy of "Advanced audio terminolgy for dummies..)
It sounds to me like you're thinking of using a standard motor, with a metal insert that goes through the coil's center.
I was actually thinking more along the lines of simply using a magnet on each side, with nothing at all through the center of the coil. Although the magnets would need to be extremely powerful, you would'nt really need any bearings or supports; the coil would be held in place by the mylar. (Or were you suggesting something else?)
I'm also not opposed to using a high-powered tweeter (which would normally be unable to reproduce bass due to size) connected to the panel through a pushrod. Not elegant or fancy, but it would give acceptable performance, especially with a pull-pull system. (Just put one tweeter on each side, and have a piece of wire going between the cones. Attach the film to the center of the bit of wire.)
Thanks for the help-I think I'll try my hand at a pair of these as soon as I'm able. |
|
|
| moray james |
Go to your library and get a copy of an article that I wrote on this design some years back. That should get you up and running. Ask for "Audio" June 1992 the article was titled "A New Type of Speaker, A Shere of Sound".
Quote - I was actually thinking more along the lines of simply using a magnet on each side, with nothing at all through the center of the coil. Although the magnets would need to be extremely powerful, you would'nt really need any bearings or supports; the coil would be held in place by the mylar.
This idea will not generate anywhere near enough force on the diaphragm. At the point in time the above article was published the business had been working on this design for almost four years. Voice coil stability in a design like this is critical and yes you will need to use ferofluid in the coil gap to keep the coil from rocking under large pulses and very loud playback. Check the patents under Babb as these are expired and they would offer excellent voice coil control. We would have liked to use this idea at the time but it was protected. We took out a US patent around 88-89 but I have forgotten the number but you could check under the company name of Highwood Audio.
I wonder if you had considered building an electrostatic loudspeaker? That project would prove a lot easier than the road you are going down now and a lot less expensive. Just a thought. Good luck Moray James. |
|
|
| Spasticteapot |
| quote: | Originally posted by moray james
Go to your library and get a copy of an article that I wrote on this design some years back. That should get you up and running. Ask for "Audio" June 1992 the article was titled "A New Type of Speaker, A Shere of Sound".
Quote - I was actually thinking more along the lines of simply using a magnet on each side, with nothing at all through the center of the coil. Although the magnets would need to be extremely powerful, you would'nt really need any bearings or supports; the coil would be held in place by the mylar.
This idea will not generate anywhere near enough force on the diaphragm. At the point in time the above article was published the business had been working on this design for almost four years. Voice coil stability in a design like this is critical and yes you will need to use ferofluid in the coil gap to keep the coil from rocking under large pulses and very loud playback. Check the patents under Babb as these are expired and they would offer excellent voice coil control. We would have liked to use this idea at the time but it was protected. We took out a US patent around 88-89 but I have forgotten the number but you could check under the company name of Highwood Audio.
I wonder if you had considered building an electrostatic loudspeaker? That project would prove a lot easier than the road you are going down now and a lot less expensive. Just a thought. Good luck Moray James. |
Why exactly is ferrofluid needed around the voice coil, as opposed to some other damping material?
I'm not building an ESL for a good reason: They're expensive, fragile, and use very high voltages. |
|
|
| dnsey |
Many years ago I owned a pair of Planax 3 speakers, made by (or more likely for) Goodmans.
They used an expanded polystyrene diaphragm about 30 X 20 X 1 inches, flat at the front and moulded at the rear into LF and HF areas. Each area had its own motor, of a conventional design, and the entire diaphragm was mounted in a hardwood 'frame', resulting in a dipole transducer.
They didn't sound too bad - the bass was a bit flabby, and the treble rolled off quite early, but the basic design seemed sound and capable of further development. |
|
|
| moray james |
if you build a driver like the Sumo Aria which was a 1.25 inch dia voice ciol mounted directly to a mylar diaphragm and then drop that into a motor assembly then the diaphragm will act like a huge top suspension to hold the coil in place. There is how ever nothing to act as a bottom suspension to prevent rocking and that is what the ferro fluid does for you. Any kind of conventional spider will generate output of its own which will sing straight through the mylar diaphragm so that is a non starter. I mentioned the Babb patent to you as it is out of date now. Babb designed vertical teflon ribs into the pole piece to act as a bushing so the coil cannot rock how ever ferro fluid would still be beneficial to reduce friction and improve thermal dissapation. Your motor top plate will need to be 3/8 inch thick or better to obtain wide band width with level. Type five neo magnets will help shrink the motor size, we used type seven ceramic as neo's were not commercially available at the time at competitive prices.
Quote:
"I'm not building an ESL for a good reason: They're expensive, fragile, and use very high voltages".
Since you have no experience in building the kind of motor structure that I am discribing and since I have built both this design as well as ESL's I will tell you again that an ESL project (one off basis) will cost you less that this project will. We spent just under two million dollars in the first year and a half getting ready for production of the "Aria" which was distributed by Sumo of California. An ESL is far les exensive to build especially if you purchase a set of say used Acoustats to use the interface and high voltage supply. I will also strongly dissagree with you that ESL's are "fragile" this is absolutely wrong. Jim Strickland of Acoustat boasted that they never had a single field failure of a panel in all the years of production. Failures were due to physical damage to the panels or connecting wires or missuse of the interfaces. Yes ESL's use high voltages but they are confined to the interface box and for all intent and purposes are a non issue. Last night I auditioned a new cable design (my own) on a set of 20 plus year old Acoustats. The panels are dead stock while the interfaces have been modified for improved performance. These speakers (0ne plus 0nes) sound spectacular. At the time that I developed the Highwood Audio speaker my reference was a set of similar Acoustats. Combined with a set of good super tweeters and stereo subs these speakers can give anything you like a serious run for thier money and better most.
So if you want to build a magnetic planar like the Highwood speaker you should at least know that you will have to design and build the motor structure prototypes all from the ground up as there is no cmmercial driver designed like what you will need. No off the shelf parts are going to get you where you need to go. You will need to design and build a very strong frame to hold the motor (magnet assembly) in place with perfect alignment. You will need to have precission machined jigs to both align and punch the voice coil hole in the diaphragm as well as similar jigs to align and mount the voice coil to the diaphragm. If you honestly believe that a project like this will be less expensive than a home brew ESL then you have never done either and I am sorry again to say that you are mistaken. I would be very happy to see you go the planar route as it is a good design and can work very well. You should know that it is really more like a tweeter with the worlds biggest suspension than a planar magnetic like a maggie. Because this speaker generates spherical wave fronts off the diaphragm (a mechanical version of the Quad 63) you will have to deal with both side wall reflections as well as floor and ceiling bounce so they need to be well away reflective surfaces as they are so non directional. This is a problem which does not exist with a line source ESL design. If you are truly interested in this project pull the Highwood Audio and the Babb patents and get familiar with what has been done and was successful (Highwood died due to business reasons not a bad design). I would be happy to share my experience on such a project. If however this is simply an arm chair design excercise I think that I have contributed about as much as is useful. Good luck and please post your progress on the speaker. Best regards Moray James. |
|
|
| bear |
Moray,
I am confused by some things you posted in this thread.
How is there any difference between the mass of a "maggie" style speaker and one where the same diaphragm is driven from a single, or even multiple "standard" voice coils??
Are you saying that the differential in mass is the applied conductors' mass?
Then what about the single point drive to a mylar diaphragm acting like a point source with a very very unstiff and very large suspension - ie. like a jump rope being waved??
The first "planar" run with voice coils that I recall was (iirc!) the Beostatic?? Or something like that out of Italy I think. Expanded Styrene diaphragm...
The original Magneplanars did use wire attached to the mylar with adhesive. They looked like a paper run through a shredder, if you put too much power in or had a parasitic oscillation! Heat + mylar = electric knife. ;)
I agree with Moray, building an ESL is a really good way to go to build a planar speaker with excellent results for not too much money. The "high voltage" is high, but is very very low current when done properly. Modest care is required. Less than with the typical tube amplifier.
There are numerous threads in the archives here on building magnetic planars and real ribbons of various sizes and shapes...
_-_-bear :Pawprint: |
|
|
| moray james |
As I said the Sumo "Aria" had a diaphragm of 150 gage mylar driven by a single point source voice coil. The entire miving mass not countint the diaphragm was 2.2 grams. In a maggie you are driving all the wire of the heavy voice coil (intentional mass especially in the bass section to lower the diaphragm resonant point) as well as all the adhesive. Maggies are for the most part fairly high mass drivers when you measure the mass per unit area that is actively being driven. It is like comparrring the mass of a large heavy woofer to that of a tweeter assembly the difference is huge.
Quote: Then what about the single point drive to a mylar diaphragm acting like a point source with a very very unstiff and very large suspension - ie. like a jump rope being waved?? Not sure that I understand what you are getting at here? The diaphragm in a driver like this has to be tight to give you efficiency. The speaker that you are referring to is probably the Bertangi (sp) which was a styrene moulded diaphragm. That diaphragm was ment to be still and rigid. Manger had to actually put off patent application as Highwood Audio had allready received a patent in the USA on a traveling wave speaker. My speaker was intnded to have waves travel across the face of the diaphragm (like the ripple in a pond effect). The traveling wave combined with the initial central pulse generater spherical wave fronts launched off of a flat diaphragm.
I hope that spasticteapot gives this design a go but I felt that he needed a heads up as to what would be involved. A handy diyer with a home machine shop could do the job. Regards Moray James. |
|
|
| Nanook |
should realize that Moray designed the Aria loudspeaker (If my poor old neurons are firing across my equally aging synaptic gap correctly). And yes, doing some reading prior to throwing out the idea may help.
As far as the development of the loudspeaker goes, believe me, Moray would be the one to ask. If he suggests that ferroid fluid is needed, he knows.
and recall Sony's attempt, the apm series several years ago. |
|
|
| el`Ol |
| quote: | Originally posted by Spasticteapot
Hmm...interesting.
Also, could someone explain how the "personal planars" speakers work? I can't figure it out.
I wish I was'nt going on a !#$!$!@#$ vacation in three days. |
Looks as if they worked by inflation. I ask myself why they don`t use compression drivers.
Does anybody know something about this planar?
http://www.sonus.de/php/p_produkte.php?produkt_id=347 |
|
|
| bear |
Moray,
I'm not sold/convinced on this "traveling wave" idea in terms of planar speakers.
The Quad concept is a "traveling wave" , but there is only a singleton wave (in principle) which exists, since the diaphragm is energized sequentially with respect to the speed of sound vs distance from the center of the diaphragm (at least that is the idea). Thus emulating as spherical source.
In the system you have just described one gets the "jump rope" standing wave effect, as far as I can see. One gets in phase and out of phase radiation from the diaphragm as the wave travels across the surface, or so it would seem.
Even if you argue that a non-repeating excitation will create a singleton, clearly a sine wave input of some frequency will create a standing wave on the surface, yes?
And, the amplitude of the wave will diminish with time/distance from the center in general. (standing waves being a possible exception)
Just curious,
_-_-bear :Pawprint: |
|
|
| moray james |
say you are standing behind a large diaphragm and just for now lets assume that there will be no edge termination reflections. If you poke the compliant diaphragm with your finger in the centre (which would be what a pulse with the voice coil does) you will create a forward traveling (in the air) pulse of air on the front side of the diaphragm. The instant your finger is removed the pulse also starts to travel outward in all directions across the diaphragm (like the ripple in the water). The leading edge of that circular (and expanding) wave is pushing air outward from the trailing edge of the central pulse you just generated with your finger. The two seperate wave fronts (one in the air going forward on the central axis and the other in the diaphragm traveling across the plane of the diaphragm) are moving at about the same speed. The combined result of these two actions generates a half spherical wave to expand on the front side of the diaphragm.
Another way to imagine this would be to think of blowing a bubble with a wire hoop that is expandable in size (from small diameter to large) as you blow. The soap film is your diaphragm and your breath is the inital pulse while the expansion in size of the hoop (and then so your bubble) is the wave traveling across the diaphragm. You end up with a great big bubble the size of which is determined by the size of your hoop (or the speaker frame). If you increase the size of the hoop at the exact same speed as the forward traveling pulse you will end up with a spherical bubble. If you expand the hoop size too slowly you end up blowing a hotdog shaped bubble and iif you expand the hoop size to quickly you end up with a flattened out buble.
You may not believe that this works but I can assure you that it does. We had our Highwood Audio speakers measured extensively at the acoustic lab department at the university of Alberta and by the JBL labs in LA and on our own system. We were able to take a 1/4 inch B&K microphone and place it at the outside edge of the speaker (at 90 degrees) and measure 10 K responce try that with an electrostatic panel and see what you measure (we did)!
A lot of hoop jumping has to be done with the diaphragm material and with its tension, diaphragm damping and especially with the edge termination damping to make this all happen and work well but it can be done. I used to like to think of this as a horn loaded speaker with no horn. The diaphragm can be made to act like a wave guide to expand the initial pulse much as a wave guide (horn) does. I hope that this helps some. Best regards Moray James. |
|
|
| lndm |
Forgive me if I'm wrong, I know you've put much work into this.
| quote: | Originally posted by moray james
The two seperate wave fronts (one in the air going forward on the central axis and the other in the diaphragm traveling across the plane of the diaphragm) are moving at about the same speed. The combined result of these two actions generates a half spherical wave to expand on the front side of the diaphragm. |
If I think about this in terms of the forward moving waves, I feel as if the result of this linear forward component and a linear lateral component will produce a conical wavefront.
| quote: | | Another way to imagine this would be to think of blowing a bubble with a wire hoop that is expandable in size (from small diameter to large) as you blow. |
Wouldn't the bubble start as a point (cone) and distort to produce a dome shape due to the air trapped behind it, not at all like a wavefront? I suppose I just feel that there could be a better angle for analogy. |
|
|
| moray james |
| Yes I did say spherical and in truth it is only half of a sphere. Perhaps my discription is poor but the initial pulse of air disturbed in front of the voice coil is essentially the diameter of the voice coil there is a slight amount of tenting happening due to the diaphragm but not much. For arguement sake the forward wave is only as wide as the coil. The rollong action of the wave expanding in the plane of the diaphragm (at approx the same speed as the wave launched into the air) expands the back trailing edge of the forward moving wave and so over time develops a half sphere. As I said polar plots confirm the dispersion. Sorry if my choice of terms may not be the most acurate but I have a flue like infection and don't feel too swift. I don't have technical training in this area so I may not be correct in the use of my terms but I think that you get the drift. All the best Moray James. |
|
|
| bear |
Moray,
I will take your representation that you measured some 10kHz response at 90 degrees... not sure that is all good or not... but does indicate that the idea works to some extent.
My concern is the anti-phase wave that is created.
I think that you are claiming a "singleton" - that is a wave that is singular, only positive? Sort of like 1/2 of a sine wave, no negative going portion.
In the case of your traveling wave on the surface, it is being "followed" across the surface by an image, negative wave, (of nearly equal amplitude) which is merely delayed in time compared to the initial positive wave. So, I am suggesting that there would be some cancellation taking place at any point in time beyond the initial stimulus??
Then too there is the question of the speed of sound in the diaphragm material playing a role in the shape of the wavefront.
And, I would think that the system is somewhat frequency limited/dependant upon the stiffness of the diaphragm material/ diameter of the VC?? The area of initial "excitation" will determine the volume of air that can be moved, so intuitively a 1/2" VC drive might make a fine tweeter, but doesn't seem terribly useful at lower frequencies since the excursion is limited??
Not sure how this idea translates into a practical speaker with useable sensitivity/efficiency?
_-_-bear |
|
|
| moray james |
| The heil unit is a piston. My design is with a stretched planar diaphragm upon which concentric traveling waves move. You could think of it as a mechanical version of the Quad 63 concept. Regards Moray James. |
|
|
| ears4hire58 |
I would not refer to the Heil driver as a piston (the diaphragm is squeezed on edge; but the planar drivers in the American Power & Light LT7 are stiff/square/carbon-fiber pistons (w/foam sur.). American Power & Light is working on a new variation on the Heil driver.
My post was an attempt to reply to the thread starter (but I couldn't find the quote box).
Is there a we site to promote this new speaker w/concentric traveling waves?
regards,
DK |
|
|
| moray james |
| I was referring to the LT7 whic is a piston driver (I had assumed that it was a Heil design and so the reference by name rather than a specific design. I have a Us Patent on the design that I mentioned but will have to look it up for you. The design was manufactured in Canada by my old company Highwood Audio and soold under the Sumo brand name then later Highwood merged (bought) Museatex and distributed under the Museatex "Melior" name. Hope this helps. Regards Moray James. |
|
|
| Paul Burton |
For those of you following this thread who may want to take a look at the patent, you can view it online at esp@cenet:
The best document is the European application, EP0296139, as it has all the drawings (called "Mosaics" by esp@cenet). The direct link below might work, otherwise you will have to do a number search:
http://v3.espacenet.com/textdoc?DB=...X=EP0296139&F=0
There may have been other IP generated after I left Highwood (I don't know), but this is the original patent.
A strange thing is that this patent actually contains two inventions: A line source and a point source. This is not allowed under international patent law (as I have since learned), but the patent examiners never noticed. Had they done so, they would have required the filing of "divisionals" and there would have been two patents. Had it ever been wheeled into court it would probably have been judged prima facie invalid on this basis.
The line source was never commercialised because it was too inefficient and didn't have enough bandwidth. In other words, it didn't really work... The existence of a patent is no guarrantee that the invention is workable, or useful.
The Sumo Arias were reviewed in Absolute Sounds (Vol. 62, I think) and there was a good explanatory article in Hi Fi News, although unfortunately I have lost my copy, so I can't give the issue number. Anybody got a back collection who could look it up, sometime in the late 80s?
There is also a description in Martin Colloms' book "High Performance Loudspeakers", including the new edition, right next to the section on Manger.
Moray is quite right, the speakers did work well (eventually, after the liberal application of $$$), and produced a flat response from 40Hz to 20kHz, with constant wide directivity. They did this without any form of equalisation or crossover.
He is also correct that they were very hard to get right, and proved James B. Lancing's (JBL) adage that "Loudspeakers are 90% glue" to a greater extent than anything else I've worked on before or since. Not for the fainthearted or shallow of pocket...
So, I'd second Moray: Electrostats have to be a much easier route to high performance home made di-poles.
Bear: As Moray says, travelling wave radiators do work. They are a true low mass "coherent" source, behaving as a virtual point source located behind the panel. You might like to contrast them with their conceptual opposite, the quasi-random Distributed Mode Loudspeaker (DML), which is an "incoherent" defuse sound source. (see the technical white paper at NXTsound.com for an introduction).
The point source is an acoustic radiator who's behaviour is easy to analyse and model, but is damn near impossible to build. On the other hand, a DML is nearly impossible to analyze and model (the maths is extremely complex), but once you have managed to do so, it is really easy to build one.
...And now there's the Balanced Mode Radiator (BMR), courtesy of Dr Graham Bank of Celestion SL600 fame. That's what this forum should be concentrating on, never mind the Heils, Mangers and Arias of 20 years ago. Much more interesting.
BMR patent: WO2005101899.
There is a recording of Dr Bank giving an AES lecture on BMR here:
http://www.aes.org/sections/uk/meetings/a0701.html
He makes it sound so obvious... But it isn't. It's very, very clever stuff. It should by rights transform the industry. But I wonder if it will? Ten years ago, we all thought NXT was going to do that, and look what happened.
Regards,
Paul |
|
|
| bear |
Paul,
I see it is your patent. :)
Myself, I never thought that NXT was going to revolutionize anything but "MUZAK" for places where people wanted better sounding MUZAK! I heard them a number of times, and yes they sounded "inchoherent"!
;)
Speaking only for myself, the Quad 63s which do much the same trick, sound nice, but there is something that to my ears sounds "wrong". Not sure what exactly. And, even though they should have good HF response off axis, they don't seem to image worth a darn left or right of the center listening position (whereas some ESLs I listen to seem a bit better on that account...)
I'm downloading the USA version for reading now... Can't stand that Euro site, too wierd for my taste!! Hard to follow, and I can't figure out how to make it print or save anything... (I must be "web challenged" ,eh?)
_-_-bear |
|
|
| Paul Burton |
Hello Bear,
Yes the patent is mine. Moray has written: "I have a Us Patent on the design that I mentioned but will have to look it up for you." Maybe something was filed after I left Highwood, but if so, I can't find it at the USPTO website (or the world-wide database used by the EPO/Esp@cenet). Moray was a founding shareholder of Highwood Audio (as were others), but he was not the inventor - which is why his name isn't on the patent.
My US patent should be treated with some caution as far as the claims are concerned (I am assuming you mean US 4,924,504) as the US examination process happened after I had left Highwood Audio, and it does not appear to have been well fought by the company. The outcome was that most of the claims were disallowed, and the patent is a bit of a hollow shell as a result. It only has two allowed claims, whereas the original Canadian one has 33 (CA 1,284,837). This often happens with US patents that are not properly supported through the examination process, usually due to the high cost.
However, the description and drawings are intact, so it is useful in that respect.
I agree with you about the ESL63s. I have a pair (currently in pieces in a box - I must do something about that...). Despite being quite heavily modified (removal of grilles, dust covers, nasty electrolytic cap, rejigging the delay line to compensate etc. - all the usual stuff you can find described online), they still managed to sound, um... overly polite. Much better than standard ones, but still not quite there.
Partly, this is because they are quite directional above 10kHz, due to the large size of the inner stator circle. I tackled this by adding a 25mm neodymium magnet soft dome tweeter in the middle, which there is room to do between the two central panels, if you nip away some of the plastic louver walls. The tweeter was a Vifa D26-55-06 (now discontinued, although there are other similar units available from Tymphany/Peerless - this was about 4 years ago). I took the tweeters apart and removed the ferrofluid, so as to get the absolute best out of them, and brought them in with a first order filter (2.2uF and an Lpad, though I can't remember how much attenuation I used). That helped a lot.
Only, it didn't really do the trick, and I soon moved on. So they can't have been that compelling... In the end, I think you are paying an inevitable price for listening to miles and miles of copper (or is it the energy stored in the transformer core?) and a load of complex electronics.
NXT: Bending wave panels can be made to sound like all sorts of things. They can perform to the very highest performance levels, if you take the trouble to get the engineering right. They can definitely be a lot better than muzak, although yes, they are always diffuse sources. But quite a lot of people (especially non-audiophiles) actually prefer this sort of presentation, finding it more natural.
The trouble with the bulk of commercial NXT products is that they are not intended to be Hi-Fi, and are usually too small and low powered to perform to hi-fi standards. I worked at NXT for six years, both as an engineer and as intellectual property manager. During that time I designed and made prototypes, and heard other people's panels, which certainly qualitied as high-end Hi-Fi. It is a pity so little made it into the commercial world...
Regards,
Paul |
|
|
| moray james |
the patent was changed after you left the company. Both myself and John Wright 's names were added to the patent. KJ did not keep up the patent maintenance and so it probably fell back to what it was prior to the beginning of the changes.
I consider the speaker and patent to be equally shared by the three of us. You may not agree but it matters little now. |
|
|
| Paul Burton |
Moray,
Just for the record (as this forum is public), I feel I have to make the following comments. I confine myself to matters of fact only:
You wrote: "the patent was changed after you left the company. Both myself and John Wright 's names were added to the patent. KJ did not keep up the patent maintenance and so it probably fell back to what it was prior to the beginning of the changes."
I don't think so. Patents are absolutely a matter of public record. Everthing relating to a patent application is published and these days, is available online. Things don't "fall back" and patent inventor lists don't normally change once filed, at least not without there being a clear record.
Specifically:
[1] Neither you nor John are named on any of the published patent documents that I have been able to find using the worldwide patent database maintained by the European Patent Office (Esp@cenet), the USPTO, or Canada (CIPO).
[2] The published documents I know about are: CA1284837, EP0296139 (A2), US4924504 (A1), JP1132300 (A), and EP0296139 (A3).
[3] Changing the inventor list on a patent aplication is rare and a serious matter. It must be carried out by the original applicant (that's me) and there must be a very good reason for it. As far as I recall I didn't do any such thing, and in any case, there is nothing I can find in the public record to suggest any such change was made. As an example of the regulations about this, I refer you to:
http://www.uspto.gov/web/offices/pa.../appxr_1_41.htm
[4] You and John are of course free to THINK anything you like about who invented things, but the fact of the matter is that you aren't named inventors on any of these documents, and there is a good reason for this: You weren't the original inventors.
[5] You both made contributions to how the invention was made to work in practice, but as far as I know, the company did not file any new patent applications relating to that work (which is how one normally adds new inventive content and inventors to a patent family, not by changing an existing application).
[6] If there were any such subsequent patent applications filed, they must have been abandoned without publication. This is a common practice, but effectively renders them as if they never existed.
You may be right to say that it doesn't matter much now, but I do assert my right to be recognised as the sole inventor with regard to these patents.
Paul |
|
|
| moray james |
"If there were any such subsequent patent applications filed, they must have been abandoned without publication. This is a common practice, but effectively renders them as if they never existed"
I all ready told you that KJ started the process but probably let it go. John and I were there and you were not. So feel free to call me a liar if it makes you feel better.
I still consider that the patent and the speaker are the result of the work of the three of us. At the very least you and I should have shared the invention title on the first application. You can argue legal all you want and be as right as you want but I don't care. Legal does not make right it just makes legal. Posting a page full of facts does not make right either it just makes factual. You see Paul your real problem is that you can't make this right. You gave up all your chances to make any of this right a long time a go.
I think that you will have to learn to live with this. You have had some practice but it seems that you will have to keep working on it a while longer. You are a very clever chap (I always thought so) just keep your spirits up and stay the course. You know it might help if you start painting again. Are you still painting? I still have that one of the two old men by the sea shore that you did. You remember I traded you a bunch of tubes for it? It has hung on my wall all these years, I've always liked it. Painting would be good for you it helps to deal with things. You could get your creative side working again and do something special. That would help.
Paul I have forgiven you for what you did but I have not forgotten. I think that you have forgotten but now you need to start to work on forgiving yourself for what you did. I hope that you do so because inside you are a very special person. You have the eyes of an eagle so you can see better than most. You are very clever so look and see, you can figure it out.
If your mum and dad are still alive please say hello for me. In the mean time I will keep you in my thoughts. Good bye Paul. |
|
|
| Paul Burton |
Moray,
Don't you think you are getting a bit personal? I confined myself to matters of fact on this public forum. You haven't.
You wrote: "So feel free to call me a liar if it makes you feel better". Okay, I will: You have gone from claiming that "I have a US Patent on the design" (Post 29) , to "Both myself and John Wright 's names were added to the patent." (Post 33), to "KJ started the process but probably let it go" (post 35)... Nope, it doesn't make me feel better.
If you thought your name should have been on the patent, you should have done something about it at the time. But it was my idea, not yours, or ours. That's just the way it was: As I recall, our first meeting was to show you a working protoype I had already built. It was pretty poor stuff compared to what came later, but it embodied the invention. It was already there.
Also, I don't think John had even joined the company when the patent was filed... but I could be wrong. It is so long ago.
The rest of what you wrote in your last post has no business being on this forum. I'm not even sure what you are on about half the time. Please refrain from further personal comments, or I shall report you.
At least you have solved one small mystery for me: That picture. I have a photo of it - but had forgotten where it was. Thanks for that. My parents are well by the way.
Good bye to you too.
Paul. |
|
|
| pinkmouse |
:cop:
Guys, chill. |
|
|
| Paul Burton |
Agreed. I shall go for a walk in the rain.
P. |
|
|
| el`Ol |
| I intended to build a segmented round speaker with balsawood segments and an NXT exciter in the center. The 20 segments with 10x30cm have been lying around for a year or so, the exciter is still missing. Do you think I should complete this project? |
|
|
| Paul Burton |
el '01,
Can you explain more about the radiator you have constructed? Are the 20 segments like slices of a cake? And you don't say if this is flat or a cone. If it is a cone, it might be interesting.
But if it is flat, in general round bending wave speakers driven from the centre perform very badly. They are symmetrical and suffer from severe tympannic modes (drumskin resonances). I refer you to my earlier post in this thread and the links there to Dr Graham Bank's BMR invention. This tackles in a very thorough way how to go about making symmetrical radiators work properly, and also smoothly manage the transition from pistonic to bending wave behaviour. The results can be spectacular...
You could follow his scheme, but be aware it is patented, the rights to it being held by New Transducers Ltd (NXT). So while you can make a pair for your own use, you cannot sell any without a license.
Paul |
|
|
| el`Ol |
Yes, Paul, I mean a flat cake. Not sure whether to connect the segments with textile stripes or not.
Here is an NXT speaker that works with wooden panels (at least partly, a report in Image Hifi said it is in compound with fiber glass. The speaker was said to be really good, not mid-fi like other NXT speakers.
http://www.goebel-audio.de/ |
|
|
| Paul Burton |
el`Ol,
If you don't fill the gaps between your segments, you will have a dipole radiator that is full of holes. This will reduce low frequency output due to leakage of the out-of-phase backwave to the front. So I should definitely fill them.
The Goebel-Audio speaker did show up at NXT in Huntingdon. I seem to remember that it sounded quite good.
Paul. |
|
|
| zelter |
Hello el'Ol,
where are you going to buy the NXT exiters?
Paul, Moray,
do you still have an idea how efficent (or how inefficient) the line soure and the point source where?
Greetings
Zelter |
|
|
| el`Ol |
| To my knowledge the SI exciters are the only ones that are available without panel, and I doubt they are high quality. My dream would be Fane, but the price for the complete Micropro is a bit high. |
|
|
| Paul Burton |
Hello Zelter,
The Sumo Aria was about 83dB. So were the later Museatex Meliors. We pulled the usual trick of quoting a 4 ohm speaker to the standard 8 ohm watt (2.83V), putting 86dB in the specification. Naughty!
The line source was only ever a prototype. It was hopelessly inefficient: Mid 70s as I recall.
Paul |
|
|
| zelter |
Hello Paul,
thank you for your answer. Is there any change to raise the efficency of the point source to over 90dB?
Best regards
Zelter |
|
|
| theAnonymous1 |
Here is a pic of what the exciter in the SI Soundpads looks like. I have four of these I would be willing to let go for cheap. I've had my fun with them; I'm over it.
 |
|
|
| Paul Burton |
Zelter,
The Arias were pretty well optimised in terms of balancing the needs for low moving mass, low inductance, efficiency and long throw, given the magnet materials we had then.
Efficiency is defined by the moving mass, the coupled radiating area, and the motor's BL product - that's B (field strength) x L (length of wire in the gap), usually quoted in Tesla Metres (Txm).
So in theory there's four ways to increase efficiency: Reduce the mass (but we already did that), increase the coupled radiating area (not so much make the diaphragm bigger, as make it stiffer), increase the L (but it then gets heavier - or shorter) or use a more powerful magnet system (possible nowadays with neodimium, but a serious engineering job).
Maybe the thing to do is to buy the most efficient full range driver you can find and cut it up for parts to make your own exciter. But most of these sort of units have very little linear throw - they try to get as many turns as possible in the gap, so you won't get much low frequency headroom. But it might be worth a try.
Those exciters look like genration one 19mm units (a guess, comparing tag size with the coupling ring). The BL probably doesn't even make 1 Txm, so they won't be very loud. You need to be looking for a motor with at least 5 Txm.
There is a range of high powered exciters with radial magnets made by Billion Sound, and similar to the Fane units. eg:
http://www.billionsound.com/eBusine...2&productid=315
Maybe you could ask for samples... But while you'd get lots of mid-band level, the coil is almost the same length as the gap, so they will be no good at all for bass.
Paul |
|
|
| zelter |
Paul,
thank you for your help. I intend to use the magnet of a prosound mid or midbase speaker to do some experiments. How much y-max did you use in the Sumo Aria?
In an earlier post Moray wrote you used 150 gauge Mylar HS.
Do you still know the dimensions of the membrane?
Up to now I didn't find any review in the web. Hopefully there is someone who has some additional information.
Zelter |
|
|
| Paul Burton |
Zelter,
They were 50 x 75 cm.
By Y-max, do you mean X-max? I can't remember, to be honest. But it was something around +/- 6mm linear throw.
There is very little about the Arias on the net. There was a feature article in HiFi News, a review in The Absolute Sound, and another feature in (I think) Stereo, done by Moray after I left. I suspect there may have been more reviews of the Melior derivatives, as they remained in production for some time, but I never saw any myself.
The reason I am vague about this is that, over the course of many house moves, I have lost the file in which I kept my copies. So I too would be interested if anybody has anything they could post on this forum.
Paul |
|
|
| stoolpigeon |
moray james said:
"We spent just under two million dollars in the first year and a half getting ready for production of the "Aria" which was distributed by Sumo of California."
Moray, can you give a rough breakdown of how the 2 million was spent. Are you just refering to tooling up or does this include R&D etc?
It seems like a lot of money.
Thanks, SP. |
|
|
| moray james |
| I don't haver the numbers on paper so I will guess for you. Over the time of a year and a half we probably blew through a half million or the best part of it on everything from bic pens to crazy glue, R&D the lot. Over time we spent more than two million that's for sure. By the time that Museatex had done with our invertors it was a lot more than that. |
|
|
| Paul Burton |
There never was $2,000,000.
Highwood (developers and manufacturers of the Sumo Aria) did have venture capital funding, but it only amounted to $500,000 (CAN), taken in two tranches. I think there was a small third round, just to keep the company going, but I had resigned by then.
Any later funding will have been related to Meitner / Museatex / Meloir. This will have been mostly electronics, not the speakers by themselves.
Moray: Stop living the lie. $2,000,000?. Try to remember what actually happened. It might do you some good. |
|
|
| Paul Burton |
I checked back through this thread and found a reference to the article I referred to in yesterday's post #50. It was in Moray's post #13. I quote:
"Audio" June 1992 the article was titled "A New Type of Speaker, A Shere of Sound". (I assume he meant "Sphere". )
Anybody got a copy?
Paul |
|
|
| zelter |
The Absolute Sound has reviewed the Sumo Aria in issue 62 according to an information I found in a review of a Manger speaker system. Anyone's got a copy?
Regards
Zelter |
|
|
| el`Ol |
I just built this magnetostatically driven speaker with balsawood diaphragm:
There is a second arch below the table that can´t be seen in the picture with a rectangular voice coil in the region where they touch eachother. The project originates in two goals that suddenly found together: Building a line-array of bending wave transducers to overcome the max-SPL and THD problems of the manger and building some speaker with wood to explore the phenomenon of "material sound". As expected there are nasty resonances. But there are other aspects that are quite unexpected:
The material has no "material sound" at all, especially no wooden one.
It is not as precisely localizable as other speakers.
Most unexpected is that the sound is almost independent of the listening position.
The level is not very dependent on the listening distance.
The CSD is quite slow even in the region above 2kHz, where it is resonance free:
Seems that I have built no bending wave transducer but a DML. What I indend to do next is experimenting with flat diaphragms made of natural materials and NXT exciters. I heard both the Göbel and the Podiumsound at the Highend this year and my first listening Impression of my elaborate tells me it should be possible to top the performances of these two.
The problem: Where getting exciters (Fane prefered) as a private person? |
|
|
| wilbur-x |
The EC Motors would maybe call for renewed thoughts ?
The EC motor have parallell coils in the stator. The rotor consists of permanent magnets. Electronically switched coils controls the rpm, making frequenze regulators obsolete..
What if.... Class D-Amps. switching direct in ribbon arrays...
Or EC motors controlling a real woofer ? |
|
|
| el`Ol |
Something like that is one of my options for this Project:
http://www.diyaudio.com/forums/show...72&pagenumber=3
Seems that the flatpanel project is dead. I think I have used Google extensively and I just found one company/person who could manufacture the diaphragm to my specs, so I didn´t dare to ask for the price. Not to speak that exciters are unavailable. |
|
|
| Charcoal |
Hallo el`Ol,
You already do not consider "Rubanoide" type for design? |
|
|
| el`Ol |
| quote: | Originally posted by Charcoal
Hallo el`Ol,
You already do not consider "Rubanoide" type for design? |
Not yet. I´ll do the less neodymium wasting project first. But I am still interested in what a polycrystaline graphite diaphragm will do. |
|
|
|
|
|
