Does a voice coil have to be round?

[Tim Link]

Dynamic speaker drivers have a voice coil embedded in a small gap between magnets, which creates a good deal of motor force. These wires are are always wound up in a circle or perhaps an oval sometimes. This is great for point source applications, but what if we want to make a dynamic line source? I see line source speakers often made of a whole bunch of little tweeters. Why not just stretch the voice coil into a really long oval? It'd be a race track essentially with a long run of wires going up on one side through a long magnetic gap, and then coming down on the other side. This could make a single driver with enough surface area to cover a wide bandwidth with a constant radiation pattern, and all the electromotive force of a dynamic driver. You could even make it into a tall compression driver with great horizontal dispersion and high efficiency.

Is there some physical or electromagnetic reason this couldn't or shouldn't be done? I'm guessing so, because nobody's doing it that I know of.

 

[DVDdoug]

Theoretically it's "just a motor" made with a coil and magnet. So no, it doesn't have to be a cylindrical coil.

But it wouldn't be an easy thing to make and there are already so many trade-offs in speaker design and that I doubt it would be better.

This could make a single driver with enough surface area to cover a wide bandwidth with a constant radiation pattern

Radiation pattern is more-related to the relationship between piston area and wavelength.

There are ribbon speakers and ribbon microphones that use a different "motor" and "generator" configurations.

 

[Roland68]

Dynamic speaker drivers have a voice coil embedded in a small gap between magnets, which creates a good deal of motor force. These wires are are always wound up in a circle or perhaps an oval sometimes. This is great for point source applications, but what if we want to make a dynamic line source? I see line source speakers often made of a whole bunch of little tweeters. Why not just stretch the voice coil into a really long oval? It'd be a race track essentially with a long run of wires going up on one side through a long magnetic gap, and then coming down on the other side. This could make a single driver with enough surface area to cover a wide bandwidth with a constant radiation pattern, and all the electromotive force of a dynamic driver. You could even make it into a tall compression driver with great horizontal dispersion and high efficiency.

Is there some physical or electromagnetic reason this couldn't or shouldn't be done? I'm guessing so, because nobody's doing it that I know of.

Physics sets the limits.
- The size of the moving tweeter area must match the frequency range
- The bobbin and dome must be stable enough
- The haunted body and dome must be light enough, XX kHz range = corresponding movements per second
- Longer coil winding = higher resistance = thicker wire = more weight
- Be aware of bundling effects for non-round domes

 

[Deleted member 21219]

Test Bench: The D8404-552000 Ellipticor Cloth Diaphragm Dome Midrange from Scan-Speak

This characterization is the sixth Scan-Speak Ellipticor driver to appear in Test Bench. The Scan-Speak D8404 is a large dome midrange with a large diaphragm area and respective voice coil diameter. For a motor system, Scan-Speak basically scaled up its currently successful AirCirc tweeter motor...

audioxpress.com

And the theory put to the test .....

Ellipticor-2F8

Jim

 

[MAB]

It can be a range of shapes. Oval for instance.

Ellipticor – Scan-Speak A/S

www.scan-speak.dk

Race-track woofers on Genelec:

Various planar drivers and speakers.

Magnepan

Magnepan is the leader in high value, high quality, flat panel audiophile speakers for stereo and multi-channel use.

magnepan.com

Dayton Audio - AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

www.daytonaudio.com

 

[Multicore]

You could use other shapes but I think the motor efficiency is optimized when round. All else being equal the magnetic flux density for a given current through a given length of coil wire is optimized when it is round. And round shapes are relatively easy to engineer mechanically. Assuming I remember any physics correctly.

 

[Tim Link]

Test Bench: The D8404-552000 Ellipticor Cloth Diaphragm Dome Midrange from Scan-Speak

This characterization is the sixth Scan-Speak Ellipticor driver to appear in Test Bench. The Scan-Speak D8404 is a large dome midrange with a large diaphragm area and respective voice coil diameter. For a motor system, Scan-Speak basically scaled up its currently successful AirCirc tweeter motor...

audioxpress.com

And the theory put to the test .....

Ellipticor-2F8

Jim

That driver had some strange distortion.

 

[Tim Link]

Physics sets the limits.
- The size of the moving tweeter area must match the frequency range
- The bobbin and dome must be stable enough
- The haunted body and dome must be light enough, XX kHz range = corresponding movements per second
- Longer coil winding = higher resistance = thicker wire = more weight
- Be aware of bundling effects for non-round domes

What's a bundling effect?

 

[Multicore]

It can be a range of shapes. Oval for instance.

Ellipticor – Scan-Speak A/S

www.scan-speak.dk

Race-track woofers on Genelec:

This photo appears to show circular voice coil. The cone is oval but I think the motor is circular.

 

[Tim Link]

You could use other shapes but I think the motor efficiency is optimized when round. All else being equal the magnetic flux density for a given current through a given length of coil wire is optimized when it is round. And round shapes are relatively easy to engineer mechanically. Assuming I remember any physics correctly.

I know that the magnet is optimized in a round motor. The long skinny thing I'm envisioning would require a lot of magnet, but not necessarily any more than a row of round tweeters of the same length. The total moving mass weight compared to the surface area and motor strength seems like it could be comparable.

 

[Tim Link]

This photo appears to show circular voice coil. The cone is oval but I think the motor is circular.

I think so, but it's interesting that Genelec is using that shape for the woofer. The racetrack surround seems to work just fine so that part has been proven to work. The question now is whether a voice coil of the same shape could be managed. I also wonder if a long skinny diaphragm might develop some strange modes along it's length. It might want to rock or twist. It would have to be held steady somehow so that the whole length would move like a long skinny piston.

 

[Tim Link]

- Longer coil winding = higher resistance = thicker wire = more weight

The resistance could be a problem. We'd have more moving weight, but a lot more surface area and motor to push it. The problem with the resistance would have to be addressed, as it would act like a bunch of tweeters wired in series. This could be solved perhaps by having multiple winding paths in parallel to bring the impedance down to the desired level.

 

[Multicore]

I think so, but it's interesting that Genelec is using that shape for the woofer. The racetrack surround seems to work just fine so that part has been proven to work. The question now is whether a voice coil of the same shape could be managed. I also wonder if a long skinny diaphragm might develop some strange modes along it's length. It might want to rock or twist. It would have to be held steady somehow so that the whole length would move like a long skinny piston.

I think that design derives from the problem of fitting it into the enclosure.

 

[LTig]

It can be a range of shapes. Oval for instance.

Ellipticor – Scan-Speak A/S

www.scan-speak.dk

Race-track woofers on Genelec:

Various planar drivers and speakers.

Magnepan

Magnepan is the leader in high value, high quality, flat panel audiophile speakers for stereo and multi-channel use.

magnepan.com

Dayton Audio - AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

www.daytonaudio.com

I'm quite sure the voice coils of those drivers are still circles.

 

[voodooless]

There are also rotary voice coils:

Rotary Voice Coil Actuators | Sensata Technologies

www.sensata.com

They may even have interesting applications in speakers, notably in subwoofers: you could move a massive sheet with these things, and have huge displacements.

 

[Tim Link]

There are also rotary voice coils:

Rotary Voice Coil Actuators | Sensata Technologies

www.sensata.com

They may even have interesting applications in speakers, notably in subwoofers: you could move a massive sheet with these things, and have huge displacements.

That's interesting. I've thought about using a bunch of piezo benders to move a diaphragm. I saw somewhere a guy had built an RC airplane that used piezo benders to move the control surfaces. Seems a lot better than a bunch of pushrod servos.

 

[Cbdb2]

The flux in non circular magnet gaps is probably more difficult to make consistent, changing in the corners, the shape of the magnet would not follow the gap. The flux from the coil would also be different in the corners. Don't know how much effect on performance. I also think about the max area from a given perimiter is a circle, so the circle pops out of many max/min equations, probably including driver efficiency. The most obvious reason, after 100 years of speakers nobody has found a better way. Im sure it can be done but why?

 

[Timcognito]

The most obvious reason, after 100 years of speakers nobody has found a better way. Im sure it can be done but why?

Time for OP to do research on DMLs and BMRs most of which use single or multiple round motors and do not have uniform pistonic motion. They act more like a ripple in a pond have very wide uniform dispersion and weak LF but allow for wide variety of planar areas

 

[Tim Link]

The flux in non circular magnet gaps is probably more difficult to make consistent, changing in the corners, the shape of the magnet would not follow the gap. The flux from the coil would also be different in the corners. Don't know how much effect on performance. I also think about the max area from a given perimiter is a circle, so the circle pops out of many max/min equations, probably including driver efficiency. The most obvious reason, after 100 years of speakers nobody has found a better way. Im sure it can be done but why?

Good points. The reason I'm thinking is to get a large surface area that can have a consistent radiation pattern over a wide range of frequencies, all the way up to above 10k. A 1" wide diaphragm that's 36" long gives you the surface area of a 6.5" cone driver, so it should disperse horizontally as wide as a 1" tweeter and be able to cover down into the upper bass at least. This is already possible with ribbons and electrostatics but it seemed to me a linear dynamic driver would have more motor force and perhaps be more robust. Indeed, the corners are weird, so maybe it'd be best to just run the wires straight through the magnets and have them not run through magnets at each end where they curve to cross over to the other side. This is basically a planar magnetic with some straight formers attached at 90 degrees that can tuck into a tight magnet gap, rather than just printing the wires on the membrane and trying to get magnets as close as possible to them. I recall reading somewhere that something like this had been done in the past using a polystyrene membrane with grooves cut it in it, and wires on the inside of the grooves. Magnets could stick into the grooves to get very close to the wires and maintain a constant flux over an excursion range, like a dynamic driver does.

I think Linaeum did something similar with their tweeter. The circuit was printed on the diaphragm, but that was placed in a narrow magnetic gap so that it moved forward and back, and then the diaphragm emerged on both front and back from the gap to round over left and right to attach firmly to the fixed sides. The flexing roll-overs were the effective surface area of the tweeter. I had some speakers with those tweeters. They tweeted, but they were just the short versions and I wasn't especially impressed by their sound. I read that if you stacked a bunch of those tweeters into a tall line source they sounded good.

 

[Tim Link]

Time for OP to do research on DMLs and BMRs most of which use single or multiple round motors and do not have uniform pistonic motion. They act more like a ripple in a pond have very wide uniform dispersion and weak LF but allow for wide variety of planar areas

I know about those. They're interesting, especially the little ones that can act like extended range tweeters with very wide and consistent dispersion. I've read mixed reviews about them, with some consensus that they can make a great midrange driver but you still need separate tweeters and woofers. We already have good options on midrange drivers.