Ask an Acoustic Engineer (me)

[Justin Zazzi]

This is an experiment where you can ask questions and an acoustic engineer will try to answer them (me).

I do not mean I am the best person to answer your question ... there are many talented people here who can help! What I mean is, I wish I knew someone like me earlier. I want to be that resource for you. You can read my professional bio over here.

I welcome questions about working in acoustics as a career, or how to go to school to prepare for this field, what enrolling in a master's degree for acoustics is like, or anything similar. I can also speak a little bit about the tools available to professionals.

I also hope this can be a place to ask those challenging questions you've been unable to find an answer to, or perhaps it is hard to find a definitive answer to something. I also hope that I can learn from you since every time I teach something, I learn something too. I'm inspired some by Andy Wehmeyer and his relentless engagement and helping people, and I am inspired by my audio mentor who has guided me along the way.

What would you like to learn?

[BigAl205]

Why do speaker drivers use a voice coil and permanent magnet instead of opposing electromagnets?

[chithead]

I'm curious about the typical speaker and subwoofer design. If sound is caused by vibrations, why are the devices producing those mostly round and slightly concave? Is this optimal for acoustics, or perhaps a compromise of strength, power handling, material cost, etc...

[dgage]

Chithead, if you mean round vs say the kicker square cones then the issue is boundaries. A round sub has one boundary to keep on the same plane for smooth waves to develop from. If you look at Kicker’s square cones, they have to keep excursion down since it is harder to keep the square cone in the same plane, which would develop the cleanest waves. Plus a square potentially has 4 boundaries corners with each having the opportunity to be slightly out of sync if not perfectly in the same plane, which means a less clean audio wave since it isn’t one clean wave but potentially 4 waves slightly distorted from each other (corners). That’s the audio enthusiast answer and I’d be interested in how Justin can correct me, not that correction is the point. But that is a good question and I’d be interested to know how much further down the deep audio hole Justin could take it.

[Jdunk54nl]

I would also think the concavity of a speaker is due to space limitations. If it was flat or convex, this would then mean that the speaker needs more space in front of it to be able to move. Ultimately it is about how much air can be pushed/moved and how much space it needs.

I would also think the concavity of a speaker has to be important in some way to the way it is moving the air, concave would mean that all of the air is being sent straight out, convex would send the air out in more drastic angles, and flat would match the in/out angles, but I could be wrong about that reasoning since I am basing that off of more how light works.

[Justin Zazzi]

Why do speaker drivers use a voice coil and permanent magnet instead of opposing electromagnets?

I've wondered this too! There might be other reasons but the ones I can think of are:

You would add another source of heat by having a second electromagnet, and speakers don't usually need more heat.

Also creating a magnetic field from a second electromagnet would be another source of energy loss, making the inefficient loudspeaker even less efficient.

A second voice coil would also be a second source of inductance (Le) so it would be challenging to get higher frequencies out of the design. It might be similar to adding a 1st-order low-pass filter onto an existing design, but there might be room to optimize it too.

You would also lose the electrical damping that is present from the magnetic field that the permanent magnet creates. The electrical damping of a speaker is substantially dominant compared to the mechanical damping. You can see this by comparing the Qes vs the Qms of a speaker:
Qms is related to the mechanical damping and this is where the "m" in Qms comes from.
Qes is related to the electrical damping and this is where the "e" in Qes comes from.
Qts is the combination of mechanical and electrical Q or the total Q, which is where the "t" in Qts comes from.

The total Q, which is related to the total damping of a speaker, is usually very similar to the electrical Qes of the speaker. A random speaker might have a Qms (mechanical) of 10 and a Qes (electrical) of 0.50, then the Qts (total) would be 0.476. So the electrical damping is twenty times more dominant compared to the mechanical damping.

So if you lose the electrical damping from not having a permanent magnetic field anymore, then the cone would tend to ring for 10-20x longer. This is very easy to test by tapping on a subwoofer cone when the amplifier is powered off and listening to how the cone rings, then turning on the amplifier and tapping on the cone again. You will hear a tremendous difference and you can even feel the presence of the electrical damping if you push on the cone a little bit.

This is also related to the "damping factor" wars of amplifiers from a while ago. If the amplifier had a poor damping factor then the electrical damping would be less effective and the cone's motion would be different than intended.

One example of a speaker that uses two electrical sources instead of a permanent magnet is an electrostatic, however they work a little differently than two voice coils like I think you are asking.

So all in all, the moving coil loudspeaker hasn't changed a whole lot in the decades it has been in use! It's one of the most amazing things that we still haven't made it obsolete with some new technology.

[Justin Zazzi]

I think this one is a little bit of "form follows function" and a lot of "form follows practicality".

I'm curious about the typical speaker and subwoofer design. If sound is caused by vibrations, why are the devices producing those mostly round and slightly concave? Is this optimal for acoustics, or perhaps a compromise of strength, power handling, material cost, etc...

The round shape has many advantages in manufacturing like being able to use a lathe to turn metal parts to size, being able to apply adhesives with a stationary applicator while the speaker spins around on a bearing like a lazy-susan, and not having to control the orientation of many parts since a round magnet mates to a round steel plate and a round basket no matter how you spin them. It is convenient to take a sheet of material and roll it into a round shape like the former of a voice coil, or the copper windings of the voice coil itself.

The motion of the voice coil is also very critical to control, especially since the clearance in the magnetic gap is very tight (often less than a fraction of a millimeter). A round shape for things like the spider and surround apply a uniform force in every direction which keeps the voice coil centered in the magnetic gap really well.

A round shape also has no stress concentrations whereas a square shape would have a stress concentration at each of the four corners. This means the stress or friction or stretching of a material is exaggerated in a few specific spots which will lead those areas to always fail first. Think of a ketchup packet. You can press and pull and squeeze along the smooth edges of those things all you want and they will never open. But if you tear really gently on the edge with ruffles you can rip them apart easily. That is a stress concentration, which is a great thing for ketchup but a bad thing for speakers (like square ones). Actually anything with a "tear here" is a stress concentration on purpose.



Round shapes are also super easy to cut into cabinets compared to square holes. Sure you can get some more surface area with a square driver but the raw convenience of making round holes is huge.

The acoustic performance off-axis of a round shape is also very easy to predict. Having the same polar pattern up and down vs side to side is not always desired, but it is super predictable and relatively simple methods can be used to design with round speakers.

Chithead, if you mean round vs say the kicker square cones then the issue is boundaries. A round sub has one boundary to keep on the same plane for smooth waves to develop from. If you look at Kicker’s square cones, they have to keep excursion down since it is harder to keep the square cone in the same plane, which would develop the cleanest waves. Plus a square potentially has 4 boundaries corners with each having the opportunity to be slightly out of sync if not perfectly in the same plane, which means a less clean audio wave since it isn’t one clean wave but potentially 4 waves slightly distorted from each other (corners). That’s the audio enthusiast answer and I’d be interested in how Justin can correct me, not that correction is the point. But that is a good question and I’d be interested to know how much further down the deep audio hole Justin could take it.

Interesting idea. I think you're talking about what happens when a cone stops moving as a single rigid body and starts to deform. Or maybe you're asking about how the corners would have a higher stress/pull on them kinda like the stress concentrations I mention above.

A square cone and a circular cone both have standing vibration modes (or breakup modes) at a high enough frequency. They are really neat to visualize and this was the entire subject of my first class at Penn State with Dr. Russell. He has an amazing website with tons of fantastic animated pictures of vibrations: everything from the textbook-style that I'm borrowing below to actual measurements he's made of hockey sticks and baseball bats and acoustic guitars.

If you have a moment, PLEASE go browse the few pages Dr. Russell has. It's mostly like a picture book of really fascinating animations that don't need any science to understand. He's one of the best teachers I've ever seen!

Image from Dr. Russell at Penn State, at his website:
https://www.acs.psu.edu/drussell/demos.html
https://www.acs.psu.edu/drussell/Dem...le/Circle.html
https://www.acs.psu.edu/drussell/Dem.../rect-mem.html

I would also think the concavity of a speaker is due to space limitations. If it was flat or convex, this would then mean that the speaker needs more space in front of it to be able to move. Ultimately it is about how much air can be pushed/moved and how much space it needs.

I would also think the concavity of a speaker has to be important in some way to the way it is moving the air, concave would mean that all of the air is being sent straight out, convex would send the air out in more drastic angles, and flat would match the in/out angles, but I could be wrong about that reasoning since I am basing that off of more how light works.

The traditional dish style cone shape is a form-follows-function thing where you need a speaker cone to be lightweight and rigid at the same time. If you use a geometric shape that is inherently strong then you can use a little less material and make the cone lighter. Think about a paper water cup that is usually found near water coolers.


The paper cup is really fragile when you first pick it up because you are holding it from the sides, and pinching it from the sides. When you fill it with water the cup has no problem holding the weight of the water because a cone is stronger in it's axial direction, or up-and-down in this case. When you're done with the cup and it is empty, it takes almost no effort to crush it from the sides and crumple it into the trash, and it weighs nearly nothing because it's essentially a thick sheet of paper with a little glue (much like a paper speaker cone!).

So the shape of the material makes it stronger in this case, but only for the intended use. If a paper cup is intended to hold water, then a paper speaker cone is intended to push and pull air in much the same way. It is a very efficient shape if you need a strong and lightweight shape that only pushes and pulls.

There are many "non traditional" shapes now that we have more exotic materials to play with. There are lots of tradeoffs in the shape of a cone and how it transmits energy to the air. A convex shape like a dome tweeter or a dome midrange has a different off-axis frequency response than a concave shape like a subwoofer cone, and some of that performance difference is due to the frequencies a tweeter plays vs what a subwoofer plays. A concave shape can also have either a straight wall which looks like a simple paper cup, or the wall can be curvlinear which is a fancy way of saying "a constant curve like the edge of a circle". Both have strength and performance tradeoffs too.

The dust cap on a speaker cone can also contribute to its strength in a big way. I'm working on a thin-mount subwoofer design that relies on the presence of the dust cap to make the structure stronger and the design would not work without the dust cap. It is really interesting to be able to model the stresses and how far a material will bend on the computer, and then making changes to see how you can improve the design without ever building a thing.

[DaveG]

Justin do you think we will ever have a single speaker cable of 20-20k? And please explain why or why not. Thanks

[JCsAudio]

Do you know anything about power supplies of amplifiers and the pros and cons of regulated vs non regulated power supplies in car audio amplifiers. An example of a regulated power supply would be JL Audio RIPS (regulated, intelligent power supply) or I think Rockford Fosgate constant power. Most car audio amplifiers have non regulated power supplies and their power changes with voltage and impedance as well as the inductance of a speaker has a great affect on power output, but dynamic power output can also be better with a non regulated power supply vs regulated. Will you hear the difference, I don’t know.

How about some commentary on speaker cone material and the certain attributes or pros and cons of each type. I have experience with some and have done a lot of research on this. I experienced the odd order distortion or breakup of aluminum cones as listening fatigue myself. Here is what I understand as sort of generalizations:

1. Paper, the most popular which is light weight, strong, and doesn’t have nasty breakup nodes or odd order distortion like metal cones but isn’t as moisture resistant. Can be used in both three ways and two ways.
2. Aluminum, has great detail, is very rigid and strong and has good pistonic action but can suffer from ringing and nasty cone breakup in higher octaves. Best used for three ways.
3. Poly, exhibits a smooth response and breaks up gradually for a smooth extended response but can lack detail, is heavier, and temperature can have a great affect on performance. Good for car audio and can be used in both three ways and two ways. Great for two ways.
4. Fiberglass and Carbon Fiber. I have a pair of Focal poly glass midwoofers that seem to sound very similar to how my paper cone mids do. Not much is known really. Can be expensive and possibly heavier than paper.

[Justin Zazzi]

Justin do you think we will ever have a single speaker cable of 20-20k? And please explain why or why not. Thanks

Ever? I think so. Soon? I'm not sure. I think it depends on how you define a "single speaker".

If you consider headphones to be a single speaker then these exist today already. They are able to get good bass response since they are, more or less, directly coupled to your eardrum. If you're listening to headphones with good bass response and then you lift them a little bit so the air seal is broken, the bass is tremendously reduced. This is because bass wavelengths are much longer than the size of the airspace between the headphone and your eardrum so any bass frequencies are directly passed along. This is a similar effect to "cabin gain" in the inside of a car where below a certain frequency the bass starts to get reinforced, seemingly for free. So headphones can do 20-20khz with a single driver but they do have a unique "enclosure" directly attached to your head that allows it.

If you mean a single voice coil and a single cone like a 6.5" driver that is supposed to fill a room with music, then it is much more challenging.

To get 20hz output you need a large cone area. You need a large cone area because it helps to move more air, yes, but you also need a large cone area because a larger cone will couple with the air better at lower frequencies meaning a larger cone will transfer energy into the air more efficiently. There is a crazy formula for acoustic radiation impedance that relies on a factor called "ka". A high ka value means the acoustic energy is more efficiently transferred from the piston to the air, and a low ka value means the energy transfer is less efficient.

The "k" part is essentially frequency. A higher frequency means a larger "k". (k is called the acoustic wave number)
The "a" part is half of the diameter of the speaker cone. A larger speaker cone means a larger "a".
So to get a really large k*a value you either need a high frequency or a large speaker cone or both.

Think about a tweeter. A tweeter has a small diameter so the "a" value is small. The only way to make a tweeter have an efficient energy transfer to the air is to have a large "k" value, which means higher frequencies. Thankfully that is what a tweeter does: high frequencies with a small cone area. So far so good.

Think about a subwoofer. A subwoofer plays low frequencies so the value of "k" must be small. The only way to make a subwoofer have an efficient energy transfer to the air is to have a large "a" value to balance the small "k" value. Because of this, subwoofers usually have a large cone area not only to move more air, but also to couple to the air better and improve the transfer of energy.

This is a really powerful concept and responsible for something I've been chasing for a long freaking time: mutual coupling. If you place one subwoofer outdoors on the left side of a stage, you will get some amount of bass. If you place a second subwoofer on the right side of the stage then you will get about twice as much bass, as expected. But if you place both subwoofers very close together at the center of the stage, you will get more than if you place them apart. This is commonly called mutual coupling. What happens is the effective cone area is increased so the "a" factor is increased so the efficiency that the energy transfers into the air increases. This also might describe why people love to have two huge 15" woofers in their cars even if they don't move much, compared to a bunch of smaller 8" woofers or something like that.

Back to your question: you'll need a large cone area to get any kind of bass energy out of it. A large cone could play treble but it will have tremendously narrow beaming at higher frequencies. A larger cone that can play bass will also have a larger and stronger motor that has a lot of inductance, and inductance is what rolls off treble response. So it's a game of tradeoffs that just doesn't work well. A different approach is needed.

The Synnergy Horn from Danley labs is one possible way to make it happen. Essentially you make a horn (flared tube) with a tweeter at the very back and then place gradually larger speakers along the tube as it gets bigger. You fire energy from every speaker into the same acoustic space (the horn) and it all exits together at the same time and it behaves like a "single speaker". I've heard a few of these and they are simple amazing. I believe one could be designed for 20-20khz range, however they are usually designed in the 50hz and up range for concerts and stadiums.

https://www.danleysoundlabs.com/technology/