The audio world is rife with theories that are the source of endless discussion and debate. Some are based on sound priciples. Others, not so much. Here are a few oft-repeated statements we can easily subject to a little critical thinking.

You’ve probably heard or seen these before:

• Home theater speakers are no good for music.
• Music speakers don’t work well for home theater.

There is a grain of thruth here but, again, it ain’t necessarily so.

When designing a relatively inexpensive home theater speaker, you must concentrate on the midrange. That is where the dialogue is and if you don’t get that right, intelligibility will suffer. This generally means using smaller drivers that are better suited to good midrange performance. Unfortunately, these drivers will not play all that deep.

This is generally not a problem in home theater since subwoofers are normally employed to augment the bottom end. But for music reproduction, the bass extension is not quite what most people would find desireable.

In most relatively inexpensive music speakers, bass extension is an important consideration. This will lead to the use of larger woofers which don’t perform as well in the critical midrange. But people casually listening to music will tend to concentrate more on the bass and the top end and may not notice that the midrange is slightly lacking. So while these speakers will not be good for home theater, they are somewhat accpetable for music reproduction.

As you can see, there are elements of truth that support these myths. However…

If you design a great “full-range” music speaker with spectacular midrange performance (where 80% of the information is in the first place), it will not only provide a great musical experience, but will be excellent for home theater as well since intelligibility will be high.

The only problem is that it is difficult to do this on a limited budget. You either have to use very high quality drivers or you have to develop a multi-way speaker. Both are more expensive.

The bottom line is that inexpensive home theater speakers are not generally good for music and inexpensive music speakers are generally not all that good for home theater use. But it has little to do with the catagory of speakers in general and more to do with the quality of the basic designs.

A great speaker will work equally as well in home theater or for the foundation of a great music system. These myths are busted (or at least explained)!

The audio world is rife with theories. They are the source of endless discussion and debate. Some are based on sound priciples. Others, not so much. Here are a few oft-repeated statements we can easily subject to a little critical thinking.

You’ve probably heard or seen these before:

• Monitors create a wider, deeper soundstage and image better than floor standing speakers.
• Floor standing speakers play deeper than monitors.

There is an element of truth to each of these statements. And therein lies problem.

Let’s examine that first statement and see how well it holds up to some critical thinking.

Superior dispersion is one characteristic that enhances off-axis response and helps create a deeper and wider sound stage. So for duscussion purposes, let’s take a typcial 5″ driver with excellent dispersion characteristics and mount it in an 8″ baffle. We’ll build a pair of rear-ported monitors and take a listen.

Wow, the soundstage is indeed very deep and wide. (That’s because we did such a good job creating our theoretical speaker.) So now let’s take those same 5″ drivers and put them in floor standing cabinets.

In our original ported design, our 5″ driver required a specific internal volume and a port of a specific length and diameter. So we will keep those and the baffle width constant. In other words, we will build floor standing cabinets tuned exactly the same as our exellent sounding monitor.

As we increase the cabinet height to go from a monitor to a floorstander, we must decrease the cabinet depth in order to maintain the same internal volume. Now, there are some pracitcal considerations here. For example, we still want to maintian some area behind the driver to allow space for it to breathe. But, for the sake of this thought experiment, let’s set issues such as this aside.

If the theory holds true, this floor standing speaker will not create the same quality soundstage as our monitors. Oh, but upon listening, we find that it does – the performance is exactly the same! That should not be surprizing. It is the same driver, same baffle width, same internal volume and the same cabinet tuning. The only difference is the shape of the cabinet.

The driver has no way of knowing that it is mounted in a tall cabinet vs. a shorter, deeper cabinet on a speaker stand. The only thing it knows is that the baffle is 8″ wide, the internal volume is correct and the port tuning is accurate. It’s performance is exactly the same as in our monitor cabinet. It can’t change. (I might point out that the floor standing cabinet takes up no more floor space either. In fact, it could actually take up less since the depth is decreased.)

The bottom line is that there is no inherant advantage to a monitor type cabinet where sound stage or imaging is concerned. Given the same driver with the same internal volume, same baffle width and same port tuning, the results, in terms of sound stage, will be exactly the same. No difference.

Quite often, we are asked if we can produce speaker “X” in a floor standing version rather than a stand-mounted version. When we ask why this might be preferable, the answer often points to a desire for deeper bass response.

So, do floor standers truly play deeper than monitors? Yes, in general they do. But not because they are floor standers. Again, if you use the same driver with the same internal volume and cabinet tuning, the bass extension will be the same.

Now, there are things that can be done with floorstanding speakers to increase bass extension. Given the same driver in a ported and transmission line cabinet, for example, the TL cabinet will normally generate greater bass extension. But you could fold that same line length into a deeper stand-mounted cabinet and achieve the same result. So there is no inherant reason a floor standing speaker would play any deeper than an appropriately designed monitor.

In the end, the only reason there is any merit to either of these statements is that the speakers in either category tend to be different by their very nature. Floor standing speakers usually utilize larger woofers which tend to play deeper. With monitors, you tend to see the use of smaller drivers with better dispersion characteritics. It has nothing to do with the specific format of the cabinets themselves.

Floor standers can image just as well as monitors and monitors can play just as deep as floor standers. It is not an issue of cabinet shape that determines sound staging, imaging and bass extension, but other aspects of the designs in question.

These myths are busted!

One of the most misunderstood topics in audio is the subject of diffraction. Diffraction, acoustic phase, and how listening rooms impact our reproduction of sound, based on what I see posted in many discussions on the internet, are subjects of much confusion. In this article I will attempt to clear some of the fog on the topic of cabinet diffraction, and hopefully, present it in such a way as to make it much easier to understand.

What is Diffraction?

Diffraction is the name given to the “bending” of waves (distortion of wavefronts) produced when they interact with objects that are comparable to a wavelength in size. This is in contrast to the much simpler phenomenon of reflection, which leaves the waveform shape intact. Both partial reflection and diffraction occur when sound waves encounter an obstacle in its path.

The description is same whether we are discussing light, sound, or waves in the water. All of these diffract in ways that are predictable and consistent. In fact, most of the early work in diffraction came from the field of Optics, as far back as before Isaac Newton, and as a result we will sometimes use terms like “illuminating” the edge, and a “shadow zone” even when discussing the diffraction of sound waves.

In order to get a good visual image of wave diffraction let’s picture a pond of still water. In this pond there is a small branch sticking out of the water, and several feet away there is a frog sitting on a rock. Suddenly the frog leaps into the water. This sets off a waveform moving outward from the frog’s entry point in concentric circles, in all directions (360 degrees). After a few seconds of travel the leading edge of the waveform encounters the branch sticking out of the water. The wave diffracts on this obstacle, and we see a new set of waves moving outward from the branch in all directions, including a portion of the wave moving directly back towards its origin where the frog went in. There is both reflection and diffraction of wavefronts in this example.

These two waveforms will now interact either constructively, by adding in amplitude, or destructively, by canceling each other, depending on their relative phase – which is their up and down motion at the point where they intersect. Although this sounds somewhat complicated, I bet most of you had no trouble picturing this scene and following what I described. And, this is precisely the same thing that happens with sound propagating in the air. When sound is propagating from a loudspeaker it diffracts when it encounters the edges of the cabinet and other obstacles nearby.

Baffle Step is Diffraction?

Yes, sort of, but first we need to understand a few fundamental principles of acoustics before this will make sense. First of all, we must understand that sound is pressure, or more precisely, it is the propagation of pressure waves in air. That’s why it is referred to as SPL (Sound Pressure Level). Second, we need to understand that this pressure is pushing a waveform that is expanding to fill the space around it in a spherical manner. In other words, it is expanding in all directions equally – just like the pressure inside a balloon is pushing outward in all directions equally. And third, we need to understand that the acoustic effects of diffraction are always directly related to the ratio of distance versus the wavelength of sound at a given frequency.

Wavelengths are inversely proportional to frequency. Low frequencies have very long wavelengths and higher frequencies have much shorter wavelengths. Therefore, for a given edge distance or baffle width the effect will be different based on the frequency discussed and its wavelength. An obstacle must be “acoustically large” before full diffraction will occur. An obstacle is “acoustically large” when its dimensions are greater than one-half of a wavelength at a given frequency, at this point there will be full diffraction of the waveform, or the obstacle will fully alter the direction and behavior of the wave. As an object gets progressively smaller than one-half of a wavelength it will have progressively less effect on the wave. Once it is around one-tenth of a wavelength in size it will be small enough acoustically to be essentially invisible to the waveform at those frequencies. Technically, the effects of diffraction asymptotically reach 0 dB only at 0Hz, but for most of this range the effects are only fractions of a decibel until the obstacle begins to become acoustically large.

Speaker drivers are usually rated with what is called “half-space sensitivity” (sometimes called 2Pi or hemispherical space – 2Pi is a geometrical way of describing half of a sphere, whereas 4Pi describes a full sphere). Because of this, we will usually describe baffle step as a loss due to low frequencies “wrapping” around the baffle into “full-space” (4Pi or spherical space) due to their wavelengths being so much longer than the width of the baffle. This description can be correct depending upon your perspective; however it really doesn’t accurately describe the phenomena in a way that allows you to see how baffle step and diffraction are tied together.

To be technically accurate we need to picture it in this way: When a loudspeaker produces a sound, this sound is in the form of a pressure wave trying to expand equally in all directions spherically (like the balloon analogy). The first obstacle that this wave encounters is the baffle face itself. For higher frequencies with shorter wavelengths where the baffle is acoustically large, the baffle causes a doubling of axial pressure into the forward hemisphere (since the pressure can’t expand spherically), much like a perfect reflector. This doubling of acoustic pressure produces a +6dB gain on axis in the forward hemisphere. A baffle with a width of about 9” would correspond to one wavelength at about 1500Hz, this +6dB gain would then be seen at frequencies above 750Hz (that half-wavelength rule. Of course, a taller height dimension pushes this a little lower in frequency in real life). At lower frequencies this gain is progressively less, dropping to near 1 dB at about one-tenth of 1500Hz, or 150 Hz (again, the taller height will push this a little lower, but you get the idea).

At very low frequencies, below 100Hz in our example, the cabinet baffle is “acoustically small enough” to become “invisible” to these longer wavelengths. As a result they have very little effect on the waveform at all; the wave is able to expand reasonably unhindered as a sphere, and there is almost no gain or ripples in the waveform due to diffraction. At higher frequencies above 750Hz the baffle is “acoustically large enough” to fully obstruct spherical expansion of the waveform, acoustic pressure is doubled, and there is a +6dB gain in the response on the forward axis. Of course, what is given up in exchange is that there is very little energy at these higher frequencies behind the speaker. There is no gain in energy here, only a redirection. In between these two frequencies there is a transition as the baffle progressively diffracts the spherical propagation of the waveform. This produces a smooth rise from 0 dB to +6dB, and this is actually what is happening in what we call baffle step. It is diffraction; or more precisely, it is moving from a state of no diffraction into full diffraction as the baffle becomes acoustically larger with increasing frequency. Diffraction is really computed from the perspective of a full 4pi – spherical space. Below what we call the baffle step frequency there is very, very little diffraction at all. The wavelengths are acoustically too large to diffract on the baffle, so the baffle is essentially invisible to the long sound waves at these frequencies. However, the baffle step gain – the rising of the step – IS diffraction.

So What About Edge Diffraction – What’s Happening Here?

OK, here’s the anatomy of a diffraction signature. Continuing with our example, let’s define our baffle as a 9″ wide by 16″ tall mini-speaker with the driver mounted centered on the baffle and only 4″ from the top (or bottom). For the sake of our discussion we will treat this driver as a point source. That places this point source at 4.5″ from both sides and 4″ from the top. This is actually a fairly common location for tweeters, and small woofers may be similarly placed at the other end.

Now, to figure the edge diffraction we need to picture the point source as a point with rays going off in all directions to the baffle edges. The point source is like the point where our frog entered the water. The rays are lines (radii) following those concentric circles moving outward in our pond. Each ray will have a specific distance before it encounters the cabinet edge and diffracts. When it diffracts, like the log in the water, the edge becomes a secondary sound source and some of the acoustic energy is reflected back toward the listener or microphone to combine with the original source with delay. How much delay depends on the frequency and the distance of the ray. Picture a triangle: One side of the triangle is the distance from the driver to the listener. The short side of the triangle is the distance from the driver to the edge of the cabinet. And the hypotenuse of the triangle is diffracted secondary waveform. Because it is longer than the direct path there will be delay and some phase shifts – Just like the phase shifts of our waves in the water combining.

Unfortunately, in our example the point source is very close to the same distance from three different edges (I did this on purpose for our example). It is actually fairly complex because the ray distances will vary continuously as you move around the baffle, encountering edges at the sides, top, bottom, and corners, but a large percentage of them will fall in the range from 4-5.5″ due to the driver placement. This means that the influence of this distance will be much greater in the final result than many other ray distances will be. This distance corresponds to a frequency range of 2.4 kHz – 3.4 kHz with a center point at 3 kHz. When the waveform of sound from the driver moves across the baffle it encounters a sudden discontinuity when it reaches the edge of the enclosure. Frequencies in this range will reach these edges, diffract, and reflect back to the listener delayed out of phase since it is exactly one wavelength. The level will be reduced so there won’t be complete cancellation, but there will be a notch in the diffraction signature centered around 3 kHz. This notch is typical in many mini-monitors due to this distance and the associated diffraction.

On the other hand, the frequencies whose wavelengths are twice this distance, so that a half-wavelength is reaching these edges will diffract to combine with the original source in-phase, but at a lower level. These frequencies will combine constructively and produce an additional gain that could reach +3 dB above the already +6dB, however, since the frequencies will be spread somewhat the gain will be slightly less than the full 9 dB peak. This peak will be twice as wide as the notch described above, and at half the frequency, so it will peak at about 1500 Hz. By the way, due to this peak, and typical crossover points for midwoofers, baffle step could actually appear to be more than the normally discussed 6dB once this hump of +2 to +3 dB is taken into consideration.

Now, we also have some longer rays going to the bottom edge of our enclosure doing the same, these are in the 12-14″ range. This corresponds to a wavelength of just over 1000Hz, so there will be a little down-ripple in the response at this frequency, and a half-wavelength of around 500Hz, so there will be a little up-ripple here due to these distances as well.

So now we have defined the typical baffle step, the peak, and the notch. At frequencies higher than this notch it is all the same mechanism that we have already discussed and applied – only the wavelengths get shorter and the phase of the diffracted sound becomes more randomized, and the ripples get narrower and shorter in amplitude. Diffraction is a form of linear distortion, because it affects the frequency response on a given axis and has a minimum phase relationship, meaning the phase is directly related to the frequency response.

I hope this explains it reasonably well. It is all about ray length, driver location, and the wavelength of the sound reaching the edge and then recombining with the original source either in-phase or some degree out of phase.

How Do You Control Diffraction?
Well, the best way would be to eliminate it, but that would involve mounting the drivers on an infinite baffle, or flush in a wall, and that doesn’t work out very well for most people. For speaker drivers mounted in a typical cabinet you can not eliminate the effects of diffraction from the cabinet, but there are several techniques that compensate for, or reduce the impact these effects significantly.

First, the most obvious diffraction effect for the typical small stand mounted monitor or the tall narrow tower type of speaker is the “baffle step” in the response that was discussed above. Fortunately, this step is fairly smooth and easy to measure on the design axis, because of this it can easily be compensated for in the crossover design. The negative side of this compensation is an apparent reduction in loudspeaker sensitivity of 6dB. The truth is that the original driver sensitivity was rated based on half-space (hemispherical) radiation, and we have adjusted everything to a flat response based on the full-space (Spherical) radiation of lower frequencies. The loss of sensitivity is traded off for flat on-axis frequency response. This trade-off is worth the drop in sensitivity. If you have listened to speakers that do not compensate for this step, the sound can be very thin in the lower midrange and bass, leaving you with a forward, bright, irritating sound.
This leaves us with cabinet edge diffraction. Several different techniques have been employed over the years to reduce the effects of edge diffraction. One of the most effective is the use of a thick felt whose tangle can effectively absorb and diffuse the sound waveform moving along the baffle before it can encounter the edge and then diffract and recombine as described above, creating irregularities in the frequency response. Despite its effectiveness few commercial loudspeakers use felt, mostly for cosmetic reasons, but there are some notable exceptions that have been very successful. Most loudspeaker purchasers though, weigh the appearance of the speaker with the sound presentation when making their selection. As humans, we are strongly visually driven, even when looking for good sound.

Fortunately, there are some techniques that work well in reducing edge diffraction effects and improve the appearance of the loudspeaker at the same time. One thing that needs to be done is to recess each loudspeaker driver so its faceplate or frame is flush with the baffle. It may not seem like it matters at first, but for surface mounted drivers the tweeter’s response will actually be impacted by the diffraction from its own faceplate edges as well as from the frame of the woofer mounted nearby. Flush mounting is an important feature that both aids in diffraction control and improves the appearance at the same time.

Sometimes you will see a tweeter offset from the centerline of a baffle. This asymmetric mounting is also a diffraction control technique. By offsetting the tweeter, the distance from the center of the tweeter to the left edge and the right edge are different distances. This means that the frequencies whose wavelengths correspond to these distances are different too. By offsetting these frequencies you can sometimes smooth the on-axis diffraction signature because the distances to each edge will produce ripples at different frequencies. If carefully designed, these can combine to smooth the response. When using this technique it is important to note that the diffraction signature is asymmetrical too and there is a greater difference in the response whether you move off-axis to the left or to the right compared to a centered tweeter that is symmetrical. Similar to offsetting the tweeter is the technique of “toeing-in” the loudspeaker so you are not directly on-axis. This has a similar effect to offsetting the driver because your off-axis position changes the geometry of how the sound recombines after diffracting off of each edge.

Another technique is to add a large radius to the edges of the cabinet. Many manufacturers will stick with a standard rectangular box with square edges as a cost savings, but the frequency response will have much more variation than it would have if the cabinet was rounded on the edges with a fairly large radius. It is costly to make baffles that are rounded or curved, but the impact on frequency response can be dramatic. The larger the radius or curve usually the better the diffraction control, and the smoother the frequency response will be. Here’s why -

When a waveform is moving across the baffle and encounters a sharp edge with a sudden discontinuity of 90 degrees, there is a very sudden change in the propagation of the wave. The sharp corner acts like an obstacle changing the direction of the wave; the wave diffracts and the edge becomes a secondary source, reradiating sound back towards the original wave, as we have discussed. When a large radius is used the waveform moves across the baffle and tends to follow the radius as it curves away from the front. There is no sudden discontinuity in its path. This does not mean that there is no diffraction, but the larger the radius the lower in frequency the disturbances lie. The large rounded radius accomplishes two things that benefit our diffraction issue: First, the smoother path around the corner of the baffle reduces the amplitude of the disturbance at specific frequencies, thus reducing the overall impact on the frequency response. This occurs because the rounded edge is seen as a “fuzzier” less defined edge, and this spreads the affected frequencies over a wider range than a sharp edge does. Second, as the wave does begin to diffract on this radius part of the energy is redirected at different angles away from the baffle, so less diffracted energy recombines with the direct sound that produce the ripples in the frequency response that I described above.

Finally, a feature that helps to control diffraction is controlled directivity. Our example above treats the loudspeaker driver as a point source. In reality this is not correct. All drivers have an effective radiating width or diameter; the larger this diameter, the greater the directivity of the driver. As a result, less acoustic energy at higher frequencies is able to illuminate the edge of the cabinet that is usually 90 degrees off-axis. (Just look at the 60 and 90 degree off-axis frequency response curves for many drivers for a good example of what I mean). If less energy illuminates the edge, then the strength of the edge source is reduced and so is the diffraction. Even a 1” dome tweeter has significantly reduced energy at 90 degrees off-axis for frequencies above 8 kHz, and for larger drivers this is even less. Properly designed waveguides and horns also control driver directivity and can significantly reduce edge diffraction as well.

Wrapping Up

For most speakers what is known as “baffle step” is usually best handled in the crossover. It is possible to avoid this step altogether in a three way system if the woofer is placed close to the floor and the crossover point is carefully selected. This allows boundary reinforcement to fill in part of the step. However, for most speakers some shaping of the frequency response is necessary to ensure flat response. When we get to the response irregularities due to edge diffraction it is left to the designer as to how he wants to deal with these. He may choose heavy felt around the drivers. He may choose to use a wave guide. Or he may choose to use rounded edges on the cabinet. He may even choose to leave the issue of diffraction unaddressed, either living with the response ripples or working them into the design in some other way. Of all of the methods used today most people seem to agree that the most aesthetically pleasing method is to recess the loudspeaker drivers flush to the face of the baffle and then shape the baffle with a large round-over, or possibly an even more complex shaping of the baffle. When you see a design like this, remember it is much more than just a pretty face – it is a very effective means of controlling cabinet diffraction and smoothing the overall frequency response.

Why Passive Radiators?

The new Salk SoundScape 10 and 12 speakers use dual passive radiators. These are obviously more expensive than port tubes, more complex to tune and require more cabinet work to implement. So why use them?

As the designer of the bass section of these two new speakers, I thought I would explain why passive radiators are perfect for this application, and why SoundScape series speakers will out-perform most other high-end speaker systems, and even many subwoofers, in low bass extension and output while, at the same time, providing extremely low distortion bass.

Achieving excellent low frequency performance is much more complex than simply using a speaker design program and arriving at a nice woofer/box alignment curve on the computer screen. This is a good place to start, but on its own it fails to address some critical aspects of performance. Unfortunately, many speakers are designed this way and that’s where it ends. As a result, the speaker’s low frequency performance meets specifications at low input levels of just a few watts, but when the volume is turned up, sometimes at levels as low as 50 Watts of input power, both low end extension and distortion are quickly compromised. This happens more often than you may know, even among high-end speaker systems. Let me explain why this is, and how we resolve this issue in the SoundScape series.

How it works

Low bass performance is all about pressurization. Since the speed of sound and density of the air are stable enough in our listening environments to be treated as constants, we find that low frequency SPL (Sound Pressure Level) becomes directly proportional to a speaker’s volumetric displacement (how much air it can move) at a given frequency.

Simply put, the sound volume a woofer can generate is directly related to the amount of air it can move at a given frequency. And the lower the frequency, the more air it is required to move in order to create that volume level.

This is where the problem lies with many speakers. It models exceptionally well at 1 Watt per Meter, but at just over a few Watts it can become limited in its ability to move enough air to maintain higher SPL’s at low frequencies. Consequently, as the volume goes up, the effective bass cut-off frequency (its  -3dB point) rises too, and usually so does the driver’s distortion because the voice coil begins to leave the magnetic gap, thus increasing the distortion. Often these speakers are ported to increase efficiency in the low bass, especially near the port tuning frequency. Unfortunately, this quickly becomes a source of limitations as well.

The first thing that must be considered in resolving the restrictions is the design of the woofer itself. Not only does the woofer need to be optimized for bass extension in its parameters, but it must be capable of moving a lot of air with very low distortion. The woofers used in the SoundScape series are custom made drivers with aluminum cones, very long linear strokes, and have specially designed motors that significantly reduce electrical distortion. These woofers are capable of a linear stroke of +/- 24mm (Xmax one-way), or approximately 2” linear peak to peak excursion.

The Soundscape 12 woofer is optimized for a “vented” enclosure and the alignment chosen has an F3  (-3dB bass cut-off frequency) of 18 Hz. As you know, most vented woofers use ports to tune the box resonance and augment the low frequency response. However, a woofer of this design forces us in a new direction for solving the limitations of simple ports. Here’s why – It moves a lot of air, and thus is capable of a much higher SPL at very low frequencies. But, this means that the port must be able to displace high volumes of air as well without creating noise or inducing flow restrictions that reduce the bass output, and what it takes to do this correctly may surprise you.

The Port Problem

Let me explain with some numbers. It is not unusual, even on many high-end speakers, when using a single 12” woofer, to use a 3” diameter port. In some cases this may be an acceptable port size if the woofer itself is not capable of displacing a large volume of air, or if the system cut-off is fairly high in frequency. However, neither of these two restrictions sounds much like a true “high-end” speaker system and, in the case of the SoundScape 12, the goal is very clean bass to below 20hz (18Hz, actually) and the capability of achieving this at a fairly high SPL if required (110+ dB at 20Hz).

So, how will the 3” diameter port perform?  Let’s take a look. In order to tune the enclosure to 20Hz the 3” port needs to be a little over 15” long. That appears doable. So, what’s the problem?

The general rule of thumb is for port air velocity to be kept below 5% of the speed of sound (approximately 17.3 Meters/Second) in order to ensure that audible “chuffing” noise does not develop along with restrictions that could limit bass output. Unfortunately, as typical as a 3” x 15” port is, in this application it reaches 17.3 M/S of air speed at only 20 Watts of input. Suffice it say, I am sure Jim and SoundScape owners would expect better performance than this.

What if we use two 3” ports? OK, let’s see. Two 3” ports need to be almost 33 inches long each in order to tune to our 20Hz frequency. However, they still reach 17.3 M/S of air speed at 80 Watts. Hmmmmm…. How about using two 4” diameter ports? Well, these won’t hit 17.3 M/S until 250 Watts of input. That’s more like it, but on the other hand they need to be almost 59” long each. Now, I need to fit two 4” diameter, 5 foot long tubes in the cabinet. If that wasn’t enough, this length results in a strong port resonance at 87 Hz, and that should be a no-no too. But, is 250 Watts enough for a state of the art speaker? Should port design be robust enough that we never really approach their limitations on the system? If so, then two 5” diameter ports will take over 500 Watts before hitting our 17.3 M/S air speed output. But again, they will also be almost 8 feet long. Are we getting a feel for the limitations here? In order for the port to be capable of output without air speed and possible noise issues it must be large in diameter, but if it is large in diameter then it must be very long – too long to fit in a reasonably sized cabinet. (Keep in mind that the internal volume of the cabinet must be increased to take the volume of the port tubes into consideration.)

The answer? Enter the Passive Radiator

Passive radiators work on the same principle as a port, even though that may not seem intuitive. The air inside a sealed box is compressible, but then “springs” back. This is called “compliance”, and mechanically we treat it as a spring. Any spring will have a resonance frequency. If you hang a weight on the end of the spring it will be oscillate slower, or at a lower frequency. Increase the weight further and you lower the frequency further.

In our vented box, the resistance of the “slug” of air in the port acts like a weight attached to our box compliance “spring”. The greater the resistance, the heavier the weight, the lower the resonance frequency – and you can do the same thing with a passive radiator (basically a non-driven cone) by attaching actual weight to it until the system resonance reaches your target tuning frequency. Makes better sense now, doesn’t it?

Now, I know what some of you are thinking: “I’ve heard some passive radiator systems in the past and they sounded boomy or muddy in the bass, and I didn’t like the passive radiator sound.” Hey, I hear ya, but things have changed. Early passive radiators were simply driver cones and suspensions without motors attached. In fact, most still are today. The problem is that these are difficult to tune low enough to reach the proper tuning frequency and they often are very limited in their mechanical parameters as well. In addition to these restrictions many designers opted to leave the system underdamped with a peak in the midbass so you could “”hear” their passive radiator and its “robust” bass.

But the passive radiators we use are different. They were designed from the ground up to be passive radiators and not just woofer cones. They are capable of handling large amounts of mass while their suspensions remain linear and centered. They also have very large linear excursions of almost  +/- 30mm. To tune these radiators to 20Hz results in over 1 kg, or almost 2.5 pounds, of moving mass on the sides of the enclosure.

Two radiators, rather than one as you often see, are used for two very important reasons. First, two provides twice the surface area as a single radiator, thereby increasing maximum output before restrictions set in. And second, with this amount of moving mass, if it was on only one side of the enclosure it could create a “rocking” issue. However, if one is placed on each side of the enclosure then they will move out together and in together and all rocking inertia will be cancelled out.

Comparisons

So how do the two passive radiators compare to the ports used in the examples above? The two 12 cones have an effective moving diameter of about 11”. With the mass added they behave the same as two 11” diameter ports with an air mass equivalent to a tube 37.9 feet long for each – air velocity is not an issue. The limitation here is the mechanical excursion of the radiator at +/- 30mm each. Calculations reveal that these two radiators reach their linear excursion limits at the 20Hz resonance frequency with a little over 500 Watts of input power. This is essentially the same performance shown above for the two 5” diameter 8 foot long ports, but with no wind noise or line resonance, or the problem of trying to find a place to fit two 8 foot long pipes.

At the resonance frequency, in this case 20Hz, nearly all of the output comes from the port or the passive radiator. If the port cannot move enough air effectively at high SPL’s then their output, and consequently the overall system’s bass output, will be compromised.

Ever hear a speaker that sounds full at lower volumes, but when you turn it up the bass just can’t keep up? We all have. In the SoundScape 12, however, at 20Hz the cone’s motion is critically controlled by the air spring of the compliance of the air in the enclosure; nearly all output is coming from the two 12” passive radiators with their large excursion capability, and these are capable of providing over 110 dB output at 20Hz (116 dB with two speakers) with very low distortion before reaching their mechanical limitations. All this while the cone is barely moving. And this is without any room gain!

Most subwoofers on the market cannot match this level of performance in the low bass.  But then again, the drivers and alignments used in the SoundScape 10 and 12 are as good or better then nearly all subwoofers already, and in this case you get two of them.  

The SoundScape 10 and SoundScape 12 represent a no compromise loudspeaker design. Their performance in the low bass is only one example of this “no holds barred” approach.

Jeff Bagby

The audio world is overflowing with broad and deeply felt
generalizations (religions?) about tubes, designer capacitors, inductors,
and resistors, bi-wiring, cables, metal drivers vs. paper drivers–not to mention the whole digital vs. analog thing. Few of these beliefs are rigorously substantiated, and many can max out your credit card very quickly.

But there is one generalization that can be substantiated, and believing it
won’t necessarily destroy your credit rating. And that would be: “No
matter how much money you throw at drivers, components, and cabinets,
you can’t hide the effects of a poorly designed crossover.” Here’s a
corollary: “A speaker with ordinary drivers and an extraordinary
crossover will sound better than extraordinary drivers mated to a
so-so crossover just about every time.” If you’ve ever visited an
upscale audio dealer and listened to megabuck speakers that didn’t sound
that great, chances are the problem was inside on the crossover board.

crossover – why we need one

Before I go any further, let’s back up and examine why we need to be
bothered with crossovers at all. In a more perfect world, we wouldn’t
and shouldn’t bother. We would simply find a near-massless pulsating
sphere that would radiate full-frequency sound evenly in every direction
and be done with it. But at the moment and on this planet, the only way
to get adequate bass response is to resort large mechanical contraptions
(woofers) that flex, ring, and beam at higher frequencies, and the only
way to get clean, well dispersed high frequencies is with much smaller
contraptions (tweeters) that can’t displace enough air to produce bass,
and that would burn up if we made them try.

let’s build one

Let’s look at two reasonably high quality drivers-a 6.5″ paper midwoofer
and a ¾” fabric dome tweeter, both from Scandinavia-and see what happens
if we just throw them in a typical bookshelf size box (19″ H X 8″ W).

Here’s the response curve for the woofer operating by itself with
nothing between it and the input signal, which in this case is a very
fast sweep tone that shows how loudly the woofer plays at each frequency
between 20 Hz and 20,000 Hz.

It is not exactly a straight line. To understand why the response
suddenly starts to increase at around 400 Hz before leveling off at
about 1200 Hz, see Jim’s excellent blog about the baffle step.
(Don’t even try to understand all of the peaks and dips below 200 Hz. At
that point the measurement software is capturing the effects of room
reflections and cancellations.) As for the mess from about 4,000 -8,000
Hz–a lot of that is due to natural resonances at higher frequencies that
are inherent in the cone material and the whole mechanical system formed
by the cone, surround, and suspension. That’s the dirty little secret of
most woofers, and, as we will see, a large reason why good crossover
design is so important.

And here’s the tweeter.

Things are in pretty good shape from 3,000 Hz on up. But notice the dip
and peak below that point. All tweeters will have a similar profile in
this box. As the sound waves at those frequencies reach the edges of the
baffle, some will be reflected back and either cancel or augment new
waves just leaving the tweeter. These “diffraction effects” are also
evident in the slight dips much further up. But what’s important is that
even excellent tweeters will not have a smooth natural roll off in the types
of cabinet baffles that are normally used. And that’s another challenge for the crossover.

running with no crossover

Now let’s make a complete mess and run both drivers with the test signal
at the same time.  (The black line is the total output, the blue is the woofer alone, and the red is the tweeter alone.)

This is not something you want to listen to. (And it will also void the
warranty on the tweeter.) There is way too much output in the midrange and treble compared with the bass range. The tweeter will be distorting badly below 2000Hz, and you will hear the woofer’s breakup peaks at full cry. 

what now?

Hopefully I’ve established that some means must be found to keep the
drivers out of each other’s way, and to restrict their operation to the
frequency range they handle best. This can be accomplished through
appropriate use of inductors (which are just tightly coiled wires of
various lengths and gauges-hence the nick name “coil”), capacitors, and
resistors.

If you connect an inductor between the amplifier and the positive terminal of the woofer, it will roll the woofer off at the top at a rate of roughly 6 dB an octave starting at a point determined by the value of the inductor. If you
connect a capacitor to the tweeter positive terminal, it will roll the tweeter off at the bottom at roughly 6 dB per octave. If you add a capacitor to the woofer circuit after the inductor, but wire it to ground instead of in series, you will get roughly a 12 dB roll off per octave. If you add another inductor in series as the 3rd component, you will add another 6 dB of attenuation per octave. Finally, if you add a second capacitor to ground, you add yet another 6 dB of roll-off, for a total of 24 dB per octave.

This kind of roll-off is called a “4th order slope,” and the 6dB slope from the single inductor is called a “1st order slope.” (No extra credit will be
given for telling me what a 2nd or 3rd order slope is.) If you replace the
word “inductor” with “capacitor” and vice versa, you get the same
results for a tweeter at the bottom of its range. Warning! These
roll-off rates don’t really work in practice, as we will illustrate, but
I just wanted you to see where the nomenclature comes from, and what the
basic building blocks of crossovers are.

At this point, let me introduce one of those religions I alluded to at
the beginning.

“Less is more. The best crossover is a simple first order
design that uses one inductor in series with the woofer and one
capacitor in line with the tweeter.”

If I wanted to thoroughly confuse you, I could elaborate on the “phase coherence” and soundstage advantages that these advocates claim. But for this blog, just accept that this approach to crossover design is touted over and over again. And it’s actually the type of crossover that many mass merchandise speakers employ-the ubiquitous “cap and a coil.”

a minimalist crossover

OK, so let’s see what happens when we use this minimalist approach to a crossover with the drivers I’ve discussed. Because the slopes will be so gentle, the tweeter will have to start rolling off fairly early (at a higher frequency)to avoid handling lots of output below 2,000 Hz. This in turn will require a relatively high crossover point for the woofer.
�

The crossover point–where the slopes from the woofer and tweeter intersect-is about 5,500 Hz. It’s evident from the big dip in the response that the woofer and tweeter are interfering with each other over the crossover region.
That’s because they are out of phase with each other in that region because of the roll-off rates that result from our simple crossover. The explanation for this is way too complex and difficult to attempt here. But a successful crossover design must take such phase issues into account.  Further, the sound will actually be worse than the graph looks.  Because we haven’t done much of anything to suppress the resonant break-up of the woofer, you will hear it ringing like a bell. 

Also notice that our “first order, 6 dB slope per octave” crossover hasn’t produced anything like smooth first-order slopes in terms of the actual output of the drivers. In particular, the woofer roll-off above the crossover point is much steeper than 6 dB per octave, because the inherent response of the driver is dropping off rapidly without any help from the crossover.  So, in addition to a ragged frequency response, our “first order” cap and coil won’t provide the slopes necessary to achieve the phase coherence benefits that are claimed for that approach. 

4th order alternative

What to do?  We have to cross the woofer lower to get the breakup region suppressed by at least 20 dB.  And that means we’ll have to resort to a steeper slope for the tweeter to avoid overtaxing it at the low end. And if we want the two drivers to sum properly to produce a smooth response curve, we’ll have to pay close attention to the phase relationships between drivers in the crossover region, where they overlap the most.

One of the most effective and popular ways to manage all this is to use a 4th order “Linkwitz-Riley” crossover design.  We can skip the details, but with this approach the tweeter and woofer are exactly in phase where they meet (although they will be one cycle apart–that’s a subject for a different blog), and roll off at 24 db per octave.  Let’s try that with our two drivers. I’m going to cross the woofer and tweeter at about 2400 Hz, which saves the tweeter from any melt-downs, and keeps the woofer in the range where it is most linear and also has reasonably wide dispersion. Fortunately, I won’t have to use as many crossover components as you might think from my earlier description of a 4^th order crossover, because I can take advantage of the natural roll-offs of the drivers to reach the desired slopes.  Here’s the result:

that’s better

That looks a lot better, and it definitely sounds better–I’ve listened to both.  Notice that the break-up region of the woofer is now about 30 dB lower in output than the overall response of the speaker, and the tweeter is out of trouble at its low end.*

To see whether the woofer and tweeter really are in phase where they cross, I can use my magic software to reverse the wiring to the tweeter, and that should produce a deep “null” where the drivers cancel each other out because they are now exactly out of phase. 

The null is about 40 dB deep, which is about as good as you can do. 

I’ve glossed over or ignored a lot of important topics–like impedance, acoustic centers, lobing, power response, and the ins and outs of phase. Oh—and I’ve only discussed 2-way crossovers.  But hopefully it will be a little clearer to you why crossovers are so important, and why I spend so much time designing them.   

*For reasons that will have to be the subject of a different blog, the roll-off of the woofer is actually less than 24 dB per octave.

Choosing the right model speaker can be a challenge. But even more trying for some people is choosing the right veneer. Sure, looking at pictures of speakers we’ve built over the years can be helpful. But there are even more options out there. So I thought that perhaps a few words about veneers and links to a few sites might be helpful.

When searching for a veneer, you need to recognize that wood is created by nature, not by man. With most woods, there is a lot of variation from log to log. I cannot tell you how often we receive emails requesting “a pair just like these on your web site – photo attached.” Sometimes it is not a problem. But we tend to search out very unique veneers and when the batch is gone, it is gone. So some of our speakers are simply impossible to duplicate.

So let’s look at some potential veneers. Here is a link to the first of 13 pages of veneer scans: B & B Rare Woods

This is one of the most complete collections of veneer photos I have come across. I often send this link to people searching out the perfect veneer. But when I do, I have to point out that these are among the best photos of the best specimens of each particular veneer. In some cases, finding batches of that specific veneer as spectacular as those shown can be impossible.

The lesson: Seeing a picture of veneer you like doesn’t mean you can get it.

Here is a link to another site: Certainly Wood

This site is different – the veneers are available. If you look at the right side of the top section on the home page, you will see a button labeled “Search Veneer Inventory.” The nice thing here is that the search results will show batches of actual veneer that ARE available. The downside is that the photos may not be all that accurate in terms of color and/or the lighting may not show the true beauty of the wood. What’s more, the veneer is raw and unfinished. When you apply finish to many veneers, the color can change dramatically. Generally the contrast between lighter and darker sections of the wood will increase quite a bit when finish is applied.

These two sites should provide a pretty good idea of what is available. If you want to know how a particular veneer might look after finish is applied, just ask. We’ve worked with most of them and have a pretty good idea of how a finished veneer would compare to a photograph of that same veneer in a raw, unfinished state.

working with burls

People looking for exotic veneers are often attracted to burls. These can be among the most stunning veneers created by nature. But there are a few things you need to keep in mind.

Burled wood is not all that common compared to plain varieties of the same species. That fact, coupled with demand, mean the price of burls can be quite high – 5-10 times more expensive. But that is just the beginning.

Sheets of burled veneer tend to be small. Which means many sheets may have to be spliced together to form a panel large enough to cover the side of a speaker. This leads to two problems.

First, you have to find batches with enough sheets to complete the project. If you need four sheets spliced together to cover one side of a speaker, you would need a batch of 16 sheets just to cover the sides of a pair of speakers and more for the tops of the cabinets. If you want a complete home theater in the same finish, you can easily see that the number of sheets you need may exceed the number available in a given batch. And you can’t simply add sheets from another batch to make up the difference as it is highly unlikely that they will match.

The second issue again relates to sheet sizes. Let’s say a sheet is 10″ wide and the side of the cabinet is 12″ deep. This means you will have to splice two sheets together to cover the side panel. So you would most likely book-match the pieces and splice them together. This splice looks best when placed centered on the panel. So you end up with four extra inches on both sides. Once you trim them off, they are too small to use for anything else.

So, not only do you have veneer that is more expensive per square foot to begin with, but you end up wasting quite a bit of it, driving up the cost for each square foot you actually use.

Nature can produce some fabulous looking woods and burls are among the most eye-catching. But keep in mind that they are costly to begin with and even more costly when you consider that much of it may go to waste.

working with crotch veneers

Crotch veneers are another type with a great deal of appeal. Crotch mahogany is a good example and has been highly sought after through the ages.

Like burls, crotch veneers present their own challenges.

Typically, you want the crotch pattern to cover an entire speaker panel. So if you have a speaker that is 40″ tall and you have crotch veneer sheets that are 38″ long, they will do you no good. You need at least 41″ or so to allow for trimming.

If you look at crotch mahogany, you will see that it is often available in sheets sizes from 24″ to 72″ or more in length. The look of individual batches can vary from somewhat interesting to truly amazing. If the sheet you like is 72″ long and 30″ wide, you have to purchase eight sheets minimum just to cover a pair of speakers. But of the 72″ length, you will only be using 40″. And you won’t need the 30″ width either.

As with burls, the base price of a crotch veneer may be 5 times as expensive per square foot as a non-crotch version of the same species. But that is just the beginning. The final cost in a finished pair of speakers will also depend on the yield, as much of it may go to waste.

splicing veneers

Some trees are large, some are small. When they are sawed into veneer, you end up with very wide sheets and very narrow sheets. So, very often, you need to splice veneer in order to end up with a panel large enough for the side of a speaker.

When we work with narrow sheets of veneer, we normally splice them in what is referred to as a book match. Imagine taking two sheets from a sequenced bundle of veneer and opening them like a book. When the two sheets are spliced together in this fashion, one half appears as a mirror image of the other half.

Depending on how dramatic the grain and/or figuring is, you can end up with some rather striking patterns.

Most of the time when we splice veneer, it is because that particular veneer is not wide enough to cover the panel we are about to veneer. But sometimes, even if the veneer is wide enough, we splice it anyway because the patterns created by a book match are so striking.

the difference between grain and figuring

People are often confused about these two terms. So I thought a few words here may help avoid miscommunication.

When you talk about wood grain, you are referring to the contrasting lines of color running through the wood. Rosewood is a great example of a generally dramatic grain structure. The blacks, browns, reds, oranges, and caramel colors create the striking grain patterns rosewood or cocobolo are known for.

On the other hand, take a wood like quilted maple. The grain in maple is not usually very distinctive. The contrast between the lighter and darker sections is not all that great. But there is a three-dimensional quilted pattern running through the entire surface. This visual aspect that makes wood appear to be three-dimensional is called figuring.

Some woods, like cocobolo mentioned above, are known for their great grain patterns. But they generally do not have much in the way of figuring.

Others woods, like fiddle-back maple, are known for their spectacular figuring, but generally do not have very dramatic grain structure.

Still others, like African bubinga, can have both a dramatic grain structure and fabulous figuring.

chatoyance

There is one other aspect of wood that may be of interest. It is called chatoyance and refers to the sparle or shimmer in the wood itself. This is due to the way light is reflected off the wood grain. Some woods, like koa or mahogany, have a great deal of it. Other woods have very little and are rather 2-dimensional in comparison.

Whether a wood has a good deal of natural chatoyance or not is normally something most people don’t consider. But that is mostly because they are not familiar with the term and don’t know to look for it. But the more it has, the more interesting the wood is when viewed close up.

Brazilian cherry and ribbon-striped mahogany can look much the same from a distance. But up close, the mahogany displays much more sparkle than the rather flat looking cherry. Both are very nice woods, but given a choice (knowing that the choice exists), I think most people would prefer the mahogany.

modifying a wood’s color

We are often called upon to add color to a wood veneer.

There are two ways to do this – stain and dye. Wood stains are often used because they are easy to work with. Stains contain color pigments suspended in a volatile medium. When you apply them to wood, the medium evaporates leaving the color pigment on the wood.

The problem with this approach is that pigments sit between you and the wood, partially obscuring the wood itself.

The other approach to coloring wood is through the use of aniline dye. This is what we use.

Aniline dye actually colors the individual wood fibers. It is a little harder to work with, but is virtually transparent and does not mask the natural beauty of the wood. If the color dye you are using is natural (as opposed to green, purple, blue or the like), there is no way you can tell the wood was not that color to begin with.

Sometimes we will shoot a light coat of reddish dye on bubinga or rosewood. This results is a much richer looking finish, but looks entirely natural. Most of the time, however, coloring a dark wood is not the best idea. It can get muddy looking quite fast.

You can add color to almost any lighter colored wood. A reddish brown dye will make cherry look like its been around for years. Straight mahogany is dyed almost all of the time. And maple will take just about any color you can imagine (just look at the finishes on electric guitars these days).

One reason many people request a dyed finish is to match a specific piece of furniture. In this case, having a sample in hand offers the best chance of getting a close match. Unfortunately, that is not always possible. So we often have customers take pictures of the pieces they are trying to match and we do the best we can. The color balance of the camera and the color balance of the computer monitor we are viewing the photos on can produce a somewhat inaccurate depiction. But most of the time, we are quite successful in matching colors this way.

narrowing your choices

With a world full of veneer choices and the possibility of adding color, choosing the right combination might seem like a daunting task. Here are a few simple questions you can ask yourself to help narrow things down. First, do you prefer light, medium or dark woods? You can eliminate quite a few possibilities once you answer that question.

Second, wood normally comes in brown/straw tones or red/pink tones. For example, a medium to dark bubinga will tend to be a deeper red, whereas European pear might be a light pink color (a very light red tone). Walnut can be medium brown while a oak might be straw colored (still brown in tone but very light). Which do you prefer?

Next, do you prefer a dramatic grain structure like you would find in rosewood or more subdued grain as in maple? As for figuring, do you prefer curly, quilted, fiddleback or no figuring? Or would you prefer both, in which case your choices become very narrow.

If you start with those simple questions, you can narrow down your options to a reasonable level. With just a little effort, you should be able to select a veneer that will result in a one-of-a-kind speaker that is perfectly suited to your taste.

Happy hunting!

If given a choice, most people would opt for a high sensitivity speaker in a small cabinet that plays extremely deep. While that may be possible (we’ll get to that shortly), this basic desire runs smack up against the laws of physics.

I’m sure you’ve probably run across that old adage, “quality, speed, low price…pick any two.” It basically states that if you want quality fast, you have to expect to pay a higher price. If you want quality at a low price, you have to be willing to wait. And if you want it fast at a low price, well…

Where driver sensitivity is concerned, there is a similar law in effect. It is called “Hoffman’s Iron Law” and it offers you any two of the following: small cabinet size, deep bass and high sensitivity.

A few decades ago, amplifiers were limited in terms of power, so speakers had to be efficient. They didn’t play particularly low and the cabinets were generally quite large. Today, most speaker manufacturers want to offer small speakers with extended bass response. So driver manufacturers put most of their R & D efforts into low sensitivity drivers that can play deep in a small cabinet. Since today’s drivers tend to be less sensitive, so do today’s speakers. While perhaps these lower sensitivity ratings are less than many people would like, the wide availability of high-power amplifiers make the trade-off a reasonable one.

In the end, speaker design is all about trade-offs. If you are willing to sacrifice bass extension and live with a relatively large speaker cabinet, high sensitivity drivers are still available. But spouses don’t have much tolerance for large speakers and most consumers prefer speakers with reasonable bass extension.

To put things in perspective, consider this example: JBL manufactures a 12″ woofer that is well regarded by the high sensitivity crowd. But it doesn’t play very deep and requires a relatively large cabinet. By comparison, it is relatively easy to produce a speaker with greater bass reach using a modern 5″ or 7″ driver. This is Hoffman’s Iron Law in action.

If you are not willing to give up bass extension and simply can’t live with an overly large cabinet, there is basically only option at your disposal – multiple drivers. When you wire two woofers in parallel, you gain about 6db in sensitivity. At the same time, the impedance is cut in half. If you hook them in series, the impedance doubles, but there is no increase in sensitivity.

So if you take a driver that is, say, 84 db sensitive and you want a speaker that is over 90 db sensitive with a impedance of 8 ohms, you need four drivers (two pairs of 8-ohm drivers that are wired in series, wired in parallel). This will get you well over 90 db in sensitivity at 8 ohms and is one common approach to the problem.

But as indicated above, there are always trade-offs. When the signals from these drivers intersect somewhere out in front of the speakers, the resulting comb filtering reinforces some frequencies while attenuating others. So you end up with peaks and valleys and a jagged looking frequency response curve.

We often get calls from people who own low power amps that require high efficiency speakers. We always try and help them out. At the same time, we always make them aware of the trade-offs they must be willing to accept.

If you find yourself obsessed about sensitivity, here are some things to consider.

• The ability of any system to produce exceptional sound will be limited mainly by the capability of the speakers
• Today’s highly-accurate, cutting-edge drivers tend to be low sensitivity
• If you want a speaker using these cutting-edge drivers, it will tend to be a low sensitivity design
• Watts are cheap – high power amps are readily available

The reality is that although there are ways around it (if you are willing to accept the trade-offs), Hoffman’s Iron Law will basically stand between you and a high sensitivity speaker in a small cabinet with deep bass response. So if you’re after the high quality, full-range sound reproduction modern drivers are capable of,  you’d be better off finding a good high-power amp to drive your new low sensitivity speakers. Don’t blame me…blame Hoffman.

While baffle step compensation (BSC) sounds complicated, it is really quite simple once you understand what happens when sound waves emanate from a speaker. Here is a slightly over-simplified explanation:

the nature of sound

Sound, by its very nature, wants to travel in all directions. When sound is generated by the woofer in a speaker, for example, that sound not only projects forward to the listening position, but also travels around to the rear of the speaker.

You can confirm this with a simple experiment. Stand behind a speaker and you will still hear sound, although note that the highs will be lacking.

Congratulations, you have just unlocked the key to understanding the mystery of BSC! Let’s look at what is happening here.

details

What happened to the high frequency sounds? In short, they were reflected forward when they bounced off of the front baffle of the speaker.

This phenomenon is related to the wavelength of the sound at various frequencies. Just so you don’t have to do any math, here is a rough table (bear with it…this will not get too technical):

I have highlighted two entries in the above table. We will see why in a moment. And you will understand the theory behind BSC when that moment arrives!

baffle width and BSC

Let’s take a look at what happens when sound is generated by speaker drivers that are mounted in the middle of a baffle that is 9″ wide.

When the 20,000 Hz signal in the table is generated by the tweeter, it tries to move in all directions. But after one wavelength, it has traveled only .67 inches. So although a portion of the energy tries to move to the rear of the speaker, it can’t. It hits the front baffle and is reflected forward towards the listener. The same is true of the 10,000 Hz and 5,000 Hz signals in the table.

The 1,500 Hz signal (highlighted), on the other hand, has a wavelength of 9″ and reaches the edge of our 9″ baffle where diffraction can actually cause a slight rise in response levels at the listening position. At 750 Hz (also highlighted), the sound actually begins to travel around to the rear of the speaker.

So basically, any frequency lower than 750 Hz will be able to travel around the speaker (creating a roll-off of 3db for two octaves at the listening position), while frequencies above that will not. You heard this in the listening experiment above.

Note that increasing the width of the baffle will simply reduce the frequency where this behavior difference occurs. So BSC must be designed for the specific baffle in question.

the problem

OK. So, in our example, most of the sound at frequencies above 750Hz is directed forward toward the listener. Sounds at lower frequencies are not only directed forward, but also pass around the speaker to the rear. In fact, nearly half of the sound pressure is lost to the rear of the speaker.

So think about this: if the tweeter and woofer generate the same volume, high frequencies will be twice as loud as low frequencies at the listening position. (While this sounds like a huge difference, keep in mind that a doubling of sound pressure is about the smallest volume differential humans can detect.)

At any rate, the result at the listening position is sound that will be thin and lacking in the bottom end.

the solution – BSC

The solution is to design a circuit in the crossover that shapes the sound to compensate. It basically rolls off the higher frequencies so that they are in line, volume-wise, with the lower frequencies at the listening position. Baffle step compensation saves the day!

You should now understand the theory behind baffle step compensation. Congratulations!

but…

What happens when you mount this speaker in a wall?

Well, you have now increased the width of the front baffle. It now becomes the entire wall surface, so you have essentially created a baffle of infinite width. In this case, even the low frequencies cannot move to the rear of the speaker.

Since you rolled off the highs with BSC, you will now have too much low frequency energy directed forward. The result will be a boomy, uncontrolled bottom end.

The same would be true, although perhaps to a lesser extent, if you backed the speaker up to the wall. In both cases, an excess of bass energy (in relation to higher frequencies) is directed at the listening position.

This is why speakers designed to be free-standing (which require BSC) should not be mounted in a wall.

congratulations!

You now have a working understanding of baffle step compensation. Your friends are bound to be impressed, but as to whether that is in a positive or negative fashion remains to be seen. You are now officially a “speaker geek.”

Browse through any audio web site and you will likely find posts with comments like, “I just purchased XXX speaker cables and could not believe how they improved imaging.”  Or, “These XXX crossover caps  dramatically enhanced the top-end clarity of my speakers.”

After reading such rave reviews, you may feel inclined to open your wallet and “drink the kool-aid.”  And if you do, chances are very good you will hear similar improvements in your system.  But the source of these improvements may not be quite what it seems.  The human mind can play a critical role in your perceptions and what you hear will very likely be  influenced by your beliefs.

For the sake of discussion, let’s assume you view yourself as an intelligent, fiscally responsible individual.  And let’s say you just forked over $1000 for a new set of speaker cables.  Further, let’s say that you wasted that money because the cables made no audible difference.

On the one hand, your self concept is that of a fiscally responsible individual.  On the other hand, you just wasted $1000.  The two statements are not in alignment as the latter is in obvious conflict with your self-concept.   Your brain is not well equipped to handle this “dissonance,” but is well equipped to resolve the imbalance. From your brain’s perspective,  one or the other statement cannot be true and it will work to reduce or eliminate the conflict.

One way you might rationalize is to say that $1000 cables are better looking or will last longer.  But this is unlikely.  The more probable outcome is that you will indeed hear improvement in the performance of your system.  It is simply your brain’s way of reducing dissonance.

So when people make claims of hearing a dramatic improvement in system performance, are they telling the truth?  Of course they are – they actually do hear a difference.  But is it real?

I, too, have often fallen victim to this sort of bias.  When switching caps or speakers cables or preamps or amps, I have often perceived rather dramatic improvements in performance.  But, on at least a few occasions, I have tested the validity (as best I could in a simple test) of my conclusions and was surprised by the results.

As an example, I upgraded caps in one speaker and not in another of the same design.  I was certain that the high-priced caps increased the transparency of the top end.  I then set up both speakers so that an associate could play either one without my knowledge of which was playing. ( This, of course, is not a classic double-blind test by any means, but it is better than no test at all.)

If the upgraded caps did indeed make a “significant” difference, I should have been able to correctly identify which speaker was playing a relatively high percentage of the time.  But after reviewing the results of repeated tests, I was only able to correctly identify the speaker being played about half the time – no more accurate than flipping a coin.  Based on this simple test, it was obvious that any improvement was not all that significant in that I could not identify it with any degree of accuracy.

Cognitive dissonance is the brain’s way of resolving differences between one’s beliefs and reality.  If they are not in alignment, human nature is to resolve the dissonance so that we are once again in balance.  In audio, the danger is that we tend to hear what our brain’s require us to hear based on beliefs we hold. If you believe speaker cables or caps can make a huge difference, you are very likely to hear that difference whether it is actually there nor not.  If you feel a single driver speaker is vastly superior to one with multiple drivers, you will not likely notice a lack of top or bottom end extension. If you feel higher sensitivity speakers have more “slam,” be prepared for a kick in the gut.

This is not to say speaker cables or caps can’t make a difference, or that a certain speaker design can’t be superior to another (that is a topic for another discussion).  The point is that one must realize human hearing is highly influenced by the brain, and what you hear is very likely to be consistent with, and influenced by, your beliefs.  If you keep an open mind and test your hypotheses, you may discover an entirely new audio reality.

Welcome to our new Audioblog.  In the months ahead, we plan to publish information on new products, tips and tricks for improved performance, essays on a wide variety of audio-related topics and more. Enjoy!

- Jim