How to buy a Subwoofer

Info: How to buy a Subwoofer

This article is a bit tough because in many cases,
especially if you’re building your first system, your subwoofer and subwoofer
amplifier will probably be bought at the same time, and your selection of
them is really pretty dependant on each other… more so than with other
components of the stereo. So, if
you’re in a situation where you’re looking for both, make sure to read this
and my amplifier article that I didn’t write yet.

This article basically picks up where the “How to buy Speakers” article
left off. If you haven’t read that yet, at least skim through it and read the section that
explains how a speaker works so that you’ll have a better understanding of what is discussed
here.

So anyway, on with the subwoofers.

Buying a subwoofer is probably the most popular and most
rewarding car audio purchase. Nothing
transforms your stereo quite like the addition of prodigious amounts of
bass. What many people fail to
realize, however, is that making your system boom is not the only reason to
consider adding a subwoofer. When I
used to sell car audio, over and over I’d hear people give me all kinds of
reasons why they didn’t want a sub, and it almost always revolved around not
wanting that booming bass-heavy drone. What they didn’t know is that I secretly switched their coffee for
Folgers Crystals… oh, wait. What they
didn’t know was that quite often I’d have a subwoofer playing when they
walked into the sound room, and as soon as they’d say something about the
boomy bass, I’d shut the subwoofer off and suddenly they’d realize that the
music is missing its entire foundation without it. They never noticed that the sub was playing
because what they didn’t realize is that bass doesn’t have to be boomy and
overwhelming unless that’s how you want it. Properly integrated bass just becomes part of the music, not an
addition to it.

Specifications

The actual data provided to a buyer can vary wildly from
brand to brand, and even model to model. Most companies will give you somewhat basic data such as power
handling, impedance, and perhaps a suggested enclosure size. Others will give you considerably more
information in the form of Thiele Small (T/S) parameters. These are the mathematical parameters that
represent the subwoofer’s mechanical and electrical properties, and are used
to determine many things, but are primarily used to determine proper
enclosure specifications.

Power handling

This spec, like most, can mean a lot of different things,
but generally you want to find the speaker’s RMS power rating. A peak power rating is worthless on a
subwoofer just like on everything else. Power handling on a speaker is a very dynamic specification that’s
hard to sum up into just one number. Like I mentioned in the “How to buy speakers” article, the power
handling really depends on the frequency at which the power resides. A speaker is going to handle a lot more
power at 1khz than at 20hz for a multitude of reasons, but due primarily to
the fact that at 20hz, that speaker will be moving a LOT, so it’s mechanical
durability becomes an issue. The
thermal capabilities of the speaker (how well it can handle heat) can be an
issue too, because with a lower frequency sine wave, the power level is near
its maximum for a longer period of time, so the voice coil can heat up more…
although it does have more time to cool as well. The type of enclosure that your subs are in
will also have a drastic effect on power handling, which is something that
I’ll go into detail on a bit later. Overall, when looking at subs from a reputable manufacturer, you
should be able to trust that their power handling specification is a
reasonable number that the speaker should be able to withstand without damage
during

NORMAL listening conditions. Passing a constant sine wave at the RMS
power through a sub may damage an awful lot of them, but since music does not
constantly stay at its maximum power continuously, you should be pretty safe.

frequency response

This can be a little bit iffy. The
frequency response of your subwoofer has EVERYTHING to do with the enclosure
and the car that they are in. You need
to know how they arrived at their number, or it doesn’t mean much. Is it an in-car number or an anechoic
chamber? (An anechoic chamber is a specially designed room that eliminates
all sound reflections so that the measured response you get is one that does
not include room interactions, which have a drastic effect on the sound,
especially at low frequencies.) Overall, the published low frequency limit of a sub doesn’t mean a
thing unless you know a lot about how that number was arrived at. It’s much more useful to see published
frequency response graphs of the sub in various enclosures. The upper limit of a sub’s frequency
response, however, can be useful if the number is a truthful number. The trend for subwoofers lately is for
massive subs with huge power handling and huge excursion, and to achieve
these goals, especially in small enclosures, you need a pretty heavy cone and
a very robust voice coil for huge power handling. The result is a huge amount of moving mass
and a voice coil with a very high inductance (after all, a voice coil is
nothing more than a big inductor). The
combination of these elements can have a severe limit on a subwoofer’s high
frequency capabilities. Since a
subwoofer is generally only called upon to reproduce frequencies below
60-100hz (typically) you may think that this isn’t a big deal, but some subs,
especially cheaper ones, have so much inductance and such heavy cones, that
their high frequency rolloff can begin at frequencies as low as 60hz!

Impedance

The impedance of a subwoofer is a fairly complex thing,
but fortunately, it’s not something that you really need to concern yourself
with, other than being aware of what the speaker’s impedance is, and what
happens when you wire multiple speakers together. Basically, what an impedance is, is a
measurement of how much resistance there is to the electrical flow. The lower the impedance is, the harder your
amp has to work, but the more power it can make. Use too low of an impedance and your amp
will go up in smoke or go into a protection shutdown mode. Use too high of an impedance, and you won’t
be able to get as much power out of the amplifier as it’s designed for. If your amp is rated for 100 watts when
driving a 4 ohm impedance, its [i]theoretical[/i] output when driving a 2 ohm
load would be 200 watts, IF it can safely handle that. Likewise, if you use the same amp to drive
an 8 ohm load, you’re only going to get about 50 watts out of it.

Voice coil
configuration

Many subwoofers have more than one voice coil. In general, I find this to be a useless
option that should be done away with in favor of having manufacturers offer
their subs in 2 or 4 ohm versions, but that’s just me. In any case, a dual voice coil subwoofer is
basically like having 2 subwoofers that share one cone and one motor
assembly. The sub has 2 completely
independent voice coils. The purpose
here is that you can wire the two coils in parallel or in series to achieve
an impedance that’s either half or double the impedance of the individual
coils. If your sub has dual 4 ohm
coils, parallel-wiring the two coils will give a single 2 ohm load. Series-wiring the coils will give a single
8 ohm load. I’m going to devote a
specific section to this later on, so if you’re lost here, don’t worry about
it. The basics that you want to
remember for now is that this gives you multiple wiring options when you use
multiple subs in order to best accommodate what your amp’s ideal impedance
load is. If you just plan ahead, you
can avoid all of the DVC nonsense by
just using the right quantity of single coil speakers that have the most
appropriate impedance in the first place, but since many manufacturers don’t
even make single coil subs in multiple impedances anymore, that idea is
pretty much gone, so you’re forced to play the DVC games. This is partly why it’s so important to
choose your sub(s) and your amplifier together. You almost always want to maximize the
output potential of your amplifier by choosing a subwoofer setup whose final
impedance matches the amplfier’s minimum allowable impedance in the
configuration that you intend to use the amp (stereo, bridged, mono, etc),
but I’ll get into that later. I’ll
explain impedance, resistance, DVC
wiring and what it all means a little bit more in a separate section.

Sensitivity

Here’s another fun number, and when accurately stated, can
be fairly useful. The sensitivity of a
speaker describes how loud it will play with a given amount of input
power. The standard convention is to
rate the SPL output of a speaker when measured at a distance of 1 meter, with
a 1 watt input. It’s also nice to know
what frequency (or range of frequencies) this is being referenced to, since
the frequency response of a subwoofer is such a touchy issue as mentioned
already. Generally you will see this
specification listed with respect to the speaker’s SPL output when measured
at a distance of 1 meter with an input of 1 watt. If either of those parameters is different,
it will drastically effect the sensitivity specification, making it hard to
compare with other speakers. While you
generally want to find a speaker whose sensitivity is as high as possible, be
aware that thanks to Hoffman’s Iron Law (explained later), if you want a sub
that goes real deep and works in a small enclosure, you’re going to get very
low sensitivity. So if somebody says
their sub has high sensitivity, and works in a tiny enclosure, they’re probably
lying about something, or their sub has no deep bass output.

Vas

The Vas is a number that represents the amount of air that
has the same compliance (springiness) as the speaker. Since the speaker will need to compress the
air, how compliant it is will have a major effect on enclosure size. Typically, the larger this number is, the
larger the enclosure will need to be for a proper response.

Qts

The qts of a driver is a number that tells us how much a
driver will resonate at Fs. That is a
useless bit of information for almost everybody reading this, but is included
for completeness. In practical terms,
the qts plays a major role in how deep a subwoofer will play, and in what
size box it will do it in.

Fs

This is the frequency at which the driver resonates in
free air (no enclosure). In general,
the lower this number is, the deeper the sub will be capable of playing,
provided the other major factors are equal (which they aren’t).

x-max

The x-max specification is the speaker’s excursion
capabilities. This number, usually in
millimeters, describes how much 1-way linear excursion the driver has. Depending on the motor and suspension
designs, some speakers can exceed their x-max and still provide reasonable
(albeit more distorted) sound. Others
may get very nasty or have audible bottoming at any excursion above
x-max. Since the maximum output of a
subwoofer is determined primarily by how much air it can move, the x-max
along with the speaker’s cone area (Sd) can help you determine a speaker’s
theoretical output capabilities, although there are many other factors that
weigh in as well.

Sd

The Sd number, usually represented in square inches or
square centimeters, represents the total usable area of the moving
speaker. This is commonly measured as
the cone’s area, plus half of the surround area. One thing that many people don’t realize,
is that these mega high excursion subs with massive surrounds actually
provide less usable cone area than a sub with a smaller surround. Since it’s also become fashionable for even
less expensive subs with lower excursion to use a big fat surround that looks
real impressive, the fact is that in many cases, a more “common” looking sub
with a smaller surround will have more cone area, and in many cases, even
with less excursion, is capable of moving more air.

Le

The Le specification represents the speaker’s
inductance. All speakers have an
inductance… after all, their voice coil is a copper winding, just like an
inductor. Most subs won’t have this
specification included, which is a shame, because while it won’t really tell
you whether a speaker is very good or not, it can tell you if it’s very bad
or not. It can also help a seasoned
speaker aficionado determine the quality of the motor. In general, stay away from a speaker that’s
got a large inductance, say 3mH or higher on a 4 ohm sub. That 3mH inductance represents the
equivalent of a 1^st order crossover at around 185hz. You want your sub to be able to play flat
at least 1 octave above the crossover frequency you select, which means that
with a corner frequency of 185hz, 1 octave below that is 93hz. By crossing any higher than that, the
natural rolloff of the sub is going to combine with the active subwoofer
crossover you use. This isn’t the end
of the world, but it’s nice if you can avoid it. Since power handling has an awful lot to do
with how much current can pass through a voice coil, it’s pretty much
necessary for high power subs to have massive voice coils, which also have
massive inductance. There are various
methods to designing the motor assembly to reduce a driver’s inductance, but
it adds considerable expense and just doesn’t exist on lower priced
subs. While this isn’t the end of the
world when choosing a sub, it’s something that you need to be aware of when
deciding “SQ” or “SPL”, because this is one of the major reasons why it’s
going to be tough to get significant levels of both in lower priced
subs. This is also going to make it
difficult to get big SPL out of an affordable sub, because it’s effectively
reducing the sub’s sensitivity at any frequency above the corner frequency
(the frequency where the high frequency rolloff begins to occur), which can
be quite a low frequency if the sub has a massive voice coil designed for
huge power handling, but hasn’t employed some of the expensive techniques for
reducing inductance. This phenomenon
is part of what makes subs with XBL^2 motors so popular for SQL (combination
of sound quality and SPL, explained below). The XBL^2 motor design helps lower inductance significantly, and does
it affordably compared to other designs.

Physical size

Make sure the damned thing fits where you want it to. Some of the new subs have very deep motor
assemblies. Most new subs have
unnecessarily deep baskets with double or triple stacked magnets, which is usually
pretty worthless, but don’t forget that sex appeal will sell more subs than
actual quality. In any case, make sure
you’ve got enough depth to mount the sub where you want. Not all 12” subs have the exact same
diameter either, so if your install includes any sort of flush-mounting or
special grills, make sure you know what the exact diameter of your sub is
too. Likewise, not all 12” subs
require the exact same 11” cutout, so if you buy a prefab box or build your
box before the subs arrive, make sure you know what size hole it needs.

suggested enclosure
size

Eh. This
specification is worthless unless response graphs are included. These days, everybody wants big bass out of
tiny enclosures, so quite often the manufacturers fib about the suggested enclosure
size. Like I mentioned above, if a sub
can play deep, you’re either sacrificing a lot of sensitivity, or it’s going
to need a massive enclosure. Stuff a
sub like that into the tiny enclosures that are often suggested by the
manufacturer, and you’re going to get a 1 note wonder with a high “Q”
alignment (I’ll explain this later too). Unfortunately, Kicker is one of the worst offenders when it comes to
this. Most of their subs sound
horrible in their “suggested” boxes, and typically should be in

boxes 3-5 times
larger.

The rest

There are plenty of other specs that may or may not be
provided with your sub, including an endless array of Thiele/Small
parameters, but most of them are not terribly important.

What kind of sub do
you want?

The first thing to determine is just what you want your
subwoofer system to do. There are two
distinct routes that you can take, SPL or SQ. SPL (Sound Pressure Level) subs are designed for one thing: LOUD. SQ (sound quality) subs tend to focus on different qualities designed
for low distortion and accurate sound reproduction. There’s a third type
that’s commonly referred to as SQL, which is basically just a blend between
the two, which considering the gray-area surrounding the issue, could mean
almost anything.

Regardless of what route you decide to take, it’s very
important to be completely honest with yourself when deciding what sub you
want. You can preach SQ all you want,
but if you’re really a boomer at heart, you might wind up ultimately being
disappointed with your SQ purchase if you can’t get the chest collapsing bass
that you really want. Likewise, many
people love the bragging rights associated with having the loudest system in
town, but once they are forced to live with it day in and day out, they can
grow tired of it. Be very honest with
yourself when deciding what your goals are.

SPL

If your goal is sheer output, and you don’t mind
sacrificing decent sound to accomplish that goal, then you’re going to want
to look for subs that are SPL optimized. What you want to look for is a sub that has high excursion, large cone
area, and high power handling. Don’t
assume that just because a sub has a massive stack of magnets and a big fat
surround that it can get loud. There
are many subs out there that look the part, that don’t act the part. If you intend to build a massive SPL rig,
then you’ll want to use a large quantity of smaller subs, say 8 10” subs
instead of 2 18” subs. This lets you
maximize your cone area for the space that you have to work with. 8 10” subs and 2 18” subs will both take up
a mounting area of roughly 20×40”. The
2 18” subs will have approximately 500 in^2 of moving cone area, where 8 10”
subs will have approximately 600 in^2. Likewise, since ported enclosures are much larger than sealed
enclosures for a given sub, your big SPL vehicles that are full of subs
usually use sealed enclosures so that they can fit more subs into a given
area. If you’re only going to use a
limited number of subs, you can get more maximum output out of that given
quantity by using a ported enclosure, but that will be explained later on. Quite often a sub designed for SPL will
have a fairly high Qts and/or Vas, and the suggested enclosure will be
smaller than what an “optimum” enclosure would be. This results in a very high Q alignment
that increases output over a narrow frequency range, at the expense of a
smooth frequency response that you’d want from an SQ subwoofer. There are a lot more variables involved
with an SPL setup than a few sub parameters, so it’s a good idea to spend
some time talking to various people (who know what they’re talking about)
about their SPL experiences, as well as taking in a soundoff or two in your
local area. Watch the DB drags and see
how people are scoring, but pay particular attention to their whole setup,
not just the brand of sub. What box
are they using, what size is the box, what is the port tuned to if it’s
ported, what amp, what kind of car, etc. All of that will effect the sub’s output abilities.

SQ

When trying to decide on a subwoofer for use as part of a
system with sound quality as the main focus, you’re going to have to put more
faith into your listening and less into published specs. The big problem with this approach is being
able to do an apples to apples comparison. If you hear 2 different subs, to have any honest insight into what one
sounds best, they both have to be in the same type of box, but one that’s
optimized for that sub. Subs will also
sound different based on the car (or showroom) they’re in, so that
complicates things even more. However,
if the manufacturer provides in-box response plots, or if you can simulate
the plots yourself, you can at least get a decent idea of what your final
frequency response may look like, keeping in mind that cabin gain (explained
later) will play a significant role in your in-car frequency response.

SQL

Like I mentioned above, SQL basically describes a setup
that attempts to strike a suitable compromise between SQ and SPL. An SQL setup is one that will provide good
sound quality and good SPL, though perhaps not the best of either. Since most high quality subs will provide
both good sound quality and good output capabilities, the line between SQ and
SQL really blurs once you get into the better quality stuff, but that’s not
necessarily a bad thing. There are a

LOT of very good sounding subs on the market that can
sound very good, and still get quite loud. One thing that you’ll want to remember too, is that thanks to the
cabin gain of our cars, even subs that aren’t geared for SPL or even SQL can
still get painfully loud. My own car
uses a pair of Oz 300L subwoofers, which are an SQ sub by almost every
account. They are not at all designed
for SPL, yet in my old Firebird, I was measured at 141.1dB with my SQ tune
and a sealed enclosure. That’s not
really as remarkable of a number these days as it was 8 years ago when I was
measured, but it is still uncomfortably loud, and more than you’ll really
ever be able to use day to day. Those
subs had an x-max of 10mm and were rated to handle 250 watts RMS. Compared to the average sub today, those
numbers are pretty modest. The point
is, in thirdgens, you can get great SPL numbers with almost anything, so even
if you want a combination of SQ and SPL, in our cars, you’ll be able to get
that with almost any good sub as long as you give it enough power.

Now that you understand SPL, SQ, and SQL, go re-read the
section on the “Le” specification. It’ll make more sense. I’ll
wait…. Ok, let’s continue. J

Impedance and wiring

Important note: Make sure you understand what bridging an amplifier is, and make sure
you understand what impedances can be safely used in this configuration. If I still haven’t written my amplifier
“how to” article by the time you’re ready to start buying/hooking up stuff,
learn it from somewhere.

One very important consideration when choosing your sub(s)
and sub amp(s) is the total impedance. As I mentioned above, a sub’s impedance is essentially its resistance
to electrical flow. The lower the impedance
is, the more current your amp can pass through the sub (i.e. the amp makes
more power), but if the impedance is too low, you’ll let the smoke out of
your amp. Things start to get
interesting when you start dealing with multiple subs or multiple voice
coils. As far as your amp (and your
wiring plans) is concerned, a dual voice coil sub can be treated like 2
single voice coil subs. From this
point on, I’m going to refer just to voice coils, because I don’t care if
they’re both in the same sub or live in separate subs.

For all of my wiring examples, I’m going to make the
assumption that all of the coils have the exact same impedance, because you
certainly should NOT be mixing and matching your subwoofers. All of your subs should be the same, which
means all of your coils will have the same impedance.

Wiring

There are two ways to connect two voice coils, you can
wire them in series or in parallel. Wiring
examples can be seen in Figure 1.

Parallel wiring is most common and will be addressed first.

When you parallel wire two coils, you’re essentially
wiring them both directly to the source. The positive terminal of each coil is directly wired to the positive
terminal of the amplifier, and the negative terminal of each coil is directly
wired to the negative terminal of the amp. See Figure 1b. This is a really easy way to wire things, but you need
to pay attention to your impedance loads. If you wire two coils this way, your total impedance is going to be
exactly HALF that of the impedance each coil has. In other words, parallel wiring two 4 ohm coils will give you a 2 ohm load.

When you parallel wire coils in doubling quantities (2, 4,
8, 16, etc), you can easily determine the impedance by continuing to cut
impedances in half ( 2 4 ohm coils gives you a 2 ohm load, 4 4 ohm coils
gives you a 1 ohm load, etc), but if you want to figure out the total
impedance with any other quantity of coils, use the following equation:

I=1/(1/R)*X

Where,

I = final impedance

R= impedance of each
coil

X= number of coils

For example, if you
have 3 coils, each with a 4 ohm impedance, the equation would look like this:

I=1/(1/4)*3

I=1/.25*3

I=1/.75

I=1.33 ohms

See Figure 1e.

Series wiring is a bit more complicated to do, but it’s
much easier to calculate your new impedance. When you series wire something, the positive terminal of the amp is
connected to the positive terminal of the first coil, and the negative
terminal of the amp is connected to the negative terminal of the last
coil. You then connect the negative
terminal of the first sub to the negative terminal of the next coil, and
continue that until all coils are connected together. See Figure 1c. When you do this, you just add up all your
coil impedances to get your net impedance.

I=R*X.

For example, if you series wire 2 coils, each having a 4
ohm impedance, the equation would be:

I=4*2

I=8.

In many cases, you will wire multiple coils in a
series/parallel combination in which some coils are series wired, then those
series wired groups are parallel wired. This is a very common practice when wiring 2 DVC subs, so I’ll use
that as an example:

We have 2 DVC subs, each with dual 4 ohm coils, i.e., we
have 4 4ohm coils.

We series wire 2 coils for an 8 ohm load, then we
separately series wire the other 2 coils for another 8 ohm load. Now you have 2 8 ohm loads, and if you
parallel wire those two, you get a single 4 ohm load. When you’re doing something like this,
always series wire the 2 coils of the DVC sub, then parallel the subs
together. See Figure 1d. You want to avoid series wiring two
different subs together whenever possible to avoid causing back-EMF issues,
which is something way beyond the scope of this article, but trust me, if you
can avoid it, you’re better off. Obviously if you’re wiring 4 single coil subs in a series/parallel
configuration, you don’t really have a choice, but with 2 DVC subs, you do.

Enclosure Types

There are many different enclosure types, but the 3 most
common are sealed, ported, and bandpass enclosures. Each has its own unique set of pros and
cons, and will be addressed individually below, but first I want to provide
some background information in order to provide some clarification on some
terms that will be used in my enclosure descriptions.

Generic enclosure
info

Regardless of the enclosure type, your goal when choosing
(or building) one is to make sure it’s rigid, free of resonances, and 100%
airtight (even on ported boxes). ¾”
MDF is the standard, and for good reason. It’s heavier and more dense than particle board, and frankly, it’s
much easier to work with. Particle
board chips and flakes all over the place, and joints are not nearly as
strong because the uneven surface doesn’t provide as much surface area for
glue to hold. You can probably get
away with ½” or 5/8” MDF for very small enclosures, but for the amount of
space you’ll save, it’s really not worth it. On larger enclosures, it’s a good idea to use bracing to stabilize
larger panels. The larger the panel
is, the more it will flex and resonate. When making big dual sub enclosures, if you don’t use a divider
between the two subs, at least stick a brace or two in there to connect the
two parallel panels that are the longest. In simple speak, if the box is 40” long, at the 20” mark, put a brace
across the box. I’ll get more into
depth on sub enclosures when (if) I write an installation guide.

Box modeling
software

If you are interested in designing your own enclosures, or
if you are interested in learning about how different enclosures effect the
sound, I would strongly suggest that you download WinISD. It’s a pretty simple program, and once you
get the hang of it, it can be a very useful tool. There are also plenty of similar programs
out there that you may find useful.

Go here to download a copy: http://www.linearteam.dk/default.aspx?pageid=winisd

Hoffman’s iron law

This is some useful information to digest before I get
into specific enclosure types. In a
nutshell, Hoffman’s Iron Law says that the efficiency of a woofer is
proportional to cabinet volume multiplied by the cube of the F3. In other words, you want a small box, high
efficiency and deep bass. You only get
to pick 2. If you want a small box
(and don’t we all J), you can have deep bass or
high efficiency, not both. Likewise,
if you want deep bass, you get a small box or high efficiency, not both. Last, if you want high efficiency, you get
deep bass or a small box. There is no
way to cheat this law until you find a way around the laws of physics, so
it’s something that you need to be aware of when evaluating your needs, and
also when evaluating manufacturer’s sometimes outlandish claims.

Cabin Gain

Another thing that I want to address before moving on is
the effect of cabin gain. This is
basically like getting a free bass boost, just for owning a car! The interior of a car is so small, that you
get massive boundary reinforcement at low frequencies, which causes a
tendency for an exaggerated bass response that increases as frequency
decreases. At just what frequency this
begins to occur will depend on the vehicle, and I’m unsure about where it
begins to occur in a thirdgen, though I hope to be able to solve that mystery
once I get my measurement tools. It
seems to be commonly accepted that it tends to become an issue somewhere
around 50hz in most coupes, though it is most certainly present even at
frequencies above that. Cabin gain
generally gives you a bass boost at approximately a 12dB/octave rate, which
is convenient since that’s around the same rate that a sealed enclosure will
roll off on the low end. Combine the
two, and you have the potential for a nice flat response all the way down to
(and below) 20hz.

Sealed

Sealed boxes are the most common, and for good
reason. Not only are they the easiest
to build, but they are the most forgiving to slight variations. If your box is a hair too big or a hair too
small, it’s not going to have a major impact on the sound. This is a major advantage when buying
pre-fabricated enclosures, because in those cases, you’re not going to have a
lot of choice in what size it is. Figure 2 shows a sealed enclosure.

Now let me explain something called the Qtc. Electronics people will know that the Q is
used to describe the slope of a filter, and this is really much the
same. The low frequency rolloff of a
speaker is an acoustic filter that behaves basically the same as an electrical
filter used in a crossover, except instead of capacitors and inductors, our
components are the subwoofer and the enclosure. The subwoofer has the Qts that you learned
about in the beginning of this article, and as you hopefully remember (I’ll
wait while you go re-read it), the Qts represents the speaker’s ability to
control its motion at resonance. Well,
you have a similar phenomenon occurring with your enclosure. The combined Q of the speaker and the
enclosure is your Qtc.

Generally, you should aim for a sealed box Qtc of about
.707, at least to start off. This will give the lowest F3 (-3dB point in the
rolloff), and will give a smooth 12dB/octave rolloff, which can compliment
the car’s cabin gain nicely. As you’ve found, as you go larger with the box,
the low frequency rolloff is not as steep, but your F3 moves up. When you go
smaller with the box, the rolloff is steeper, the F3 moves up, and you start
to get a nasty bump in response right below the dropoff. Unfortunately, with
the small box craze the way it is, all too often this is what you get when
you use a manufacturer’s “suggested” enclosure. Kicker is notorious
for suggesting unrealistically small boxes, but the boomer crowd loves it
because they get that big boost in output, even if it is in a narrow range.

Attached is a graph (Figure 3) with the same subwoofer plotted with a Qtc of
.707, .5, and 1.0.

Figure 3

The yellow plot (Figure 3) has the Qtc of .707, and you can see that this one
has the lowest F3, and has a nice smooth 12dB/octave rolloff, which will help
compliment the car’s cabin gain nicely.

The orange plot (Figure 3) illustrates a Qtc of 1.0, which means that the box
is too small. You can see that your F3 moved way up, and that the overall
response really isn’t very good. You get a bump in output, followed by a
sharp dropoff. This is sometimes preferred by the boomer crowd, provided the
bump in output isn’t at too high of a frequency.

The red plot (Figure 3) shows the Qtc of .50, which is what’s called
“critically damped”. You can see that the F3 moved up, but the
rolloff is more gradual. When combined with a car’s cabin gain, this will
likely sound a bit lean in the upper bass, but will have the strongest deep
bass of any sealed alignment.

One thing that’s interresting to see here is that your frequency response
really doesn’t change all that drastically even with tripling the box size,
however, cutting the box size in half really has a profound effect. What you
can’t see here is how the box size effects cone excursion. A smaller box will
have a much larger resistive force on the cone motion, which will help increase
power handling (physical power handling, NOT THERMAL). However, this
increased power handling is at the expense of low frequency sensitivity, so
you’re still not getting additional output, just requiring more power to
achieve it.

Equations

For those who want to play with a calculator instead of
WinISD, here are some equations for you. All equations are shamelessly copied from www.diysubwoofers.org because I’m
not going to spend time reinventing the wheel when somebody else already did
it for me.

Qr = Qtc/Qts

Else dBpeak = 0

where,

Vb = net box volume (litres)

dBpeak = maximum peak or dip in system response

Figure 4

Ported

Ported enclosures offer some advantages and disadvantages
over sealed enclosures. A ported
enclosure can provide a flat response to a much lower frequency (i.e. a lower
F3), but the rolloff below the F3 is much steeper. Thanks to the added low bass potential in a
car from cabin gain, a ported enclosure can wind up giving you a bottom-heavy
response in which the lowest couple octaves are exaggerated compared to the
upper bass range. However, having the
ability to get much more deep bass out of a speaker can be very handy, like
in cases where high SPL is desired, or in cases where you’ve got an equalizer
handy in order to flatten the response back out, giving you the same sound
with less power, or more output without sacrificing the sound quality. It helps you extract the most out of
smaller speakers. Figure 4 shows a
ported enclosure.

One major drawback to ported enclosures is that they are
obviously more complicated to design and build, and they are also much less
forgiving to variations in size, and variations in port tuning. This really makes it difficult to buy a
pre-fabricated ported enclosure and be able to get good results unless you
just happen to luck out. This is where
box design software like WinISD really comes in handy, because plugging away
with ported box calculations will take forever. Just be careful, because if you get too far
away from the “expected” or “common” t/s params, and try modeling some really
unorthodox stuff, WinISD can get a little bit misleading. But, for normal usage, it’s really a
fantastic tool. If your box is tuned
improperly, the result can be exaggerated output over a very narrow frequency
range (usually centered around the tuning frequency). This may be a benefit if you’re an SPL
competitor, but not if you want good sound quality. A box with a huge peak at one frequency is
known as a 1-note-wonder, because you really only hear a very small portion
of the music.

One last drawback to ported enclosures that needs
mentioning, is cone excursion. At
frequencies above the port tuning frequency, cone excursion will be very well
controlled, and will actually be lower than that of a sealed enclosure, but
at frequencies below the port tuning, as there is no internal box pressure to
control excursion, and the result is that the sub can easily reach it’s
excursion limits with very little power. This isn’t a big issue with tuning frequencies down in the 20-25hz
range, because unless you’re a test tone junkie, you’re not likely to come
across much content down there (unless you play movies… movie sound effects
can get very deep). However, start
tuning in the 35-40hz or higher range, and now you have the potential for
significant content to be below your tuning, and that can kill your sub in a
hurry. Many subwoofer amps come with a
subsonic filter, which is designed to block these lowest frequencies, and it
is a very good idea to have one if you intend to run a ported box.

Figure 5 shows the first ported example. I’ve only got 2
alignments here, the “standard” alignment shown in green , and the
“EBS” alignment in blue.

The EBS (extended bass shelf) alignment allows you to get a lower F3, but
it’s at the expense of a smooth response. You can clearly see (Figure 5) that
it shelves off for a while, then falls below the port tuning frequency. This
type of arrangement really isn’t used all that often in a car environment,
though it can be useful if you’ve got the space, because it helps push the
port tuning frequency lower. Cone excursion below the tuning frequency in a
ported enclosure is very, very high, and can easily cause damage to the sub
if not controlled. For that reason, some people may intentionally tune the
system low, and compensate with equalization, although generally once you
factor in a car’s cabin gain, you need heavy equalization to get rid of the
excess bass, not to add to it. Since very little music has much content below
around 35-40hz, it’s usually a bad idea to tune your port above that area
unless you run a sub amp that’s got a subsonic filter that you can set just
slightly below the tuning frequency.

The green plot (Figure 5) shows the standard type of ported tuning in which a
flat, smooth frequency response is desired. As with the EBS alignment, cone
excursion is very high below the tuning frequency. This standard alignment is
what WinISD will default to when you load a ported setup, but it can easily
get confused if the t/s parameters are outside of a certain range. Once the
Qts of a woofer gets too large, it becomes impossible to get a reasonable
“standard” alignment without a massive box volume, and if the Qts
gets even higher, it will be impossible to get a “standard” alignment
regardless of box size. Maybe I’ll make a separate graph to illustrate that
situation.

Figure 6 shows a couple ported enclosures that are too small.

First off, the yellow plot (figure 6) is the optimally damped (.707) sealed
enclosure, just for comparison. Notice that the volume is about 1 cubic foot,
just like the ported configuration I chose for this example.

The green plot (figure 6) is the optimally damped ported enclosure, used as
another comparison.

The orange and blue plots (figure 6) represent ported enclosures, each 1
cubic foot in volume. Their difference is the port tuning frequency.

For the orange alignment, the tuning frequency is about 23hz, just like the
optimized green plot. This gives you a very good idea of what happens when
you change just the box volume but not the tuning frequency. Overall the plot
really doesn’t look bad, in fact, it’s a very nice compromise between a
conventional sealed alignment and an optimized ported one. The problem with
this alignment isn’t immediately noticeable, and that’s port length. When
using a 3″ diameter port, the length would need to be 34 inches. Good
luck fitting that into anything reasonably sized. The smaller your box is,
the longer your port needs to be to achieve the same tuning frequency.
Re-read that, it’s very important. The same tuning frequency in the 1.8 cubic
foot plot (green) is achieved with a port only 16″ long! If you build a
complicated slot-port, or even if you decide to use a million elbows to fit
the port into the enclosure, since the port’s volume is NOT considered part
of the enclosure volume, you’d have to make the box large enough to fit all
of that port volume in addition to the 1 cubic foot of airspace. The result
is a box that’s bigger than if you just built the “standard” ported
box to begin with.

So, we move on to the orange plot. This one shows the same 1 cubic foot, but
with a more manageable tuning frequency. This port length is only 13″

long because it’s tuned to 35hz instead of 23hz like the other two. But, you
can see the disaster of a frequency response. This is a nice setup if you’re
a boomer who just wants the most output possible from your setup, but if you
want any kind of sound quality, forget about it. It’s a 1-note-wonder.

Equations

Please note that this set of equations is just going to
give you the “optimal” ported alignment. The equations change quite a bit for different alignments, such as an
EBS alignment. You’ll wear out your
calculator trying to model a bunch of different alignments, so just get
WinISD.

Vb =
20*Qts^3.3*Vas

dBpeak = 20*LOG(Qts*(Vas/Vb)^0.3/0.4)

where,

Dv = internal diameter of port (cm)

Vb = net box volume (litres)

dBpeak = maximum peak or dip in system response

Bandpass

A bandpass box has some very unique qualities, but also
has some pretty significant drawbacks. A bandpass box is basically a sub that’s got a sealed enclosure on one
side of it, and a ported one on the other side. Figure 7 shows a bandpass enclosure. A bandpass enclosure does just as its name
implies… it only allows a certain range of frequencies to pass, since all of
the sound is directed through the port.

Drawbacks to this configuration are significant. First of all, they are very difficult to
design properly. Forget about
crunching numbers, let simulation software like WinISD figure it out for you. Fool around with changing things like
enclosure volumes or port tuning and see how even the smallest of small
changes can completely screw up the frequency response. These boxes are also more difficult to
build. Not only do you have a much
more complicated enclosure to build, but you’ve got to design a way to be
able to remove the woofer, which usually means a removable panel on the
sealed side of the enclosure. On top
of that, you’ve got to be precise, because if you’re off a hair here and
there, the resulting changes to the enclosure sizes and/or port tuning will
have a noticeable effect on the sound. Due to the very precise nature of the bandpass alignment, forget about
finding a prefab that’ll ever work. If
the box isn’t just right, you’re going to get the king of 1-note-wonders.

Another major drawback to the bandpass box is that with
many subs, the high frequency rolloff may be at too low of a frequency. For example, in figure 4 you can clearly
see that the F3 on the high side is only 65hz. This isn’t going to work real well in an
application that requires the sub to play up to 80 or 100hz.

Yet another drawback to this alignment is also one of its
advantages, and that’s the fact that you get an acoustic high pass
filter. I’ll explain its benefits
below, but it’s drawback is that it becomes impossible to hear the audible
distortion that you get from overdriving a sub until it gets insanely
bad. Once you start hearing
distortion in a sealed or ported enclosure, it’s a sure sign that you’re
overdriving something (either the sub or amp) and that if you don’t turn it
down, you’re going to destroy something (usually the sub). Since the bandpass enclosure filters out a
significant amount of this distorted sound, you’ll usually never know that
you’re destroying things until it’s too late.

The most significant benefit to a bandpass enclosure is,
once again, the fact that it provides an acoustic high pass filter, which is
nice for a few reasons. Harmonic
distortion in a speaker results in sound at frequencies where there shouldn’t
be any. For example, a 50hz tone has a
2^nd harmonic at 100hz, a 4^th harmonic at 200hz, and so
on. This is called even-order harmonic
distortion, and can actually be pleasant sounding, but that’s another story
all together. For the sake of this discussion,
we’ll just say it’s bad. Even order
harmonics can be somewhat difficult to detect because since they are
occurring at multiples of the fundamental, they blend nicely with that
fundamental. This is also partly why
it can sound good… it can add a richness to a tone, but since that wasn’t
there on the original sound, we don’t want it there when it’s
reproduced. A speaker will also have
odd order harmonic distortion. That
50hz tone will have a 3^rd order harmonic at 150hz, a 5^th order at 250hz, and so on. Odd order
harmonics sound nasty no matter what, because they’re occurring at
frequencies that won’t be masked by the fundamental. Since this distortion occurs in the
speaker, if you run an active (or even passive) filter, you’re attenuating
the higher frequency stuff BEFORE it gets to the sub. Your sub is still going to produce these
harmonics, which if bad enough, will be audible. In addition to sounding bad, these upper
frequencies will make it easier to localize the direction of the bass. Our ears can’t determine direction of low
frequencies, but if that sub starts making noises at higher frequencies, even
if they’re too low for us to be noticeably audible, our ears can interpret
what they’re hearing and localize where the sound came from. When you have a bandpass enclosure that’s
filtering out the higher frequencies AFTER the sub creates them, you
eliminate a lot of that harmonic distortion, which can provide a cleaner
sound, as well as “up front” sounding bass, because with the absence of those
upper harmonics, your ears won’t be able to tell where the sound is coming
from, and the upper harmonics that are naturally in the music, which are
coming from your front speakers, will make your ears believe that it’s coming
from up front.

Figure 8 shows a typical bandpass response (in blue) along
with the same sub (also the same sub in all of these other examples) in a
conventional ported enclosure for comparison (in green).

Another benefit to bandpass enclosures is that they can
look very nice, especially if you implement plexiglass so that you can see
into one of the chambers. A lot of the
commercially available bandpass enclosures offer this, which is why they
continue to have some level of popularity.

Equations

Yeah, right. If
you’re really up for the punishment, get them from www.diysubwoofers.org, but trust me,
WinISD or some other type of simulation software will help you keep your
sanity. The most valuable thing I’ve
learned from getting my Engineering degree is when to give your calculator the
big middle finger and reach for simulation software instead! None of the bandpass equations are
complicated, but there are a lot of them, and at some point, you’re going to
swap a number or invert a fraction.

Plates/ Infinite
Baffle

Everybody repeat after me, “PLATES SUCK.”

Now that I have that off my chest, let me take a step back
and explain what Infinite Baffle (IB) is. An infinite baffle enclosure is basically a sealed box with an
infinite enclosure size. Obviously in
most applications, especially in cars, there’s no such thing as an infinitely
large enclosure, but an enclosure that’s very, very large, can behave
remarkably close to it. Generally any
enclosure whose volume is at least 2 or 3 times (depending on who you talk
to) the size of the speaker’s Vas is considered infinite baffle. A great example of this is a home theater
that has the subs mounted on a baffle that utilizes another room (such as the
basement or attic) as the enclosure. Another great example is one that uses a board behind the back seat of
a sedan in order to utilize the entire trunk space as the enclosure. In both examples, it is vitally important
that the rear chamber is completely sealed off from the listening area. In a house, this means no heat vents or
open doors connecting the other room to your listening room. In your car, it means that you must
completely seal off your trunk area from the interior.

When choosing a sub for an infinite baffle installation,
you’re going to want a speaker with a Qts at or near .7. Since you won’t have an enclosure, your
final Qtc is going to equal the sub’s Qts. In most cases you’re also going to want to find a sub that’s either a
very robust design, or one that’s specifically designed for IB use, because
without the compliance of the air mass in the enclosure, nothing is going to
keep the cone excursion under control. A sub designed specifically for IB tends to have a very stiff spider
and a very stiff surround in order to control the excursion. Some high excursion subs have very robust
designs anyway, so they can work well in IB situations provided their
specifications are appropriate, and you take it easy with the power.

So, all of this sounds pretty cool. You get great bass, without building a box,
and you get all of your trunk space. Sweet! Well dummy, you’re
driving a thirdgen, not a Camry. You
don’t have a trunk, you’ve got one big interior space that’s all
connected. The only way to get a true
IB installation in your car is to cut a hole in your roof and mount the sub
to it so that the rear chamber is the entire free world. Don’t laugh, I’ve seen it.

So, let’s get back to the plate idea. There are a lot of people who seem to think that a plate
installation is just like an IB installation. Those people are all wrong. The biggest problem with a plate
installation is that you aren’t properly isolating the rear wave from the
front wave. Even if you spend your
time doing a real good job shaping that plate, and making sure that you get a
nice air-tight seal all around the plate, you’re neglecting the fact that the
whole back part of your interior is made of ill-fitting plastic panels. There’s nothing stopping the rear sound and
pressure from getting around (and through) those plastic panels where they
meet the floor, and where they meet each other. Likewise, once it’s back there, it will get
through all of the cracks around the plastic and into your interior, causing
massive cancellation with the sound coming directly from the sub.

So let’s say that you’re going to spend a bunch of time to
reinforce all of those plastic panels, and seal off all of the gaps between
them and between your floor. Neat. Unfortunately, to do this
right is SO much more work than just building a box out of MDF or fiberglass,
and even after you do all of that, you no longer have an IB setup because the
airspace is too small… you just have a regular sealed enclosure that’s
probably way too small for your sub if you bought one designed for IB.

You should have just built or bought a sealed box in the first place.

Equations

Plate=SUCK

What subs work with
what enclosures

So now you know all about the different enclosures, but
how do you know what type of sub is best suited to each? There’s no hard and fast rule here, but
there are a bunch of guidelines that can help simplify things. In all cases though, if you’ve got WinISD,
I would encourage you to model the sub in both sealed and ported arrangements
to see how they behave. That being
said, there’s a quick way to get a rough idea of what type of enclosure the
box is going to work best in. This is
known as the EBP, or Efficiency Bandwidth
Product. This number defined as Fs/Qes. Generally if the EBP is less than 45, the
sub is best suited for a sealed enclosure. If the EBP is above 65, it’s probably best in a ported box.

There are other clues that you
can take from looking at the various t/s specs, but that’s a bit beyond what
I want to get into here. If you become
knowledgeable with WinISD, as you model more subs, you’ll begin to be able to
get a good estimate of what a sub’s plot will look like before you do it. Generally you’ll find that a sub with a
lower Qts is going to require a smaller enclosure, but will have a higher F3
than a comparable sub with a higher Qts. Often a sub with a lower Qts will work better in a ported enclosure in
order to help you get that deeper bass out of it. Subs with a higher Qts will be able to get
deeper if you can afford the additional enclosure space, but once the Qts
gets too large, you may find that a ported enclosure may just be too large,
plus that higher Qts will help the sub get plenty deep in a sealed box.

Another thing that you’re going
to want to be careful of is port length. Many of the subs, even those that will work great in ported enclosures,
require a small enclosure, but with a relatively low port tuning
frequency. This can be a problem,
because the smaller your enclosure is, the longer the port needs to be in
order to achieve a given tuning frequency. With many subs, it is not at all uncommon to require a tiny box with a
ridiculously long port. Obviously this
is going to be a major challenge, and is something that you’ll want to
consider when deciding what sort of alignment you want. Generally, you’ll find this situation
exists with subs that have a small Vas and a low Fs.

Bandpass applications tend to
work with the same speakers that work well in sealed applications… after all,
a bandpass enclosure is just a sealed box with another chamber that acts as
an acoustic filter.

For infinite baffle
applications, you’re going to want to look for a subwoofer with a Qts of
around .5. I would try to steer clear
of subs with a Qts above .7, but for a typical car IB application (usually in
a sedan with a trunk), the closer to .5 you get, the smoother your frequency
response will be. In a home theater
application where you’ve got a truly massive IB chamber (like an attic or a
basement), subs closer to .7 will likely yield better results. Just remember that there is no good way to
use an IB sub in a thirdgen!

Sound quality vs.
sub size and box type

There’s a common misconception out there that different
size subs sound different. While there
is a sliver of truth to it, it’s mostly all crap. Smaller subs aren’t “faster” “tighter”
“punchier” or whatever people like to call it. Any difference that a person hears from one
size sub to the next is based on one thing: Frequency response. Any sub
that can play a 200hz signal sure as heck isn’t going to play a 50hz signal
“faster” or “slower” than any other sub. While it’s technically true that different designs may be capable of
accelerating from a standing position more rapidly than others, the variables
are insignificant in cases where the sub is at least capable of playing
frequencies significantly higher than you’ll be using. People often confuse transient response for
the “quickness” of a sub. The
transient response represents how long it takes a speaker to react to an
input signal, which when you do the math, boils down to the amount of current
going through the coil and not much else. Since a coil with a higher inductance is going to have an effect on
current storage, your transient response is more dependant on your sub’s
voice coil inductance. However, the
extent to how audible the transient response is can be debated forever. For more info on this, check out Adire’s
white paper: http://www.adireaudio.com/Files/TechPapers/WooferSpeed.pdf

Anyway, back to my rant about sub sizes. What people tend to refer to as “faster”,
“tighter” “punchier” or whatever, in my opinion, is simply a lack of lower
bass output. When you have a little
bit less of the lowest octave or two, the bass sounds “quicker” because the
sound is made up of more high frequency (relatively speaking… say 50-60hz
instead of 40-60hz) content. Since
smaller subs tend not to play as deep as larger ones (a vast generalization),
people attribute the differences they hear to the size of the sub instead of
what’s actually going on. If you EQ a
10” and 12” sub to have the same frequency response, you’ll never hear the
difference.

So why did I lump box type into the same discussion as sub
size? Because it’s the same
issue. The sealed vs. ported debate is
just as hotly discussed as the 10” vs. 12” debate, and it boils down to the
same thing. Ported enclosures have
more low frequency output, which is why they are described in the same way as
larger subs. “slow”, “sloppy” blah
blah blah. It’s all frequency
response. Ported enclosures also get a
bad reputation because there are so many poorly designed (or implemented)
examples out there.

Recommendations:

This section isn’t as clear cut for subs as it is for
other things. Your final result will
be heavily dependant on the combination of your sub, your enclosure, and your
amplification (at least for SPL purposes), but I’ll give a few options based
on your interests. Please note that I
have not personally heard all of the subs on this list, so many of these recommendations
are based on the opinions of trusted individuals, combined with overall
reputation.

SQ recommended:

Dayton
Reference Series (HF or HO)

Dayton
Titanic Series III

Peerless XXLS

Ascendant Audio

Soundsplinter

Oaudio

Adire Audio

Boston
Acoustics

Image Dynamics

JL Audio

Focal

Arc

SQL Recommended:

Ascendant Audio

Soundsplinter

Adire Audio

Image Dynamics (IDMAX)

JL Audio (W6, W7)

Alpine (Type R)

Resonant Engineering

Diamond Audio

SPL Recommended:

Resonant Engineering

Kicker

Digital Designs

Not Recommended (the usual suspects):

Audiobahn

Volfenhag

Pyramid

Pyle

Sony

Boss

Legacy

And plenty of others!

Conclusion

Well, if you managed to make it all the way to the end, I
hope that you learned a thing or two, and that you’ll continue to learn from
this by re-reading it as you learn more through your own experiences.

While this article probably made you feel a bit
overwhelmed by just how complicated buying a subwoofer can be, just remember
that it’s only as complicated as you make it. There are literally thousands of subs out there to choose from, but as
long as you take the time to audition them, you’ll find one that suits your
needs. A lot of the information here
goes way beyond what most people will need or want to know about subs, but
don’t think that you need to know
all of this to buy a sub that you’ll be happy with. In the end, make sure you listen to what
your ears are telling you, and when it’s all said and done, sit back and
enjoy the tunes!

For more information, give these sites a try:

1. http://www.diysubwoofers.org

Hope
this helps.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8