Hope this is the right group to ask such a question.

I am trying to build my own speakers from a design I found here...
http://www.fane-acoustics.com/public...closure_12.pdf

Looking at the size of the vents in this design I am unable to
reproduce the way they established the length of the vent ports. Can
anyone shed some light on this?

According to my own calculations (using a formula I found on internet
and in some books) I calculate a length of the vent ports approximately
double that of what the speaker manufacturer puts in his design.

jansb000 wrote:
Looking at the size of the vents in this design I am unable to
reproduce the way they established the length of the vent ports. Can
anyone shed some light on this?

According to my own calculations (using a formula I found on internet
and in some books) I calculate a length of the vent ports approximately
double that of what the speaker manufacturer puts in his design.

The port length required depends upon a just a few parameters,

* the frequency to which the enclosure is tuned,
* the internal volume of the enclosure
* the diameter of the port

So those are the three parameters you must know in order to
determine the length of the port. If any of them are missing,
you can't calculate the length.

The tuning frequency is constrained by how the overall system
is "aligned," and that's determined by the driver and the desired
system response function (these two, however, could very well
be mutally exclusive).

Since in a helmholtz resonator (which is what a ported system is),
the resonant frequency is:

Fb = 2 pi sqrt( 1/(Cab*Map) )

where Fb is the resonant frequency of the enclosure, Cab is the
acoustic compliance of the box, and Map is the acoustic inertance
of the port, a little algebra will show you that:

Map = 4 pi^2 / (Cab Fb^2)

Now you have the acoustic inertance required to achieve that
frequency with that enclosure volume.

The port supplies the acoustic inertance, and is determined as
follows:

Map = pl' / S^2

where p is the density of air, l' is the end corrected length of
the port, and S is the cross-sectional area of the port, which
is, of course, determined by:

S = pi d

where d is the diameter of the port. Combining these last
two:

Map = pl' / (pi^2 d^2)

(note, all measurements in the mks syste, e.g., meters,
kilograms, seconds, etc.).

Again, a little algebra tells us that

l' = pi^2 d^2 Map / p

And finally, l' is the 'end corrected' length: the port is, in effect,
slightly longer than the physical length, in a way that's dependent
on the diameter. Assume that one end of the port terminates
flush with the cabinet wall, and the other is free (even if inside
the cabinet), that end correction is:

l' ~= l + 0.73d

where l is the physical length. One more slight of algebraic hands
then gives us:

l ~= l' - 0.73d

So that's how we get there.

Now, to repeat, you HAVE to know the intended enclosure
resonant frequency, the volume of the box and the diameter
of the port. Without those, you can't determine the length.

And, without many more details on your particular case, there
is no way to evaluate if the numbers you are getting are right
or wrong.

Could be that since there are two ports in this design, each one would be
half as long as if there was only one.

James. )

wrote in message
ups.com...

According to my own calculations (using a formula I found on internet
and in some books) I calculate a length of the vent ports approximately
double that of what the speaker manufacturer puts in his design.

James Lehman wrote:
Could be that since there are two ports in this design, each one would be
half as long as if there was only one.

James. )

Actually, doubling the number of ports would require doubling the
length of each to achieve the same effective acoustic inertance, in
much the same way that putting two inductors in parallel would
require each inductor to be twice that of the desired total inductance.

Another way to look at it is that two ports whose diameter is d
have a toal corss sectional area twice that of a single port, and
wince the length . Since the inertance goes as the length and
as the reciprocal of the area, then to achieve the same inertance
with twice the area requires twice the length.

"jansb000" wrote in message
ups.com...
Hope this is the right group to ask such a question.

I am trying to build my own speakers from a design I found here...
http://www.fane-acoustics.com/public...closure_12.pdf

Looking at the size of the vents in this design I am unable to
reproduce the way they established the length of the vent ports. Can
anyone shed some light on this?

According to my own calculations (using a formula I found on internet
and in some books) I calculate a length of the vent ports approximately
double that of what the speaker manufacturer puts in his design.

I have a function generater connected to the amp that is connected to the
speaker getting tuned.. I connected a digital VOM to speaker and measured
the ouput at various frequencies. Since the room acustics and recording
speaker's response vary at different frequencies, the test is only good for
comparing different port lengths at a given frequency. From that, I was able
to tune the ports to a smoother roll-off. Too short gave a boomy response
around 50Hz, too long and 40-60 Hz is weakened and too much response down in
the nearly unused 30Hz area. The result is smooth efficient bass with usable
output down to 30Hz.

The 4" ports I used were not very expensive, so I bout an extra to cut it
down for the test.

John

Can this process be reversed: Knowing the dimensions of the port, the
cabinet and the thiele-small parameters of the driver and from there
calculate the resonant frequency?

It strikes me that the cabinet design can be used for two drivers from
FANE, namely the Colossus 12MB and the Cresendo 12MB. These drivers
have different thiele-small parameters and therefore would result in
different tuning of the driver/cabinet combination.

Can this process be reversed: Knowing the dimensions of the port, the
cabinet and the thiele-small parameters of the driver and from there
calculate the resonant frequency?

It strikes me that the cabinet design can be used for two drivers from
FANE, namely the Colossus 12MB and the Cresendo 12MB. These drivers
have different thiele-small parameters and therefore would result in
different tuning of the driver/cabinet combination.

jansb000 wrote:
Can this process be reversed: Knowing the dimensions of the port, the
cabinet and the thiele-small parameters of the driver and from there
calculate the resonant frequency?

The enclosure resonant frequency is a function of the port dimensions
and the cabinet volume. That's it. So if you have a cabinet of
such-and-
such dimensions and it has port(s) of so-and-so size, that's all you
need to determine what the enclosure resonant frequency is.

However, whether that cabinet is at all suitable for any given driver
you pick is another issue altogether.

If you have a driver in hand, the parameters of the driver and a
knowledge
of what the desired system response function is supposed to be then
determines what cabinet volume is suitable and what the enclosure
resonant frequency should be. Thus, it's altogether possible to have
a driver and a cabinet and a set of ports that have no business being
put together.

It strikes me that the cabinet design can be used for two drivers from
FANE, namely the Colossus 12MB and the Cresendo 12MB. These
drivers have different thiele-small parameters and therefore would
result in different tuning of the driver/cabinet combination.

Yes, and the fact that they have different Thiele/SMall parameters
could very well mean that they need a different cabinet volume than
what you have.

Selecting the proper cabinet volume is as important a part of the
system design function as selecting the cabinet resonant frequency.
You can't just stick the driver an any cabinet and try to tune your way
to an optimum system response.