[Dick Pierce[_2_]]

Doug wrote:
I'm making a stereo that is to be mounted on the back of my bike.
It's in a box that has internal dimensions of about 12" x 6" x 6" and
inside this you'll find a T-amp (it's rated at 100 watts/channel, but
I'm not going to power it anywhere near that high) and two 7 Ah lead
acid batteries. Input will come from an iPod that's outside the box.
...
As I see it, there's a few advantages for porting --

-- better low frequency response

Only if the drivers and enclosure together are suitable for
a ported design.

-- make it more efficient -- use less power for the same volume
(the power limited is limited, so this is a concern.)

Porting an enclosure WILL NOT make the system any more
efficient. Not in the least. Ported systems CAN be more
efficient, not because they have ports, but because they
have drivers suited for the enclosure that result in
electromechanical parameters that mke the DRIVERS more
efficient to begin with. A more efficient driver designed
for a prted enclosure will result in a system with
exactly the same reference efficieny whether the port is
there or not.

-- if the batteries do vent some hydrogen when charged, it gives it
an easy place for it to escape. It also lets the amp cool a bit
easier, though it's so efficient and the power relatively low so
I don't think that's a big concern.

But the stuff on calculating port size is quite confusing. I imagine
I could calculate the frequency of the box (though the batteries and
amp inside the box are likely to confuse that calculation) and guess
at the Xmax values and go from there.

Wrong.

I could just guess, put in a port that's about as large as I have room
for, though if I understand it correctly, I want to tune the
diameter/length so that the wave coming out of the port is in phase
with the wave coming out of the front of the speaker -- so it *does*
matter.

Wrong.

But wouldn't this calculation only be valid for a specific
frequency?

The calculators I find online talk about calculating things for
woofers and subwoofers. I guess the main speaker *is* a woofer. But
when the calculators as for a frequency ... is it the Helmholtz
resonator frequency for the cabinet I should use?

Maybe I just need to bite the bullet and find a book on speaker design
-- since it seems that that's exactly where I'm headed.

You simply cannot just take a pair of speakers, jam them
in a box, stick a port in it, and expect 1) for it to work
reasonably well, 2) for it to have "better low frequency
response and 3) more efficiency. It simply does not work
that way.

A speaker is a SYSTEM composed of drivers, enclosures and
amplifiers.

If I were hired to do such a job, the first thing I'd
do is sit down and try to come to an agreement about
what constitutes "efficient" and "good low frequency
response". The third factor is enclosure size, but you've
already constrained that, so the only two variables you
have left are efficiency and low frequency cutoff. Things
like XMax play no role at this stage: it's on;y relevant
in determining the MAXIMUM output.

And the conversation would go something like this: you
have presented me with a MAXIMUM enclosure size of 0.25
cubic feet, and maybe more like .15 cubic feet considering
the batteries. That's all of 5 liters. And you're going to
put TWO drivers in there, meaning the effective volume is
half that, 2.5 liters. And, at best, you're talking 5"
drivers. I'd stay with 4"

Now, the classic Thiele-Small efficiency/bandwidth/size
rule takes over and constrains reality. SP you have 2.5
liters per driver. The MOST your are likely to find in
the efficiency for a 4" driver is on the order of 89dB
SPL @1W/1m. That means, the absolute BEST you can expect
for a low-frequency cutoff is about 90 Hz. And that's
only with a driver PERFECTLY suited to the application.
Not just any 'ol driver. It would have to have T/S
parameters like and Fs of 90 Hz, Vas of 2.6 L,
Qt of around 0.4, and that would require a cone mass
of nearly 12 grams with magnet Bl product of 7 N/A.
That's a REAL tough driver to build.

And yes, then you'd stick two of them in the box, and
you'd stick two ports in, let's say you made them out
of 1" PVC, each about 2 7/8" long.

It's doable, but not the way you want to go.

Then again, just sticking what you have in the box
may well be good enough for a bike. That's for
you to decide.

--
+--------------------------------+
+ Dick Pierce |
+ Professional Audio Development |
+--------------------------------+