This article is copyright © 2006 by Audio Amateur Corporation. PO Box 876, Peterborough, NH 03458, USA. All rights reserved.
To design a double chamber system, just start by selecting a reflex alignment for the driver in use, taking into account all the known procedures to obtain a well sounding reflex system. To help you deciding, you can use one of the numerous software available; I suggest Unibox that is found in the Download folder. Then divide the total volume of the reflex (V_{B}) so to obtain a partition that divides the box into two unequal chambers, with the bigger one the double of the smaller one; the woofer is mounted in the bigger chamber. For example if your reflex V_{B} is 60 liters, divide it by 3 to get the smaller chamber volume (V_{2 }=20 liters) that is multiplied by 2 to get the bigger chamber volume (V_{1}=20x2=40 liters). As we have seen in the Impedance Comparison chart before, we can assume that F_{B}=F_{2}_{ },_{ }therefore we can go on calculating F_{1 }and F_{3 }using the seen formulas. But what happens when the chambers volumes are not kept in the 2:1 proportion? I investigate it by leaving the V_{2 }volume always the same and changing V_{1 }so to have a one liter bigger chamber and a two liters smaller one, in comparison to the ideal 2:1 volume; the DCR total volume (V_{1}+V_{2}) was of 35 liters. In the impedance chart below you can see the results: while the F_{2} is almost the same for the three curves, the F_{3 }shows some notable differences in the two liters smaller than ideal curve.

To better understand what the differences found in the
impedance chart could bring, I compared the near field responses of
the three systems; these measures were taken without adding the ports
responses since I wanted to see the systems tuning frequency (F_{B}=F_{2})
given by the typical response notch. The Near Field chart confirms what we
already saw in the impedance comparison, as it should be; the three
curves show an F_{2 }difference inside 0.8 Hz; after 70 Hz
the green and
yellow curves are similar in shape while the red
one starts falling down at 83 Hz
drawing a deeper and wider dip, for sure more audible. My
conclusion is that is better to keep the volume of V_{1 }never
less than 2xV_{2} , as it can cause a more evident
typical DCR dip; even using V_{1 }> 2xV_{2}_{
}(inside a 5% increase) doesn't create problems but just
a deeper F_{2 }in comparison to the
ideal 2:1 response.

To calculate the ports
dimensions, knowing the chambers volumes and the desired F_{2
}, just use the volume of the larger chamber (V_{1})
that tunes at F_{1 }; remember
that F_{2 }= 1.126 x F_{1} therefore F_{1
}= F_{2 }/ 1.126. Then use the founded port
dimension for all the three ports, that are all the same. Another issue I investigate
is the position of P_{2} : should the tube be inside
V_{1} or V_{2} ? I measured the impedance of the DCR
once with the P_{2} tube inside V_{1}
and after with it in the middle of the two
chambers: as you can see below, F_{2} is the same
for both the curves, meaning that the tube position is irrelevant.
However the P_{2 }tube inside V_{1 }phase_{ }(red)
shows more spikes, quite for sure caused by it's closer distance
to the P_{1 }tube, therefore is better to place it in the
middle of the two chambers.

As you have seen, the design
of a DCR system is pretty easy, just pay attention at the wood cut to
get the right net volume of the two chambers. Now it's time to take a
look at Auri.