Fun with Foam Core
I got a crazy idea for a sat + sub build with the W5-2143 and the ARGSUB840. I know this combination sounds pretty good with the W5-2143 in 12 liter ported boxes tuned to around 55 Hz. The major problem is the limited Xmax of the W5-2143 tuned this low in a ported box. I also know that the W5-2143 is loosing a lot of its mid-range magic if crossed as high as 320 Hz. So how low can I cross it without hitting Xmax too soon? It would also be nice to use a sealed box to limit excursion and try to get it to roll off naturally without the need for a high-pass filter.
The use of foam core to build speaker prototypes is well documented by xrk971 at DiyAudio. I got a stack of 100 x 70 cm foam core sheets and I am going to use one of them to quickly measure the W5-2143 in sealed boxes of different volumes.
I use a 4 mm sheet in this case with two layers forming a baffle.
I glue the two pieces together with hot glue.
The W5-2143 was then fixated in the baffle with nuts, bolts and big washers.
I folded the larger piece of the sheet into a 20 x 20 x 70 cm tube and taped it together. Each 10 cm of this tube now holds 4 liters of air.
The baffle was then taped to one end of the tube, the element sealed to the baffle with even more tape and each 10 cm segment of the tube marked up. It was now just a matter of shopping the tube to a proper length and tape a bottom to it. Make measurements, take 10 cm off, mount the bottom, rinse and repeat.
This is what it looks like when your done...
I was mainly interested in near field measurements so I took them in the garage with the tube on my workbench. I also took some far field measurements but those figures have to be taken with a grain of salt given the lousy conditions. It is why I applied heavy smoothing on the curves. It doesn't really matter since I am only interested in the bigger picture at this stage. First the near field measurements made in 16, 12, 8 and 4 liters (ignore absolute SPL since I did not calibrate with an SPL meter).
Not much difference between 16 to 8 liters. The biggest difference is between 8 and 4 liters. Next up was near field measurements of the ARGSUB840 with its original plate amp and enclosure of 14 liters.
Measurements was taken with the built in crossover at 50 (min), 100 (mid) and 150 (max) Hz. The gain is kept the same between measurements. It is a bit funny how SPL changes with the crossover frequency (ignore absolute SPL since I did not calibrate with an SPL meter).
Next up is the gated far field measurement of the W5-2143. A notch filter is probably needed for the peak around 10 kHz but this is on axis and I know it is not as bad a bit off axis. Also keep in mind that this is in a 4 mm foam core box without any bracing or stuffing.
My next step was to export these measurements from REW and import them into VituixCAD. There I first used the Merger tool to merge the near and far field measurements for the W5-2143. Here I compare the response from the sub with the response from the 8 liter box ...
... vs. the 4 liter box.
They look pretty similar and it looks doable to get a good integration within the range of the builtin crossover and the natural roll off in the sealed box. I now turned to another simulation software, Basta!
Here I made simulations of the ARGSUB840 and the W5-2143. The plate amp is rated as 80 W. It equals around 21 V max and that is what I used in a 14 liter sealed box for the simulation of the sub. I also applied a second order low-pass filter to the sub in Basta!
Next up was the simulation of the W5-2143 in a 8 liter sealed box (blue curve). I had to apply 14 V or 28 W and tune the crossover to 80 Hz to get a flat response (black curve is the summed response). But what I was looking for in this simulation was cone excursion. It is 5 mm at 14 V. Not so good...
Next up was the same simulation but in a 4 liter sealed box (purple curve). Crossover had to be adjusted to 100 Hz to get a flat response and it actually looks a bit better at 13 V with a cone excursion of less than 3 mm. But F3 is up quite a bit compared to 8 liter.
This was simulated in a heavily stuffed box for the W5-2143. Sealed boxes around 4 to 6 liters would keep them from Xmax all the way up to where the sub is running full power. It should also be enough with a 40 W amplifier driving the pair of W5-2143. This actually looks like a doable build. :)
