El Cuerno

[erjee]

Waveguides and multiple subwoofers are very popular these days, thanks to Earl Geddes for sharing his knowledge with the community. So after playing around with fullrange dipoles and dipole satellites with 4 subwoofers I decided, back in October 2010, to build an active cardioid / waveguide system with four subwoofers: El Cuerno. For the cardioid part I was inspired by the work of MKB (Adventures in cardioid).

Signal processing is done on an old computer running Linux/BruteFIR for filtering and MPD (Music Player Daemon) for music playback. El Cuerno is actually a complete system, but I believe that the rightful place on this forum is in 'multiway speakers' because the DSP and amplifiers eventually ‘serve’ the multiple electro-acoustical transducers (obviously the point of view of a loudspeaker designer )

Satellites
The drivers used are the 18Sound XT1086 waveguide with a NSD1095N compression driver and a B&C 10PS26 midwoofer. The mid plays between 200 and 1000 Hz and the waveguide above that. The satellites are active, mid and high are each driven by a Hypex UcD180.



The baffles are made of 20 mm thick pinewood, 32 cm wide and 90 cm high. The waveguides are damped with 25 PPI open cell foam (not shown here).



Using some containerboard and foam it is possible to create a supercardioid radiation pattern for the mid. I’m still looking for a solution to upgrade the looks





The gated directivity sonogram of the horizontal plane (1/12 octave smoothing) shows a very smooth transition between mid and high. The directivity is fairly CD, a slight increase in directivity towards higher frequencies can be observed. The filter is currently a very steep FIR filter, it is expected that vertical directivity measurements will show less lobing than a standard LR4 or LR8.
The satellites are toed-in at an angle of 45 degrees, so the listening axis will be approximately 20 degrees off-axis.

Subwoofers
The four subwoofers each have a Peerless SLS12 in a 25 liter closed cabinet. Two of them play in a corner-woofer stereo configuration between 120 and 200 Hz. The range of 20 to 120 Hz is mixed down to mono and sent to all four subwoofers. A room-correction curve is based on a spatial average.



Listening position
The satellites are placed along the long wall of the room (3.8 x 9 m).



And this is how it sounds at the listening position (spatial average of left, right and left+right combined).



El Cuerno has more precise imaging than the dipoles and they can play much louder (> 105 dB at listening position). One reason is the higher than normal efficiency (less thermal compression, lower excursion). Another, perhaps more important reason, is that the narrow directivity provides a much higher direct/indirect ratio of sound at my listening position compared to the dipoles or normal-directivity speakers. There is a tendency to play 5 to 10 dB louder, which I attribute to the higher ratio of direct sound and which is only apparent after measurement with a SPL meter (or an attempt to have a conversation ).

[markus76]

Nice and thanks for sharing. Did you get the directivity plots from measuring outside?

[erjee]

The measurement was done inside, at a distance of 1 meter, so gating is applied. Outside would be better, maybe I can take them to Keyser (again )

[markus76]

I think for such a design you should measure at 2m. The max. distance I get inside is 1.3m.

[erjee]

I agree and I will do that when I have time But I would not expect a major difference, only for frequencies below 500 Hz or so. From the spatial average at the listening position one could argue that the range above 500 Hz should be attenuated by 1 or 2 dB.

[keyser]

Erjee, nice of you to show your system here as well. It was your system that inspired me to design my latest speakers.

Originally Posted by erjee

The measurement was done inside, at a distance of 1 meter, so gating is applied. Outside would be better, maybe I can take them to Keyser (again )

You are most welcome!

[erjee]

Thanks, I am planning to redo the containerboard with something more substantial and get the response right using gated measurements and a spatial average at the listening position. Then it would be nice to confirm te response with outdoor measurement, hopefully it does not rain in your 'anechoic chamber'

I think we were both surprised by the sound quality and dynamics of the waveguides.

[gedlee]

Very respectable work.

[454Casull]

I see you like your bass!

[ttan98]

erjee,

Nice work, at the same time the system costs a fraction of an equivalent commercial system. Should sound very good even without listening to it.