Thank you for that link.
I have used Hi-Fi engine and Vinyl engine previously, but not that site.
So the whole panel radiates bass, but narrower strips to the side radiate progressive high frequencies as well as bass to prevent beaming. However as you have found out it still beams.
The only solution is the late Peter Walker's of Quad. The secret is to build circular delay lines into the membrane, so the membrane becomes a spherical radiator. His system works.
Anyhow, the socket you are referring to when the passive subs are connected cuts out the bass to the panels at frequencies below 100 Hz and sends them to the subs.
They claim they you will get 8 db increased output. With a 100 Hz crossover that is just plausible. So you will get a modest increase in volume.
So, do not use the socket on the speaker to connect your sub.
Connect the sub to the LFE output.
Now set the left and right mains to small in the speaker set up of the receiver. Set the crossover to 100 Hz.
With the crossover that high, you really should use two subs and set each one beside a panel.
That will give you the best result.
By the way the sub they marketed for these speakers, was somewhat like the Gradient sub. The Gradient was an open backed di-pole. The Acoustat sub, had multiple front and rear firing woofers. I assume the front and rear speakers must have had their own sealed enclosure space. The front and rear drivers would have to have been wired out of phase, otherwise the rear drivers would have been out of phase with the rear radiation of the membrane. So these subs would have been di-poles like the Gradients.
If you want to copy this to obtain optimal integration it would require a DIY solution.
