CY have made an active version of his B&W 802D with the DCN28 Digital Pre-amplifier / Digital Crossover.
The passive crossover have simply been bypassed to power the individual drivers with it's own power amplifier channel.

At first - to gain confidence in setting up the active loudspeaker CY started on an inexpensive DIY loudspeaker to avoid destroying the expensive drivers in the B&W 802D speakers. On the left you can see the initial testing on a DIY loudspeaker box.

System set-up:

• System AC power generator: PurePower 2000
• CD Transport: Esoteric K-03
• External Clock: Esoteric G-03x
• Pre-amplifier / Digital Crossover: DCN28
• High Frquency (4300 Hz+) Amp: SIT-2 (10w x2. Single transistor. Zero negative feedback. Poor dampling factor)
• Higher-Mid Frequency (600-4300 Hz) Amp: Ayre V-1x ( 200w x2. Zero negative feedback)
• Lower-Mid Frequency (200-600 Hz) Amp: Red Dragon Leviathan (500w mono. ICE power. Dampling factor: 2000)
• Woofer Frequency (<200 Hz) Amp: Red Dragon M1000 MkII (550w mono. ICE power. Dampling factor: 1000)
• Speaker: B&W802D (converted to 4-ways without crossover network. Another two pair of terminals added outside the speaker)

Reports from CY:

"Clipping Noise:
After a few repeated testing, the problem of clipping like noise from the "CANTATE DOMINO" from "Proprius" track #13 have nothing to do with the DCN28 master volume setting. Previously I reported that with the Ayre PreAmp providing 8 dB gain to compensate the low gain of SIT-2 and the gain volume of DCN28 can set 8 softer for the channel for Tweeter amp SIT-2 and the noise reduced for the 18 sec occurrence. After prolong repeated testing, the minor clipping noise during 18 sec remain the same regardless with or without external Ayre Preamp or whether DCN volume set at -7 dB or -15 dB for the tweeter. So you are right, this noise is not from the Op Amp clipping."

"The 4-ways active digital crossover system takes full advantage of the principle of multi-drivers speaker system without all the problems and distortion caused by the passive crossover between the amps and the drivers. The improvement exceeded my expectation."

"Music is pure and direct. But at times it reveals the over-load distortion of some recordings. But to me this is what Hi-Fi is all about."

"Currently the crossover is at 200, 600 and 4300 Hz. The main consideration is to target for higher resolution for human voice, as well as limited power and control of the SIT-2 (reduce the tweeter amp operating range). This is just an initial trial set-up, I will fine-tune it later."

"Although there is still a lot of fine-tuning required for my system, comparing to 3-ways active, 4-ways active produce significantly higher resolution on male vocal voices. The male vocal have never been so real, warm and have so much details and resolution.
Yes! The sound is very warm too. Partly because voice is very real and driven directly by the 500w high damping factor class-D mono amps without any passive crossover network. It is the quality and not quantity that really count. And each driver only needs to operate and do their best in a small frequency band."

"I am glad to receive e-mail from you. I have not made any change of the DCN28 set-up recently. Because I am spending my time to appreciate the quality sound coming out from the system - and I also know that there is many improvements can be done when I have the time to do it."

The total simulated frequency response of the active B&W 802D
Comment: As this is an early simulated response of the active version of the B&W 802D based upon the measurements of the individual drivers in around 1m - one have to make more measurements and adjustments to get the right tonal balance in the listening position and evaluate what the dip around 200Hz really is caused by. The 100-200Hz dip is often caused by the distance from loudspeaker front (bass driver) to wall behind loudspeaker or the distance to a side wall.