Volt radial woofer

gregm,
Actually, giving some consideration to a custom build. Am inspired by the Verity Lohengrin. Alas, I will never have $80K to invest in loudspeakers.
I like the non-square cabinets of the Lohengrin, and from an engineering perspective, understand the purpose they serve.

My idea is the Volt RV3143 (12") or RV3863 (15") woofer in the lower cabinet, coupled with Seas Exotic W8 X2-08 (8") woofer, the Volt VM752 (3") midrange dome, and Seas T29CF001 or Scan-Speak D2904/7100-03 tweeter in the upper cabinet.
Yes, I fully understand these are $$$ drivers. I can design the cabinets, and am thinking 30mm internally braced MDF with front baffle angled to maintain time alignment.
Probably have the cabinets cut/fabricated on CNC router. Definitely need assistance with crossover design.
Design goals are wide frequency response, excellent micro/macro dynamics, and relatively high efficiency. I will be using a SUMO Half Power for amplification, but prefer the load to be an easy one, so as to make maximal use of the amplifier.
Your thoughts/comments are appreciated!

Hi if it's any help to you I built a pair of speakers a couple of years ago using the RV3863, VM753 and Scanspeak D2904/710003. The cabinet was a Wilmslow Audio TL15 transmission line and the only change to it was to specify all panels (internal as well as external) in 25mm mdf. I followed their exact recommendations on line stuffing.
The sound quality is astonishingly realistic, I think mainly due to the RV3863's uncoloured lower midrange and the VM752's clarity. Distortion doesn't exceed -45dB across the range until the lowest frequencies. Bass is of course very extended but at first listeneing seems "dry" until you realise that there is no emphasis. Then an organ pedal or bass drum shakes the room! I'm a big fan of transmission lines.
I run all-Bryston electronics and am happy with the setup.
Some advice:
1. Buy Dayton Audio's Omnimic and Test System v2. You will need this to obtain the data for the crossover design.

2. Mount the drivers in the completed cabinet and measure the response of each individually to create FRD and ZMA files for each driver using the test CD provided. Take care with volume levels so as not to damage the mid and tweeter.

3. Install XSim on your PC. It's free, easy and intuitive to use and will give you hours of fun playing with different circuits and values.

4. When you are happy with the results you can order the components and build the crossovers. I used 2nd order designs for the bass low pass, the mid low pass and high pass and 3rd order for the tweeter. Don't be too ambitious and try to design 4th order circuits like PMC do. They are complicated and extremely difficult to get right. They use this design with its steep slopes to get very high power handling in commercial installations. You don't need that at home if you value your hearing.

5. Make sure you order the Volt drivers in 8 ohm spec. The Scanspeak D2904/710003 is a 4 ohm unit and is extremely sensitive so will need padding resistors. The bass and mid are perfectly matched and don't need any balancing.

6. For initial trials keep the crossover outboard. The system may measure flat but it may not sound right and you could have to make changes - especially adding resistors to balance the drivers. XSim will help with this as it is so interactive and you can see the results as you play with different values. It may be helpful to use switches to select alternative values at this stage. It took me two years of listeneing to be satisfied with the crossovers.

7. Don't be afraid to cross the bass to mid at 500-600Hz. The RV3863 is very clean up to higher than this and the VM752 needs to roll off below 500Hz if you want decent power handling. I chose 550Hz and 3.8kHz for my design.
I'm sorry if this is a bit long winded. I hope it helps.

Correction: The midrange unit which I used is the VM752, not VM753.
The VM752 matches the sensitivity of the 15" RV3863 at 94dB/w@1m

The 12" RV3143 is less sensitive and matches the VM753 at 91.5/92dB/w@1m. They both have smaller magnet assemblies than the above.