Update Maggie MG12 with sub and new X-over

[marnixmorsink]

Hoi from the Netherlands,

I'm new in this community and have two questions, from which I hope somebody can provide my with an answer:

I’m busy at the moment to update my Maggies MG12. This update is a result of my recently build OTL (Tim Mellows design) and the possession of my two Maggies MG12. The idea is to connect the Maggies with a small subwoofer to enforce bass. After a lot of reading the sub will going to be a sealed box with a relatively high Qts to be able to follow the fast Maggies. Although it is best fit for a ported housing, I made the choice for the Audio Technology C-quenze 23SD speaker, with a lot of TS-parameters suitable for my goal. But this thread is not about the sub…..

For reason of the combination with the sub I also decided to update the X-over, at least with better components. Basically the 2-way X-over will be transformed in a 3-way (2.5 way?) system. The goal of the C-quenze will be to support the MG12 bass section. So the MG12 bass which will keep the full bass input. The sub acts as an extra supporting, down firing bass section.

I took a lot of time in several calculators and simulation programs to find a good design. In the Passive Crossover Designer of Jeff Bagby I started to bring in the 2-way original MG12 X-over, both in “in phase” configuration as in (tweeter) out-phase configuration. Although my Maggies don’t sound worse, the simulation was far from optimal. I noticed in the simulation a large overall SPL dip at 600Hz (in phase configuration) or a worse phase behavior in the out-of-phase configuration.

In phase….. and

out-of-phase…

Because I love 6dB x-overs, I tried a simulation with only the original coil in series with the woofer (also without the parallel 50uF capacitor) and the existing 33uF capacitor for the tweeter. Much improvement made in simulation mode. Nearly flat SPL response and very nice phase behavior. Only a slight dip 450Hz.

After several simulation runs I noticed a full flat, ideal frequency response with a single coil of 2.2mH for the woofer and the original 33mF capacitor. Low X-over was now 550Hz.
Because foil coils are not cheap and I had two Mundorfs of 1.8 mH at home I used them at a X-over frequency of 705 Hz with a small raise of 1dB at the X-over point.

After this I combined an extra single Mundorf feron T200 transformer coil of 5.66mH for the C-quenze at 225Hz X-over in phase with the MG12 bass section. By that a 6dB SPL raise starting at about 2500Hz is obtained, compensating nicely the roll of of the C-quenze in a sealed enclosure. So again a 6dB X-over configuration with a nearly perfect phase behavior. Together with the substantial higher efficiency of the C-quenze (91 dB against 86 dB for the Maggie) a large “baffle step” compensation is obtained for the down firing subwoofer.

So, after a long introduction, finally my two questions:

First, can somebody tell me if high and bass section of the original MG12 are in phase or out-of-phase?

And secondly; has somebody any clue why Magnepan brought in the 12 dB configuration (so with the 50uF capacitor parallel to the woofer) for the bass section, although all simulation programs provide a far better phase behavior in 6dB X-over mode. Could it be that the bass section of the Maggies also have a kind of break-up behavior at higher frequencies?

Thanks in advance,
Marnix