I'm trying to use Master Q in a 3 band crossover matrix (3 group tracks). I'm using the low- and highpass filters for this (12dB/oct). Group 1 has a lowpass filter at 120, group 2 has a highpass filter at 120 and a lowpass filter at 5,45kHz. Group 3 has a highpass filter at 5,45kHz. The Q in all filters are at the default value. In a fft analyzer I see a notiable drop around the crossover frequency. i can tweak this by introducing more overlap with the bands. Is there a more accurate way to do this or am I missing the Q aspect here?

Thanks,
Rune

Is a drop after summing bandwidth on a master track?
Actually the MasterQ and other EQs are know are not built to work as a cross over but you can of cours try to do it.
Please note that low pass and high pass filters are not complementary in terms of phase and level so probably you should try to slightly overlap bands to achieve the result you are looking for.

Regards,

Hello Rune,
the problem of phase cancellation is known from crossovers for speaker cabinets. Each 12 dB slope causes phase shift of 90 degrees. At the turnover frequency, the lower band turns 90 degs is one direction, the higher band 90 degs in the other one. That means for you: Your problem should be solved by changing phase in your mid-band!

Regards,
Turntablerocker

Then I would have to use the 24dB/Oct slope on both the high and lowpass filters? Otherwise the low and the high band would be 180 degrees apart, I guess?

No, leave your low band and high band as they are and turn phase in mid band by 180 degrees. This is the way it is done in 3way/12dB per oct. crossovers for speaker cabinets. Check it out by viewing a pink noise signal on your analyzer. You shouldn't have phase cancellation at the turnover frequencies any more.