Two way, using Accuton C95-T6 & C23-6

[acoustic_design]

Two way, using Accuton C95-T6 & C23-6

This is my next project:
2 way monitor loudspeaker using Accuton ceramic drivers. This time I decided to use electronic crossover ( Linkwitz-Riley, 24dB/oct) and bi-amplification. The question is which is the optimum crossover frequency between C95-T6 and C23-6.
If it's too low, there is harmonic distortion issue with the tweeter, but the dispersion and resulting off axis frequency responce of the loudspeaker will be better.
If it's too high, the roll off at break-up frequency (3,5 - 5,5kHz) is not enough to properly tame the resonance of midbass driver.
I suppose 1,9kHz is OK. Am I right?
Any other suggestions?

 

[LaMa]

Everything between 1.5 -2 kHz (beaming of the c2-95) will do. I never used the c2-23 though the specs (factory distortion plots) suggest low distortion from ~1.3 kHz upwards and fs of 350 Hz. In short this will be quite easy execution, do not forget to notch the resonance at 3.8 kHz. Unfortunate going active will make this a bit more troublesome.

In my experience the C2-95 is easily integrated when tweeter allows for cross <2 kHz. For your reference Kharma crosses ~1.7 kHz.

 

[acoustic_design]

According measurments of John Krutke (http://www.zaphaudio.com/tweetermishmash/hd.html) C23 have high third harmonic distortion below 2,5kHz, despite the fact the manufacturer"s measurments don't indicate it. This may lead to sterile and cold sounding in that region, and all my observatios lie at the roof of this speculation.
If I go with 1650Hz will I avoid the need of notch filter? The filtered signal will be at least 24dB down at break up resonance of the cone or it wont be filtered enough to avoid deleterious effects it is going to cause.

 

[acoustic_design]

Any ideas for schemes of good active notch filters with medium Q factor appropriate for this design?
Thanks.

 

[LaMa]

In my opinion 24 dB down is not enough, it's still audible. Take a look overhere http://www.klippel-listeningtest.de/lt/

 

[acoustic_design]

You are absolutely right. Even 36dB down is not enough!! So I must definately use notch filter. But the problem is I don't have ANY experience with active notch filters...
I found the following sheme http://www.national.com/ms/LB/LB-5.pdf but is it appropriate for this task??

 

[Taco]

I cannot help you with the active x-over, is it not possible to do a passive filter. It's only a 2-way.

 

[forr]

Maybe this could help :
http://www.diyaudio.com/forums/showthread.php?s=&threadid=6655&highlight=

 

[acoustic_design]

I cannot help you with the active x-over, is it not possible to do a passive filter. It's only a 2-way.

Some examples of same drive units and implementation of passive crossover:

Talon audio Hawk - http://talonaudio.com/products/productframe.html

Marten design Duke -
http://www.marten.se/base.html

Avalon mixing monitor -
http://www.avalonacoustics.com/pro_frameset.html

Audio Acoustics Sapphire Ti-C SE -
http://audioacoustics.co.uk/models/audioacousticssa.html

Margules audio Orpheus M -
http://www.margules.com/home.htm

Metal sound design MoonII -
http://metalsounddesign.com/down/moon/moon8.pdf

Most of these loudspeakers have crossover frequency of 3000Hz (!!!) and passive crossover 12dB/oct

All of the reviews are positive. But the break up of C95 is still there although low in level cause the notch filter at 4kHz.

 

[Taco]

You can suppress the breakup with a notch filter. Also I totally agree that you have te cross <2000 kHz.

 

[acoustic_design]

I desided 1.7k would be optimum. I'm going to use Vincent SP-991Plus as an amplification for C95. C23 will be powered by 16W single ended class A solid state amp for cleanest treble possible.
Enclosure will look like this: