• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Passive crossover design advice needed for a 2-way

Aijan

Member
Joined
Jul 13, 2021
Messages
26
Likes
66
This is my first time designing a passive crossover, and I'd appreciate your advice on it. I'm designing a 2-way bookshelf speaker with a 6.5" pro audio mid-woofer and a compression driver + horn combo.

The on-axis in-room measured response of the CD+horn is below (1/12 octave smoothing):

cd_horn_raw_response.png


I used a first order crossover @ 8000 Hz for the CD to get a relatively flat frequency response so that I can get away with a single notch filter centered at 1000 Hz with Q value of 1. There's also an L-pad to reduce the CD level. The crossover for the woofer is a simple second-order at 1100 Hz. The actual acoustic crossover point is around 1850 Hz. The power response and directivity index look relatively smooth to me for a passive crossover.

The woofer rolls off early because there is no port in the simulation. It should be flat down to 100 Hz or so with the addition of a port. Subwoofers will be used below that point. Sensitivity is around 89 db/2.83V.

crossover.png


I'm a bit concerned about the impedance plot though. Both drivers are nominal 8 ohm drivers. The black trace is the impedance after the crossover applied. The red and green lines are the raw impedances of the woofer and CD respectively (please ignore the wiggles; they're artifacts of the automatic conversion of impedance plots from driver spec PDFs).

Finally, the purple trace is the EPDR, and it looks a bit low at around 3 ohms for most of the mid-range. Is this normal?
impedance.png

Please let me know what you think about the crossover. Any advice will be appreciated.
 
Your impedance is low because your lpad is before the x-over. You want the attenuation after the xover so the resistors don't see the full power of the amp.
 
Your impedance is low because your lpad is before the x-over. You want the attenuation after the xover so the resistors don't see the full power of the amp.
Rookie mistake. I've changed the placement of the L-pad as you suggested, and the impedance plot looks much better. I also had to adjust the values of the capacitor and resistors a bit. Should be a relatively easy load for an amp now. Thanks!

revised_crossover.png
 
Last edited:
Your impedance is low because your lpad is before the x-over. You want the attenuation after the xover so the resistors don't see the full power of the amp.
GOOD CATCH!!
 
I have a question - Why the notch filter at 1K? What purpose does it serve?

(I have never designed an XO, only followed plans, none of which have had or discussed a notch.)
 
Found this, read up, answered own question.


I think my question arose from the thought of needing a notch at 1K when no peak was apparent. Does it's inclusion generate anti-phase harmonics to damp the wiggle at harmonics of the notch?

Passive crossover tuning is an amazing thing.
 
Excellent, thank you very much. I have a set of Grund Audio GT870 which are very much like yours. I have them crossed at between 120 and 250 hz with the subs within a foot of them. Still figuring out what sounds best. Since they're 8" and rated to 80Hz, they shine much like you describe yours once the bottom end is supported by subs.

Very nice build.

Next time I have them open I'll get a schematic of the XO.
 
Back
Top Bottom