Layout help for series crossover

[rabbitz]

I used PCD 5.x and SCD 1.x. For these to work you need frd and zma files.

 

[mV8]

I never noticed the the original post had no cap. This is the one from 2003 that I started with.

View attachment 1300338

This crossover scheme with a capacitor is quite alright. It already gives some design flexibility and, in my opinion, is such a minimum for a sensibly designed series crossover. Of course, you can't do it properly without measuring the speakers and simulating the crossover. For simulation I recommend VituxCAD.

 

[mtidge]

As an exercise I used this XO design on a simulation I did for a new member. This was just a PoC using downloaded Dayton factory data, I was surprised at the quality of the results after adjusting the values. I have no idea how it would sound, I'm guessing both drivers are pushed into distortion, but the frequency response looks nice. It was in the recent thread.

https://www.diyaudio.com/community/threads/first-3-way-crossover.412155/page-3#post-7673003

 

[wolf_teeth]

I think you need a cap in series with the tweeter too, try 15uF.

 

[perrymarshall]

This thread is relevant to the current discussion:
https://www.diyaudio.com/community/...over-vs-parallel-for-subs-open-baffle.404680/

 

[rabbitz]

Going by past experience you could get a better result starting off with L1 at 0.12mH and L2 > 1mH. This would roll off the tweeter better but I do not know the drivers and target.

 

[mtidge]

I think you need a cap in series with the tweeter too, try 15uF.

There is, C1 at 27uF

@rabbitz This was just an exercise to play with a serial XO, I have no plans of building the design.

 

[wolf_teeth]

No, I mean immediately adjacent in series, on the top line as far as position goes. The tweeter is rolling off too gradual and may benefit from upping the order. It will also help align the acoustic phase better.

 

[mtidge]

Got it, the change pushed the XO point up to 3500. I did some more fooling around with serial XO circuits and I think they are of very limited use. The woofer and tweeter need to have the same impedance for any chance of success. Maybe using a cone tweeter that can play very low or a small full range for the high frequency might work. I even tried modeling a second order serial with some of my working designs that use modern drivers and all I could end up with was a hot mess. There seems to be to much interaction between the low pass and high pass circuits. I will stick to parallel XO circuits.

 

[rabbitz]

Different impedance for tweeter and woofer is no problem and a lot of mine have 4 ohm tweeter with 8 ohm woofer crossing over around 2.4 kHz. Driver selection makes a big difference with this type of crossover as there is not the flexibility of a parallel crossover.

 

[AllenB]

as there is not the flexibility of a parallel crossover.

Well, there is that flexibility, it's just a different set of compromises. Besides, series does have better adaptability to different impedances for those times when redesigning is not convenient.

 

[mtidge]

Let me put it this way, serial XO's are beyond my skill level.