Three way crossover question

Hi all,

The attached sketches are of outboard 3 way crossovers I designed and built for use with B&W P6 speakers. I'm hoping someone can tell me what order it is or what it would be called. And does it seem a reasonable design? The speakers are connected to a system including an OPPO BDP-95, Audilble Illusions M3A and McCormack Deluxe Edition DNA 0.5 amp.

P6 XO diagram.jpg
P6 XO drawing.jpg
Wait a minuet, you designed them, but don't know what they are called? How did you design them is you know nothing about filters, don't have the .frd and .zma files, and did not simulate them? How would anyone else assess them without said files? Why would you want them outboard? Why do you think what amp you use has anything to do with it?

It is not a 3-way, it is a 2.5.
It's a 2 1/2 way crossover with nominally a third order high pass for the tweeter, second order low pass for the mid-woofer and third order low pass for the woofer. I say 'nominally' because what is important is the system response, not the crossover on its own which is pretty much meaningless.
Thanks for the reply. About ten years ago I built outboard XOs for my Sound Dynamics 300Ti speakers. First step was to make a schematic of the original XO's and then the research started. I found specs for the SD 300Ti drivers and plugged info into some online XO design aps. I then bought hi-end caps, inductors and resistors with values from the aps. I built the XO's and the speakers sounded good. I thought there was room for improvement so after more reseach and listening I replaced some of the the caps and inductors with different values. I was very happy with resluts. Some years later I bought B&W P6 speakers and decided to make outboard XO's for them. They sound pretty good but was hoping to know the the XO order was.
The speakers are 3 way and my design shows the three drivers. So what makes my design a 2 1/2 way and not a 3 way? And after doing more research trying to learn about crossovers I've changes my design. I would appreciate your comments on the attachment.
P6 XO final.jpg
Joined 2004
Paid Member
Ah, thanks. I thought 15mH seemed excessive. :)
High pass could be as simple as a capacitor, or could also have inductors, like the filter on your tweeter.
As it is drawn, it's likely your woofer and midrange have significant overlap. That is usually avoided, but can work fine depending on the drivers.
You cannot reuse/repurpose the same values if you change their position without putting everything 'in the picture'. Usually 2 and half ways exhibit two same midwoofers, each contributing in reproducing the lows and only one meets the tweeter...
A 3 way classically has a woofer (big) a midrange (little) and a tweeter- minuscule... All dimensions related to the wavelengths to be reproduced, and the crossover values according to that.
My last speaker has 4" woofer, 2" mid and a Dayton mini-8 tweeter.
You can use 1.5mH for midrange, in parallel after 22uF series for a roughly 800 Hz frequency crossover. Low pass for the tweeter should be well under 0.5 mH
  • Like
Reactions: 1 user
The bass filter has one element, a series inductor, that acts as a first order low pass.
The mid driver has two elements, a series inductor that blocks high frequencies getting to the driver, and a shunt capacitor that bypasses high frequencies around the driver. Together they act as a second order low pass filter. There are no elements in the mid driver circuit to roll off low frequencies. That would require a series capacitor and/or a shunt inductor to create a high pass filter section that would be cascaded with the existing mid driver low pass section.
The tweeter has four elements, a series resistor that acts as an attenuator, two capacitors in series blocking low frequencies and an inductor tap between the capacitors as a shunt through which low frequencies bypass the tweeter. Together these four components act as an attenuated third order high pass filter.
Last edited:
The original crossover for the P6 relies on the natural rolloff of the bass mid driver as the low pass crossover element of the midrange section. Adding an additional low pass element with a cutoff frequency ~1kHz creates a dip in the response until the tweeter comes in at ~4kHz. This will create a more than one octave-wide hole centred on ~2kHz that will be 6dB deep in the middle, compared to the original alignment.
Another point: the sensitivity of the system, which can be (misleadingly) called efficiency (but in this case and others can be assumed as similar in explaining the phenomenon), which must be in the 'low' region (85 dB?!). The 1.5 mH suggests that the upper mw is heavily equalized in order to meet the level of the lower one- 8mH is a value for subwoofers! - so the baffle step compensation happens gradually and spreaded between the two midwoofers.