CM5 B&W upgrade crossover

You always start by doing the homework. Service manual here:
B&W Group North America Service & Support - Home

More details here:
Bowers & Wilkins CM5 loudspeaker | Stereophile.com

Minimal crossover is 1.4mH aircoil on 6.5" Kevlar bass, 4.7uF Mundorf capacitor to 1" metal tweeter. Possibly with 0.5R level adjust in series.

I expect you could do something a bit more sophisticated. Often you notch a 6.5" bass at 5kHz to take out some breakup. A higher order tweeter would be feasable too.

Depends how interested you are, and how willing to learn about these things.
 
From what I recall of my own measurements of the air core in its predecessor (CM1) that inductor gets rather lossy above ~1kHz due to proximity effect. You might be able to ameliorate the 1k-5k apparent 'suck out' in the FR plot with an air core using multiple strands of narrower wire, preserving the DCR.
 
I am interested as well, because these are my second set speakers that I could tinker with. All depends on whether it is worth changing. I often heard/read that b&w uses proper xovers, at least parts.
It does look like a very simple crossover to begin with.

Screenshot_20200603_130207.jpg
 
Last edited:
The drivers in this speaker will be engineered to work effectively with a minimilist crossover by incorporting their own mechanical roll-off slopes.

The result, as described in the review, is excellent detail resolution and transparency.

I would leave the crossover alone as it is a distinct design feature.
 
There's definitely a downside to a single series cap for the tweeter, and that's tweeter excursion. Lynn Olson writes : The Ariel, Part II

All direct-radiators (tweeters, mids, woofers, etc.) have diaphragm excursion that increases with decreasing frequency at a rate of 12dB/octave. (Drivers are constant-acceleration devices.) To prevent the excursion from actually increasing below crossover, you must use at least a 2nd-order (12dB/Octave) crossover ... even this rate simply keeps the excursion constant below the crossover point.
 
It may be that the diaphragm excursion problem is avoided in the CM5 because the crossover (natural driver slopes in combination with electrical slopes) is effectively second-order Linkwitz-Riley.

In combination with a crossover frequency of 4kHz, this would result in the tweeter being at least 24dB down at its fundamental frequency of resonance.

There's an interesting discussion on 1st order tweeters here:

1st order tweeters
 
Useful thread, thanks. So what I'm getting here is that even though there's no electrical correction shown for the tweeter impedance rising at resonance, potentially there's an acoustic fix in that the tweeter may have a 'Nautilus tube' on its rear which flattens the impedance curve?