Real measurements of B&W CM10 s2

[Arc78]

I recently read in a Swedish forum that a majority of the new b&w speakers measure strangely due to a 6db/octave crossover. This is all new to me.
I was wondering if this strange response also was valid for my cm10s2 so far I have been unable to read any good measurements of this speaker but I suspect this is one reason it sounds little light and bright.
Anyone seen any measurements of this speaker cm10s2?

Or someone can give me a easy list of things I need to be able do this from a PC and a beginners guide how to make this measurements?
I borrow this measurements he done on 686 s2 as illustration of what I am looking into
I think this is quite interesting

 

[Ageve]

Here’s a CM10 S2 measurement from audio.com.pl:

BOWERS & WILKINS CM10 S2

Wprowadzając nowe serie czy poszczególne konstrukcje, choćby tylko lekko zmodyfikowane, ich producenci podkreślają znaczenie zmian i udoskonaleń ...

audio.com.pl

Red = on-axis
Brown = 15 deg horizontal
Yellow = 30 deg horizontal
Blue/green: +/- 7 deg vertical
Black: With grille

Google Translate:

”The wide range of cooperation of the speakers and the unevenness of their individual characteristics, especially the midrange characteristics in this range, as well as the phase relations changing with the angle in the vertical plane, cause a small "mess", which, however, does not get completely out of control and from the framework of the general balance - the characteristics in the range of 200 Hz - 20 kHz, from the main axis, we can fit in the path of +/- 3 dB, and from the axis of +/- 7 degrees - in the path of +/- 4 dB.”

Edit: 40dB vertical scale makes it look a bit worse compared to the standard 50dB range.

 

[Arc78]

Thank you!!
do you think Dirac might have a chance to correct part of this or how bad would you translate it?
There is a real downturn in the area of 1050 to 1300 why is that then sharp up in really high tones.. I guess I have allot more to learn about this.
I would say that this speaker is light very little bas and a quite sharp very detailed top.
Why you think that is? To get the other frequencies in good volume I heart my ears.
Room is not treated and is a square of almost 4x4 m
And I have not yet got any Dirac or similar

 

[Marin]

Hvala!!
misliš li da bi Dirac mogao imati priliku ispraviti dio ovoga ili koliko bi loše preveo?
Postoji pravi pad u području od 1050 do 1300 zašto je to onda oštrije u jako visokim tonovima.. Pretpostavljam da moram još puno naučiti o ovome.
Rekao bih da je ovaj zvučnik lagan, vrlo malo bas i prilično oštar vrlo detaljan vrh.
Zašto mislite da je to tako? Kako bih dnk a parametric EQ would be very helpful for this speaker, Dirac RMS even better.

I'm always surprised when I see B&W speaker measurements, especially after seeing installations that have both Having visited the Bowers & Wilkins Research Center, in Steyning, nestled beneath England's South Downs, and been impressed by the depth of the company's engineering expertise, I am always puzzled by the fact that B&W speakers don't measure as well as they could.
Yes parametric EQ would be very helpful for this speaker, Dirac RMS even better.

 

[Juhazi]

Nearfield measurements in post #1 has obviously both drivers playing during each measurement. This causes huge interference dips and peaks that we see mainly in woofer's upper band. Individual responses should be measured with other drivers decoupled.

In general yes, most B&Ws of this millenium measure awfully bad. Nautilus was well done, but then they lost Lawrence Dickie. Shallow xo is one reason, egg tweeter and mid is other. Poor crossover design...

 

[Arc78]

Well the speaker itself seem impressive build quality and elements, the older B&W absolutely fantastic. Were is the cross overfailure? Can bi amping with external crossover help? Or is it mid and tweeter that causes this problem?

 

[Arc78]

I so much want this speaker to live up to its potential. Not just because i invested 2000$ in the pair of cm10 but also have the whole setup with cm8 och cm center but also because i think this speaker has such a potential.

 

[Arc78]

I’m no super audiofile but I hear that something is not correct with the heights treble something sharp.
I would like to say it like when the volume is perfect for all the rest of the audio like mid and bass a part of the higher spectrum is too high up in volume or distortion wise.

 

[Speedskater]

Nearfield measurements in post #1 has obviously both drivers playing during each measurement. This causes huge interference dips and peaks that we see mainly in woofer's upper band. Individual responses should be measured with other drivers decoupled.

Disconnecting other drivers with a passive crossover, will change the response of the measured driver.

 

[Ageve]

Nearfield measurements in post #1 has obviously both drivers playing during each measurement. This causes huge interference dips and peaks that we see mainly in woofer's upper band. Individual responses should be measured with other drivers decoupled.

You probably misunderstood the measurements above. There were no interferences in the near-field. The small "window" with an arrow was just to show how the woofer near-field response affects the far-field response. In other words, it explains the uneven response that can be seen in NRC-measurements of 685 S2, and Stereophile measurements of 683 S2.

Bowers & Wilkins 686 S2 Spinorama measurements (CTA-2034)

Here are measurements of the Bowers & Wilkins 686 S2 bookshelf speaker. It was released in 2014, and discontinued in 2018, and it's two generations behind the 607 S2 measured by Amir. I Think the MSRP was 399.99 USD / pair. My measurements are quasi-anechoic, with near-field port+woofer...

www.audiosciencereview.com

When doing quasi-anechoic measurements, you have to measure very close to the woofer to get an accurate result up to around 800 Hz (usually 5-10 mm depending on woofer size). At such a short distance, there won't be any interference from the tweeter (the woofer sound dominates).

I measured the response at a slightly longer distance for the image posted above, 3 cm (5-10mm while ideal for quasi-anechoic bass response, may look worse than it really is at higher frequencies). Remember that these near-field measurements are not meant to show an accurate response, but to show port and/or woofer resonances, unevenness, crossover implementation, cone breakup etc.

I still have the speakers, so I brought one of them up to my living room, and measured at 1 cm and 10 cm, with the tweeter disconnected (bi-wire terminals). It turned out to be sample 2 from the original measurements.

As you can see, the tweeter has no effect on the woofer near-field response, and when moving out to 10 cm, it looks even worse (higher peaks compared to the overall level):

 

[Arc78]

i see what you wrote even with tweeter turned off we have an very uneven mid.
Why do you think that is?
What you think about what I am talking about with sound character? It’s very different from the really really old 601 s1 also with Kevlar woofer. They are neither harsh or low on mid/bass for their size

 

[Arc78]

Do you think Dirac is best or only way to help this speaker to stop lifting some frequencies and lower others? What about distortion? What can be the reson for this sound problem? The speaker is not so bad at all just this is not optimal since i can’t get an even volume so to speak. I would like to hear bass and mid at their best without getting soar ears

 

[Ageve]

i see what you wrote even with tweeter turned off we have an very uneven mid.
Why do you think that is?

The peaks are probably caused by cone breakup. The kevlar cone doesn't work well at high frequencies, but if it had a proper crossover (at least 12dB/octave) it would look much better.

 

[Arc78]

The peaks are probably caused by cone breakup. The kevlar cone doesn't work well at high frequencies, but if it had a proper crossover (at least 12dB/octave) it would look much better.

How I solve this error and how come they not see this tthem selfs? I mean they been building top notch speakers for decades. Maybe just need to rebuild the filter board after a drawing to prevent mid to go to high?