So much money for such poor performance! They look coolThis is a review and detailed measurements of Bang & Olufsen (B&O) Beolab 20 speaker. It was kindly drop shipped to me by a member and costs US $15,000.
View attachment 272656
You expect out of box industrial design from B&O and that is what they deliver. The Beolab 20 is an active, 3-way, sealed design. The woofer and mid-range are behind the cone cloth. The tweeter is unusual in the way it fires up into an acoustic lens of sorts:
View attachment 272657
The back side is made out of some kind of rubber (?) with a panel that removes to expose the inputs, power, etc.:
View attachment 272658
I was disappointed in the look, feel and component selections there. As you may be able to see, there is dust on the lower shelf. The RCA connectors are the ultra-cheap looking tin coated ones instead of gold. The dip switches work but again, don't give the feeling of luxury. On the positive from there is support for wireless streaming in the form of WISA (but sadly not bluetooth). And digital input. This requires synchronization between the two speaker which they are doing with a toslink out to the other speaker. And then you would need some kind of volume control.
The measurements you are about to see are generated by the Klippel Near-field Scanner (NFS). Critical measurements are made anechoic with others being in-room. Temperature of the measurement room was rather low at 57 degrees F. Reference axis was that of the tweeter. Speaker was tested with he switches as you see with the Free-space selection. Even though I have two speakers, listening tests per research was performed using a single one.
Beolab 20 Measurements
Let's start as usual with our anechoic, CEA-2034 standard compliant frequency response measurements:
View attachment 272659
Wow, I don't think I have measured a speaker that goes essentially flat down 20 Hz and especially one in this compact size! That is the good news. The bad news is that there is fair amount of variations, resonances and disturbances along the way. The plurality of these will make it difficult to develop an EQ for this speaker. We can see clues to some of these issues in our near-field measurements of each driver:
View attachment 272660
Note that I could not visually located the mid-range so that response may be approximate. But the woofer is not and we can see resonant peaks. The mid-range's upper spectrum also has a lot of ups and downs. And there is peaking in the tweeter as well. I am assuming DSP is used in this speaker. If so, surprising that they have not attempted to quell these variations with better crossovers and filtering.
Back to CEA-2034, here is our early window response which again, doesn't look all that nice:
View attachment 272661
So naturally our combined predicted in-room response has issues:
View attachment 272662
There is large variation in dispersion of the speaker:
View attachment 272663
I was surprised and disappointed to see the super narrowing of the tweeter response above 8 to 10 kHz.
View attachment 272664
Vertically we have a very messy situation:
View attachment 272665
So much going on outside of our solid areas indicating secondary diffraction and such.
Distortion is not bad at 86 dBSPL but gets worse at an odd position in response:
View attachment 272666
View attachment 272667
I looked at woofer response and it was not the cause of the peaking around 800 Hz. Instead, it seems to be the midrange that is complaining:
View attachment 272668
It is also responsible for multiple resonances up high which we saw in near-field response of the driver.
Waterfall display shows resonances:
View attachment 272669
Finally, here is the step response for fans of that measurement:
View attachment 272670
Beolab 20 Speaker Listening Tests
I placed the speaker as you see in the intro shot in my main listening room. I had just tested a $500 speaker with the famous Stevie Ray Vaughan track, the Tin Pan Alley. I expected it to sound even better on Beolab 20. Instead, I found the experience the other way around. It sounded dull and unexciting. So I skipped that and went to my standard reference tracks of female performances. Fidelity was improved here with a bit more bass presence that I typically hear from small speakers. As I progressed through tracks, I realized that the Beolab was activating my room mode at 105 Hz so started to EQ that and made more corrections after that:
View attachment 272671
As I noted in the review, building an EQ by eye is going to be difficult and it was. I *think* I made an improvement but I am not sure it translated across all tracks. The highs needed a boost so I put in the shelving filter initially at 2 dB but had to knock that down as it was a bit bright. A less lazy approach with multiple filters would have avoided that perhaps.
The initial filter set did not have that high-pass filter. I jumped into my tracks with sub-bass (< 30 Hz) content expecting great performance. The Beolab played this range but what I heard was quite distorted and odd sounding. The above filter reduced that some but what I was hearing was still abnormal and I preferred not having that sub-bass than what I heard. We see high distortion in bass frequencies in the measurements confirming what I was hearing.
On the positive front, high-pitch notes at times resonated around the room in a way I had not heard before. They would sparkle against different points in 3-D space which was a nice effect! So that tweeter design is doing something.
Overall, the response as I noted was variable. On some tracks such as Deadmou5, the speaker sounded wonderful. But on others as I noted, it was a let down. Maybe more work with positioning and filtering helps.
B&O is known for innovating in design department when it comes to looks of the products. At the same time, they have extensive engineering and research facilities. The latter gave me hope that they could tame the restrictions that the industrial imposes on design. Alas, it doesn't seem to be the case here. I think driving the speaker to have 20 Hz response but with distortion is unwise. There are a lot of resonances and interference patterns between the three drivers which I think could have been dealt with better. Clearly what we have is nicer than some just going for looks. But as a hi-fi product, it doesn't fit the bill quite right.
I am sad to not be able to recommend the B&O Beolab 20 speaker. Performance is too uneven given the very high cost.
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/