"BBC dip" does it sound like Quad ESL57?

[Electrostat]

I suppose this is my continuous quest to find some speakers for another room that has some of the qualities of the ESL57 that I have been in love with forever. They are the only piece of hifi equipment that has left me this impressed for this long with how real they sound through the midrange. Lots of my listening is to solo piano and chamber music.

From all the measurements I've seen the ESL57 is pretty much dead flat throughout the midrange. But when I do searches for speakers that other ESL57 owners like I frequently see the classic BBC speakers being brought up. Now this might be because most ESL57 owners are in the UK and these BBC speakers are much more common in the UK than the rest of the world. They also have some qualities that might draw ESL owners to them like they can be used in small to medium sized rooms.

I also have Harbeth SHL5 which I don't think have the very pronounced BBC dip (I get the impression Harbeth has been on a path to modernize their speakers for some time) and they are superb in the midrange, more so than dozens of other speakers I've heard but I still greatly prefer the ESL57.

I'm curious what others think.

 

[Gervais Cote]

To me, the ESL is dead neutral and very realistic compared to a small BBC monitor that is a bit more colored but also more lively, specifically with female voices.......but that’s just me !
The small BBC monitors are more exciting for me.

 

[chartz]

My OTA 57's can be lively with the proper material.

 

[Gervais Cote]

My OTA 57's can be lively with the proper material.

You might be right about the proper material, I heard these panels with Quad 405 amplifier only so rather limited listening and it was a long time ago.........

 

[chartz]

Sorry, by "material" I meant music and recordings.
On bright CDs or records, treble can bite on my ESLs too, but in an elegant way, ie not unpleasant. And the Quad 34 tilt button can do wonders here.

 

[tuga]

The BBC- or Gundry-dip is a broad-band relaxation roughly located in the presence region (around 3kHz):

The ESL57s are flat on-axis in that range and shouldn't produce such dip when toe'd in towards the listener:

 

[tuga]

The ESL57s seem to produce a very wide dip between 200 and 500Hz, probably a floor-bounce cancellation issue (though in theory this shouldn't show in a free-field plot).

The lower end of this very broad region is associated with a perceived sense of "thinness/fullness" and "coldness/warmth" (of which the ESL57s should be found lacking) but the upper reaches of the range are linked with "muddyness/clarity" (that should be plentiful with the ESL57s).

More info in the links below:

http://alexiy.nl/eq/

https://www.teachmeaudio.com/mixing/techniques/audio-spectrum/

 

[tuga]

In this graph we can see the extemelly beneficial effect that raising the ESL63 has in the lower mid-range response, practically eliminating the aforementioned floor-bounce cancellation:

 

[tuga]

I would like to add that non-waveguided speakers paired with large woofers will tend to create a BBC-dip at the listening spot (the sound reaching the listener is a sum of direct and early reflection) anyway, even when the anechoic response is flat on-axis, because the woofer have narrow dispersion at the crossover frequency whilst the tweeter's dispersion will be at its widest, an effect made more extreme in narrow rooms and/or by not pointing the speakers towards the listener (toe-in):

 

[tuga]

I also have Harbeth SHL5 which I don't think have the very pronounced BBC dip

The orignal (non-Plus) version had a more pronounced dip, but as I've mentioned in my previous post having an 8" woofer crossing into non-waveguided tweeter means that the response off-axis will produce a dip in the presence region at the listening spot, more or less pronounced depending on how close the speaker is placed to the side-walls, the amount of toe-in and whether or not the side wall are naked or if they provide some amount of diffusion or absorption.

The Stirling LS3/6 is an identical but more affordable alternative to the SHL5+ (shameless sales pitch, I know ).

 

[Electrostat]

I would like to add that non-waveguided speakers paired with large woofers will tend to create a BBC-dip at the listening spot (the sound reaching the listener is a sum of direct and early reflection) anyway, even when the anechoic response is flat on-axis, because the woofer have narrow dispersion at the crossover frequency whilst the tweeter's dispersion will be at its widest, an effect made more extreme in narrow rooms and/or by not pointing the speakers towards the listener (toe-in):

So does this mean speakers that have this dip on axis in an anechoic chamber will be even more pronounced at the listening position?

Regarding the price on Harbeth vs Stirling, at least in the US Harbeth has a huge dealer network and they come up for sale used far more often. I've watching the Stirling LS3/6 go from $4000 to $5000 to $6000 where it is at now.

 

[Electrostat]

I don't think these Youtube video clips are very good for evaluating sound, but here he is switching directly between two speakers that sound like they are matched in level. The LS50 sounds closed in and thick but the KEF LS3 sounds more open, now to my ears that is because of a roughly 1-2 KHz range emphasis at the expense of sacrificing some body.

 

[ToTo Man]

I don't think these Youtube video clips are very good for evaluating sound, but here he is switching directly between two speakers that sound like they are matched in level. The LS50 sounds closed in and thick but the KEF LS3 sounds more open, now to my ears that is because of a roughly 1-2 KHz range emphasis at the expense of sacrificing some body.

Also note that reflections from the desk will be colouring the frequency response of both speakers in unpredictable ways, not a good method to fairly compare the tonal balance of two speakers IMO.

 

[tuga]

So does this mean speakers that have this dip on axis in an anechoic chamber will be even more pronounced at the listening position?

Sometimes a dip on-axis is compensated by a peak off-axis but I think that this is generally the result of a resonance or some other problem, so not a good thing.
It happens a lot with Wilson Audio speakers for some reason. The Duette:

 

[Electrostat]

Sometimes a dip on-axis is compensated by a peak off-axis but I think that this is generally the result of a resonance or some other problem, so not a good thing.
It happens a lot with Wilson Audio speakers for some reason. The Duette:

Interesting observation, thanks. I also have JBL LSR 306 MKII which are about as "perfect measuring" as it gets with regard to polars, especially with how little they cost. But they don't sound remotely as engaging as the ESL57 or the Harbeth. My only guess is there is some breakup with the relatively cheap tweeter/waveguide combination that the brain doesn't like. In comparison the ESL57 is incredibly clean (measurements by Lynn Olsen)

With the SHL5 I can't recall how they sound firing straight ahead, when I bought them I subconsciously went straight to toeing them in and only tried various degrees of toe in.

 

[Ali T]

Having had 57’s in the past, I’d suggest a decent single driver speaker should get close to what you are looking for.

 

[tuga]

Having had 57’s in the past, I’d suggest a decent single driver speaker should get close to what you are looking for.

I disagree.

Single driver speakers produce significant amounts of intermodulation distortion (some call them "girl with banjo speakers" for a reason), increased directivity with frequency (which can only be slightly compensated with toe-in), a tonal balance that is tilting the wrong way (upwards from bass to treble), terrible high frequency performance (unless they're very small and produce no sub-bass), the cone break-up resonances are not filtered out, and they usually have limited SPL capabilities (or a very significat portion of the treble rolled-off).

 

[Ali T]

A sweeping statement. Not all drivers are the same.

Also, show me one speaker that does not exhibit increased directivity with frequency.

 

[tuga]

A sweeping statement. Not all drivers are the same.

Also, show me one speaker that does not exhibit increased directivity with frequency.

I'll do one better show you 5:

Dutch & Dutch 8C (wave-guided cardioid narrow baffle with rear woofers)

Gradient Helsinki 1.5 (waveguided coax open baffle)

MartinLogan Montis (planar dipole above 300Hz)

Avantgarde Uno Nano (horns above 350Hz)

MBL Radialstrahler 101E Mk.II (omni)

 

[tuga]

This is the Rethm Maarga, a back-loaded whizzer'd single driver:

And the Quad ESL-2805: