Bowers & Wilkins 683 S2 loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Bowers & Wilkins 683 S2's frequency response in the farfield; and an Earthworks QTC-40, with its small, ¼"-diameter capsule, for the nearfield responses.

The 683 S2 is specified as having a sensitivity of 89dB/W/m; my estimate of its voltage sensitivity was slightly lower, at 87.7dB(B)/2.83V/m. The impedance, specified as 8 ohms, drops below 4 ohms in the lower midrange and in the top audio octave (fig.1), with a minimum magnitude of 2.92 ohms at 124Hz, very close to B&W's specified minimum of 3 ohms.

Fig.1 B&W 683 S2, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

A small discontinuity in the impedance traces just above 200Hz implies the presence of some kind of cabinet resonance in that region. However, when I investigated the enclosure's vibrational behavior with a plastic-tape accelerometer, the primary resonance mode that I found, on all surfaces other than the top, lay at a high 531Hz, with a lower-amplitude mode at 348Hz (fig.2). These resonances are of fairly high Q and, taking into account their frequencies and amplitudes, unlikely to lead to any reduction in sound quality.

Fig.2 B&W 683 S2, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel level with midrange unit (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The red trace in fig.3 shows the output of the 683 S2's front-panel port, measured in the nearfield. It peaks between 20 and 40Hz, with the corresponding minimum-motion notch in the woofers' output (green trace) lying at 27Hz, as expected from the impedance graph. The two woofers behave identically above 50Hz, though with slightly different notch frequencies below 50Hz. Small blips are visible in the woofer and port responses just above 200Hz—the region where, probably and not coincidentally, the glitch in the impedance traces occurred.

Fig.3 B&W 683 S2, acoustic crossover on tweeter axis at 50", with nearfield responses of midrange unit (blue), woofers (green), and port (red), respectively plotted below 350Hz, 450Hz, and 600Hz.

Higher in frequency in fig.3, the crossover from the woofers to the midrange drive-unit (blue trace), measured in the farfield on the tweeter axis, occurs at 500Hz rather than the specified 400Hz, with steep filter slopes. The 683 S2's higher frequency response on this axis is disturbed by peaks and troughs of approximately equal amplitude—these are also evident in the overall response averaged across a 30° horizontal window centered on the tweeter axis, with the complex sum of nearfield responses plotted below 300Hz (fig.4). The rise in the B&W's upper-bass response is almost entirely an artifact of the nearfield measurement technique; the speaker is basically tuned to be maximally flat, with a –6dB point at the port tuning frequency of 27Hz.

Fig.4 B&W 683 S2, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield responses plotted below 300Hz.

Fig.5 shows the 683 S2's lateral dispersion, normalized to the tweeter-axis response. This graph shows that the on-axis notch between 1 and 2kHz deepens off axis, which will give the speaker a slightly polite balance—as Kal Rubinson noted, the presence of voices was not fleshed out to his complete satisfaction—though the similar notch an octave higher fills in to the speaker's sides, which is why experimenting with toe-in is important. In the vertical plane (fig.6, the traces again normalized to the tweeter-axis response), a sharply defined notch appears at the tweeter/midrange crossover frequency above the tweeter axis, while this graph also suggests that the optimal height for a seated listener's ears is actually slightly below the tweeter axis, which is 38" from the floor with the speaker sitting on its plinth.

Fig.5 B&W 683 S2, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis, reference response, differences in response 5–90° off axis.

Fig.6 B&W 683 S2, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–10° below axis.

This is confirmed by the step response on the tweeter axis (fig.7), where the decay of the tweeter's step (the sharp spike at 3.7ms) doesn't blend quite smoothly with the start of the midrange unit's step. This graph also reveals that all four drive-units are connected in positive polarity. The cumulative spectral-decay plot on the tweeter axis (fig.8) shows an impressively clean decay in the tweeter's passband, but some residual energy at 3640Hz, the frequency of the on-axis peak. I note that KR found the 683 S2 to have a smooth balance of midrange and treble, so perhaps this behavior measures worse than it sounds, though it would add to the sense of detail conveyed by the speaker.

Fig.7 B&W 683 S2, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 B&W 683 S2, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

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. But its measurements do reveal that, other than that lack of upper-midrange energy, the sound quality of B&W's 683 S2 has been carefully managed and better than one might expect for just $1650/pair.—John Atkinson

B&W Loudspeakers
US: B&W Group North America
54 Concord Street
North Reading, MA 01864
(978) 664-2870

chrisdemarse's picture

I found a deal on a pair of the 685 S2's and couldn't pass it up. I've experience the same uncanny experience of centralized vocals and very competent overall sound, but I can't seem to find any reviews anywhere that unpack them in detail with a more competent amplifier (i'm driving the modestly efficient speaker at 87db with a NAD D3020). Anyone else have a pair that cares to share their thoughts?

fetuso's picture

Chris, see my comment below

Anon2's picture

Besides the excellent "product of year" designation and accompanying video review of the B&W 685 S2 by What Hi-Fi, here are some other recent reviews:

CNET: (seems to be the successor site to, which gave a "Sehr Gut" rating to the 685 s2):

Numeriques (France):

Nextmedia (Australia):

I hope, still, that we can add Stereophile's review of the B&W 685 S2 to these soon.

This speaker, while perhaps not the best stand-mounter overall, is an outstanding product. I have heard the 685 on Marantz, Rotel, and Arcam amplification. Its easy impedance load and good sensitivity make it a versatile speaker with a variety of gear.

The 685 s2 projects a massive presence for its modest size. It has the characteristic, kind of laid-back "B&W sound." This sound can be laid-back for some, but there is little listening fatigue for this product. Many stand-mounters crossover to the tweeter around 2,800 Hz. Other, more expensive stand-mounters, cross over a lower frequencies of around 1,600 - 1,800. B&W with the 685 S2, as with their other stand-mount speakers get a good result, to my ears, by crossing over to the tweeter at 4,000 Hz.

Weight-wise, the B&W 685 s2 is not heavy at around 15 lbs per unit. There appears to be no inner-bracing of the cabinet. However, another leading manufacturer of speakers is asking almost $3,000 per pair for a pair of speakers that weigh almost the same as the B&W 685 S2 (makes you wonder where the additional $2,300 is going).

An interesting thing I have noticed is that the B&W 685 S2 is such a great deal at $700/pair, that the product at times seems to get limited exposure, by retailers and publications alike. I am sure that many would gravitate toward this product, over much more expensive alternatives, if they knew the sound at which this product was capable, compared to more margin-rich, but not much better products. A major retailer in my major city conspicuously has the B&W 685 S2 on display, but not connected to any amplification or source equipment; they are on display to be seen but not heard.

This is my amateur-hour appraisal of the B&W 685 s2. Read the professional reviews relayed by me for more polished insights.

Obrej's picture

I have had several more expensive pairs of speakers through my place lately looking for something that pairs well with my "new to me" Rega Elex-R, and the 685 S2's are staying put, as they sound fantastic - and they haven't even been run in properly yet. And I'll admit that I kind of looked down my nose at them because of their low price, and that they were being sold at Bestbuy, but man was I wrong. Believe the hype - these speakers are giant killers.

johnnyangel's picture

An interesting review of speakers I might be in the market for. Nice to read it on the web at least but I'm a paying subscriber whose Zinio copy of the edition from which this review was taken still hasn't shown up. Is there any end in sight to this ongoing problem?

John Atkinson's picture
johnnyangel wrote:
I'm a paying subscriber whose Zinio copy of the edition from which this review was taken still hasn't shown up. Is there any end in sight to this ongoing problem?

The Zinio edition of the September Stereophile was made available a week ago. (As a check on their schedule, I pay for a subscription and downloaded my issue last Saturday.)

You need to contact Zinio customer service. (As Zinio is an independent company, I am afraid there is nothing I can do.)

John Atkinson
Editor, Stereophile

johnnyangel's picture

Thank you, I will contact them. And meantime I will investigate the alternate subscription mechanism through the iTunes store (now, roll on, rumored large-screen iPad Pro).

In any case Stereophile is to be complimented for only offering limited servings of what has appeared in print on its website. Many titles now (particularly automotive) now provide almost every bit of their content on the Internet before readers can possibly have received their copies. Nice for the freeloader, but really an insult to paying subscribers!

fetuso's picture

Chris, the review of the 685s2 on recommends pairing them with the Arcam FMJ A19. For what it's worth, I recently upgraded from the D3020 to Peachtree Nova65se, and it's made a huge difference. I'm currently using Wharfedale Diamond 220's, which are 86db, and 8 ohm nominal. The 685's are on my list of possible upgrades.

spacehound's picture

Unlike that dopey Gamut thing. Its grille looks awful and serves no practical function, not being cat, dog, and child proof.

The B&W is made for the real world.

Dr.Kamiya's picture

...and its largely decorative grill: A system with a $39,000 speaker is likely to be in a room where cats, dogs and children aren't normally allowed in.

spacehound's picture

On whether you consider $39,000 a lot of money of course :)

I don't. But I would not buy the Gamuts as they CERTAINLY won't sound 20 times better than the B&Ws. Once you get beyond a 'bound to be rubbish' price point, way below the B%W price, HiFi price and performance are not often connected, particularly with speakers.

russm535il's picture

I have RevelF12 floor standing speakers since I have no way to audition speakers here in Pittsburgh This review is so good should I consider selling my Revels and moving to the 683 S2 ?
Thank you for your input !
Russ DeJulio
Pittsburgh PA

Kal Rubinson's picture

Whew! I cannot help you there. I loved the F12s but I last heard them back in 2006. I would have bought them then but they were a bit too tall to fit under my display.

russm535il's picture

Thanks Kal yes I have had. Mine awhile also they are definitely tall and heavy !

lifeliver's picture

I really stuck between 3 speakers,
Sonus faber venere 3
B&W CM10 S2
And Dali Rubicon 6
Can someone give me advise please ?


traynor3's picture

I plan to pair the B&W 583 S2 with Creek 50A. Could these 50wpc adequately power these 683 S2 in a 11' x 19' room?

ergonomico's picture

and compared them both to the 804d3 and the monitor audio silver s8 and s10 but, at least for my ears, they are way below the 804 and below the s8-s10. The B&W 683 s2 have excellent midrange for the price but that's it. The bass and highs are not extended and on many recordings they actually flat the instruments out. That goes for the highs but of course its not as evident as with the low frequencies. Compared to the MA S8 they sound less transparent, less extended and slightly moer boxy :/ Too bad because WAF was pretty high on the B&W and i really wanted to try out the brand but the 804 are way too expensive.

CJ's picture

Thank you for such an as-always articulated review. I am a fan of B&W and highly interested in this particular speaker thanks to your review. You mentioned "Certainly compared to my relatively ancient Paradigm Reference Studio/60 v.3s, the B&Ws had a much smoother midrange and treble, more stable imaging, and tighter, more resolved bass. In short, no contest." You also mentioned that the Studio 60 V5 is significantly improved over V3. That leaves me wondering how is BW 683 S2 compared to Paradigm Studio V5, in terms of those sound attributes in particular bass? Thank you.