Bowers & Wilkins 705 Signature loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Bowers & Wilkins 705 Signature's frequency response in the farfield, and an Earthworks QTC-40 mike for the nearfield and in-room responses. I measured the impedance with MLSSA, checking the results with Dayton Audio's DATS V2 system. I used the formula in a 1994 JAES paper by Eric Benjamin to calculate with Microsoft Excel what UK writer Keith Howard has called the "equivalent peak dissipation resistance" (EPDR, footnote 1).

Bowers & Wilkins specifies the 705 Signature's sensitivity as 88dB/2.83V/m; my estimate was within experimental error of that figure, at 87.6dB(B)/2.83V/m. The 705 Signature's impedance is specified as 8 ohms. The impedance magnitude (fig.1, solid trace) stays above 8 ohms in the midrange and mid-treble region, with minimum values of 5 ohms between 169Hz and 210Hz and 4.55 ohms above 20kHz. Though these impedance magnitudes are higher than the specified minimum value of 3.7 ohms, the electrical phase angle (dashed trace) is occasionally high when the magnitude is relatively low. For example, there is a combination of 6 ohms and –33° at 130Hz. As the 705 Signature has minimum EPDRs of 2.63 ohms at 46Hz, 2.4 ohms between 120 and 130Hz, and 2.18 ohms between 8.2kHz and 9.6kHz, it will work best with amplifiers that are comfortable driving 4 ohm loads. With the 705 Signature's port blocked, the impedance magnitude trace had a single peak in the bass, reaching 38 ohms at a relatively high 73.5Hz (fig.2).


Fig.1 Bowers & Wilkins 705 Signature with port open, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).


Fig.2 Bowers & Wilkins 705 Signature with port closed, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The traces in figs.1 & 2 are free from the small discontinuities that would imply resonances of some kind. However, when I investigated the enclosure's vibrational behavior with a plastic-tape accelerometer, I found a strong resonant mode at 516Hz on the sidewalls and top panel (fig.3). This mode has a high Q (Quality Factor), which, in combination with the relatively high frequency, might work against it having audible consequences.


Fig.3 Bowers & Wilkins 705 Signature, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The saddle centered at 55Hz in the impedance magnitude trace suggests that this is the tuning frequency of the flared port on the rear panel. The minimum-motion notch in the woofer's output (fig.4, blue trace), which is when the back pressure from the port resonance holds the woofer cone still, lies close to that frequency, at 57Hz. The port's output (fig.4, red trace) peaks between 40Hz and 90Hz and its upper-frequency rolloff is relatively clean, though there is some low-level liveliness between 500Hz and 1kHz.


Fig.4 Bowers & Wilkins 705 Signature, acoustic crossover on tweeter axis at 50", corrected for microphone response, with the nearfield responses of the woofer (blue) and port (red), respectively plotted below 300Hz and 1kHz.

Higher in frequency in fig.4, the woofer's farfield response on the tweeter axis is fairly even before it crosses over to the tweeter (green trace) at 3kHz. The crossover slopes seem to be low order, ie, fairly shallow, but the woofer's rolloff above 3kHz is well-behaved. The tweeter has a little too much energy in the octave above 6kHz, with a subsequent dip in its output just below 20kHz. The 705 Signature's farfield response, taken without the grille and averaged across a 30° horizontal window centered on the tweeter axis, is shown as the black trace above 300Hz in fig.5. The balance is nicely flat and even up to 3kHz, with then a slight dip between 3kHz and 6kHz and up to 5dB too much energy in the high treble. Repeating the measurement with the vestigial grille covering the woofer in place gave an identical result.


Fig.5 Bowers & Wilkins 705 Signature, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the complex sum of the nearfield woofer and port responses plotted below 300Hz (black) and the nearfield response of the woofer with the port blocked plotted below 200Hz (red).

The black trace below 300Hz in fig.5 is the complex sum of the nearfield woofer and port responses, with the latter compensating for the fact that the port is on the rear panel. There is the usual peak in the upper bass, which is almost entirely due to the nearfield measurement technique. The B&W's ported woofer tuning is maximally flat, in textbook manner. With the port fully blocked, the woofer's output (fig.5, red trace) rolls off below the upper bass with a 12dB/octave slope.

Fig.6 shows the 705 Signature's horizontal dispersion, normalized to the response without the grille on the tweeter axis, which thus appears as a straight line. The loudspeaker's radiation pattern narrows a little at the top of the woofer's passband and in the tweeter's second octave. Other than those issues, the contour lines in this graph are relatively evenly spaced. In the vertical plane (fig.7), again with the off-axis response normalized to the tweeter-axis response, the small dip just above the presence region flattens out 5°–10° below the tweeter axis. A major suckout in the crossover region develops 5° above the tweeter axis, however. As I found in my auditioning, the 705 Signature must be used with stands that place the tweeter axis just above the listener's ears to get the smoothest mid-treble response.


Fig.6 Bowers & Wilkins 705 Signature, 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.7 Bowers & Wilkins 705 Signature, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45–5° above axis, reference response, differences in response 5–45° below axis.

Fig.8 shows the B&W 705 Signatures' spatially averaged response in my room, generated by averaging 20 1/6-octave–smoothed spectra, taken for the left and right speakers individually using a 96kHz sample rate, in a vertical rectangular grid 36" wide by 18" high and centered on the positions of my ears. This tends to average out the peaks and dips below 400Hz that are due to the room's resonant modes. While the low bass gets some reinforcement from the lowest frequency mode in my room, the midbass region is shelved down a little. A loudspeaker that offers a flat on-axis response and well-controlled lateral dispersion gives a gently sloped-down treble in the spatially averaged room response, due to the increased absorption of the room's furnishings at high frequencies. By that criterion, the 705 Signature produces too much energy in-room between 6kHz and 11kHz. The responses at the listening position of the left and right 705 Signatures were closely matched above 1kHz.


Fig.8 Bowers & Wilkins 705 Signature, spatially averaged, 1/6-octave response in JA's listening room.

The excess of high-treble energy can also be seen in fig.9, which compares the B&W's spatially averaged response (red trace) with that of the Marten Oscar Duo, which Michael Fremer reviewed in November 2020 (blue trace), and that of the KEF LS50 that I reviewed in December 2012 (green trace). (The Martens were in the same positions as the B&Ws, the KEFs in similar positions.) The behavior of all three loudspeakers is very similar in the midrange, but the B&Ws have the most high-treble output, followed by the Martens, then the KEFs. The LS50s, however, have more presence-region output than either the Oscar Duos or 705 Signatures. At the other end of the spectrum, the Martens offer the best bass extension, with the B&Ws exciting the room's lowest-frequency resonance to a greater extent than the KEFs.


Fig.9 Bowers & Wilkins 705 Signature, spatially averaged, 1/6-octave response in JA's listening room (red), of the Marten Oscar Duo (blue), and of the KEF LS50 (green).

Turning to the time domain, the 705 Signature's step response on the tweeter axis (fig.10) reveals that the tweeter and woofer are both connected in positive acoustic polarity. The decay of the tweeter's step doesn't quite smoothly blend with the positive-going start of the woofer's step on this axis, which reinforces the implication in the vertical dispersion graph (fig.7) that the optimal blend of their outputs occurs just below the tweeter axis. Some ripples in the decay of the woofer's step correlate with a small amount of delayed energy just below 2kHz in the 705 Signature's cumulative spectral-decay plot (fig.11). Other than that, the B&W's waterfall plot is clean overall.


Fig.10 Bowers & Wilkins 705 Signature, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.11 Bowers & Wilkins 705 Signature, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

The Bowers & Wilkins 705 Signature offers generally excellent measured performance—better than that of the floorstanding 702 S2—though it is fair to note that the original 705 performed even better on the test bench.—John Atkinson

Footnote 1: EPDR is the resistive load that gives rise to the same peak dissipation in an amplifier's output devices as the loudspeaker. See "Audio Power Amplifiers for Loudspeaker Loads," JAES, Vol.42 No.9, September 1994, and
B&W Group Ltd.
US distributor: Bowers & Wilkins
54 Concord St.
North Reading, MA 01864
(978) 664-2870

tonykaz's picture

Every pair has its own unique grain pattern, they say.

The Wood is an Architectural product that comes in a wide range of species but only one specie is available from B&W.

No-one seems willing to say that these Loudspeakers ( and stands ) are made in China, even the very few internet images of its rear label panel are blurred enough to conceal the printed "Made in China" declaration while all the rest of the imaging remains beautifully clear & sharp. hmm. B&W does seem proud that they actually make a few 800 series somewhere in Europe.

So, the loudspeakers are drop-dead gorgeous, probably more beautiful than anything I own. ( I do have some French Impressionist Paintings in Storage that are NOT as glitzy as these 705 Signatures )

Country of origin is a big thing now-a-days

Tony in Venice

John Atkinson's picture
tonykaz wrote:
No-one seems willing to say that these Loudspeakers (and stands) are made in China . . .

I do note that fact on the Specifications page of this review, Tony. I have always felt that including the country of manufacture is relevant information for the readers.

John Atkinson
Technical Editor, Stereophile

tonykaz's picture

Thanks for writing

I missed it,

I only looked at "Manufacturer",

I didn't think to look at Seriel Numbers.

I applaud you, yet again.

Still, I did a quick Internet Search of all things B&W only to find a lack of accurate information. ( which is the anxiety source of my observational comment here ).

Made in China listed under Serial Numbers leads to the possibility that differing Serial numbers will have differing countries of origin, wouldn't it? ( perhaps that is the case, or is it? ) .


Mr. Micallef's report on that Austrian Integrated Amp might be the best and most captivating Audio writing I can remember. I hope you guys double his word budget, he might be the best that's out there. "lift my jaw up with a shovel" ! ( I'll have to double check but I don't think that he uses ly words or superlatives, for gods sake )

Tony in Venice

ken mac's picture

Much appreciated.

tonykaz's picture

are you the Mr.KM that I'm complimenting?

There is a type of complementing that would apply to what you are doing for Stereophile .

Tony in Venice

BillBrown's picture

I was interested to see your measurements, the "tailored" response clearly seen, this consistent with all of the measurements I have seen of modern B&Ws. Someone is certainly voicing them with a common goal. Clicking on the measurements of the Diamonds from years ago shows that wasn't previously the case, when my impression was that they were frequently used as recording monitors and much more neutral. I'm not sure that this change in direction is a good one.


MZKM's picture

You don’t even have to look at the Diamonds; J.A. linked his measurements of the original 705 (which costed just $1500), and the measurements are miles better than this modern Signature model (and also higher in sensitivity).

The prevailing assumptions online (especially due to all the news about executives jumping ship and being sold) is that this tailored sound is to make them stand out in show rooms, that the excess HF energy could be seen as added detail.

BillBrown's picture

I certainly think your second paragraph is true and almost mentioned it, especially since B&W expanded more into the "big box" stores (Best Buy, etc.).

JHL's picture

...before comments inevitably turn to rank assumptions about amplitude response, I see that this speaker's amplitude is superb at distance, the only metric that could matter, assuming amplitude is all that relevant. It also significantly betters the arguably over-rated small KEF.

Other than that we'll have to contend ourselves with what the expert reviewer heard and now reports, always a point of contention among the "objective" reader whose unique talent in spotting lines on graphs none of the rest of us can expect to achieve. Hidden in plain sight, as it were.

BillBrown's picture

I don't think that "rank" is a needed term or the word "objective" being placed in quotes. The opportunity to discuss the desired room response and correlation with the extensive literature re. preference and perceived neutrality by skilled and unskilled listeners doesn't need to be a bad thing. JA has certainly worked very hard over many years to provide measurements in the effort to correlate subjective impressions with objective data.

My impression is that I might find these initially impressive as alluded to above, but that for long term listening I would probably find them bright and fatiguing. I think this would likely be music-source dependent. I simply prefer a different in-room response. If someone else loves and enjoys them long term that is great!

I do think that B&Ws speakers now are less suitable to monitoring (I picture the original 801s seen in classical recording monitoring environments). The recordings monitored on their modern speakers I suspect would end up "dull."


JHL's picture

...with measured data include an almost insurmountable bias that they and virtually they alone describe sound. This is simply not true. *Rank assumptions* is quite descriptive, especially when sound itself isn't the evident aim of so-called Objectivist practice. Abstracts are his aim, abstracts and criticisms of listeners who look past data.

No one said it's a negative to discuss data and so-called subjective sound, Bill. It's it's just fundamentally incorrect to assume that data describes sound while doing so.

Likewise that JA has provided measured data for decades. That's a free-standing thought - he certainly has - and that data is also free-standing, which I'm trying to point out. But it's subtext as I see it, not pretext, *especially* when simple amplitude response is the Objectivist's nearly universal data bias for sound when it *isn't* sound.

Audio is for hearing and data is presumed to aid, not define it. It's especially not to refute or deny it.

Whether B&W are suited for monitoring is just not shown in this data. In fact they have been used for professional monitoring successfully for a very long time, and apparently for good reason. Here again, the instinct to *assume* sound from data is a bias.

The terms objectivist and subjectivist are reversed. The real Objectivist is the open-minded listener. The pseudo-objectivist simply assumes or projects sound, many times while not having the terms to adequately describe it. He may even disallow "subjective" prose entirely, urgently denying it. That's hardly objective.

BillBrown's picture

I understand what you are saying, and I do agree that there are "radical objectivists," but also "radical subjectivists." I am not in either camp and think both are missing out, dare I say misguided. I don't understand why people limit their data input and not incorporate both measurements and the subjective impressions of skilled listeners that we (?I) have grown to trust. In this case I think JA perfectly described subjective impressions that correlated with his measurements, while also nicely characterizing aspects of the speakers' sound that may not be reflected in measurements.

I am equally uninspired by the subjective comments of the "measurements only" camp as I am by subjective-only comments (especially in speakers as I do believe that there is some correlation between measurements and perception, and that in some designs the choices made can be predicted to lead to certain results that will be described subjectively in a predictable way). I will take the Stereophile way, rather than, say, the ASR or TAS way.

I don't find any of it worth fighting over, though (not that you and I are).

I will disagree with you on these speakers as an effective tool for monitoring. While it is certainly possible that a skilled recording engineer could produce recordings that translate well to different systems, I think they do have flaws in this regard.

Best wishes,


JHL's picture

I'll add that of the flaws associated with both camps, if they are camps, if given one choice, I'd defer to the subjectivist side. Audio is for hearing, not measuring.

Data is an abstract. It is a generation removed from pertinence. It can be seen but it cannot be heard. I'd miss data but the point I'm making doesn't require such an extreme.

The objectivist side has, in addition to its tremendous subjectivity toward and with data, the enormous problem of cynicism. It simply assumes that without data nefarious motives and directives must apply. That is the most subjective of all the flaws of objectivism. It's personal.

Add the two together and we have exactly what we see today: Objectivism is a brew of technical assumption and personal presumption. Not only does it not hear, I doubt it can. And it appoints itself your authority.

The subjectivist, meanwhile, just spins another record. He's completely harmless and if experience is a guide, he evangelizes for the high end while objectivism demeans it. That's another strike against objectivism as a movement. Not against *data*, against abusing it.

BillBrown's picture

Especially re. leaning toward the subjectivist side (particularly with the reviewers whose ears I have grown to trust over 30+ years). But please know I started at 18 :).

I think you describe the negative aspects of radical objectivism perfectly. The cynicism, designations of conspiracy and motive, dogma, and "talking down," contempt, etc. are tiresome. I don't like know-it-alls, as I don't think anyone, certainly not me, knows-it all.

Thank you for the nice exchange,


JHL's picture

Some people reject the simple handle "The High End". Some reject subjective language - it's not subjective, of course; it's standard common human vernacular relating to a very real experience - because it's not "objective". And many assume data is not just *more* descriptive, but that it actually displace experience. You can't hear that. You must hear this. And so on.

We have it backwards. It's nuts. Nowhere else in life I can think of do we reject the personal to this degree.

Yet audio is for *personal* enjoyment, and that's borne out in the fondness so many have for the well-executed review. The prose resonates.

We have to ask why we do audio. It's either to become reconnected with performance *or* it's to talk about aspects of the abstracts that literally have nothing to do with that performance and instead - and get this part - have to do with what we've preconditioned ourselves to listen for.

Not listen to. Listen for. We've elevated the singular focus of our assumption about sound, which is simple loudness uniformity, to the point that we hear only an excess couple of decibels at 3kHz. Or insufficient output at 1kHz. We've recalibrated ourselves to hear audio as a tone control. It's astounding.

That's bias and that's data bias. Data bias is not only prevalent, it's tracked back into engineering and changed the nature of the hardware. That doesn't mean an amplitude-centric component can't sound authentic; it means that we're potentially preventing ourselves music-centric components that aren't flat.

The way I hear it great audio gets down to one thing: Enough suspension of disbelief that you question how artists could be there in your space, transparent, playing a session recorded in 1973. I am transported and I see, hear, and feel the experience. Time machine.

Everything else isn't worth the effort and cost and it certainly isn't worth the years spent online arguing about a couple decibels here or there.

At some point you have to ask why you're doing this. It's plainly obvious it's for any number of reasons, yet when I read a great review I identify with the real sound in a real space or I realize the component in question is just parts in a box. And I've done that well before I ever get to the data.

Charles E Flynn's picture

From the manual at (bold style added):

For bi-wire connections or bi-amplification (above right), the terminal links should be removed and each pair of terminals connected to the amplifier or amplifiers independently. Bi-wiring can improve the resolution of low-level detail.

jimtavegia's picture

I am looking at the graph and see in Fig 9 that the B & W is at 0db at 10khz and the others are below which would indicate to me that they have less HF energy than the B&Ws. I do see about +3db at 150hz, 250hz, and 6-7khz, but I would not think this is a major issue as it is all going to change in most rooms and placements. I would agree that fig 5 and 9 are quite different.

It is normal for a speaker to measure this different between these plots? Maybe this is something I have missed in reading measurements over the years?

I would think this would be a very likable speakers and with a sub, might be an excellent choice for a small or medium sized room.

AJ's picture

Fig 5 is the (quasi) anechoic response of the speaker, i.e., the initial waves launch of the speaker itself, measured at around 36-50" away.
Fig 8 & 9 are in room, at LP averaged responses, so several feet away.
They would include sum total reflections from boundaries, distance attenuation from air/boundary absorption, room modes in the lower end, etc. All are "pressure at omni microphone" measures.
> 500hz +/- we(2 ears/brain) essentially/dominantly hear the speaker response, below, the room/speaker response.
So Fig 5 is far more representative of what will be heard at high frequency, for those with normal (audiometrics) hearing. "Flat" response in room several feet away, would tend to sound a bit bright.
Luckily, older men, aka, audiophiles, will have quite a bit of HF hearing "attenuation" to compensate.



jimtavegia's picture

I generally look at the in room response of where ever JA1 measures them and take that as to what they would sound like. You are right, as for me and my HF loss, flat does not help me as much, but I think this speaker measures pretty well.

I am still amazed at what Andrew Jones can do with much less money. Glad he can though.

Another poster has talked about what has happen at B & W corporate and where the company has been heading. It is a company with an illustrious history, and regardless this pandemic is not treating many people or companies well at all. I have listened to more music in the last 8+ months of lockdown than I have in the last 10 years I think. I am now into the middle of season two of the Crown and am enjoying it. The TV is showing us human beings at our best and worst on just about every channel.

TJ's picture

"...fair to note that the original 705 performed even better on the test bench. - John Atkinson". A recurring theme ever since they replaced the 800 Nautilus series. The 801N and 805N were so exceptional in both sound and test results.

Axiom05's picture

I agree that B&W seems to have changed their profile goals after the 800 Nautilus series from a maximally flat response to one that is "voiced" to have certain characteristics. Having owned N801, N803 and 802 Diamond speakers, the evolution is clear. However, I still think the Achilles Heal of the system is the FST midrange. Removing the surround leads to the requirement of a large diaphragm in order to have sufficient excursion capability at lower frequencies but gives poor off axis performance at higher frequencies. Of course it doesn't help that they make the mid/tweeter crossover so high in frequency. The two-way speakers seem to do better as the mid/woofer has a traditional surround.

TJ's picture

... which set a new standard for its on-axis response but was constrained off-axis. Likewise the FST midrange. As dazzling and mighty as the 801N was, the 805N somehow sounded more natural and life-like inside its dynamic range. I hope the new company owner takes note of KEF's impressive R&D team, rethinks its branding and gets back to its roots.

DougM's picture

I agree with you. The N805 Signature was, and still is, my dream speaker. I look at what has come after, with the loss of the awesome Kevlar wwofer/mid driver, and the poor measurements of the far more expensive Diamond series, and it has me scratching my head. I had pairs of both 601S3s and 602S3s, but unfortunately they needed to be played louder than my ears and small room can tolerate before they wake up and sound balanced, so they were sold. But, I still long for the times when I would turn them up and revel in their magic. I haven't owned anything that sounds as good since, and I've tried a lot of different speakers searching for something that could get that magic at lower volume levels, and I think Monitor Audios came the closest.

AJ's picture

Congrats John! ;-)

robertbadcock's picture

I sincerely miss the aesthetic of the original 705 - the slope of the front panel made it an item of art; and sound.

Of country of manufacture; that to me is also an item of reference - are the employees purchasing homes? Gathering 401Ks, or other benefits from their employ? Can I readily go visit the factory; or at a minimum; tour it online; as I can with Rega, Focal, Pro-Ject, and others?

Respectfully; and not unlike art (most of art - unique items can originate from horrid areas); I'm not one to purpose my economics towards an area of the world that may, or may not more or less be an epi-center of environmental, and human disposability.

Love my B&W items - 601S1s, ASW750, ASW 700, ASW600 - blocks of quasi-industrialist art in the cases of the subs; and of the 601s; well. Icons from um; me, near 20 years ago. Focals have replaced the 601s some time ago fwiw.
Classe amps, Ayre pre-amp, Luxman (Japan days) tuner, etc.
IDK, maybe it is a 'me' thing. Being able to speak of, and listen to; my gear is important.

'Braggin rights'; could be a modern day Greensbergism I suppose.

I do adore these beauties. Way far out of my reach; and I most certainly mean no harm with my words; and I also understand that many items would cost multiples more if made other than in China.

robertbadcock's picture

Just listening (lol, pun?) to this guy makes me want to buy a third TT.


John Atkinson's picture
John Atkinson wrote:
Back in June 1994, I reviewed the Bowers & Wilkins John Bowers Silver Signature standmounted loudspeaker. . . It was the best-sounding speaker I had used in my Santa Fe listening room . . .

The most recent time I heard a pair of B&W Silver Signatures was in February 2017, when I visited an audiophile in Chicagoland for Kyomi Audio's launch of the Vivid G1 Spirit loudspeaker. Before I interviewed Vivid's Laurence Dickie - see - our host Mario took me down to his basement listening room, where he had a pair of Silver Signatures set up. They sounded as fantastic as I remember from my Santa Fe days.

John Atkinson
Technical Editor, Stereophile

tabs's picture

Sorry to do this here, but is there any chance you can provide a hint when we might see the LS50 Meta review and measurements? I remember reading somewhere that you already had the Meta passive version in-hand or en route or some such. Thank you for all the hard work!

John Atkinson's picture
tabs wrote:
Sorry to do this here, but is there any chance you can provide a hint when we might see the LS50 Meta review and measurements?

My review of the KEF LS50 Meta will be published in the January 2021 issue of Stereophile, which will hit newsstands and mailboxes around December 12.

John Atkinson
Technical Editor, Stereophile

tabs's picture

Wunderbar. Thank you!

Charles E Flynn's picture

The USPS delivery of the January 2021 issue has been completed at an undisclosed location in New England. As proof, I offer this spoiler for the cover: There is a small image of the LS50 Meta, and the text reads: TWEAKED! KEF LS50 META