Tannoy Revolution XT 6 loudspeaker Measurements

Sidebar 3: Measurements

I measured the Tannoy Revolution XT 6's farfield behavior with DRA Labs' MLSSA system and a calibrated DPA 4006 microphone. I used an Earthworks QTC-40 mike for the speaker's nearfield responses. Tannoy specifies the Revolution XT 6's sensitivity as "89dB," which I assume is for 1W at 1m. My estimate was 88.8dB(B)/2.83V/m, which is both within the margin of error and a little higher than average. Props to Tannoy for an honest sensitivity specification.

Although the Revolution XT 6's nominal impedance is specified as 8 ohms, the magnitude (fig.1, solid trace) is greater than 8 ohms throughout the treble. The impedance magnitude drops below 8 ohms in the midrange, with a minimum value of 5 ohms between 185Hz and 230Hz. The electrical phase angle (dashed trace) is generally benign, as when it has a high value, the magnitude is also high, reducing the speaker's demand for current from the partnering amplifier. The shape of the magnitude trace in this graph, with a value of 12 ohms and above in the treble but averaging 7 ohms in the midrange, suggests that the XT 6's treble will be exaggerated with tube amplifiers that have high output impedances.


Fig.1 Tannoy Revolution XT 6, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

Slight discontinuities at 520Hz and 1400Hz in the traces in fig.1 imply the presence of panel resonances. However, when I investigated the enclosure's vibrational behavior with a plastic-tape accelerometer, the only modes I found on the sidewalls were lower in frequency, at 168Hz and 266Hz (fig.2), and were also low in level. The latter mode was stronger on the top panel; equally strong was a mode at 1400Hz, corresponding to the highest-frequency discontinuity in the impedance trace.


Fig.2 Tannoy Revolution XT 6, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of sidewall (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The saddle between 50Hz and 60Hz in the impedance magnitude trace suggests that the tuning frequency of the Tannoy's downward-firing port lies in this region. This was confirmed by the fact that the nearfield response of the woofer (fig.3, blue trace) has its minimum-motion notch at 50Hz. (The back pressure from the port resonance holds the cone stationary at the tuning frequency.) The nearfield response of the port (red trace) peaks broadly between 40Hz and 80Hz with a generally clean upper-frequency rolloff.


Fig.3 Tannoy Revolution XT 6, 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 350Hz and 1kHz.

The XT 6's manual specifies the crossover frequency between the woofer and tweeter as 1.8kHz; this is corroborated by fig.3. (The tweeter's output is the green trace.) The woofer's upper-frequency rolloff is disturbed by a series of high-Q resonant peaks, and the tweeter appears to be balanced between 3dB and 5dB too high in level. (Bearing in mind my earlier comment about the shape of the impedance curve potentially exaggerating the XT 6's treble, I took these quasi-anechoic measurements using a Krell KSA-50 amplifier, which has an output impedance of 0.13 ohms; the effect of this low output impedance on the frequency response will be minimal.) Note also the pattern of narrow peaks and dips in the tweeter's output. I suspect these are interference effects due to the concentrically mounted tweeter's symmetrical acoustic environment. They should not affect sound quality.

The black trace below 300Hz in fig.4 shows the sum of the Tannoy's nearfield woofer and port outputs, taking into account acoustic phase and the different distance of each radiator from a nominal farfield microphone position. The excess of upper-bass energy will be due to the nearfield measurement technique, which assumes the radiators are mounted on a baffle that extends to infinity in both horizontal and vertical planes. The XT 6's reflex alignment appears to be tuned to be maximally flat, in textbook fashion.


Fig.4 Tannoy Revolution XT 6, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response (black), with the complex sum of the nearfield woofer and port responses plotted below 300Hz.

The Tannoy XT 6's farfield response, averaged across a 30¯ horizontal window centered on the tweeter axis, is shown as the black trace above 300Hz in fig.4. The midrange is flat, but the region covered by the tweeter is too high in level, and a significant peak is visible between 2kHz and 3kHz. (This measurement was made without the vestigial grille; repeating it with the grille in place didn't change the response to any significant degree.) In an email, Herb did note that there was excess energy at "about 2.2kHz."

The plot of the Revolution XT 6's dispersion in the horizontal plane, referenced to the response on the tweeter axis, is shown in fig.5. The contour lines in this graph are relatively even up to 10kHz, which, all things being equal, will correlate with stable stereo imaging. However, as can be seen in fig.6, which shows the actual responses to the speaker's sides, the peak between 2kHz and 3kHz in the on-axis output continues all the way to 90° off-axis. In the vertical plane (fig.7), the XT 6 maintains its on-axis response over a wide window.


Fig.5 Tannoy Revolution XT 6, 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 Tannoy Revolution XT 6, lateral response family at 50", from back to front: responses 90–5° off axis, on-axis response, responses 5–90° off axis.


Fig.7 Tannoy Revolution XT 6, 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.

In the time domain, the Tannoy XT 6's step response on the tweeter axis (fig.8) indicates that the concentric tweeter is connected in positive acoustic polarity, the woofer in inverted polarity. The decay of the woofer's step is overlaid with regular undulations, which correlates with two strong ridges of delayed energy between 2kHz and 3kHz in the Tannoy's cumulative spectral-decay plot (fig.9). I suspect that this behavior is due to break-up modes in the woofer's paper cone that have not been suppressed by the second-order low-pass filter used in the crossover. The decay in the midrange is clean, though other lower-level ridges are apparent throughout the treble.


Fig.8 Tannoy Revolution XT 6, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.9 Tannoy Revolution XT 6, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

While there were some good aspects to the Tannoy Revolution XT 6's measured performance—the higher-than-average sensitivity, the easy-to-drive impedance, the optimally tuned low-frequency alignment, and the clean, even midrange output—I was bothered by the excess of energy in the presence region, which I could hear with the MLSSA pseudorandom noise signal when I was performing the measurements.—John Atkinson

Tannoy Group Ltd.
US Distributor Upscale Audio
2058 Wright Ave.
La Verne, CA 91750
(909) 931-9686

Bogolu Haranath's picture

I agree ...... Ortofan is an 'astute' reader :-) .......

davip's picture

"...Imagine sound that's thin, metallic, herky-jerky, dull, and rolled off completely below about 90Hz. I repeat: rolled off completely below 90Hz..."

That, Herb, was what I heard from the Tannoy DC2000s that I stupidly bought online in 1989. Unlike your experience, they Never broke in and sounded as sh1t on Day 300 as they did on Day 1. I could never understand why a speaker with two 7-8" woofers in an IB enclosure over 1-m tall could sound so anaemic in the bass. By Day 300, I didn't care and sold them for 1/3 purchase to someone to sit aside his Sony Profeel monitor.

I kid you not when I say that I have heard loudspeakers the size of a 2lb bag of sugar with better bass-response than those Tannoys, and I eventually chalked that absence of bass and crap, shouty sound quality down to simple lack of engineering -- anyone who understood what they were doing in Tannoy is long-dead and gone. I will never buy anything from this junk brand again -- and will happily shout my experience for anyone who's thinking of putting money their way.

It might not seem like a particularly "astute" comment, but when you start a review with "...sound that's thin, metallic, herky-jerky, dull, and rolled off completely below about 90Hz. I repeat: rolled off completely below 90Hz...", it's questionable that you should finish that review with the words "Highly Recommended". This isn't 'break-in', it's Acceptance, and quote 1 mirrors my experience of the Tannoy Dual Concentric driver exactly...

beave's picture

It's a good thing you don't hold a grudge for long! ;-)

davip's picture

I'll remember you said that... X-/

beave's picture

Believe me, I was afraid to submit that post! :-)

emir's picture

Yes certainly Tannoy is not the brand it used to be when they introduced their coaxial drivers back in the four ties and their huge loudspeakers Westimster for example.Now they are just a name uses by some international group more interested in profit than in high quality.Note as some engineers left the brand and started a new brand Fyne Audio to continue the legacy

AJ's picture

The true nature of the beast.


beave's picture

I definitely like seeing the non-normalized plot along with the normalized one.

Bogolu Haranath's picture

It would be nice if Stereophile includes non-normalized response curves on a routine basis ..... The ripple effect between 2kHz and 3kHz is better shown in the non-normalized graphs for example, with these Tannoy speakers ....... The FR jumps up almost +7 db from 2kHz to 3kHz ..... Non-normalized graphs are some what similar to the 'Spinorama' graphs obtained at Harman International :-) ......

Erin's picture

Some prefer one method. Others prefer another. I provide the data in both ways just so people won't raise pitchforks. For example:

AJ's picture

Others prefer another.

Well, one shows what hits the walls, one doesn't.
To each their own.
Nice site.



Erin's picture

Absolutely. Just saying that I wish more would provide both to catch both sides of the coin.

funambulistic's picture

I know you from the AH site! Loved your reviews of the Buchardt and Jamo speakers.

Bogolu Haranath's picture

It would be interesting to see the measurements of MC Audiotech and Endow Audio speakers ....... They were shown in Florida Audio Expo 2020 and RMAF 2019 ........ Both JM and JVS liked their sound during those shows :-) .......

Bogolu Haranath's picture

Nice listening room ..... BTW, are those copper infused socks? :-) .......

Erin's picture

Yes, yes, they are copper infused socks. Not because of their ability to help stuff ports properly. But because of their conductive properties when it comes to blood flow. Gotta keep those toes a-wigglin'. :D

Bogolu Haranath's picture

JA1 routinely provides port FR measurements ...... See, Fig.3 red trace in measurements section for this Tannoy speaker model :-) .........

Bogolu Haranath's picture

Take a look at Stereophile port FR measurements of Totem Skylight speakers ....... Fig.3, red trace in measurements :-) .......

Bogolu Haranath's picture

Take a look at 22 song playlist Rolls Royce uses to fine tune their car audio systems, posted on Sound & Vision magazine website ........ How come you don't have 'Hotel California' on your playlist? ...... Rolls Royce has it on their playlist :-) ........

tonykaz's picture


the factory is supposed to have Quality Control burning everything in, with a final test sign-off! ( one more china problem )

Keven Deal should've checked these Loudspeaker's performance before he let them out the door.

Deal and Tannoy got very lucky that this reviewer persevered , put the drivers thru a full 100 hour initial break in. I doubt that I would've had the patience of JOB like Mr.HR displayed and I doubt that buyers in today's marketplace will put up with such a defective 'First Impression'.

I can't recall ever getting a "Highly Recommend" on any product that didn't dazzle right out of the box.

I'm wondering: How long will this broken-in Loudspeaker System perform adequately before it's hidden problem re-surfaces. We need accurate, no-bs explanations from Mr.Deal ( don't we? ) and from the Tannoy people considering that Stereophile is the highest integrity Audio Journal.

Mr.JA1 described the complexity of this design and highlighted a persistent problem lurking in it's presence region.

Tony in Venice

ps. these things might sell quite well on the shelves of Best Buy, probably discounted down to some attractive 'Street Price'

er1c's picture

Herb you work so hard to find redeeming qualities in things, ie, critical comments followed by a reason to praise; and I find this endearing. And you write like a painter (Oh yeah, you are that) I own nos tubes purchased at Upscale, a few great a few so so in my Rogue Sphinx v2 and now v3. With Kevin (and I very much enjoy his enthusiasm) everything is fabulous as if he's partaken of very high quality sacred herb just prior. I mean no offense, just saying. I am acquaintance of two very high profile recording engineer/producers who use early Tannoy Monitor Golds with custom cabinets, as studio monitors and since they can afford and obtain anything they want, these must actually be fabulous. I mean they use damn original RCA 44s on drums for crying out loud (I am actually crying) Judging by the artists they've recorded and how amazing the records sound I have no doubt in thier choice of gear. So there must be TANNOY and then tannoy Thanks Herb, I own Spinx v3 and love it whether you do or not, it is a wonderful sound in my economic reach.

Herb Reichert's picture

of the Sphinx V3 (Followup to follow) as well Tannoy Monitor Golds. Also...

I am a forever fan of Tannoy Arden and Tannoy Red or Black drivers on open baffles - just sayin'


Bogolu Haranath's picture

Is it possible to get measurements of Sphinx V3? :-) ......

John Atkinson's picture
Bogolu Haranath wrote:
Is it possible to get measurements of Sphinx V3?

Herb's followup appears in the August issue and is accompanied by a full set of measurements. And yes, he does evaluate the headphone output.

John Atkinson
Technical Editor, Stereophile

Bogolu Haranath's picture

May be HR could also review the new PS Audio Stellar Strata integrated amp ($3,000) :-) .......

Bogolu Haranath's picture

Hope HR evaluates the headphone output of Sphinx V3, also :-) .......

beave's picture

It would be interesting to have measurements for both a "broken-in" speaker and one straight out of its box. If not measurements, at least listening comparisons might be worthwhile, to see how much the break-in is the speaker changing and how much is the reviewer's perception changing.

Bogolu Haranath's picture

Breaking news! ........ After 100 hours, the speakers have finally broken-in! :-) ........

Bogolu Haranath's picture

Friend: Herb, is there a new construction going on in your listening room? ..... I hear jackhammer sound ....
Herb: No, those are my new speakers breaking-in :-) .......

Ortofan's picture

... better off by instead submitting for review the Revolution XT-8F?

Measurements accompanying the Hi-Fi World review indicate that the low frequency response peaks at about 60Hz, which would likely address HR's comments regarding a lack of output below 90Hz for the XT-6. Foam plugs for the ports are available if the low frequency output is too great. Also, the ~7-8dB peak in frequency response at about 3kHz, shown in JA1's test data, appears to be absent in the XT-8F. Note that the HFW reviewer remarked that the XT8 "had an expansive, immersive sound that encouraged me to extend listening sessions just a little bit longer."


In another review, the XT-8F was declared a Hi-Fi Choice group test winner, besting competitors from DALI, Epos, GoldenEar, KEF and Q Acoustics.


Certainly the $2,300 XT-8F costs more than double the price of the XT-6, but perhaps that added expenditure might yield more than proportional gains in sound quality. In between, there is the $1,700 XT-6F, which received a 5-star rating from What Hi-Fi.


Bogolu Haranath's picture

KEF R5 ($3,000/pair) competes well with Tannoy XT-8F :-) .......

Ortofan's picture

... make that determination?

Bogolu Haranath's picture

KEF R5 was reviewed by TJN for S&V magazine (along with other KEF HT models) ....... Judging from the KEF R3 and R11 models reviewed by Hi-Fi News, most likely the R5 is also a good value for the money :-) .......

Bogolu Haranath's picture

Because XT-8F is a 2 1/2 way design, they may provide slightly more low bass extension than the KEF R5 :-) ......

Ortofan's picture

... the 2-1/2 way XT-8F might provide slightly more low bass extension than the 3 way KEF R5?

Bogolu Haranath's picture

Tannoy XT-8F has two 8" mid-bass and bass drivers ...... Since XT-8F is a 2-1/2 way design, both 8" drivers operate into bass frequencies, up to the port tuning frequency ...... KEF R5 is a 3-way design with two 5.25" drivers operating in the bass frequency up to the port tuning frequency ...... So, the XT-8F could have as much as 10 Hz low bass frequency extension than R5 ......

It is possible that the R5 could have less TIM distortion in the midrange with better midrange resolution than the XT-8F, because R5 is a 3-way design :-) ........

Kal Rubinson's picture

...... Since XT-8F is a 2-1/2 way design, both 8" drivers operate into bass frequencies, up to the port tuning frequency ..

Both 8" drivers will operate down to (and below) the port frequency as well as up to the LP filter for one of them.

Bogolu Haranath's picture

Yes, agreed ......There would be rapid fall-off of bass frequency, below the port tuning frequency :-) ......

Kal Rubinson's picture

I was too subtle. They both perform down to the port frequency, not up!

Bogolu Haranath's picture

Yes, the lower woofer rolls-off above 250 Hz :-) .......

Bogolu Haranath's picture

Ha, I it see now ...... I should have said 'both drivers operate 'down' to the port tuning frequency' ....... Thank you for the correction ....... I should work on my English language :-) .......

Bogolu Haranath's picture

Aha ..... KR's next reviewing project, Tannoy XT-8F :-) .......

Bogolu Haranath's picture

HR could review the NHT C3 ($1,000/pair to $1,100/pair) ...... Was reviewed and measured by Home Theater HiFi :-) .........

Bogolu Haranath's picture

SVS Prime Pinnacle ($1,600/pair) is another floor-standing speaker favorably reviewed by S&V magazine and Hi-Fi News :-) ........

Ortofan's picture

... in regard to (apparently high) levels of distortion, over on this site:

Roger That's picture

…in an relatively recent online interview, break-in should apply mainly to the woofer(s), and it should be accomplished in a matter of seconds by applying a very low frequency signal (that makes the cone and its suspension traveling in high excursions without applying too much power to the voice coil/thus less heat).

He even defends that the parameters don’t change significantly after those seconds of hard work (except for the mechanical wear and tear that time, years and/or operation will unavoidably bring along), and that a loudspeaker that isn’t used for a very long time can benefit from that process again.

I don’t have any credentials to confirm or deny statements by someone like Andrew Jones, although I can say that I’ve measured the resonant frequency and impedance curve of used drivers (woofers) that were stored for quite some time before and after that exercise mentioned by AJ, and they definitely changed.

I ended up not taking the extra steps in my measurements and go through the Thiele-Small parameters, but I’d almost bet that it would likely be different before and after that exercise (I used a sine wave around 12 Hz or so for less than a minute).

I assume that this was a pure anecdotal experience, but it does correlate with what a very respected loudspeaker and driver designer says (he doesn’t seem to have much faith in “breaking in” tweeters and other smaller drivers for a long period of time, and honestly, neither do I in my experience*).

And all of this to say that breaking-in these specific Tannoys won’t ever change one of the following facts:

  • Either there was a problem on manufacturing and these units were not up to specifications, or
  • This is a poorly designed loudspeaker which shouldn’t be sold by this brand at any price, let alone $1000;

None of these scenarios is positive, because one would suggest a terrible quality control at the manufacturing stage, and the other would suggest just plain wrong engineering (which doesn’t make sense in a brand with these resources).

But that FR is way too flawed to sound anything close to “right”, let alone neutral. That is a ridiculous response any way we look at it.

That “thin, metallic” sound does correlate to what JA measured, and the “rolled off completely below about 90Hz” wouldn’t be surprising if the nearfield rise in level is removed, and other room-related issues combine in a way that’s not favourable to this underachieving loudspeaker.

In the end, almost any loudspeaker can sound “ok” if a combination of factors happens at the same time (the music used for testing, the interaction with the room and partnering equipment’s, and even reviewer bias (we all have our own, whether we realize it or not).

Comparing these measurements with almost any reputable British manufacturer (as reviewed by Stereophile) paints a very clear picture, even at lower price points.

In all fairness, JA (in all its amazing diplomacy) is very clear on his final assessment, which is definitely not something worth a “Highly recommended” measured performance.

I would like to see the in-room response in the reviewer’s room at the end of the review, in order to see if we could make some sense of what made this poorly engineered loudspeaker deserve such a high accolade by the reviewer.

I cannot praise Stereophile enough for having measurements at the end of a review, because (without any disregard for the reviewers opinion at all), this is why people (myself included in the past) would buy a recommended product from an audio magazine, only to find out that the harshness (or any other problem) would never go away.

By making the “complex sum” of the reviewer words with JA’s measurements, things tend to become a lot clearer (or harsher -literally in this case).

*To be clear, my personal experience and beliefs are only mine, without any pretension to be right.

trynberg's picture

It's appalling this poorly engineered speaker earned a "highly recommended" designation. The frequency response is so awful and the woofer rings like a bell. This kind of performance shouldn't be excused on a $300/pair speaker, let alone a $1,000 pair. Pathetic.