KEF LS60 Wireless Just Announced

[Purité Audio]

I have the kiis/D&D here and have taken Gen 8351s in part exchange, I no longer have the Gens but tonally the LS60s are not dissimilar as you would expect ,what specifically would you like to know?
Keith

 

[harkpabst]

Interesting ! .
The crossover for the uniQ driver seems to be slightly lower in frequency than the spec .- in this review it seems to be at around 280 Hz .

No, not really. Obviously the woofers have been measured independently (near-field?). Summing up their output means pushing up the measured graph. You cannot simply look at the point where those lines are crossing.

 

[Tangband]

No, not really. Obviously the woofers have been measured independently (near-field?). Summing up their output means pushing up the measured graph. You cannot simply look at the point where those lines are crossing.

I looked again and yes, you are absolutely right.

 

[Tangband]

Test Streamingbox KEF LS60 Wireless - LowBeats

Es ist nicht lange her, da hat mir der Test der 8351B von Genelec im wahrsten Wortsinn die Augen geö

www.lowbeats.de

Sound comparison with Genelec 8351b.

 

[stren]

Stereophile review with some measurements is up

 

[delta76]

Stereophile review with some measurements is up

KEF LS60 Wireless loudspeaker Measurements

Sidebar 4: Measurements

www.stereophile.com

Save you a few clicks.

As I said earlier this remains my wet dream.

 

[Tangband]

KEF LS60 Wireless loudspeaker Measurements

Sidebar 4: Measurements

www.stereophile.com

Save you a few clicks.

As I said earlier this remains my wet dream.

Yes .
As one can see, the response is very flat.

 

[delta76]

Yes .
As one can see, the response is very flat.
View attachment 265329

I believe with it was rated as 8.0 without a sub, and 8.8 with the perfect sub, from its previous spin data. Only if I have a dedicated theatre room where I can put my Focal set . this will be only my main speakers in living room.

One can dream, you know.

 

[vanwalsum]

I know this can be a silly question and the answer is subjective. I’m now playing with a set of dali Epicon 6 paired to a Marantz model 30 and sacd30n. I’m considering the ls60 as a replacement. Would you say it could be an upgrade or a downtrade.

 

[sifi36]

I know this can be a silly question and the answer is subjective. I’m now playing with a set of dali Epicon 6 paired to a Marantz model 30 and sacd30n. I’m considering the ls60 as a replacement. Would you say it could be an upgrade or a downtrade.

If you don’t have a very large room or play very loud the LS60 will extend a little deeper. At very high volumes or at longer listening distances, the KEF rolls off on the bass side of things with its built in limiter to a point where eventually the Dali will go lower, probably around 95db at 1 metre, if the Dali specifications are to be believed.

I can’t find a spinorama, though the absolute sound review references a dip around 1-2kHz for the Epikon 6 and the Stereophile review of the Rubicon 8 and Ikon 6 show the same behaviour. If you like a “smiley face” curve, the KEF won’t be for you as it’s basically flat bar an on-axis dip around 10kHz. This is according to Stereophile, whose measurements are not as reliable as carried out here, though they tend to agree broadly. Ignore the peak around 100Hz for both, it’s a measurement artefact.

Dali Rubicon 8:

LS60:

TLDR: if your room isn’t massive and you want a largely flat response, then the KEF is a big upgrade. As would be the Buchardt A700 and the equivalent Dynaudio Focus speaker.

 

[witwald]

Well, lets see how the coming review in Stereophile looks like . There is enough measurements now that shows that the frequency response of ls 60 is not ruler flat.

KEF's own measurements are shown below. Would it be entirely reasonable to assume that KEF knows how to measure loudspeakers in general? Their results look much, much flatter (as expected) than those produced by some others. The results presented by Stereophile seem to be quite similar, although less comprehensive.

 

[witwald]

The Klippel shows that DI is not flat up to 150Hz, contrary to the data supplied by KEF.

Can anyone offer an explanation as to what a negative DI value means? I thought DI couldn't drop below 0 dB, which corresponds to omnidirectional sound radiation.

 

[sifi36]

Can anyone offer an explanation as to what a negative DI value means? I thought DI couldn't drop below 0 dB, which corresponds to omnidirectional sound radiation.

KEF measure their speakers in their own, relatively small anechoic chamber. As such, their measurements become less accurate at low frequencies, so I’d take their representation of the measured performance below around 300Hz with a grain of salt. For the R series, their own spinorama wasn’t plotted below 300Hz.

Below is a comparison of the KEF spinorama with Erin’s (who used a Klippel NFS) for the Reference 1:

And the on-axis response alone:

The KEF data is the dotted lines in both charts, which are from spinorama.org (an excellent resource created by @pierre). In both cases you can see that KEF’s data underrepresents how flat the Reference 1 is through the low frequency area and the extension isn’t quite as low as what KEF’s data shows. I’d say the Stereophile on axis response is more favourable than KEF’s own for the LS60 and wouldn’t be surprised if the same were to be found once someone gets it on a NFS.

 

[thewas]

Can anyone offer an explanation as to what a negative DI value means? I thought DI couldn't drop below 0 dB, which corresponds to omnidirectional sound radiation.

It is an artefact of the measurement when there are partial cancellations so that the total sound power is lower than the direct sound.

 

[vanwalsum]

If you don’t have a very large room or play very loud the LS60 will extend a little deeper. At very high volumes or at longer listening distances, the KEF rolls off on the bass side of things with its built in limiter to a point where eventually the Dali will go lower, probably around 95db at 1 metre, if the Dali specifications are to be believed.

I can’t find a spinorama, though the absolute sound review references a dip around 1-2kHz for the Epikon 6 and the Stereophile review of the Rubicon 8 and Ikon 6 show the same behaviour. If you like a “smiley face” curve, the KEF won’t be for you as it’s basically flat bar an on-axis dip around 10kHz. This is according to Stereophile, whose measurements are not as reliable as carried out here, though they tend to agree broadly. Ignore the peak around 100Hz for both, it’s a measurement artefact.

Dali Rubicon 8:
View attachment 272615

LS60:
View attachment 272614

TLDR: if your room isn’t massive and you want a largely flat response, then the KEF is a big upgrade. As would be the Buchardt A700 and the equivalent Dynaudio Focus speaker.

Thx for the extensive answer. Maybe I will buy then before I sell mij Epicon 6 and Marantz and look how the can satisfy me.

 

[NTK]

Can anyone offer an explanation as to what a negative DI value means? I thought DI couldn't drop below 0 dB, which corresponds to omnidirectional sound radiation.

The definition of the sound power (or early reflections) directivity index is the ratio of the listening window response to the sound power (or early reflections) response. When in dB scale, it becomes the listening window response minus the sound power (or early reflections) response.

If for some reasons the listening window response at certain frequencies is lower than the sound power (or early reflections) response, you'll get a negative DI. We see that often for speakers with rear reflex ports. When the port is providing significant output, the speaker output pointing toward the back can be significantly higher than toward the front, and may result in a higher sound power number (the speaker output averaged in all directions) than the listening window number.

Using the Revel M105 as an example, below are the polar directivity plot at 51 Hz and the Spinorama plots. You can see that the output pointing toward the rear is stronger than toward the front, and the "eyeballed" average (≈ sound power) is about 2 dB higher than the listening window value (≈ on-axis in our case at the low frequencies). The SPDI value from the Spinorama graph is ~-1.6 dB, which is close enough to our eyeballed number. Also note that the calculation of the early reflections response is heavily weighted toward the front, and therefore the ER response is closer to the listening window response (than the sound power response), and that gives an ERDI much closer to 0 (but still negative).

 

[witwald]

The definition of the sound power (or early reflections) directivity index is the ratio of the listening window response to the sound power (or early reflections) response. When in dB scale, it becomes the listening window response minus the sound power (or early reflections) response.

If for some reasons the listening window response at certain frequencies is lower than the sound power (or early reflections) response, you'll get a negative DI.

Thank you very much for the explanation. The negative results now make perfect sense, based on the method that has been used for their calculation.

We see that often for speakers with rear reflex ports. When the port is providing significant output, the speaker output pointing toward the back can be significantly higher than toward the front,

That's quite clear as well. However, something seems a little unusual about the results as presented.

At low frequencies, say around 51 Hz in your subsequent Revel M105-based example, the port and the woofer are acting as monopoles, albeit ones that are spaced apart by a small fraction of a wavelength. The wavelength of sound at 51 Hz is 6725 mm, so even a 250mm-diameter woofer is only 3.7% of that wavelength.

The far-field radiation of a monopole such as the one just described is omnidirectional, isn't it? The fact that the measurements (see below) show a higher output towards the rear (180°) than to the front (0°) would seem to indicate that the measurement distance from the acoustic center of the source simply was greater for the measurement at 0° than at 180°.

View attachment 272885

Using the Revel M105 as an example, below are the polar directivity plot at 51 Hz and the Spinorama plots. You can see that the output pointing toward the rear is stronger than toward the front, and the "eyeballed" average (≈ sound power) is about 2 dB higher than the listening window value (≈ on-axis in our case at the low frequencies). The SPDI value from the Spinorama graph is ~-1.6 dB, which is close enough to our eyeballed number. Also note that the calculation of the early reflections response is heavily weighted toward the front, and therefore the ER response is closer to the listening window response (than the sound power response), and that gives an ERDI much closer to 0 (but still negative).

Beranek and Mellow (2019) state that: "The directivity factor is the ratio of the intensity on a designated axis of a sound radiator at a stated distance r to the intensity that would be produced at the same position by a point source if it were radiating the same total acoustic power as the radiator."

As a result, I don't really understand why the front-facing axis has been chosen as the reference, as it would seem that, according to Beranek and Mellow (2019), "usually, the designated axis is taken as the axis of maximum radiation". That way, it's not possible to get a DI value less than 0 dB, and the comparison offered by this single number makes (intuitive) sense. In the above example, choosing the front-facing axis of radiation as the reference axis produces a confusing result, at least to me.

Reference:

Leo Beranek, Tim Mellow (2019). Acoustics: Sound Fields, Transducers and Vibration (Second Edition). Academic Press, ISBN 9780128152270.

 

[NTK]

Thank you very much for the explanation. The negative results now make perfect sense, based on the method that has been used for their calculation.

That's quite clear as well. However, something seems a little unusual about the results as presented.

At low frequencies, say around 51 Hz in your subsequent Revel M105-based example, the port and the woofer are acting as monopoles, albeit ones that are spaced apart by a small fraction of a wavelength. The wavelength of sound at 51 Hz is 6725 mm, so even a 250mm-diameter woofer is only 3.7% of that wavelength.

The far-field radiation of a monopole such as the one just described is omnidirectional, isn't it? The fact that the measurements (see below) show a higher output towards the rear (180°) than to the front (0°) would seem to indicate that the measurement distance from the acoustic center of the source simply was greater for the measurement at 0° than at 180°.

View attachment 272885

Beranek and Mellow (2019) state that: "The directivity factor is the ratio of the intensity on a designated axis of a sound radiator at a stated distance r to the intensity that would be produced at the same position by a point source if it were radiating the same total acoustic power as the radiator."

As a result, I don't really understand why the front-facing axis has been chosen as the reference, as it would seem that, according to Beranek and Mellow (2019), "usually, the designated axis is taken as the axis of maximum radiation". That way, it's not possible to get a DI value less than 0 dB, and the comparison offered by this single number makes (intuitive) sense. In the above example, choosing the front-facing axis of radiation as the reference axis produces a confusing result, at least to me.

Reference:

Leo Beranek, Tim Mellow (2019). Acoustics: Sound Fields, Transducers and Vibration (Second Edition). Academic Press, ISBN 9780128152270.

I used Amir's NFS data to create the polar directivity plot. Below is an animated one showing the directivity plots from 20 - 1000 Hz. The data from the NFS showed that the radiation at the lowest frequencies weren't omni-directional, but started becoming so by around ~70 Hz. I haven't dug into the reason why.

In this case only at the very low frequencies, due to the port, that the reference axis is not where the M105 gives its maximum output. The reference axis is usually the same as the preferred listening axis. I don't think it make a lot of sense to have a moving reference axis that depends on the frequency.

 

[witwald]

I used Amir's NFS data to create the polar directivity plot.

Thank you very much for going through all that trouble. I appreciate your efforts in making all of this extra data available. It's really great and helpful.

Below is an animated one showing the directivity plots from 20 - 1000 Hz. The data from the NFS showed that the radiation at the lowest frequencies weren't omni-directional, but started becoming so by around ~70 Hz. I haven't dug into the reason why.

That is a curious result indeed. I suspect that the NFS algorithm may be having some difficulties in fitting the spherical harmonics to the raw data.

In any case, by 101 Hz we get a really good omnidirectional result:

while the result at 35 Hz looks decidedly misshaped:

For the measurements of the M105, I assume that the Klippel Near-Field Scanner (NFS) was set up with the vertically-aligned axis of rotation more or less placed at the center of the front baffle.

At 101 Hz, the main contribution to the acoustic output would be from the woofer. Measurements to the front and to the rear would be equidistant from the woofer. Hence, we could reasonably expect a very good circle for the response plot, and a well-centered circle is what we get.

Based on the M105's impedance curve, the port resonance is at around 60 Hz. The polar sound pressure response plot is shown below. It also is very circular, with just a slight emphasis towards rear-facing radiation. As the contribution from a woofer at the port resonance frequency is usually quite low, this shows that the port seems to be an omidirectional radiator. The results are a mirror image of each in the 0°–180° axis.

The reference axis is usually the same as the preferred listening axis.

It makes sense to do that when the main radiation axis is front-facing. However, at lower frequencies where omnidirectionality is the expected behavior for a monopole system, alignment with the dominant radiation axis also makes sense, as it will give an indication of any directionality that is actually there.

 

[Soniclife]

The data from the NFS showed that the radiation at the lowest frequencies weren't omni-directional, but started becoming so by around ~70 Hz. I haven't dug into the reason why.

Is it possible we are looking at some out of phase cancellation between the port and and driver?