KEF LS50 Meta, LS50 Wireless II User Manual

KEF R&D

LS50 Meta

LS50 Wireless II

CONTENTS

CONTENTS CONT’D

Introduction

Philosophy

LS50 Meta

Tweeter

Metamaterials

Tweeter Metamaterial Absorption Technology

Coupling the Absorber to the Tweeter Dome

Tangerine Waveguide

Tweeter Gap Damper

Motor Design

Bass/Midrange

Motor Design

Crossover

Industrial Design

Specication - LS50 Meta

Appendix 1 - Cabinet

Diffraction

Rigid Bracing/Constrained Layer Damping

Appendix 2 - Ports

Resonances inside the cabinet

Organ pipe resonances

Turbulence

Appendix 3 - Uni-Q

Tweeter

Diaphragm

Tangerine Waveguide

Bass/Midrange driver

Diaphragm

Z-ex Surround

LS50 Wireless II

Overview

Inputs

Connection between Primary and Secondary

Streaming support

DSP Processing

EQ Settings

Desk & Wall Modes

Treble Trim

Phase correction

Bass Extension

Adding a Subwoofer

Syncing with Vision

Firmware

Power Amplication

Summary

Specication - LS50 Wireless II

Summary

References

Introduction

“If I have seen further, it is because I stand on the shoulders of Giants.”

This famous quotation, attributed to Sir Isaac Newton,

illustrates how progress is made in virtually all

scientic disciplines - progress is evolutionary, rather than revolutionary. So it is in the development of

speakers and techniques developed for one model are carried through to subsequent designs and added to.

That small loudspeakers could be serious hi-

reproducers was proved in the 1960s with Laurie

Fincham’s Maxim design for Goodmans. Grossly inefcient, with a cone tweeter and rudimentary

crossover, this speaker was outdated by the 1970s, but nevertheless inspired the BBC to design the

larger, but still diminutive LS3/5. Originally designed

for ¼ scale auditorium acoustic modelling, this design was soon recognised for its suitability as a monitor in the cramped surroundings of BBC Outside Broadcast vans. The design was subtly changed to become the

legendary LS3/5A when the KEF drive units, with which the speaker was equipped, were modied.

As time went by and loudspeaker design progressed, it was again time to review the capabilities of the

LS3/5A and in 2012, to mark the 50th anniversary of KEF, the original LS50 was born. As the name suggests, the LS50 Meta is a development of the original LS50, and again redenes what a small monitor loudspeaker

can do.

These words set the company’s philosophy from day one and they are as relevant today as they have ever been. Continuing research into materials and

engineering has always driven KEF’s quest for

innovation, with the added element of ensuring that the resulting designs work within a typical domestic environment. Everybody deserves great sound.

LS50 Meta

Both LS50 Meta and LS50 Wireless II feature a 12th

Generation Uni-Q driver array featuring an industry-

rst - Metamaterial Absorption Technology (MAT).

This particular Uni-Q driver, whilst being inspired by

that of the R Series, has been engineered with regards

to the overall system design.

Firstly, the addition of MAT has necessitated considerable redesign of the driver structure, allowing

KEF engineers to further rene technologies such as

the tweeter gap damper.

Secondly, where R Series features 3-way designs, LS50

Meta is a 2-way loudspeaker, meaning the midrange driver must cover a much wider bandwidth, down into

bass frequencies. Whilst excellent for integration, this

does provide challenges in regards to performance.

With this in mind, KEF engineers made a number of

changes to the motor system, suspension, surround and cone of the midrange driver to deliver as smooth a performance as possible.

Tweeter

LS50 Meta - Carbon Black

Figure 1

Philosophy

“Of all art, music is the most indenable and the

most expressive, the most insubstantial and the most immediate, the most transitory and the most imperishable. Transformed to a dance of electrons along a wire, its ghost lives on. When KEF returns music to its rightful habituation, your ears and mind, they aim to do so in the most natural way they can…

without drama, without exaggeration, without artice.”

Raymond Cooke, KEF founder

With any drive unit, as much sound is generated at the

rear of the unit as at the front (see gure 2) and this

radiation is unwanted and needs to be absorbed.

Figure 2 - Generic driver

The most important innovation in this loudspeaker is the near-perfect absorption of the unwanted rear sound generated by the tweeter dome. One might legitimately ask, “Why is this important?”

It’s true that, if the tweeter is listened to on its own

without the contribution of the bass/midrange driver

- and here we are talking of frequencies above about 3kHz - very little seems to come out of it. More revealing is to listen to any loudspeaker system

without the tweeter connected. The sound is mufed

and, although one can discern that it’s a piano or a violin playing, it’s impossible to hear the difference

between a Steinway and a Bosendorfer, a Stradivarius

and an also-ran violin.

The true music lover or audiophile wants to hear this amount of detail. More than that, one should be able to imagine the striking of the keys or the way the bow

is stroking the strings. In short, it should be the closest

one can get to a live performance without actually turning up.

the mathematical expression between the target

absorption spectrum and the physical realisation, which has a minimum thickness only 1/10th of the

wavelength at the lowest frequency (620Hz in this case).

Tweeter Metamaterial

Absorption Technolology

Figures 3 and 4 respectively show a computer model

of the absorber and the real thing. Its design borrows

from a room acoustics absorber in that it is based around ¼ wavelength resonators. Looking rather like a maze, what you can see are 15 separate tubes,

folded up to make a circular object. In fact, there are 2

layers, so there are, in total, 30 tubes, each tuned to a different frequency.

Each tube has a very high Q, which means high absorption over a very narrow band, and the only damping, which is the way the energy is dissipated, comes from the friction between the air moving in the tubes and the tube walls. The frequency spacing is designed so that, when the effects of all the tubes are combined together, the absorption is uniform over a wide band.

Figure 5 shows the pressure response, measured at the closed end of each of the 30 channels, together with the absorption spectrum of the whole absorber.

This sort of detail comes from the tweeter and the better the tweeter, the more the enjoyment of listening to the music.

Metamaterials

Metamaterials are probably most familiar in the eld

of optics, where synthetic materials may be realised that have properties that cannot be found in nature.

For example, a base material may be infused with

another in varying density such that the refractive

index varies throughout the material. Thus things like at lenses may be constructed that are much easier to

produce than grinding glass to a precise shape.

The term “meta” has since gained the more general

description of any material that exhibits characteristics foreign to the solid form. In this case, ABS has been

moulded into a shape that is an almost ideal broadband absorber.

The design of the absorber was a joint project between

KEF and AMG (Acoustic Metamaterials Group). With the help of Professor Ping Sheng (a worldrenowned expert on metamaterials), AMG derived

Figure 3

Computer model of absorber

Figure 4

Actual absorber

Top: pressure response at closed end of each separate absorber channel.

Bottom: absorption spectrum of whole absorber.

Figure 5

Figure 7

Top: Amplitude and Phase responses of transfer functions

V1 (green), V2 (blue) and V1+V2 (red).

Bottom: Amplitude and Phase responses of input impedance

These are steady-state measurements and anyone

who is familiar with lters will know that, although the amplitude response of narrow-band lters may

well add to unity, this usually hides the presence of allpass phase responses and some time smearing occurs. The system is said to be non-minimum phase. This is the case with the summing of the transfer functions of the separate channels of graphic equalisers and the outputs of two or more drivers fed by a passive

speaker crossover network. Such time smearing can

readily be shown by impulse measurements, where a

Generalised 3rd-order Butterworth constant-resistance

Figure 6

crossover. Units are Ohms, Farads and Henries. fc is cut-off

2 3

change in prole between the exciting signal and that generated by the device(s) indicates time smearing.

Coupling the Absorber to the

Tweeter Dome

The absorber sits at the rear of the Uni-Q™ driver and is coupled to the tweeter dome by a slightly tapered conical duct, which acts as a waveguide. This waveguide passes through the centre poles of both the tweeter and bass/midrange drivers and has involved a complete redesign of the tweeter magnet assembly to accommodate the wider diameter required for the duct to work properly. The difference in the motor

assemblies is shown in gure 9.

not being totally absorbed. It exposed a lesser, but still important layer of residual colouration. It turned out

that the tangerine waveguide and the dome surround support - both plastic mouldings - were physically

deforming at high frequencies. (gure 11)

Tweeter Gap Damper

One of the problems with constructing a combination driver array like Uni-Q is dealing with the gaps that separate the constituent parts. There is a narrow channel – an annular gap – between the moving midrange voice coil and the static start of the tweeter waveguide. This channel acts as an organ-pipe-like

resonator, and is excited by the tweeter output. The

resonances modify the response of the tweeter, adding a series of glitches that are not present if the gap is closed off – simulating a perfectly smooth waveguide.

Impulse response of tweeter dome velocity (red), overlaid with

Figure 8

ideal response (dotted blue)

However, it must be realised that what is being used here is the acoustic impedance, not a transfer function. To take an electrical analogy, consider a passive,

constant-resistance network such as that of gure 6.

While the individual low- (V1) and high-pass (V2)

transfer functions are minimum phase, the sum of the

transfer functions (V1+V2) is not. It is a 2nd-order allpass. However, the total input impedance (Zin), is a

pure resistance, R. As such, it is minimum phase. There

is no all-pass (gure 7).

To illustrate that the absorber is indeed minimum-

phase and introduces no time smearing, gure 8 shows

the impulse response of the tweeter dome movement overlaid with an ideal, minimum phase response. There is virtually no difference.

The tweeter dome movement was chosen because it can be measured close-to, which increases the signal

to noise ratio. As it is only valid well below the rst

break-up mode of the dome and the absorber only works above a certain frequency, both signals were equally band limited to avoid errors.

A small amount of porous material is placed in the duct, which has the dual effects of reducing the amount of

ripple at high frequencies and ne-tuning the knee

of the absorption spectrum. Figure 10 shows the absorption spectrum immediately behind the dome.

Figure 9

LS50 (left) and LS50 Meta (right) tweeter motor assemblies.

Figure 10

Absorption at the entrance of the conical duct, immediately

behind the dome diaphragm.

Tangerine Waveguide

Many acoustic engineers would be so pleased at designing the almost perfect tweeter absorber that they would have rested on their laurels. But speaker design is rather like peeling an onion - remove one

layer and there is another one exposed. So it is with the

removal of colouration caused by the rear radiation

FEA simulation of exaggerated deformation of tangerine

Figure 11

waveguide and surround support at 12kHz

Strengthening ribs were added to both components,

which reduced the deformation. Figure 12 shows

the actual modied parts (viewed from the rear) and gure 13 illustrates the reduction in displacement. It should be noted that the modied parts have an

area comparable to the tweeter dome itself and any movement will add audible colouration to the overall.

For the basic operation of the tangerine waveguide,

featured in previous models, see Appendix 3.

Figure 12

Modied surround support and tangerine waveguide viewed

from the rear to show added ribs

Figure 13

Simulated displacement of tangerine waveguide

Obviously, this annular gap is necessary to allow the LF/MF cone and voice coil to move, so the solution was to create a cavity between the midrange and tweeter

magnets to which this annular gap connected (gure

14). Adding damping to this newly-created cavity was

found to be effective in taming the resonances in the annular gap and the removal of the response glitches was immediately apparent as an improvement in

detail clarity (gure 15).

This feature - the tweeter gap damper - was rst introduced through the R Series (2018), and was

a major development that led to birth of the 12th Generation Uni-Q. The addition of MAT and the

Figure 14

Uni-Q driver with damped, optimised gap

change in tweeter system structure for LS50 Meta,

however, necessitated a redesign of the tweeter gap damper. Additional work was carried out in regards to the shape of the cavity and the placement of the wadding material - now comprising of two rings - to further improve performance.

4 5

+ 11 hidden pages