EXHIBIT 2012
`
`EXHIBIT 20 1 2
`
`

`
`1512 (D-4)
`
`A BANDPASS LOUDSPEAKER ENCLOSURE
`
`L.R. Fincham
`KEF Electronics Limited
`Maidstone — England
`
`Presented at
`
`the 63rd Convention
`May 15 through 18, 1979
`Los Angeles
`
`A u o I o
`
`W
`
`This preprint has been reproduced from the author’: advance
`manuscript, without editing, corrections or consideration by
`the Review Board. For this reason,
`there may be changes
`should this paper he published in the Journal of the Audi'o
`Engineering Society. Additional preprints may be obtained by
`sending request and remittance to the office of Special Puiblil
`cations, Audio Engineering Society, 60 East 42nd Street, New
`York, New York 10077, USA.
`< Copyright 1979 by the Audio Engineering Society. All rights
`reserved, Reproduction of this preprint, or any portion thereof,
`is not permitted without direct permission from the office of
`the Journal of the Audio Engineering Society.
`
`AN AUDIO ENGINEERING SOCIETY PREPRINT
`
`THX Ltd. Exhibit 2012 Page 1
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 1
`IPR2014-00235
`
`

`
`A BANDPASS LOUDSPEAKER ENCLOSURE
`
`L. R. FINCHAM
`
`KEF Electronics Limited
`Maidstone ~ England
`
`Abstract
`
`A loudspeaker enclosure whose acoustic response has a
`second order bandpass filter characteristic is described.
`The theoretical trade off between bandwidth. efficiency
`and box volume is given.
`
`It is shown that when the system is fed through a suitable
`first order bandpass electrical filter. a third order
`acoustical bandpass characteristic can be achieved which
`is of practical value in certain sub-woofer applications.
`
`Introduction
`
`Recently there has been a renewed interest in the design
`of so called sub—woofer loudspeaker systems which are used
`to augment the low frequency response of very small
`enclosures.
`These systems normally cover a range of
`1-3 octaves and are fitted with a low pass filter whose
`upper cut off frequency is usually less than lO0Hz.
`The design of this low pass filter, using passive
`components,
`is often difficult because in practice the
`filter is not
`terminated by a pure resistance but by a
`loudspeaker system, whose input impedance varies
`significantly with frequency. particularly close to its
`resonance frequency.
`Even when satisfactory results
`are achieved,
`the resulting network requires large
`values of inductance and capacitance which make it
`both bulky and expensive. This problem can be avoided,
`at the expense of some additional complication and cost
`by using active or passive low level dividing networks
`placed ahead of separate power amplifiers for the sub—woofer
`and satellite systems.
`
`This paper describes an alternative approach in which
`a second order bandpass response is achieved acoustically,
`and by the addition of a single first order bandpass
`electrical filter an overall third order response shape
`is obtained.
`
`THEORY
`
`The total acoustic output from a reflex or vented enclosure
`is given by the vector sum of the separate contributions
`from the drive unit and the vent. and this is shown for
`a lossless B4 alignment
`in figure 1.‘ It can be seen that
`the output from the vent has a second order bandpass
`characteristic, with its centre frequency equal to f
`resonance frequency of the vented enclosure.
`
`the
`
`3,
`
`If the output from the drive unit is isolated from that of
`the Vent.
`then the resulting system has a natural bandpass
`characteristic whose response exceeds that of the original
`reflex enclosure below f
`.
`The general arrangement for such
`a system is shown in figure 2,
`together with its impedance
`type acoustical analogous circuit.
`The vent has been replaced
`by a passive radiator which, although it does not significantly
`affect the theoretical analysis, has certain advantages in a
`practical design.
`The system is seen to be a modified
`reflex enclosure of volume V 2 in which an additional enclosure
`of volume VB
`has been placed over the rear of the drive unit
`so that sound is only radiated from the passive radiator.
`
`Analysis
`
`For the purposes of analysis the following simplifications
`are made:
`
`the acoustic losses in the passive radiator and
`2.
`and R
`R
`cavity 2,A%re assumed negligible.
`
`other circuit resistances are combined to give a single
`resistance, RAT
`
`BI 11
`
`‘1)
`Rg + RE) 53 * Rns + RAB1
`"here RAT =
`is assumed
`The acoustic compliance of the passive radiator C
`to be very much greater than C
`,
`the acoustic cdfipliance
`of cavity 2, and may be neglectgd.
`
`CAS and CAB1 are combined to give a single compliance CAT
`C
`C
`AS A31
`CAS * CABl
`
`where CAT :
`
`(2)
`
`The simplified circuit is shown in figure 3.
`
`* Ref.
`
`1
`
`THX Ltd. Exhibit 2012 Page 2
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 2
`IPR2014-00235
`
`

`
`Before analysing the circuit, it is useful to define
`a number of system parameters.
`
`the free air resonance frequency of the drive unit
`f ,
`ig given by
`
`fs = 1/2w
`
`the volume of air having the same acoustic compliance
`,
`V
`aésthe drive unit suspension is given by
`=
`1
`V
`90° CA5
`
`AS
`
`(3)
`
`(4)
`
`is the density of air in kg/m’ and C is the velocity
`where o
`of souna in m/s
`
`Letq=£=_E
`
`the resonance frequency of the drive unit when loaded
`f 1,
`Bl
`b§ the volume of cavity 1, V
`is given by
`
`fcl = 1/ZHJEREEXE
`
`from equations (3) and (6)
`
`fcl = fs
`
`1 + gfigl
`
`%
`
`or fcl = ES (1 + a)%
`
`(5)
`
`(6)
`
`(7)
`
`the acoustic mass of the passive radiator is arranged
`,
`M
`tgpresonate with the acoustic compliance of cavity 2, C
`2:
`at a frequency f
`.
`In order that the bandpass characté¥istic
`may be symmetricgl
`fC2 is made equal to fcl.
`
`Hence EC2 = fcl = l/ZWJMAPCAB2
`
`aT.
`
`the total compliance ratio is defined from equations
`(6)
`(2) and (8) by
`
`G
`
`T
`
`___ E _ CAT
`“As
`CABZ
`
`7 avg
`CAB2
`
`1
`(1 + “)
`
`_ 3 -
`
`(8)
`
`(9)
`
`V
`
`AS
`_
`°‘ “T ‘ VB;
`
`V
`c
`_ B1
`1
`(1 + u) ‘ V52 (1 + a)
`
`(10)
`
`when the acoustic compliance of the drive unit, CA . is
`negligible compared to the acoustic compliance of gavity 1,
`CAB1,
`then equation (10) simplifies to
`
`(11)
`
`(12)
`
`(13)
`
`V
`o:T s V—‘“—
`B2
`
`Then the total system volume VT = VB1 + VH2
`
`_
`— VB1
`
`l
`1 + G
`
`The total Q of the drive unit mounted in cavity 1,
`QTcl,
`is given by
`1
`
`°'rc1 = 2n£C1cM_RAT
`
`Freguengy Response
`
`from
`The total system response is due only to the output
`the passive radiator, and the sound pressure at a distance r
`is given by
`
`[prl = E? ‘Up!
`
`where Up is the volume velocity of the passive radiator
`
`It is convenient to define a reference sound pressure lprlref
`
`(15)
`
`THX Ltd. Exhibit 2012 Page 3
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 3
`IPR2014-00235
`
`

`
`is the reference volume velocity of the drive unit
`IUDI
`in cgsfty 1, with cavity 2 and the passive radiator removed,
`at a frequency where m’M’
`>> R‘
`i.e.
`the drive unit is
`mass controlled
`AS
`AT
`
`pg
`Iublref ‘ mg
`
`e E1
`
`where p
`
`= ———3——-T~
`(R9 + RE SD
`
`g
`
`The frequency response of the system is given by
`
`|Pr,
`pr ref
`
`:
`
`lu l
`UD ref
`
`Substituting for IUDI ef and determining the value of
`Up from the circuit ii
`figure 3 gives
`
`lp,1
`pr ref
`
`= Zvffihs
`pg
`
`pa
`2nfMAP
`
`l: +
`
`1
`—
`Y _ ;
`
`where Y =
`
`fCl
`f
`——— — ———
`fci
`f
`
`1
`
`:“T
`
`and noting that q in a bandpass filter corresponds to
`Q in a low pass filter. and is a measure of the degree
`of peaking before cut off.
`(Ref.
`3 p.406,7
`
`the Q of the series tuned circuit in figure 3
`Q C ,
`1; felated to q
`q
`
`where QTC1 = 7;;
`
`(15)
`
`(17)
`
`(13)
`
`(19,
`
`(20)
`
`(21)
`
`From equations (9 )
`
`(19) and (21)
`
`=
`
`lp I
`I
`pr ref
`
`M
`A5
`MAP
`
`l
`°TQrc1
`
`i
`
`*
`+
`
`_ A I
`V
`
`Y
`
`= 1
`[IT
`
`1
`a Q
`T TC1
`
`v - —
`Y
`
`%
`
`5
`
`+
`
`*1
`
`is plotted in figure 4 for various values
`Equation (23)
`of a , for a maximally flat or Butterworth response,
`wherg q = 0.707
`
`i.e. Q§C1uT = 0.5
`
`At f = f
`
`Cl’
`
`Y = 0
`
`P
`“d J'—p—‘+
`r ref
`
`=
`
`i—
`T
`
`(22)
`
`an
`
`(24)
`
`(25:
`
`At the upper and
`Y = 1 and the res
`passive radiator
`From equation (20
`
`and EL,
`f
`lower cut off frequencies,
`ponse is -3dB with respect tg the
`output at fcl for a Butterworth response.
`)
`
`E
`fcl
`
`-fc1
`f
`
`2 i /E; at EH and fL
`
`and f
`
`Cl
`
`: if f
`H L
`
`Hence fig
`fL
`
`=
`
`fciI
`fL
`
`=
`
`ff;
`F-c1
`
`From equations (2
`
`6) and (28)
`
`153::-H
`f_
`H:["212
`
`=
`
`fC1
`fL
`
`:
`
`
`“E; *
`“T * 4
`2
`
`_ 6 —
`
`(26)
`
`(27)
`
`(28)
`
`(29)
`
`THX Ltd. Exhibit 2012 Page 4
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 4
`IPR2014-00235
`
`

`
`for a Butterworth response are given
`Values of Q
`in table 1 ?S§ various values of a ,
`together with
`bandwidth ffl/fL, relative sensitivity and total box
`volume VT.
`
`The relative sensitivity is expressed in dB as
`
`spl of double cavity system at f=fCl
`spl of drive unit with cavity 2 removed at f>>fCl
`
`n
`
`n
`
`rel’
`
`the relative efficiency is defined as
`
`re
`
`of band ass s stem
`1 Efficien
`Efficiency of equivalent closed box system
`I
`
`E = fcl
`double cavity at
`pr
`°r “rel = 7nTmf1’
`
`(30)
`
`It can be shown from equations (l2),(25),(29) and [eq.(24)Ref.4]
`7
`(/5; + ‘“T + 4)
`1
`1 + J1+4/am I
`nrel' —s"rrI71:Wq
`' UT/Q
`2
`
`‘31’
`
`Rel. Sensitivity Total Volume
`20log10aT
`
`: 20 1ogloaT
`
`Table 1
`
`Q
`
`T
`
`0.5
`
`l.0
`
`2.0
`
`3.0
`
`4.0
`
`Q
`
`TC1
`
`l.0
`
`0.71
`
`0.5
`
`0.41
`
`0.35
`
`Bandwidth
`fH/fL
`
`2.0
`
`2.62
`
`3.73
`
`4.79
`
`5.85
`
`n el is evaluated for various values of QT and shown
`ih table 2. These values of n
`represent the maximum
`values obtainable and n
`will
`n practice be somewhat
`lower due to losses in Efié acoustic lining in V
`which
`have been ignored in the analysis for the sake Sf
`simplicity and clarity.
`
`Table 2
`
`mm (max)
`
`lologmnrel
`
`increases.
`From Table 2 it can be seen that as a
`and the system bandwidth increases, sg the relative
`efficiency decreases. There is still a useful
`theoretical gain of 70% for s
`= 2 where the bandwidth
`.
`T
`is nearly 2 octaves.
`
`THX Ltd. Exhibit 2012 Page 5
`|PR2014-00235
`
`
`
`+6dB
`
`0dB
`
`-6dB
`
`—9.5dB
`
`—l2dB
`
`increases the
`From Table 1 it can be seen that as a
`system bandwidth also increases at thg expense of lower
`relative sensitivity.
`
`Efficiencx
`
`to compare the efficiency of this bandpass
`It is useful
`system with that of a closed box system having the
`same lower cut off frequency and box volume and a
`Butterworth high pass response.
`
`THX Ltd. Exhibit 2012 Page 5
`IPR2014-00235
`
`

`
`Third order response with Auxiliary Electrical Filter
`The electrical equivalent circuit of the double cavity
`bandpass system fed through a first order bandpass
`electrical filter is shown in figure 5, and has been
`derived from the circuit given in figure 3 using the
`relationship
`(B1)‘
`
`is the impedance of an element in the impedance
`where Z
`type acéustical analogous circuit, and Z
`is the
`impedance of the corresponding element in the electrical
`equivalent circuit.
`
`In figure 5
`
`CMES corresponds to MAS
`
`LCET corresponds to CAT
`
`RES corresponds to RAS + RABl
`CEB2
`
`corresponds to CAB2
`
`L
`
`EP
`
`C
`
`corresponds to MAP
`
`REL2 corresponds to RAL2
`
`is an additional element representing the
`where R
`acousti§L%esistance of leakage losses in cavity 2.
`
`can be
`For the purposes of analysis, or simulation, RA
`adjusted to provide a convenient frequency-invariant
`approximation to the actual acoustic losses in cavity 2
`due to absorption R
`2,
`leakage R
`, and the passive
`radiator suspensionA§AP, which inA%§actice all vary with
`frequency (Ref. 5).
`
`L and C are the external inductor and capacitor used to
`provide the first order bandpass electrical filter and
`are chosen so that
`
`LC : LCETCMES I Lcsszcnsp
`
`1
`_
`— V
`
`(32)
`
`Again for simplicity this circuit is analysed assuming
`that RE
`= ”
`and RE 2
`= O.
`In practice of course,
`these gsses cannot
`Q ignored and their effect will be
`included in system simulations shown later in the paper.
`
`Analysing the circuit shown in figure 5, shows that for
`a Butterworth response
`_
`4
`0m - Wm
`The external electrical components may be calculated from
`Series inductance, L = LCEB2
`3
`
`Series capacitance, C = BCMP
`
`The new —3dB frequencies EH‘ and fL' are given by
`
`f
`f
`———
`C1
`
`fci
`f
`— —-
`
`= 1 ¢2a
`
`T
`
`at r
`
`L
`
`' and r
`
`H
`
`'
`
`Hence
`
`.
`
`k
`
`=
`
`H
`fL
`
`.
`
`fa
`fcl
`
`=
`
`fci
`L
`
`=
`
`r-
`2“T * ‘2“T * 4
`2
`
`As before {see page 7),
`as follows
`
`the relative efficiency is defined
`
`nre1' Efficiency of third order bandpass system
`Efficiency of equivalent closed box system
`It can be shown that
`
`.
`
`nrel
`
`_
`
`3/5
`
`1
`
`' T TTTT/‘E71. [o
`
`1 + /1 + 2/
`
`’
`
`‘33’
`
`(34)
`
`(35)
`
`(35)
`
`(37)
`
`(W
`
`for a third order Butterworth
`The required values of Q
`response are given below in Table 3 for various values of cT,
`I
`together with the resulting bandwidth fH'/fL', and the
`relative efficiency,
`n
`rel
`
`Bandwidth
`
`fH'/fL
`
`Comparing the values shown in table 3 with those given previously
`in tables 1 and 2, it can be seen that the third order bandpass
`system is always more efficient than the second order, and
`increasingly so for higher values of aT.
`For g
`= 3 the bandwidth is nearly 3 octaves. and the system has
`a thegretical efficiency of more than twice that of the
`equivalent closed box.
`
`_l0-
`
`THX Ltd. Exhibit 2012 Page 6
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 6
`IPR2014-00235
`
`

`
`EXPERIMENTAL RESULTS
`
`A third‘order bandpass system having a total internal
`volume of 65 litres was constructed from 300mm diameter
`cardboard tube having a wall
`thickness of 12mm.
`
`System Parameters
`
`cavity 1
`
`lined with bonded acetate fibre wadding
`
`Val
`
`fcl
`
`QTCI
`Drive unit
`
`area, SD
`
`29 litres (300mm diameter x 400mm long)
`
`44Hz
`
`0.98
`0.022m’
`
`35 litres (300mm diameter x 480mm long)
`VB2
`Passive radiator D.033m’
`area
`
`Total compliance
`ratio, QT
`Series L
`
`0.65
`
`11mH
`
`Series C
`
`1200uF
`
`Freguengy Response
`
`The frequency response of the bandpass system was measured
`using the nearfield sound-pressure method (Ref. 6) and
`also using the equivalent method of measuring the
`acceleration of the passive radiator diaphragm by means
`of a miniature accelerometer bonded to its front face.
`The equivalence and accuracy of these two methods was
`confirmed by means of free field pressure measurements
`taken at a distance of 1 metre out of doors with the
`system mounted on a small platform raised 10 metres above
`the ground.
`
`The frequency response of the system with no lining in
`cavity 2
`is shown in figure 6,
`together with the Calculated
`response. There is very poor agreement above l4OH2.
`The assumption in figure 3 that the compliance of cavity 2
`can be represented by a single capacitance C
`2,
`is only
`justified for frequencies where the wavelenggg of sound
`is greater than sixteen times the length i.e. where
`f < 45Hz (Ref.
`2 pl29LCavity 2 may be represented by a
`series mass and compliance for frequencies where the
`wavelength of sound is greater than eight times the
`smallest dimension i.e.
`f <
`l44Hz (Ref.
`2 p217).
`
`_
`
`11...
`
`It can be seen from the measured response in figure 6 however,
`that there is still significant output above l44Hz with
`high-Q resonant peaks at 330Hz, 66OHz and 990Hz, so an even
`more complex representation of cavity 2 seems to be required.
`A quasi-distributed impedance—type acoustical analogous
`circuit for cavity 2 is likely to be of the form shown
`in figure 7.
`The system frequency response has been
`re-calculated from the simplified circuit shown in figure 3.
`using the distributed representation for cavity 2
`in place
`of the simple capacitor and this is shown,
`together with
`the corresponding measured response,
`in figure 8.
`
`In an effort to damp out these resonances, cavity 2 was loosely
`filled with 300mm discs made from 50mm thick glass fibre.
`The absorbent material was selected for its high absorption
`at frequencies above l50Hz and it was hoped that by allowing
`the discs to move at
`low frequencies that the loss of efficiency
`of the system in its pass band would be minimised.
`
`is shown
`The system response, with glass fibre discs in cavity 2,
`in figure 9 and was obtained using a constant sinusoidal input
`of 6V rms.
`The response shape appears nearly ideal, having a
`flat pass band with -3dB frequencies at 26 and 76Hz, and a
`smoothly falling characteristic above 14082,
`the previous peaks
`having been damped out by the lining.
`The system response was
`then rechecked,
`this time using the fast Fourier transform to
`calculate its frequency response from a digital measurement
`of its impulse response taken in the nearfield (Ref. 7).
`To facilitate comparison,
`this is shown super-imposed upon
`the analogue measurement in figure 9.
`
`It can be seen that the curves differ considerably below
`50Hz.
`the digital measurement peaking by ldB at 4OHz before
`cutting off nearly 1/3 octave earlier at 3282. Removing the
`lining from cavity 2 and remeasuring the response using both
`sinewave and impulse excitation revealed no difference, so it
`was deduced that the discrepancy must have been caused by the
`lining.
`It was suspected that at
`low frequencies the sinewave
`excitation was high enough to overcome the static friction
`between the glass fibre discs and the inside of the tube.
`so that part of the lining moved with the passive radiator
`thereby adding to its effective mass.
`The impulse excitation
`however, had insufficient energy to overcome the static
`friction, so that the effective moving mass of the passive
`radiator was lower, causing mistuning of the system.
`This
`hypothesis was later confirmed, for when the glass fibre discs
`were constrained by interleaving them with fixed wire grids,
`the response curves, measured by both methods agreed, as
`shown in figure l0.
`
`THX Ltd. Exhibit 2012 Page 7
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 7
`IPR2014-00235
`
`

`
`Given in figure 11 is the complete electrical equivalent
`circuit which uses the quasi-distributed representation
`of cavity 2,
`including its lining, and contains an
`r
`additional electrical element L
`corresponding to C
`the acoustic compliance of the Eggsive radiator suspeggion.
`Finally the calculated and measured response curves
`for the complete third order double cavity system. given
`in figure 12,
`show excellent agreement from 20—SO0Hz.
`
`Conclusions
`
`The theory and design of a third—order bandpass loudspeaker
`system which employs only a first-order bandpass electrical
`filter has been described. Theory shows that even for
`bandwidths up to three octaves,
`this system still has a
`pass-band efficiency of more than twice that of the
`equivalent closed-box having the same lower cut—off
`frequency and total internal volume.
`
`It has been shown that the usual representation of the
`internal volume of an enclosure by a simple capacitor in
`the impedance—type acoustical analogous circuit fails to
`predict the presence of high Q resonances, which in practice
`appear at the output of the bandpass system at frequencies
`where the wavelength of sound is less than eight times
`the smallest dimension of the second cavity.
`
`These resonant peaks, although usually obscured for the
`reasons given below, have been observed before,
`in
`experimental nearfield measurements made on vented
`loudspeaker systems,
`the discrepancies between the
`measurements and the theoretical predictions for the vent
`output being incorrectly attributed to cross talk from the
`drive unit
`(Ref. 6).
`
`the general vent output at high
`In a vented enclosure,
`frequencies is usually sufficiently below that of the drive
`unit that these high Q resonances are unlikely to be detected
`from a visual examination of the steady—state amplitude
`frequency response characteristic, although their presence
`often imparts a distinct colouration to the perceived sound
`quality. This colouration can only be adequately attenuated
`by using so much internal lining material that the overall
`low-frequency efficiency of the system is so impaired that
`many of the theoretical advantages of reflex loading are negated.
`
`Attempts to restore the low frequency efficiency by allowing
`the internal lining to move can result in a form of dynamic
`non-linear distortion,
`in which the frequency response shape
`changes significantly with level. Transient stimuli,
`such as bass drums and tympani, will then appear to have a
`restricted low frequency response with some peaking, resulting
`in overhang, whereas nearly steady state signals such as
`organ pedal notes will not reveal this problem.
`
`_l3-
`
`Acknowledoement
`
`The author gratefully acknowledges the many theoretical
`and practical contributions made by P J Baxandall
`(Audio
`Consultant), E Cecconi, M E Gough, and C J Moore (KEF
`Electronics Limited)
`to this project.
`
`References
`
`l.
`
`J E Benson “Theory and design of loudspeaker enclosures
`Part
`1 — Electrical relations and generalised analysis"
`A.W.A. Tech. Rev. Vol.l4 No.1 p.47 (1968)
`
`L L Beranek Acoustics McGraw-Hill, New York:
`
`l954
`
`F E J Girling and E F Good "Active Filters Part 2 — basic
`theory: lst and 2nd order responses" Wireless world
`Vol.75 No.1407 pp.406,7 (1969)
`
`R H Small "Closed box loudspeaker systems Part 1 — Analysis"
`Journal of the Audio Engineering Society Vol.20 No.10 p.802 (1972)
`
`R H Small "Vented-box loudspeaker systems Part l - small
`signal analysis“ Journal of the Audio En ineerin Societ
`Vol.2l No.5 p.367 (19735
`
`D B Keele "Low-frequency loudspeaker assessment by
`nearfield sound-pressure measurement" Journal of the
`Audio Engineering Society Vol.22 No.3 p.154 (1974)
`
`J M Berman and L R Fincham “The application of digital
`techniques to the measurement of loudspeakers" Journal of the
`Audio Engineering Society Vol.25 No.6 (i977)
`
`THX Ltd. Exhibit 2012 Page 8
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 8
`IPR2014-00235
`
`

`
`LIST OF SYMBOLS
`
`Bl
`
`Force factor
`
`Velocity of sound in air
`
`compliance of cavity 1
`
`compliance of cavity 2
`
`ABI
`
`AB2
`
`Acoustic
`
`Acoustic
`
`Acoustic
`
`Acoustic
`
`Acoustic resistance of port or passive radiator
`
`Acoustic resistance of drive unit suspension losses
`
`Drive unit dc resistance
`
`Amplifier output resistance
`
`Effective area of drive unit diaphragm
`
`compliance of passive radiator suspension
`
`Volume velocity of the drive unit in cavity 1
`
`compliance of drive unit suspension
`
`Volume velocity of passive radiator
`
`Total acoustic compliance of drive unit and cavity 1
`
`Open circuit output voltage of amplifier
`
`Volume of cavity 1
`
`Volume of cavity 2
`
`Frequency
`
`Resonance frequency of drive unit in cavity 1
`
`Resonance frequency of passive radiator in cavity 2
`
`Upper cut off frequency of bandpass system
`
`Lower cut off frequency of bandpass system
`
`Free air resonance frequency of drive unit and air load
`
`Acoustic mass of diaphragm assembly and air load
`
`Acoustic mass of passive radiator including air load
`
`Strength of acoustic pressure generator
`
`Acoustic
`
`pressure at a distance r from the loudspeaker system
`
`Total Q of drive unit at f
`
`C1 due to all system resistance
`
`Distance
`
`Acoustic
`
`Acoustic
`
`Acoustic
`
`from sound source to measuring point
`
`resistance of absorption losses in cavity 1
`
`resistance of absorption losses in cavity 2
`
`resistance of leakage losses in cavity 2
`
`_]_5..
`
`Volume of air having same acoustic compliance as
`drive unit suspension
`
`Total volme of double-cavity system (VB1 + V32)
`
`Compliance ratio CA3/CAB1
`
`Compliance ratio CAT/CAB2
`
`Efficiency
`
`Density of air
`
`-16-
`
`THX Ltd. Exhibit 2012 Page 9
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 9
`IPR2014-00235
`
`

`
`fig 1
`Frequency response
`of reflex system
`showing separate
`contributions made by
`drive unit and vent.
`
`D—‘ C)
`
`%‘
`
`Q’
`'3
`-‘.3:
`Ti
`E
`
`0
`
`I
`
`'3q).1
`
`02
`
`6.5
`
`1
`
`2
`
`5
`
`10
`
`Normalised frequency f/f3
`
`fig 3 Simplified analogous circiut for Double Cavity System.
`
`I-‘ O -20
` Cavity 2
`various values ofuT.
`
`Volume VB2
`
`Passive
`Radiator
`
`Drive Unit
`
`Cavity 1
`
`::3::::1_
`
`10
`
`O
`
`%
`
`‘*3
`T —:
`iianO9-!
`
`O
`(V
`
`-10
`
`-20
`
`-30
`
`"‘°o.1
`
`0.2
`
`0.5
`
`1
`
`2
`
`5
`
`10
`
`Frequency
`
`f/ fc 1
`
`fig 4 Buttemorth responses of Double Cavity System for
`
`fig 2 Double Cavity bandpass loudspeaker system and impedance
`tv'De acoustical analogous circuit.
`
`THX Ltd. Exhibit 2012 Page 10
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 10
`IPR2014-00235
`
`

`
`M1+ M2+ M3
`
`C1+ C2+ C3+ C4 * CAB2
`fig 7 Quasi-distributed impedance-type acoustical analogous
`circuit for cavity 2 (without lining).
`
`fig 5 Simplified electrical equivalent circuit of third-order
`Double Cavity loudspeaker system.
`
`I)
`U--1
`E
`r—d
`
`E
`
`0
`
`5
`
`no
`
`%
`
`*3
`I-4
`.2’
`
`30
`
`20
`
`10
`
`0
`
`
`= 0.1 MAP
`
`
`0
`
`~
`
`50
`20
`Frequency
`
`Hz
`
`100
`
`200
`
`500
`
`lk
`
`fig 6 Frequency response of third-order Double Cavity system
`......calculated - using simplified equivalent circuit
`measured - experimental system (cavity 2 unlined).
`
`so
`‘lo 20
`Frequency Hz
`
`100
`
`200
`
`500:1k
`
`fig 8 Frequency response of third-order Double Cavity system
`With CflV1tY 2 Unlined
`- - - - --°31°U15t°d ' Using qU551'di3tTibUted T¢PY¢3€“t3t
`Of C5V1tY 2
`measured — experimental system.
`
`i
`
`°n
`
`THX Ltd. Exhibit 2012 Page 11
`|PR2014-00235
`
`THX Ltd. Exhibit 2012 Page 11
`IPR2014-00235
`
`

`
`.....analogue measurement
`using sinewave
`
`_____digital measurement
`via FFT of measured
`impulse response.
`
`Amplitude pc excitation.
`
`u:0
`
`dB
`
`30
`
`20
`
`10
`
`C1+ C2+ C3 = 0.1 CMEP
`
`R1 to R7 correspond to absorption
`
`Ll+ L2+ L3+ L4 = LCEBZ
`
`and leakage losses in cavity 2.
`
`500
`
`1k
`
`_
`_
`_
`fig 11 Complete electrical equivalent circuit for third-order
`Double Cavity system.
`
`1 4
`
`1-?
`3-v-(
`E
`5
`
`Q 50
`
`40
`
`30
`
`20
`
`10
`
`
`
`100
`
`200
`
`O 20
`50
`Frequency Hz
`.
`fig 12 Frequency response of third-order Double Cavity system
`......ca1culated - using complete electrical equivalent
`circuit
`measured - via FFT of measured impulse response.
`
`500
`
`lk
`
`THX Ltd. Exhibit 2012 Page 12
`|PR2014-00235
`
`50
`020
`Fr¢qUenCY Hz
`fig 9 Frequency response of third-order Double Cavity system
`with unconstrained lining in cavity 2.
`
`100
`
`200
`
`.....analogue measurement
`using sinewave
`excitation.
`
`_____digital measurement
`via FFT of measured
`impulse response.
`
`€50
`o
`E
`:1 #0
`a
`E
`
`30
`
`20
`
`10
`
`0
`
`fig 10 Frequency response of third-order Double Cavity system
`with constrained lining in cavity 2.
`
`S0
`20
`Frequency Hz
`
`100
`
`200
`
`l
`500
`
`1k
`
`THX Ltd. Exhibit 2012 Page 12
`IPR2014-00235