PAPERS

Improvementsin Monitor LoudspeakerSystems*

DAVID SMITH, D. B. KEELE, JR., AND JOHN EARGLE

James B. Lansing Sound, Inc., Northridge,

CA 91329,

USA

As the recording industry enjoys the benefits of both digital and advanced analog recording technology, attention is appropriately focused on the use of compression driver and horn designs which are some 25-30 years old. Evolutionary improvements in woofers, compression drivers, and dividing networks combined with new constantcoverage horn designs have resulted in a frequency response more consistently uniform at all coverage angles (yielding flat power response) along with lowered distortion and increased acoustic power output at the frequency extremes.

0 INTRODUCTION High-quality studio monitor loudspeaker systems have evolved out of the theater traditions of the 1930s and 1940s. These systems, whether of two-, three-, or fourway design, invariably make use of compression drivers in their mid- and high-frequency sections for greater reliability at elevated output levels. These design traditions have been eschewed by the "audiophile" segment of the consumer market, who have in general preferred the relative smoothness and low distortion (at moderate levels) of cone and dome direct radiating systems. As audiophile record productions take on a more conspicuous profile, and as digital recording technology promises higher orders of performance in the studio, we once again examine and attempt to reconcile the apparent differences between that which the dedicated audiophile feels to be a state-of-the-art approach to loudspeaker design--and that which experienced recording engineers require for their specific needs, Among the chief performance parameters we have identified are uniform polar response and directivity, smooth power response, andlowdistortion. Asecondary requirement is for accurate stereophonic imaging at close-in listening positions in the studio control room. A new family ofconstant-directivity horns has formed the basis of a new approach to monitor design, and we will now describe two monitor loudspeakers emb6dying them. * Presented at the 69th Convention of the Audio Engineering Society, Los Angeles, 1981 May 12-15; revised 1983 March 29. 408

I THE CONSTANT-DIRECTIVITY

HORN

Horns used in previous monitors were of the radial type or, alternatively, straight exponential with a horizontally divergent acoustic lens. Either type offered wide but uncontrolled horizontal response at the expense of narrow vertical response at the highest frequencies and constantly rising directivity with frequency. Recent developments in horns (Keele [1], Henricksen and Ureda [2]) yield very uniform vertical and horizontal coverage patterns which change little with frequency. These horns provide surprisingly constant angular coverage along with very stable directivity over their operating range. This alone cures the worst complaints about previous horns and makes systems using such horns truly a new generation of monitors. Optimal horn parameters for monitor use are as follows. 1) Constant coverage angle with consistent polar patterns, both horizontally and vertically controlled over the total operatingrange(1000Hzto 16000 Hz). 2) Coverage angles wide enough to mate at crossover with a cone woofer (90-100 square or a directivity index [DI] of about 8 dB). 3) Faster flare than previously used, for lower secondharmonic distortion. 4) Shorter length to place woofer and horn in the same acoustic plane. The first version of a constant-directivity horn (Keele [1]) was basicallya muchimprovedradial horn with end flaring to combat midrange narrowing and maintenance of high-frequency beamwidth by elimination of the typicalradial horn neck. A secondversiondeJ. Audio Eng.Soc.,Vol.31,No.6, 1983 June

PAPERS

IMPROVEMENTS INMONITOR LOUDSPEAKER SYSTEMS

scribed by Henricksen and Ureda [2] brought vertical angle control down to a lower frequency by flipping conventional horns 90 onto their sides, which allowed a much larger vertical height without the width growing too excessively. Flat surface flares with hard transitions were used for manufacturing ease. A third version by Keele combines the better points of the previous two with several improvements: the vertical flare is fed by a diffraction slot, which can be made narrow enough to feed a wide horizontal angle up to any desired frequency. The primary and end flares have rounded transitions as on the previous Keele horn, but rather than being arbitrarily rounded, the side contours are defined by a three-term mathematical expression: y = a + bx + cxn where b determines the initial wall angle and the cx" term determines the amount of mouth flaring (see Fig. 1). The performance of such a horn is compared to that of a prior art radial horn and of an exponential horn with acoustic lens in Figs. 2-4. Its -6-dB coverage angles and directivity index are quite consistent. 2 THE COMPRESSION EQUALIZATION DRIVER AND ITS

dB per octave above 3 kHz(Newman [3]) would require a horn with a reciprocal increase in directivity index. The compression driver would then be acoustically equalized, but only on axis, as shown in Fig. 5. When this compression driver is loaded by a constantdirectivity horn, the axial response follows the power response of the driver, as shown by Fig. 6. The very high midband efficiency given by horn loading allows passive equalization for both flat axial and, at the same time, flat power response. A typical network configuration is shown in Fig. 7. The midband sensitivity is reduced by the L-pad and dividing network. The highs are shunted across this attenuation by a second-order bandpass filter tuned to the highest operating frequency. Independent control of the midrange and high end are an added benefit. When the high end of the horn-driver combination 'is properly equalized [Fig. 8(a)], we can return the compression driver to the constant impedance plane wave tube and verify that its power response is now flat [Fig. 8(b)]. 3 THE WOOFER Woofers for monitor use have been slowly but constantly evolving over the past 20 years. The ideal woofer must have: 1) A smooth response curve with the required midband sensitivity 2) Controlled directional characteristics 3) High output at low distortion levels

Previous horns were judged on their ability to generate flat response on axis with typical compression drivers. This persisted, even though compression drivers were known to fall off in power response above the midband. A typical driver whose power response rolls off at 6

b B

,t=,z.1 +,x+
'

'r
LO_HER LI_IT 0_: _IIERT.

9.o +

,,
,/'

/
/ LO_IER LII_IT _O_Z.. 1BE. ll_tli_'tl 01['I'1i%5 t.z) l

'

11;

Fl'lEO

l_l_NtO'l['li.A _ Tt_t_

C01Nr_I_7,Z_L O_!,tStOt,,I

_-'_IPONE_t-I['I_L JOtNtN_ 51Ecl'tO!4

_ -_

COqNI'_OL 1FtltEO I_ 0tl_E._'_tOINk (_Z.

9o'r_,_
tO0" r..XlNEqiZI_E-

zoos

,
:

_,_t" (,>oJ,',L_-,?,-_,,_,z--/?

/
01;' Z5.4 4_./v_ LI/',iNIT5 _IEI2T_,CP, L ll_e-/v_tll>'l[li& ODNT_,OL '1'0 17..1_ IKII&z.. _q:)_.. %00* _E[glN%l[)1'lJi TO t_,ooo _;_ ',

(a) Fig. 1. Keele constant-coverage biradial horn (JBL 2344), computer-generated cross section. (b) Horizontal cross section.
J. Audio Eng. Soc., Vol. 31, No. 6, 1983 June

(b) plot. All dimensions in mm. (a) Vertical

409

SMITH AL. ET

PAPERS

4) Freedom from dynamic offset problems It is commonly thought that 380-mm(15-in) woofers cannot be used up to 1000 Hz due to poor frequency response and ragged polar characteristics. Fortunately this need not always be the case. If the cone stock is well chosen and properly terminated at its surround compliance, then its response will be smooth, both on and off axis, to the 1-kHz region. Fig. 9 shows the beamwidth and the directivity index of a 380-mm woofer mounted in an enclosure of typical size (0.17 m3, 6 ft3). The directivity rises smoothly, until at 1000 Hz it becomes an even match for a 100 by 100 horn.

The high price of Alnico V magnets has forced most manufacturers to use ferrite magnet structures, even though these were previously regarded as higher in distortion. Improved geometry and the use of flux modulation canceling rings has reduced their distortion levels to less than the equivalent Alnico structure (Gander [4] and Gilliom [5]). An added benefit is the elimination of the Alnico structure's tendency to demagnetize itself under high-power low-frequency pulses. Woofer dynamic offset is a problem long known about but seldom discussed or treated. With high input power

_,-

.,

, '-...

:... /% '-.

)_.

.. ,?:,, _,,,::,:(.

',.'

,,;_

:;:/

-.

, :-:'

'", ..

,o_ii !__;_i ..... .,,:.

,"
, >

', !:" _ ',",,':i', :'':_.i ' ' _,,_,1, .::'_,__/, '__________'_t_",_t "__' !\'': __1_, __, W,/l//i,, __, ,,
', , '-. ?/'

-_---., _' -,w :-'-_-___ ___:',._. _

(a)
?IGO ......... -_CoO

200:
IN

...................... ,oo_:
-12_

I-I-[:
_

...........
___z_'*< *'_;-_-_ 7'

- ;: :"-221::2
-gO

IoO--_ _!L]:: _-'_L 50-

_
o :_o

.': =-_L= 'Z': 2 =----]_

-._/.... _r_ _\ __
j

r7 _,--, :}:
IO000

90'

%'

.......

_ ............... i { !....

! ....

25 - _ ...... _O

___. 50 100

L_

............................. 200 _dO t-_,ec,_t m_c d

i
Itt_

-2000 50o0

IOOO

22.5" 2do0(I

(b)

20

50

lO0

200

.500

Id00

g'O00

5000

ID000

2Oooo

.F'F?EOUEIdCY

]_z

(c) Fig. 2. Polar response, beamwidth and directivity of constant-coverage horn shown in Fig. 1. (a) Composite plot of onethird-octave bandwidth polar curves at all one-third-octave center frequencies from 800-1600 Hz (1 dB per minor division). (b) Beamwidth versus frequency. (c) Directivity versus frequency.
410 J. Audio Eng. Soc., Vol. 31, No. 6, 1983 June

PAPERS

IMPROVEMENTS IN MONITOR LOUDSPEAKER SYSTEMS

at low frequencies, many woofers tend to shift their mean displacement forward or backward until the coil is nearly out of the gap. This is most likely to happen just above each low-frequency impedance peak of a system. The result is a high level of second-harmonic distortion and subjectively a bass character that loses its tightness at high acoustic output levels [4]. The cure for offset, as shown by Wiik [6], is a restoring spring force that increases in stiffness at high displacement in an amount that counterbalances the reduced B field at the extremes of voice coil travel. Such a nonlinear spider will in fact reduce distortion and eliminate the tendency to offset,

4 DIVIDING NETWORK CONSIDERATIONS The main goal of good network design is to produce the flattest response over the widest range of listening angles. The three-dimensional position dependence of the frequency response of a loudspeaker system, as controlled by the network, is generally overlooked. Response at off angles and total power response are of major concern. Each trial network with good axial curves must be measured at many angles, up and down, left and right. Those that pass this phase of testing will of course be further tested for power response and directivity characteristics, and will then be taken to the

--',,

/, :'v---', -- :_I -/:

.. _ .. -_.--_:_<_.._._ .____

..'._-;.:'! '/ ; .." ,..'

_ '_ '_..." . ., .. ./_ :.-.

(a)

20

.50

,tOO

200

_'_0

]PO0

2000

.5.-000

_o000

2_0o,0

r-rcucl_m,qc.' _[c t

(b)

20

.....

_0

I00

200

500

I000

Z'O00

8000

tO000

20o00

F'[_ffOUEklCY

_lz

(c)
Fig. 3. Polar response, beamwidth and directivity index of radial horn (Altec 51 lB). (a) Polar composites 800-16 000 Hz.

(b) Beamwidth versus frequency.

(c) Directivity versus frequency.

411

J. Audio Eng. Soc., Vol. 31, No. 6, 1983 June

SMITH AL. ET

PAPERS

studio for measurement and exhaustive listening tests, Crossover nulls that appear at off-axis angles are an inevitable consequence of the finite driver spacing. In noncoaxial designs the spacing is usually in the vertical

oL = half-angle nc = wavelength

between

nulls frequency

at crossover

plane, and it causes the woofer-to-listener and tweeterd] = center-to-center spacing (vertical array asto-listener distances to vary as the system axis is tilted, sumed) Linkwitz [7] shows that the angle between nulls is roughly defined by the-wavelength of sound at the For example, a lO00-Hz crossover frequency and a crossover frequency and the vertical spacing i_ee Fig. & ,.spacinggf 0.4 m yields 10). It is given by a = arcsin(_) dl = 0.4m(16in)

nc = 0.34m(13.5in) a = 25or 2a = 50 (arcbetween nulls)

where

....__._:_-_,_3_....__, -"L:-- _3:: / ": _'._:::_'_-L

-? " , '_':_' .... ._ / _x. '

.... k

_ -:-_4 / _ _ '

-r _t'5-'o; _ .' --: " ' _' '

:-::,,
........ :. '._ .. .,

.:>---_...,,._

....

.. ,,

_ ',',

......... " ,_ .... .....t_'_,,_ ...... _.'-'_ : ,

_ ---_

"":' "_ _' "i'!:'_' "" ! __-'"-_"_ t "".

, .....

-:,,_'.>...' ' '/;I ,: ,t..' '"""

-%._ '._._ : , _ : /

(a)

_o ......... 2O0.
I00 to .... 50 _o .................................... _o ........................ _ .................. ....................................... n' .50 IO0 2 O0 5_0 I ..............................

.__ %:1

'-.

3_o o

_ / _=: -- -_? _ -::'<;_--_

"

. _o ..... io (}0r_ )h, 2_ 22.5'

_s

V;ftc_;):m % _
1000 2 otto 500 O

p,g[ c_,_ Ir,-,Ir,,_ tit_

(b)

,oo i] l' '1

1]lIl{tltl '

'

' jJ ]IILo

20

! l IiilIItiI,lillt Itt;tt
50 lOO 200 500 lO00 o00 &O00 I0000 2Cooo F_E"OtlEMCY }4z

(c) Fig. 4. Polar response, beamwidth and dire ctivity index of straight exponential horn and divergent lens (JBL 2307/2308). (a) Polar composites 800-16 000 Hz. (b) Beamwidth versus frequency. (c) Directivity versus frequency.
412 J.AudioEng. Soo.,Vol.31, No.6, .1983 June

PAPERS

IMPROVEMENTSMONITOR IN LOUDSPEAKER SYSTEMS

The arc between nulls can be made more useful if it is tilted upward. That is, floor-standing systems should be optimized for response at angles on axis and above, If the system is to be used above ear level, then inverting it will once again yield the greatest latitude of listener positioning. It is also important that crossover nulls in the offaxis frequency response be as narrow and unobtrusive as possible. This is usually assured by higher order network transitions with minimal overlap, 5 PERFORMANCE A two-way monitor using mm (15-in)-nominal-diameter which meets all the previously 4430). Its performance in a measured in comparison to and other companies, the new horn and a 380woofer was designed mentioned criteria (model variety of tests has been previous designs of this

The measurements most vital in revealing the performance of this type of monitor are the beamwidth and directivity curves. Use of a computer-operated measurement system reduced the tediousness of these measurements and calculations (Keele [8]). Figs. 11-13 show comparisonsof the beamwidth and directivity of the 4430 (Fig. 11) to two previous monitors. Our two-way 4331 model (Fig. 12) had wide horizontal coverage but poor vertical coverage and constantly rising directivity. The DI versus frequency of the coaxial design (Fig. 13) appears smooth, yet the directivity index of the horn at higher frequencies is greater than that of the woofer by 4 dB, revealing a poor horn-woofer match. It is doubtful that a wideangle horn with good midrange control could be built into the available space of a typical coaxial design. Beamwidth versus frequency of this horn shows an interesting but undesirable trait, in that coverage angles vary in a complementary fashion, which maintains

(a)

,oo HtlItllI1 [ft]ttlllqIII [nTl][ 11111 f3 Ii [Ht IDIBHt mltlil]IHt][ .,o

I0 -+lo

, !l!WJl J JtiJJJJJ!H JtlJ[I!I IJ!JJJJ to

20 .50 I00 200 500 lO00 _000 $000 I0000 20000 F'R_EGUE_IC Hz

(b)

'

3'

0 i00

15'

-5

6 78910

'3'._J inco

15

'_

$ 78910

15 10100

(c) Fig. 5. Acoustic equalization of compression driver by beaming horn. (a) Power response of compression driver (terminated tube response). (b) Directivity index of horn. (c) Response of horn and driver. Power curve = axial curve - DI curve. Axial curve is flat, but power curve rolls off.
d. Audio Eng. Soc., Vol. 31, No. 6, 1983 June 413

SMITH AL, ET

PAPERS

consistent DI but does little to help off-axis response. Figs. 14-16 show the normalized off-.axis frequency response curves of the three systems. These curves represent those that would result from equalizing the axial response flat. Power compression versus level is plotted in Fig. 17. As the power levels were increased, the chart recorder gain was decreased a like amount. The degree to which the curves coincide shows the system's freedom from the effects of compression. These curves were

_,cn .8rrass

-q

Fig. 7. High-end equalization network.

.....

:i i'

e 70't

I i'il mmliI_L,1'1 _1'[ '_k, i_l '_;ElL _,'14 5 ,,6 78om ._._.__'1;I_[l 15 _ _5
Ioo 1 ooo

7 i '_ i:i2 a 5 6 789m 15 inooo

(a)

I
_0

-_0

I00

200

,
500

,
lO00 _000 6000 10000 200o_

O _z

F'IS'EOUEkICY

(b)

r: !;_j
il i _'; ' _ . .

ii,l C
: :ii_i
15 _ 3 4 5 ioooo 2 3 4 7 8 910 moo

1!i....,11,1
,t io mc}

(c) Fig. 6. Axial and power response of constant-coverage horn. (a) Power response of compression driver (terminated tube response). (b) Directivity index of horn. (c) Response of horn and driver. Power curve = axial curve - DI curve. Power and axial curves run parallel.
414 d.AudioEng. Soo.,Vol.31, No.6, 1983 June

PAPERS

IMPROVEMENTS

IN MONITOR

LOUDSPEAKER

SYSTEMS

run using a narrow-band tracking filter. The purpose of this tracking filter is to remove distortion components which would lessen the apparent compression as energy is transferred from the fundamental to harmonics. Conventional 18 distortion curves are also shown in Fig. of the 4430, earlier c_cm_,E__ a, =cE_m_ rac_e_ c_=A_ _,_Z_.

,;.w_v_ F_,_r_ 4,.,_--.T--.T--.T--._ / _ '_:_ / , ,' ' '*.r,_,

_<_c /

_,.._ _n_ec_r,. ,_L, s.u_,J_o._ _,cec, ^r

t,0tCC lqtFT S

/-_% ///(/ /_ J'/ i// '_7(_? '_"_ / f \; ', I \. _ _ ,_ _'_

l-- _'_.. _0__,s ?-__ _ I I "_' IIr_ \ _ / L I

The group delay characteristics

4331, and a popular constant group delay monitor are all plotted in Fig. 19 versus the Blauert and Laws criteria the constant group delay monitor excels, although threeeasilyfall wellbelowthecriteria

,o, _udib,_ ,9,.._,_ _*_''., m,n,m.m t,me d.,a, di,_,epa.ci_, '-._ '
all l_u,_, L,P. LOAgEPJ_Xlt_ F For uses where even greater low-frequency output capability with an attendent reduction in distortion is

_,._ ,_
'

Fig. 10. Geometric causes of vertical off-axis response nulls.

';"::::: , 1 Ti"i ' ::_!!][ _ti::li!_'! ! il t I_''_!'! : : :' _ i'_: ::-i::il.... ,_' I. !i' _

li
8 ',

i,
(a)

....15
.

' t! t :I I1 ii
'.._[[MLI.I?_I)/I ,,_..'_ ,il . 4I'L J_dJiI...!_ t F_,, _, 1 ._.'.... 2 ..,_, 5 s 7 _ _}i'_r_o 15;1__:: :,_,0 LJ 'f,._ r_ !'l' ln0,_

.......,f:Ir.......'' i:,.....I'll t' Itt....' ;_:1;_1,1" .... ........ ' , l: _lii:::

(h) Fig. 8. Equalized horn response and resultant driver power response. (a) Compression driver and horn electrically equalized. (b) Equalized compressmn driver on terminated tube Both axial and power responses are flat.

I00. 1lttlIi_ 10_U . 2 W_Iltlt_T , LLIJ_It

Ill'_''

20

50

IOO

00

._00

lO00 _z

_000

tSO00

I0000

20000

F_EOUEHC

.0 .o o
415

Fig. 9. directivity
J.AudioEng. Soc., Vol.31,No.6, 1983 June

index of 380-mm woofer.

SMITH ET AL.

PAPERS

required, a double woofer system has been designed (model 4435). Directional characteristics have been left intact by bringing the second woofer in below 100 Hz only. The maximum output before thermal or excursion limiting has been raised by 4 dB and extended

on the low end by half an octave. This is shown in Fig. 20. Note that this is not a response curve but instead is a curve of maximum reverberant field sound pressure level generated at the excursion limit or long-term power limit of the two systems in typical monitoring condi-

2SO

SO

,50

100

200

500

I000

2000

5000

IooOo

2tlooo

}:r:C,_dENC_ill

(a) __ucc-.c_St ,_',Fl<u-ot_E_tc) 51 _l-: t

o !' l ,l,H'it: 'Il tI,t "it:lt

20 50 lO0 200 500 I000 _'o00 5000 IOOr,o 20aoo Fh_EOOEI4CY ttz

(b) Fig. 11. Beamwidth and directivity index of constant-energy index. response monitor (JBL 4430). (a) Beamwidth. (b) Directivity

_ __
30o .............................

3too

ji
,_ t_o.............. ,,,................ -lO
.......................... 60 ............................

=_ ,. _c_,_V _ --

........

.....
180 _','x._ ....... 2a.5 o tooo _,, 2o_n

--'

25 -20

-- ----_...... . ,50 I00

200

5'00 I000 _: -eQUEN(.,J_[__ F

200o

I.... 5'000

(a)

20

50

IOO

200

500 ,c'_E(_IJEt4CY

I000 Mz

8000

6000

lO000

20000

(b) Fig. 12. Beamwidth and directivity index of two-way monitor (JBL 4331) using a 380 mm woofer _nd horn-lens assembly. (a) Beamwidth. (b) Directivity.
416 J.Audio ng. oc., 31,No. , 1983 E S Vol. 6 June

PAPERS

IMPROVEMENTS INMONITOR LOUDSPEAKER SYSTEMS

Zoo 2SO

................................ , ............................. ................ t,_ .==_r._ .................................... [_0

200.
t\tlGLE' Iz

'"

,_

I"-

__'_

_:i...CTZ /

- /i ;71CZ Z7

.......... _0 50 I00 200 500 I000

I.............. 2000 5000 Io000

22.5 2_oon

(a)

;20

50

}00

_'00

500

lO00

_O00

_O00

IOOOO

_Oooo

F_E_UE_C

Iqz

(b)
Fig. 13. Beamwidth and directivity index of a popular coaxial-style monitor (UREI 813). (a)Beamwidth. (b) Directivity

index.

I_- _
0 ?

O cul_,qr_
: :

15 g<Tq,_LILF--D
: : 7 . :

TO FL/_T:
: : 0

_N.F, TI4_- _I[-K[IL/_L

........

Ot:r---,t_15
[

C1.1Rx1.5 /_12, : E

o::,I:':
I'_EN TI-IE 140_-._LONY/_L

-_i [::i::i i
01:V-/_XV5 CLIF,-_C--_AR_:

tio"- s

zo' -

-_/.O-$
lO a -I0/ U,p

* '

-5
-lo

_,eL

_._.

....
tO"

I I I

50

mo

too

5oo

iooo

zooo

5000

_o, oo zo, _o o o

Fig. 14. Normalized off-axis frequency response curves of coaxial monitor (813). Note: Even coaxial mounting of horn does not prevent off-axis crossover dips; also, it exhibits high-end beaming and response deviations due to horn.
d.Audio Eng. oc., 31,No. 1983 S Vol. 6, June

_'5

I 40[ -_5 z_

_,o

_eo

z0o

_o

tooo z_o

_o

_,oc_ zo,o_o
417

SMITH ET AL.

PAPERS

IF TI46 O CLIgVC

16 Cqu_I_D

TO

FEAT:

I:M,t, T_i_ID

'q_RTI, CAL

O$_-/_._15

CLI_q

'ES _RE'..

:,_!oJ .... .... :i==]==J='_Ji=i::[:i:i=:l::l _o..-._[ I _2'-I ...... J '-'_"f............ I" ...__ J-tS_ I I ) I _ I llA[l'4, l'14E- g_O_.%ZDuT/_L 0_4:-._,,t CLII:_C--5 /M_,[-: _ _.-',x_ _

'_ /% / /,
. .

up

,o]
-IO ..... ,[_n ........... '"

- t5

-I

?Jo

5o

_oo

zoo

50o

_ooo

_oo

5ooo

_o. oo _o,ooo o

Fig.

15.

Normalized

off-axis

frequency

response

curves beaming. horizontal

of

ao_q -_0 _o too zoo 500 _oo za:x> 5 ooo -oo,-_,lo _ ooo

curves but poor vertical lens horn with much high-end curves monitor with acoustic (4331). Note: Good

IF

T_6

0 cuT-qr-

t5

r=qUt_L_Lr=['9

TO

F-LAT:

/_NO

T_[=

_IE_TICAL

Or-5-hxl5

u1_,,/[5

AIDE:

o-:I '[ _.,J

1-14EN TI4g HORIZ-ON/_L -lOt-!Z_-*
-tO;

' ["[':i"_,
OFF-A')(I5 CLeRiC-5 h_g: ZOe up --' o . _0 ............ u_ - _ -: i ,

--.

constant-coverage monitor (4430). Note: Smooth and even curves with aberrations confined to crossover region. Vertical response optimized for on axis and above (20 up curve surpasses 10 down curve), 418

Fig.,6. o,.m ,,zedo _a ,, r quencyresponsecurve,

a,,,,o -tS t zo 5o too zoo 5oo moo zooo

50oo

_o,0oo zo. oo o

d.Audio Eng. Soc., ol. 1, V 3 No. , 1983 6 June

PAPERS

IMPROVEMENTS INMONITOR LOUDSPEAKER SYSTEMS

tions. In the very low-frequency range of 20-30 Hz a stereo pair of the dual woofer systems can generate some 115-120 dB sound pressure level under these conditions. Both low-frequency drivers in the double system are identical to the driver in the single system except for lightened cones, which yield a 3-dB increase in midband sensitivity,

6 RAMIFICATIONS APPROACH

OF THE NEW DESIGN

6.1 The Room Curve Will Be Flatter; Equalization Will Be More Accurate Studio monitors are generally equalized as a matter of course. Control rooms are rarely as smooth at low frequenciesas may be desired, and mountingconditions

i i

_'_

ii! _ !_!!i_ii

!!'

i ,i:i.

Fig. 17. Power compression at input levels of 1, 10, 100 W: 93, 103, 113 dB at 1 m (4430).

_'rT._:Tr_].l,l.l_T_. 1fliT,l_,,lilj_ ;III N! ,.: :::,, _ll: '_ r,, '[!I:I,T, T , ii* II , ]; ,: ........ t,,,
112 "il , ; I = ] :i ! " ':a ![,: ..... r

,;

r{'

.m'.

............
loc

i
i030

.........
io0oo

3Z 'o

(a)

: i :2 !I_: . r. ]!!_1:! =. !

_ ;

':

[:!i

{_ii

i:{-i;F;I 1:/
1oo

......
[ooo lo(mo

.,_x
(b)

Fig. 18. Harmonic distortion versus frequency at two input powers (4430). (a) 5W input. (b) 50W input. Distortion curves raised by 20 dB. Second and third harmonic distortion components measured to 20 000 Hz only.
J.Audio Eng. oc., 31,No. ,1983 S Vol. 6 June 419

SMITH ET AL.

PAPERS

for the monitors are not always ideal. Further, the user's concept of monitor balance may not agree with that of the manufacturer. Even though we believe that constant coverage monitors will require less equalization than previous designs, the need for equalization may still exist, One of the more curious aspects of the recording art is the high-frequency tailoring of playback monitors, Amplifiers are flat to one-tenth of a decibel; the ideal microphone is supposedly flat; tape recorders are aligned with great care in order to have as flat a response as possible--yet control-room monitors are traditionally rolled off, typically as much as 3 dB per octave above 4000 Hz (Schulein [ 10]). If this is not done, the response is often thought to be overly bright. Recent studies have shown that equalizing to such a rolled off curve is merely a roundabout way of arriving at a flat direct sound field by allowing for the effects of increasing directivity and decreasing reverberant field intensity at high frequencies. In effect, we have been-equalizing the reverberant field but listening to the direct field (Queen [11], Bridges [12]). When measured in the reverberant field, the typical house curve exhibits a rolled off high end because previous monitors shared a similar power response rolloff. The degree of success with which a monitor could be properly equalized depended

on its direct field being flat when its power response was adjusted to this empirical "house curve" (Fig. 21). Deviations in the power response from this house curve would be equalized to yield complementary response errors in the direct field. For those cases where the powerresponse did not properly followthe direct field response, equalizing would make the direct response worse, and hence degrade the perceived balance. The high degree of parallelism between the axial and the power responses of the new monitor design means that less high-end equalization will be required and, of greater importance, that equalization will always be an improvement and never a degradation. 6.2 Stereo Imaging Will Be Improved The frequency response of the new monitor design (Fig. 22) is _tuite uniform, even at angles sufficiently off axis both horizontally and vertically to be unlikely listener posihons. However} this results in more uniform room reflections, which contributes to a stable virtual source that does not change with frequency (Queen [13]). In addition, increased toe-in can be used with no degradation of the direct sound field. If enough toein is used for the axes ofthe systems to cross somewhat in front of the listener, then the level precedence effect

r EL 21Il f[il/I
/ _30 '

I I/ll
, I / !lltl

IC" HL2-I\1 ,

[lt lll!/[ll!llflX[ JI
?o Hz 5O foo 20o 5OO t0O0 _C_ 5000 000O 2

,I_"Rg.._ U Z"IV Y c

#Z

Fig. 19. Group delay versus Blauert and Laws criteria.

i' i!r ...... I1 ! 1,lll ' " ,'.:.l,,i'!'", .1,i,r0 ,IT

: [ , I I ir: ' '
I

' '
!

i '
_

! :
i

_IF i Acousr,c
OUTPUT

,o1)[ /
2rllR I 0 0 , _ [

: ':
_ : : il
[ [ ] : [

i
!i
i
I; I i I

,i
i :Ii
: : i I
:

,
' I ] ; I _ ,

to.+
] o.o,

! 'I::
I : I _ I I ' i ; '

(_REYA'g.i/_f'-iVore
FIELD, ]

RADmrE .a. Aco,,s,-,c Sn_ ,_dB

r_:r ou:: wooFs're $rs r__t_ c:N I
i;'--;--'[ ;';l;l' ' -' : [ _ I _ I _ _ I'

',

J
I 4 [, I .......

: 'i
' ' I

1
0 . 00 +

90

lU,)

,_c:"R'-_OUE,VC /'[Z-

h,o0

h,_uo

Fig. 20. Maximum continuous output (4430, 4435).

420 d. Audio Eng. Soc., Vol. 31, No. 6, 1983 June

PAPERS

IMPROVEMENTS INMONITOR LOUDSPEAKER SYSTEMS

can partially offset the time precedence effect (Haas [14]). This contributes to a more stable stereo image as the listener's position varies along the length of the control board. 7 CONCLUSION As all other parts of the recording chain are improved, the playback monitors must follow suit. Previous mon-

itors had adequately flat axial frequency response and high acoustic output. By paying attention to power response and off-axis response, a monitor with fewer "colorations" and improved stereo effects can be realized. The use of a constant-coverage horn allows the designer to create a two-way monitor that surpasses three- or even four-way monitors in several of these important aspects. 8 REFERENCES [1] D. B. Keele, Jr., "What's so Sacred about Exponential Horns?" presented at the 51st Convention of the Audio Engineering Society, J. Audio Eng. Soc. (Abstracts), vol. 23, p. 492 (1975 July/Aug.), preprint 1038. [2] C. A. Henricksen and M. S. Ureda, "The MantaRay Horns," J. Audio Eng. Soc., vol. 26, pp. 629634 ( 1978 Sept.). [3] R. J. Newman, "The Constant Directivity White Horn White Paper," Electro-Voice, The PA Bible, no. 6 (1980).

IFt4_Omra_,b.I. [- ........................... _3LLS OFF S_::OTt_L',/... _ _ 'i/ _

_ _"' ""'"'" Xl _",*,,ff I "%1 _,_T 3,,_,, -.ro_,_,,ou,,_?__,,,o,v.... ] _ P, ..... _,,,, r.o[ .,,,.,_ L_ ...........................

'II_Ekl _L_,LlZtflGW--................................ 'n4_p.r_isr_r! RE_-_' 'th_"_'S_Co_g",/tELl_ A /!4_T, ' _'_

QF,_f:L'T 5rXlklD_.//

___B

iFI_.l.Sr,rs[' t l 6_mn4,,,

r- .................. ;//__ ,// _ ?' _-..........................

[4] M. R. Gander, "Moving-Coil Loudspeaker Topology as an Indicator of Linear Excursion Capability," J. Audio Eng. Soc., vol. 29, pp. 10-26 (1981 Jan./ I Feb.). [5] J. R. Gilliom, "Distortion in Dynamic Loud3 speakers Due to Modulation of the Permanent Field," _.... _""_!x_: presented at the 42nd Convention of the Audio Engi_'x_ neering Society, J. Audio Eng. Soc. (Abstracts), vol. _'_,'"_N,x I q ] 20,[6] T. H.(1972 June). p. 430 Wiik, "Transient Distortion Caused by Nonlinearities in Driving Force and Suspension of a Loudspeaker," presented at the 56th Convention of the Audio Engineering Society, J. Audio Eng. Soc. (Ab[7] S. H. Linkwitz, i ,
I

_r_OALt2lflO 5--...................... . _EVEt_._f_AkI'F O:,,Ut.V__Te ll-tg _c'T '" "' _zvE"_t6m TO"_xr-_ptt__ /_'-'--' __._o.__,j_, e. sco,r-_, e i L............ Fig. 21. Reverberant field equalization.

"Active

Crossover Networks

for

Noncoincident Drivers," J. Audio Eng. Soc., 1205 . stracts), vol. 25, p. 336 (1977 May), preprint vol. 24, pp. 2-8 (1976 Jan./Feb.). [8] D. B. Keele, Jr., "Automated Loudspeaker Polar Response Measurements under Microcomputer Control,'' presented at the 65th Convention of the Audio Engineering Society, J. Audio Eng. Soc. (Abstracts), vol. 28, p. 374 (1980 May), preprint 1586. [9] J. Blauert and P. Laws, "Group Delay Distortions vol. 63, pp. 1478-1483 (1978 May). in [10] R. B. Schulein, "In Situ Acoust. Soc. Am., Electroacoustical Systems," J. Measurement and Eng. Soc., vol. 23, pp. 178-186 (1975 Apr.). [11] D. Queen, "Relative Importance of the Direct and Reverberant Fields to Spectrum Perception," J. Equalization of Sound Reproduction Systems," J. Audio Audio Eng.Mar.). (Project Notes), vol. 21, pp. 119121 (1973 Soc. [12] S. Bridges, "Effect of Direct Sound on the Perceived FrequencyResponseof a SoundSystem," presented at the 66th Convention of the Audio Engineering Society, J. Audio Eng. Soc. (Abstracts), vol. 28, p. 552 (1980 July/Aug.), preprint 1644. [ 13] D. Queen, "The Effect of Loudspeaker Radiation Patterns on Stereo Imaging and Clarity," J. Audio Eng. Soc., vol. 27, pp. 368-379 (1979 May). [14]H. Haas, "The Influence a SingleEchoon of the Audibility of Speech," J. Audio Eng. Soc., vol. 20, pp. 146-159 (1972 Apr.).
421

555'6_ _'73--_

4oo[Is?__'_q

_-

I
905.7

--'-"-[3532---_2] .... _

I1" :1 :. '----_-_ \ ..... Fig. 22. 4430 monitor.

J. Audio Eng. Soc., Vol. 31, No. 6, 1983 June

..

....

SMITH ET AL.

PAPERS

THE AUTHORS
c

David Smith received a B.S.E.E. degree from Purdue University in 1977. He began work as a designer of loudspeaker drivers and home loudspeaker systems for Cletron division of Essex (now Harmon-Motive) in Cleveland. In 1980 he moved to California where he worked for JBL. He was responsible for the design of the first four monitors of the 4400 series. Early in 1983 he left JBL, and is now doing acoustical consulting, D. B. (Don) Keele, Jr., recipient of the AES second Publication Award in 1975, to "an author under 35 years," was born in Los Angeles, California, in 1940. After serving in the U.S. Air Force for four years as an aircraft electronics technician, he attended California State Polytechnic University at Pomona, where he graduated with honors and B.S. degrees in both electrical engineering (majoring in electronics) and physics. Mr. Keele worked three years for Brigham Young University in Provo, Utah, as an audio systems design engineer in the electronic media department. There, too, he received his M.S. degree ih electrical engineering in 1975 with a minor in acoustics, From 1972 to 1976, Mr. Keele was with ElectroVoice, Inc. in Buchanan, Michigan, as a senior design engineer in loudspeakers, concentrating on the design of high-frequency horns and low-frequency direct-radiator vented-box loudspeaker systems. He is the primary designer of their "HR" series of constant directivity high-frequency horns on which he holds the patent. In 1976, Mr. Keele worked for Klipsch and Associates in Hope, Arkansas, as chief engineer involved in the company's commercial line of industrial loudspeakers. In October 1977, he joined James B. Lansing Sound, Inc. in Northridge, California, as a senior transducer engineer, and is currently working on high-frequency horn and transducer design, with

heavy emphasis on computer-aided methods for acoustical design, analysis and testing. He also holds the patent on JBL's "Bi-Radial" series of constant directivity/coverage HF horns which he received early in 1982. A member and fellow of the Audio Engineering Society, Mr. Keele has presented and published a number of papers on loudspeaker design and measurement methods, among them the paper for which he won the AES Publication Award, "Low-Frequency Loudspeaker Assessment by Nearfield Sound-Pressure Measurement" (J. Audio Eng. Soc., vol. 22, p. 154, 1974 April). Mr. Keele has been an invited speaker at many AES section meetings and loudspeaker conferences. John Eargle holds degrees in music and electrical engineering from the Eastman School of Music (B.M.), University of Michigan (M.M.), University of Texas (B.S.E.E.), and Cooper Union (M.E.). He is a member of the Acoustical Society of America and Society of Motion Picture and Television Engineers, a senior member of the IEEE, and a fellow and honorary member of the AES. He is a past president of the AES and is active in reviewing papers for the Journal. He worked for RCA Records and Mercury Records in a variety of positions during the sixties. During the early seventies, he worked for the Altec Corporation and formed the consulting firm of JME Associates. Since 1976 he has been associated with JBL IncorporatedofNorthridge, California, in the areas of product development and product application. Since March 1982, he has expanded his activities in the areas of recording engineering and producing. He is a copious writer, having published more than forty technical articles and record reviews. His books, SoundRecording and The Microphone Handbook, are widely used as texts at the university level.

422

J.Audio ng. E Soc., 31, 6,1983une Vol. No. J