september - American Radio History
SEPTEMBER
1947
Why
the took- 4/Tube is at home in FM
ULTRA -HIGH FREQUENCIES
... HANDLED
WITH
EASE
The Sylvania Lock -In is the tube specifically
engineered to more than satisfy the requirements of Frequency Modulation handles
ultra -high frequencies with ease! Some
reasons for this electrical superiority are:
short, direct connections
fewer welded
joints -less loss; getter located on top . . .
shorts eliminated by separation of getter
material from elements.
-
...
STAYS PUT IN
SOCKET... MECHANICALLY RUGGED
Specially designed "lock -in" locating lug
on each tube keeps it in place- assuring
firm socket contact. Improved tube mount
keeps elements ruggedly supported on all
sides. There are few welded joints and no
soldered joints -the elements can't warp
or weave.
COMPACT... MADE TO FIT SMALL SPACES
This famous Sylvania product is ideal for
use in modern sets, where the tendency has
been toward more compact units -has reduced overall height and weight. Has no
top cap connection
overhead wires are
eliminated! See Sylvania Distributors or
write Radio Tube Division, Emporium, Pa.
...
NIA
ELEC RIC
SYLVA
SYLY1NIA3 LOCK-IN TJBE...
tie rado tube whose electrical aid -mechanical superiority
makes it the ideal choice
for
FM
and television, equipment in the
air, on the road, marine radar.
MAKERS OF RADIO TUBES; CATHODE RAY TUBES; ELECTRONIC DEVICES; FLUORESCENT LAMPS,
FIXTURES,
WIRING
DEVICES;
ELECTRIC LIGHT BULBS
John H. Potts, Editor
Sanford R. Cowan, Publisher
Member of Audit Bureau of Circulation
C. G.
McProud, Managing Editor
S.
Edgar E. Newman, Circ. Mgr.
Editorial Advisory Board
Representatives
Howard
A.
James C. Galloway
816 W. 5th St., Los Angeles 13, Calif.
Chinn
John D. Colvin
Successor to
RA D I O
C. J.
LeBel
J. P.
Maxfield
Dale International Publications, Ltd.
105 Bolsorer St.
London W. 1, England
Harris & Floyd
297 Swanston St., Melbourne
C. 1, Victoria, Australia
George M. Nixon
S.
CONTENTS
Established 1917
Adv. Director
L. Cahn,
H. N. Reizes, Adv. Mgr.
Ann Ellis, Asst. Circ. Mgr.
Louisa Dresser, Edit. Prod. Mgr.
Young White
SEPTEMBER, 1947
Vol. 31, No. 8
Letters
3
Editor's Report
4
High -Fidelity Volume Expander- Norman
Multi -Lingual Interpreting
C. Pickering
Systems -C. A. Tuthill
High- Frequency Equalization for Magnetic Pickups -C.
Ultrasonics in Liquids -S.
10
G. McProud
13
Young White
Design of Electronic Organs, Part
A -C
7
16
II- Winston
Wells
19
Voltage Stabilizers-Leo L. Helterline
Planning
a Studio
Installation, Part
Record Revue -Edward Tatnall
Musical Acoustics, Part
23
D. Colvin
25
29
Canby
IV-Benjamin F.
Tillson
30
Improved Method of Locating Operating Point -Arthur Plowright
32
Audio Design Notes
34
Technicana: Pilotuner, Cathode-Follower Precautions, Sound Motion
Picture History, Improving Speaker Damping
36
New Products
38
Chart: Loss Due to Cable Pair Working Between Impedances of
Advertising Index
5
Ohms
.
40
48
COVER ILLUSTRATION
Rear view of Broadcast Recorder's new mobile broadcasting unit, showing recording machines,
amplifiers, and table microphone in foreground. Chief engineer Joe E. Otis is using apparatus.
AUDIO ENGINEERING (title registered U. S. Pat. Off.), is published monthly at 28 Renne Ave.. Pittsfield, Massachusetts, by Radio Magazines. Inc., J. H. Potts,
President; S. R. Cowan. Sec y- Treas. Executive and Editorial Offices at 342 Madison Avenue, New York 17. N. Y. Subscription rates- United States. U. S. Possessions
and Canada. $3.00 for I year, $5.00 for 2 years; elsewhere $4.00 per year. Single copies 35c. Printed in U. S. A. All rights reserved, entire contents Copyright 1947 by
Radio Magazines. Inc. Entered as Second Class Matter. under the Act of March 3, 1879.
AUDIO ENGINEERING
SEPTEMBER, 1947
1
BARGAINS
JOBBERS
WHOLESALERS
MANUFACTURERS
LOS ANGELES. CALIF.
Cole Instrument Co.
1320 S. Grand Ave.
Large inventories of valuable electronic tubes,
devices and equipment are being offered by the
WAA Approved Distributors listed herewith
for your convenience. Alert commercial buyers
are taking advantage of this big bargain opportunity. Why not fill your present and future requirements from these available stocks. Act
now -while inventories still permit wide selection.
Purchase of this surplus equipment has been
greatly simplified. The Approved Distributors
appointed by WAA were selected on a basis of
their technical background and their ability to
serve you intelligently and efficiently. Write,
phone or visit your nearest Approved Distributor for information concerning inventories,
prices and delivery arrangements. You'll find
you can "Save with Surplus."
OFFICE
OF
AIRCRAFT
AND
Hoffman Radio Corp.
3761 S. Hill St.
NEWARK, N. 1.
National Union Radio Corp.
57 State St.
Standard Arcturus Corp.
99 Sussex Ave.
Tung -Sol Lamp Works, Inc.
95 -8th Ave.
NEW ORLEANS, LA.
Southern Electronic Co.
512 St. Charles St.
NEW YORK, N.
BOSTON, MASS.
Automatic Radio Mfg. Co., Inc
122 Brookline Ave.
Technical Apparatus.
165 Washington St.
BUCHANAN, MICH.
Electro-Voice, Inc.
Carroll & Cecil Sts.
CANTON, MASS.
Tobe Deutschmann Corp.
863 Washington St.
CHICAGO,
ILL.
American Condenser Co.
4410 N. Ravenswood Ave.
Belmont Radio Corp.
3633 S. Racine Ave.
EMPORIUM PENNA.
Sylvania Electric Products, Inc.
FORT WAYNE, IND.
Essex Wire Corp.
1601 Wall St.
HOUSTON, TEXAS
Navigation Instrument Co., Inc.
P. 0. Box 7001, Heights Station
ELECTRONICS
Y.
Carr Industries, Inc.
1269 Atlantic Ave., B'klyn.
Electronic Corp. of America
353 W. 48th St.
Emerson Radio & Phonograph Corp.
76-9th Ave.
General Electronics, Inc.
1819 Broadway
Hamm W.
St. Co., Inc.
460 ,. 334th ga.
Johann, rl Keegan Co., Inc.
62 Pearl St.
Newark Electric Co., Inc.
242 W. 55th St.
Smith - Meeker Engineering Ca.
125 Barclay St.
NORFOLK, VA.
Radio Parts Distributing Co.
128 W. Olney Road
ROCHESTER, N. Y.
W. & H. Aviation Corp.
Municipal Airport
SALEM, MASS.
Hytron Radio & Electronics Corp.
76
Lafayette St.
SCHENECTADY, N. Y.
General Electric Co.
Bldg. 267, 1 River Road
WASECA, MINN.
E. F. Johnson Co.
206 2nd Ave., S. W.
DISPOSAL
1284
WAR ASSETS ADMINISTRATION ._ la
Wei,A
moll.
s't,A1/
Offices located at: Atlanta Birmingham Boston Charlotte Chicago Cincinnati Clevel and
Dallas Denver Detroit Fort Worth Helena Houston Jacksonville Kansas City, Mo. Little Rock
Minneapolis Nashville
New Orleans
Los Angeles
Louisville
New York Omaha
Philadelphia
Portland, Ore. Richmond St. Louis Salt Lake City San Antonio San Francisco Seattle Spokane Tulsa
2
AUDIO ENGINEERING
SEPTEMBER, 1947
JI
.l).Il 1lF. hf=
ui fh/vyñferuiei
AD
REDÚCED-RECO1RD-PRÉS¡SÚ_RE_39
l
MORE ABOUT PITCH
Sir:
I should like to take exception to the
viewpoint which you expressed in your
editorial on Pitch, in the August issue of
Audio Engineering- namely, that in the
interest of achieving higher fidelity in
reproduction, the concert pitch used
during a composer's time be used
today in performing his works.
First, a fundamental truth about music
is that it is the inter-relationship of tones,
rather than the individual tones themselves, which give "meaning" to music.
Concert pitch is merely a technical convenience, useful as a mode of reference in
any specific musical era, and established
mainly by performers, conductors, musical instrument- 'makers and physicists.
The "content" of any musical composition is not affected by the relatively small
historical variation in concert pitch; it
is this "content" which the composer is
anxious to convey to the listener.
Variations in concert pitch extend, historically, from A =373.7 cps
( "church pitch," 1648, ref. Mersenne)
and A =402.9 cps ( "chamber pitch,"
same date and ref.) to our present A =440
cps. The basis for the widespread
variation is a complex one, involving
such factors as hearing, room acoustics,
scale temperament, the "temperaments" of performer and conductors,
musical instrument construction and
organs),
(e.g., church
durability
style in musical composition, and virtuousity in performance. The attempt
to standardize concert pitch at A =440
cps (International Conference on Pitch,
May 1939, London) was the inevitable
conclusion of physicists and musicologists who were aware of the necessity
for such standardization in modern
society, whose technology makes cultural interchange both desirable and
relatively easy.
Second, even in a specific era, different composers were accustomed to different, or similar, pitches, according to the
instruments they used and the cities in
which they happened to reside. Consider
the travels of Mozart- France, England
Italy, Germany; and consider that known
concert pitch ranged from A =395 cps,
(organ at Trinity College, Cambridge,
1759) to Handel's tuning fork, A =522.5
cps (London, 1751). How should one
even begin to make a specification for
the use of an "historical" concert pitch?
Third, many factors other than "historical" pitch are involved in any attempt
to recreate the historical setting of a
musical composition; so many, in fact,
that such an attempt becomes more the
work of a museum rather than a concert
organization. For example, the violin of
Bach's time was played with a bow
whose wooden arch curved away from
the bowhair; whereas the modern bow
uses an inverted arch, permitting in[Continued on page 5]
AUDIO ENGINEERING
.
NEW
PRESSURE
22 GRAMS
OLD
l
32
só
29
28
27
25
24
23
22
PA RA -FLUX
REPRODUCER
gives less wear on record . .
lighter impact of stylus ... and
improves a well-known tone quality
e
PRESSURE
35 GRAMS
It's the low mechanical impedance designed into
the improved PARA -FLUX . . . the special refined
metals and other components now obtainable . . .
that enable reducing the record pressure of all
35 grams to 22
R -MC Reproducer Heads from
grams. And all three types: Vertical only, Lateral
only, and Universal maintain the correct weight for
permitting the pressure of 22 grams on the record.
From our knowledge, we believe that PARA -FLUX
Vertical only and Universal are the only heads
obtainable today, which operate on commercial
with
Para -Flux Reproducer
interchongeable Heads:
service at a pressure of 22 grams. This improved
Universal
feature means less wear on records, and lighter
Lateral only
Vertical only
impact of stylus when inadvertently dropped.
R -MC engineering skill applied to reproducer design gives all the advantages that discriminating users demand: More realistic reproduction of
transcriptions . . . a reproducer of precision -build, sturdy construction, with
finest materials obtainable . . . embodying up -to- the -minute features, including convenient finger lift for preventing slipping of Reproducer when lifted
off record. A highly polished aluminum alloy center -piece of tone arm and
head enhances the attractive design of Reproducer.
This new lightweight Head, either Vertical only, Lateral only, or Universal,
functions correctly with all R -MC Tone Arms now in service. Therefore it is not
necessary to change arm in service when ordering the new Head.
Whenever you may need a new PARA -FLUX Head, your R -MC Jobber will
in exchange for
immediately
supply you with the new lightweight head
your old one, in accordance with our standard replacement policy and
.
...
...
exchange price.
Available through authorized jobbers. Descriptive, illustrated Bulletin PR 4.
RADIO -MUSIC CORPORATION
EAST PORT CHESTER
SEPTEMBER, 1947
CONN.
3
EDITOR'S REPORT
JUST the other day, one of the most prominent
broadcast engineers remarked that our magazine
had brought to him a realization that there is a
great deal more to audio engineering than he had
formerly thought. This is quite understandable, because so little about the audio field is published elsewhere. The r -f men have been the glamor boys of the
radio -electronic industry. It is their work which has
received such disproportionate emphasis in both radio
and general publications. Their accomplishments in
making radio waves lie down, roll over, and bounce
around have been oh'd and ah'd by the general public
as well as the engineering professions.
But anyone who listens to an unmodulated radio
wave gets awfully bored in a very short time. Nothing
so forcefully points up the fact that an unmodulated
wave is a carrier, and nothing more. Insofar as the
broadcasting industry is concerned, the audio man has
by far the toughest job. He must pick up sound, often
under the most adverse circumstances, compress it
within the dynamic range limitations of the r -f transmitter, keep distortion of all kinds to a negligible
level, hold the esthetic qualities of symphonic music,
and bear the criticism of tin-eared listeners using poor
radio receivers. The audio man is the unsung hero of
the broadcasting industry.
NATIONAL ELECTRONICS CONFERENCE
On my desk before me is a program of the forthcoming National Electronics Conference, to be held in
Chicago, Nov. 3, 4, and 5 inclusive. This Conference
is sponsored by the Illinois Institute of Technology,
Northwestern University, the University of Illinois,
the American Institute of Electrical Engineers, and
the Institute of Radio Engineers -all in co- operation
with the Chicago Technical Societies Council. Of the
74 technical papers scheduled for presentation, only
two are within the audio engineering field. H. H. Scott
will again discuss his dynamic noise suppressor, presenting, we understand, a more detailed description of
this useful device. The only other paper is by Warren
and Hewlett, and deals with intermodulation tests.
Fortunately for those who want to get in and get out
as quickly as possible, these two papers are the first
to be presented. The program then veers off into a
hodge -podge of instrumentation, microwave technique, nuclear physics, general electronics, and basic
science, finally getting around to television and FM
on the last day.
4
Most of these lectures are so highly specialized that
those of interest to one group can have little appeal to
others. If all discussions which deal with radio were
subdivided into their proper categories and presented
consecutively, the Conference would render a far
greater service to those busy engineers who must travel
appreciable distances to attend, and who have no
time to waste. As it now stands, studies of nuclear
reactions, the dynamic properties of the infra-red
cesium arc, and scintillation counters, are sandwiched
in between lectures on microwave technique, FM,
and television broadcasting.
Just how the I. R. E. can justify its sponsorship
of this program is difficult to understand. The
Rochester Fall Meeting of the I. R. E., which follows
right on the heels of this Chicago affair, is devoted
entirely to radio and is always well attended. There is
no earthly reason why Chicago shouldn't be able to
offer a similar program-and audio, too, if you please
-without including a dose of physics.
PROGRESS REPORT
Since our first issue in May, AUDIO ENGINEERING has
enjoyed the most phenomenal growth in circulation
of any journal in the radio- electronic field. A circulation audit during August showed that our paid circulation had nearly reached 10,000 -almost double that
of the first issue -and is growing at an even faster
rate. Because we employ no subscription salesmen,
circulation promotion concerns, or cut -rate offers to
stimulate circulation, this growth must be due largely
to the support given us by our readers. Let me take
this opportunity to thank you one and all for your
co- operation. It makes it easier for us to give you a
better magazine.
To improve our service on the West Coast, Harry L.
Bryant, Chief Engineer of Radio Recorders, will report on audio engineering developments in the Hollywood area. In October, Joseph P. Maxfield of our
editorial advisory board will also establish himself in
Hollywood as a consultant, and will contribute regularly to our magazine.
Forthcoming articles include one on the Magnetophon, studio model, about which there has heretofore
been little written. A new principle in pick -up design
is discussed by another writer. A complete series of
H. P.
p -a articles will start in the next issue.
J.
AUDIO ENGINEERING
SEPTEMBER, 1947
creased flexibility and virtuosity in performance, along with greater refinement
in timbre. Would you, then, suggest
that the interest of higher fidelity be
served by requiring the use of the nowoutmoded bow, in the performance of
music by Bach and his contemporaries?
May I also mention the following
objections:
Most instruments could be tuned
to the required historical pitch; but
what should be done with the piano,
the organ, the xylophone, glockenspiel, etc.?
Today, the hearing phenomenon of
absolute pitch is characterized by a
generally reliable response to the reference pitch of A =440 cps- perhaps because of the conventional use of this
frequency by piano manufacturers.
The musician with absolute pitch (and
there are many) hears and thinks of
music in terms of the reference pitch
(A =440 cps), whether he plays or composes. The mental gymnastics required of such musicians, if required to
perform according to historical pitch,
would be highly complicated, if possible
at all.
I should be pleased to discuss this
matter at greater length.
May I take this opportunity to congratulate you on your new magazine, and
to express my hope that it will contribute
LISTENER PREFERENCE TESTS
With the advent of radio, music is now
used chiefly as a background. People have
developed an aversion to silence and often
have the radio on from morning till night
with only scattered periods of real listening.
Under these circumstances, the presence
of high frequencies is attention compelling
and therefore irritating, but if highs are cut
heavily, a pleasant, unobtrusive rumble
results which provides the required noise
with minimum distraction.
In tests performed by Dr. Olson, the experience was presumably of short duration
and the subject especially intended to listen
to the music, which may account partially
for the result.
I do not mean to imply in the foregoing
that the other factors listed are not of
[Continued on page 471
Sir:
Dr. Olson's "Listener Preference Tests"
seem to be by far the most valuable yet performed and apparently reverse the findings
of the vast number of previous similar experiments. It seems to me that one important factor has been omitted in the reasons
given for preference for a restricted range
high-frequency sounds are more attention
arresting than those of lower frequency.
Prior to the advent of radio, listeners
seldom heard music except when they made
a considerable effort to do so, such as attending a concert. Under these circumstances,
they attended to listen to the music, not to
use it as a background for conversation or
-
other activity.
7»ordarstw r
NEW 10 WATT PHONO- AMPLIFIER
to greater co-operation and understanding
between the musician and the scientist?
I should like especially to commend you
on the series called "Musical Acoustics,"
by Mr. Tillson.
H. L. Robin
Director,
Acoustics Department,
Juillard School of Music
120 Claremont Ave.,
New York 27, N. Y.
GUIDING FISHES
Sir:
With regard to Mr. J. T. Barnaby's
letter in your August issue. Answering
his salmon problems, may I suggest that
he look up "Forced Movements, Trophisms and Animal Conduct" by Jacques
Loeb, M.D., Ph.D., Sc.D., (J. B. Lip pencott Co., Philadelphia, 1918). In
Chapter IV on "Galvanic trophisms," he
shows that a large number of animals,
such as shrimp, tadpoles, etc., when put
in a trough of water through which a
galvanic current flows, have a strong
tendency to go to the anode, because they
assume a position which causes them
least pain. He also stresses the importance of showing the correct current density. In his experiments with shrimp he
found that a current density of one
milliampere per 1400 mm2 rms best.
May I suggest that Mr. Barnaby try a
direct electrical current through the
water to guide his small salmon into safe
channels past the Columbia River dams?
B. J.
Typical of the quality built into the Thordarson line of
Hi- Fidelity Amplifiers is this new 10 watt phono -amplifier.
Designed for use with the Meissner AM -FM Tuner, or with tuners
of comparable performance, this unit may be used either as the speech
amplifier in an amateur transmitter or as the amplifier section in
recording units. Separate bass and treble controls (with both
accentuation and attenuation action) assure correction of output to all
acoustical conditions. Production costs have been materially lowered
by mounting on the simple chassis shown here.
Complete specifications on all
Thordarson Hi- Fidelity Ampli-
fiers available on request.
AVAILABLE FOR
Miessner,
Miessner Inventions, Inc.
Van Beuren Road,
R.F.D. 2, Morristown, N. J.
AUDIO ENGINEERING
SEPTEMBER, 1947
IMMEDIATE
DELIVERY
ELECTRONIC DISTRIBUTOR AND
INDUSTRIAL SALES DEPARTMENT
MAGUIRE INDUSTRIES, INCORPORATED
500 WEST HURON STREET
CHICAGO 10, ILLINOIS
EXPORT -SCHEEL INTERNATIONAL, INCORPORATED
4
2
l
7
N
.
L
I
N C O
L
N
A V
E
N U
E
,
C
H
I
C
A
i
O
18
,
I
L
L
I
N O
I
S
C
A
B
L
E
-
H A
N
5 C
H
E
E
l
5
/
/
\\
i
; PRESENTING
ì'635\
HIGH FIDELITY DYNAMIC MICROPHONE
i
vr Std v
_
1
1
NEW
ard gallote
C
1.
I
`L
G
4),
1
1
\
Chuck Acree, well -known
radio personality, shown
using E -V 635 on net-
audience particpotion programs.
work
t
da11y r,abor
o of
to
dlity
y Approvedgropdcast
promin ent
Certifie6
Stations
N
dc
service
Engineered for the exacting requirements of
.. this new E-V 6°3:5:i s proving its quality and versatility on
important radio network programs. Substantially flat, wide
range response (60 cycles to 13 kc, plus or minus 2.5 db) conforms
assures faithful
to modern FM as well as AM standards
reproduction of speech and music. Omni -directional-below 2000
effective
c.p.s., becoming directional at higher frequencies
for group or individual pick -up. Output -53 db.
The 635 is excellent, dependable for announcing, audience
participation, applause pick -up, roving reporter, sports, special
and for high quality sound amplification
events, recording
systems. Can be used on a stand or in the hand-indoors and
outdoors. Recessed switch in microphone stud gives
ready selection of 50 or 250 ohms impedance. Fully
guaranteed. Try it ... test it ... see and hear itfor yourself!
Model 635 Broadcast Microphone. List Price ..$60
...
...
...
Send now for Bulletin No 135
AUTHORIZED DISTRIBUTORS EVERYWHERE
No Finer
Choice Than
ELECTRO- VOIÇE, INC., 6UCIHI: 10,',!, MICt.IC7AN
Export Division: 13 East 40th Street, New York 16, N.Y., U.S.A.
6
AUDIO ENGINEERING
Cables: Arlab
SEPTEMBER, 1947
Front
view
h
h- fidelity
i
g
of
volume expander
High- Fidelity Volume Expander
NORMAN
ALMOST ALL devoted listeners to
recorded music, after they have
gotten their playback systems to the
point where the sound is pleasant to hear,
feel that the loudness range on records is
much too limited. To the listener who is
concerned primarily with the aesthetics
of the matter, the compromise can be
made in either of two ways: the volume
level can be set high enough to achieve
satisfactory fortissimos
which setting
the pianissimos are too loud and too
noisy; or the volume can be adjusted for
maximum tolerable surface noise-in
which case frustration sets in when the
crescendos level off at a feeble mezzoforte.
A considerable improvement in loudness range can be made by the intelligent
use of a well-designed volume expander.
The great difficulty is, of course, in the
design of the expander itself, which is far
from being a simple job. There have been
many circuits offered in the past, all of
which have at least one of the following
defects:
1. The distortion is excessive
2. The expansion is not smooth
3. The rate of action is too slow
4. There is no positive limit to the
amount of expansion.
The development of the present circuit
was instigated by David Hall, the eminent
authority on records and sound reproduction. The equipment had to pass the most
stringent tests from the standpoints of
both technical and musical listeners. This
it seems to have done very satisfactorily,
and it is believed that this is the most
musically satisfying volume expander yet
developed.
PICKERING*
Presenting for the first time
In order to keep the distortion to substantially zero level, and at the same time
provide a definite limit to the possible
increase in gain, it was decided to attack
the problem by decreasing the gain of a
amplifier
high -quality low- distortion
stage in a controlled manner, by a method
which, in itself, would not introduce
distortion.
a
volume expander in which all
forms of distortion are reduced
to negligible proportions.
-at
Fig. 1.
Schematic diagram of the high-fidelity volume expander.
OK
EXPANSION
INDICATOR
-MA/,
6AF6
G
RELEASE TIME
3
2
6SL7
M
EXPANSION
CONTROL
*Pickering & Co., Inc.,
.309 Woods Ave., Oceanside, N. Y.
AUDIO ENGINEERING
C.
+
300
V.
B
SEPTEMBER, 1947
" NO EXP
12á1V
db
7
E
te
EININ11I.E11OAEII
OM0A//
MOMMMOMM\!,M'A©I
EOEE/IH.MIOt11
6
II
111111111111111111
PLATE CHARACTERISTIC
CURVE OF TYPE 60
CONTROL
MONDE
RAMS Of OPERATION
MM'A\!4MOE.ANlI
OMO
MMO MI,EAEI1\`7A1IIII
MIII,\'C.\IIMI'AOM
M. MV nOI,
Noaw
E
11AIIti:mE'AE'S-A0TBAL
MINEIAI1'A01/MI.MlyE rHxt
MMMA1AI.OMA\TVi,I/MI
E E110CMAIIEEIcEE1A\: .;M/I ABnML
RurE'
I010 110101011
EPAIAAE'AI/MMI00111,A
OMMA1/I,I
'AOV/O:!EMMEMM
MMM/HOMAM11/IM/,MEI
ium'AmOO
MAI1MAAINEAOMIIt,'II,M
E7M
MMOMIOr1Iib/OOA3.rA/,MC!HPO
1101501MIO'ArA11bOI
1:4MO
A
IM/EIMMA'Iry\IAI,A1.I
EOOA11I
10
'A/4
'AE
MMM
OMMMM
RANEE
c1vrNEHEN
TUBES
PLATE CURRENT VS. INPUT
TYPE SP5 (TB)
51BNAL
10
tRxHYGE
AI
IIIIJ IMMÌÍUM.I
IIIU/:iE/.E/EE
MO/
ME
/
INIAA/AMMAINIiM7
M1/
E AM I A I 'IM%MOr/MMOLM
EE/E/AEI
IUUM
Irlt:fiIiME'II.
UEEUEIIAE/MMIEIE .
I/P %I./
I
a
I
ORMI AI.AI.EI,I Ì4'"iA6M
I%I.MO/A//R%I,IMi/E/I.%
PiM i IUM
UI/AErMMt:MSr'AiEMOSCiMMMM
/.MI.M/.
o
too
100
Fig. 2.
The use of variable -mu tubes was
abandoned at the outset, as soon as
inherent intermodulation distortion of the
order of 15% was discovered. From the
standpoint of low distortion there is still
nothing better than a triode of low plate
resistance.
The gain of an amplifier tube is given
by the following equation*:
Zi
Voltage across load- -
"
Rp
+Zt
p= amplification factor
where
et= input signal
Zt =load impedance
Rp =plate resistance.
Since we have already ruled out the
advisability of a variable-mu tube, we
must find another way to vary the voltage
across the load for a constant input signal,
es. It is apparent that variations in the
load impedance Zr will change the output,
especially if Rp can be varied in the
opposite direction at the same time.
Referring to the circuit diagram of
Fig. 1, it can be seen that the controlled
amplifier is essentially a single stage of
push -pull amplification. In addition to
the plates of the amplifier tubes (6.J5) the
output transformer supplies current to
the plates of the control tubes (6P5).
The amount of plate current, and hence
the plate resistance, of the control tubes
can be determined easily by the instan4Terman-Radio Engineering, p 173.
EO
10
5
o
-S
crease in grid bias.
The actual operation is shown in the
curves of Fig. 2. The output transformer
is loaded with its nominal load of 500
ohms, which reflects a normal load of
15,000 ohms across each amplifier tube
(control tubes cut off). This is indicated
by the load line A -A across the curves.
The quiescent operating point is shown
by P in the diagram. The heavy line, of
which P is the center, shows the region of
actual operation. When a plate current
of approximately 10 ma per tube is
allowed to flow through the control tubes,
their plate resistance drops to 10,000
ohms. This causes the effective load to
be:
Rp Z! _ 10.000 x 15,000
Input signal level
vs.
At the same time the effective increase
in cathode current causes the operating
point to drop to point Q. This condition
of operation is indicated by the load line
X- X passing through point Q.
The net result of the two effects is to
reduce the amplification as the plate current of the control tubes is increased. The
only thing which limits the amount of
gain reduction possible is the approach of
expansion curves.
EEM
MEMO[ MOMO
O
EOEEMIIIMIMM
EE0
MEO
EMEMMOIPM4INIEMOEEO
E
EE
EMEUU
EUEEE
ME
E
OMMIMiaMM
EEE[EOM
EOM
E
PMd
M
MIN
MEMO M51MME
MEMO
.,'
\Ht
EXPANSION (LEFT NANO SCALE)
w
,,,smoN
-60
-40
-50
INPUT
SIGNAL
- th
IDELCO
SNW
-30
5004
-to
point Q to the non -linear region of the
plate characteristic, with resulting increase of intermodulation distortion. It is
interesting to note that the limitation in
permissible distortion is on the low-level
or unexpanded condition of operation.
There is no possibility of running into
distortion when the amplifier is wide open.
With the control tubes cut off, the gain
of the amplifier stage is approximately
20 db. This will be the maximum possible
gain, regardless of the amount of expansion used. A large number of listening
tests indicated that, if smoothly applied,
a maximum of 12 db expansion is permissible on well -monitored records. On
records where there are noticeable "steps"
in the volume changes due to too-rapid
changes of gain during recording, 8 db is
the maximum tolerable amount of expansion. All of these results were based on
the premise that the listener must not be
able to detect the actual operation of the
volume expander.
With the foregoing information as a
basis, the type of control tube was chosen,
and the operating point was selected to
give a loss in gain of approximately 8 db
when the control grids were at zero potential relative to ground. It was then found
that a grid voltage of -21 volts was sufficient to reduce the plate current to 2
microamperes. A positive potential (relative to ground) of 3 volts, applied to the
Intermodulation distortion
at various
input signal levels.
MEEEEMEEEEMMMEMMMEMMMMMMMEMMEMOEMM
%.011[
E EE E!4EE
MOMEEEE
EEEMOIIII
MONO
MEEINE EMME
EEEEE
PEE
E
EEEEMMEEMEEEEEEEEEE
M
MEMEINEEMME
E
EEEEEEE
MMEgMEEMEE=EEEE
-t0
-so
LEVEL-db
MIME
MMME
Fig. 5.
MEO
MM
O MOOMMOO
ME01E-EE
ME
MO
EP.u EE
IriMimEEMM
B
O
6000 ohms
10,000+15,000
iR7f-Zi
.t
Plate current vs. input signal curves of control tubes.
taneous grid potential. At the same time,
the additional plate current flowing
through the common cathode resistor
causes the plate resistance of the amplifier tubes to increase because of the in-
EE MEMOOMMOMOOEEO
MEE
8
Fig. 3.
uuu
M
Fie. 4.
-40
-50
-BO
SUT
of expander tubes.
,113 P4
o
400
soo
Curves showing operating range
f
a
EOMOMOEOEEMOOE
MEEEMEEEEEEEMINEE
EEEMEEEMEEMEEEEMMEEEM
EMEMEEEEEEE
PAEMMMEEEEMMOMEEEMEM
llMM
EEMEEEmirt_aim
.i
MMLír.°[A
MEniMEEEEEMENMEEE)LMEEEMMOM
IIMEMEttiCÜ=Er"ORMIUDIUUDEMEM3'EE
_
o
I
NMEEEEEMM0EMEEEOEMMEEE11
il.í:u::.1MEMMEMEEEMEEEEEEMEME
MEMMEEMEEMEMMEEMEEMEMEMEEMM
MEINEMMEMEEMEMEEEEMEEEEO
M(t r
EEEEMEMEEMEMMEEMMEEMEEEEEMEE
MEEEEEMMEEEEMMEEMEMMEEEEM
-60
INPUT
-60
SIGNAL
- db
BELOW
-40
500A
-30
-t0
6 NW
AUDIO ENGINEERING
SEPTEMBER, 1947
control grids, decreased the gain by an
additional 4 db. It is necessary merely to
switch this positive bias on and off in
order to change the expansion range from
8 db to 12 db. An additional switch position is provided to ground the control
grids, making the unit a fixed -gain amplifier with zero expansion.
Time Constants
The next problem is that of the attack
and release time constants of the system.
Again it was necessary to resort to listening tests to decide on the optimum conditions. It was quickly decided that an
instantaneous attack was necessary to
provide unnoticeable operation. In other
words, a loud sustained tone reaches its
full level at once, and doesn't build up
over a noticeable time interval, as so often
happens with volume expanders. On the
other hand, it is necessary to delay the
release time so that the expander will not
follow individual cycles at low frequencies.
Furthermore, the release time adds to the
normal decay of the sound, and, if too
rapid, will effectively decrease the amount
of reverberation. For this reason it is
necessary to have at least 2 seconds delay
for normal program material. In the
laboratory model of the expander, values
of 0.5, 1, 2, 5 and 10 seconds were provided. It was found that 5 seconds gave
good results on almost every type of
record.
Inasmuch as it is necessary to charge
the 2 mfd. capacitor in a matter of 1 millisecond or so, it is apparent that a considerable amount of power is required. A
6V6 is used to supply power to the rectifier. The rectifier, type 6H6, is momentarily called upon to supply several times
its peak rated current of 10 ma, but because of the extremely short duty cycle,
this has no adverse effect on the tube life.
It was necessary to do a considerable
amount of testing to determine the optimum frequency response of the amplifier
used to supply power to the diode. If a
wide- band -30- 10,000 cps-frequency
range is used, it is apparent that the con-
trol voltage will be influenced by the
strongest signal currents within that frequency range. Furthermore, low -frequency rumble and thumps, and high frequency surface noise and clicks, will
produce false control and consequent gain
increase. It is best, therèfore, when using
a single hand of frequencies from which
to derive the control signal, to limit the
response to from 500 to 3000 cycles. The
effective loudness is determined by this
band of frequencies, anyhow, so it is
logical to use the same region for the controlling voltage.
An interesting effect was achieved by
splitting the control amplifier into two
channels, each supplying a separate
diode. The outputs of the diodes were
connected in series, and the individual
amplifiers were arranged so that they
would overload when supplying half the
AUDIO ENGINEERING
Rear vie
of high -fidelity volume expander.
necessary control voltage. Une amplifier
passed the frequency range from 100 to
500 cycles; the other from 600 to 3000
cycles. Either amplifier alone could only
produce half the total expansion, regardless of the energy in the pass band. It required energy in both bands to produce
full expansion. This condition prevails in
full orchestra, organ, or band music, at
which time full expansion is required.
This system prevented "blasting" when a
single instrument or voice momentarily
overbalanced the full ensemble. The results were very good, but on most records
the improvement over the single -channel
system of restricted frequency range did
not justify the circuit complexity.
Expansion Indicator
A 6AF6 electron -ray tube is used as an
indicator of the amount of expansion. It
has a considerable advantage over a
pointer-type meter, because of its freedom
from dynamic error. It is adjusted so that
the eye is just closed when the 6P5 control tubes are cut off.
The last important problem is the
linearity of expansion versus input signal.
It is very important not to have any
"steps" in the expansion control. The
plate -current variations with input signal,
with the control voltage derived from the
previously described rectifier system, are
shown in Fig. 3. The change in plate
resistance is not exactly the type of curve
needed to produce a linear expansion.
This condition was greatly improved by
making the total range of expansion about
6 db more than desired, and reducing the
over-all gain bymeans of inverse feedback.
Since the gain reduction is a function of
the amplifier gain, it is apparent that the
amount of feedback will vary as the
amplifier gain is varied. The gain will be
reduced more at full amplification than at
low levels. This has the effect of straightening out the expansion curve and reducing its slope, thereby accomplishing the
desired end. This is shown in Fig. 4. The
feedback also contributes somewhat to
the over-all low distortion obtained with
this device. The curves of intermodulation distortion versus input signal level
are shown in Fig. 5. The harmonic distortion is barely measurable.
Fig. 6 shows the main amplifier circuit arranged to be inserted in a high impedance amplifier. The action is identical with that of the low- impedance unit.
It will be noticed that this circuit can
be used as a remotely operated gain con[Continued on page 3.91
Modifications needed for high -impedance input circuit. Input signal should be about
Fig. 6.
2 v. rms max.
6J5
6P5
6J5
PLATE
-
TO -
GRID
500
K
1000
.05
500
50
K
SEPTEMBER, 1947
K
ee
_
500
K
10
50
K
500
K
1000
O
6,15
6P5
O
+
300
V.
CONTROL
SIGNAL
9
Multi-Lingual interpreting Systems
C.
Describing
a
XCEPT IN OUR STATE DEPART ment. little is known in this country about simultaneous translation
from one into several languages. In
Europe two adjoining states-such as
Pennsylvania and New Jersey here
-
speak two different tongues. To countries
on the continent it is an old story, but
until the United Nations invaded these
shores little concern was ever had hereabouts for complex translation.
Until audio interpreting systems were
installed in three conference rooms last
fall at Lake Success, L. I., many thousands of man-hours were lost during
United Nations debates. More than one
hundred fifty delegates and their aides
had to sit restlessly through draggy, annoying interpretations of a speech. The
meetings would come to a complete
halt while trained interpreters repeated
*2345 Broadway, New York 24, N. Y.
A. TUTHILL*
unique and most useful application of audio equipment.
in French or English the twenty- minute
or half -hour speeches. The Russians generally spoke in their own language for
home shortwave consumption. That
called for two translations. That caused
the clock hands to travel three times as
far as necessary with simultaneous
facilities.
This nuisance was partially eliminated
one year ago when the electronic was
coupled to the human with such success
that five major languages were poured
back into the ears of the delegates as
rapidly as their colleagues spoke. As
soon as a speaker concluded, the other
delegates knew what he had said. Today
the representative from China watches
his friend from Paris orate in French while
he hears the speech in his own Chinese.
Larger conferences and committees now
schedule their meetings so as to obtain a
room fitted with multi -lingual facilities
rather than endure the imperative delay
of verbal repetition in rooms not yet
equipped.
This great time -saver proved its worth
more than fifteen years ago in international conferences. After the old
League of Nations settled in Geneva,
nine IBM -Filene-Finlay Translator audio
channels were installed there. More
recently the prosecutors of the Japanese
and German War Trials have used this
same equipment. It has been roughly
estimated by one of the judges participating at the Nuremberg trials that those
hearings would have consumed five or
more grueling years were it not for the
multi -lingual channels.
Conference Rooms
At United Nations headquarters, two
large conference rooms served by the
multi- lingual equipment seat over fifty
delegates each around an oval- shaped
table. Selector switches and a headphone to the right of each national offers
him a listening choice of five languages.
Any delegate understanding the spoken
tongue need wear no headphone, since
both the conference table and outer areas
of the room are covered by a low -level p-a
system utilizing 48 speakers, with those
adjacent to microphones being killed
automatically by relays when the microphones are keyed into the circuit. Outside
the inner circle the advisers, press, radio
and any public overflow are also serviced.
Selector switches and phones are mounted
to the sides of chairs strapped mechanically together in banks. This construction
protects cable feeders and expedites
trouble-shooting through bank isolation.
Across the end of the room opposite
the chairman, and elevated enough for
good vision, are five sound-proof, air-conditioned booths. Each seats three interpreters who relieve each other in the handling of one language since there can be no
interruption of service for human failure. In rotation laterally these booths
handle: Chinese, Spanish, Russian,
French and English. Adjacent to these is
an open booth housing rack-mounted
Original IBM nine- channel rig used at League
of Nations, Geneva, Switzerland.
(IBM
Filene -Finlay Translator.)
10
AUDIO ENGINEERING
SEPTEMBER, 1947
variable-gain amplifiers, rectifiers, patch
panels and spare equipment. An engineer
checks the entire system well in advance
of operation and constantly monitors the
channels at this point.
System Operation
Delegates in all major UN meeting
rooms have microphones before them
which normally feed all radio, television,
newsreel recording, and p-a apparatus.
It is from these microphones that the
interpreting channels start to function
(see block diagram, Fig. 1). From preamps on through a keying mixer and its
boosters a feed is linked with the multilingual input. Here it is balanced, amplified, monitored and distributed to the
headphones of fifteen highly capable male
and female interpreters.
The keying- mixer, Fig. 2, is the audio
nerve center of the entire system. The
two vu meters shown centrally above the
mixer panel indicate the behavior of the
basic regular and emergency channels.
Starting from the left of the panel, six
vertical banks of five keys each cut on or
off thirty microphones. These accommodate sixty delegates, each microphone
serving two people. The knobs directly
beneath the rows of keys pre-set and control levels. The two keys above the
seventh knob from the left are for two
interpreters' microphones when the room
is not equipped for simultaneous interpretation. The knob above these keys
controls a two -position, low- roll-off speech
equalizer. The eighth knob and its microphone key allows individual handling of
the chairman. The notable feature here is
that the gain controls for thirty -three
microphones are boiled down to eight
Fig. 1.
Fig. 2.
(Courtesy The Langevin Co.)
UN Lake Success control unit.
knobs. The white knob and key above set
the monitoring level, and provide instant
comparison if either bus behavior is in
question. The lower right knob is the
final master. Its key instantaneously
flips the output to an alternate line amplifier in case of trouble. Above the master
is the p-a control.
A strategic factor in this arrangement is
key designation. The engineer must find
the correct key for an active nation in a
split second. This calls for distinct labelling of all keys-two nations per key.
For that reason a light -colored main panel
is topped by a translucent plastic mat.
This permits rapid markings and erasure.
Full committees of fifty or more nations
are often replaced by a similar one in the
same room within fifteen minutes during
busy days.
The alertness and good judgment of
the control engineer contribute greatly to
the smoothness of the show. The chairman frequently fails to recognize delegates and they oftep do not wait for
recognition. They do not stand in con-
ference while speaking. Since some are
seated with their backs to the engineer,
and are often entirely inanimate, he must
recognize their voices to cut in the proper
microphone. When argument reaches
heated cross -fire proportions, he has
difficulties indeed.
Interpreters
Referring to the block diagram, Fig. 1,
we see the human factor linking the trans-
lating audio equipment with the original.
As in the performance of preceding units
along the line; each is dependent upon the
other. Thus, at a second point, the entire
system depends upon the human element.
The men and women who translate
these international orations perform with
the precision of pulsing circuits. Justice
to their achievement cannot be done here.
Their concentration and patience is
laudable; many are tested but few are
chosen. To relay truly the original speech,
they must introduce no feelings of their
own as they translate rapidly. Instead
they must convey the genuine intent
-
Block diagram of distribution system used for multi -lingual operation.
,-
LISTENERS
HEADPHONES
-
ORATORS
MICROPHONE
PRE -AMP
32
MIKES
OTHER
e\MIPie.7MIIIMefe7e1ei711
Niiele)AIIMMINa7faelMIMIIM7MMVUe1
1
INTERPRETED
..MAIM
KEYING
elie71e'a17Ilda1
MASTER
BOOSTER
MASTER
TO
MOVIE,
RADIO,
RECORDING,
VISION
B
BUS
FEEDS
CHINESE
SPANISH
BOOSTER
BOOSTER
BOOSTER
AMPL
AMPL
AMPL
RUSSIAN
FRENCH
AMPL
eti
15
PWWV
INTERPRETERS " THE
V V
VV
3
PROCEED-
OTHER
PRESS,
DELEAND
ENGLISH
LANGUAGE
FOR EACH
HUMAN LINK"
vV
MULTI - LINGUAL
F
60011
TO
Ali
ISOLATION
AMP
TELE-
PA.
BOOSTER
VC.
LINE AMP
60011
BOOSTER
AMPL
INGS
GATES,
RADIO
nn
AMPLIFIER
15
INTERPRETERS' HEADSETS
MONITOR
LINES-
AUDIO ENGINEERING
SEPTEMBER, 1947
11
supported by enough bona-fide expression to hold attention.
Literal word -for-word interpretation
being an impossibility on account of
idioms, these multi -tongued people lean
upon their own judgment for an equivalent phrase which will convey the identical
meaning in another language. Despite
all this, these human time -delay relays
lag but slightly. They often finish within
a second or two of the original speaker.
These interpreters do an admirable job.
The audio channel picks up again as the
interpreters talk into breast -type ribbon
microphones. From this point forward
the speech is boosted conventionally
through variable -gain amplifiers constantly monitored. From the rack mounted regular and spare amplifiers the
interpretations are distributed over audio
Iines to several hundred selector switches
and their headphones. With all seats in
the house covered, the audio system is
completed.
Radio System
An innovation of the above method
deserves some treatment here even though
it is intended to employ audio systems at
the permanent UN headquarters in New
Fig. 4. IBM Filene -Finlay delegate desk mike.
York. This is a small radio installation,
as shown in Fig. 3.
In several IBM -Filene- Finlay installations and in the one which has served
the World Radio Conference at Atlantic
City, N. J., low- powered frequency -modulation transmitters have been installed.
In the above case ten crystal -controlled
low-frequency FM transmitters, all
coupled to one loop antenna encircling
the auditorium, are operated with carriers spaced twelve and one-half kilocycles apart. The channels run from 100
kc upward and provide adequate field
strength although operated at low power.
In this installation a great amount of
audio wiring is avoided. Only the microphones and signal lights need be wired.
Instead, 1800 FM receivers weighing only
one and one-half pounds each are provided for the delegates. They resemble
and are worn like a flat camera, the
shoulder straps doubling as loop antennas. The receiver dimensions are actually
1M x 434 x 5M in. Not necessarily worn,
they may even be buried under the delegate's papers while he works at his post.
Each set has an off-and-on switch, a
selector dial for the desired language and
a volume control Each employs three
midget tubes and affords two stages of
radio amplification. a -v-c diode detector,
and a pentode output stage permanently
connected to headphones. Internal high capacity war-developed batteries will
handle one week of daily operation without replacement.
Delegates enjoy the freedom of being
able to move about the hall, comparing
notes with colleagues while still hearing
speeches interpreted. When they want
to talk, however, they must return to their
seats and use a desk-type microphone,
[Continued on page 46]
TRANSLATORS' BOOTHS
RADIO TRANSMITTERS
SPEECH AMPLIFIERS
TRANSMITTING AERIAL AROUND CEILING
IP
TRANSLATORS' SIGNAL LAMP
TO SIGNAL SPEAKER TO PAUSE
I.
RADIO RECEIVER
FOR EACH MEMBER
i
Y
TABLE MICROPHONES
SPEAKER'S MICROPHONE
'T;
CHAIRMAN
S
OR
SPEAKER S LECTER
Fig. 3. Diagram showing operation
of multi -lingual translating system. Speaker's
words are conveyed by radio to interpreters, each of
whom translates the speech into a different language.
12
AUDIO ENGINEERING
SEPTEMBER, 1947
High- Frequency Equalization
For
Magnetic Pickups
C.
G. McPROUD'
Methods of using shunt capacitance to provide low-pass filter action
and NAB compensation for users of Pickering and G -E pickups.
ONCE THE OUTPUT of a magnetic
pickup has been equalized so that
the response up to the transition
frequency is a realistic reproduction of
the original recorded material, the listener
next turns to the other end of the spectrum in an attempt to eliminate noise or to
match the recording characteristic. The
methods of equalizing the low-frequency
response have been covered previously t,
and such methods are equally applicable
to both the General Electric Variable
Reluctance pickup or to the Pickering
cartridge. These pickups are of the high impedance type, designed to work directly
into the grid of a vacuum tube. The
electrical characteristics are similar, and
the same methods of high- frequency
equalization can be applied to either pickup.
In order of importance, the three
principal reasons for equalizing the high frequency end of the audio band are:
1. To reduce surface noise
2. To eliminate distortion
3. To compensate for the recording
characteristic
The reasons for this order will be apparent from the discussion.
The first of these listings is the one most
noticed by the layman, who is often the
person for whom the equipment is being
designed. He is apt to be conscious of
noise much before his notice of distortion,
and favorite records-usually played
more than others -are likely to be the
most worn and consequently the greatest
offenders.
The second reason becomes more apparent as the ear acquires training. The
effect is noticed on many records as a
definite "muddiness" and "smearing" in
the high -level, full- orchestra passages,
probably as a result of cutting at an
average level which is too high for the
amplifier equipment, the cutting head, or
the record material. Consequently, when
peak passages are reached, system distortion rises considerably. A higher over-all
distortion which remains constant is less
* Managing Editor, Audio Engineering
t Audio Engineering, p. 10, July, 1947
AUDIO ENGINEERING
than low distortion
objectionable
throughout parts of the recording, with
great distortion on high -level passages.
Compensation for the recording characteristic is likely to be considered by
many as the most important of the three
listed reasons. Without diverting into a
discussion of recording characteristics
-
_
Ro
T
Ro
(A)
Cartridge.J
(B)
Fig. 1. (A) Conventional L- section low -pass
filter configuration. (B) Circuit of magnetic
pickup shunted by capacitor C and terminated
by resistor RI., feeding grid of amplifier tube.
which are definitely not standardized
among the various manufacturers, or even
in one manufacturer's own studios-let it
be said that some degree of high- frequency
equalization is generally employed in the
recording studio, as measured from the
microphone to cutting head. However,
this does not necessarily mean that the
individual record that you buy is going
to reproduce with that increased h-f
response. It is recognized that one
stamper is capable of producing about
1,000 records before it is discarded. The
first record may be excellent, but what
about the 999th? One more record and
the stamper will be discarded. In the
writer's opinion, very few commercial
records are on the market today with any
excess of high- frequency response-cer-
tainly not to the extent that they cannot
be compensated for easily by usual "tone
SEPTEMBER,
1947
control" circuits. Therefore, with the
exception of compensation for the NAB
characteristic, no "sloping" circuits will
be discussed. Reproduction from the
record will be assumed to be correct as
far as frequency response is concerned,
and since both of the high -quality magnetic pickups in growing use are capable of
performance up to 10,000 cps, it will be
further assumed that the electrical output from the pickup is directly proportional to the velocity of the stylus tip.
Correction Methods
The simplest method of reducing the
response of any electrical circuit to a
desired maximum frequency is by means
of a low -pass filter. In proper designs,
practically any shape of curve can be obtained. However, in high -impedance circuits the series inductances reach large
proportions, and they do pick up hum.
Consider a simple low-pass filter such as
that shown in Fig. IA. It consists of a
series inductance and a shunt capacitor.
The equivalent circuit of a magnetic
pickup is essentially a generator in series
with an inductance equal to that of the
coil, and with a resistance equal to the
resistance of the coil. Thus, if a capacitor is shunted across the output of the
pickup, as in Fig. IB, the net result is that
a low-pass filter section is formed. It is
only necessary to choose the correct
value for the capacitor and the correct
value for the terminating resistance,
after the cut-off frequency is selected.
Filter Circuits
The configuration for the low -pass filter
used in connection with these magnetic
pickups is now seen to consist of the
inductance of the pickup itself shunted by
a capacitor, with the combination working into a suitable load resistance. Without considering the actual cut-off frequencies, the general circuit is developed
first from the formulas for a constant-k
low-pass filter, which will provide the
values for C when the inductance L and
the cut-off frequency are known. The
cut-off frequency, fe, is that at which the
response is down 3 db.
L -P
13
0.1
given by the simple relations
R (ohms) = 1.2 fc for the Pickering,
.07
0.I
.05
.07
and
R (ohms) =0.9 fc for
the General
Electric.
The response is flat up to approximately
80 per cent of the cut-off frequency, is
down 3 db at fc, and decreases at the rate
of approximately 15 db per 'octave
above fc.
Any resistance in series with the capacitor does not affect the curve up to a frequency of approximately 2fc. Above this
point, the curve tends to flatten off. Therefore it is recommended that the capacitor
be connected directly across the pickup,
and shunted by the correct load resistor.
For flexibility, and to provide for wide open response when record surfaces and
distortion content permit, a switch may
be arranged to connect either of two or
more capacitors across the circuit as
desired, each capacitor having its own
resistor permanently connected across it.
Using this arrangement, cut-off frequencies anywhere in the band can be
chosen at will. For ordinary use on commercial shellac pressings, a three -position
switch providing cut-off frequencies of
3,500 and 6,000 cps, together with one
wide -open position, has proven quite satisfactory, although a more elaborate arrangement would obviously give a wider
range of control.
.05
.03
.03
.02
.02
i
IA.
W
.0I
.007
.01
.005
.007
U
H
.005
.003
.003
.002
.001
2000
fc
Fig. 2.
-
3000
FREQUENCY
5000
of
10000
NAB Roll -off
DROOP (cps)
Chart for determining capacitance to be shunted across Pickering Cartridge or General Electric Variable Reluctance Pickup to provide low-pass filter action.
By further rearrangement of the formulas, the value for the terminating resistance is given in terms of L and C. To
simplify the determination of these capacitance values, they are shown in chart
form in Fig. 2. The inductance of the
Pickering cartridge is approximately 120
mh, while that of the G -E Pickup is approximately 100 mh. To use the chart,
determine the frequency at which a 3-db
droop is desired; follow the frequency line
upward until the diagonal line is reached.
For the Pickering, read the correct
capacitance value from the left edge of
the chart; for the G-E, read the values
at the right.
Response Curves
A typical response curve obtained by
the use of a properly terminated capacitance shunt across the output of a magnetic pickup is shown by the solid curve of
Fig. 3, while the dotted curve shows the
effect of operating the combination into
a load resistance equal to three times the
optimum value. The smoother response
of the solid curve is much to be preferred,
and the use of a filter with sharper cut-off
characteristics is generally undesirable
for high quality audio circuits. Empirical
values for the terminating resistance are
14
3 -db
7000
+5
o
5
-10
-15
20
-25
1000
3000
5000
10000
FREQUENCY- CPS
Fig. 3. Shape of response curve for fc = 3850
Dotted curve indicates peak due to
cps.
improper terminating resistance.
SOURCE
tT
Z1
'f
Z2-
LOAD
cuit
is
following:
75 (Zr +Ze)
ZiZ,s
ZO
Fig. 5.
Professional users of these pickups
often have need for a circuit which will
provide the proper roll -off for the high
frequencies to permit satisfactory
reproduction of transcriptions employing
the NAB recording characteristic. This
curve is shown by the solid heavy line of
Fig. 4, together with the equalization
provided by the low-frequency equalizing
circuits previously described, and the
simple capacitance droop to correct for
the high end. The resulting range of
control provides for a response at 100 cps
anywhere between +6 and -2db, together
with an approximately flat response
above 1,000 cps.
The droop resulting from the shunting
of a capacitor across a line of a certain
impedance is a gradual deviation from flat
to a loss of 3 db at approximately three
times the frequency at the point of deviation, with a further droop at arate of 6 db
per octave above that point. However,
this does not apply to capacitor shunted
directly across an inductive source such
as a pickup. The capacitor must be shunted across a circuit which is essentially
resistive. A simple method of determining
the correct capacitance to provide equalization for the NAB curve is given by the
Impedance Zo of any point in a circomposed of Zi and Z2 in parallel.
where Zi and Z2 refer to the impedances
indicated in Fig. 5. This is a modification
of the standard formula employed in some
AUDIO ENGINEERING
SEPTEMBER,
1947
broadcasting plants which is C = 100/Z0,
where Z0 is the impedance at any given
point in a circuit. Thus, the impedance
at any point on a 600-ohm line equals 300
ohms, since the source and load-each
being 600 ohms-are in parallel. The
value of 100 gives a roll -off which is somewhat greater than required to provide
flat response.
Adaptations to Preamplifier
The preamplifier suggested for low-frequency compensation is readily adapted
to provide the required roll -off to compensate for the NAB curve. The logical
place to insert the shunt capacitor is
across the output circuit, at which point
the impedance consists of the load resistance shunted by the plate resistance,
and both shunted by the grid leak of the
following stage. Assuming that the grid
leak is 1 meg. or more, the impedance at
this point is numerically of the order of
40,000 ohms, resulting in a shunt capacitance value of 0.0018µf.
When the connections from the preamplifier to the succeeding circuits must
be fairly long, the capacitance of the
shielded lead is apt to cause a droop in
the response. This may be corrected
easily by the addition of a capacitor across
+20
w:ií ,"`''''
/
i/
EQUAL ZER
BOOST
,'.
gal
ME
awil llgl
_i?'
ill
-P211111-----111
IIIII
IIIIIMP'
III
III
o
10
-20
1!
"" (`,\11111
RANGE OF.
L-F CONTROL
50
70
100
200
1IIl11I
300
FREQUENCY
Fig. 4.
RESULTANT
G
111 °a°
iIll
NPe
500
-
2000
1000
CYCLES
PER
°4p4°,r4F
R°op
3000
5000
10000
SECOND
NAB recording characteristic related to equalized preamplifier and to capacitance roll -
of to
provide control over low- frequency response and
R4 in the circuit shown in Fig.2 in the July
issue. Assuming a value of 0.27 meg for
this resistor, the capacitance to be placed
across this resistor is equal to 0.59/î,
where f is the frequency at which the response is down 3 db. This same method
of equalization may be used to bring the
a
flat high- frequency response.
high-frequency response of the G-E pickup
up to flat at any desired frequency by the
application of this formula, if the particular cartridge droops appreciably. The
output of Pickering cartridges is held to
a tolerance of ± 2 db to 10,000 cps, and
further correction should not be necessary.
The new Western Electric 1304 type reproducer is shown here being used in conjunction with Western Electric 25B speech input equipment.
Ultrasonics in Li uids
S.
YOUNG WHITE
The author discusses important indussrial applications of ultrasonics in liquids.
IN
GOING FROM GAS to liquids in
our consideration of ultrasonic energy,
we find a markedly different picture.
The pressures increase over fifty times,
and the motion is about one-fiftieth
that in gas. We cannot hope to obtain
many millions of "G ", but must be
happy with perhaps a maximum of
100,000 or so, corresponding to about a
thousand miles/per sec /per sec. for a
few kilowatts per cm2.
Provided we can generate these powers
and effectively couple them to a load in a
production line, what will be the desirable
effects?
Emulsification
One of the most pronounced is emulsification. Practically any liquid can be
so mixed with another that stable mixes
are formed, especially if there is a small
amount of gas in one liquid, such as
mercury and milk,* aluminum and lead
(in melted form), oil and water and so
on. This is a high -power action, and in the
clearest case of mercury and milk is
rather easy to visualize. The mercury
stays on the bottom, of course, by
gravity separation. Intense short waves
are set up on the surface of the mercury,
like waves in the sea. These work up
to a sharp peak in the milk, and break
off in little dollops, as a sailor would say.
If you try this, very often about 20%
will separate after a few days, the rest
remaining in suspension.
A very ingenious suggestion has been
made, to use this effect for laundering
clothes. The grease will emulsify into
the water without soap, and the solid
dirt particles will be separated by the
terrific accelerations. This is the type
of ingenuity we must cultivate in our
thinking of ultrasonics, as the possible
fields of use are limited only by our imagination.
The pronounced effects noted in emulsification teach us any two liquids of different densities will be thoroughly and
quickly mixed, one probably being re-
*The homogenization of milk is an example of the emulsification power of ultrasonics, and lead and aluminum will stay
mixed after cooling if "emulsified" this
way. It is obvious that chocolate and
many similar substances can be emulsified
to give a finer grained, table mix.
16
duced to micron size at once, and then
these particles will be violently agitated
within the other liquid. This probably
explains the almost universal speeding up
of chemical reactions noted between two
liquids. If they are of different sonic
impedances, enormous work can occur at
the boundary layers, with marked effect
on the electric charges of the interfaces,
and the production of high temperature
locally in the boundary. It must be noted
again that our knowledge of these boundary layer conditions is not too good in
most cases, so again the best test is to try
it out.
While high power will tear things up,
especially with cavitation, at quite low
powers the reverse sometimes takes
place -colloidal size solid particles are
formed into relatively large groups.
This reverse effect can be looked for either
at reduced power or often by going to a
lower frequency at the same power as
impedance of this layer was about the
geometric can between the solid magnetJstriction transmitter and the water,
so it formed an impedance match between
them, and much more energy was coupled
into the water, giving us about four times
the usual signal. This energy density was
sufficient to release gas from the water,
however, so in ten seconds or so the gas
layer became thick enough to form an
almost perfect energy reflector, and the
signal fell to zero. When we boiled the
water, the gas layer did not form.
This coagulation effect should have
wide commercial use. While it seems
possible to coagulate paper pulp on a
flat plate to make a very dense paper,
the adhesion is so high that it would
be almost impossible to remove it from
the plate. If you wished to remove the
particles by coagulating them out of the
liquid you could continuously scrape
them off a diaphragm. In some cases,
the layer only forms to a certain thickness and does not increase thereafter,
Steel
so you could continue to put energy into
spring
the liquid through such a diaphragm,
and let the coagulates settle after they
RBioc
had flowed through the apparatus.
If there is considerable gas in the liquid,
\Oscilloanother interesting effect takes place.
ti on
When you release pressure on such a
liquid, gas will bubble out. The rarefaction portion of an ultrasonic wave will
likewise release bubbles, as that portion
of the wave lessens the over -all pressure
Fig. 1. Gear and biased spring.
locally. Now if fine particles are in the
liquid, a little gas overcoat will form on
each reduces acceleration. This coagula- each particle. If the small particle is a
tion can sometimes be a serious nuisance, bacterium or yeast cell in beer, for inwhen diaphragms are used, as a coagulate stance, it will in many cases be inhibited
forms on the diaphragm which is difficult and stop growing, as it must be in contact
to scrape off.
with liquid to absorb food and excrete
The writer was given an interesting waste. Many bacteria are unaffected by
lesson in impedance matching while this, however, so we would have to go to
measuring the transmission loss of sig- cavitation to control them. Apparently
nal through a two-foot path of water no organism can withstand high-level
containing about 1% paper pulp, the cavitation, being torn to pieces rapidly.
liquid being tap water (not degassed). The gas bubble may be useful in flotaWork had been done at an input signal tion processes, however, such as ore
level of about 50 milliwatts per cm2 and separation. If the bubbles are allowed to
checked quite well. The power was then escape, the liquid may be degassed, of
increased to about 2 watts /cm2. The course.
received signal rose to a very high
Another useful effect is that the enorvalue for about a second, and in ten mous pressures and accelerations will
seconds fell to zero. Investigation showed force liquids to penetrate solids; for exthat a layer of dense pulp had rapidly ample, the Germans found that dyes
formed on the diaphragm. The sonic were forced into the grains of photo-
AUDIO ENGINEERING
SEPTEMBER, 1947
graphic emulsions so they were colored
to a much higher degree. Perhaps our
wives can enjoy the doubtful blessing
of bright purple nylon stockings. But,
after all, most inventions have some
drawbacks.
The Germans also teach us that grain
size of photographic film emulsions can
he reduced by a factor of six, which
means six times the resolution in the
picture. While the emulsion is being
manufactured the silver haloids tend to
agglomerate into groups. Ultrasonic acceleration breaks these up.
The main application of ultrasonics
to metallurgy is while the melt is cooling.
This is so important that we shall delay
discussion until the next article which
will be on solids.
1800 beer lines in the country, and many
thousand canning lines, so this is a good
market.
We must introduce our energy over
a large area of a glass container at once,
as too high an energy density will cause
the glass to flake off in pieces a few
millionths of an inch in size. The doctors
may give us glass-lined stomachs some
day, but it is doubtful if they will use this
method. The tin may be broken loose
from the steel in the "tin" can if we hammer it too hard in one place. The can is
used only once, but the bottle may go
through the line ten times, so we must he
more careful of the bottle. We cannot
Magnets
Sterilization
Sterilization of food in containers is
a very practical problem. Canned food
is usually placed in pressure cookers and
the inmost particle brought up to 250 ° F.
or so. This breaks the continuity of the
production line. This also requires girls
to load and unload the rack which must be
handled by an overhead crane, which then
inserts it in the pressure cooker, which
must then be bolted shut and steam
applied. Every other operation can be a
continuous process.
While some help in sterilization may
come from the formation of bubbles
around the bacteria, the vast majority
of materials must be cavitated. Energy
of about ten or twenty
watts /cm2 are required. This must be
applied so that every part of the liquid is
cavitated. To do this, a very confused and
erratic wave pattern with multiple reflection at all angles must be used, or
the high- frequency pattern must be
shifted by moving the container or by
frequency-modulating the generator.
In the case of a usual cylindrical "tin"
can the analysis of the wave pattern is
fairly simple, and we have two perfect
diaphragms at either end to work with.
Bottles are oddly shaped, and we may
have to rotate the bottle end for end. We
cannot well take more than a second to
obtain our results, as we must handle
several hundred a minute. It costs little
to break the line into several parallel lines
and then bring them back into one line
again.
The type of food we can sterilize is
probably limited. Simple liquids such as
orange juice should be easy, but cans
containing large solid objects like plums
may give difficulty. We would tend to
homogenize the contents so the solids
would not settle.
densities
A particular advantage is that heat
not necessary. All beer is now pasteurized at 145 degrees minimum after
bottling. The taste is somewhat affected
by the destruction of enzymes, so many
is
people prefer keg beer to beer in containers. This we would avoid. There are
AUDIO ENGINEERING
)2
FGop
i_
=
_
.002
(/
Mognets
Fig. 2.
Rotary magnetic generator.
depend on the wall thickness of the
bottles in calculating, as they vary two to
one in thickness. Tin cans are very uniform, of course.
We have several factors that help us
in generating and handling the ultrasonic
energy in liquids. From an impedance
match view, we can insert a solid diaphragm in the path of the energy when
we are working with liquids. The center
of the diaphragm formed by the tin can
will reflect only about 10 or 15 per cent of
the energy, if we have water on both sides.
If we hit it with sonic waves from a
turbojet unit the reflection would be
about 99.9%. Of course, it does not follow
that reflected energy is lost
has to go
somewhere, and both steel and water have
Q values in the thousands under ideal
conditions. So the energy in the steel or
water would simply wash back and forth
as wattless component and the field
strength would rise until the load took
all the energy, except energy lost in the
walls of the chamber.
The Q of the load is of great importance. We use Q in the radio sensethe ratio between the energy returned by
elasticity and the energy lost in heat and
friction. In sea water 24 kc will travel
about 100,000 feet before falling to half
value, so in an enclosed space with perfectly reflecting walls the resonant buildup would be very high. So, for high Q
loads, the energy density in the load can
be very high with rather small driving
power. There is bound to be some loss in
the walls of the chamber, though.
SEPTEMBER, 1947
-it
A low Q load example is cornstarch
freshly wetted in water. With about
40% cornstarch the Q was about 6 at
200 kc, so the resonant build -up is small.
This is the worst case in the writer's
experience. It showed the energy was
going into work at least in the boundary
layers of the starch particles, and was
probably doing something to them.
Now we have shown a few attractive
possibilities of commercial use in liquids,
how do we generate and apply the power?
Quartz Crystals* *
Quartz crystals will give a hundred
watts or so. They are a natural product
and are very subject to breakage when
operated at above 100 watts. YoÙ also
need a one-kw transmitter to 'drive them,
and they must be immersed in oil. As a
serious power source they are useless.
The PN (Primary ammonium tartrate)
are more uniform, being artificial, but
are rather limited in power also, because
of their low dielectric constant.
Magnetostriction seems promising until
you try it. At high power, the tubes
crystallize and break. The magnetostriction rods must be in the form of tubes so
they may be water -cooled. They must be
slit for eddy currents, and the slit filled
with some cement so the water will not
leak out. Nickel tubes from old Edison
storage batteries are pretty good. For
maximum activity, heat for four hours in
a hydrogen furnace at 1485 ° F. and let
cool all night. Then if you put them in a
lathe and polish them with crocus cloth
they lose about half their activity, as they
are very sensitive to mechanical shock.
While special magnetostrictive materials have been observed to have an
efficiency of 9 %, in general you must
put in 30 kw to obtain 1 kw out, from
many magnetostrictive tubes, and then
the short life makes them a nuisance.
Let us take a look around for some
other phenomena that we can utilize with
the hope of securing a kilowatt or so, and
with possibilities of a hundred or a thousand. It is entirely possible that the
fields of application of ultrasonics are so
wide that a whole series of different types
may be the answer, rather than one
universal type.
The simplest type to consider is shown
in Fig. I. This is a gear of, say, 80 teeth
* *While
a steel diaphragm will be
cavitated by a power density of two watts
/cm2, quartz will operate under oil at ten
watts with no sign of cavitation. This
is evidently due to the much higher compressibility of oil, which must be employed in any case to provide insulation
to prevent flash -over. It seems reasonable
that quartz would cavitate even more
readily than steel if used in water, which
is of course impracticable. So oil may
well be used as coupling to a bottle or
other load that cannot otherwise take
high energy densities.
17
rotating 300 rps- practically a duplicate
of the rotor in the turbojet unit. We can
then take a spring steel strip and hold it
against the periphery of the gear. The
gear teeth are only capable of giving a
push, and the elasticity of the spring must
provide the return part of the cycle. We
do not have to try this, as common experience tells us such enormous power
density will be applied to the spring the
end will burn or be hammered off. The
gear teeth will not be damaged very much,
as the work is spread among 80 of them,
but is all concentrated in a single spring.
Magnetic Units
The next obvious thing is perhaps a
magnetic unit similar to that shown in
Fig. 2. Here we place 80 Alnico magnets
on a rotor revolving at 300 rps as before,
and then arrange 80 similar magnets
around the outside, and have a small gap,
say two mils clearance. Each time the
magnets line up they will pull toward
each other. The idea is to allow the outer
magnets to mechanically oscillate radially, make them, say % wavelength
long so they will resonate and allow us to
take the energy off their outer ends.
When we take two Alnico II bar magnets M inch long and one-eighth inch
cross -section and measure their pull with
a 2 mil gap we obtain the disappointing
figure of three ounces. Since one watt
per cm2 is over 100 pounds per square
inch in metal, this doesn't look too
promising.
The average quarter -horse power motor
runs at about two watts/cm2 of rotor
area, but some water -cooled highspeed
1600-cycle induction motors run at
40 watts /cm2. Perhaps by going to
electromagnetic fields and long rotors
we could generate one or two kilowatts
this way, but it is difficult to mount the
outside magnets with sufficient rigidity
to allow precise clearances and yet
transmit lengthwise oscillations without
excessive loss in the mounting. There
are large lateral forces also that must
be controlled, and under this type of
mechanical oscillation many magnetic
materials lose their magnetic properties.
This looks a little discouraging at this
time. The writer has the uncomfortable
feeling some bright young man is going
to make this work some day.
Jets
It is so easy to obtain high power
densities in the turbojet unit, let us
consider jets for a while. In a perfect
gas orifice if we start at zero gauge
pressure and gradually increase the
pressure, we find that at 13.6 pounds
in air we have attained sonic velocity
in the jet -1100 feet 'sec. No matter
how much we increase the pressure, our
velocity remains at that value.
Taking 15 -pound pressure to allow a
little for losses and imperfect orifices, we
find a one inch circular jet will discharge
over 400 cubic feet a minute at 15 lb.
gauge. This is 21.5 h.p. adiabatic, a very
respectable amount of power. Now we
see what the aircraft boys are up against
when they want a sonic velocity wind tunnel-50 feet diameter requires 7,700,000
h.p. at 100 per cent efficiency in the compressor, or ten million in practice.
If we consider putting liquid through
the turbojet unit some rough calculation
shows it would take about 300 h.p. to
turn it over, with instantaneous pressure of several hundred thousand pounds
per sq. in.-obviously impossible at the
velocities we must use to obtain the high
frequencies of ultrasonics.
Push only
About
(B)
V3
cycle
4- --,
)Y4
.
Water bullet
(A)
Fg. 3. A shows nozzle N designed to convert water pressure energy into kinetic energy. B, to
inject energy into a sine wave by pushing, force must be applied for about one -third the cycle.
18
There is no difficulty in designing
a nozzle for liquid, and little trouble
pumping the liquid at a pressure of
several hundred pounds to give us liquid jet velocities of several hundred
feet per second from that nozzle. When
we direct such a stream of liquid on a
load it will of course give us a d.c., or
unidirectional, push. If we could inter-
rupt this stream at regular intervals and
high frequencies, we would be shooting
bullets at the load giving a regular series
of pushes. The restoring force would
come from the elasticity of the load,
which would be the transformer necessary
to turn pulsating d.c. into a.c.
We have found we cannot do this
with a solid rotor, but perhaps there is
another attack. We have all heard a
water tap giving a 500-cycle note when
at some critical point when almost turned
off. We have also seen a mixture of
steam and water come out of a pipe,
and been impressed with the enormous
energy shown in the gurgling noise it
erratically gives out, and the velocity
of the random slugs of water it shoots
out. Perhaps we can discipline this effect in some manner.
Figure 3A shows a nozzle N designed
to turn all the energy in the water pressure into kinetic energy or flow energy.
This flow is continued in the small pipe
P, say 60 mils (1 /16th inch) diameter.
The air input A is shown coming in the
side, fed by the turbojet unit TJ.
The turbojet can be considered as
metering or measuring a known slug
of air each cycle, say 24,000 per second.
This slug can be made to come out at
almost any velocity and pressure we
want. Thus we inject a disciplined
bubble into the liquid stream each cycle.
The water and the bubble travel
downstream together, the air separating
the water into bullets. Figure 3B also
shows that to inject energy into a sine
wave by "pushing" only, we must push
only approximately a third cycle. Sc)
the bullets must be separated by twice
their own length.
This means the water comes out of
the nozzle at 100 ft,'sec., and after
being chopped by the gas, must come out
at 300 ft /sec., so about all the energy
must come from the gas since the kinetic
energy in the water bullet is proportional
to the square of the velocity.
Concentric Injection
Since we want a rapid and definite
insertion of the bubble, we can resonate
both the water solid stream and the air
injector pipe, so that when the wave
in the water is minimum pressure, the
wave and slug of air, is at maximum
pressure. We should continue the water
pipe far enough so the bubble can expand to twice the dimensions it had when
inserted -this takes about an inch of
pipe.
[Continued on page 45]
AUDIO ENGINEERING
SEPTEMBER, 1947
Design of Electronic Organs
WINSTON WELLS*
In this series
PART II
of articles, the author presents
a
thorough
discussion of the design and operation of electronic organs.
THERE ARE TWO BASIC METH ods by which musical tones may be
produced. One is by means of oscillating systems; the other involves the use
of scanning devices.
All the existing non-electronic musical
instruments, as well as the vocal apparatus of animals, consist of an oscillating system with a source of energy, a
filter network and some form of radiator
for communicating the oscillations to the
atmosphere.
In addition, most musical instruments
are provided with means for varying the
pitch of the sound, in order that the intervals of the musical scale may be played.
This is accomplished either by varying
the natural period of the oscillating
system, as in the case of the flute, or by
providing separate fixed oscillators for
each note of the scale, as is the practice in
the organ and piano.
Percussive instruments, such as the
piano, generate damped waves. The
hammer supplies a pulse of energy to the
resonant system formed by the string
under tension, and the latter continues to
oscillate with diminishing amplitude
until the energy is expended.
Wind instruments are continuous wave
generators. String instruments, such as
the violin, may usually be regarded as
continuous wave generators when played
with a bow, since the latter supplies a
power pulse at the same phase of each
oscillation of the string.
Timbre
Timbre, the complex quality which
distinguishes the sound of one instrument
from that of another, is the combination
of effects produced by:
1. The frequency and amplitude of the
partials as related to the fundamental
frequency of a tone.
The dynamic characteristic of the tone
throughout its duration.
3. The deviation from the mean pitch of
the tone, throughout its duration.
4. Changes in amplitude and character of
partials, throughout the duration of
the tone.
5. Beats and higher order resultants,
produced through the interaction of
partials.
Of all these factors, the dynamics and
harmonic structure of the tone are of
greatest significance. (The term "harmonic structure" may be applied, even
2.
*307 E. 44th St., New York City
AUDIO ENGINEERING
when the tone contains partials which are
not harmonic.)
Aside from vibrato and celeste effects,
it is seldom desirable to purposely introduce deviations of pitch in a note. Some
musical instruments, particularly horns,
are characteristically incapable of holding
the same pitch through the inception and
termination of a note, but the musician
makes every attempt to do so in the
DYNAMIC
CHARACTERISTIC
KEY
KEY DEPRESSED
FREQUENCY
RELEASED
DEVIATION
.5Y
stop which will yield a tone having the
average characteristics of the instrument
to be represented, but there is little or no
means for modifying these characteristics
during the rendition of a passage.
It is important that we bear the foregoing facts in mind when designing an
imitative stop for an organ. We must,
first of all, familiarize ourselves with the
character and purpose of the instrument
to be imitated, then analyze the effect
produced. We may then pick out the
principle characteristics responsible for
this effect, and devise means for simulating it. Fig. 1 shows a convenient form of
notation to use in the analysis and synthesis of musical tones.
Design Requirements
Not all of the stops on an organ are
imitative of other instruments; in fact,
-55
ABUNDANCE
OF
PARTIALS
TIME
IN
SECONDS
HARMONIC
F:o
STRUCTURE
F30
Fso
Unique chart shows complete tonal
character of a note, at a glance! This particular
chart was made of a tibia clausa pipe, tuned to
middle C, speaking on 15 inches of wind, and
with the vibrato on. It will be noted that the
frequency shift caused by the vibrato is accompanied by a concurrent change in amplitude
of signal, and in harmonic content.
Pg. 1.
majority of instances. The slide trombone and instruments of the violin family
are capable of changing their pitch
through a continuous spectrum, rather
than by steps, but it is obviously impossible to perform these "slurs" and "slide
glissandos" upon a keyboard.
When attempting to imitate another
instrument upon the organ, it is desirable
to reproduce the effect of that instrument
as faithfully as possible, yet there are
limits beyond which we cannot go in
practice. The violinist is free to render
his artistry upon a single melodic passage
(one note at a time), using both hands to
shape the tone in endless ways. The
organist, on the other hand, may draw a
SEPTEMBER, 1947
quite a few organs have no imitative
stops at all. A well -designed instrument
must possess a primary group of stops
which sound good when full chords are
played upon them. In addition to these,
it must have some stops of sufficiently
interesting and contrasting tone color to
be useful in the rendering of melodic
passages against a background of the
first group. A third group is sometimes
added, the stops of which are seldom
played alone but, rather, in combination
with other stops for the purpose of modifying their timbre. And, finally, there are
the imitative stops, which are most frequently used for orchestral effects.
Since the organ is most often played
with several stops speaking simultaneously and, since the full organ is necessary for
climatic effects, it is of utmost importance
that all except the solo stops be capable
of blending in tone when played in
unison. This subject will be discussed in
detail at a later date. For the present,
it will suffice to say that a good ensemble
is most easily attained through the blending of tones whose partials are true harmonics, the partials diminishing in amplitude with their order.
It is apparent from the foregoing discussion that there is no element in the
production of musical sounds by accoustical means which cannot be simulated
electrically. Furthermore, the electrical
approach gives us considerably greater
control over the characteristics of the
tone producing elements, since we may
19
work with relatively pure circuit elements
(inductance, capacity and resistance)
rather than lumped circuits. Whether we
build an exact electrical counterpart of an
acoustical instrument, or whether we
utilize a novel circuit to yield the same
tonal effect, is more often a matter of convenience and economy than of physical
necessity.
In the case of non-electronic musical
instruments, the basic signal is generated
by an acoustical or mechanical oscillator,
its harmonic structure being modified by
the resonator, which behaves as a filter
network and, usually, as an impedancematching device between the oscillator
and the atmosphere. The violin is an
Circuit diagram of
Fig. 2.
A4
an
ideal example of such a system. The
flute is an instrument in which the
resonator serves as the oscillator tank
circuit, filter network and radiator, all in
one.
We may produce a perfect sustained
violin tone by feeding the output of a
vacuum tube oscillator into an electrical
filter, then into an amplifier and loud
speaker. We may provide an ideal
vibrato by modulating the oscillator plate
supply at about seven cycles per second.
We may key the note by means of a
switch, placed anywhere in the audio
circuit although, with most oscillators,
greater realism may be attained through
keying the oscillator plate supply. This
experimental electronic organ.
OSCILLATOR
PITCH
440
=
CPS
6N7
I
73 K
1M
FTC-
UPPER
*-MANUAL
100 K
.005
.005
MANUAL
BUS
BAR
.005
l
10
10K
713
K
1M
10K
4K
.25
q
50K
CONTROL
.1
50K
100 K
10
.SM
3K
BALANCE
.5
100K
50 K
1.5 K
20
50K
#1251120
750 f1
VIBRATO FREQUENCY
1M
K
SWELL CONTROL
10
10
100
M
6J5
6J5
M
.5
CONTROL
TUNING
LOWER MANUAL
OSC. PLATE
1
6N 7
SUPPLY)
6L6
1
10K
01
er
50K
100K
100 K
2W
100K
20
M
20
50K
50 W
VVVWVV
40
150f1
2.5 K
15w
o
T
--+f--
TWO
30
IIllll.i
5U4 -G:
5H
5
.
20
-15
H
SWINGING
-
=
100
111
5H
11
loo
takes the "edge" off of the attack and
decay of the tone, and more closely
simulates the bowing of a violin.
If a number of these oscillators, (tuned
to the steps of the equally tempered
scale), are coupled to the filter and amplifier system, and if they are keyed through
contacts linked to the keys of an organ
manual, we have an excellent violin stop.
Melodic and harmonic passages, played
upon such an instrument would be indistinguishable from identical passages rendered by violins, as long as we restricted
ourselves to "straight fiddle playing" and
avoided techniques incompatible to the
keyboard.
Figure 2 shows the circuit for a modified
instrument of this type. The values of
the components were taken directly from
one of our experimental models and,
while operative, do not necessarily represent the ideal.
The oscillators are resistance -capacity
tuned, and the output is keyed instead
of the plate supply, since it was desirable,
in this case, to form a separate stop for
two manuals from the one rank of oscillators. Only one oscillator is shown, the
others being coupled through their respective key contacts to the manual bus
bars, which are common to all the keys
of one manual. The filter, corresponding
to the body of the fiddle, is a simple resistance- capacity network of the "tone
control" variety, and it serves as the
coupling between the first and second
stages of the audio amplifier.
The Vibrato
The vibrato consists of a low-frequency
oscillator, followed by a buffer amplifier
which drives the power stage. The plate
of the power output tube is connected
to the common plate supply terminal for
the tone generating oscillators, the
vibrato output forming a variable shunt
across the power supply of the tone
generator. Sufficient resistance is introduced, between the main power supply
and the terminal feeding the tone generators, to enable the swing of the vibrato
output to modulate the power fed to the
tone generating oscillators.
The frequency of the vibrato oscillator
is continuously variable from four to
twelve cycles per second. The amplitude
of the beat is continuously variable to
give a maximum frequency deviation of
about seven per cent to each side of the
mean tone generator frequency. The most
pleasing effects for general use are obtained with the vibrato set to seven
cycles, with a frequency deviation of two
to three per cent.
In the use of vibrato, it is important
that the frequency of the note be shifted
by the sanie percentage of deviation to
each side of the nominal frequency, since
the ear fixes the pitch of the note at a
point, midway between the extremes.
Any departure from this practice will
result in an apparent change in pitch,
when the vibrato is turned on or off.
AUDIO ENGINEERING
SEPTEMBER, 1947
Photograph of two-octave, two -manual experimental organ (circuit shown
in Fig. 2). The assembly includes
entire instrument with the exception
of the power supply and amplifier.
Knobs on front left or panel control
balance of volume between manuals
and high frequency cutoff, respectively. Knob at extreme right regulates
oscillator plate potential, permitting
the pitch of the instrument as a whole
to be raised or lowered.
view of the experimental
model, showirg chassis layout.
The shafts between the rows of
tubes are For the tubing controls,
one for each oscillator. The two
knobs on the rear of the chassis
control the vibrato frequency and
and the depth cf beat, respectively.
Rear
It should also be noted that vibrato
circuits of the type described can only
be used with tone generators whose frequency is a function of the applied
voltage.
Circuit Design
Figure 3 shows the essential circuits of
a complete electronic organ, designed by
the author. The instrument, as shown,
is provided with five stops, available on
both manuals and the pedals, in addition
to octave and sub octave couplers.
All the stops are derived from a single
rank of oscillators, a separate filter being
used on each note, for each stop. Through
the proper choice of filter circuits and
constants, it is possible to reject or
accept the partials in the oscillator's
signal in any combination desired. This
gives us the choice of a wide variety of
tone colors, since the signal from the
oscillator used is exceptionally rich in
upper partials.
AUDIO ENGINEERING
There is no theoretical limit to the
number of stops which ma-v be derived
from a single rank of oscillators in this
fashion. However, changes in loading
affect the pitch of these oscillators, and it
becomes increasingly difficult to maintain constant loading as more stops are
added. This difficulty might be surmounted through the use of buffer amplifiers or constant impedance keying devices, but the additional cost would be
better justified, if used to provide an
extra rank of oscillators.
This would permit the use of separate
vibratos for the two sets of stops, and
would yield a better ensemble on full
organ. Furthermore, the two ranks of
oscillators could be adjusted differently,
to provide partials of distinctly different
types, thus providing further contrasts of
tone color. It would be highly desirable
to use five or more ranks of oscillators in
instruments where expense and bulk
could be disregarded.
SEPTEMBER, 1947
The borrowing of stops from one
manual to the other is a bit reminiscent
of the'practice of duplexing stops on the
pipe organ. There is nothing objectionable about it, however, when used in the
manner indicated, since the stop may be
had at different strengths on different
manuals, and the corresponding notes on
separate manuals are additive in strength,
when played together.
The octave and sub-octave couplers
also differ from those found on the pipe
organ, in that their notes are additive,
when duplicated upon the manual. The
couplers are derived through the use of
three contacts under each key. One is
connected to its nominal note, another to
the octave above and the third, to the
octave below. It would, of course, be
possible to couple any other intervals together in a similar fashion, as well as
various registers of individual stops, thus
unifying the instrument.
21
OSCILLATOR
STOP NO.1
2
3
,N-J
5
STOP
3
ft
MANUAL
UPPER
FILTERS
\BUS
BAR
LOWER
MANUAL
tivw-J
;STOP
TABS
--wwvVnn.
t-,
(A)
BC
PEDAL
TC
NMI
TO POWER
OSC.
TO AMP.
GI
STOP
TO BALANCE
CONTROL
COUPLER
1
2
3
TAB
Cil.
C3
Ca
C5
C6
Ci
Cs
(B)
FILTERS
11111M11
2
3
MANUAL
16'
UNISON
KEY
OFF
CONTACTS
4.
MANUAL BUS BARS
Fig. 3.
Essential circuits of complete electronic organ, designed by the author.
Frequency Drift
Oscillators of the type used in these
experiments are subject to considerable
frequency drift, unless stringent measures
are taken against it. The power supply
must be well regulated, and the circuit
components mounted rigidly. The resistors and condensers must either possess
a near zero temperature coefficient, or be
graded in sets, the condensers having an
equal and opposite temperature coefficient to the resistors, over the normal
range of operating temperatures. The
oscillator tubes must also be tested for
stability, since changes in transconductance affect the frequency.
The performance of organs of the type
discussed compares favorably with that
of the better pipe organs, as far as tonal
resources and quality are concerned.
Instruments with but a single rank of
oscillators are lacking in ensemble, but so
are the smaller type organs. When several
ranks of oscillators are used, a grande
celeste is produced which equals that of
the largest orchestra or pipe organ.
The continuously variable vibrato
offers a considerable improvement over
anything encountered in the pipe organ
field. Numerous attempts were made to
develop a device of this kind, but none
that we have heard of were entirely
Bottom rear view of chassis, showing layout of components. Resistors and condensers are
mounted in tiers on movable frames, permitting access to those underneath. The manual key
contacts are shown. It is necessary to use electrostatic shielding between all contacts, since the
associated circuits are of extremely high impedance.
22
successful.
It would be absolutely impossible to
duplex the stops of a pipe organ in such
a manner that the corresponding notes on
different manuals would be additive.
[Continued on page 44]
AUDIO ENGINEERING
SEPTEMBER, 1947
l:c>:!.:>:tr'.
t ,rw
ftf
'i
e
ii'}, .O«l«->.
"q rnäi= t-4.S4:
f>:3+.;..
r
:
,
a
..
z
.
>
Ki
N
+ rn
f'171.
rVsr
su
_
CO
: ,
«G
'
a¡
.
.
.
.
.
.
)
O
n
.4:a:C:C:C C:C:C.N;`p
._r,=<
A -C
Voltage
Stabilizers
LEO L. HELTERLINE, JR.*
The Model 150 line- voltage regulator, designed for
mounting on a standard rack panel or to be used in its own
housing. This unit provides good regulation at low cost for
a wide range of audio and allied equipment.
Fig. 1.
IT
NEVER DOES an engineer any
harm to examine his assumptions. One
of these assumptions which is often
made is that audio equipment will be used
under the conditions prevailing in the laboratory, or under conditions prevailing in
the particular manufacturing or processing plant in which this equipment is made.
A good example of the effect of changed
conditions is presented by a study of
distortions. Normally operated high quality amplifiers are rated at 1 or 2 percent distortion at a line-voltage input of
117 volts. Reducing the line voltage by
5 to 10 volts will often more than double
these distortion figures.
Another assumption is made regarding
pure waveforms. Most a-c lines have a
reasonably good waveform, unless highly
reactive loads or those with transient
peak requirements are supplied from the
same transformers. A good waveform is
a necessity when certain types of voltage multiplying circuits are used, since the
performance depends upon the crest value
of the line voltage. Such circuits are
limited when the waveforms are markedly altered, and the effects of surges are
more deleterious than the gradual fluctuations that are experienced in a normal
line. Generally speaking, most audio
amplifiers do not require precision regulation of voltage. However, to attain the
ultimate in performance, voltages must
*Chief Engineer, Sorenson & Co., Inc.,
Stamford, Conn.
AUDIO ENGINEERING
be held within four or five volts of their
design center.
D -C Amplifiers
D -c amplifiers used in laboratory work
and in many other modern applications
of electronics are perhaps the most critical of supply voltages, and many such
amplifiers are completely unworkable
unless their filament supplies are fed from
sources comparable to batteries. Where
motor speeds are related to voltage, the
control of that voltage again becomes
important. Equipment destined for the
foreign market must be planned to be
as free as possible from the effects of
changing voltage, as well as frequency,
at the input terminals. It is not sufficient
that some form of voltage regulation be
supplied which is workable at the nominal
frequency but which becomes completely
unregulated if the frequency shifts appreciably.
T1
REGULATED
AUTOTRANSFORMER
AC OUTPUT
T2
A.C. INPUT
SATURABLE-CORE
REACTOR
10 K
O.1
6
MEG.
L6
V2
HIGH
VOLTAGE
POWER
fN\
SUPPLY
25
K
T3
1R
BB.
AC
F
g. 2.
SEPTEMBER,
Schematic of Model 150 Sorenson regulator. V, is special diode used
sensitive control for the 6L6 current through a saturable reactor.
1947
as
the voltage-
23
The solution to the operation of electronic equipment on lines furnishing
variable voltage and variable frequency
is to use some form of voltage regulation
which is independent of frequency.
Needless to say, the regulator which will
fulfill these requirements is also capable
of functioning correctly on lines supplying constant frequency, so that it is not
necessary to make any change in equipment even though the frequency of the
supply he changed, provided the design
of transformers is adequate for the lowest frequency encountered.
One such regulator is manufactured by
Sorenson & Co. Inc., and the degree
of control is well within the limit specified by very critical applications. The
regulation curve of a typical line -voltage
regulator is shown in Fig. 4, and the schematic of the regulator is shown in Fig. 2.
The voltage-sensitive element of this
device is a special filamentary diode.
also manufactured by Sorenson for this
particular use, and held to close limits.
This tube resembles an ordinary vacuum
tube, but has some unique characteristics.
The filament is of pure tungsten, and the
vacuum is considerably higher than is
usual with ordinary radio tubes. The
principal characteristic of importance is
that the plate resistance is a highly retraceable function of the filament voltage.
Fig. S illustrates the appearance of the
tube itself.
Diode Operations
To understand the action of this diode,
assume a change in the output voltage in
the upward direction, resulting from a
reduction in load or an increase in line
voltage. This change causes an increase
in the filament voltage, and moves the
operating point along the plate resistance-filament voltage curve in such a direction that the plate resistance is
lowered by an appreciable factor. This
plate resistance forms a portion of a
bridge circuit, and the unbalance voltage
is fed into the grid of a beam power tube.
The beam tube draws its plate current
through the d -c winding of a saturable
reactor, the a-c coils of which act as a
variable reactance in the line circuit.
From an inspection of the a-c circuit,
it is seen that the voltage appearing
across the output terminals is dependent
upon the line voltage, the step-up ratio
of the autotransformer, and the reactance
of the a -c -coils of the saturable reactor.
As V2 draws more current, the reactance
of thea-c coils is lowered, and consequently
Fig. 3. A close-up of the filamentary diode,
Sorenson 2AC15. The control function depends upon the plate resistance of this tube,
which may be energized by either a -c or d -c
voltages.
a higher voltage is applied to the autotransformer. The converse is also true,
and the bridge circuit may be so connected that the change in the plate resistance of Vr serves to compensate for line
voltage changes.
The gain of this particular arrangement is of the order of 50,000 when
measured as the ratio of voltage change
across the d -c portion of the saturable core reactor divided by the change in
output voltage. This degree of control
is sufficient to maintain the output within
much less than 0.5 per cent over a wide
119
Input vs
voltages
of Model 150
regulator. Slope of
regulated
porFig. 4.
118
output
voltage
117
t
i
o n
of
t h
i
s
characteristic may
be varied to
compensate
for
116
resistance of
leads
ulator
80
90
100
LINE
110
VOLTAGE
120
130
140
from regto load.
fluctuation of supply line voltage and
load current.
Proper design can also compensate for
the resistance of the leads from the regulator to the load. This is particularly
important when the load fluctuations are
great and the leads are long. Thus, it is
possible to so arrange the controls that
the voltage at the load remains constant,
even though the output voltage changes
slightly in a reverse direction when
measured at the output terminals of the
regulator.
Since there are no resonant elements
in this regulator, the action is completely
independent of frequency over a range
of plus or minus 30 per cent or better.
The manufacture of the tungsten filament diodes is a critical operation if the
best possible characteristic is desired.
Since the diode is the true voltage reference, it must be independent of all
factors except the voltage across the filament of the tube. Other possible sources
of dependency arise from thermal ambients, change in filament emission
with time, and changes in other characteristics with time. All these problems
have been investigated in the manufacture of this type of diode.
It has been found that high vacuum
is necessary to produce a truly stable
source. Compensation for ambient temperature is accomplished by careful design of the elements and the correct
choice of metals with respect to their expansion coefficients. The order of stability of these units constitutes an improvement over voltage regulator tubes
of the gaseous discharge type by a
factor of at least 10. The life characteristics of these units when employed at
low plate- current values exceeds that of
practically any commercial vacuum tube.
A simple analogy of life comparison is that
of running an incandescent lamp at approximately one-half its rated voltage.
As established by Langmuir and others,
life is a multiple- powered function of
filament temperature.
Stability is of utmost importance and
can only be obtained by careful techniques. Assume, for example, that there
is a plate voltage of 500 volts across the
diode elements. A change of 0.1 volts in
the filament supply to the diode will
produce a change of 300,000 ohms in the
plate resistance of the tube, and if this
tube is in series with a 1.0-megohm resistor, it will change the plate potential
approximately 100 volts. If this change
is impressed on the grid of the beam power
tube, it is obvious that it will more
than control the tube over its full range
of grid voltage characteristic.
Rectified A -C Supplies
This type of circuit is also adaptable
to rectified a -c supplies employed for
furnishing filament and relay voltages.
Since the sampling voltage may be either
[Continued on page 43]
R4
AUDIO ENGINEERING
SEPTEMBER, 1947
Planning
a
Studio Installation
J. D.
PART Ill -This
is the
third of
a
COLVIN*
series covering broadcast studio installa-
tions. The methods outlined are also suited to large public- address projects.
BEFORE AN INSTALLATION can
be made and put into operation a
considerable amount of material
must be purchased. The greater the
amount of this material that can be purchased and had on hand before the construction actually begins, the faster the
work will proceed and the least amount of
temporizing will take place because the
exact item required cannot be obtained
at the moment. A complete list of materials required for the rack and console
can be obtained, since these have already
been laid out in detail. The total amount
of conduit and a good approximation of
cable for the installation can be determined from the conduit layout. There
are, however, a number of items that do
not show on the diagram and layouts
made so far that must not be overlooked
and a list of such items follows as they
occur to the author.
Additional Items Needed
sheets list every terminal to which a
connection is made and indicate the
connections to be made from terminal to
terminal. By making the connections indicated on the running sheets line by line,
the rack becomes completely wired when
the last line is checked off.
Before the running sheets can be compiled, it is necessary to prepare a pictorial sketch of the rear view of the rack
showing the location of all terminal
layout must also be shown. Nothing
should be left for the wireman to figure
out in the way of circuits or levels-his
only contribution to the finished racks
should be perfect soldered joints and
neatly formed, tightly laced cables.
Running Sheets
A very convenient form in which the
wiring information can be compiled is
known as "running sheets." These
Fig.
Pictorial sketch of the rear view of the rack.
1
IB
SUPPLY
6 -BA- IA
PRE-AMPS
........
I
`s,..s,.s_
.aws..
A6
A7
`Iicrophones- velocity, pressure, direc-
tional; microphone stands -floor, desk,
boom, hand; microphone plugs, receptacles and extension cords; turntables,
tubes for all amplifiers, spare tubes,
spare fuses; clocks, touch-up paint to
match the equipment finish; cable shield
termination (such as Thomas and Betts),
loudspeakers -talkback, monitoring, office
speakers, remote gear-amplifiers, mikes,
stands, cables, carrying cases; headphones,
patch cords, attenuater cleaning solutions, work bench and tools, circuit
check meter, telephone equipment-lines,
instruments. There may be and probably
are other items that must be taken care
of before the job is 100% complete,
but if all of the above are considered, the
remainder will be very minor (or so
obvious as to be over -looked until required).
Having completed the mechanical
layout of the equipment in the rack the
next step is to prepare information in
some form that will guide a wireman to
connect the individual components to the
proper points, as determined by the
block diagram. Not only must the exact
electrical connections be shown, but
cable runs containing circuits within the
same level groups as used in the conduit
POWER
IIBX
A4
AS
A3
A2
2-BA-IA
SPARE
REG.
_
All
15
D
A8
A9
A10
0
METER PANEL
o
o 2
t
2
3
I
o 3
0 4
0 5
0 6
-0
-O
4
5
0-
6
7
o
7
8
9
JK
24
ß
9
JK
BA4A
P2
I
PI
Al2
IA13I
IT3j
SHELF
MON. AMP.
I
o--
MOUNTED
DN
f
BA 4A
I
A15
CUE
AMP
T2
A14
I
RL4
RL3
T
1i17m
RL
ITS3I
a
---{ TS4I
r
0
1
1
l/N
TT
o
SWITCH 8
m
FUSE
*Audio Facilities Engineer, American
Broadcasting Co.
AUDIO ENGINEERING
SEPTEMBER, 1947
25
FROM
TO
Terminals Cable Equipment Terminals
Pair
Equipment
1
Jack row
2
IT
21
A
"
"
3T
A
PI
3
n
n
4
"
I
TS 2
TS
6T
71
"
0
"
6
"
9T
I
A
I
1
A
81
5
1-
I
I I
T
"
I2T
"
13T
"
8
9
15T
"
161
"
17T
"
181
n
f9T
II
"
21T
"
22T
23T
"
24T
Fig. 2.
II
TT
out
1
2-3
2 -4
TT
3-1
TT 3 out
2
out
Mixers out
3-4
A
PI
A
out
Divide pad reg out
A
PI
B
out
Divide pad emg out
3
Mic
n
A
RL
1
1
relay out
6
Form used in making running sheets.
blocks on amplifiers, transformers, pads,
etc., and a designation for each terminal
block assigned. Such a sketch is shown
in Fig. I. It will be noted that the terminal blocks on the amplifiers are designated
as A 1, A2, etc. Transformers are referred
to as T1, T2. Terminal strips carry the
designation of TS. Pads are P. Jack
rows are numbered from top to bottom
and the jacks in each row are numbered
from right to left as viewed from the rear.
(This gives a normal left -to-right numbering from the front of the rack which is of
greater convenience for operation.
There are sometimes components that
do not have a definite, stenciled terminal
board, such as rotary selector switches,
lever keys, etc. In such cases it is well to
make a separate sketch of the component
in sufficient detail and assign terminal
numbers to the various solder points.
These sketches can then be checked
by the wireman while working on the
rack to determine the particular point
on the unit being referred to in the running sheets. The most commonly used
method of numbering terminals on standard 40 or 60-pair telephone type of terminal blocks that are used in making the
external connections to the racks is to
refer to each row by number counting
from top to bottom (usually 1 to 20) and
the terminals in each row by number,
counting from front to back (usually 1 to
26
out
Mic 2 out
3 -3
A
20T
10
1
3 -2
"
141
1
-4
2-2
II
7
-2
-3
2 -2
A
Mic
I
II
IO T
LL Trunk
I
Il
"
A
Circuit
-2
10
4 T
5T
10-
the apparatus of which they are a part
and are numbered by the manufacturer
of the equipment. Since there is a choice,
a definite attempt should be made to
group terminals for circuits within the
1 to 6). With this arrangement, any
particular terminal is referred to by row
number and its number in the row. For
example, the very ,first terminal on the
block would be 1 -1. The very last terminal on a 40-pair block would be 20-4.
Jacks are referred to by row number, the
number of the jack in the row and whether
the connection is to be made to the tip or
normal of the jack. A standard jack row
consists of twelve pairs of jacks but each
individual jack in the row must be numbered from 1 to 24 in order that the proper
phasing of circuits can be maintained
throughout the entire wiring of the rack.
For example, the tips on the first pair of
jacks in the first row would be Jack Row
1, IT and 2T. (T standing for tip). The
last normals of the last pair in the second
row would be Jack Row 2, 23N and 24 N.
Although Fig. i shows all cable runs in
place, these are not drawn in at the time
the terminal blocks, etc., are located but
are developed as the running sheets are
prepared, and it is determined how many
of these are required and where they go.
4 or
External Connections
Of all the ter,nival blocks shown on the
rack sketch, the only ones at which a
choice of terminals can be made are those
used to make external connections to the
rack. In this case they are TS1, 2, 3 and
4. All others are definitely determined by
same level grouping as used previously.
No trouble from cross -talk should be experienced by running cables of low,
medium and zero levels to the same terminal block. High levels and control
circuits should be placed on a separate
block. Variations to this arrangement are permissible if several rows of
unused terminal are left between level
groups of more than 30 db. It is also
advisable to attempt as far as possible to
keep the normally used program circuits
on the same terminal block. These would
be such circuits as microphone outputs,
preamp input and outputs, faders in,
mixers out, program amplifiers in and
out, line in, etc. Other miscellaneous or
seldom used circuits as trunks, spare
jack circuits, control circuits and spare
terminals for any future additions should
be put on a separate block. The purpose
of such segregation is to keep the block
having the permanent connections mentioned free from those circuits most likely
to be changed and where additions might
be made. The less a terminal block is
worked on after it is once put in the less
likelihood of trouble from broken or
shorted wires later on. This is not a hardand-fast rule since often the number of
terminals used up by the program circuits are so few it would be foolish to put
in a second block to carry what odd circuits that might exist. A general rule to
follow is to start at the top of the terminal block with the low-level microphone
and turntable circuits and end up at the
bottom of the block with the higher
levels. Also, where groups of similar
equipment are used, such as microphones,
preamplifiers, etc., connections to the
terminal block should run in consecutive
order as Mic i, Mic 2, Mic 3 or Preamp I
In, Preamp 2 In, etc.
Since the jack field is the most common
point of all circuits in the rack it is the
most logical place to start the running
sheets. The form used in making the
running sheets is shown in Fig. 2, and
since it is sheet 1 it indicates all the connections from the jack tips of Jac! Row 1.
Connections are being indicated as going
from the jacks to their respective designations. Thus the first step is to list under
the "From" equipment and terminal
columns all terminals in the rack starting
with the jack rows and the remaining
equipment listed alphabetically -as A
for amplifier, P for pads, etc. The next
equipment to be listed after the last jack
row would be the amplifier whose terminal
block is designated on the rack sketch as
Al. All terminals on the Al block would
be listed in numerical order. A2 would
follow with its terminals and so on. A
total of 24 pages were required to list all
terminals for the rack under discussion.
AUDIO ENGINEERING
SEPTEMBER, 1947
Terminals
Having listed all the terminals from
which connections are to be run, the
next step is to determine and list the
equipment terminal blocks and terminals
on the blocks under the "To" column.
This work will determine the actual
wiring and placement of the wiring in the
rack and it is necessary to have a copy of
the block diagram and the jack field layout at hand as the source of circuit information. An examination of the jack
field shows that the first pair of jacks or
terminals 1T and 2T in the first jack row
are for a low -level trunk. As such it will
be connected to a terminal strip at the
bottom of the rack. However, as it was
discussed previously, it is desirable to
start at the top of a terminal strip with
microphone connections and therefore
the terminal designation for this and
similar pairs of jacks will be deferred
until the main program carrying circuits
in the jack field are finished.
The next six jack pairs are microphone,
turntable and mixers outputs and go to
the terminal block. They are therefore
assigned to TS1, 1 -1 and 1 -2 up to and
including 3-3 and 3 -4. The title of each
circuit as Mic I out, Mic 2 out, etc. is
filled in the column headed "Circuit,"
and are for future convenience in locating
terminals. The next jack pair having
terminals 15T and 16T are indicated on
the jack field as being "Divide Pad Reg.
Out." This jack pair on the block diagram shows it being connected to the
output of the divide pad feeding the inFig. 3.
put of the regular program amplifier and
the rack sketch shows this pad as being
designated Pl. Accordingly, Jack Row 1,
15T and 16T will be indicated as being
connected to equipment "Pl " and the
"A out" terminals. Similarly the remaining connections are determined and are
recorded. It will be noted that terminls
17T and 18T as well as 21T and 2.07' of
Jack row i show no connection. This
is due to the fact that these jacks are multiples of the divide pad regular and emergency outs. Such multiple connections
on the jack field can be shown on the
same type of form as used in Fig. 2, or by
a sketch shown in Fig. 3. Use of the
sketch seems to be preferred by wiremen
since it will show all normal and multiple
connections pictorially on one sheetwhereas possibly six or more of the form sheets
would be required.
Numbering
Referring again to Fig. 2, the pairs
should be numbered consecutively in the
"Pair" column. This pair numbering
will be found useful in finishing the running sheets and is sometimes used by the
wireman in marking pairs in cables.
Indication of the "Cable" in which each
pair is to be run must now be indicated,
both on the running sheets and on the rack
sketch. It will be observed from a study
of the block diagram that all pairs listed
on this first of the running sheets come
within the limits of the "Low" level
grouping and therefore can go in the same
cable. Since it is convenient to label
cables alphabetically, the letter "A" is
therefore assigned and entered in the
column head "Cable." On the rack
sketch, a line starting at jack row one is
run down to TS1. This line at the moment represents the cable consisting of
the two microphones, the three turntables and the mixer out circuit. However, the running sheet shows two pairs
going to Pl, and therefore a turnoff from
the line already drawn will be run over to
Pl. Likewise, a line will connect cable A
to RL1. The additional connections into
the A cable as shown in Fig. 1 over those
mentioned above are the result of other
pairs being assigned to this cable in latter
running sheets. In this manner, the
running sheets are built up and the cable
runs added to the rack sketch.
Other Checks
When each pair of terminals is indicated as being run from one equipment to
another, it is advisable to go to the sheet
on which the "To" equipment is listed
and indicate that those terminals are to
be run back to the point from which they
came. For example, on Jack Row 1, 1T
and 27' are shown going to TS2 10-1 and
10-2. In Fig. 4, the sheet showing that
portion of TS2 which contains 10-1 and
10-2, it will be noted that it is shown going
to Jack Row 1, 1T and 2T, that it carries
the same pair number, cable letter and
circuit name. It will also be noted that
an asterisk is placed in the box containing
the cable letter. This asterisk is to indicate to the wireman that this pair has
already been run in and should not be
duplicated. An examination of Fig. 4
Sketch showing multiple connections required in the jack field.
111111111111111111111111111
-
III
Ili
2
3
4
I/ AV INIIIIMINE
EMI
1111
;
1111111111.111111
AUDIO ENGINEERING
SEPTEMBER, 1947
1111111
111111111111111111
=Ell
8
27
FROM
Pair Equipment
73
TS 2
74
75
Terminals Gable Equipment
-I
8
"
8 -2
"
8 -3
"
8 -4
"
9-1
9 -2
9 -3
"
o
ULTRASONIC MAGNETIC
RECORDING
TO
jack row8
B
Terminals
T
I
I
Circuit
Spare
121
B
13 T
Spare
14T
B*
151
Spare
I6T
9 -4
I0 -I
"
I
Il
12
"
u
II11
99
I
11_3
o
II -4
"
12-1
12 -3
"
12
"
13 -I
"
13_2
"
13 -3
"
13 -4
"
14 -I
"
14
"
14 -3
"
14
Fig. 4.
3T
1
LL Trunk
2
4T
"
B" Jack row 8
T
1
PGM
line to MC
2T
B
"
5T
"
6
RL 3
7
"
9
Spare
Intercom relay
control
-4
D'`
RL
I
13
RL
1
coil
14
D
-2
RL 2
13
"
14
RL 2 coil
-4
Sheet showing that portion of TS2 which contains 10-1 and 10 -2.
reveals that every pair shown is a return
indication as evidenced by the asterisks.
Preparation of running sheets for the
console is done in the same manner as for
the rack. A rear-view sketch is made of
the panel components, letters assigned
to the terminal blocks and equipment and
the running sheets made. In a console,
the terminal block intended for making
the external connections is usually the
most common point for all circuits and
may be used as a starting place for the
running sheets. As the preparation of the
running sheets progresses, cable runs are
to be drawn on the console sketch.
Interconnection Sheets
Interconnection or cross- connection
sheets show how to make the external
connections between the rack and console,
between the studio control room and
master control and runs to all external
units as loudspeakers, turntables, door
lights, etc. Their preparation is similar
to running sheets for racks and consoles
and the same form is used. The sketch
used to aid in their making is the preliminary conduit layout previously made.
Instead of making reference to cable runs
28
LL Trunk
2T
4"
2
"
IT
1
"
-2
12
97
Jack row
10 -4
o
71
i
10
"
70
-2
10 -3
A
by letter as in the rack sketch, reference is
now made to the conduit number in which
the cables are to be run. To start, all terminal blocks involved and their terminals
are listed on the forms. Then, with the aid
of the block diagram, the "To" point of
each "From" point is determined and
written on the sheets. As with the running sheets, those pairs that are indicated
as running back upon themselves should
be so marked with an asterisk.
All that is necessary to complete the
conduit drawing is to count the number
of pairs shown on the interconnection
sheets for each conduit and to list these
quantities in the conduit table. The conduit drawing showing the points between
which conduits are to be run and the conduit list that gives the individual conduit
sizes, and the number and level of pairs
each is to contain, gives the installer all
the information needed to pull in all interconnecting cable. These cables, when
connected in accordance with the cross
connection sheets, completes the electrical
part of the installation.
[To be Continued]
Characteristics and details of magnetic recorders essentially flat to 50,000
cps were revealed at a recent IRE section
meeting in San Francisco. The speaker
was Al Isberg, Chief Engineer of f -m station KRON, and during the war a project
supervisor in the Airborne Instrument
Laboratories on Long Island.
Security requirements permitted him
only to say that the units described were
used for the recording and later visual
analysis of signal material of only a few
seconds duration derived in part from
radio receivers with i -f frequencies in the
5-mc region. Success of the project
hinged on the fact that waveforms are
not distorted when recorded in modulation envelopes, regardless of whether distortion in the recorder is independent of,
or proportional to, frequency.
At the conclusion of this presentation,
Mr. Isberg introduced Jack Mullen, San
Francisco sound engineer, and Col. R. H.
Ranger, who discussed the original discovery of Magnetophon tape recorders in
Germany. Col. Ranger pointed out that
military intelligence knew nothing of the
recorder and its outstanding operational
characteristics because the Germans had
not labelled it secret. Mr. Mullen's
"liberation" of several of the units called
it to the attention of Col. Ranger, who
returned to Germany again for further
details and is currently active in the
manufacture of both recorders and tape.
The meeting was concluded by demonstrations of a number of the available
commercial wire and tape recorders.
NEW R.M.A. PRESIDENT
Max F. thLlcum, 'rice President and
Treasurer of Sylvania Electric was recently
elected president of the Radio Manufacturers' Association. He has been active in
R.M.A. for the past twelve years, serving
Max
F.
Balcom
as vice president for two separate terms
and as a board member for the past five
years.
AUDIO ENGINEERING
SEPTEMBER, 1947
In this
department the author, who
is a
very well -known record critic, will review
monthly record releases of outstanding technical,
as
well
as
musical, quality.
EDWARD TATNALL CANBY
IT HAS BECOME increasingly apparent
since the first introduction of vinylite
records in the commercial field that
somehow, some time, a compromise of
sorts would be necessary between the shellac type of record and the pure vinylite.
Unfortunately such a compromise is not
easy to manage-the problem is a little
like compromising between water and oil!
From the listening point of view, however,
it has become fairly clear that though the
plastic record has the advantages of
lightness, unbreakability, and lower hiss,
it has numerous disadvantages as compared to the shellac, especially the high quality laminated shellac disc. Most
important from the listener's viewpoint
is the type of noise that is heard on
vinylite. One type, the tiny irregular
click, though insignificant as a meter
indication, turns out to be just as annoying, if not more so, than the even, steady
hiss of good old- fashioned shellac. And
worse, it can't be filtered out, since it extends to lower frequencies. Any irregular
sound attracts the attention far more than
a continuous one. An even worse type of
plastic noise is the lop -sided hiss that
comes from slight defects in material or
pressing. Hardly a set is made without
at least one such side. With nine-tenths
of a record surface virtually silent, even
a very faint hiss over a square inch or so
of surface is extremely annoying- whereas the same hiss, uniformly distributed,
would be quite harmless. Other difficulties that listeners experience with plastic
are its amazing ability to attract dust
particles (a wet cloth is the only solution),
and its sensitivity to scratching. (Look
at an "unbreakable" children's record
after a week of play!)
Experiments in various plastics and
AUDIO ENGINEERING
and particularly in combinations of
vinylite with other materials have been
going on most energetically, and it is
likely that very soon there may be some
improved record materials on the retail
market that in every sense approach
a halfway point between the pure vinylite
and the pre -war shellac. It looks as
though only some such new development
would make it possible for us to play
records at high fidelity, "wide open,"
bringing through highs to ten or twelve
thousand cps, yet with a minimum of the
psychologically objectionable type of
noise, and without fear of the quick wearing that spoils so many plastics when
something less than the ideal pick -up is
used.
Here are some recent recordings of outstanding interest technically:
Dvorak, Symphony #1, Opus 60. Cleveland
Orchestra, Leinsdorf..... Columbia M687
A seldom heard but most enjoyable work
by the composer of the familiar "New
World" Symphony. Large orchestra,
good resonance to hall. Wide -range
recording with good surfaces. More of an
"over-all" pickup than with Columbia's
New York Philharmonic records, it seems.
Gershwin, Rhapsody in Blue (Arr.)
Selection of Hebrew Dances
Ambrose and His Orchestra (Embassy
Club, London)
London Decca, F5454, F5285
"ffrr" recording in a popular equivalent.
This is British salon jazz, not particularly
to the U. S. taste. Quite wide -range
recording, more liveness than in most
American populars, but considerably less
than the big classical ffrrs. (Two samples
from a larger series.)
Debussy, Sonata #3 for violin and piano
(1917)
Zino
Francescatti, Robert
Columbia X280
Casadesus
A new wider range recording with excellent
liveness, fine balance between violin and
piano. It supplements older version
(pre -war) by Szigeti and Foldes, Colum-
SEPTEMBER, 1947
bia X242. Interesting comparison technically. The old set is minus highs, violin
is more prominent, closer. But over -all
liveness is excellent, too.
Purcell, Abdelazer Suite. Vox Chamber
Vox 199
Orch. Edvard Fendler
Numerous recordings are still being
issued with the pre -war type of restricted
range -nothing above 5 -6000. This is
good example of it -small string orchestra, excellent liveness, good recording.
But no highs in the record, and surface is
poor. Sounds fine on standard console
phonographs!
Liszt, Mephisto Waltz. N. Y. PhilharColumbia X281
monic, Rodzinski
Apparently this was made before recent
improvements in Columbia's Philharmonic recording technique. Wide range,
but upper part of "curve" seems lower
less rising characteristic than in later
issues. Also liveness is different; more
over -all effect here, apparently less accentuation. Sounds a bit tinny, "narrow."
-
Turnover point is evidently higher than
on newer Columbias, making for less positive bass. Compare with Kahchaturian,
Gayne Suite or Rimsky-Korsakoff, Russian Easter Overture.
Sir Lancelot, Calypso Carnival. Trinidad
Serenaders; The Volunteer
Crest Recordings CT -3
Excellent highs for this type recording;
snare drum, etc. come through clearly,
also sibilants in vocal. Heavy recording,
tends to "buzz" a bit.
Shostakovitch, Symphony #9.
a) Boston Symphony Orchestra,
Victor M1134
Koutsevitsky
b) New York Philharmonic, Efrem
Columbia M688
Kurz
The Columbia set has the better surfaces,
and frequency response appears to extend
to the 10,000 -cps range. Doubtful if
Victor set exceeds roughly 7,000 cps.
Columbia recording has greater breadth
of sound than Victor, showing a legitimate
area of difference in technique; both use
overall type of pick -up, but Columbia
uses more accentuation. This is clearly an
and constitutes an
arguable point
interesting comparison. (Both sound
equally well on average phonographs.)
.
29
MUSICAL ACOUSTICS
BENJAMIN
F.
TILLSON
This
IV
APPRECIATION of music may
result from either or both of two
approaches: the emotional and the
intellectual. The former is not subject to
a definite analysis as it depends upon the
personality of the individual, his environment, his mental serenity, health, disposition, associated ideas, heritage, education, experiences, and still other factors.
Although we do not know why, most
of us probably do not find the same
appeal in a melody and a harmony, but
particularly the latter, as it is transposed
in different keys. There have been
elaborate discussions concerning the emotional and esthetic suitability of each key.
It is not likely that all persons would hold
the same opinions concerning the expressiveness of each key.
Someone might question whether such
descriptive variants in musical keys were
not due to the tempered scale causing in
the different keys a different sequence of
intervals divergent from "just intonation."
But, playing a phonograph record at
different speeds will maintain the original
interval-frequency ratios and will change
the keys. A listener may thus determine
for himself whether the theme and harmony holds for him the same emotional
appeal in all keys.
A composer selects some certain key
or keys as the best vehicle for the musical
expression of his message or poem. Most
lovers of classical music fail to realize
that they do not hear it reproduced today
as the composer intended it to be heard.
The conductors, arrangers, and publishers fail to make allowances for the
changes of pitch which have occurred in
the years since composition. Then, too,
the timbre of musical instruments has
also changed.
Vocal music of over 150 years ago must
be transposed a whole tone lower than
present standard pitches in order to
duplicate the emotional sound effects for
which it was then composed.
The vocal music of Handel's time, the
music of Mozart, and probably that of
Haydn and Beethoven, should correspondingly be transposed one -half tone
lower, because "C" then varied from 500
to 513 cycles per second. Handel's tuning
fork of 1751 gave a "C" of 510 cycles/sec.
The pitches were different in Germany,
France, Italy, England, and Belgium,
T11E
Consulting Engineer, Montclair, N. J.
30
is
a series of articles on music theory,
written especially for sound engineers.
the fourth of
affecting the operas therein; and during
130 years the pitch of "C" rose from 467
to 546 cycles/sec. There was a lack of
standardization in either the pitches, the
laying of the temperament of the scale in
tuning the scale intervals of instruments,
and in the timbre of instruments during
the golden age of classical music. So
there was not adequately uniform ear
training in music.
Since then, during the past fifty years,
through the medium of the phonograph
and radio, the public's listening opportunities have been intensified and extended more than heretofore in the realm
of music. But the defective fidelity and
distortions of such music programs have
trained ears to become accustomed to
them, and have destroyed a finer appreciation.
Listener Preference
Most listener reaction surveys in
recent years have concluded that the
public did not wish to hear the higher
frequencies and even that the violins of
the old masters were superior because
they lacked the high frequencies. However, some such tests have not determined and reported the complete technical data necessary to make an intelligent analysis of their conclusions; nor
have they avoided the psychological pitfalls which are possible in the statistical
sampling of human reactions, opinions,
and experiences.
Much more is involved in music than
the qualitative and quantitative analysis
of its component tones and their harmonics. The duration of time for each
harmonic, or conversely its damping,
makes a great difference in musical
pleasure. With the damper pedal of the
piano one can hear how displeasing is the
jumble of sustained tones from a piano.
The merit of the violins of the masters
probably rested in their uniform resonant
response and a lack of extreme damping
of tones, and lack of harmonic modulations. Such aberrations are predominant
in the higher frequencies. With imperfect
instruments and reproduction, relief from
them is most easily attained by doing
away with the higher frequencies; and the
public would naturally prefer such devitalized music rather than distressful
distortions.
The best of lateral -cut commercial
phonograph records or radio transmitters
rarely give undistorted frequencies above
5,000 cycles per second and often not
;
better than 3,000 cycles with the loud
speakers commonly supplied. A frequency band two octaves wider (up to
11,000 cycles) is claimed for British hilland -dale disc recording and Western
Electric Co. recording of vertical -cut
records. Sound films have a range to
9,000 cycles, as has also been claimed for
magnetic recordings, which more frequently are below 5,000 cycles.
AM radio transmitters do not exceed
7,500 cycles because of the limitation in
the width of licensed wave bands to
avoid crowding the commercial radio
spectrum. FM radio transmitters are required to be adequate for a range of 50 to
15,000 cycles; but lack of financial
resources, coupled with the restrictions of
musician .unions, rarely permit programs
of such frequency range because the radio
stations are forced to use phonograph
records and telephone lines limited to
5,000 cycles, or possibly 8,000 cycles if
the increased cost can be afforded.
Most a-m radio programs also pass
over telephone lines from the studio to
the transmitter with the same restriction
of 8,000 cycles. Although the telephone
companies say that they can offer a
service range up to 15,000 cycles at a
premium cost, the source of this information thought that no one was leasing such
high -frequency service.
Musical Education
Education should promote an appreciation of the finer things in life. That
requires training. The most profitable
period for training is in childhood, before
habits have been set. Music appreciation
has been accepted as one of those finer
things proper in public school education;
but, while its content has been given
attention its quality has been too greatly
ignored. Ear training in the schools
offers the best method for counteracting
such a low grade aural environment as
has been thrust upon our youth by juke
boxes, phonographs, and radios.
During the first century A.D., Plutarch
stated in his "Opera Moralia": "Whoever it be that shall give his mind to the
study of music in his youth, he will be
sure to applaud and embrace that which
is noble and generous, and to rebuke and
blame the contrary, as well in other
things as in what belongs to music. And
by that means he will become clear from
all reproachful actions, for now having
reaped the noblest fruits of music, he may
be of great use, not only to himself. but
AUDIO ENGINEERING
SEPTEMBER, 1947
to the commonwealth; while music
teaches him to abstain from everything
that is indecent, both in word and deed,
and to observe decorum, temperance,
and regularity."
However, times change; and if Plutarch
heard some of the indecent songs now
"plugged" on the radio under the guise
of music his idealism might have been
tempered by the realization that such
tools of great potential service to mankind may be prostituted to evilly destructive uses.
There exists to the reproducers of
music and the makers of reproduction
devices a great opportunity and challenge
for public service, as well as commercial
profit, which they have failed to realize
to the full during the past sixty years
since the phonograph was developed.
By greater attention to the higher
musical fidelity of their products and performances, and by more earnest cooperation in cultivating the finer appreciation
of music by our youth, they will open
markets of vaster importance than their
present conception. They will halt the
dying listener interest in phonograph and
radio music. It seems irrational for them
to believe their reports that listeners prefer not to have high fidelity. If that were
so they and others would avoid the personal performances of musicians and
orchestral concerts.
Perhaps, as in other fields, there is a
need for an Institute of Sound-Reproducers to integrate an approach to the
public relations and research problems of
their industry, with a staff free from the
inhibitions of past habits and experiences
in the phonograph and broadcasting
fields, and guided by one with technical
imagination and broad interests.
The Acoustic Problem
Like most important projects, the
problem is not simple. But is it receiving
the attention and effort it warrants? Or
has the tendency been to avoid the
problem by rationalizing a conviction
that it does not merit solution, because
the vast majority does not want what it
has had no adequate opportunity to
learn to appreciate -high fidelity.
There would have been no progress in
the world if such an attitude had prevailed in other fields. Most new developments have been opposed by the majority, even the advent of the Machine Age.
Public education, itself, is still compulsory.
The development of the science of
electronics was probably the greatest
incentive in much-needed research for
improvement in the acoustics of phonographs. But its apparent magic has, perhaps, its drawbacks if it encourages the
that fundamental, primary
faults can be compensated for and alleviated electronically; and, therefore, there
is no incentive to rectify the sources of
trouble.
The phonograph and radio arts need to
develop a true reproduction of live
programs, and then any further circuits
and gadgets can be added to permit the
attitude
listener the self-expression of altering the
true to suit his fancy and musical ignorance.
It is probably generally true that whenever harmonic electric oscillations are
subjected to electric or magnetic fields or
impedances, whether capacitive or inductive, they have the hazard of distortion.
The more such means are employed to
correct aberrations the greater the opportunity to lose certain attributes of fidelity,
liveness, and that naturalness whose lack
promotes psychological fatigue. The
deficiency may be only in transients
which the usual methods of testing fail to
reveal in average or root -mean- square
observations.
Unfavorable listener-reaction studies
of high fidelity programs may be due to
such transient or intermodulation deficiencies in true quality as were not
appreciated by the listeners or those who
conducted the tests.
Of all the branches of physics, sound
appears to have been most neglected in
college science courses; and it is generally
ignored by musicians and architects.
Distortion
The following classification of distortions accords with the effects produced:
1. External noises or sounds which affect
the electro-mechanical systems for detection, transmission, or reproduction of sound
Causes of
programs. These chiefly occur from microphonic vibrations of electron tubes without
shock -mounting or acoustic insulation from
the loudspeaker output.
[Continued on page 431
Operator at the center control console of CBS-KNX.
AUDIO ENGINEERING
SEPTEMBER, 1947
31
Naked o f
7
Locating Operating Point
ARTHUR PLOWRIGHT
VARIOUS TECHNIQUES which may
be utilized to find the operating point
for the circuit of Fig. 1 have been
presented during the past few years.*
All assume constant gain or require
plotting of curves. The improved method
to be described necessitates only a
draftsman's scale or machinist's rule,
involves no assumptions of linearity, and
requires no construction lines.
e
Constant Gain Technique
A method in which constant gain is
assumed is shown in Fig. 2. A load line
is drawn with a slope equal to the negative reciprocal of (RL+Rk). At its inter-
ep
e
h:::ü:
I//Q:M7
I/11INIIqMI
INI,
I/IM11
IIh II/
ItIM
MII::..
loo
75
Ii\
IltCN
III,\.p ,
11I
11II
IIIII
II/IMI
IAINI
110
III.I
AIMINIII
/N/
1111UMIN'
MANN/
50
25
r
n'1,
ebb
111111
I
000f1
RK
'
-
N
IH1/
11I
.r,...,........I...1/
iÌ.w...lura..
III/I/MNMININA
III --yIINIVIrI
-N:,rnIrrni
A."/INrrI.rw1
i.i
v/I`WIII'
IIIN%NA
r
L
I
MEW
N
MI
60001L
R
700
E
°b
111111111
/N
-60
NM
-70
-a0
-90
AMMO'
/I
,`IIMI,II7IINIM %GMI7
IMIANM
IM11i\II
:ú01N.NOI'uN
iIINI/%N%NN
\..411A/NI
/
i11\11%INÌANN
I ./C1111I11I.om%rN
MI
o
o
I
250
500
PLATE
750
1000
Cathode characteristic can be graphed, using ec -ir, coordinate frame, and will be
accurate for one particular circuit.
18
6 J 5
16
O
I.ME
CATHODE SELF BIAS RESISTOR LINES
ib
ib
ec
ib
ib
N
tiy,
14
12
10
.,
I
,°c
4
2
2000
0
0
4
2.7
2
10.6
6
4.0
3
2.4
6
4.8
0
-4
-6
n
MOO
.1000
0
,,..,.
I/!iI
a
2500
O
6
'
6.7
0
2
IIIIMI N" A
MEOW
.,I 1'./I,II
I
om
b
//'í'D' we %
II /,WM!M/1MI/MI
ff
°
,
I
RK
.
1000 SL
8. 1500 11
C 20001L
D. 2500 1L
I~
m
mmiumm/.dm
,tia
"
M
.O,,,
iZ.MMEMM,MMMEiE,
I,MM//,'II,.,I.E
04
AMME,,,,. OM'/,
50
100
150
200
PLATE
32
RK
/
°,.I
IIMEN
NOME
R6
6
I,,
250
section with ee =0, a perpendicular is
dropped to the point A'. Then a somewhat lower value of plate current than
that at ee =0 is chosen and its drop across
-Rk is computed. This point is located at
B (for ip =50 ma), and a perpendicular:is
erected to intersect with this value of
current at B'. A straight line joining A'B'
intersects the load line at the approximate
operating point. The point thus located
is not quite accurate because of the underlying assumption of uniform gain.
VOLTS
Fig. 2. Assumption of constant gain incurs certain error. Constructions are required.
Fig. 3.
Conventional schematic diagram including information usually supplied.
Fig. 1.
300
VOLTS
350
400
450
500
Cathode Characteristic Technique
A method which eliminates the foregoing inaccuracy, but which requires
plotting of points and curve construction is shown in Fig. 3. Here the drop
across Rk is plotted in an ec -ip frame of
coordinates, so that the intersection of
the cathode characteristic with the load
line may be used as a criterion of equilibrium and thereby a specification of the
operating point.
This method is capable of giving accurate determinations provided it is
applied to a specific situation. In this
case, the slope of the load line is chosen
as for the first demonstration and the
location of the operating point involves
no approximations. In Fig. 3, curves are
drawn to present a general situation involving a range of cathode resistance and
load resistance values. Since the cathode
resistance cannot be conveniently included in the load line, determinations on
this base are correspondingly inaccurate.
It must not be overlooked, however,
that when applied to a specific circuit
*Arthur Shach, Electronics, Feb. 1944
Paul Hunter, Electronic Industries, Nov.
550
1945
John F. Rider, "Inside the Vacuum
Tube," 1946
AUDIO ENGINEERING
SEPTEMBER, 1947
Fig. 4. Improved method of locating operating point involves only the application of an
arbitrary linear scale.
with Rk included in the load line, that
the method is capable of precision. Note
that A, B, C, and D are slightly concave
upward rather than straight lines, as
required by the coordinate reference
frame.
Direct Determination Technique
The new method next described requires the computation of one point and
the application of a linear scale, such as
a draftsman's rule. In the diagram of
Fig. 4, the load line for the circuit is
drawn as before, for the circuit of Fig. 1.
The cathode resistance is 1000 ohms, and
the drop for 10 ma, for example, will be
10 volts; the "10" on the rule is placed at
10 ma, while the zero of the rule is placed
at a convenient point along the ep axis.
Since the condition of equilibrium of
the circuit is that ec = ek, the operating
point is now found by matching values
of ec along the load line with values of
ek along the rule. This may be done running the eye along the horizontal coordinates, or a sheet of paper may be slid
down the chart until matching values are
seen along the edge. For the case illustrated, this condition evidently takes
place at ec = -7.9 volts and the problem
16
It.
I4l1
l
?l
W/..
ó
12
Ay.. fne1
1141,
R
...aig14
I,
_
&
W
6
VARACTERISTICS
3
.
19.000 1L
R
1,000
E.
400
" .
,
rip,.
V
..,,.0
200
.
i
y`'
,
,r._e_
600
T00
VOLTS
3=
----
_-. ¡NM
::'
-
4fNß
w
ß
500
g
i
12
a
300
PLATE
16
..r
ß
,
.e.
loo
o
_
--- _
.r®r
.IS"
NM
I/,EI
_~_
A
1
_-
ß
.ß
o
'ry
Specified Grid Bias
One of the nio: t difficult procedures in
non-linear analyses is to determine desired
operating conditions. We wish to find, for
example, the most favorable value of
cathode resistance for the situation of
Fig. 4. Using techniques of plotting the
cathode bias characteristic as shown in
Fig. 3, the labor is multiplied as it becomes
necessary to make several plots (preferably for the RMA values) and then to
select that which most nearly meets the
desired operating condition.
This selection may be made readily
with the new technique as follows: Let
it be specified that Rk shall place the
operating point as near to ec = -10 as
practicable. This means that ek must
likewise be equal to 10 for equilibrium.
It is usually desired to maintain RL at
its original value, changing only Rk to
meet the bias requirement.
In the event, however, that the sum of
RL and Rk may be maintained constant,
it is only necessary to note that a current
of 6.4 ma flows at the intersection of the
load line with ec = -10, or the cathode
resistance is 1562 ohms correspondingly.
The nearest 10% or 20% RMA value is
®
MAE
o
Finding operating point when
37,500 and R1, 2500 ohms.
AUDIO ENGINEERING
100
200
12
4
,-
500
EK
15
o
100
°
l
1
6Ç5
VLATE
AVERAGE
CHARACTERISTICS
I'm.,
I,..
ir.iri
/../.I, ,, .
,.I,
iAUJXbM:f
m
6
-
200
_
A,7,-,
400
300
VOLTS
.
rim
.I I,M
`/LIII°
lIIII/I/I
16
V,
_G
300
400
500
o
When R1 is constant, and
5 (Center).
operation is desired at a specified grid bias,
Ri. may be found as illustrated.
n.«
PLATE
[Continued on page 391
Fig.
y
ßo
I
ovaordoropploorr
'
.grifilir
4
.
,C
5
C
r
M
is solved.
Fig. 6.
.
' .1f
,
6
AVERAGE EPLAT6
A
A.3
/
/
100
REr
37 5001L
25001L
RK
°
.,I r
I.'.' l.,
IIIIII
-
4,
s
'
4,
J
:
.,
r,
200
PLATE
300
VOLTS
:
-
tio
a
y
ArW
o
ff
_
i
.
500
R,
SEPTEMBER, 1947
33
AUDIO DESIGN NOTES
DATA FOR MATCHING PADS
WITH MINIMUM LOSS
T and
R
..
PADS FOR
H
500w
LINE
R,
1
R1
=iiiiiiiiiiiiiiiiiúriiii
iiiMUM --.r//
.121
NNII11I11MINIIIIIIIIIIIII
iiIiiiüi iiiiiiiiiiiiii riiiiii
FOR ANY OTHER LINE IMPEDANCE MULTIPLY
NM=
and
R1
R2 BY
Z0/500
N\
1I
IIIII
NN.IIIIIII IIIII
ME
f,\I
IIIII ----=_____
---- .....----:M
__
--------------------------1-------------------...1...------....----.-.......-1...V--.. -----------M--M.
=P--1-MIMOM.MEE..
'..I.iMEI1
111
ME.
iM
1-1M..E
O
1..
\11NMNN=--NIMM7iMu
..
71.M
E. MEIN
NNE
MIINO1ME
M.
1111 N N MNINIM
nMI1Moi.
MIME
NI1.\I11I1111111111111
MA
I1111111111111111111111
II\1111111I11111111111M
1111111111'11
EI= 111
I11111
M
M
MM.
11111111
----..,.
-------------.,.------------.------_-------.:
--------------i
--.
.....
..ii.- R2(20-60db)
----11iil:l:l:=.
-----M
i-20db
., /M.NME-MM1I.MME.111..
..111MIME
----.m.
---\.....0I11I
E.
-MM.\..1E..I
IM-MIMIM...101I1
--.1M.
NIIIIIINIIII
MN11111111,1NNNNII111111111
=MIM11:1111111
ÁIMIIII1111IIIIIIIII.111
=NINIi\IIINII Ì1111111111111111111111
dill
NII1I1I1I'.NIIIIII1111111M
MUM
IIIIN
---_____
----------------------------------
1111111
MINIM
1111111
111111111111
1111111111111
1111111111111
..1 I
i01w
11111111111
\111111
IN
1111111111
...
1.
:
.. GEM
,
-------=
4320 co
IMMUNE
1'
..,,.-
1
NMI ME ME
.Mn===
EEEOI.O 11111.
MN
N\\
--....
-II1
i
I
MN
111:11
MIIIIII:SWIM
MIIINII111I1111111111
IIIIIIII\\Ví1111111111mIII1II111111
N11.11111
1111111111111
NEM NMI
'HMI
EEN111111
EN
11111111111
P
11111111
1111111111
MM
M
)
ELM'S
11111111111
------
-IMEO
2 (
E11111u111111
MM
MEOWS
.
/ENE
O ISM
Ó
O
X'
íllllllll
II I
--'
__
--------=I.
11
.
--IiN EII M.E I
I
_
\ -M..11111
1 I
1111I11111111
M111111111111111\\1I1í11111111111111 111111111111111MI1111111111111111M
111111111111111111M111111111I1II1111111 1111111111111111111111I111111111111111I1111111
M. I1
II MUM MIEE11111111NE
=i51
1E111111111111
IMINIIIIIII/ANN6U111111I11NM.NIIIIIIINEII1E1111I111111
.IIIIII!1\1II111111MN
M11.1111111
1111111111111
IN1111!AI111:11111111
1111N11111111 MI
AMMO= 1111q11111 MI111I11
11111111111111
-_-_--'--_____
`...
_MIME.
--- ' :=_..--...::
::'
----M.===
...:M.
----...
-----MOI
SOM..1
M.
----------------...
..-...-------..
-------..--....MI.M.----------.....
-----------.....
-----.M
M.11M------MMi
i
NM
I 11
--MANM' 'MIME EMII.Iu111n1111
m-1
II1111111111111111MM11111111111111111111 111111111n1111111111111.11111111111111111
AM 11.1
MS
--.1II.NME
MM.= 11=11=1110101=11
NOMMEIIU1111 PANNE
MOON
1111111111
MIIIIIII
-.----
\\
-----
MM
MA= =-=1.
71.11111111
--IMI
II111
.MM
1
=11=1=5===M1111
II.N.
M
11
MMI,/IIIIIIIII111111111111\\111111111111
11111111111
ININ1111111I111111111111\
11111111
NMI
!1111111111
NIAI1111111
MORI NIIIIIIIIIIII
11111111
11111111111
--.
INI11111
I.111I1111111
11IIIII
111
M
u
u!!
---.
------T1-----°..
---..
---'-----...
-----....
---..
SM
----..1
.M
uM.-M..1-OM.SM
-----...
...
..
=-MIMMM
-------=-M\MM.
....EM.1
.M
=MOMS=
.M
M-MI\1M
M.
N..
..1I..IHai
M.
--n1.
N..
=MUM
111 M.
OWN
--MIOSMNN E.1111
1..1II
N.
IBM
MN
=-I.MN..
E.1.1I
M..
.1I1.1I
M
...
M.
--.IM.
----MI
MNIIIIIINIIIIIIIIIIIINN
=IIM\NE11111111
EN111111111111N
MERE IIIIIIIINNI.IIIIIIIIIIMEEN M1\IIN.111111 MEMO MEN
IMNINNNIIIIII
IIIIIIIIIIINN
NE\INEI
IIIIII N1II11111111
M
111111
11111111111
11\\11 111111
MEMOS
1111
MEIN= M
--MI.
.
IN
IIIIIIIII
.1
MM
1111111111 MM
i
M---...111..
NM
IIM
o
-EMI
.
1
IN NMI
NM=
1
I .
IIIII
III1111I11
,
i
1
II ON
I1I11
IM
IIIIIIIIIIII
11111111111I11111111111111 MI\\111111
M\ I11111111I11
11111111111M
1=111111111111111111
111111
1I11111111111 11II11111111111111111II II1111111111111 11111111111111111
111111111111111111111111111111111111111111\111
1111111111111111
11111111111111111111111111111111III11111111111'C 101w 1111111111111111±11
1
4
5
6 7 8 910
2
3
40 50 60
20
30
80
111111111
LOSS
34
-
DB.
AUDIO ENGINEERING
SEPTEMBER, 1947
do more
it%tI\S
than
cost
at lower
ordinary reproducers
STORMPROOF
ACOUSTIC &
P -M
Unit
Marine Speaker
MATERIAL
Cellular Horn
Acoustic
with Bacon
method giving
makes speakers
patented
BACON
by
a
Only
the outputt of
processed
Cloth which is wall, thereby increasing
vibration. Supplied
a
due
on all straight
without
hornsl,
ho
the
a
reen antStormproof types
(hal
asap when so ordered. weather and tempe rature,
hntee
t
cl matr
regardless of
art
ADVANCED
A
ENGINEERING
Re- entrant Trumpet
engineerin sound reproducer
ou
leadership
pee
almost
BACON'S
for
decad
recognized for
ing has been units have
excess of any
driver
far
BACON
watts,
performance
capacity 30 driving
and continuous watts.
and
speakers
brands-continuous
as city 60
BACON sp they denver more
capacity
Straight Trumpet
& DESIGN
High Frequency Speaker
yet
input
P
energy.
ut.
units repmre less
reproduction output. cutoff frequencies
sound
efficient
BACON s to power quenli es
claims made by
of speakers,
w driver dohs
of
lengths
and acoustic
frequency range
recogand
laboratories
efficiency
at
made
capacity.
by tests industry.
are substantiated
the
the foremost in
nized as
Armored Projector
Dwarf. Re- entrant
COMPLETE
FROM
LINE TO CHOOSE
trumpet or speaker
the
driving unit,application-also
aPP
reThere is a BACON
sound
be
may
that
every conceivableand housings
that
know
(brackets
es. Soundmen
accessories special
line that is complete.
paired
quired fcr
use a speaker
reproducer
and
kind
pays to
of sound horn
makes
auditorium
Yes-BACON
down
length
foot
speaker-from
the giant 7 inch
eve
driver
intercom cone
tiny
4
the
to
to the small P.M. driving unit
the super giant
for paging horns.
Paging Horn
Radial Re- entrant
SEND FOR OUR NEW FREE CATALOG
Bacon Elec. Co., Inr.
19th St., Nev, York 3, N. Y.
Gentlemen: Please send me a copy
of your new free catalog.
SMART BUYERS
52 E.
use
Name
Nair
RACON ELEC. CO., INC., 52 E. 19 ST., NEW YORK, N. Y.
AUDIO ENGINEERING
SEPTEMBER, 1947
Address
City & State
35
FM PILOTUNER
While not strictly within the field of
audio, this tuner provides a means of obtaining an audio signal from FM signals, as an
element in a high- fidelity radio receiver
system.
The Pilotuner is a compact FM tuner,
employing only five tubes and a selenium
rectifier. The circuit consists of a tuned r-f
stage, a combination first detector -oscillator,
two stages of i-f amplification, and a ratio
detector. It is ruggedly built on a 6 x 8 inch
chassis, and enclosed in a wood cabinet.
The tuner has a sensitivity of approximately
25 microvolts for an output signal of 0.1
volts, which is sufficient for most audio
factory in New York City where interference
and building absorption is particularly
pronounced.
The tuner is equipped with a combination
power switch and signal transfer switch, so
that the output of the tuner is connected to
the phonograph input terminals of an AM
receiver and the phonograph pickup leads
are connected to the tuner. When the tuner
power switch is actuated, the output is fed
to the receiver through the shielded lead,
with the phono switch on the AM receiver
in the phonograph position. When the tuner
is turned off, the output of the phonograph
pickup is fed to the receiver.
The half-wave selenium rectifier circuit is
followed by a resistance capacitance filter
which effectively precludes any trace of hum
in the tuner output.
amplifiers. The radio detector provides perfect quieting, and the audio response is extended up to 12,000 cps, with the circuit
being designed with somewhat less deemphasis than is required by transmitter
standards. This gives some equalization to
compensate for drooping high- frequency
response in existing AM receivers.
One feature of the Pilotuner circuit, Fig. 1,
is the inclusion of two r-f chokes in the leads
to the power transformer, and the 50-µµf
capacitor for coupling the power line to the
antenna coil. This arrangement eliminates
the need for an outside antenna when the
tuner is used in areas having a fair signal
strength -and it is, in fact, perfectly satis-
6
16
11.
i1
.
CATHODE -FOLLOWER
PRECAUTIONS
Users of miniature tubes are cautioned
against employing the 6AG5 and 6AK5
types in cathode -follower circuits, since
[Continued on page 46]
¡
o
-
M
1
-J
a
6BA6
ó
=
Ó
o
6846
l
p
1500Vytl
J
6ÁL5
6BA6
1500
ISOOyyf
PPf
1
+
IOOwf
u#0
Ol
or
T
TñK - 2yr
005 out
L
50 vu
T. 1500yyf
470n
1111
AAAATI.
40 Pf
o
T
T
of
40yf
IOO'125 7
ANT
j
41
6
71
FM
LINE
INPUT
60^,
Fig. 1.
5
V PHOONO
OUTPUT
Schematic diagram of the Pilotuner
PROFESSIONAL DIRECTORY
J. LEBEL
C.
AUDIO CONSULTANT
Winston Wells
Custom -Built Equipment
370 RIVERSIDE DRIVE
NEW YORK 25, N.
Y.
U. S.
1121
INSTRUMENTATION
PSYCHO ACOUSTICS
SOUND.RECORDING
ELECTROMEDICAL
SUBMINIATURE AUDIO
Recording Co.
Vermont Ave., Washington
PRODUCT DESIGN
36
,
District
1640
5, D. C.
Designer and Consultant
Acoustical and Electronic Research
307 East 44th St.
MU 4-3487
New York 17, N. Y.
AUDIO ENGINEERING
SEPTEMBER, 1947
Permission Peter Arno
®The New Yorker Magazine, Inc.
PUZZLE
Of all the ways of saving up a tidy
sum of money, one of the easiest and
safest ways is to invest your money in
U. S. Bonds.
You can buy Bonds either through
the Payroll Savings Plan at your place of
business-or if the Payroll Plan is not
Station Scene, 1957. Find the man who is getting a steady income from
U.S. Savings Bonds. He was smart enough to start buying, back in 1947.
available to you, but you do have a
checking account, through the Bond-aMonth Plan at your local bank.
Both ways repay you $4 for every $3
you save, by the time your Bonds mature.
Choose the sum you can afford from the
chart on the right. Start saving today l
Save the easy, automatic way
...with
Save
each week
AND YOU WILL HAVE
In
1
Year
In 5 Years
In 10 Years
$1,440.84
$2.50
$130.00
$668.97
3,75
195.00
1,004.20
2,163.45
7.50
390.00
2,009.02
4,329.02
12.50
650.00
3,348.95
7,217.20
1&75
975.00
5,024.24
10,828.74
U.S. Savings Bonds
Contributed by this magazine
in co- operation with the Magazine Publishers of America as a public service.
AUDIO ENGINEERING
SEPTEMBER, 1947
37
NEW PRODUCTS
PICKUP ADAPTER
Technical Products International, 453
West 47th St., New York 19, N. Y., has
recently announced the Vibromaster Type
M adapter, designed to permit the use of
the new Pickering Cartridge or the General
Electric Variable Reluctance Pickup with
the standard Western Electric 5A arm.
VERTICAL REPRODUCER
CARTRIDGE
Pickering & Co. Inc., 29 W. 57th St.,
New York 19, N. Y., has recently announced
the new Model 131 Vertical Cartridge,
identical in size, appearance, and mounting,
with the lateral cartridge.
The vertical unit has an output of
approximately 20 millivolts, and is designed
to work into a grid circuit, with the necessary equalization furnished by the preamplifier. The frequency response is fiat on a
velocity basis, and excellent response is
obtained to upward of 10,000 cps. As with
all Pickering Pickups and Cartridges, the
intermodulation and harmonic distortion
are held to a minimum.
Tracking pressure for the vertical cartridge is approximately 1 oz. This value
may appear to be high in the light of modern
lateral pickups, but considerably greater
pressure is required with vertical reproducers in order that the entire up and down
motion of the stylus be translated into
effective movement of the magnetic structures. The pickup arm should show no
tendency to follow the vertical modulation
on the record.
The Model 131M Cartridge is supplied
with a 0.002" sapphire, with diamond styli
being available at an extra charge.
D -C
AMPLIFIER
Specifically developed to accommodate
the General Electric variable reluctance
magnetic pickup without the use of any
additional preamplification or equalized
circuits.
Equipped with a built -in specially designed low -noise and low -hum pre -amplifier
and fixed pre-equalizer to fully compensate
for the characteristics of the G. E. variable
reluctance pickup. It contains, in addition,
a variable high-frequency equalizer for
compensation of pre-emphasized recorded
and radio programs, as well as a low frequency equalizer for full compensation of
constant amplitude recordings.
Filtered dc, having less than .03%
ripple, is applied to the heaters of the input
tubes through a regulating ballast resistor
38
to stabilize heater voltage and insure minimum heater to cathode leakage.
Utilizes a new signal self-balancing and
current drift -correcting direct -coupled
output circuit. Response is 20 to 20,000
cycles +1 db. Develops 23 watts with less
than 1% total distortion. Less than % of
1% is present at a 12 watt level. Overall
gain; 117 db. Hum and noise level;- -40vu.
An additional independent input of 500,000
ohms is provided. Balanced output terminals are provided for 8/16/20 and 500
Ohms. In-between terminals provide the
following additional output impedances;
2/4/5/10/80/125 /160 and 175 ohms.
Entire input section is mounted on a
floating plate.
Manufactured by the Amplifier Corp. of
America, 398 Broadway, N. Y. 13, N. Y.
ROBOMAT
Eastern Amplifier announces production
of a completely automatic sound reproducing system for industrial organizations,
stores, offices, restaurants, ships, churches,
cemeteries, mortuaries, skating rinks, country clubs, swimming pools, etc. The new
product is called the "Robomat."
This unit can be used to replace programmed or continuous wired music installations. "Robomat" includes a microphone
for instantaneous paging. A built-in powerful AM radio receiver for special events
gives excellent results even on stations with
a weak signal.
The record changer is heavy duty and
automatic. It stacks 20 records and plays
both sides or a total of 40 discs. This record
changer can be operated for continuous
periods on either a planned program, such
as one record every three minutes, or on a
continuously operating program. All you
do is flip a switch. The "Robomat" is
substantially shock mounted so that it can
operate under extreme vibration conditions,
such as encountered on ships, trains, etc.
The output capacity is 90 watts to assure
coverage of 50,000 square feet in noisy
indoor areas, 200,000 square feet for background purposes and a one -quarter mile
radius outdoors when used with proper
speakers. The "Robomat" goes into operation automatically in the morning and shuts
down automatically at night! It requires no
attention other than to change records once
a day. If records are not changed it merely
repeats the same records the following day.
WAGNER- NICHOLS RECORDER
In response to many requests, the Wagner-Nichols Recorder Mfg. Co., 67 W. 44th
St., New York 19, N. Y., has set aside a
limited number of recording units for those
interested in assembling a complete recorder using their own amplifier and speaker
equipment. Recent improvements on the
original design have resulted in a reliable
unit employing a unique lead -screw arrangement which records at 416 lines per inch.
Fifteen minutes of program time may be
recorded on each side of a single 4 %" Vinylite blank.
These basic units are available complete
with 60-cps, shock -mounted 331/grpm motor
free of wows or vibration, recorder chassis
and mounting plate, crystal embossing and
reproducing head, and a supply of blank
disks. Full instructions regarding their use
will be enclosed with each unit.
Less than one watt of audio power is
required to drive the embossing head, and
the output level from the playback is
approximately 0.25 volts. For embossing,
the head is coupled to the output plate of
an amplifier through a capacitor, while
conventional circuits are used for the playback connection. A complete recording
channel may thus be assembled with the
addition of an amplifier, speaker, microphone, and switching facilities. For recording radio programs, the connections
may be made to the output stage of any
receiver capable of supplying the necessary output voltage. An important innovation in this device is the fact that the same
stylus and head are used for both recording
and reproduction.
This recorder, first announced in the May
issue of this magazine, has been improved
considerably since its first showing, and is
now ready for distribution to technically
inclined users. The complete recorder, including amplifier, microphone, and speaker,
all mounted in an attractive case, will be
available in a few months.
[Continued on page 48]
AUDIO ENGINEERING
SEPTEMBER, 1947
,
TRIODES
Volume Expander
[from page 9]
trol. It is merely necessary to substitute
for the control signal a 22% -volt battery
and potentiometer. The volume can then
be varied smoothly over a range of about
15 db, and the connecting cables can be
of any length whatever, since they carry
only direct current.
A little care must be exercised in using
the volume expander. It should never be
used on program material where the
source of sound is inherently incapable of
a volume range of more than 20 to 30 db.
This applies to solo instruments (other
than piano and organ), solo voices, string
quartets, and so on. On orchestral, choral,
and organ music it can be used on almost
any record with excellent effect. The
actual manner in which the original re-.
cording was controlled determines
whether 8 db or 12 db of expansion can be
used. Paradoxically enough, the wider
the volume range on the original recording, the more expansion can be tolerated.
In Fig. 4, it will be noticed that in the
12 db position the input signal necessary
to cover the entire range of expansion is
about 29 db. This is about the volume
range of a good modern recording. When
playing such a recording it is best to 'set
the expansion control so that the eye of
the indicator tube fully closes on average
peak levels. The expansion will then be
completely off on very low-level signals.
On records of more restricted range, it is
best to set the expansion control so that
surface noise just does not operate the
indicator. This will then give the maximum increase of volume range.
It is good practice to install the expander with a gain control following the
unit. The expansion control can then be
left in the full -on position, and the input
gain control used to adjust for input signal
peaks. The output gain control then controls loudspeaker volume, and all of the
output peaks will be at the same level,
regardless of the actual level on the
recording.
The superiority of a
triode amplifier
is
most apparent in the
final test ...listening
*TRIODE
4 -7A4,
TUBES
2
used
throughout.
-7N7, -5Ú4G, -5Y3G.
1
2
- 6B4G,
*AUTOMATIC BIAS CONTROL greatly increases
undistorted power at moderate cost.
1
*Gain -55
*Three push -pull stages preceded by an inverter stage.
-
Two -stage
compensation
tapped condenser -resistor networks.
*Interstage transformer insures good push-pull
balance.
*Input-38,
150, 500 /600 and 500,000 ohms.
Output-1.5 to 30 ohms and 500 ohms.
*Power available for other units -250 volts,
*Flat within
power and
1 db to 25 cycles at full
to 4 cycles at reduced power.
*Flat within 0.2 db to 30,000
*0.6% harmonic distortion
to 120 db in various models.
*Bass and treble
cycles.
0.090 A DC; 6.6 volts,
and 0.2% inter modulation distortion at 5 watts.
*Rated power -30 watts at 21/2 % total
distortion.
5
A.
*Fuses -Main power and 684G plate line.
*Attractively
finished chassis. High
components. Finest workmanship.
quality
The BROOK HIGH QUALITY
AUDIO AMPLIFIER
Designed by LINCOLN WALSH
BROOK ELECTRONICS, Inc., 34 DeHart Place, Elizabeth 2, N. J.
SOMETHING NEW-APEX CUEING DISCS
A
"must" for every broadcasting station
Smoother "pick- ups " -longer lasting
of dramatic bridges.
Apex Cueing
- AM
and FM
records- accurate cueing
18"
Discs are the answer.
-
plastic discs flocked on top side to prevent record slippage
polished on under side for minimum drag on turntable motor.
$4 :50
Overhang standard 16" turntables by
1
". Already
in use by
networks and many leading stations.
are exclusive distributors - order as many as you
require from this ad. Immediate shipment from stock.
We
JACK STRIPS
KI?LITt; FR ME
Y:lix34NA: IiatXtitétc t,C.
DURAL FRAME
B
Operating Point
[from page 331
1500 ohms, which would therefore be
chosen, leaving 18,500 ohms for RL, for
which the nearest 10% value is 18,000
ohms. The result obtained is precise, even
if the limited range of commercial resistance values prevents use of the exact
value.
Returning to the condition that RL
remain constant, we may make the following summary of a particular problem: RL = 32,500 ohms, Rk is unknown,
Ebb = 410 volts, cc = -15. Required to
RACK SPACE
si
RACK SPACE 2's"
Type 124D Double -Row $
Jack Strip for rack mounting.
Equipped with 48 jacks and
one designation
[Continued on page 41]
AUDIO ENGINEERING
1h"
Type 114D Single -Row Jack
mounting.
Strip
for rack
Equipped with 24 jacks and
one designation strip.
SEPTEMBER,
1947
.50
grounded to frame of jack strip through
mounting screws. No need for grounding
bus from
jack to jack.
Equipped
EACH
with
48
jacks
$
.50
EACH
and two designation strips.
strip.
AUDIO
923
EACH
RACK SPACE 1;4"
Type 134D Double -Row Jack Strip for
rack mounting.
Jack frames already
50
EQUIPMENT SALES
EIGHTH AVENUE
-
NEW
YORK
19, N.
Y.
39
z1 \i
\Ì a,.
........\EN....EMNN.EI.M.N11..1.1.
.I.EN.N....E...N1..E..1.....l'..........I...EMNN.N..NME
I.N.........INENE.1..NEEENME1
=ME MIMI EEMEEEE
MN=i ii
IME i iiiiiiiii
N
i
miiiiiiii
ii
i
o i
ii ooN
1.
.t\E
NEEEN 1 CSNEEN.E IE pM,1M E......1'E p ...IE
NN,
iiiiii 11 iiii pE
iiiiiiiii
.N..NEE1
..
.........1
ii
oi
iiiiiiiiii
i
li
iii
! iiii
i iiiiiiiiii
E.1
V ..1 EE1 E1I iiiii
..EIS
E.EEEE
.ENI'
ME11
ENE.EN
i
a...N.I
1t
EE
E\
ENNNE
EEENEE
1EEME.EM1ENNNEM.\M.....EIMEEMMENME
1E.N.E....ME.E.t1NEMN.EN.11NMEE.NE.1E..E1.N.NEE
1.............,1EE.E.EE.EM1 uNE.E...
um
EN W
Mi
c0
Ì .E
EINS
,\\
INN
Z
1E W
Ì Z MEW Z
Ó
MN.11 o
IY
Z
INE IL
o
\ mum
E
1'.
1\
...
1'...............1
1...1,,11.
mom
1
MINIM
a,
MUMMER a,
MEN NEI
.11\MEt1.ENNE.ÍEMNN.EEEN
.,1.\\..1EM11ENEEIENM.NN
NE..I.EE..Et\MNE.EE.1EEN1. EMI N
\1.
o
o
MEE
o MEMO=
EIMER= INMEN.NEt\ MUMS
1.MEEEE.1\.ENE1E..EE.E,I.......NI............
o
MUM I.ME.MEIEEE
\NEENEEE11EEE1.ENEENE
o
MEW=
EEENEE
MEMO MM. .....1\..E.NNE...\...E....EtVIM
EE.1.ENEN.M.N.E o
.l'EMNE.1'.1\NE1.1N..EEE
MOM= \.\..t1ENMM1EMEMENE..N
,1EMENENM..NM1NEt\ENENNINNE
......\1EE.NEE\.E..M..E,1EEEII.........
MEN
MEN
MEN
MEN
NEE
1ME=
MEE
.1..E1...M.11.MNNE
EEE.E11.E.EE.E
'ME1MEMM\\E1EE1M
1\.1\E...E.IEEN.EE1.NN
1,.EENENMl'ENNN,1E
.E.EEEEE,1N....E....E
1\.1E..NN...t\NMM.1
MIME EME..E
ENN
t1.EN.MN...\N.ENENE,INE1.MEI
IENE
t1ENN1.EM111MEE
t\...N..
....
.EEMEE.E.E1.NN..EEE1E..EtME1.......
MIME= 1 MEN MEN E1\E.EEEt1..Eil.11.EEOM
MtIM.1ENE
MEE MOM
'E.EEN..NNNM1I...N..,I...Et
=MC 7WNEE'E. MEMO. MEER=
INE1.NE
......,EEEMNEM1\WMEt1EEE1
=MIME I.NMEE EME o
...EEE...M.E....1\.E.E..N.E.N.,INMEENEE1EEEEII.EEEN
..,1............
EE....M1\NN...t\..EE.11..
Emu o
.....1\EM.
MIN
NE,1.
MEE
1ME1EMEEM
.,IE.ENEMNNNE17NNE1
M1EE.1...
1\MEEEMEt1.EN.E1...E..1 MOM 1...E.E.
...,IMEEEEE..MEE...NEE
W
MEE
ENE
NEE
MEER
NNE
\\.t\MI'EE1
.....1I..E..
.........\\.\11IE1NIINEEN.
EM7,
NEE.E
1111,1NEE.E.1EE.t1N..1...11.........
1EE7E11IN
.\\EM\ENE...1
=mg IN...
...\\t,1M1MEEEEII
NENM1N1
E1..E.E.E.EEEMIENME.E.EI......11EE...
NNNNN
NEE
..EEENM.N1EEE1EE
1EE
..EEEEN.E..N..I..E..EE,I..E..\..E1
MEINENEpEE..EEE.EVEEShM1'EM.t11
E1.ENE
IEM 1NN.....1..ENE.\1.........N.
EEE,
\E1 EE.
1EE\
E.NEt.1EN,1NE.11E
N
N
E
E, NE N11ENl'EN E
i 1EE.
1
MENE
..,1EMNEE.\N.
I..E
N.
E.
1........
I.EE.M...
I1
E.
1M.E.\1NEEE1\ENE,IEENE1
.ENNE
E
n.E.\\ENEE,INNEEl'EE
1MNENER'MEME\
1.
ENEM
V.NE.M.L IMM1ME11IMM.
IEE..\\EE.NE.\.E1
MINE
ENE
ENE
NEE
=UMW
=NEE =Mk
MEE
3
MERMEN
N
......N
.......0
M
N
E
.
M
MENE.
.N
ENNE
N
ENE
.ENEM
ENE
.ENNE.
ENNE E.
MENE
.E.NEN
.EM
.. ...
N
NN
.....
E
EE
ENE
MEN E.EN
..
E.
.......
E
..r.
EM
E
ME
MEE
.......
MEN
ENNE
MMN
MEN
MENEM
MN IMO
.NNE
M
M
MINE
MIME
MEER MEN=
VMS= MEMBERS
MN=
MEE
I..... MENE I N = N\
momREM EOM
mmum
NNE
IN
NEME
co
MENEE
MEN
NEE
EE
N.
NEI
ENE IEEE
NEE
LNENNN
INNEN1'EEE
oI1
ti
z
w
z
o--I
ow
o
EME
l'NEE
MENE
oI-
MOM
RIM E.\\.ENNEEE\\,........EII.N....
'N.N..ME.M.
'MEE NNEMN
MINIM IMME EMMA
RE= MEMO o1nEEE.EEE
MOM 1E1NEE z
nM.I\\11M
1
.
1\
1M.E...ECEE..N...M.M'MEN.E I..N.....1 MENE.EN J
IEM...EEE'EE..EN\..uENNEmom
mom
EMILE'\E.CEAEaEt1NM111EIMME
MIMI
EMEN
I....E.E..NEMM..O.
MEN
..E.11.IMIMEN
1EE..EE.t1EEENE\\EEE1EE111MEN
1..EE.E......E.EE\..E.EN.N
NN.EE.'EE.NNIIEE.1'.EE.
o
1NE.NEEEEE1 MESON M\\Et\E1
MENI NNEE o
SEEM MEN IEMIME
1M'\71l'
1..............NE1'E1Et\.WOE=
1NMMEN\\E\
1E..EEEEE,1EE.t
\EE1'..1 NM 1E
INE
1N.E..
MEE E..E\1...'EEE.\.Et11MEN
1ME..NE.EE.N\
MENE.!
EMEI..E11.WEN
1NNEENME'EME=11MI'MEEN
INEEEEEEEENNE1.EEN\1N
\E1EEEE
1.........NN.........E.E\\.E.....'mom
1M.mom o
INEENENEN..E7M\IE\\1
o
EMME.EMNNE..1E
;_:_:'=:1::::
M111E
,`. 1 1'11
1ENEEEEENM..INMt
,10!M'
I.N.EEEE.EEEE.NEEENEE
1MME\ M
lemma
INE MIME
1.0
1
`
I
.
.
UWE
IMENE
.1
::::::::::::::::::::='IBIÌ:=S':.limi
::::IC::CC
1......MN......MN.....E.1M.E,
IMO
\....NE o
:::::
C'.::CC:C:::S:Q=C::C?I:d::::::::
1MN....NEN.MNE.N.EEE..\...N\1.\
1.N..E
E..MEN.MEEE\E7E,1E1N
IE
MNM\\t.I1'NI'tNE
N
M'E
1EMMMEMM\1M\
1MENEM\NM71E111
MEE
EN
=MEMO
MENE
=II
1.III
ME
1
\7
\Et, M l.
MOM MOM MEN MEEMEE
WEN .MEEEENNNN.N\E7EU'11
E.NNEEEEEEEE..EEEEE.EEEEEE.\\Et.IN111
EMS
NNE=
MOM
Rim
EE..M
OHM MM. .UI
/WM NM/ MEN
M.EWE
MIME MEN MOM EIIMIgI1
NUM MEE
MOM REM MIR NNEEE..EEMEEM
MEN
E\\\\,\11
LEM OEM MEN EENNNNME.NE EMS MIN
NNN\IM11'.I
.EE M
EEEM
...M.EE
MMNE\\1E11
.EEEN
Il
Eh.
E.EEME.NMEE!EEEE.NEEEE..E..NME,\'I
NICEREEENNEENENINWV
ER\NNE\.
r)
N
8a
40
NI
SSO-1
AUDIO ENGINEERING
SEPTEMBER, 1947
[Continued from page 39]
determine the necessary value of Rk
precisely. Since the final slope of the
load line is likewise unknown, we must
start with the available locus, which is
that of ec = -15. It is desired to travel
down this locus to the point of equilibrium. When this point is found, ip
will be known as well as its drops across
Rk and RL.
The condition of equilibrium is that
Ebb = ep+eL+ek, and that ek =ec. This
latter fact allows a solution, since we may
write Ebb = ep- FeLd-ec, or Ebb =ep FeL+15.
Since eL =ipRL, a linear scale may be
placed across the coordinates as shown in
i
n Fig. 5.
Writing
ipRL =Ebb -15 -ep,
another linear scale with intervals equal
to those of the plate volts axis may be
placed to graphically subtract 15+ep from
in Fig. 5. All figures read
on this reversed scale are possible values
for the right side of the foregoing equation; all figures on the oblique scale are
possible values for the left side of the foregoing equation.
To find the equilibrium condition, the
left and right sides of the equation must
be equal. Therefore, slide a draftsman's
triangle or a sheet of paper with its comer
on ec = -15, keeping the right -angled
device square with the coordinates as
shown in Fig. 5, until the same value is
read on both scales.
This evidently occurs at ip =2.75 ma.
To obtain 15 volts drop at 2.75 ma obviously specifies that Rk =5450 ohms,
and the problem is solved precisely.
Ebb, also shown
Unbypassed Cathode Resistor
This technique is further adapted to
finding integral values of input voltage
along the load line for Rk, unbypassed
(see Fig. 1). In this event, an a-c component appears across Rk which is of
opposite polarity to et. We observe that
at any time, ec = et -ek; when et =0,
ec = -ek, which relation was utilized to
locate the operating point.
With current feedback, and et = -1,
ec = -1-ek, or the rule in Fig. 4 may
be slid parallel with itself to make "1"
fall on the ep axis instead of "0 ". Then
the point of equilibrium is found for equal
values of ee and et, as before. But since
the scale for ek has been moved one
volt, the input voltage is precisely one
volt greater at this point.
This is illustrated in Fig. 6 for RL =
37,500 and Rk =2500. The operating
point is found as before, and falls at
ec = -10. Let it next be desired to find
where et = -10 falls. The ek scale is accordingly displaced 10 volts as shown, and
it is seen by matching equal values on the
two scales that et = -10 falls at ep = -15.5.
As before, this is a precise determination
within the limits of graphical accuracy.
The process may be rapidly repeated
for any integral values of et which may
be of interest for the circuit under
analysis.
AUDIO ENGINEERING
tl
"New Times -New Modes ", says old proverb.
These new
attenuators were born to meet new war -created demands
They represent a new medium frame size: Type 800 121/4"
dia.) and a larger size: Type 900 13" dia.). The Type 800
is supplied as potentiometer, rheostat, ladder and T -pad up
to 20 steps. The larger size Type 900 is similarly furnished
with up to 45 steps. Write for new bulletin.
`
S)
SJ
Manufacturers of Precision Electrical Resistance Instruments
337 CENTRAL AVE.
JERSEY CITY
7,
N. J.
THE COLLINS 14'/AM TUNER
These tuners are custom built on direct order available with extra
added accessories on special request such as: bass and treble tone
controls, precision calibrated tuning dial, tuning meters, 500 ohm
output transformer and many other features suited to your particular needs. You may request your tuner the way you want it.
The tuner is also available in attractive leatherette covered
cabinets in a variety of colors to match any interior or furniture.
Mahogany and walnut finishes are also available.
The COLLINS FM/ AM tuner is a precision instrument which will
give you laboratory performance from both FM and AM broadcasts. Do not be satisfied with less than a COLLINS tuner for full
enjoyment from your favorite programs. COLLINS means
QUALITY and PERFORMANCE.
Write for descriptive literature on the COLLINS FM /AM
Tuner and matched amplifiers.
COLLINS AUDIO PRODUCTS CO., INC.
126 Park Street
SEPTEMBER, 1947
Westfield, N. J.
41
What
do you
need...
If it's your store's future, you
not only need attractively priced
merchandise; you also need your
community's good will, plus the
assurance that your community will
have a reservoir of savings, to insure
continuous sales.
If it's an individual's future, he
needs funds to buy tomorrow's necessities and a reserve to meet financial emergencies.
If it's
the country's future, we all
need a government and a nation of
people with the money to operate
efficiently.
YOU CAN HELP wrap up all
these futures securely -by promot-
ing sales of U. S. Savings Bonds.
Devote a generous portion of your
advertising schedule to Bonds. Include in your copy an explanation
of the "Bond-a- Month" plan (see
details in box). If you prefer to utilize the official Treasury ads for this
purpose, mats are available through
your newspaper or your State Director of the Treasury Department's
Savings Bonds Division. He can also
help you build up your Payroll Savings Plan -the plan that gives employees a feeling of extra security
and thus enables them to be happier
and more efficient in their work.
Every $ 3 put into Savings Bonds pays
$4 at maturity.
THE
"Bond -a- Month"
PLAN
For self-employed people and
others to whom the advantages
of the Payroll Savings Plan are
not available, the Treasury Department and the banks of America have established facilities for
systematic Bond buying through
monthly transfer of funds from
checking accounts.
Use some of your advertising
space to tell your community
about this new opportunity to
invest in the future!
The Treasury Department acknowledges with appreciation the publication of this message by
AUDIO ENGINEERING
This is an official U.S. Treasury advertisement prepared under the auspices of the Treasury Department
and The Advertising Council.
4R
AUDIO ENGINEERING SEPTEMBER, 1947
Voltage Stabilizers
[from page
241
a-c or d -c, because it is used only to
heat a filament, it is equally obtainable
from the output of a selenium rectifier
stack and the necessary filter, thus offering a high degree of regulation to lowvoltage d -c supplies. This type of instrument is also available under the trade
name "Nobatron," and an equal degree
of voltage regulation is obtainable.
Certain types of thyratron-regulated
supplies are noted for the poor waveform,
largely due to the steep wave fronts
resulting at the firing point of the a-c
cycle. These disturbances are usually
measurable throughout an entire building
in which such supplies are used, and very
often seriously affect the performance of
other equipment. With the regulation
provided by this type of regulator, it is
noted that the changes are comparatively more gradual, and the input waveform is not affected. The waveform
of these regulators is well within 5 percent harmonic content, and the correction is applied within approximately 6
cycles of the line voltage, or one-tenth
second on 60-cps lines.
Where audio circuits are line -voltage
operated, considerable improvement can
often be obtained by the use of a regulator such as that described here. Inasmuch as the advantages of regulation
can be obtained without the sacrifice of
a reasonably good input waveform, and
at a relatively low cost, it may often he
the means of improving distortion and
output level characteristics of a sound
system. For those users who are limited
by an uncontrolled line frequency, this
type of regulator offers an inexpensive
solution.
frequencies which do not appear as periodic
vibrations in the stimulus. The reverberation of rooms may be an influence. Phase
differences are usually not acoustically important because, except in extensive filter
circuits, the time lag is too short (less than
180 °). But even with the most simple
mechanical system some transient distortion
is, theoretically, inevitable. This is acoustically harmless if non-oscillatory; but if
oscillatory at an audible frequency with a
long period then objectionable foreign components are heard. The reduction of mass is
important to reduce transient distortion, as
it is to improve frequency distortion.
7. Other varieties of distortion may come
from: playing phonograph records at incorrect or variable speeds; introduced variations in volume level (fading in radio reception, or standing waves in rooms); and
transmission or reproduction at an unnatural volume level.
The poor acoustics of many audi-
toriums give audiences little chance to
hear a high quality of live programs. So
their hearing gets inured to extraneous
noises and reverberations and becomes
trained to perform a certain amount of
directional discrimination between them.
Although they may not be aware of the
cause this produces some aural -mental
fatigue. The added strain of aural attention might be compared to the visual
strain of flicker, although not analogous.
The binaural hearing of the listener also
adds a certain spaciousness to the live
program which is lacking from loudspeakers supplied from a single channel.
This is lost to the single microphone
and in no way can it later be restored.
Then, too, a microphone has directional
limitations. The two ears gather a large
percentage of the sounds from reflective
surfaces in all directions. Perhaps a
multi -directional composite group of
microphones spaced according to the
listener's ears and located at a favorable
listener position, instead of near the
sound source, would give an improved
result. But it could not equal stereophonic broadcasting and reception.
However, the microphone itself lacks
fidelity and uniform response over the
desirably broad range up to 15,000 or,
possibly 20,000 cycles. It must receive
and suffer from high electronic amplification and the balancing of faulty response
-electronic tampering.
The very minuteness of the voltages or
currents produced by microphones or
phonograph pickups make them more
difficult to discriminate from the inherent
noise due to the thermal agitation of electrical circuits in amplifiers. The threshold of hearing is only about 11 db above
the noise by thermal agitation of electrons, so small amplifications make the
latter noticeable.
Thermal Agitation
The square of the effective value of
such noise voltage components (E) produced by thermal agitation, where (R) is
[Continued on page 44]
RADIO ENGINEERS! HAMS!
If you are interested
radio, you'll want
month.
Musical Acoustics
handy
in
amateur
"CO"
every
Subscribe now.
Use the
below.
Attach
coupon
[from page 31 ]
2. Internal noises, often caused by a
variable impedance in an electric circuit, or
from an induced e.m.f. Noisy carbon granule and condenser microphones are also
guilty.
3. Amplitude distortions from failure of
the r.m.s. (or peak) values of response to be
proportional to the stimuli under steady state conditions and at different levels of the
intensity of the stimuli.
4. Non-linear distortions which fail in
proportionality between instantaneous values of response and stimulus, phase differ-
ences being ignored.
5.
Frequency distortion from the failure
of proportionality between r.m.s. (or peak)
values of response to stimulus at different
frequencies.
6. Transient distortion. During a transitory period a sound may be subject to any
of the foregoing distortions, but it may also
have a duration of transitory period which
may exceed that of the stimulus, and which
may differ in extent for different frequency
components. Moreover, transitory oscillations may be produced in the response at
AUDIO ENGINEERING
your remittance and mail today.
-
-------CUT
CQ -RADIO MAGAZINES, INC.
342 MADISON AVE., NEW YORK 17,
Sirs: Here is my
of CO.
year.)
check (or
-24
12 issues $2.50
Please indicate:
NEW
i
OUT -MAIL TODAY
Call
N.Y.
money order) for
S
Send me
issues $4. (Foreign subscriptions are
issues
$1.00 higher per
RENEWAL
Subscriber's Name (print carefully).______ ._.
Address
City
SEPTEMBER, 1947
Zone
State
43
[Continued from page 43]
the hot cathode (filament) temperature
the resistance component of the im- T1 is given by the more complex formula:
pedance which produces voltages of
RP
thermal agitation, and is a function of the
=
R° (ToR, +TiRo) (f1)
frequency (f) which may be expressed as
(4 +Ro)2
the integral of Rdf between the limits of
It is evident that both may be factors
f' and f, may be expressed when R is
constant by the formula E2 equals where many stages of amplification and
compensating electronic circuits are in4 K T R (f' -f) where T is the absolute
temperature in degrees Kelvin (equiva- volved. But because of subsequent
lent to degrees Centigrade above -273 ° C.) amplification of such extraneous noise the
and K is Boltzmann's constant of 1.374 first stage is most important. If the
electron emission is not uniform from
(10) -23.
The following tabulation of the values oxide-coated filaments or is insufficient
for a few conditions will indicate the for an adequate space charge such thermal
magnitude of the voltages produced by agitation noises may become troublesome.
Other noise sources from electronic
thermal agitation for a temperature of
amplifiers are:
80° F. (300° K).
1. Hum induced from electromagWhen R is expressed in megohms,
netic or electrostatic fields of
instead of ohms, and E is expressed in
power line frequencies into filamicrovolts, instead of volts, then
ment or heater leads, filter chokes,
E= 0.12841/VR(f' -f) where f' -f is the
or amplifier or power transformers;
frequency range in cycles per second.
or where there is high impedance
to ground (grid circuit of AF
Frequency Range
E in
microvolts for Resistance
amplifier open to grouud or with
(in cycles)
in megohms where R equals
high resistance to ground).
'/2
2
2. Microphonic noises from the jar5,000
6.42
9.08
12.84
ring of tube elements by acoustic
10,000
9.08
12.84
18.16
vibrations from the loudspeaker.
15,000
11.12
15.73
22.24
3. Regenerative changes in amplificaThe thermal agitation voltage E
tion and frequency response from
appearing across the plate load resistance
feed -back coupling in an amplifier.
Ro as a result of thermal agitation in R0
4. Overloading of amplifiers, causing
and temperature To of load resistance
intermodulation aberrations.
and in plate resistance R9 of the tube at
(To be continued)
E
1
*
making eugiuF'erinb. miracles
(ome true
Electronic Organs
[from page 22]
The same is true of coupled notes, yet
both of these things are accomplished in
the electronic instrument.
The attack and decay of the tones of
our instrument is very abrupt, as is true
in all cases where the output of the tone
generator is keyed directly. This has the
advantage of permitting great rapidity
and accuracy in the execution of musical
passages, but it is also a disadvantage, in
that it sometimes introduces key- clicks.
A characteristic of this type also places
a greater demand upon the skill of the
performer, since he must play legato
passages (consecutively connected notes)
with a slight overlapping of consecutive
notes, to accomplish a smooth transition
from one to the other. However, the
effect of a sharp attack diminishes as the
reverberation time of the room increases,
and is scarcely noticable when the period
exceeds one second.
The oscillators used in our electronic
organ will maintain their frequencies
within plus or minus 0.25% over a period
of months; through normal variations in
room temperature. A tolerance no greater
than plus or minus 0.1 % would be highly
desirable, but would entail the use of
oscillators that were less adaptable to our
instrument.
The best of pipe organs is notoriously
poor from the standpoint of frequency
stability. "Flue pipes" (those whose frequency is dependent upon the length of
the air column), which make up the bulk
of an organ's stops, are subject to a one
per cent increase in frequency for every
nine degrees F. rise in temperature. Reed
stops, on the other hand, usually exhibit
a negative frequency-temperature characteristic. The change per degree is less
than for flue stops, and is largely dependent upon the temperature coefficient of
the material from which the reed tongue
is made.
Changes in Pitch
The really important new radio and electronic products
usually are manufactured by men
with vision
backed by components with a long record of dependable performance. Among
-
MAIN
OFFICE FACTOar
such components, small but
mighty El -Menco capacitors are
generally rated first.
MOTIVE MFG. CO., Inc.
WiIIimantic, Conn., U. S. A.
THE ELECTRO
MOLDED MICA
44
EFifienc
CAPACITORS
Send for samples and
complete specifications.
aMICA TRIMMER
In addition to frequency drift caused
by temperature changes, the stops of a
pipe organ will change pitch with changes
in atmospheric pressure. Humidity has a
pronounced effect upon wooden pipes,
due to the swelling of the wood, and its
increase in density with the absorption of
water. It is possible for these effects to be
additive and, when this is true, a pipe
may be as much as a half of a chromatic
interval out of tune. Most pipe organs
are tuned every three to six months, and
are often tuned immediately before an
event of importance in which the instrument is to be used. Normally, a wellconstructed, well- maintained pipe organ
will remain within plus or minus 0.5% of
the frequencies to which it is tuned,
barring severe changes in the weather.
AUDIO ENGINEERING
SEPTEMBER, 1947
Most of the stops will change pitch by
the same amount, so that the relative
deviation between stops, and between
notes will be slight. The reeds excepted!
The subject of pitch tolerances is a
highly controversial one, but is of sufficient importance to warrant the concerted attention of everyone associated
with music.
A person having "absolute pitch," or a
well developed sense of relative pitch,
can distinguish between consecutively
played notes differing by as little as
0.2%, under some circumstances.
When a single stop (or a group of stops
derived from one rank of tone generators)
is played, the stop as a whole may be two
or three per cent off of its nominal pitch,
without offending the average musically
trained listener; this, providing the intervals between the notes of the scale are
accurate within about 0.5%.
If, however, two stops differing in pitch
by several per cent are played together,
the resultant sound will be extremely discordant and unpleasing. The apparent
pitch of a note will be midway between
the two, and a warbling effect will be produced by the beats between them.
If several more stops are added, and
are tuned to frequencies somewhere between the two extremes, the total effect
will not be unpleasant, since a grande
celeste will have been produced. Generally, however, a grande celeste should not
have a band width greater than about
two per cent of the nominal frequency.
It may be seen from the foregoing that
considerable deviation in pitch can be
tolerated when either a single rank or a
large number is played in unison, but
that the tuning must be critical when
two or three stops are used together.
This applies, in particular, to the upper
registers of the scale, where a slight detuning of a note can cause it to beat with
its partner at an objectionable rate. The
same percentage of detuning, several
octaves lower, would yield a beat too
slow to be noticeable during normal
playing.
Since an organ must be capable of being
played with any combination of stops,
and in combination with other instruments, it should be designed to meet the
most critical conditions which it is likely
to encounter in fulfilling its purpose.
This would call for a tolerance within
plus or minus 0.1% for each note, with
some stops purposely detuned to improve the ensemble. Whether we adhere
to this standard, or accept a reasonably
satisfactory compromise, is determined,
mostly, by the price range within which
the instrument has to sell.
We have seen that the functions of the
pipe organ can be successfully duplicated
in a purely electronic instrument, and
that introduction of electronic technique
to the art offers means for improvement
in performance, which would otherwise
be impossible.
AUDIO ENGINEERING
The electronic organ described in this
issue is but one example of the many
types now in existence and under development. Others will be discussed in
their turn.
Ultrasonics in Liquids
[from page 18]
The construction shown is merely for
discussion of the principle. A concentric
injection of the air would meet the op.
crating conditions better.
Each such cylindrical slug of water
weighs 6 micro- pounds, and the total
energy per jet just under 400 watts.
Since our present turbojet unit has 80
jets, we could emit 41 h.p. of bullets.
This looks definitely promising.
One of the difficulties is the high
power density, being about 200 kw /cm2,
so in many cases a splitter or diffuser
would be employed to spread the energy.
A great advantage in many cases is
the fact that you have no boundary
layer problem. A very difficult problem
is putting energy into a hundred-pound
sand casting. It is difficult to imagine
any type diaphragm working into the
part of the casting exposed in a pouring
gate unless it was inserted while the
casting neck was still liquid, and even
then the tendency to form gas at the
diaphragm would probably reflect all
SEPTEMBER, 1947
the energy. Shooting liquid bullets at
it insures a continuously clean surface,
and even if the neck wears down an
inch or so in the twenty-minute application, it is thrown away anyhow.
The impedance match problem with
bullets is rather unique. If they arrive
at a uniform rate they soon set up
oscillations in the load due to the elasticity of the load. If they are a third
cycle long they give a properly timed
push each cycle. Now if they give up
their kinetic energy in their own length,
as a liquid bullet must when hitting a
solid load, the impedance match is automatic, as it is simple matter of them
developing a push.
A bullet made of air would give up
all its energy, as would one of mercury,
if both had the proper velocity, frequency and length. Since the mercury
would weigh about 13,000 times as
much as the air bullet it would have that
much more energy.
There are some points of similarity
with a bullet fired from a rifle. Here
we have gas coupled to a very dense
solid, and from one point of view the
coupling is by time. If the barrel is 18
inches long it takes about a millisecond
for the bullet to be accelerated to 3000
feet a second. During that time the
powder is generating about a thousand
kw. In a pistol with a 3-inch barrel, the
[Continued on page 46]
THE MOST
EXPENSIVE...
INDISPUTABLY
ACTUAL
SIC!
The Pickering Cartridge provides
the cleanest reproduction ever
achieved, with linear response
to the limits of audibility.
It tracks with only 15 grams
pressure and fits practically
any arm. It is acknowledged to
be the finest record reproducer.
PICKERING & CO., MC., 29 WEST 57TH STREET, la. Y. C.
45
GREATEST ACHIEVEMENT
geaptd t
iati
ct
(+1
MQM
0IXlroa4
NO WIRE
NO TAPE
NOT A CYLINDER
INSTANTANEOUS RECORDING!
!
[Continued from page 451
coupling is much less effective as it has
less time to work, and we may only reach
500 feet a second.
There are undoubtedly an infinite
series of liquid and solid generators
waiting to be discovered. It seems to be
a good rule to look around for phenomena that are now considered parasitic and troublesome, and analyze their
possible application to the generation of
ultrasonic power. It is possible almost
every such discovery will fill some need
in the wide horizons of the ultrasonic
world.
Multi -Lingual Systems
!
!
The hit of the I. R. E.
Show at Grand Central
Palace! The sensation of
the R.M.A. Chicago
parts show!
10" wide, 13" long
8" high
Approx. 15 lbs. packaged
$7950
Records
a full 30
minutes on a 4!4
plastic disc! Costs less
than tac per minute!
Full
range high-fidelity music and voice
recording!
YOU DON'T HAVE
TO WAIT- Experimenten! Technicians! Hams!
It will be many months
before t h e complete
Wagner -Nichols unit
with speaker and amplifier-will be available to the general
public.
-
However,
there are a
limited number of
"knockdown"
units
immediately available
for recording experts.
Attachment to your
own equipment is a
matter of only a few
minutes!
Read:
TELE-TECH,
June, 1947, pgs 55 -56
-AUDIO ENGINEERING, May, 1947, p. 24,
ff.- SCIENCE DIGEST,
August, 1947, back inside cover.
SCIENCE
ILLUSTRATED OCT.
1947, back inside cover.
f. o. b.
}
N. Y.
Expert's Unit Includes:
Dual- function crystal recording and play-back
recorder head (frequency
response 50-5000 cycles
flat within +2 D.B. signal
to surface noise ratio,
35 D. B.) SPECIAL Sap phire Embossing stylus
(.0005 radius) for both recording and reproducing
foolproof precision tracking mechanism (embosses
416 lines per inch
33 /a
G.I. gear reduced motor
set in shock-proof mounting for minimum vibration. D. B. meter mounted
and wired directly on
panel. Completely wired
switches and special attachments, c i r c u i t diagrams and descriptive
-
I
boo
1
klet-PLUSIdot.
434" vinylite
recording
discs. Additional discs
supplied at $2.00 per doz.
20
%deposit required on
C.O.D. orders. We pay
shipping charges if payment accompanies order.
[from page 1.21
of the type shown in Fig. 4. It is a 50ohm ribbon unit similar to those used by
interpreters. In its base is a push- button
which sets up a light in an annunciator
strip before the chairman indicating which
delegates wish the floor. Hidden under a
lucite bar directly beneath and parallel
to the ribbon microphone is a signal light.
This light may be flashed slowly by an
interpreter whenever an enthusiastic
delegate starts racing along too rapidly.
The progress achieved through the use
of the facilities described above is commendable if we reflect back to the days
before such facilities were available. In
earlier times, neighboring nationals participating in trade or treaty conferences
scarcely ever left their luxurious hotels.
Not familiar with the many tongues
spoken, they spared themselves embarrassment and discomfort by assigning
deputies. Huddling together after sessions these deputies unscrambled their
findings and at long last presented their
superiors with fairly complete interpretation. The United Nations delegate
of today has no need to participate "by
proxy" --he is able to keep himself posted
and up-to-the -minute on its proceedings.
Distributed By
SOUND EMBOSSERS, INC.
P. 0. Box 503, Times Square
Stalin,
New York
18, N. Y.
Technicana
[from page 38[
VERTICALLATERAL
9A, 9B, D-93306, MI -4856
Reproducers Reconditioned
Only new parts
used-includes repainting and
replating:
Response checked with advanced Laboratory
equipment
FAST SERVICE
VIBRATION SYSTEMS, INC
1040 W. Fort
Detroit 26, Mich.
ROTARY CONVERTERS
D, C. TO A. C.
Input: 115 Volts D.C., 2.7 Amperes, 3600 R.P.M.
Output: 110 Volts A.C., 1.14 Amperes, 60 Cycle.,
Single Phase, 125 V.A.
Filter Equipped, Manufactured by
ELECTRIC SPECIALTY CO., TYPE
R-182 -N E W EQUIPMENT -Sur-
plus-In Stock
PILOT MARINE CORP.
29 B'way, N. Y. 6, N. Y.
46
Whitehall
3-3734
the internal shields of these types are connected to the cathodes, and are not therefore capable of being grounded, as such
shields should be, according to an application note in Radiotronics.
Recommended types for cathode-follower serivce are the 6AU6 and the 6BA6,
both of which have the internal shields
connected to the suppressor grid. Optimum connections for these tubes are with
the suppressor grids at the same d -c
potential as the cathode, but with an
effective by -pass to ground.
SOUND MOTION PICTURE
HISTORY
For those who are interested in the
history of sound motion pictures, the
series now running in the Journal of the
AUDIO
Society of Motion Picture Engineers will
be found of considerable value. This
series, started in the April issue with parts
1 and 2, and continued in the May issue
with parts 3 to 7, bringing the reader up
to 1930, has been compiled by E. I.
Sponable of Twentieth Century -Fox Film
Corporation, and will presumably continue in succeeding issues.
Not only does this listing comprise the
technical developments with which the industry has grown from the early parttalkies to a full-grown, highly artistic
enterprise, but it also includes most of
the commercial steps, licensing agreements, and the other data necessary for
a complete familiarity with the art, arranged in chronological order. Obviously
it is impossible to abstract this information for the benefit of our readers, but
those who are interested are referred to
the Journal.
In addition to this history from the
viewpoint of the production of sound
pictures, another listing gives the contributions of the theatre service organizations to the development of present day
picture quality. This material is also in
chronological order, as seen by E. S.
Seely, Chief Engineer of Altec Service
Corporation, and appears in the May
issue of the Journal. Between the two
articles, a very complete history of both
divisions of the industry is presented.
IMPROVING SPEAKER DAMPING
The following method, long used in
disc and light valve recorders, has been
found very effective in smoothing out the
bass response and improving the damping
of a cone loudspeaker.
For example: Connect an 8-ohm resistor
(of the proper wattage) across an 8-ohm
speaker voice coil, connecting the resultant 4-ohm impedance to a 4-ohm output
transformer tap. In this case, 50% of the
power is being dissipated in the resistor
and the output stage should be capable
of supplying this loss and still give the
required dynamic power range. Lower
values of shunt resistance may be used,
making a further improvement in damping, but the resultant parallel impedance
must always be matched to the proper
output transformer tap to reflect the
proper load impedance to the output
tubes.
All loudspeakers display a large rise in
impedance at the cone and cabinet
resonant frequencies and this causes them
to continue oscillation after the electrical
excitation has stopped. The shunt damping resistor damps this oscillation as it
absorbs power generated by the speaker's
free oscillation and thus loads it mechanically, arresting the free oscillation.
This system is particularly effective on
output stages without inverse feed -back
loops, making the output tubes considerably more independent of reflected L. S.
impedance variations and improving L. S.
damping. Charles L. Benson.
E NGINEERING
SEPTEMBER, 1947
Letters
VIBRO MASTER TYPE K
[from page 5]
importance. High-fidelity music is quite
probably an acquired taste which will gain
acceptance when the public is given an
opportunity to hear it more frequently.
In connection with the matter of radio
and distraction, an unpublished study made
at Wayne University about ten years ago
indicated that radio entertainment was generally distracting, but in the case of certain
well- conditioned subjects, both quantity
and quality of work were increased in the
presence of radio entertainment. These
tests were made with college students and
used simple addition and a content test on
fiction reading as the work to be performed.
The fidelity of reproduction was comparable
to the average home receivers.
It would be interesting to conduct a
similar test with variable frequency response
to scientifically test the thesis of distraction
by high frequencies.
VIBROMASTER TYPE K
PROFESSIONAL ARM
R. B.
for General Electri: Variable
Reluctance Pick -ups Or the New
Pickering 120M :artridge
Resonance characteristic flat plus or minus
Db 40 to 1000 cycles measured in 10
1/3
cycles steps throughout resonant range).
22 grams pressure when used with General
Nottingham
Sales Manager
Acoustic Products Division,
The Brush Development Comnr .ly
PROFESSIONAL ARM
FOR
I
... will
2
... resonance
r\I)IO(
!
>O;,I;)
AUDIO ENGINEERING
.
1/2
characteristic flat
plus or
DB 40 to 1000 cycles.
3
... balances
4
morning after handle prevents record
damage when cueing.
GE cartridge at 22 grams and
Pickering 120M at 15 grams stylus pressure.
...
$34.50
NET PRICE.. (less cartridge)
S1i':
.
CARTRIDGES
neither introduce nor amplify lateral distortion.
minus
see your distributor or write
Electric variable reluctance cartridge.
Flatwise bearing coil spring maintains pressure even under abnormal ambient chan_es_
"Morning After" handle prevents record.
damage when cueing. Picksp head will not
move until entire weight is lifted.
Metal bearing race so const -ucted that bearing pin actually floats free.
Universal "One- hole" mount permits ins'ollotion in any turntable, with hairline height
adjustment.
Permanent stop prevents overswing to injure
connections. Internal loop prevents dam :ge.
For any lateral recording up to and including 16 inch, 78 or 331/3 rpm.
.
120M
Designed by VIN BROYLLS
NEEDLE TALK
From time to time we have noticed
references in the literature to a phenomenon
associated with record reproduction, popularly known as "needle talk." Some of the
statements which have been made do not
at all agree with some of our own findings,
and for what it is worth, we would like to
comment on the subject.
Sound radiation from a pickup- record
combination can originate either in the
pickup mechanism and /or arm, or from the
surface of the record itself. The radiation is
a function of the projected area of vibration
'walk" even on 30 cycle
Arm will not
and the amplitude and frequency with
modulated grooves.
which the radiating body vibrates. In pickPrecision made and individ.ally serial MAP-.
ups with a rigid stylus mounting and high
bered for constant reference.
lateral and vertical stiffness, considerable
Finish is brushed aluminum and black to
match any existing turntable.
radiation is likely to take place from both
Tests indicate that in conjunction with a so-the arm and the record surface because of
mal General Electric Cartr dge and a load
the lateral forces involved. This is accomequalizer, overall response will be excctlir
as the record is cut. The pick -up arm will
panied by a high rate of record wear and is
neither introduce nor ampi fy lateral distorresponsible for the association between
tion and has no self -vibra ion.
record wear and "needle talk." Pickups in
net less .artridge $34.54
PRICE
GENERAL ELECTRIC CARTRIDGE..5 4.77
this class can be made to have much less
515.00
PICKERING CARTRIDGE
sound radiation by decoupling the stylus
GENERAL ELECTRIC SPX -001
point from the large masses of the arm and
s 5.95
PRE AMPLIFIER
pickup mechanism. The familiar bent
GE Electronic Switch, Type YE -9. This instrument was designed for special elec- ic_I
shank needle accomplishes this to a large
refrequentstudies of wave -form, phase,
degree. It is obvious, of course, that this
lationship, etc. It will show the simultaneo.n
method of reducing "needle talk" also reobservation, for comparison, of two or more
independent signals on the screen of a catFduces high- frequency response.
$5.1.95
ode -tube oscilloscope.
Wit A good lateral pickup must have enough
Note: All prices are Net,
vertical compliance to be able to respond to
N.Y.C. and are subject to change
without notice.
the unwanted vertical modulation on lateral
records which reaches an appreciable velocity only at very high frequencies. The tiny
parts used to support the moving system of
LO. 3-1800
reproducers of this type can become fairly
Telephone:
efficient sound radiators at frequencies of
the order of 5,000 cycles and above. There
are two ways of reducing this radiated
sound: either to increase the vertical stiffL ,1
1
f`
I'
ness to reduce the amplitude of motion, or to
damp the spring suspension heavily in order
-13 West 43rd St., New York 18, N. Y.
to accomplish the same result. Either of
these means imposes additional wear on the
VARIABLE RELUCTANCE
GE
AND PICKERING
TECHNICAL PRODUCTS
453
INTERNATIONAL
West 47th Street, New York City 19
Featuring Outstanding Values in:
-a vaut,
This is it -the new 1947 Concord Catalog
in Radio and
complete selection of everything
items available for
Electronics- thousands of from
CHICAGO or
IMMEDIATE SHIPMENT
now available for the first
ATLANTA- hundreds latest
1947 prices. See new
Ume -featuring new,
RADIOS,
LOWER prices on RADIO SETS, PHONOPLAYERS,
RECORD CHANGERS, RECORD
SOUND
PORTABLES, AMPLIFIERS, COMPLETE
SYSTEMS, FESTERS. See latest listings of standard, dependable lines of radio parts and equip ra nsformers.relaps.etc.
ment- tubes,conden sers,t
Write for FREE COPY -NOW Address Dept. F -97.
!
TIME PAYMENTS: Write us for details of time payment plan
on Communications Receivers, Amplifiers, Test Equipment,
Radios, Phono- Radios.etc.
`I
\I'(Iiá
SEPTEMBER, 1947
R
A
SOI
C
D I O
7,
ILL.
N. Jackson Blvd.
O N
O R P O R A T I
1AfAYETTE RADIO
CHICAGO
CORPORATION
ATLANTA
255
3,
GA.
Psochtrso St.
47
record, and it appears to be unwise to sacrifice performance in order to quiet the
operation.
Since this sound radiation is an indication
that the pickup is performing its function of
A revolutionary development in amplifiers cleverly
designed to defy obsolescence and amazing in its
performance. New circuits, new materials and new
processes are actually combined in this one amplifier
to produce the most satisfying musical amplifier the
world has ever known. If you ore one of those discriminoting persons for whom anything less than the best is
a disappointment, you are one for whom the ACAI00DC was designed. Send for technical literature.-:
AMP
IER CORP. of AMERICA
398 -4 Broadway
New York 13, N.
Y.
"riding over" vertical modulation, it is not
necessarily a sign of undue record or stylus
wear. It may be noticed that in several
wide-range reproducers the surface noise
radiated directly from the pickup head is
considerably greater than that which is
transmitted through the electrical system.
This indicates that the reproducer is not
converting the vertical motion into electrical signal output, thereby giving an improved signal-to-noise ratio. At the same
time it is responding to lateral vibration of
the stylus at frequencies beyond 10,000
cycles per second.
Address Changes
Subscribers to
ENGINEERING
AUDIO
should
notify our Circulation Dept.
at least 3 weeks in advance
regarding any change in
address. The Post Office
Dept. does not forward
magazines sent to a wrong
address unless you pay
additional postage. We cannot duplicate copies of
AUDIO ENGINEERING
sent to your old address.
Old
and new address
'
MUST be given.
Circulation Dept.
Norman C. Pickering
AUDIO ENGINEERING
New Products
[
from page
381
Radio Magazines, Inc.
342 Madison Ave.
New York 17, N. Y.
FILM RECORDER
Back Copies
of "Radio" available
hand some
copies of the January
We have on
and
February -March
1947 issues of "RADIO"
which may be purchased
at 35c per copy. Please
include remittance with
order and send to:
Circulation Dept.
AUDIO ENGINEERING
Radio Magazines, Inc.
342 Madison Ave., New
York 17, N.Y.
AUDIO ENGINEERING SCHOOL
practical 9 months' course in Sound Fundamentals, Recording, and Sound Transmission
measurements; in a laboratory containing transmission sets, oscillators, square wave generator
and intermodulation analyzer, and other equipment.
Complete recording studios assimilating broadcast, motion picture and commercial sound recording.
This training available under "GI" Bill.
For information write
H. M. Treuraine, Director
SCHOOL OF SOUND RECORDING AND
TRANSMISSION ENGINEERING
1655 -A Cherokee
Hollywood 28, Calif.
A
f
PA"
LMGRAPH
ReCOrders
nerence
ECONOMICAL
UNINTERRUPTED
Longtime (up to I2 hoer,/ Conference
8 Telephone Recordings on Safety Film
Models for Dictation "TALKIES"
PERMANENT
INSTANTANEOUS
PLAY-RACK
MILES REPRODUCER CO.,wc. 812 BROADWAY,N.Y.3DeptAE
48
Filmgraph Model "11K" permanent recorder and instantaneous reproducer employs 2 reels 16 mrn. size 14" in diameter.
Capacity up to 5000' of M2 film on which
100 sound tracks are put across the width
of the film. Each sound track has continuity
up to 3 hours. Selection of recording for
playback is facilitated by track and zone
(footage) indicators. Machine starts recording automatically the instant the telephone
bell (when used with Miles Telemike Inductor) starts ringing or as soon as sound is
picked up by the microphone. Designed for
long time recording of any voice programs,
volume dictation, interviews, piped dictation, telephone recordings, hearings, etc.
A single machine can serve an entire organization by remote control.
For further data, write the Miles Reproducer Co., Inc., 812 Broadway, New York 3,
New York.
COMPLETE MAGNETIC PICKUPS
Full scale production of Amplifier Corp.
of America's recent development of the new
and highly popular General Electric DL
1R.M 6C Variable Reluctance Magnetic
Pickup, complete with tone arm and permanent jewel- tipped needle is now under way.
Now all owners of sound equipment may
acquire this cartridge complete in a pickup
unit, available in Studio and standard
models, for all sizes of records.
These arms offer the user (1) Less record
wear because of soft needle suspensions; (2)
Less distortion because of the soft spring
jewel mounting; (3) Less needle scratch
because of response only to lateral vibrations; (4) Less needle talk because of the
very light and small stylus assembly.
Additional features in the Studio Transcription Model (160GE) include a new type
of swivel base for raising or lowering to desired height; low frictional pivots for constant needle pressure; low vertical inertia
for ideal tracking, even on poor records;
easily maintained spring adjustable needle
pressure of 1 oz.
For literature address inquiries to Amplifier Corp. of America, 398 Broadway,
New York 13, N. Y.
ADVERTISING INDEX
Amplifier Corp. of America
Audio Engineering School
Audio Equipment Sales
Brook, Electronics, Inc.
Collins Audio Products Co.,
Inc.
48
48
39
39
41
Concord Radio Corp
47
Electro Motive Mfg. Co., Inc. _44
Electro- Voice, Inc.
6
Harvey Radio Co., Inc...
LeBel, C. J. ._ -- --
__47
36
-
Maguire Industries, Inc.
Miles Reproducer Co., Inc.
Pickering & Co.,
Inc.____
Pilot Marine Corp..
5
_
_48
45
-_ -_
_
46
Racon Electric Co., Inc.._
35
____
Radio Music Corp..
3
Simpson Electric Co.
Cover 4
Sylvania Electric Products Inc.
Cover 2
Tech Laboratories .... -_
41
Technical Products International
47
U. S. Recording Co
36
Vibration Systems, Inc.
Wagner -Nichols Sound
46
Recording
War Assets Administration
Wells, Winston
46
AUDIO ENGINEERING
SEPTEMBE
2
36
.,
1947
Join
An Outstanding Group!
Subscribe NOW to
AUDIO ENGINEERINGthe ONLY Journal for
Sound Engineers
Each issue covers 5 or
more of these subjects
1- Broadcasting
2-Transmitter and receiver
Our editorial policy has aroused tremendous
interest and acclaim among engineers prominent in the industry We are proud to announce
that co- operating with the editors of AUDIO
ENGINEERING are such outstanding authorities as:
.
manufacturing
3- Sound -on -film equipment
4- Recording (disc, wire and tape)
5- Public address
6- Industrial sound equipment
HOWARD A. CHINN
Chief audio engineer for the
Columbia Broadcasting System
JOHN D. COLVIN
Audio Facilities Engineer,
American Broadcasting Company
and applications
7- Acoustic
J. P. MAXFIELD
Authority on sound engineering,
Bell Telephone Labs.
GEORGE M. NIXON
Ass't Director of Technical
Development
National Broadcasting Company
who are members of the editorial
treatment of studios, rooms,
advisory board.
auditoriums, etc.
Because of the acute paper shortage we are compelled to limit the number
we
of subscriptions we can accept. If you are not already a subscriber,
on
this
form
subscription
the
NOW
mail
and
out
recommend that you fill
page. AUDIO ENGINEERING is R.váilable only to subscribers. Make
certain you get your copy every month.
AUDIO ENGINEERING
RADIO MAGAZINES, INC.
342 Madison Avenue, New York 17, N. Y.
Sirs:
Here
is
my
check (or
the next
money order) for
S.
Enter my subscription order to
AUDIO ENGINEERING for
issues.
Subscription Price: In U.
Please check one:
S.
A.
and Canada
New subscriber
-12
issues S3
-24
issues $5.
Foreign subscriptions are $4 annually.
Old "Radio" subscriber
Name (please print)
Address
Occupation (give title)
Firm Name
City
State
open
SIMPSON Model 260
. .
The world's finest high sensitivity set tester certainly
deserves the best in carrying cases. So we decided to give it
just that by building the tester into the case to make Al
integral unit of case and instrument. Here's -how we do it:
we take the standard Model 260, place it inside a housing
of heavily molded bakelite, and permanently fasten it
there. Instrument and case become one unit. Beneath the
instrument is a compartment for test leads. Over the face
of the instrument a roll top (of molded bakelite, too) slides
up to open, down to close, the case. With a flick of the
*The regular Model 260, without Roll Top Safety
`
Case, is always
3
j
Volt -Ohm Milliammeter
, eat.i Roll ?oft Sairetf eade
finger you roll it up and out of sight and the instrument
is ready to carry, and fully protected. With the Roll Top
Safety Case you cannot leave your carrying case behind.
It is never in the way. And you have constant, im;,ortant
protection to your 260 from damage, whether in use or not.
Just remember this fact, always: You cannot touch the
precision, the useful range, or the sensitivity of Simpson
Model 260 in any other instrument of equal price or in
some selling for substantially more.
available, of course,
Simpson 260, High Sensitivity Set Tester
for Television and
Radiyicing
At 20,000 Ohms per volt, this instrument is far more sensitive t
other instrument even approaching its price and quality. The pr
negligible current consumption assures remarkably accurate f
voltage readings. D.C. current readings as low as 1 microampere
y
ly
case
d up
to 10 amperes are available.
Resistance readings are equally dependable. Tests up to 20 megohm.
and as low as t/z ohm can be made. With this super sensitive instrument
you can measure a wide range of unusual conditions which cannot be
checked by ordinary servicing instruments.
Model 260-Size 51/4" x 7" x 31/4"
53895
Model 260, in Roll Top Safety Case -Size 5%43" x 9" a 43/4"
.
.
.
$43.75
Both complete with test leads
0
=g
2.5
Éo
É;-'
2.5 V.
10
100
V.
100
500
.e
2
2.5
10
50
10
50
250
1000
5000
250
1000
5000
10
50
V.
250 V.
1000 V.
5000 V.
0 -2000 (12 ohms centers
0-200,000 (1200 ohms center).
0 -20 megohms (120,000 center)
Amperes
D.C. (5 Decibel ranges:
-10
to +52DB)
10
SIMPSON ELECTRIC COMPANY
5200 -5218 West Kinzie Street, Chicago 44, Illinois
In Canada, Bach- Simpson Ltd., London, Ont.
o
I
N
S
T
R U
M
ASK
YOUR
JOBBER
E
N T S
T H A T
S
T A Y
A C C
U R
A T
E
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement