null null

The Transactions of the South African Institute
of Electrical Engineers.
Founded June, 1909 ; Incorporated December, 1909.
Vol. XXVI, December, 1935.
Editor: H. Denehy, M.I.E.E., A.Am.I.E.E.
Proceedings at Two Hundred and
Sixty-second Monthly General
Meeting
20th December, 1935, at 8 p.m.
Mr. Joseph White, M.G. (President),
was in the chair, and there were present
37 members and visitors, and the Secretary.
MINUTES.
The Minutes of the Monthly General
Meeting, held on the 28th November, 1935,
were confirmed.
Part 12
Mr. Joseph Aubrey Moore (Associate
Member).
These candidates will stand for election
at the Annual General Meeting in January.
The Council has admitted Messrs. John
Richard Gauntlett, Reginald William Ken
dall and Joseph John White as Associates,
and Mr. Gerald Graham Edwards as a
Student Member of the Institute.
An application for transfer was received
from Air. John Harry Rogers (Associate),
and the Council has graded him as an
Associate Member. Mr. Rogers will stand
for election at the Annual General Meeting
in January.
MEMBERSHIP.
Messrs. G. J. Privett and W. M. Winstanley were elected scrutineers of the
ballot for the election of new members, and
the following gentlemen were declared duly
elected:—•
Mr. Thomas Henry Hurst (Associate
Member).
Mr. Norman Dean Kennedy-Potts
(Associate Member).
Mr. Kenneth Byres Findlay (Mem
ber) .
(Transfer from Associate
Membership.)
For Full Membership: William Ainsworth.
For Associate Membership: William
Arthur Odgefs.
The Secretary announced that applica
tions for membership were received from
the following gentlemen, and the Council
had graded them as follows :•—
Mr. John Joseph Patrick Dolan
(Associate Member).
Mr. Alexander William Lineker
(Associate Member).
GENERAL BUSINESS.
Visit to Delta Sewage Disposal
Works.
The President: Under the heading of
" General Business," I would like to refer
to the visit paid by the Institute to the
Delta Sewage Disposal Works a couple of
weeks ago. We had a most entertaining time,
and I am sure all who were able to come
along will agree that the trip was well
worth the trouble. I would like officially
to express thanks to Dr. Hamlin and his
staff for the very excellent arrangements
made and to the City Council of Johan
nesburg for their abundant hospitality.
Associated Scientific anp Technical
Societies' New Year's Dance.
The President: I would also like to draw
attention to the New Year Dance under the
auspices of the Associated Scientific and
Technical Societies. I understand the book
ing is heavy, so those who wish to attend
had better get their names in early.
322
The Transactions of the S.A. Institute of Electrical Engineers.
DECLARATION OF ELECTION OF
PRESIDENT FOR 1936.
The President: My next duty, ladies and
gentlemen, is a very pleasant one; I think
it is an open secret, but I have to
announce offi cially to-night that your Presi
dent elect for 1936 is our senior YicePresident, Mr. C. T. Cocks. (Applause.)
I have had the privilege of serving on your
Council with Mr. Cocks for over 12 years,
and I can testify to the great interest that
he has always taken in the work of the
Institute. I am person all y very grateful
to Mr. Cocks for the self-sacrificing way in
which he has supported me during my year
of offi ce as your President, and I can safely
congratulate the Institute on securing the
services of Mr. Cocks as President for next
year.
Ml'. C. T. Cocks (President Elect) : Mr.
President, ladies and gentlemen, at this
stage all I wish to say is thank you for
the confidence you have placed in me. I
shall do my best to emulate the good work
that has been done by my predecessors,
those worthy gentlemen who have been in
the Presidential chair of this Institute. As
I shall be expected to speak at far greater
length at a later date, I will content myself
by thanking you again. (Applause.)
ANNUAL BALLOT FOR COUNCIL.
The President: I will now call for
nominations for scrutineers for the annual
ballot of Council.
On the proposal of Air. J. llussell, the
following gentlemen were declared elected
scrutineers for the annual ballot for Coun
cil:—Messrs. Jos. White, C T. Cocks, H.
Oenehy, E. V. Perrow, W. D. Wheeler and
Lt.-Col. J. Stewart Ross.
The President: I will now call on Air.
Borthwick to give us Air. Dorte's paper
on—
" SOME ELECTRICAL ENGINEERING
ASPECTS OF TALKING PICTURE
PRODUCTION."
Mr. R. J. Borthwick: Air. Chairman,
ladies and gentlemen, this is the paper
which Mr. Philip Dorte, recording engi
neer for the Gaumont (British) Picture
Corporation, has asked, me to read: —
[December, 1935.
SOME ELECTRICAL ENGINEERING
ASPECTS OF TALKING PICTURE
PRODUCTION.
By PHILIP H. DORTE, A.M.I.E.E., Recording
Engineer, Gaumont-British Picture Corporation, Ltd.
Read by R. J. BORTHWICK, Recording Engineer,
African Film Productions, Ltd.
Air. Chairman, Ladies and Gentlemen,
In expressing the honour which I feel on
being asked to present this paper to you
this evening, I must emphasise my very
genuine regret at not being able to be in
Johannesburg to present it personally, and
I must record my sincere appreciation of
the kindness of Air. Borthwick, of African
Film Productions, in reading the paper
for me.
In the short space of time at our dis
posal this evening we have so much ground
to cover that you will appreciate the
impossibility of describing in anything like
detail the many varied pieces of electrical
apparatus which are in use in a modern
film studio. Where, therefore, such
machines as motor alternators, motors,
transformers, etc., have to be mentioned,
I propose to merely indicate their horse
power or electrical rating, and leave the
rest to your imagination.
The modern film studio may cover a
ground area of anything from one to
thousands of acres, and may provide
facilities for the making of one to ten
pictures simultaneously through the medium
of one to twenty " floors " or " stages "
treated with acoustic materials for the dual
purpose of providing correct acoustical con
ditions for recording, as well as for ensuring
that no sound will penetrate into the studio
from outside. In the great majority of
studios these stages are erected at groundfloor level, so as to enable sets to be only
partially dismantled for transportation
from the carpenters' shops (where they are
made) to the stages themselves. Further,
if stages are built one upon the other,
difficulty is experienced in efficiently isolat
ing them acoustically. Each stage is
equipped with its own administrative
offi ces and dressing rooms, and common to
these stages are a number of editorial
rooms and review theatres, and, of course,
workshops, carpenters' shops, plasterers'
shops and film laboratories.
December, 1935.]
Tlie Transactions of the S.A. Institute of Electrical Engineers.
In considering, then, the electrical
equipment necessary to operate a film
studio, one must appreciate the necessity
of a power, station capable of supplying
•sundry workshops, as well as the heavy
loads of the studios themselves, with which
we shall deal in a minute, and when it is
realised that a large set may be lit by a
combination of lamps drawing in all
some ten or twelve thousand amperes at
110 volts d.c, it will be seen that a studio
sub-station is no mere electrical toy.
Some studios generate their own power
by means of five or six hundred horse
power heavy oil engines directly coupled
to d.c. generators, but the usual practice is
to receive bulk supply (in England gener
ally 50 cycle 3-phase at 6,600' volts) and
transform it down to 230 volts to drive
rotary converters giving a 110; volt d.c.
output. This supplies only the necessary
power for lighting, and by lighting I mean,
of course, the lights used on the set for
photographic purposes, house and general
lighting being 220 volts a.c. In a sub
station, in addition, there will be numerous
smaller machines providing sundry voltages,
both a.c. and d.c. ; as an example, a
230 volt, 24 cycle supply is usually adopted
for bi-pole machines for interlocking sound
and " picture cameras and/or projectors,
while, as a further example, large repulsion
starting induction motors operating off,
say, 415 volts may be used to drive large
fans for ventilation purposes.
T have already mentioned that the main
lighting supply is nearly always 110 volts
d.c, and this supply is carried to the
stages on heavy busbars and thence to
various distribution panels complete with
breakers, which are usually situated in the
gantry or otherwise near the roof of the
stage. From there it is led by means of
heavy flexible cable to nests of six sockets
scattered about on the spot-rails and on
the floor itself, and into which the various
lamps are plugged direct. Alternative
arrangements are met with in different
studios, and I am personally acquainted
with several studios where a 3-wire system
is operated (which has the disadvantage of
needing continual supervision for balancing
the load), and in another studio which I
know, all the incandescent lights are oper
ated from a.c. ; the two main disadvantages
of this are that the arc lamps cannot be
plugged into any available socket, and
323
that the a.c. fields from the cables fre
quently interfere with the sound recording
apparatus. In the days of silent pictures,
arc lamps were used for studio lighting to
the almost entire exclusion of incandescent
lamps, but when sound was introduced, it
was found that the arc noise interfered con
siderably with recording. Consequently,
incandescent lighting was resorted to in the
main, and even though by means of chokes
and modern carbons, arc noise has been
reduced to a minimum, they are used
mainly for such effects as daylight stream
ing through a window and for general
lighting on night shots when the blue con
tent of the light is very valuable. The
modern studio arc lamp varies in sizes
suitable to accommodate 300 to 1,000 milli
metre reflectors, and may draw as much as
300 amperes, thus dissipating nearly 30 kilo
watts. The incandescent lights—by which
1mean the incandescent bulb complete
with lamp housing—may be divided into
two types, known as spots and floods, and
ranging in power from 200 watts to 10 kilo
watts. Spot lights are of the projection
type filament, and are generally of either
2, 3, 5 or 101 kilowatts rating. The next five
slides show you, respectively, a 5 kilowatt
projector-type bulb, a 3 kilowatt lamp
housing and mounting, a " baby " spot
with 8 in. condenser, a typical spot-rail
array of lamps (this photograph was taken
during the filming of " Britannia of
Billingsgate ") and a general view while
shooting a scene for the film " I was a
Spy."
Flood lights, on the other hand, have
ordinary spiral filaments and rarely exceed
2kilowatts, their lamp housings varying
from ordinary tin cans to elaborate rifletype floods. The next slide shows you a
bank of overhead flood lights such as many
cameramen use as the first step in lighting
a set. Note particularly the wire netting
to catch bulbs (or pieces of bulbs) in the
Lamp
event of breakage or explosion.
housings are so constructed that the lamp
can be partially or entirely screened by
diffusers consisting generally of one or two
thicknesses of gelatine, silk or cheesecloth.
Alternatively, glass, either frosted, tinted
or clear, may be used ; the latter, incident
ally, being frequently employed in front of
an arc lamp in order to reduce hum and arcbubble noise. Other contrivances, such as
snoots and gobos, are used to cut off
different quantities of beam, and spill light
324
The Transactions of the S.A. Institute of Electrical Engineers.
eliminators are frequently used on 3 and
5 kilowatt lamps.
The next four slides demonstrate a simple
set with one artist lit in four different
manners. The first shows lighting with
front lights only; the second, three-quarter
back lights only; the third, front and back
lights, but with the back lighting too strong
or insufficiently diffused, and the last one
shows us all the lights properly balanced.
While it is impossible to lay down rules
for lighting even small and simple sets, one
would guess that the room in the last
picture was lit by half a dozen 2 to 3 kilo
watt spot lamps from a height of 12 feet,
that the artist was picked out with another
2 kilowatt spot, while a diffused baby spot
of some 1,500 watts was trained on her face.
In addition, a tin can or rifle flood was
probably placed at approximately camera
level each side of the camera.
Lighting a particularly complicated set
is a job which calls entirely for experience,
and is probably the only job in existence
where a non-technical man, in the electrical
sense, has control of thousands of amperes;
this lighting expert (or chief cameraman)
must have something of the artist in his
composition, for on the disposition of his
[December, 1935.
the film. The two negatives are ultimately
printed, on to the one positive for projection
in the cinema (Fig. 2).
Fig. 1.
The picture negative is of a very high
speed panchromatic film stock sensitive to
the whole of the spectrum range, except
the extreme ends, while recording film
stock is of a very contrasting nature
(usually orthochromatic positive), having
an extremely fine grain in order to reduce
halation to a minimum.
lights depend entirely the high lights and
shadows which make a set look on the
screen a black and white counterpart of
either a wish-wash water colour or the work
of an old master.
In the silent days the picture camera was
a noisy contraption, using virtually the
whole width of the film for its picture,
which it shot at the rate of 16 per second.
When sound was introduced, space had to
be made on the edge of the film for the
sound track, the camera had to be silenced
and the speed of the film through the
camera had to be increased in order to make
it possible to record and reproduce the high
frequencies which constitute the overtones
and harmonics of speech and music. The
speed ultimately fixed upon was 24 pictures
per second.
The next two slides show the dimensions
of standard 35 millimetre film and illustrate
where picture and sound are accommo
dated (Fig. 1). The sound tracks are of
the variable density type, which we shall
discuss later. The picture negative is shot
with a mask over the portion of the film
devoted to the sound track, while the sound
film is shot exposing only this portion of
Fig. 2.
Here is a slide showing a picture camera
complete with flexible drive from a
synchronous motor, which is usually rated
at 220 volts, 48 or 50 cycles, 3-phase, which
drives the camera at the correct speed for
synchronising with the sound camera driven
by a motor of similar design.
When talking pictures were first intro
duced, the camera was silenced by placing
it, complete with operators, in a small
booth with an optical glass front, but apart
December, 1935.]
The Transactions of the 8.A. Institute of Electrical Engineers. !
from unwieldiness, there was so much
agitation for reform from camera operators
that various forms of sound-proof covers
were applied to fit closely round the camera
itself. These are termed blimps. The next
slide depicts a scene being shot in the early
days of the camera booth.
Many of you, on going to the pictures
(which I hope you sometimes do), may
have wondered at the ease with which
scenes are apparently shot in different parts
of the world. In practice, it is very seldom
that these scenes are shot in the place
which you view on the screen. The answer
is a process known as back-projection,
whereby the background only is actually
photographed at the place in question, and
the film is then brought back to the studio
and is projected on to a transparent screen,
graduated, as a matter of interest, from the
centre outwards, in order to eliminate a hot
spot from the projector arc in the middle
of the screen; the modern back-projection
screen is actually made of some secret
acetate solution, which is sprayed on to a
piece of flat glass and, when hardened,
peeled off. The artists are placed in front
of the screen; that is, on the other side of
the screen to the projector, and the camera
is placed in front of them. The camera
and projector are electrically interlocked,
so that the respective shutters are open at
the same time, and the developed negative
will then show the artists gazing at a moon
lit sea or looking out of a taxi window in
London, Paris, Johannesburg or New York;
any other situation which a director may
desire can be similarly created, dependent
only on the visit of a cameraman to the
location of the desired background. In more
simple cases, where the background is, say,
a house or a range of hills with no move
ment called for, a photographic backing,
being merely an enlarged still photograph,
may replace the complicated back-projection
paraphernalia. In these days the size of a
good backing reaches gigantic proportions,
very good detail still being maintained on
an enlargement 100 ft. by 60 ft.
The mechanical construction of a picture
camera does not strictly concern a body of
electrical engineers, as the only electrical
part of it is the motor driving it. Suffice
to say that mechanical construction calls
for a given length of film to be held abso
lutely rigid at a distance from the lens
which must be accurate to a ten-thousandth
of an inch for l/48th of a second, and then
325
moved on during the next l/48th of a
second to allow the next unexposed frame
to come into position. The lenses are
usually mounted in turret fashion, so as to
ensure the lens being changed without loss
of time. The modern camera is usually
equipped with a 25 or 28 millimetre lens
for long shots, 35 or 40 millimetre for
medium shots, 50 millimetre for close-ups
and 75 millimetre for what are termed big
head close-ups when the head of an artist
will fill the whole picture. Additional
equipment will include 100 millimetre lens
for very long focus shots in the open, and
this may even be used for shooting a closeup in the studio from the back of the set.
News-reel cameras, which may be called
upon to shoot, say, a close-up of a test
match batsman at the wicket from the roof
of the grandstand, may have a lens of even
48 inches focal length. A comparatively
new innovation is the Zoom lens—a lens
with variable focal length, but as the
aperture alters according to focal length
change, its use is very limited, since its
exposure will alter too.
Camera equipment also includes a series
of gauzes and filters—pieces of optical glass
ranging in hue from light yellow to dark
orange, and from green to red. The
object of this, of course, is to filter out the
different colours of the spectrum, and con
sequently to emphasise or to delete various
coloured objects in the picture. The
gauzes consist of open-work net, giving a
softening effect to the subject being photo
graphed, and these are normally used only
on close-up pictures.
In modern film technique, the director
wants to be continually moving his camera
about during the course of a shot; for this
reason, cameras, complete with blimps,
are mounted on small wagons running on
pneumatic-tyred wheels, and can then be
pushed about without undue vibration.
Cameras are often mounted on cranes, a
device (often worked pneumatically) which
makes it possible to raise and lower the
camera, keeping it at the same time on a
constant plane^
Now we come to what you, as electrical
engineers, will probably find the most
interesting department of motion pictures
production; I refer to sound recording.
The first talking picture shown in the
United States of America in 1926 " talked "
by the simple means of synchronising
standard picture film with a disc recording,
326
Tlie Transactions of the S.A. Institute of Electrical Engineers.
and the only difference between these discs
and standard gramophone records was that
the former were of 16 in. diameter and
revolved at 33J revolutions per minute
instead of 78. It was soon apparent, how
ever, that disc recording, as applied to
motion pictures, had two severe drawbacks,
quite apart from any question of sound
quality.
They were: (1) Editorial difficulties, in
that it was diffi cult to cut the picture and
duplicate the requisite takes of the original
wax on to a new master wax and keep the
whole in synchronisation ; and (2) the
constant trouble of the picture film break
ing in the theatre, one or two frames being
lost in splicing, resulting in the latter part
of the reel in question being out of syn
chronisation with the sound record.
Thus sound on film came into its own.
This method of sound recording relies for
its reproduction on the properties of the
photo-electric cell, which will generate
varying voltages in accordance with the
change in the amount of light falling on its
electrodes. If, therefore, we can, by some
process, take the output of a microphone,
amplify it and somehow translate its
electrical impulses into some sort of photo
graphic wave-form on a continuously
moving strip of film, which is subsequently
passed at the same speed in front of a
photo-electric cell, we should receive a vary
ing voltage output from the photo-cell
following fairly accurately the original out
put of the microphone, and which voltage,
when amplified and connected to a loud
speaker, should faithfully reproduce the
original sounds applied to the microphone.
After considerable argument, the picture
people agreed to surrender to the recording
engineers a tiny portion of the space on the
film then devoted to their picture; in fact,
a strip 100 mils, wide between the righthand side of the picture and the sprocket
holes. Two alternative methods of record
ing now- presented themselves. The first
was to retain the complete 100 mils, width
and vary the exposure in accordance with
the electrical input from the microphone,
and the second was to maintain a constant
exposure and vary the area of the film
exposed to the light up to a maximum of
the 100 mils, available. These two methods
became known as variable density and
variable area recordings, and were cham
pioned by Western Electric and B.C.A.
[December, 1935.
Photophone respectively, and although the
two types of sound track vary considerably
in character, they both rely, as indeed any
light modulating system must do, on very
narrow slits focussed by optical means on to
moving film; in practice, the actual slit is
100 mils, long, and is on average approxi
mately 5 mils, in width, giving, through
optical reduction, a line on the film
reduced some seven times, resulting in
something between 0-5 and 1 mil. In
variable area, recording the slit image is
modulated by varying the length of the
illuminated portion, thus producing a strip
of constant density and varying area, which
to the eye will appear as an oscillogram,
while in the case of variable density
recording the slit image is constantly
illuminated along its whole length, the
exposure being varied from point to point
along the sound track; the transmission
thus varies in linear relationship to the
pressure or velocity of the sound wave to
the microphone, dependent on whether
pressure (carbon or condenser) or velocity
(ribbon or dynamic) type microphones be
used. Theoretically, of course, the slit
image should be of infinitesimal width, but
in practice this is obviously impossible, and
so the slit is kept as narrow as is practic
able in order to reduce fundamental attenua
tion consequent on the width of the slit
image being a measurable fraction of the
wave length of the recorded frequencies.
In variable area recording, the width of this
slit image also introduces harmonic distor
tion, making it all the more necessary to
keep its width as small as human ingenuity
can attain.
In the case of variable area recording,
modulation is effected by means of an
oscillograph similar in design to those used
in electrical laboratories for making sundry
electrical measurements, but so designed
that it is sensitive to an audio-frequency
range of some 40 to 10,000 cycles.
In the case of variable density recording,
where the image illumination is variable,
two systems are in general use ; the first
makes use of a glow-lamp, which has the
property of varying the brilliance of its
emitted light in accordance with the a.c.
potentials applied to its electrodes, the
resultant varying light being focussed via
the slit on to the moving film. The second
method is patented by the Western Electric
engineers, and utilises a light valve which
employs two ribbons vibrating at a speed
December, 1935.]
The Transactions of the 8.A. Institute of Electrical Engineers.
corresponding to the frequency that is to
be photographed, and through the gap
between these ribbons a light beam passes
to the film via the usual optical system.
To treat the systems described above in
somewhat greater detail. Fig. 3 shows the
schematic layout of the variable area
system sponsored by the R.C.A. Photo-
Fig. 3.
327
these fundamental differences: The mask,
instead of being semicircular, is a triangle
in shape, while the galvanometer mirror
turns about a horizontal instead of a
vertical axis. This triangle of light is so
imaged on the slit that the apex and base
are equidistant from the slit (Fig. 5c).
Under these conditions the slit is illumin
ated about half its area. Modulation
causes the triangle of light to vary up and
down, thus illuminating' the slit between
maximum (base) and minimum (apex)
according to current values applied to the
galvanometer. It produces an image trace
as in Fig. 5d.
I
Fig. 5.
Fio. 4.
phone. A source of light of constant
intensity is projected through a lens system
on to a mask, which consists of a metal
surface in which a " half-moon " shaped
opening has been cut. This semicircle of
light is projected on to the mirror of the
oscillograph galvo and reflected through a
suitable lens combination on to a mechani
cal slit. This slit is in turn imaged, by
means of a microscope objective, on to the
film. Under conditions of no modulation,
half the slit is illuminated, as in Fig. 5a,
by the semicircle of light. When modula
tion current is applied to the galvanometer,
the mirror rotates accordingly about its
vertical axis, causing the semicircle of light
to move to and fro in a horizontal plane.
Thus at any instant of time the length of
slit illuminated may vary from zero to
maximum. As the film is moving past the
recording point in a vertical direction, this
produces an image trace as Fig. 5b.
Another variable area system is that of
British Acoustic, shown in Fig. 4. Similar
in general principles to the previous, it has
Fig. 5e.
Fig. 5e shows examples of these two
systems (negative tracks).
To pass on to the variable density
systems of recording. The Western
Electric " light valve " method is shown
diagramatically in Fig. 6. As stated above,
the light valve consists of two metallic
ribbons stretched parallel to each other and
separated by a gap of 1 mil. Viewed
against the light, this gap appears as a slit
1 mil. wide by 256 mils. long. Placed in a
powerful electro-magnetic field, this light
valve acts as an electrically operated
shutter, the current alternations traversing
the loop formed by the two ribbons,
causing the gap to narrow or widen its
(1 mil.) width in sympathy. As the
maximum opening is 1 mil. (after optical
328
The Transactions of the 8.A. Institute of Electrical Engineers.
reduction of 2), it not only acts as a
shutter to vary the amount of light reach
ing the film, but acts to form the slit image
that is necessary in all present-day sound
on film systems.
OBJECTIVE
|
LECTRICflUy
OPER^TED SHUTIER
1)
OOI inch
Fig 6.
The second example of variable density,
shown in Fig. 7, is the glowlamp system.
Here a variable intensity light source is
imaged on to a fixed mechanical slit, which
in turn is imaged on to the film plane. The
varying intensities of light produced by
variation of modulation potentials across the
glowlamp terminals are thus translated into
lines of varying density on the film.
V^RIABLE ,
IKKHSITY C~
LIGHT
SOURCE
f
LENS
JC
J>
FILM
'SLIT
Fig. 7.
The next slide will show you a typical
Western Electric light valve yoke (Fig. 10),
while the succeeding slide shows you a
complete Western Electric recordingcamera. Note the film magazine and the
fly-wheel which ensures the film passing
at a really constant speed over the record
ing drum. This is one of the chief responsi
bilities of the recording engineer, to see
that the film is propelled past the record
ing point at an absolutely constant speed.
As the film has no great inherent mass like
a gramophone disc and turntable, in the
early days of sound films, this presented
a surprising amount of difficulty. There
are three main mechanical systems
employed in this connection: Sprocket
recorders, slit block recorders and drum
recorders. All are similar, in that they
have a feed sprocket engaging the perfora
tions in the film driven through a chain of
gears, and the recording head is isolated
mechanically from this, firstly, by two
loops of film, and, secondly, in that it is
itself driven through some system of
mechanical filter, or in the case of the drum
recorder, an electro-mechanical filter con
sisting of a type of eddy current disc
[December, 1935.
actuated by a revolving electro-magnet
driven by the feed sprocket. This is so
arranged that just sufficient torque is
applied to the drum to overcome friction
losses, and so that the drum throws the
necessary film loops on either side.
It will be appreciated that, with any film
recording system, a period of silence, such
as is met with between sentences, between
words and even bfetween syllables, will
result in the track corresponding to those
periods being 50 per cent, unexposed (that
is, permitting 50 per cent, transmission),
resulting in half the developed negative
being, at those periods (in the case of
variable area tracks), quite transparent.
The result of this will be that, after two or
three projections of the positive, dust and
dirt will have collected on the transparent
portion of this track, resulting in a series
of clicks and hisses being emitted by the
loud speaker; these very unpleasant sounds
are known collectively as " ground noise."
To overcome this, apparatus for ground
noise reduction has been developed which,
in the case of variable area recording,
automatically reduces the exposed part of
the track to a minimum width when no
modulation occurs, and which entirely
blacks out the positive track in the case of
variable density recording.
Fig. 8.
The operation of this ground noise reduc
tion is more fully set out in Fig. 8. Here
we have the layout of a typical recording
system (variable area), consisting of micro
phone, amplifier and galvanometer. A per
centage of the output modulation a.c,
however, is tapped off and rectified, and
this rectified a.c. or pulsating d.c. is led
through a low pass filter, which passes a
maximum of about 20 to 25 cycles. The
output of this is, therefore, a varying d.c.
current, which approximates the envelope
of modulation current. It contains no
audible frequency variation, but varies in
ratio with the average value of the modula
tion a.c.
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
329
In the E.C.A. Photophone system, this
pulsating d.c. is made to actuate a
magnetically operated shutter (Fig. 9a),
which cuts off some of the illuminated
portion of the track (on low modulation),
producing a sound track similar to Fig. 9b.
Again, in Fig. 9c, we have the method
of ground noise reduction used in the case
of British Acoustic. Here the triangle of
light, in the non-modulation condition, is
set towards its apex, instead of, as before,
half way. The d.c. component is used to
bias it towards the half-way position, in
accordance with the average value of
modulation current, resulting in a track
like Fig. 9d. In both these instances the
amount of transparent track surface has
been cut down to a minimum. It should
be noted that, whereas' Figs. 5b and 5d
Fig. 10.
Fig. 11 shows examples of variable
density and variable area positive film.
I am not, at this point, going- to be
drawn into the old argument as to which of
the two systems (density or area) produces
the better sound; I will merely say that
variable density sound tracks are the more
difficult to photographically process (with
variable area, one merely has to deal with
Fig. 11.
(It is of interest that the variable density film
shown here is a test film from the Gaumont
British production " Rhodes " for the filming of
African scenes, for which the African Film Produc
tions' sound equipment was used.)
are representations of negative sound track,
Figs. 9b and 9d are positive. In variable
area, the difference between positive and
negative is, of course, one of the position
of the two densities. The blacks in the
negative are whites in the positive, and
vice versa.
black and white), while variable area
recording systems have the disadvantage
that it is not safe to let the track occupy
the whole of the 100 mils, allotted on the
film for sound, as many projector slits are
not lined up correctly, so that a hundred
mil. track runs the risk of having its peaks
330
The Transactions of the S.A. Institute of Electrical Engineers.
eut off in the projector, thus causing very
bad distortion. In consequence, a maxi
mum " full track " of 70 to 80 mils, is all
that is considered safe with variable area
systems, causing, in comparison with
variable density, a maximum volume drop
of from 2 to 4 decibells.
Considering for a moment how the sound
impulses to be recorded actually reach the
recording camera, I want you to look at the
next slide, which shows you a typical
microphone boom; on this boom is mounted
one or more microphones, and its mechani
cal construction enables the boom operators
to move the microphone about the set in
both vertical and horizontal planes in order
to keep it at a constant distance from a
moving artist, and at the same time out of
view of the picture camera.
[December, 1935.
installed a monitor equipment enabling the
recording engineer to hoth see and hear
what he is doing to the volume of the
sound which he is controlling.
In some studios this booth is situated on,
and is movable about, the floor of the
studio, while in others it is built perman
ently into one of the walls, thus enabling
the engineer to look down upon the whole
set; such an installation is shown in the
next slide (Fig. 13).
From the mixing booth the impulses are
carried to an amplifier room and thence to
a recording camera such as you have
already seen (Fig. 14).
Fiq. 13.
Pio. 12.
The next slide shows you a typical
microphone position relative to the picture
camera position for simultaneously photo
graphing and recording a stationary artist,
the area inside the rectangle only being
photographed. (Fig. 12.)
The output of the microphones are led by
cables to the mixing booth, where they are
connected across constant impedance
volume controls, and where there is
Sound., film reproducers consist of a
picture projector (immediately underneath
which is mounted the photo-cell, exciterlamp and optical-slit assembly) and high
power amplifiers and loud speakers. Due
to the fact that the film passes through this
sound-head after it has passed through the
picture head, the sound corresponding to
a given frame of picture is not printed
immediately opposite to it, but 14^ inches
ahead of it.
The next four slides depict a typical
sound and picture projector, a typical
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
theatre exponential-horn loud speaker and
a diagram of a typical auditorium
installation.
Some of you may have wondered how it
was that speech was obviously super
imposed later over music in films you have
heard. Part of the equipment of every
film studio comprises what is known as a
re-recording or dubbing channel. This
channel consists of a standard sound record
ing camera, amplifier and mixer, but
instead of the mixer being fed from micro
phones, it is connected to a series of what
may well be termed " film phonographs,"
which are virtually sound film projectors
331
this paper to delve also into the intricacies
of television.
Suffice to say that until very recently it
was not possible, using strictly television
methods, to televise really, satisfactorily
more than the head and shoulders of an
artist, or to reproduce on a screen of more
than a few inches square, but recent
demonstrations by the Baird Television
Company have shown amazing improve
ments in this respect; quite large scenes
have been satisfactorily televised on to a
very reasonably sized screen; and when
you consider the fact that to obtain picture
definition approaching the standard of,
Fig. 14.
minus picture head. These machines are
electrically interlocked, so that various
sound tracks may be reproduced from them
in strict synchronism, and their outputs
mixed on to a new negative in an orthodox
manner. Alternatively, one or more sound
tracks may be mixed with the output of
one or more microphones, and it is interest
ing to note that while, in modern motion
picture production, all dialogue is recorded
simultaneously with the shooting of the
picture, all music and sound effects are
recorded on to the dialogue tracks at a
later date.
When your Secretary asked me to write
this paper, he also mentioned that you
would like to know something about tele
vision. Unfortunately, we have devoted so
much time in dealing with motion picture
production that it will not be possible in
say, 9-5 millimetre film, it is necessary to
work at a scanning speed of at least 2 to
300 lines per second, you will appreciate
some of the problems with which television
engineers are faced.
Beception is effected by viewing the
picture on the screen of a large cathode-ray
tube, this tube being operated from
amplifiers having a frequency response of
the order of one million cycles—a frequency
range which has to be maintained through
out the whole transmitter, receiver and
connecting link. You will, therefore, see
that transmission of high definition tele
vision entails the use of connecting lines of
an electrical standard far above practical
costs (I believe it works out at something
like 4,000 per mile), and that a radio link
necessitates the employment of a radio
frequency corresponding to something of the
332
The 'Transactions of the S.A. Institute of Electrical Engineers.
order of 5 metres in order to accommodate
the terrific sideband spread without inter
ference.
By various means, however, these latter
difficulties c^n, and are, being overcome,
and the limits attached to the size of the
subject being televised and the dimensions
of the picture at the receiving end can, and
are, being met by the use of intermediate
film. Tiie scene to be televised is shot
(picture and sound) on film which is pro
cessed in the remarkable short space of
time of 35 seconds, and this developed film
is then presented to the scanning apparatus.
Actually a camparatively short loop of film
is used for this, as, after leaving the
scanning gear, the film travels via a
de-emulsifying bath into a re-emulsifying
bath, and is then introduced again to the
picture camera. At the receiving end a
similar process is adopted, the cathode-ray
tube being focussed on a continually moving
band of film which photographs the sound
and picture images, and after 35 seconds
development passes through a standard
picture projector (whence it may be viewed
on a normal screen) and then on through
de-emulsifying and re-emulsifying baths as
in the case of the transmitter.
On behalf of Mr. Dorte, I must express
my thanks to Western Electric, Ltd., and
the General Electric Co., Ltd., for the loan
of slides, and to the African Film Produc
tions, Ltd., and Mr. Albrecht, general
manager of that firm, for the loan of equip
ment, particularly the complete sound
camera equipment which I have on hand
here, and which I am now going to demon
strate. It is a variable density glowlamp
recording portable equipment used for newsreel work and is different from the systems
previously described, in that both sound
and picture are recorded on the film in the
same mechanism. This is a compromise in
the interests of portability; it is not an
ideal method.
Thank you, ladies and gentlemen.
A demonstration of glowlamp sound
recording equipment was given at the con
clusion of the paper.
A number of figures were shown on the
screen which have not been reproduced.
The President: Ladies and gentlemen,
I think you will agree with me that we
have listened to a most fascinating paper.
It was indeed good of Mr. Dorte to prepare
this paper for us, especially as I under
[December, 1935.
stand, after he had finished the '' Rhodes "
film he went Home and took up another
hobby—in other words, got married and
only had at his disposal a very short time
in which to prepare his paper. Our thanks
are also due to Mr. Borthwick for the veryable way in which he has read the paper
and conducted the demonstrations. We
have also to acknowledge our indebtedness
to Mr. Albrecht, -the general manager of
African Films Productions, and to African
Films Productions for affording facilities
to Mr. Borthwick to give the paper.
(Applause.)
Mr. R. H. Gould (Past President) : I
fear it is a bit late to-night, there is a verylong agenda, and I think I would be doing
better service to the Institute if I were to
not speak oij the subject now.
The President: I will now call on Mr.
Winstanley to give us his reply to the dis
cussion on his paper.
MERCURY ARC RECTIFIERS.
By W. M. WINSTANLEY (Member).
(Journal, April, 1934.)
Reply to Discussion.
Mr. President and gentlemen:
In presenting this paper, my intention
was that it should serve as an introductory
paper setting out the elementary principles
governing the functioning of the mercury
arc rectifier, together with a description of
the apparatus. It is to be hoped that the
opportunities presented by the installation
of rectifier sub-stations for railway traction
service in this country will lead to more
practical papers being presented to this
Institute dealing with actual operating
experience.
I should like to thank those members who
have contributed to a very interesting
discussion.
In reply to the questions raised: Mr.
White raises the question of the space
charge. The formation of a space charge
in the high vacuum thermonic valve is due
to the fact that electrons only are present
in the space between the anode and
cathode. In the gas or mercury vapourfilled rectifier, as I briefly mentioned in the
paper, positive ions are formed by bom
bardment of the gas or vapour by rapidly
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
moving electrons, and any space charge
which may tend to form is immediately
neutralised except in the immediate vicinity
of the cathode, the operating anodes and
the control grids on a grid controlled unit.
An exception to this is the case of a cold
rectifier called upon to pass a very heavy
overload current. Under these conditions
all the vapour in the anode stream may be
ionized without providing suffi cient positive
charges completely to neutralise the elec
trons present, resulting in a low potential
space charge.
The existence of this space charge is im
mediately made evident by an increase
in the rectifier arc drop, which may have
twice the normal value, the arc itself
becoming unstable, resulting in severe
surges on the transformer windings.
The remedy is to design the rectifier so
that even when cold there will be sufficient
vapour present to carry the full rated over
load, or alternatively, by the provision of
special devices to ensure the rapid warm
ing up of the unit.
Mr. White also asks whether harmonics
on the D.O. and A.C. system can be
smoothed out by tuning the grid circuit.
The control grids of a mercury arc recti
fier are used to prevent the striking of the
arc to its associated anode until the
required instant. After the arc is estab
lished, a space charge is built up around
the grid, which renders it. incapable of
further influencing the anode current. If,
therefore, a particular wave form is im
pressed on the grid it will not be reproduced
in the anode current, as is the case with
the thermionic valve, so that such a method
cannot be employed for the elimination of
ripples and harmonics in the A.C. and
D.C. mains.
Control of Electronic Flotv by Action of a
Magnetic Field.
In general, the application of a magnetic
field to an electronic flow results in a
change of direction of flow rather than a
change in magnitude of flow. As far as I
am aware, all experimental attempts to
control the value of an electronic current
have been confined to small laboratory
models, and the results obtained suggest
that the production of large commercial
units operating on this basis would not
prove practicable.
333
Temperatures.
The operating temperature of the main
cylinder and top plate is generally 40-60
C, and the discharge water temperature
50-70 C. For all designs there is a par
ticular temperature at which the arc drop
is a minimum, and if other considerations,
such as liability to back-fire, permit, the
unit should be run at this temperature.
Poiver Factor.
I would refer Mr. White to the paper by
H. Rissik, " The Influence of MercuryArc Rectifiers upon the Power Factor of
the Supply System," I.E.E. Journal, Vol.
72, page 435.
The deviation of the power factor from
unity is due to:. ....
(1)The magnetising current taken by
the transformer and auxiliaries.
(2)The distorted or non-sinusoidal cur
rent drawn from the supply system.
Both these factors produce a lagging
power factor, and the only way in which
a leading power factor could be obtained
would be to arrange for the anode current
to be transferred from one anode to the
next, before the natural instant of com
mutation had arrived, by means of grids
which could control the stopping of the arc
as well as its initiation. As I have already
explained, this is impossible with a normal
control grid. I understand one manufac
turer claims to have produced an experi
mental model in which this can be done,
but no details are at present available for
publication. This, however, is a possible
future development.
Parallel Operation.
Rectifiers possess an inherent shunt
characteristic, so that they will always
operate satisfactorily in parallel with them
selves or with plant having a similar charac
teristic. If parallel operation with plant
having an over compound characteristic is
required, the rectifiers must be provided
with grid control, induction regulators, or
on-load tap changers. The only special
precaution usually taken is to provide each
rectifier with a reverse current high speed
circuit breaker, which operates in case of
back-fire, thereby minimising the resulting
shock to the A.C. and D.C. systems.
Rating of Transformers.
The two transformer connections gener
ally employed are those shown in Fig. 14
334
The Transactions of the S.A. Institute of Electrical Engineers.
of the paper. Such transformers are rated
where P = rectifier output in kW + losses in
plant.
[ Primary ... 1-05 P
Fork connection ' Secondary ... ,1-79 P
{Average
... 1-42 P
Interphase^Primary
... 1^05 P
transformer|Secondary ... 1-48 P
connection(Average
... 1-26 P
[December, 1935.
It will be seen that the A.C. three-phase
is converted to D.C. and the D.C. con
verted to A.C. in the inverter, the fre
quency of the single-phase depending upon
the speed of the contact maker which
applies voltage to the grids at the appro
priate instants.
The equivalent size of the interphase
transformer at 50 cycles is 8-5 per cent, of
the main transformer for the second case
cited.
The figures, of course, remain the same
if the primary is delta-connected in place
of star.
Pre-heating of Rectifiers.
The modern rectifier will take a load up
to full rated overload when cold without
trouble being experienced, but under this
condition the arc drop is higher than nor
mal, so that complete load sharing in
parallel with units already on load is not
secured until after a few minutes running.
This applies to both new and old units.
Probably the suggestion that only new units
require pre-heating arises from the fact
that new units, if they are left standing
idle for long periods after manufacture, will
in all probability require " forming "
before taking load for the first time, as
explained in the paper.
Fig. 20.
Although conversion from three-phase to
single-phase is shown, three-phase to threephase at a different frequency can be
achieved by the same method.
In method (b), which is only suitable for
three-phase to single-phase conversion, the
arrangement is as shown in Fig. 21, which
shows two rectifiers each fitted with grid
control.
Back-firing.
Mr. White suggests that ultra-violet
radiation from the main arc on to an anode
during its idle period may cause a back-fire.
With the steel tank rectifier such radiation
is prevented by the anode shield, and in
the case of the glass bulb unit the filtering
action of the glass is sufficient to reduce the
radiation to a safe value, so that this effect
is not regarded as being of much importance
in normal rectifier design.
Mercury Arc as Frequency Changer.
An attempt to enter into details of the
methods used to employ the rectifier as a
frequency changer would entail a lengthy
explanation. Very briefly there are two
main methods:
(a)By the.use of a D.C. link.
(b)The direct converter.
" A " and " B " operate alternatively as
rectifiers. When " A" is passing current,
the control gi-ids on " B " have a negative
potential in relation to the cathode so that
no current can flow in " B."
In method (a) a rectifier is used in con
junction with an inverter, the connections
being arranged as shown in Fig. 20 for
three-phase to single-phase.
The current from " A " flows to bus
bar " D " and back through the load to
busbar " C," and then to the transformer
neutral. The voltage thus given on the bus-
Fig. 21.
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
bars is obviously an ordinary rectified D.C.
voltage, busbar " D " being positive. This
rectified voltage can, of course, be kept on
the bars for any time interval desired, after
which the control grids on "A " are given
a negative charge when the current flow in
"A" ceases at the end of the half cycle
on the anode carrying current. If at the
same instant the negative potential on the
grids of "B" is removed, rectifier " B "
will carry current as a rectifier. Busbar
" G " then becomes positive, and the direc
tion of current through the load is reversed.
By alternate operation of " A " and " B "
an alternating voltage is supplied to the bus
bars " C " and " D."
The resulting busbar voltage has a wave
shape far removed from a sine wave, as
denoted in Fig. 22.
Fig. 22.
It is obvious each rectifier must operate
for equal times, so that the positive and
negative half-waves are of equal duration.
For conversion from 50 cycles to 16f cycles,
as mentioned in the paper, with six-phase
rectifiers, each rectifier must obviously
operate for 1^ complete cycles of the 50
cycle supply.
There are several methods of correcting
the wave form to bring it more nearly to a
sine wave on the low frequency side. One
method is to arrange the transformer to give
different voltages on the secondary wind
ings ; for instance, phase 4 would be wound
for 100 per cent., phases 3 and 5 for some
thing less, 2 and 6 for something less still,
and phase 1 for, say, 30 per cent. This
results in alteration of the anodei voltage
curve and consequently the voltage wave.
See Fig. 23.
This explanation is admittedly very in
complete and skethcy.
II. Bissik, in an article in The Electrical
Review of June 12th, 1931, stated that a
2,500 kVA frequency changer of this type
was being supplied to the German Federal
335
Railways for single-phase supply at 15,000
volts, 161 cycles. I have no information in
regard to this unit in service.
Dr. Randall's discussion does not really
call for any comment from me except per
haps to say that 3,000 volts D.C. is quite
standard for railway electrification, and
voltages of 12,000 and hihger are in use for
radio communication.
Fig. 23.
Mr. Badham emphasises the necessity of
the ignition point controlled by the grid
being 100 per cent, perfect. The problem
is complicated by the variations which may
occur in the point of ignition due to varying
arc drop and variations in temperature of
the rectifier. This has led to the adoption
of impulse excitation of the grid where the
voltage impulse can be made suffi ciently
high to ignite the rectifier at the correct
instant despite variations in arc drop and
temperature. This peaked impulse voltage
necessitates the use of the saturated trans
former method or the synchronous contact
method, as described in the paper. The
sinusoidal voltage method of biasing thegrids mentioned in the paper does not meet
this condition adequately.
In reply to Mr. Brinkworth, my brief
comment on the overload capacity of recti
fiers certainly requires amplification.
Ordinary electric machines or trans
formers, owing to the large masses of metal
which are used in their construction,
usually take at least six hours before the
temperature reaches a steady value when
operated on full load starting from cold, and
if then placed on overload will again take
an appreciable time before the temperature
rises to the maximum safe value. With a
rectifier the governing factor is usually the
temperature of the anode, and the anode is
really quite a small mass of graphite, there
fore the anode very quickly reaches its
maximum steady temperature for any par
ticular current. If, therefore, when opera
ting at full load the anode operates at its
-'136
The Transactions of the S.A. Institute of Electrical Engineers.
maximum safe temperature, any sustained
overload will raise the anode temperature
beyond the safe limit, and back-fires will
result. Under these conditions only
momentary overloads are possible. If a
rectifier is built to carry 50 per cent, over
load for one hour, as mentioned in the
paper, then as far as the anode is concerned
it must be built to carry this load continu
ously, which means that at full load it is
running much below its maximum output.
A short circuit can be carried without
trouble, as the circuit will be intercepted by
the protective devices before the anode tem
perature can rise to a dangerous value.
Under similar conditions, a commutator
machine would probably dash over before
the circuit breaker could operate.
In designing a rectifier for a particular
overload, care is taken that the design of
the water-cooling system is suffi cient to take
away the heat developed on overload, and to
ensure that the temperature of the cylinder
and consequent rise of mercury pressure are
kept within safe limits.
It may be of interest to note that one
manufacturer has placed on the market
small air-cooled units which resemble in
shape and general proportions a glass bulb
rectifier. No glass, however, enters into
their construction, steel replacing it.
The anodes are placed in arms, and this
construction is made possible only by the
use of the special vitreous enamel seal, as
described and illustrated in Fig. 5 of the
paper.
These small rectifiers are air-cooled in a
precisely similar manner to the air-cooling
of the glass bulb rectifier.
Actual experience so far obtained, ex
tending over the last three or four years,
seems to point to the strong probability that
it may be possible to dispense entirely with
the air pumping equipment and treat these
units in an exactly similar manner to the
glass bulb type. With installations supplied
so far, as experience with this type is com
paratively limited, it has been considered
advisable, as a form of insurance, to install
an air pumping equipment, for it is felt that
it is not possible to guarantee that these
units will be absolutely leakless, say, over
20 years' service. This pumping system
would certainly only be required for a few
hours at a time, once or twice a year, and
the pumping equipment is arranged as one
unit, common to all rectifier units installed.
[December, 1935.
The loss of vacuum would not be due to
leaky joints, but it is felt there is always
the possibility of air seepage through the
actual metal itself.
This rectifier would seem to hold out
great possibilities as, being of all-steel con
struction, it will carry overloads in exactly
the same way as the water-cooled, steel
type rectifier; as Mr. Brinkworth points
out, the overload capacity of the glass bulb
type is severely limited, owing to the lack
of heat capacity of the glass used in its
construction.
For servicing an installation of the small,
steel-clad type, it is anticipated that no airevacuating plant being required, the units
would be treated exactly as a glass bulb
unit, and replaced in case of failure.
Obviously, the steel-clad unit can be easily
re-conditioned and put back in service.
Mr. Brinkworth states that there are
many B.C. commutating machines givjng
500 kW at 12,000 volts per commutator. I
should be very interested to know where
these are installed. It is apparently some
new development about which nothing has
been published, and I am sure we should
all like a little fuller information.
In mentioning 8,000 volts as the limit for
the commutator of a D.C. generator, I was
aware that as far back as 1886 gramme
machines had been built for 6,000 volts per
commutator, with closed coil armatures.
Open coil armatures giving 3,000 volts were,
of course, quite common at the same date
for series arc lighting.
I still maintain that about 3,000 volts is
the maximum voltage per commutator for
a commercial machine of normal con
struction.
The high tension D.C. motor generator
sets first supplied to the B.B.C. each con
sisted of a motor driving two generators.
The D.C. output was 160 kW at 12,000
volts. Each generator armature had a com
mutator at both ends, and the four commu
tators were connected in series, giving
3,000 volts per commutator.
Perhaps Mr. Brinkworth has in mind the
H.T. D.C. generators used for electrostatic
precipitation plant. With this type of
plant, voltages of the order of 15 kV to as
high as 50 kV may be required, but with
50 kV the current would be something less
than a quarter of an ampere, the
output of the machine being less
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
than 10 kW. Here again it is usual to
employ two commutators per armature, and
connect suffi cient commutators in series to
keep the voltage per commutator below
8,000 volts, generally less. As mentioned,
the current to be dealt with is usually a
fraction of an ampere, and the machines
may be considered as entirely special. If
12,000 volts, 500 kW on one commutator is
now common practice, the present limit of
1,500 volts in Great Britain, and 3,000
volts elsewhere, for D.C. traction would
seem unnecessarily conservative, and H.T.
D.T. traction an immediate possibility.
The possibility of automatic control of
rectifier sub-stations was mentioned in the
paper. Those interested will find an article
dealing fully with this subject in the
G.E.C. Journal, Vol. IV., No. 3.
I do not agree with Mr. Brinkworth that,
with a rotary, surges on the D.C. side are
largely taken from the kinetic energy of
the machine. This effect can only be slight,
for the rotary is a synchronous machine.
Kinetic energy can only be given up by
change of speed, and only a very small
variation of angular velocity is possible if
the rotary is not to fall out of step.
As mentioned in my reply to Mr. White,
a reverse current high speed circuit breaker
is the best protection against back-fires.
Since dictating the above, I have seen in
the Journal of the I.E.E., Vol. 77, October,
1935, a description of the Droitwich Broad
casting Station. On page 450 a description
is given of the M.G. sets installed. These
are of 300 kW capacity, designed for any
voltage between 7,000 and 12,000 volts.
Each generator, which runs at 500 revs.,
has two commutators connected in series,
so the voltage per commutator in this par
ticular machine is from 3,500 to 6,000 volts.
The weight of the armature alone is 6^
tons, which is only about 1 ton lighter than
a complete normal 300 kW D.C generator
at this speed. The armature is obviously
very special, and the machine very heavy
for the output. Incidentally, of course, on
these high voltage generators the armature
core is insulated from the earthed shaft.
The President: Ladies and gentlemen,
our thanks are due to Mr. Winstanley for
his very able reply. You will notice that
on the agenda we have two more Replies
to Discussion; one by Mr. Littlewood, and
337
one by Mr. Rochester. I would ask these
gentlemen to bear with us, as the hour is
late, and to save their Replies to Discus
sion for a future date, or alternatively, to
permit up to publish them in the Journal
without first reading them at a general
meeting. We have a further discussion by
Mr. Montgomery on Mr. Mitchell's paper,
" The Technical Development of the Tele
graph Service in the Union of South
Africa." I think in this case also the same
procedure should apply.
That, ladies and gentlemen, brings us to
the end of our proceedings, and I would
remind you that refreshments will be
served, as usual, downstairs. This being
the last meeting of the year, I would like
to wish you a very happy Christmas and a
happy and prosperous New Year.
Mr. C. T. Cooks (President Elect) :
Ladies and gentlemen, I feel we could not
definitely close this meeting without my
being first permitted, on behalf of all
present, to extend to our worthy President
and Mrs. White the compliments of the
season and to wish them many happy new
years to follow. (Applause.)
Mr. President: Thank you, Mr. Cocks
and ladies and gentlemen.
The meeting then terminated (10 p.m.).
IMPROVEMENTS IN MODERN TURBO
TYPE ALTERNATORS.
By H. M. ROCHESTER (Associate Member).
(Journal, April, 1935.)
Reply to Discussion.
Surge Distribution. '
According to the remarks of Mr. Badham,
any type of alternator with normal end
windings is not capable of giving uniform
surge distribution. The speed of propaga
tion of a surge through the end windings of
an alternator is about 600 feet per micro
second, while in the slots it is only onetenth of this value. Knowing the length of
one turn of the winding, the time for a
surge to traverse one turn can be calcu
lated, and when taken in conjunction with
the rate of rise of voltage will give the volt
age between turns, the maximum value so
obtained being on the end turns.
The normal type of strip wound alter
nator having several turns in series per slot
838
The Transactions of the S.A. Institute of Electrical Engineers.
may be subject to high interturn stresses,
but it is fundamental of a concentric con
ductor winding that successive turns are in
adjacent slots and the interturn stresses can
only exist in the end windings, when the
clearances are very liberal.
In addition, reflections of a surge can be
set up at a transition point where the end
windings join the conductor or enter the
slot. Again, the fact that successive turns
are in adjacent slots renders the concentric
arrangement more or less immune from
such troubles.
It is well known that the slope of surge
wave fronts may correspond to frequencies
of the order of 1,000,000 cycles per second,
at which frequency the impedance due to
the inductance of the windings becomes
much greater than that due to the capacity.
The surge current will thus pass directly to
earth through interturn capacity, and if this
is uniformly distributed through the wind
ings then the surge voltage will distribute
uniformly over the windings.
It will be clear that the concentric con
ductor winding has a great advantage over
the ordinary type of winding by reason of
its inherently high capacity between turns
which form a capacity chain direct to earth
and present a relatively low impedance to
a surge.
The following extract is taken from the
Journal, A.I.E.E., January, 1934, by J. F.
Calvert. As a result of cathode ray oscillo
graph studies, the investigator states: —
" The winding using concentric turns
(one actually inside the other within
one slot) ' which was introduced in
Europe for high voltage machines
should give adequate protection to the
turn insulation."
A voltage surge may give rise to exces
sive voltage difference in the windings of
an alternator, transformer or similar appara
tus, as follows: —
1.A surge of high amplitude causes ex
cessive voltage between windings and
earth, which may be accentuated by
reflections from the neutral point or
from the junction of two sections of
the windings having different charac
teristics.
2.A surge having a steep wave front
causes a high voltage between the end
turns of a winding where the capaci
tance of a turn to earth is appreciable
[December, 1935.
in comparison with the interturn
capacities—this is so. for wow-concen
tric conductor winding.
3. The slope of the wave front of a surge
is gradually modified owing to the
absorption of energy as it passes
through a winding; there is then the
possibility of the slope of the wave
front at some part of the windings
corresponding'to its natural frequency,
and an excessive voltage due to reson
ance may be sent up.
The protection of machines against volt
age surges due to lightning and other causes
has received much attention, but there still
appears to be some controversy as to the
best measures to adopt. These must depend
a great deal on the conditions to which the
plant is subject.
On a number of systems lightning
arresters are fitted in order to reduce the
amplitude of the voltage surges; some engi
neers are of the opinion that the operation
of lightning arresters is uncertain and that
this apparatus is itself a potential source of
breakdown.
Surge absorbers have been fitted on a
number of systems, and while they are
effective for surges with steep wave fronts,
a surge with a less severe wave front may
reach the windings of terminal apparatus
and cause a high voltage between turns
owing to resonance.
(It may be of interest to members to
know that a combined surge absorber and
excess voltage discharge has been developed
by Messrs. Merlin and Gerin, Grenoble,
France, who are associates of Messrs. A.
Reyrolle & Co., Ltd., Hebburn-on-Tyne.)
Experience and discussions with a num
ber of enginers indicates that probably the
best safeguard for general application is to
provide a long cable connection between
the apparatus and the high overhead lines.
The length of cable generally recom
mended is about 400 yards. On the English
North East Coast the practice is to use
250 yards for a 20 kV line and about 70
yards for a 66 kV line. The corresponding
length for a 35 kV line would be 150-200
yards.
It is recommended that the cable should
be steel wire armoured in order to absorb
the energy of the surge. Where the ter
minal of the line is close to the station it is
December, 1935.]
The Transactions of the 8.A. Institute of Electrical Engineers.
339
usual to coil the excess length of cable in
the ground.
stand any transient voltage occurring in a
concentric conductor winding.
In order to protect the high voltage alter
nators not of the concentric conductor
design, having a large number of turns per
slot, American practice favours the use of
surge absorbers at the machine terminals,
and, if the insulation of the winding is
graded a lightning arrester is connected to
the junction of the high and low voltage
sections of the winding.
The end windings in a concentric con
ductor design are amply spaced and insu
lated, and the arrangement is such that
there is no danger of breakdown due to a
It has also been proposed that condensers
should be connected between tappings taken
from machine windings in order to assist in
distributing the surge potential uniformly
throughout the winding. There are obvious
disadvantages in bringing out a large num
ber of tappings at high voltages and in
accommodating and insulating the con
densers.
The Parsons concentric conductor type of
high voltage winding has many advantages
under transient conditions. An important
feature of this winding is the high capaci
tance between sections of the winding form
ing a direct path to earth for a surge.
Further, as opposed to the small values
obtained in non-concentric windings this
capacitance is high compared with the elec
trostatic capacity of the turns to earth, and
for this reason a surge voltage even with a
steep wave front is uniformly distributed
across the windings from the bull conductor
to earth. The transient voltage between
sections and between turns is thus kept down
to a moderate figure. The ratio of the
transient voltage between sections or turns
to the normal voltage difference at these
points cannot be greater than the ratio of
the maximum amplitude of the voltage
surge to the terminal voltage of the alter
nator ; whereas, in a non-concentric wind
ing having a large number of turns in one
insulating tube, the relation of the capaci
tance between turns to that of the turns
to earth is such that the transient voltage
may be many times greater than the normal
voltage between turns. (Such turns may
be provided with the full thickness of in
sulation, but this arrangement occupies a
large amount of space and is uneconomical.)
In the concentric conductor alternator
each successive conductor is situated in a
different slot and is, therefore, fully
insulated. The insulation between sections
is equal to a third of the total insulation
to earth and is more than ample to with
surge.
The possibility of resonance due to a volt
age surge is also remote with a concentric
conductor winding, owing to the high
capacitance to earth.
A list of papers and articles relating to
the j^rotection of rotating machines from
surges is attached.
Concentric conductor type alternators are
in operation in Africa and England without
any form of lightning protection, and, as far
as is known, no special measures are taken
against lightning. The high voltage alter
nator at East Band, in addition, operates
in conjunction with Petersen coils.
Heat Gradient.
In his remarks, Mr. Badham has stated
that for a given current density in the con
ductors the temperature of the bull in a
concentric winding would be greater than
that of a single conductor, but in his argu
ment he has assumed that the total heat
from, the three conductors has passed
through the total thickness of insulation.
This is not so. The heat passing through
any layer of insulation is only that of the
conductor or conductors enclosed therein,
and consideration of a triple concentric con
ductor shows that the heat from the bull
only passes through the first layer of insula
tion, that from the bull and. inner through
the second layer of insulation, and all of
the- loss through approximately one-third of
the total insulation. This permits suitable
grading of current densities to secure the
most favourable distribution of temperature.
For the same loss per unit of cooling
surface the temperature difference between
the bull conductor and surrounding slot is
approximately 30 per cent, less than for a
single conductor with equivalent insulation.
Corona.
In conjunction with the consulting enginers (Messrs. Merz & McLellan) for the
H.V. generators for the Salt River Power
Station, tests were carried out for corona.
A canopy was erected over one end of the
alternator, and after 15 minutes had elapsed
for the spectators to accustom their eyes- to
340
The Transactions of the S.A. Institute of Electrical Engineers.
the darkness, the alternator was excited to
normal volts. No corona was visible. A
further test was carried out at over-voltage
and photographic records obtained. Again
no corona was visible.
Experience has shown that corona is not
likely to occur on alternators of normal
constructions up to 11,000 volts.
Sine of Alternator.
The maximum size of concentric conduc
tor H.V. alternators in service are installed
at the newly-opened power station at Swan
sea, where two units, each of 37,500 kVA,
are installed. As far as I am aware, the
maximum size of any other H.V. alternator
under construction is considerably less than
twice this rating and is not yet in service.
Reactance and Stresses.
It is a feature of the design of a concen
tric type H.V. alternator that a high
reactance is obtained, a value which is diffi
cult and not economic to obtain on a L.V.
type of alternator. In addition to this, the
arrangement of the end windings and the
clearances available reduce the electro
magnetic forces to very low values.
In the course of works tests, H.V. alter
nators with concentric windings have with
stood repeated sudden short circuit tests
when excited to normal volts on open circuit
without the windings showing the slightest
movement; also, H.V. alternators in service
have been subject to severe system disturb
ances, and subsequent examination showed
no signs of electrical or mechanical injury
to the windings. This provides an illustra
tion that in making the utmost of technical
developments safety is not necessarily
prejudiced.
PAPERS AND ARTICLES RELATING TO THE
PROTECTION OP ROTATING MACHINERY
PROM SURGES.
" High Voltage Alternators," by J. Rosen. Journal
of the J.I.E. Vol. XLV, Part I, Oct., 1934.
" Connecting Rotating Apparatus Direct to the
Line." S. L. Henderson, Electrical Journal.
Vol. 27, 1930, p. 647.
Voltage Oscillations in Armature Windings Under
Lightning Impulses." A.I.E.E. Trans.. Vol.
49, Oct., 1930, pp. 1587-1607.
Effects of Lightning Voltages on Rotating
Mahcines and Methods of Protecting Against
Them." P. D. Fielder and E. Beck, A.I.E.E.
Trans.. Vol. 49, Oct., 1930, pp. 1577-86.
" Surge Protection for Rotating Machines." J. E.
Calvert, A. C. Monteith and E. Beck, Electrical
Journal, Vol. 30, March, 1935, pp. 91-4 and 116.
[December, 1935.
" Protecting Machines from Line Surges." J. E.
Calvert, Electrical Engineering (A.I.E.E.
Trans.), Vol. 53, Jan., 1934, pp. 139-46.
" Transition of Lightning Waves from one Circuit
to Another Through Transformer." K. K.
Palueff and J. H. Hagerguth, A.I.E.E. Trans.,
Vol. 51, Sept., 1932.
THE TECHNICAL DEVELOPMENT OF
THE TELEGRAPH SERVICE IN THE
UNION OF SOUTH AFRICA.
By J. A. F. MICHELL, B.Sc. (Eng.) (Member).
(Journal, May, 1935.)
Discussion by T. MONTGOMERY (Member).
I feel I should like to contribute a few
observations before Mr. Michell's valuable
paper is closed to discussion.
I have been associated with the telegraph
service for 40 years. During the first ten
years I saw little progress. Everything was
standardised—the last word had been said,
and it only remained to exploit the competi
tive spirit amongst the operating staff to
secure the highest possible traffi c output.
We looked askance at the, power engineer ;
he was not to be relied upon, and so, rather
than take unnecessary risks, we dej)ended
on primary cells to supply whatever volt
ages we required. However, as time went
on the power engineer gave proof that he
could " deliver the goods," and then began
the conversion of the telegraph batteries at
important stations from primary to second
ary cell working.
It is remarkable how one innovation leads
to another. Before many years had passed,
the conversion of our wheatstone trans
mitters from weight drive, necessitating
frequent attention to the task of winding up
the weight, to the much steadier motor
drive, had become an accomplished fact.
The mechanisation of the telegraphs during
subsequent years went on without undue
haste, but with relentless persistence, and
to-day we have to acknowledge that it is the
machine and not the skilled manual tele
graphist which is the sheet-anchor of the
telegraph service.
Referring back some thirty to forty years,
there was throughout the Capo Colony a
wonderfully useful chain of single current
repeating stations where the incoming
signals were amplified and passed forward,
not always free from distortion.
The
December, 1935.] The Transactions of the S.A. Institute of Electrical Engineers.
change-over to double current working and
the introduction of fast speed apparatus,
with suitable repeaters at a few selected
points, displaced these primitive devices
during the reconstruction period which fol
lowed the Anglo-Boer War.
It has been said that the telegraphs have
become a by-product of the trunk telephone
service. It is not strictly true, at least not
at this stage, but, regarded, from the point
of view of inductive disturbance, it is highly
desirable that the telegraph signalling cur
rents shall be of approximately the same
magnitude as those used for telephone
speech. This condition is secured where
both telegraphs and telephones are worked
by the carrier current method described by
Mr. Micheli.
Although Mr. Micheli does not favour
voice frequency (300-3,000 cycles) for tele
graph working, a compromise may be pos
sible without having recourse to the expen
sive equipment entailed by the present high
frequency carrier method, which is only
justifiable between points requiring a large
number of channels. Research seems to be
proceeding along these lines, and a solution
may, therefore, be looked for.
Mr. Michell's reference to the new baby
of the telegraph service, viz., the tele
printer exchange, gives one food for much
speculation. We have seen in our day the
gradual encroachment by the telephone on
what was at one time regarded as the
domain of the telegraph. I refer to long dis
tance communication. May we not now be
entering on a new phase where the tele
graph typewriter or teleprinter will win
back the ground temporarily lost. The pos
sibilities of the new application are pro
digious when it is remembered that up to
eighteen telegraph channels can be worked
by the carrier current method within the
frequency band normally allotted to one
telephone channel. It is possible that
within ten years the bulk of the business
telegram traffi c will be handled direct by
the teleprinter method between the offi ces
of the firms concerned through teleprinter
exchanges.
A point which might, I think, be stressed
is the present-day dependence of the com
munications engineer on the power engi
neer, or, to put it in another form, the milli
watts on the kilowatts. A similar position,
of course, obtains in the industrial sphere,
but the temporary failure of a public utility
service is always regarded as an intolerable
341
state of affairs. Fortunately a cessation of
telegraph communications following a
breakdown of the power supply has never
occurred in South Africa, though on occa
sions battery reserves have fallen danger
ously low. The elimination of the duplicate
battery, which is foreshadowed in Mr.
Michell's paper, definitely increases the
dependence of the telgraph service on the
local power supply, and creates a need for
emergency plant to ensure the continuity
of telegraph and also telephone communi
cations.
The operation of the belt conveyors for
collecting and distributing the telegrams in
the new Johannesburg telegraph office will
be watched with interest. The time interval
known as " offi ce drag" will at all events
be calculable, being no longer wholly
dependent on that erratic quantity
" modern youth."
In conclusion, I should like to thank Mr.
Micheli for his very instructive paper.
VISIT TO DELTA SEWAGE DISPOSAL
WORKS.
By the courtesy of the City Council of
Johannesburg, approximately fifty members
of the Institute and their friends paid a
visit of inspection to the Delta Sewage Dis
posal Works on Saturday morning, the 7th
December, 1935.
The visitors were welcomed by Dr. E. J.
Hamlin, the Johannesburg City Engineer,
and after refreshments had been served
they were shown over the works in small
parties, each under the guidance of one of
the Council's officials, who explained the
various workings.
The Council wishes to place on record
its appreciation of the Johannesburg City
Council's' invitation to visit the works and
for its hospitality, and its sincere thanks
to Dr. Hamlin a"nd the members of his
staff who so lucidly explained the works.
DESCRIPTION OF DELTA SEWAGE
DISPOSAL WORKS.
By H. T. CLAUSEN, A.M.Inst.C.E.
(Superintendent, Sewage Disposal Works,
Johannesburg).
The Delta Sewage Disposal Works,
which were put into operation on the 21st
April, 1935, will deal with the domestic
sewage and trade waste from.that portion
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers,
of Johannesburg comprising the north
western drainage basin. The flow at
present is approximately 300,000 gallons
per day, but it is estimated will eventually
be 2,000,000 gallons per day which these
\vorks, with extensions, will be capable of
treating.
The sewage, on entering the works,
passes through a hand-operated bar screen
With one inch openings, through a small
detritus pit for catching the heavier solids,
and then flows to the sedimentation tanks,
^svhere the solid matter is settled out. All
these units are enclosed in a chamber which
lias forced ventilation. Electrically pro
duced ozone for oxidising impurities in foul
air is mixed with the air exhausted from
this chamber before being discharged to the
atmosphere.
^ To the effl uent from the sedimentation
tanks is then added activated sludge, bear
ing bacteria, which, acting probably as a
catalyst for the oxygen in the air, oxidise
313
the impurities 'in the settled sewage, this
being achieved in the aeration tanks, where
air is blown through diffusers into this mix
ture of settled sewage and activated sludge
for the purpose of keeping the activated
sludge in suspension, and supplying the
necessary oxygen to the biological growths.
. 1'he mixture then flows to sludge separat
ing tanks, where the activated sludge
settles out, and is drawn off continuously
to be returned by air-lifts to the influent to
the aeration tanks to do further work.
There is a constant increase in the quantity
of activated sludge, so that a certain amount
has to be drawn off constantly, and this is
pumped to the incoming sewage to be
settled out in the sedimentation tanks.
The effl uent from the sludge separating
tanks may then be run on to land, or may
be passed through sand-straining filters,
chlorinated and discharged into a nearby
stream, this latter course only being
resorted to if necessary during storms.
Pig. 3.
Sludge Digestion Tanks with Floating Steel Gas Collectors, Deita.
December, "935.]
The Transactions of the S.A. Institute of Electrical Engineers.
345
Fro. 6.
Gas Engine Alternator Set, Delta.
The solid matter, consisting of the
screenings and the sludge from the sedi
mentation tanks, is pumped to the digestion
tanks, where anaerobic alkaline fermenta
tion process reduces a large quantity of the
volatile matter to gas, and also causes the
remaining solid matter to lose its offensive
properties. The residue from these tanks
is later run on to under-drained drying beds
of sand. The dried solids are then removed
and used" on land.
Gas, consisting of approximately 70 per
cent, methane, the remainder being princi
pally carbon-dioxide, is generated during
this process of fermentation, and is
collected into floating steel gas collectors,
from which it is led to a 156 b.h.p. gas
engine direct coupled to a 105 kW alter
nator. The electrical energy derived from
this local source is utilised for driving air
compressors for supplying air to the aera
tion tanks, and for driving sludge pumps,
ozone producing apparatus, etc. The waste
heat from the gas engine is utilised via a
heat exchanger and hot water coils for heat
ing the raw sludge, as the optimum tem
perature for the sludge fermentation process
is 80 Fahrenheit.
As an external electricity supply from
the municipal system is also utilised,
because the local supply is not sufficient,
the controlling switchboard has two sets of
busbars, one for the local and one for the
external supply. A busbar coupler is
included on the switchboard, so that
parallel running of the two supplies is
possible by closing this. Each of the out
going supply panels is . equipped with
change-over links to either set of busbars,
so that any unit may be run on either of
the supplies. Included on this switchboard
are electrically operated meters and
recorders for the sewage flows, the air
supply and thermo-couple thermometers in
the digestion tanks; also pilot lights for the
ventilating and ozone producing apparatus.
The offices, laboratory, machinery room
and sedimentation tank chamber are all in
one building.
346
The transactions of ihe ^.A.' Institute of Electrical Engineers.
[December, 1935.
INSTITUTE ANNOUNCEMENTS
AWARDS.
The Council is empowered each year at
its discretion to award the Gold Medal of
the Institute, together with a certificate
and, if considered desirable, a premium not
exceeding 10 10s., to a member of any
grade for an original paper of outstanding
merit submitted during the year.
It is also authorised to award at its
•discretion the Bronze Medal of the Insti
tute, together with a certificate and, if con
sidered desirable, a premium not exceeding
5 5s., to a member of the Student grade
for an original paper of outstanding merit
•submitted by a member of that grade.
The Institute is indebted to the Manage
ment of the Victoria Falls and Transvaal
Power Company, Limited, for an annual
-donation of 25 for the purpose of award
ing premiums, not exceeding 10 10s. in
value for any one award, for the purchase
•of scientific and technical books or instru
ments for .papers and contributions to the .
discussion of papers, which, in the opinion
of the Council, merit special recognition.
The papers and contributions of Student
Members are eligible for awards both in
their own Section and in the Senior
Section, and all awards are at the dis
cretion of the Council, which may vary
them in amount according to the merit of
the papers and contributions submitted
during the year.
The Institute is also indebted to the South
African Cable Makers' Association for an
annual donation of 5 5s. for the purpose of
awarding one or more premiums each year
to the total value of 5 5s.
The conditions governing this Award are
the same as those for V.F.P. Awards, with
the exception that all Papers and Contribu
tions read before the Institute are eligible
for the Award irrespective of whether the
authors are members of the Institute or not.
Secretary at least one month before the
date of the meeting at which it is to be
read.
Where symbols or abbreviations are con
sidered necessary, if is desirable that
definite standards should be used.
Attention is drawn to the publications cf:
the British Standards Institution dealing
with " terms " and graphical symbols used
in electrical engineering, with the sugges
tion that these be employed.
DIAGRAMS FOR TRANSACTIONS.
It is requested that thefollowing
rules be observed by memberspreparing
illustrations
for
publicationin the
" Transactions ": —
Drawings accompanying papers should
be made in India ink on white paper; hand.
sketches, if they are bold and heavy in
• o utline, are suitable for reproduction in the
Journal.
Drawings or photographs to be shown on
the screen by means of the epidiascope
should not exceed 5 in. x 5 in. in size, and
slides must be 3^ in. x 3^ in.
All drawings and photographs require to
he numbered to ensure their use in the
correct order.
ADDRESSES WANTED.
Addresses are wanted for the following:—
L.
L.
B.
S.
C. Lawson.S. Lawrence.
Graham.A. Leek.
W. Fowler.A. H. Buitenweg.
G. E. Geering.
Kindly communicate with the Secretary,
F.O Box 5907, Johannesburg. Telephones,
33-5248/9.
PAPERS
LETTERS OF DESIGNATION.
Members presenting papers before the
Institute are requested to forward a copy
•of their paper, in final form, to the
The attention of members is drawn to
Rule 26 of the Constitution and Rules of
the Institute, which reads:—
December, 1935.]
The Transactions of the S.A. Institute of Electrical Engineers.
" The authorised letters designating the
class in the Institute to which a member
belongs shall be as follows: —
For an Honorary Member,
Hon. M.(s.a.)I.E.E.
For a Member
M. (s.a.)I.E.E.
For an Associate Member,
A.M.(s.a)I.E.E.
For an Associate,
Associate (s.a.)I.E.E.
For a Student, Student (s.a.)I.E.E.
In each case the letters S.A. shall be in
brackets and be less in size than, but
not less than half, the size of the other
letters."
TO COUNTRY MEMBERS.
The Council is particularly desirous of
receiving written contributions from mem
bers resident in the country on any of the
papers up for discussion as set forth in the
Agenda.
Members who are unable to attend the
Ordinary General Monthly Meetings should
avail themselves of this opportunity.
In this connection members should
bear in mind the Institute, V.F.P.. and
Students' Awards which are available for
members presenting a paper or contribu
tion of suffi cient merit.
Kindly address all contributions, etc.,
to the Secretary, P.O. Box 5907,
Johannesburg.
MINUTES OF THE BRITISH
STANDARDS INSTITUTION.
The Minutes of the above Institution
are regularly received by the Institute, and
may be obtained for reference purposes on
application to the Secretary.
BRITISH STANDARDS
SPECIFICATIONS.
H.M. Acting Senior Trade Commissioner,
Johannesburg, has advised the Institute
that he has received from the British
Standards Institution, of London, a com
plete set of British Standards Specifications.
These Specifications are filed in his
office, and may be consulted there when
required.
The Specifications are for consultation
only, and cannot he loaned from his offi ce.
347
BOOK-PLATES.
Book-plates for publications purchased
from Award Premiums may be obtained
from the Secretary of the Institute.
TECHNICAL PUBLICATIONS.
By the kindness of Doctor H. J. van der
Bijl, the chairman of the Electricity Supply
Commission, the Institute is favoured with
copies of the weekly library extracts of
current technical literature prepared by the
Librarian of the Commission.
The Council desires to announce that the
Extracts are available for reference by
members on application to the secretary,
and, in future, a copy will also be available
on the bookshelves in the lounge of Kelvin
House.
W.P.C. AMERICAN COMMITTEE.
The American Committee of the World
Power Conference has kindly extended an
offer of assistance to South African
engineers visiting the United States, and
is willing to advise on places of interest
to visit, itineraries, etc. The offices of the
Committee are at 1419-21, Chrysler Build
ing, 406, Lexington Avenue, New York
City.
Intending visitors are invited to com
municate with the Committee at the above
address.
RECIPROCITY ARRANGEMENTS.
Reciprocity arrangements exist between
the South African Institute of Electrical
Engineers and the Institution of Electrical
Engineers, London, the American Insti
tute of Electrical Engineers, the Societ^
Fran^aise des Electriciens, Paris, the
Association Suisse des Electriciens, Switzer
land, and the Verband Deutscher Elektrotechniker, Germany. It is hoped that
members proceeding overseas will avail
themselves of the facilities offered. Full
particulars are obtainable from the
Secretary.
CLIPPINGS FROM TECHNICAL
PRESS.
Members are invited to send in any
clippings from the Technical Press which
may have bearing on papers up for d^s
cussion, or may be of general interest.
Please state source. These clippings will
be submitted to the responsible editor, who
will pass for publication such items as can
be accommodated in the Transactions.
348
The Transactions of the S.A. Institute of Electrical Engineers.
WORLD POWER CONFERENCE.
The next Chemical Engineering Congress
of the World Power Conference will be held
at the Central Hall, Westminster, London,
between June 22nd and June 27th, 1936.
Any information in connection with the
Congress may be obtained from the Secre
tary of the South African National Com
mittee World Power Conference, P.O. Box
1091, Johannesburg.
[December, 1935.
FOR SALE.
A Foundation Member of the Institute
has a complete set of the Transactions for
sale, the first fifteen volumes of which are
bound in half-calf. Condition as new.
What offers?
Apply to the Secretary, S.A. Institute of
Electrical. Engineers, P.O. Box 5907,
Johannesburg.

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