Produced by The Online Distributed Proofreading Team at
http://www.pgdp.net (This file was produced from images
generously made available by The Internet Archive)









                          The Thompson-Houston
                           System of Electric
                               Lighting.

                   Thesis submitted for the degree of
             Bachelor of Science in Mechanical Engineering,

                  to the Faculty of Purdue University


                               June 1887

------------------------------------------------------------------------




“In its power to assume always that form of energy which happens to be
the most useful lies the great importance of electricity.” This
importance has been brought to the notice of the public by means of the
many recent exhibitions. Public interest has been roused and there is
everywhere a desire for information and a guide through this far
reaching field for discovery and invention. And, although there are many
works treating on electricity and electric light, people specially want
a short and concise though thorough description of the various schemes
by which electric light is produced. In this thesis the object is to
give a brief treatise on one of the many schemes of producing light by
electric currents viz—The Thomson-Houston System.

In pursuing the subject of electricity, the first thing noticed is the
analogy and difference between the dynamo and its older and more
powerful rival the steam engine. The resemblances are, First as in the
development of the steam engine, but few of the improvements and
inventions in electrical machines were made by mathematical leaders.
Watt ran across the idea of the seperate condenser while repairing the
Newcomen model and applied the expansion of steam to the steam engine by
a mechanical accident rather than by his own ingenuity, and so we find
the first designers of the dynamo were mechanics rather than
philosophers. Secondly the tendency to disregard old methods and
instruments because of new discoveries and inventions has, as in the
steam engine, hindered the advancement in electrical science. As an
example it has become customary to regard frictional and statical
electric machines, for practical purposes, as obsolete, but recent
discoveries seem to hint that they may yet be utilized. Lately Prof.
Dodge has shown that dust and vapor whirling in the air may be settled
by a discharge of electricity consisting of a continuous series of
electric sparks. This has been utilized to clear the atmosphere in lead
smelting works from the fumes of volatized lead and with its application
comes the invention of Wimhurst which produces with a minimum of
mechanical labor a continuous series of electric sparks and works
admirably.

The differences between the engine’s and dynamo’s developement are:
_First_ the marvelously rapid developement of the dynamo as compared
with that of the steam engine. Since 1867 when the term “dynamo electric
machinery” even to scientific men had but little signification, the
dynamo has been brought to a very high degree of perfection. _Secondly_,
the development of the dynamo has reached a much higher degree of
perfection than that of the steam engine. Among the best steam engines
twenty per cent effeciency is considered as very good while a good
dynamo gives out in the form of electricity, ninety per cent of the
mechanical energy put in it. But the class of people who improved and
made the steam engine what it is were as well educated in one sense as
were the men who brought out the dynamo. While it is true that in Watt’s
time the knowledge concerning steam was very meagre, yet the practical
men who _made_ the dynamo, did it by themselves as nearly all the
teachers of electricity knew nothing except what may be called
electrical tricks. As has been said[1] “The teachers and writers of
textbooks, practically did not know that there was anything in common
between the electricity from a rubbed glass machine and voltaic
electricity, or to be brief, that there was a science of electricity as
distinguished from mere natural history.” In fact as late as 1870 there
were really no textbooks on electricity. Even now electrical knowledge
is so meagre as to warrant the same writer’s expression, “We can not
imagine a mechanical engineer mistaking a few inches for a few miles or
a grocer compounding an ounce of sugar with a carload, but this gives
too truthful an idea of the vagueness that still exists.”

In the distant future, electricity will be used for electric lighting
only as subordinate to other uses to which it may be applied such as
heating houses, taking place of stoves for cooking, being used as a
substitute for the steam engine. In fact the motor is rapidly becoming
of as much practical use as the electric light. The principle of the
motor is just this; a certain amount of mechanical energy say thirty
four horsepower per minute into the form of electric currents, which by
the way gives enough current to run 45, 2000 candle power lamps, send
the current and distance through suitable conductors and attach them to
similar dynamo or dynamos but in such a manner that the current in the
second set of dynamos flows in the reverse direction to that of the
first; when, the armature of the second dynamo or dynamos will revolve
and at the pulley or pulleys of the dynamos, aside from friction, will
be given out 95% of the thirty-four horsepower, the loss being due to
the resistance of the conductors. Now in practice a motor is placed on
the arc light circuit the same as a lamp, for energy less than twelve
horsepower. It does not affect the lights and is a clean, neat way of
obtaining energy.

But however true the foregoing may be, the greatest present use of
electricity is to start and maintain light. There are several so-called
systems, embracing dynamos, lamps, regulators, etc, from which I select
the Thomson-Houston as the one for the purpose of describing for several
reasons, _first_, it is at least as good as the average system of which
there is a mushroom growth; _second_, valuble information was kindly
offered by the parent Company; _third_, a good plant is near to which
free acess was given, and _fourth_, we have at the Mechanical Hall of
this University, a dynamo, loaned by the parent Company, which affords
information without any inconvenience. As each part of the system comes
up to be described a little of its history will be given. As the first
part of a system necessary to be produced is the current generator we
will first describe




                     _The Thomson-Houston Dynamo._


In considering the current generator the first thing to be decided upon
is the definition of the term dynamo. The following is thought to be a
correct definition,—A dynamo or dynamo electric machine is a machine
which is used to convert energy in the form of mechanical motion into
energy of electric currents, or _vica-versa_. Those used to generate
currents of electricity are called dynamos, those used to generate
mechanical motion are known as motors.

In attempting to make clear the theory of the dynamo, we will recall
some simple experiments. In Fig. 1, send a current around B from right
to left. Now A being free to move vertically either up or down, connect
its binding posts to a _galvanometer_ (that is, an instrument used to
tell the direction of a current and also used to test the _relative_
strength of two or more currents) and move A up suddenly when a current
will be generated in A whose direction will be the same as that of the
current in B. Now this current is not created energy, because in
lifting[2] the coil A, work is expending against the attraction between
the coils, as between two currents flowing in the same direction there
is an attraction. If we pursue this experiment in its various forms we
will find the following statement known as Lentz law is true, viz: “If
the relative positions of two conductors A and B be changed of which B
is traversed by a current, a current is induced in A in such a direction
that by its electro dynamic action on the current in B it would have
imparted to the conductor a motion of the contrary kind to that by which
the inducing action was produced.”

The theory of this law is that around every wire carrying a current
there is a magnetic whirl (Fig. 3). Now if the conducting wire be passed
through a hole in a horizontal plate of glass and iron filings be sifted
upon the latter they will arrange themselves, as shown in Fig. 2., along
lines, radial in this case, known as lines of force, which arranging is
due to the magnetic attraction of the current in the wire upon the iron
filings. Now in B. Fig. 1, every portion of the wire has just such a
whirl and just such lines of force, or magnetic field, and when A is
moved each part of the wire of A cuts one or more lines of force of the
many magnetic fields making up the magnetic field of the entire coil B.
Now when the wire of coil A cuts magnetic field of B a current is
generated in A acording to the following statement known as Faraday’s
Law; “When a conductor in a field of force moves in any way so as to cut
the lines of force there is an electromotive force produced in the
conductor in such a direction that supposing a figure swimming in the
conductor to turn to look along the positive direction of the lines of
force (in Fig. 1, toward axis of B), and the conductor be moved to his
right, he will be swimming with the current so induced.” Hence in Fig.
1, the current generated in it will be from left to right.

Practically Faraday’s principle means just this: by moving a wire across
a space where there are magnetic lines, the motion of the wire as it
cuts the magnetic lines sets up around the cutting wire a magnetic whirl
or in other words sets up a current in that wire.

The foregoing laws are the “principles of the dynamo,” yet after their
deduction, the progress of the evolution of the dynamo was slow and
attended by many dificulties. Between 1860 and 1870 however, a working
knowledge of these laws became the property of thousands of mechanics,
and by comparing the number of inventions before and after that date
(1860) the present generous growth of systems, dynamos and lamps, prove
that inventions were almost in proportion to the number of people who
had any electrical knowledge. In 1866 Wilde produced a toy
magneto-electric machine for giving shocks, in which he used excited
electromagnets. In the same years Varley and others produced a machine
which excited its own field magnets the type of all machines used in
practice. With this principle of Varley’s and Pacinnotti’s ring, Gramme
produced in 1871 his since famous continuous current generator, one of
which the second dynamo electric machine ever brought to this country
can now be seen at the engine house at Purdue University. In 1877 Silas
Brush brought out his famous dynamo and it may be interesting to know
that he designed and had one made without experimenting in the least. In
the following year a patent was issued to Messrs. Elihu Thomson and
Edwin J. Houston, Professors of electricity in Philadelphia on the
present though much improved _Thomson Houston Dynamo._

[Illustration:

  _Fig. 1._

  _Fig. 2._

  _Fig. 3._

  _Fig. 4._

  _Fig. 5._

  _Fig. 6._

  _Fig. 7._]

To go back to Lentz and Faraday’s laws and carefully consider them we
can but assent to S. P. Thompson’s “fifteen propositions on the dynamo”
which are:—1. A part of the energy of an electric current exists in the
form of a magnetic whirl surrounding the wire.

2. Currents may be generated in a wire by setting up these whirls.

3. We can set up these whirls by increasing or decreasing the relative
distance between magnets and wires.

4. To set up and maintain these whirls consumes power.

5. To induce currents in a conductor there must be motion between them
so as to alter the number of lines of force (Fig. 4 to 7).

6. Increase in the number of lines of force in the circuit produces a
current of the opposite sense to decrease (Fig. 7).

7. Approach induces electromotive force in the opposite direction to
that induced by retreat.

8. The stronger the magnetic field the stronger the current.

9. The more rapid the motion the stronger the current.

10. The greater the length of the conductor which cuts lines of force
the stronger the current.

11. The shorter the conductor not so employed the stronger the current.

12. Approach being a finite process the approaching and receeding must
give alternating directions to the current.

13. By the use of a commutator all the currents can be turned in the
same direction.

14. In a steady circuit it makes no difference what kind of magnets are
used to procure the requisite magnetic field whether permanent or
electromagnets.

15. Hence the current of the generator may be used to excite the
magnetism of field magnets.

Now the Thomson-Houston dynamo comes under that class of dynamos in
which there is a rotation of coils in a uniform field of force, such
rotation (Fig. 6.) being affected round an axis in the plane of the
coil. Of course this dynamo is made like all others of its class to have
_first_, as powerful field-magnets as possible, _second_, the armature
or rotating coil has as great a lenght of wire in it as possible the
wire being thick to offer little resistance and _third_, built to stand
high rotative speed.

The simple theoretical dynamo is shown at Fig. 8, consisting of a single
rectangular loop of wire rotating in the magnetic field formed by large
magnets, and in order to take the current so generated from the loop so
as to give a continuous current, we use a two part commutator (Fig. 9)
consisting of a metal tube split in two and mounted on wood, each half
connected to one end of the loop. The current is taken off by brushes
which lead to the main circuit. But manifestly this dynamo would give no
appreciable current becase it has a very small length of wire on the
armature, so a great number of loops were used which at present
constitute the so-called _drum armature_.

We may rotate the loops of wire in Fig. 8, on one of its sides as an
axis or even push it farther from the center of revolution than that. To
do this, wrap the wire around a ring and connect both ends to a two part
commutator (Fig. 10). If instead of the ring in Fig. 10, being solid it
be a number of coils of wire and if instead of there being one coil
around the ring there be thirty we will have Pacinnotti’s ring before
spoken of. If we used four to ten coils or “bobbins” of large size which
is shown diagramatically at Fig. 11, we would have the Brush dynamo.

So with exceptions we may say that there are practically two types of
dynamos as regards armatures, the _ring_ type as Brush, Pacinnotti’s
Gramme, and the _drum_ armature (page 20).

The Thomson-Houston dynamo is like the rest of that dynamo, unique. To
quote S. P. Thompson; “The Thomson-Houston spherical armature is unique
among armatures, its cup shaped field magnets are unique among field
magnets, its three part commutator is unique commutators.”

[Illustration:

  _Fig. 8._
  Simple Dynamo.

  _Fig. 9._

  _Fig. 10._

  _Fig. 11._

  _Fig. 12._

  _Fig. 13._]

An armature of a dynamo is the rotating coil or coils which generates
currents of electricity by moving in a magnetic field of force. It is
the most important part of a dynamo as it is literally the current
generator. So we first consider the Thomson-Houston




                               _Armature_


It is spheroidal in shape as is noted for the fact of its very seldom
_burning out_, i. e., the electricity heating the wires of the armature
to such an extent as to destroy the insulation or fuse the wires, either
rendering the armature useless. It is made by keying two dish-shaped
iron disks _SS_ Fig. 12, to the shaft x and putting ribs _dd_ about ten
in number in the twenty-five light machine, and over the whole putting
varnished paper. Then at stated intervals, pegs JJJ are driven into
suitable holes in the disks and ribs to help in winding wire on the
shell. Next three insulated wires of equal length are joined together at
_h_ Fig. 13, and the three wires are then wound over the shell in the
following peculiar manner: one half of No. 1 is wound so as to form a
zone of a sphere of which the shaft is in the same plane as the center
circumference of the zone. The armature is then turned on the shaft as
an axis 120° and one half of No. 2 is wound in the same manner as the
first half of No. 1. The armature is moved 120° more and all of No. 3 is
wound. The armature is then turned back, 120° on the shaft as an axis
and the remainder of No. 2 is wound. Lastly the armature is turned back
120° more and the rest of No. 1 is wound. They are bound by wires _gg_
Fig. 13 to hold them when rotating. The object of this rather
complicated winding is to get the three coils equi distant from the
shaft in order that each coil will generate practically the same
current. Now as will be seen the overlapping wires will form a nearly
spherical armature. The armature is mounted on the shaft _x_ as an axis
which extends far enough out from its bearings to put a pulley on the
end _H_ and a commutator on the other end to the three parts of which
are fastened the three wires marked one, two, three, Fig. 13.

It has been urged that the repairs of this armature will be larger than
on any other armature. If there should be a “burnout” it would
necessitate the taking apart of the dynamo and sending the armature to
the factory to be rewound. But it never burns out except through
positive carelessness and it will be found that the repairs on this
armature is less than on the armatures of its several powerful rivals
taken separately even though they be of simpler construction.

When the Thomson-Houston armature is rotated between the cup-shaped
fieldmagnets alternate currents are generated in each coil in turn and
now the next point to be considered is the




                              _Commutator_


which incites the alternate currents so formed into one continuous
current. The commutator as before stated is fastened on the shaft at the
end one, two, three Fig. 13. It consists of three copper plates in the
form of a cylinder each segment _A´A´A´_ covering 115° of the dotted
circle Fig. 14. They are screwed to rod _CCC_ and _DDD_ which are
insulated by wood and gutta-percha plates _EE_ from the iron mounting
_E´´_ which is in turnescrewed to shaft by set screws shown. The wires
one, two, three, have respectively red, white and blue insulation and
are put in binding posts _DDD_ marked one, two, three at the factory and
if not so placed may work badly. The current enters _D_ goes to _B´B´_
which there have direct contact with _A_.

[Illustration:

  _Fig. 14._
  Half Sec. A.B.

  _Fig. 15._ _Air blast Nozzles._

  _Fig. 16._ _Air blast Mechanism._

  _Fig. 17._ _Section of Dynamos._
  Spherical Armature in Fig. 13.]

Now in a three-part commutator the spark occurring as the segments pass
under the brushes would very quickly destroy the surface and interfere
with the currents in the coil. This difficulty is overcome by blowing
out the spark by an air blast given at just the right place and time.
The manner in which the blast is delivered is as follows: the segments
of the commutator

are separated by gaps of about 5° and in front of each of the leading
brushes there projects a nozzle, Fig. 15, which discharges an air blast
alternately three times in each revolution. The blast itself is supplied
by an ingenious piece of mechanism known as the




                          _Thomson Air Blast_.


It consists of an elliptical box II whose sides have perforations II
where air can enter while inside of this rotates a steel disk keyed to
the armature shaft and having radial slots in which slide three wings
RRR of ebonite which as they fly around drives air into the holes JJ
leading to the nozzles Fig. 15. The result is that, since the spark is
done away with, oil can be supplied to the commutator in limited
quantities but still amply sufficient to reduce the wear on the
commutator to such an extent that the life of a segment is greatly
increased. The air blast is fastened to the dynamo frame just behind the
commutator and can be see in Fig. 23.




                          _The Field-Magnets_,


as may be seen from Fig. 17, consists of two flanged iron tubes _AA_
whose end consists of a convex segment of a sphere accurately turned to
recieve the armature. Coils of wire _CC_ which are in the outside
circuit and through which the entire current flows are wound upon the
tubes. After the armature is placed between them the two tubes are
bolted together by heavy wrought iron bars _BB_ and the whole carried on
the frame work _PN_ shown also at _PN_ Fig. 23. Now a little magnetism
only remains in the wrought iron bars and iron frame works when the
armature first revolves, but the current even though slight, going
through the coils makes an electromagnet out of each tube and heavily
magnetizes the wrought iron bars and in two or three seconds after the
armature first rotates it is entirely surrounded by a heavy magnetic
field. One of the good points of these field magnets is that but very
little magnetism is lost as compared with most other dynamos and since
it takes power to maintain a heavy magnetic field, this dynamo is in
this respect very economical.




                     _The Thomson Regulating Gear_


Later on we will show that pushing the brushes together or pulling them
apart alter the strength of the current, but for the present just accept
the fact and we will show how the brushes are varied. It is accomplished
by the mechanism shown in Fig. 18. The brushes are fixed to the levers
YY and Y_{2}Y_{2} united by the lever _l_. The automatic movement is
obtained by the electromagnet _R_ while a dashpot _J_ prevents too
sudden motion. Suppose the brushes to be in the position shown when the
current would get too strong owing to lights being cut out. The
electromagnet R getting stronger would raise _A_ and reduce the current
taken off until current came to normal. If, instead, some lamps were
thrown in the current would become weak and the electromagnet _R_ would
become weak, drop A which would increase current and this will continue
till current reaches normal.

The foregoing regulating gear is used on small dynamos and old style
large ones. On the large new style dynamo a more delicate regulating
gear is used, the current which operates it being shown at Fig. 19.
Normally the electromagnet _R_ is short circuited by the wire _r_ and
only acts when this circuit is broken. At some point in the main circuit
is a _wall controller_ or _controller magnet_ shown in Fig. 19, at ST,
consisting of two electro magnet, Their yoke supported by a spring and
the yoke operating the contact lever S. If the current becomes too
strong the controller magnet circuit is broken and all the current of
the main circuit goes through the electromagnet _R_ which by its sudden
increase of strength quickly raises _A_ and thus alters the brushes.
This only exists for a moment until the yoke of the controller magets
fall because of their decrease of magnets strength, when current again
flows through wire _r_ because when yoke drops contact is made. This
decreases the strength of electro magnet _R_ thus dropping _A_ and
increasing current. Hence _S_ will again raise and break contact and _R_
again rais _A_. This is continually repeated.




                             _The Brushes_,


of which there are four in use on all machines, are made of a broad
strip of springy copper having six slits two thirds the distance up, and
thus touching at several points. They are held by clamps shown at Fig.
23 which also shows the brushes. The brushes are held to the commutator
by their own springiness and the variation of position due to strength
of current. The brushes are set by a gauge sent with each dynamo which
shows length from the end of brush to the holder. The holders are set at
the correct angle by a gauge of brass of the shape of a right angled
triangle the short side having a wide flange curved to fit the
commutator for which it is sent, while the second side as regards length
must fit to the holder when swung to it on the commutator as an axis.

After describing the details of the dynamo, we will at once proceed to
find how the




                   _Thomson Houston Dynamo Operates._


In the diagram Fig. 19 the rotation is as in practice against the hands
of the watch when seen from the commutator end of shaft. The three coils
of the armature are represented by three lines _A_, _B_, _C_, united at
their inner extremities each being joined to a segment of the
commutator. There are two positive brushes _P_ and _F_ and two negative
ones _P´_ and _F´_. The current delivered to _P_ and _F_ goes round one
of the field magnet coils, then to the outer circuit consisting of
regulating gear, lamps, motors, etc., through the other field magnet
coil to brushes P´ and F´. Now from Fig. 20, we observe that supposing
the loop to be rotating against the hands of the watch in a magnetic
field the diagram represents by arrows the direction of the
electro-motive forces induced in those loops. The action is a maximum
along the line of the resultant magnetic field m m´ and the minimum
along the line n n´ which is at right angles to m m´. The reason that
m m´ is not horizontal is that the induced poles of the armature is in
advance of the poles of the field magnet and is constantly tending to be
drawn back. Applying Fig. 20 to Fig. 19, we see that there will be an
outward current in _B_, an inward one in _C_, _A_ generating no current
for that moment.

Now the following pair of brushes _F F´_ are shifted backward three
times as far as _P P´_ is shifted forwards. When the current is the
greatest possible the brushes P and F and P´ and F´ are 60° apart thus
leaving _P_ and _F´_ and _P´_ and _F´_ just 120° apart and since the
segments of the commutator are each 120° in length[3] there will always
be two coils in parralel with one another and in series with the third.
Taking one sixth of a revolution and continuing all the way round we
find the following tabulated statement showing brushes in contact with
coils, to be true viz:—


                           { P - C }   { P´ }
 From external circuit     {       } B {    }  to external circuit
                           { F - A }   { F´ }

                           { P }    { P´ - B }
  ”      “        ”        {   } A  {        }  ”    “         ”
                           { F }    { F´ - C }

                           { P - A }   { P´ }
  ”      “        ”        {       } C {    }   ”    “         ”
                           { F - B }   { F´ }

                           { P }   { P´ - C }
  ”      “        ”        {   } B {        }   ”    “         ”
                           { F }   { F´ - A }

                           { P - B }   { P´ }
  ”      “        ”        {       } A {    }   ”    “         ”
                           { F - C }   { F  }

                           { P }   { P´ - A }
  ”      “        ”        {   } C {        }   ”    “         ”
                           { F }   { F´ - B }


Now suppose the current to become to strong owing to any cause, the
following brushes are made to recede. This can but shorten the time that
the brushes are in contact with the commutator when the coil is passing
through that position in which it is generating the maximum amount of
current and also hasten the time when it goes into parralel with a
comparatively idle coil. If the current is to weak then the brushes are
made to close up thus reducing the time that the most active coil is in
parralel with one less active and also makes the brushes be longer in
contact with the segment when the coil is generating its maximum amount
of current. The motion of advance and retreat of the brushes is
accomplished by the _Thomson Regulating Gear_ before described. On Fig.
23 can be seen all the dynamo’s details except the _Controller magnet_.

[Illustration:

  _Fig. 18._

  _Fig. 19._

  _Fig. 20._

  _Fig. 21._

  _Fig. 22._]

As regards the _Thomson-Houston Dynamo_ it will be found to produce the
steadiest and most uniform current of any dynamo now in use. It
regulating gear is the simplest and most natural one ever used. In its
ability to reduce the current simaltaneously to one tenth of its former
quantity inside of one or two minutes _without injury to itself and
lamps_ it stands alone, in practice.


                               _Fig. 23._


               Your engraving representing “_Dynamo
               Electric Machine with Thomsons Spherical
               Armature_”

               —Taken from one of your catalogues, and
               pasted on a sheet of this paper—

In a system the most important thing next to the dynamo is the lamps.
The first experimenter who produced an electric glow was Otto von
Guericke. But neither the glow nor electric spark have been used to
produce electric light for practical purposes, this was left to the
voltaic arc on the one hand and the incandescent lamp on the other. Davy
in 1800 mentions experiments in which electric light was obtained by
electric sparks between two carbon points. He showed the arc[4] light
for the first time in 1810 at the Royal Institute, which with Foucalt’s
hand regulator (1844) Deleuil lit the Place de la Concorde, Paris.
Thomas Wright in London (1845) devised the first apparatus which
automatically adjusted the carbons. W. C. Staite used the electric
current for the regulation of the carbons in 1848. In 1855 Serrin
constructed a lamp which would have been used on a large scale had it
not been for the cost of generating electricity. In 1876 Paul
Jablochkoff invented his electric candles and in 1881 there were 4000 in
use, but as their use increased their defects were found out. Regulated
lamps were again brought into use and with them experimenters again
endeavored to solve the problem of dividing the electric light. In 1877
Tschikoliff solved the problem in a very simple manner. He reasoned
that, if the current be divided and part go through the carbons and make
the arc and the rest go through an electromagnet and regulate the arc
and the the current unite and when another light is wanted the current
be again divided and reunited, the current may be divided any number of
times and the scheme work nicely. When put in practice it worked very
nicely and is used on most lamps at present. Suppose there be a lamp
placed in the circuit. The current divides and the larger half goes
through the carbons, as here there is no resistance as the carbons
touche, while the remainder, going through a spiral of high resistance,
is small. When the carbons burn away a little the arc is formed and the
resistance increasing brings the regulating gear into operation. Now the
strength of the current is the same after it has gone through the lamp
as before because the current is going to get through either one way or
the other, hence any number of lamps may go on in series, depending only
upon the tension of the current.

Incandescent lamps were produced as early as 1859 but not till 1879 when
Swan, Edison, Sawyer and others were they ever in a practical form. The
first glow lamp Edison constructed had platinum wire to be heated. He
however examined the properties of organic substances and finally fixed
on bamboo fibre. The bamboo is divided into fibres one millimeter in
diameter and twelve millimeters long. These fibres are pressed in
U-shaped moulds and baked in ovens where they are allowed to become
carbonized. The carbonized filament is attached to platinum wires which
are fused in a glass vessel from which the air has been exhausted. We
will speak more fully of the incandescent lamp when describing the
Thomson Houston System’s incandescent lamp.

The Thomson Arc lamps was used by the Thomson Houston System since its
begining till about two years ago when they stopped manufacturing them,
only furnishing broken parts. The arc lamps at present used is




                      _The Thomson Rice Arc Lamp._


They are manufactured in two styles the single lamp used for stores,
buildings etc., and the double lamp used for street service, all night
work, etc. The light is produced by the voltaic arc between two carbons,
the negative pole or lower carbon burning away about half as fast as the
positive pole or upper carbon. The outside view of the single lamp is
seen in Fig. 21 and of the double lamp in Fig. 22.

The regulation of the double lamp is diagramatically shown in Fig. 24,
which is a plan of the lamp with cover removed, showing only a plan of
cylindrical part of the lamp. The wires marked _a b c d_ run along the
top in order to be out of the way. In Fig. 24 the current comes in at
the binding post and is at _A_ divided into three currents _A_, _B_, and
_C_. The current _a_ goes to the yoke _I_ of the electromagnets _h_ and
_i_ and when the yoke is not held down by magnets _h_ and _i_, it goes
out wire _a_ to binding post _B_. This only continues a moment until the
current _b_ which goes through the carbons and at the start has almost
no resistance offered it, attracts the yoke _I_ thus breaking contact of
curcuit _a_ until the current ceases or till both carbons burn away,
when in the latter case the resistance of _b_ becoming very high as
compared to _j_ and _k_ but little current goes through _h_ and _i_ and
_I_ is raised by a weak spring not shown, thus making contact of circuit
_a_, and since current _a_ has little resistance as compared to _b_ or
_c_ most of the current goes through it, thus practically making a
cut-out. The current _b_ goes round the electro-magnets _h_ and _i_,
then to the “bed” through screw J, the “bed” being a cast iron bottom of
the cylinder _E_ Fig. 22. From the bed it goes down carbon holder _C_
(or _H_) through carbons and arc to frame bed _A_ Fig. 22. From there it
comes up a wire by the side of frame _C_ Fig. 22 and joins other
currents at _B_. The third current _a_ goes through electromagnets _j_
and _k_ and joins other currents at _B_.

This is when switch F Fig. 22 and _M_ Fig. 24, is turned _on_. Now since
the dynamo will regulate all differences in current the lamps can be
turned _on_ or _off_ at will by any one. This is accomplished at the
lamp by turning _off_ the switch. When the switch is turned _off_, the
current goes through _d_ to screw _K_ which is then touched by metal L
(in contact with binding post B and worked by _M_).

It will be perceived that any disorder in a lamp cannot affect other
lamps in the circuit and will right itself or if not the lamp can
immediately be switched out of circuit.

Now as to the regulating gear. The two carbon holders are held up, _H_,
by clutch operated by springs (not shown) till end _N_ of lever _ON_ is
permanently held down, and _C_, by the raising and falling of yoke _D_.
There is only _one_ arc burning at a time in a double lamp and the
so-called positive carbon _C_ burns first. When the lamps are trimmed
the switch is first turned _off_ the carbons put in and the switch
turned _on_. This will draw the upper carbons up about a quarter of an
inch.

When the current is turned on the circuit _aa_ is almost instantly
broken and most of the current goes through _c_ as the distance between
carbons being a quarter of an inch the arc has a _very_ large
resistance. The electromagnets _j_ an _k_ attract _D_ which lets loose
_C_, which falls to lower carbon, and the resistance being almost
nothing, most of the current goes through _b_. This weakens _j_ and _k_
which lets _D_ up while _D_ takes _C_ up with it thus establishing the
arc. The current _all_ goes down _C_ till the enlarged end of _C_
strikes lever _ON_ thus letting _H_ drop and also putting it in
electrical contact with “bd,” which it was not in before.
After a short time the carbons burn away, the arc becomes longer and
establishes itself and the resistance becoming greater in passing from
carbon to carbon and a correspondingly less current flows through _b_
and a greater one through _c_. This makes the electro-magnets _j_ and
_k_ strong enough to draw _D_ to them in spite of spring _Q_. When _D_
is attracted by _j_ and _k_, _C_ (or _H_) falls and again the arc
lengthens, always being kept about 3/32 inch long. This is frequently
and continually repeated, the delicacy depending upon the strengh of the
spring _Q_ as compared to the electromagnets strength.

[Illustration: _Fig. 24._]

When the carbon in carbon holder _C_ burns to a length of about two
inches in attempting to fall to maintain arc’s length, an enlarged port
at the top of the carbon holder _C_ strikes and holds down lever _ON_
pivoted at _O_ (and end _N_ held up by a spring _P_) thus letting loose
a clutch by which electrical contact is made between _H_ and “bed” and
letting _H_ fall till it touches lower carbon when an arc is established
and regulated just as for _C_.

The Thomson-Rice single lamp has the same gear with the exception of
having only carbon holder _C_, _H_, lever _ON_, and spring clutch and
spring _P_ being absent. The single lamp will burn eight hours and the
double lamp fourteen hours continuous running.

These lamps are intended only for a steady current and will not cut out
of circuit if the current gets too strong. But with the Thomson Houston
dynamo the current never gets too strong and because of this there are
less power absorbing mechanism and as anything’s functions decrease the
remaining functions are increasedly better. As the Thomson-Rice lamp has
less functions and power consuming machinery, it can but be the most
economical, delicately adjusted and steadiest lamp extant. They are made
to stand a current of five amperes above the normal current for a short
time, as, when forty lights are simultaneously cut out of a forty-five
light circuit, the current runs up about four amperes above the normal
current for about one half a minute.

Prof. Thomson has gotten out a divided arc lamp which supplies a light
of moderate candle power for locations where a 2000 candle power lamp
gives more light than can be economically utilized. It is specially
suited for factory and mill use where looms or other tall machines are
liable to cast disadvantageous shadows. It is said that these lights are
supplied cheaper per candle power than the standard lamp and up to date
is sucessful.

He has also arranged apparatus by which arc lamps are run in multiple
series, series or multiple arc. It is said that divisions, redivisions
and reunions are practicable. This is also sucessful as far as we can
find out.




                  _The Sawyer-Man Incandescent Lamps._


As before stated Edison fixed upon carbonized filament of bamboo. The
Sawyer-Man company however applied for a patent on carbonized filament
for incandescent lamps on January 19^{th} 1880, and after five years
litigation with Thos. A. Edison they were granted a patent No. 317,676,
on May first, 1885, covering their invention. The Sawyer-Man lamp Fig.
25, consists of a carbonized connected to platinum wires fused in a
glass tube from which all the air possible had been extracted. The light
is produced by the glow of the filament and heat of gases given from
filament. The life of a lamp is from 1000 to 1500 hrs. and requires a
current of 1¼ to 1.3 amperes and give 20 to 25 C.P. When the filament
becomes brittle and breaks the tube is unscrewed from the key Fig. 26,
and a new one screwed in. They are run on the arc light circuit by the
use of an individual distributor Fig. 27 which consists of a brass case
containing a magnet in the circuit of the lamps and a resistance coil
automatically substituted in case the lamp should break or is turned off
by key Fig. 26. The scheme of arranging lamps so as to get the right
current is shown at Fig. 28. the number of lamps in a group depending on
the current.

Prof. Thomson has gotten out a lamp Fig. 29 in two styles one for 6.8
amperes current and one for 10 amperes current. Three lamps of different
candle power, due to different potential differences at binding post of
lamp, are use on the same current. The method of connecting them is
shown in Fig. 30. It will be perceived that the lamps carry the full
current yet have a life of 1000 hrs. or more. This is a great invention
indeed doing away with a great loss of power due to high resistance
coils. It will be noticed however that a 125 C.P. incandescent lamp uses
as much energy as a 2000 C.P. arc light, the 65 C.P. lamp one half as
much and the 32 C.P. lamp one fourth as much.




                           _General Remarks_


The Thomson Houston system also furnish lightning arresters, ammeters,
hanging boards, switchboards, hoods, insulators, lamp arms, etc, but,
though in some respects many of these miscellaneous articles are
ingenious and novel, yet they are not distinctive of the Thomson Houston
or any other system. Be it said however that all these articles fill
their proper places. The company also furnish a motor to go on their
circuits but for the double reason that of the motor not being strictly
related to electric lighting and of being unable to obtain a description
of it, it must remain undescribed as far as this thesis is concerned.

After describing all the parts of the system it may be interesting to
know how a plant is arranged. The last plate is a photograph of the
LaFayette Gas Company’s Plant of the Thomson Houston System taken at ten
oclock one night. It shows the engine, dynamos, the wall controller on
the left wall, and a view of the lamps which had hoods put before them
to prevent the polarization of the negative.

On the accompanying page will be found a table showing experiments with
an old style dynamo given Purdue University by the Thomson Houston
Company, which dynamo is now in the engine house of the Mechanical Hall.




           _Experiments with Three Light T-H. Dynamo No. 79_


The dynamo was run by a large pulley (about four and one half feet in
diameter) on the same shaft as the fly wheel and beside the latter. Two
lamps were put in circuit with a Deprez-Carpentier ammeter and a volt
meter of the same make was put in between the brushes. First one lamp
(old Thomson style) was switched out of circuit, the dynamo started and
when speed was reached the circuit made. The following readings were
taken when the engine made 139 & the dynamo 1122 revolutions per minute.


 +-----------+----------+----------+---------------+----------------+
 |           | One Lamp | Two      | When 2^{nd}   | When 2^{nd}    |
 |           |          | Lamps    | lamp was      | Lamp was       |
 | Readings  |          |          | switched in   | switched out   |
 | At End of +----+-----+----+-----+-------+-------+-------+--------+
 |           |Amp.|Volts|Amp.|Volts| Amp.  | Volts |  Amp. | Volts  |
 |-----------+----+-----+----+-----+-------+-------+-------+--------+
 | 1 second  | 10 | 55  | 10 | 110 |  6.7  |  55   |   14  |  110   |
 | 30  ”     | 10 | 55  | 10 | 110 |  10   |  75   |   10  |   85   |
 | 2 Min     | 10 | 55  | 10 | 110 |  10   | 109   |   10  |   60   |
 | 3   ”     | 10 | 55  | 10 | 110 |  10   | 110   |   10  |   55   |
 +-----------+----+-----+----+-----+-------+-------+-------+--------+


                               _Fig. 25_


          consisting of drawing of Sawyer Man lamp cut from
          catalogue, and trimmed to contour of drawing


                               _Fig. 26_


          a drawing showing action of key in Sawyer Man lamp,
          cut to contour


                               _Fig. 27_


          a drawing of the Thomson Rice Individual distributor
          cut from cataloug and pasted in.


                               _Fig. 28_


          a drawing cut from pamphlet showing “Method of using
          Thomson Rice Individual distributor”


                               _Fig. 29_


          a drawing cut from pamphlet showing “Prof. Thomsons
          incandescent lamp—series incandescent lamp”


                               _Fig. 30_


          Drawing showing “method of using the Series
          Incandescent Lamp manufactured by the
          Thomson-Houston Elec. Co.” cut from your pamphlet
          and pasted on a similar sheet.

                  ------------------------------------


          A photographer of La Fayette photoed the Gas
          Company’s plant of T & H in this city one evening at
          10 o’clock when several lights were burning in room.
          I had a large one printed and pasted on a piece of
          bristol board of the same size as this sheet, and
          put in my original copy.




                              _Footnotes_


-----

Footnote 1:

  The foregoing statement is quoted from Dr Urbitzkany’s work
  “Electricity in the Service of Man.”

Footnote 2:

  Gravity does not enter, as a current is generated in lowering A.

Footnote 3:

  Each segment is really only 115° in length but the brushes are set at
  a distance from the holder far enough to just reach over the five
  degree gap by the gauge above described.

Footnote 4:

  An arc light is a light produced by the use of the voltaic arc, which
  is made by the sparks passing between two poles of a powerful battery
  which are brought together and then seperated a little.

------------------------------------------------------------------------




                          _Transcriber’s Note_


The source for this e-book was a hand-written thesis.

Footnotes have been moved to the end of the book.

The captions for Figures 23, 25, 26, 27, 28, 29 and 30 are reproduced,
however, the original drawings were not bound with the published thesis
and are therefore not part of this e-book.

The author’s spelling has been maintained, Some standardization of
punctuation was done to improve readability.

The following proper names as used by the author are reproduced here
with their more commonly used spelling:

                 Author          Standard

                 Thompson        Thomson
                 Wimhurst        Wimshurst
                 Dr Urbitzkany’s Alfred von Urbanitzsky
                 Lentz           Lenz
                 Pacinnotti      Pacinotti
                 Foucalt’s       Foucault’s

Phrases which the author portrayed as underlined are presented by
surrounding the text with _underscores_. Some standardization of these
was also done particularly with regard to the presentation of
illustration captions.