WA !

12 SI

Ifefory a: Sled Mway Rails

Pivil FnotaePTiiw*

9 9 I

•
OF
]) i,.!,: •
•UAI'.y
 The person charging this material is re-
sponsible for return to the library from
its
which it was withdrawn on or before the
Latest Date stamped below.
Theft, mutilation, and underlining of books are reasons
for disciplinary action and may result in dismissal from
the University.
To renew call Telephone Center, 333-8400
UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN

PUILDING USE: ONLY.

JUW 8 |gg f

ONLY.
BUILDING USE

L161— O-1096
:>
^\ iz^^i^^^ ii^^I^^k t^^i^r^ I^ ^K^^^l# ^i^rs\ l^s^^I^t^^i^S!^
E
 Digitized by the Internet Archive
in 2013

http://archive.org/details/historyofsteelraOOwals
 THE HISTORY OF STEEL
RAILWAY RAILS
BY

WILLIAM JOSEPH WALSH

THESIS
FOR THE

DEGREE OF BACHELOR OF SCIENCE

IN

CIVIL ENGINEERING

m THE
COLLEGE OF ENGINEERING
UNIVERSITY OF ILLINOIS

PRESENTED. JUNE, 1909

wl\6
 UNIVERSITY OF ILLINOIS

June 1, 190 !

THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY

WILLI AM J OSEPH WALSH

.

ENTITLED . THE HISTORY OF STEEL RAILWAY RAILS

IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE

degree of B aehelor of S el ene e in Civil Engineering.

...

Instructor in Charge.

APPROVED:

head of department of (21x11 Engineering.

145020
vvit>

CONTENTS

Introduction -------------page i

Development ------------- 3

Weight ---------------- 5

Length of Rails ----------- 7

Composition ------------- 9

.Durability - -- -- -- -- -- -- - Eo

Growth Steel Industry -.-------25

Conclusion --------------£9
Bibliography ------------- 30
,«"*N
 .

IFT.RODUCTION

The railroad rail has received more study and more careful

attention at the hands of engineers than any other track

material, and it has been greatly improved and its cost cheap-
ened, nevertheless, it is still of vital interest to the

engineer, on account of the wonderful growth of railroads, and

the urgent demands of the traveling public for higher speed.

For the above reasons, as well as for the safe transportation of

passengers and freight, a very efficient and substantial rail

is required. Moreover, there is a great future for the engi-

eera who are able to design and produce rails which will

satisfy the present requirements.

Furthermore, the steel rail is of interest to the world in

general; on account of the large number of people employed in

its manufacture; the necessity for safe transportation of pass-

enger and freight traffic; and because of the vast amount of

commerce which is being "carried on" at the present time.

It is also true that many of the steel rail mills are owned

and operated by the great steel trust. This accounts partly

for the high prices of steel rails, and the excessive tariff

rates demanded by railroads, both for passenger and freight

service

In order to enumerate some of the things to be observed in

the manufacture of 3teel rails, it would be well to begin with

the primary rail and trace its development down to the present

time. In the following pages, then, an attempt will be made

 2

to follow each phase of development from the primary rail down

to the modern steel rail.

 3

DEVELOPMENT

In the development of the steel rail, there has been a con-

stant evolution since its introduction. This was due largely to

the extensive and necessary growth of railroads, and to the de-

mands both for heavier trains and greater speed.

The first rails ever laid were wooden stringers, which

were used on short roads around coal mines. The surface of

these stringers were protected by wrought iron strips. Later,

cast iron tramways were introduced, in which angle shaped rails

were nailed to wooden stringers and fla*ngele3S rails laid on

these. In 1820 the fish-belly rail was introduced and a few

years later the bridge rail. In 1844 the pear-shaped rail was

invented, but it was unsuccessful, owing to the difficulty in

obtaining a good form of joint. The present flange or T rail

section was invented by Colonel Robert L. Stevens in 1830. About

the same time, the bull-head rail was introduced into England

with the idea that, after one head had been worn out, the rail

could be reversed and its life practically doubled. Experience

has shown, however, that wear in the chairs is so great that when

one head has been worn out by traffic, the whole rail is generally

useless. It is also claimed that a track built of this special

type of rail is better for heavy and fast traffic; but is more

expensive to build and maintain. It is, however, the standard

form of track in England and Europe at the present time.

Until a few year3 ago, there was a multiplicity in the

design of T rails in this country. Nearly every prominent rail-
road had its own special design, which perhaps differed from that
of some other road only in very minute and insignificant details,
but which, nevertheless, would require a complete set of new

rolls. This certainly had a very appreciable effect on the cost

of rails. Because of this fact, the American Society of Civil

Engineers, in June 1893, after having obtained the opinions of

the best experts of the country, decided upon a series of rail

section, drawings or sketches of which will be shown in the suc-

ceeding pages. Consideration was given to the manufacturing

details of rail making, as well as to the design of those forms

which would be best adapted to meet the various requirements of

traffic.

To have a smooth track, it is absolutely necessary that

the surface and line of rails shall be without bumps and kinka.

To obtain such rails, it is of importance that the area of the

cross-section in the head and flange shall be as nearly equal

as economy will permit. This allows' the hot metal in the rail

to cool with the least internal stress, so that when it becomes

cold its form is nearly straight. This prevents excessive gagg-

ing in the final straightening. The importance of this increases

with the amount of metal in the section. The quantity of metal

in each part was decided upon -as follows: 42 percent for the head

21 percent for the web; and 37 percent for the flange. The

other constant factors were.- the top radius of head, 12 inches;

top corner radius of head, 5/16 inches; lower corner radius, 1/16

inch; 3ide radius of web, 12 inches; top and bottom radius of web,

l/4 inch; angles at underside of head and top of flange, 13°.

The ohief features of disagreement among railway men relate

 5

to the radius of the upper oorner of the head and to the slope of

the side of the head. For it is agreed that sharp rail comers

wear the wheels and produce sharp flanges which are liable to

cause derailments at switches. Moreover, the tires of engine

wheels are required to be more frequently turned down to their

true form. On the other hand, it is generally believed that

rail wear is much less rapid when the area of contact between

the rail and wheel flange i3 small. When the rail-head has worn

down fas it invariably does) to nearly the same form as the

wheel flange, the entire rail wears away very quickly. The

American Society of Civil Engineers has decided upon standard

rail seotions for the various weights of rails, and the railroads

are gradually adopting them. There are a few other types such

as the Dudley and Sayre sections, but they are closely related to

those of the American Society's standards. (See Fig. 8).

*

WEIGHT

The weight is another important matter in the construction

of rails from a financial standpoint. The weight of primary

wooden stringers, used as rails, was 13 pounds per yard. On

account of the introduction of locomotives, and the increase in

weight of trains, the heavy iron rail was introduced, the first

of which weighed from 40 to 50 pounds per yard. Some years ago,

Mr. Sandberg began a crusade against the policy of using very

light sections of steel rails. He maintained that the constantly

increasing weight of rolling stock and the high speed of trains

demanded a stronger track system. This, however, is not to be

obtained by merely increasing the number of ties, as has been

 too I b
90 lb.
D/STRIBVTfON flfiTAL

Sase 57 y»

„ 5TAHDARD MIL SECTIONS or the

Am 5oc Civil Enqrs.
J i 1

80 lb.

ft* 6

4o ibf -5
 90 lbs.
Fie-" 8 — Sayre Type. Fi»* 31
l
.
—
type of American
Society of Civil Engineers.

K - 4j -
X
7E lbs. 100 lbs.
Fig, , &) .
— band berg Kail Section. MQi 'i'l . — Dudley Type.
Rail Sections.

U ~
5? Pi

Fig. 08 .
— Hunt Rail Section. Frgr-&4. — Rail Section of Great Northern Ry.

90 lbs.

.
—
Bull-head Rail and Cast-
iron Chair; London & North-
western Ry, (England).

n3 E -

r rg.-4. Cross Section of 90-lb. Rail for Chicago &

South Side Elevated Railway.
U»"V
 6

done in some oases, but rather by the adoption of a heavier rail.

In this country, there are still many miles of rails too light

for safe and economical transportation. The tendency of all

roads, however, is toward an increase in weight, rendered neces-

sary by the increase in the weight and capacity of the rolling

stock. Moreover, the price of rails has been so reduced that

it is both better and cheaper to obtain a more solid and durable

track by increasing the weight of the rail, than by attempting to

support a weak rail by an increased number of ties. It should

be remembered that in buying rails the mere weight is, in one

sense, of no importance. The essential things to be considered

are the strength and stiffness. Let us assume that all weights

of rails have similiar cross sections, which is nearly true.

It is also true that in beams of similiar cross-section, the

strength varies as the cube of the homologous dimensions and the

stiffness as the fourth power, while the area (and therefore the

weight per unit of length) varies only as the square. From this,
it follows that the stiffness varies as the square of the weight,

and the strength varies as the 3/2 power. A3 an illustration,

the use of an 80 pound rail instead of a 75 pound one, adds only

6 2/3 per cent to the cost; but 14 percent to the stiffness and

nearly 11 percent to the strength. This shows why heavier rails

are more economical and are being adopted, even when not absolute-

ly needed on account of heavier rolling stock.

The heaviest rail generally used in this country is the

110 pound rail of the Ohegnecto Railway in Canada, but 100 pound

rails are in use on the Hew York Central; Pennsylvania Railway;

 7

Hew York, New Haven and Hartford; Chespeake and Ohio Railway,
and other lines. The minimum economical weight in ordinary-

service is 65 pounds; but many roads have used lighter rails,

even under considerable traffic.

With regard to the weight of rails in the future, they can-

not be increased much without causing inconvenience in handling.

Since it is not likely that much heavier rolling stock will be

introduced, the writer does not expect a very marked increase in

the weight of rails in the next few years. He looks forward to

the time, however, when most of the railroads will be using the

100 pound rail.

LENGTH OF RAILS
Another item of interest in the history of rails is the

standard lengths that have been adopted from time to time. The

primary rail- -the wooden stringer- -was IS feet long. The oast

iron rails introduced later were only 3 feet in length. Still

later, the 18- foot wrought iron rail was brought into use. In

1855, when steel rails were introduced, the standard length was

30 feet. Just why 30 feet came to be universally adopted as the

standard length is not a matter of record, but it may readily be

surmised that convenience of transportation had considerable to

do with it. For many years the ordinary length of flat cars

was such that a rail longer than 30 feet could not be conveniently

handled thereon in shipment; and it is also probable that in the

early rolling mills, a sufficient quantity of metal for making a

rail longer than 30 feet could not be handled in a single ingot.

However, such difficulties no longer exist, but the 30- foot rail
 .

remains the standard.

In recent years, many roads have been trying the 45-and 60-

foot rails. The argument in favor of longer rails, is chiefly

that of reduction in rail joints, whioh are costly to construct

and maintain, "besides being a fruitful source of accidents. Mr.

Morrison of the Lehigh Valley Railroad, in his report to the

Roadmasters Association of 1895, declares that as a result of

extensive experience with 45- foot rails on that road he finds

that they are very much less expensive to handle, on account of

their length, and can be laid around sharp curves without being

curved in a machine as is necessary with the shorter rails. The

great objection to longer rails lies in the difficulty experien-

ced in allowing for expansion, and the inconvenience in transport-

ing and handling them. Both the Pennsylvania and the Norfolk and

Western Railroads have a great many miles of 60- foot rails. Al-

most all the street railways have adopted this length, as there

is no difficulty encountered in allowing for expansion, the rails

being welded together and made continuous.

Rails 33 feet long are extensively used on many roads, and

an increase in the standard length to 33 feet has been recommended

by the Roadmasters and Maintenance of Way Association, and by the

American Railway Engineering and Maintenance of Way Association.

Rails longer than 33 feet have been used on comparatively few

roads

In Germany, the old standard length for rails was 29 l/2

feet, and the experience derived from increasing the length seems

to have resulted more satisfactorily than has been the case in

this country. An increase in 33 1/2 percent, or to 39 feet 4

inches, seems to have met with general approval, whereas on some

roads the standard length has been increased to 59 feet.

Until some method which will allow for expansion is found,

there will probably be no increase in the standard length adopted

in the future. Moreover, it is rather difficult for a few men

to replace 45 or 65 pound rails, because it is often necessary

to haul rails more than a quarter of a mile on a hand car, which

is very inconvenient and dangerous. It is more than likely,

however, that the 33-foot rail will be adopted as the standard,

since it is less expensive than the shorter rails.

COMPOSITION

As has been stated before, there has been a marked change

in the composition of rails. The first rails were composed of

wood, usually a high grade of oak or pine. At this time, timber

was plentiful and hence the best was used. With reference to the

composition of oast and wrought iron rails, little is known. For-

merly little attention was paid to the chemical composition of

the rail. Early rails, which wore remarkably well, showed on

analysis a wide variation in chemical constituents. Carbon

ranged from .£4 to .70 percent and phosphorus . ud to .16 percent.

The mill processes, however, were performed slowly and carefully.

The rail heads were small, and the rails were worked at so low a

temperature that they came out hard and tough. It mattered little,
wiierefore, about the carbon or the other chemical constituents.

As time passed, however, the demand for rails increased, prices

fell, and the mill outputs grew rapidly. For this reason, the

manufacturers became less careful. Rails were furnished hotter,

 .

10

and were of an inferior quality, and then the chemistry of the

rail assumed more importance.

About 1818 Dr. C. B. Dudley, chief chemist of the Pennsyl-

vania Railroad, announced that soft rails were best. His con-

clusions were based on £5 samples of rail, taken out of the

track, and his critics maintained that this was too small a

number to serve as a basis for a general law. In 1880, he came

forward with a second paper, a long, carefully written and in-

genious report to the superintendent of motive power. In this,

he based his conclusions on 64 samples, and he advocated the

same theory as before, namely, that rails low in carbon were

the best. The practical outcome was that this proposi tion was

accepted and supported for years. He recommended to the Penn-

sylvania Company a specification regulating the chemical con-

stituents as follows • carbon .25 percent; phosphorus .10 percent

or less; silicon .40 percent and manganese .30 to .40 percent.

It is doubtful if carbon is ever placed as low as .30 percent,

The Dudley specification was succeeded by one requiring .30 to

.40 percent of carbon and giving a wider range to the chemical

constituents

But the Dudley specification made people timid about a high

carbon rail. The theory, however, was not born out by exper-

ience, nor was it demonstrated by Dr. Dudley's figures. The

rails, which he considered, were not of high carbon content.

Of the 64 samples, only one had over .60 percent of carbon, and

only four contained more than .50 percent. But today, 50 per-

cent of carbon is regarded as only a little above the average.

 11

More carbon is usually specified and a rail having only .45

percent would not be ranked as hard; while twenty years ago a

carbon rail of .50 peroent was considered ample in the United

States. In French practice, it would have been considered

low. For the French consider no rail hard unless it has more

than .50 percent of carbon, and they often use 1.00 peroent.
r
.Ir. Conard, a French engineer, found from a study of rails

turned out by various French mills that their composition aver-

aged as follows: carbon .86 percent; manganese .69 percent;

silicon .15 peroent; and phospherus .05 percent. He compared

these with a number of rails manufactured by five German mills

averaging .31 peroent of carbon and .09 percent of phosphorus,

and found that the French rails wore about twice as long as the

German rails.

In the present rails, carbon ranges proportionately as

follows.* 60 pound rails .40 to .45 percent; 70 pound .45 to .50

percent; and 80 pound .45 to .60 percent. The maximum is

.65 to .75 as specified for the 10O pound rail of the New York

Central Railway, but it is rarely, if ever, that .70 percent is

exceeded. The object of carbon is to give hardness to the

steel, but it is liable to render it brittle unless special

care is taken in proportioning the other chemical constituents.

The analysis of 32 specimens on the Chicago, Milwaukee and

3t . Paul Railroad shows the proportion of chemical composition

as follows:

Carbon l.Zll to 0.52 percent;

Silicon (
0. 13 to 0.256 " ;

Phospncrus 0.055 to 0.181 percent:

>
 12

Manganese 0.3 5 to 1.63 percent.

}

Carton up to one percent increases hardness and tensile

strength but decreases the ductility and toughness. It was

formerly used in small proportions, but in late years one tend-

ency is for more carbon to increase the hardness which is re-

quired to meet the increased wheel pressure.

Jlanganese is necessary to take up the oxides of iron while

the metal is in the molten state. In sufficient quantity, it

assists toward a more uniform distribution of the carbon through

the iron. It also facilitates the chemical combinations of

the iron at high temperatures, and tends to prevent separation

into graphite as the iron cools. If not used to excess, it

imparts strength and toughness. If the iron is low in carbon,

the effect of manganese is similar to that of carbon alone; but

diminishes the ductility in less degree. It may, therefore,

replace the carbon to some extent, and is considered an effective

antidote for sulphur. When used in high percentage, it has

the effect of making the rail hard and coarsely crystalline,

and its tendency is to brittleness when it exists in unnecessary

quantity. We, therefore, see how important this constituent

has become. Its use varies according to circumstances from

.70 to 1.4 percent.

Silicon is another useful ingredient and it is gradually

becoming recognized as an important element. It acts as a flux,

and like manganese tends to prevent injury by oxidation of the

iron. When of just sufficient quantity, it gives added tough-

ness, but any increase beyond this tends to produce brittlenes^.

It has a hardening effect, and to a limiting extent may replace

 13

carbon. Its present use ranges from .10 to .20 percent.

Sulphur and phosphorus are both objectionable elements, but

are difficult to extract from the metal. They perform no use-

ful service in combination which is not secured by other elements

Their quantity, however, must be kept at a minimum; sulphur not

to exceed .07 percent, and phospherus not to exceed .085 percent.

R^ils which have a high percent of either should not contain a

large amount of carbon. Prior to the last ten or fifteen years,

these constituents have received little attention, only recently

has it been advocated that a small amount of these adds to the

strength of the metal.

Traces of copper have been found in rail steel, and it

sometimes runs as high as 0.8 percent, altho specifications do

not usually require it or place limitations upon the percentage

used. In small quantities, it slightly raises the tenacity

and elastic limit without tendency to brittleness, but reduces

the toughness. This effect, however, is not pronounced when

the quantity is small.

Chemical specifications are at present considered only an

approximate guide, because much depends upon the mechanical and

heat treatment of the metal during the process of manufacture.

In late years, chemical specifications have not been considered

as important as they once were. More stress is now being laid

upon the production of rails to meet certain tests of strength

and stiffness. There is more attention paid to a guarantee of

serviceability for a stated period than to the chemical compos-

ition. It has become the custom of many roads to leave the

 14

cnemical composition within wide limits, or entirely to the

discretion of the manufacturer.

The process of manufacture of steel rails is becoming more

important. The first iron rails were made from straight puddled

"bars. These were about one inch thick, and were placed one

unon another until a pile of sufficient weight and height was

formed. The pile was then reheated and rolled into rails, and

it was to* the formation of that pile that inventive genius was

applied.

At one time, a rail with a puddled steel head - or rather

with the top bar of the pile of puddled steel - found much favor,

but owing to the difficulty of obtaining unif oranL j good welc. .

,

the results were not satisfactory.

Another nlan, upon which much money was spent, was to ham-

mer a puddled ball or to weld two puddled balls together under

a steam hammer. These were then drawn into a slab two to two

and one-half inches thick, which was placed on the top of the
'

rail pile. The Pennsylvania Railroad ordered several thousand

tons of these rails, but their service was somewhat disappoint-

ing and their use was abandoned.

Luter the railroads adopted the system of having the old

rails rerolled into new ones. At first a certain percent of

new iron was specified, but as the necessity for immediate

economy increased, that demand. was no longer made and new rails
were composed entirely of old ones.

From the many reworkings of rails , the cheapening of the

process of manufacture, and the necessary demands of traffic,

the ability of the iron rails to resist wear became more and
 15

more unsatisfactory. It then seemed that from this cause e"i one

the limit of railway development had been reached. The solu-

tion of the difficulty was the introduction of the Bessemer process.

*,7e all realize that without such an innovation as Bessemer' s,

the subsequent tremendous expansion in railway development would

have been physically impossible. The first rail ever made by

the process was placed on the Midland Railroad of England in

1857 and remained until 18 73, some sixteen years, during which

time about S, 250, 000 trains passed over it.

The material essential to anything like a comprehensive

description of the various processes of manufacture of steel

rails would swell convenient limits of space in this work; hence

only the changes which have taken place from time to time will

be mentioned.

The first Bessemer steel was produced by an experimental

plant at Wyandotte, Michigan, in 1864. It was first introduced

by A. L. Halley, Its invention led to a greatly increased

facility of manufacture and a consequent decrease in the cost of

steel rails. This process with Halley' s improvements and the

invention of the Siemens-?ilartin method of manufacture, and the

consequent introduction of rails at reasonable prices, were great

factors in the enormous railway development of the last £5 years.

The cheapness and rapidity of manufacture, however, have to some

•extent received more attention than the composition.

In 1867, George Fritz, then chief engineer of the Cambria

Iron Company, Johnson, Pennsylvania, invented the blooming mill

on which steel ingots are rolled instead of reducing them by

hammering, as had "been the practice. Mr. Fritz "built his first

regular three-high blooming mill in 18 71. This departure from

the old practice (added to Halley's modified converting plants)

greatly helped to increase production.

During the Civil war there was a great demand for cheap

steel. An answer to this universal demand was made "by the in-

ventive genius of two men - William Kelly, an Irish- American of

Pittsburgh, and Sir Henry Bessemer an Englishman of London.

They devised a new way to refine iron, which has since been

known as the Bessemer process. Their discovery was an entirely

new idea, and one which at first seemed absurd to every other

steel-maker. Within a few years, however, it was universally

adopted, revolutionizing the iron and steel trade and providing

the world with a cheap and abandunt supply of its most useful

metal.

In 1851 the first converter was built - a square brick

structure, four feet high, with a cylindrical chamber. •

The

bottom was perforated for the blast. The greatest difficulty

was to obtain a sufficiently strong blast.

The second converter was made with holes in the sides, and

after this, one improvement followed another. Kelly built

seven converters in a backwoods hiding place and his eighth, a

telling converter ; was built in the Cambria Iron Works. With

this converter, he showed that cold air does not chill molten

iron but refines it with amazing rapidity if blown through it

for the proper length of time.

The new process was perfected by a third inventor, Robert

Mushet, who solved the problem of how to leave just enough carbon
 jVQS

ioo mk&0

. Jucf noii

W9n orlT

Job oriw
 17

in the molten steel to harden it. Instead of trying to stop

the process at the right moment, Mtishet asked; "Why not first

turn out all the carbon, and then pour back the exact quantity

you need?" This was a simple device but no one had thought •

of it before. Since then, other improvements have been added

by Halley, W. R. Jones, Reese, Gilchrist ; and Thomas.

The new metal was 30on called "Bessemer steel." Strictly

speaking it was not steel but much like wrought iron. It was

not hard enough to serve for all purposes.

The modern converter is a huge iron pot about twelve to

fourteen feet high. It swings on an axle so that it may be

tilted up and down. About 30,000 pounds of molten iron are

poured into it, and a strong blast of air is then forced through

the perforated botton, 7hieh rushes through the metal with great

rapidity. Then yellow sparks are sent out at the top many feet

into the air. The converter roars like a volcano in eruption.

The impurities in the iron - phospherus , silicon, sulphur and

carbon - are separated from the metal. The sparks change from

red to yellow; and then suddenly become white. At this juncture,

the converter is tilted sideways. A workman now charges it

with several hundred pounds of carbon mixture to restore the

necessary elements that have been blown out. It is then tilted

further, and the molten iron is poured into a swinging ladle and

from this into a train of huge clay pots, pushed into place

by a little locomotive.

Another improvement in the manufacture of steel rails was

the mechanical appliances added to the rolls by Robert Hunt.

Until Ttfarch 1884, all American rails were fed by the use of
 18

hooka and tongs. Three high trains required from fifteen to

seventeen men for a production of 300 tons in twelve hours,

numerous inventors up to that time had sought to accomplish

this work by machinery which would be automatic in its action,

but none had been actually built. In 1884 tables were driven

in front of the finishing rolls of the rail train of the Albany

and Rensselaer Iron and Steel Company of Troy, New York. This

worked so well that an automatic arrangement was put in front

of the roughing rolls.

It was not until 1899 that the commercial manufacture of

rails from basic open hearth steel was introduced. Aside

from their comparative merits, the quuntity of acid and basio

open hearth rails rolled in the United States in 190f is small

as compared with the Bessemer tonnage. There have been two

attempts to make steel in America by the bj.sie Bessemer process.

Eoth were technically successful, but owing to the character

and oost of the obtainable iron, failed commercially.

T£ost of the present rails are made by the Bessemer process.

It is true, however, that a large number of rails fail in service

and it is probable that the number of broken rails would be con-

siderably reduced if they were made of open-hearth steel.

The question, therefore, arises why more rails are not

made of this material, and railway engineers occasionally come

forward with inquiries to that end. It may be well to say,

therefore, that the making of open-hearth rails is purely a

commercial question and involves immense sums of money. 'lost

of the rails made in America today are made by the Bessemer pro-

, cess, and each rail making plant must be regarded as a unit.

7
 19

The converting department is one factor, its capacity and whole

Bcheme of operation being designed for the one purpose of sup-

plying the blooming mill with just the right quantity of ingots,

at just exactly the right size. It may happen that at a given

rail making works there is no open- hearth furnace plant at all.

Many are aware of the fact that small lots of open-hearth

rails may be made, but their production on a large scale means

a plant laid out with that end in view. Moreover, if this

plant does not have a regular line of business extending over

many years at an inoreased price, it will be a loosing venture.

Within the last few years it has been clearly shown that a great

improvement may be made by certain modes of heat treatment.

Much care is now taken to finish the rails colder than formerly,

and to do a great deal of work on them while they are at a mod-

erately low heat. By so doing, a much better grain is attained,

and this renders possible the use of a higher content of carbon

than was formerly thought advisable. This question of finishing

temperature and all the associated problems of wear and toughness

are being thoroughly threshed out, and it may be well to await

the results of experiments now under way before starting out

into untried fields. The open hearth rails seem to be of better

Quality than those made from Bessemer steel, but the future

rails are likely to be made by the Bessemer' process owing to the

reasons given above.

The hardening and toughening effect of alloying steel with

nickel, so successfully practiced in the manufacture of armour

plate, has naturally suggested a like treatment for rail steel.

Such experiments are now being conducted on a small scale with

i
 .

steel made both by the Bessemer' and open hearth processes. On

the Pennsylvania lines, some nickel- steel rails were laid on a

five degree curve, and after four years of service were said to

be wearing better than rails of ordinary steel. Experimental

use of nickel- steel rails for six years seems to indicate that

rails of such metals will out wear three or four sets of erarbon-

steel rails. Nickel- 3teel rails laid in a freight track of the

Pennsylvania Railroad at Kilanning, Pennsylvania, lasted three

years, as against 10 months for ordinary steel rails.

In addition to the roads ~ientioned, the following are ex-

perimenting with nickei -steel rails; Bessemer and Lake Erie

Railroad, the tinier Railroad, and the New York Central and

Hudson River Railroad.

The chemical composition of these rail3 was founr 1

to "be as

follows: carbon, .418 percent; silicon .102 percent; manganese,

.79 percent; phosphorus, ,094 percent; and nickel, 3.38 percent.

Some frogs and switch points of the s^rae steel are also under

t ri al

DURABILITY
An investigation was begun some time ago at the test depart-

ment of the Philadelphia and Reading Railway to determine the

elements in steel rails which caused fracture or relatively

rapid wear; and to work out the means necessary to reduce these

to a minimum.

A considerable number of rail-sections which were fractured

in track or which were removed owing to very rapid wear were

examined in order to gain the desired information. A consider-

 :

able number of rails which gave good results in service hi ve

also been tested in the hope of finding fundamental character-

istics which might acoount for the durability or the failure,

as the case might be. Also in the regular rail inspections

at the mills, sections representing the various methods of manu-

facture have been taken, and particular attention has been given

to investigation of the differences between rails which failed .

under the drop test and those which passed it successfully.

It was believed then, as it is now, that a proper chemical

composition is one of the essentials for obtaining the greatest

durability in steel rails. This theory was confirmed by other

investigators, and they proved that physical character and struct-

ure have at least an equal influence upon the final outcome.

Up to the present time, about 200 defective rails have been

examined. In some cases complete chemical analysis have been

made; and in others the loose, coarse grained fracture or other

physical character - such as piping - showed the cause at a

glance; and in still others, a rather elaborate investigation

was necessary to prove the matter to a certainty.

In a general way, the results of analysis in this investig-

ation have merely confirmed the previous opinion, and have proved

beyond question that specifying chemical composition alone in-

sures neither a durable nor inefficient rail. From the start,

a marked difference in structures was found, and the following

characteristics in the rails which gave defective service were

noted

(a) Coarse, regular, granular struoture; and

(b) Excess of foreign matter, such a3 oxides, slag and

i

I
 .

22

enolosed gases. Either of these resulted in relatively poor

service

On the other hand, in rails of the same general composition

giving satisfactory service, a generally fine interlocking

"broken granular structure with relative freedom from foreign

matter and gas was found. By a comparison of the above results

with those obtained in other mill inspections, complete accord-

ance was found. The rails similiar to those of the first type

proved exceedingly fragile under the drop test of 20C0 pounds,

falling 20 feet. In contrast to this, rails of the latter

class showed a marked toughness under the drop test, and one

stood 14 blows of the drop test without fracture, being turned

after the first blows, ~nd succeeding odd numbers.

These marked differences confirm the work of others. Mr.

J. E. Stead for instance saya, "It is clear that the junctions

(of the grains) are a safe guard, and the more junctions there

are, the more reliable will the steel be." And again, "it would
seem to follow, then, that the smaller and finer the grain the

safer- the structure." This coincides with the researches of

Brinell ,
Sauveur, and others.

It is a common saying that the wearing qualities of rails

made during the late years, particularly the rails of heavier

section, have been disappointing; that they do not compare with

the service obtained from the 5C- )Ouna rail rolled about 1880.

The reasons explanatory of this, experience have been discussed

until the situation is quite generally understood. Making due

allowance for the effect of the largely increased wheel loads

and train speeds, the currently accepted views, of the situation

 23

may be summed up "briefly in the following statement. Competit-

ion and the desire of the railway oompanies to purchase rails

at the lowest possible price have forced the manufacturers to

resort to quicker and cheaper methods of handling the metal and

to cheapen the cost in the process of rolling. The result was

that rails were finished at too high a temperature to obtain

the benefits of the rolling action on the steel, or the "working"

of the steel as it is called. In course of time the competition

largely disappeared, and the manufacturers fixed their own price

for rails, but the quality of the metal was not improved.

Foreign matter in the rails decreases its durability. As

to the origin of these impurities in steel, it is evident that

they must be due to defective mill practice, either in the manu-

facture of the steel, including teeming and settling; in burning

of steel in soaking pits, or in blooming furnaces 11 the latter

are used, or in cropping blooms or rails. In any event' the

short life of those rails is an unquestioned proof of defective

mill practice.

In order to insure the most durable rail of a given composit-

ion, the above investigations indicate that there must be,-

(1) Freedom from brittleness,

(2) Absence of unsoundne

(3) Fine granular 3tru.ctu.re.

Chemical and mechanical tests are both necessary for a

thorough determination of the value of the rail. Until the

introduction of the steel rail, there was little, if any, test-

ing made upon rails. The chemical test has for its main object

the determination of those minute quantities of chemical elements

 24

which have such a marked influence on the rail for good or bad.

The mechanical tests consist of the usual tests for elastic

limit and ultimate strength. The elongation at rupture is

determined from pieces cut out of the rails. One of the chief

tests used is the "drop test". This consists in dropping a

weight of 2000 pounds from a height of 16 to 20 feet upon the

centre of a rail, which is supported on abutments, placed 3 or

4 feet apart. The number of blows required to produce rupture,

or to produce a permanent set of specified magnitude, gives a

measure of the strength and toughness of the rail. in order

to pass the test, the butt is supposed to withstand one blow

from the falling weight without breaking. The test also

requires that the rails shall not bend more than 6 inches, and

upon revcraal must stand, without breaking, a blow on the convex

side from the 2000 pound weight falling through half the standard

height, or 8 feet to 10 feet.

In this country the testing of rail metal for tensile

strength is not in general praotice but test pieces from rail

steel of good quality will usually show an elastic limit of

55,000 to 65,000 pounds per square inch, and an ultimate strength

of 110,000 to 120,000 pounds at breaking. The test piece should

have an elongation of 12 to 15 percent in a specified length

usually 8 or 10 inches and a modulus of elasticity of 29,000,000

to 30,000,000 pounds.

In this country, moreover, no tests for hardness are im-

posed, but the physical hardness may be quite closely determined

from the chemical hardness, namely by the determination of the

amount of carbon and phospnorus present. Some attempts at

 25

measuring degrees of hardness have been made by an observation

of the indentations on the rail of loaded knife edges of hard-

ened steel; but the use of such tests and others of a different
character does not seem to have passed the realm of experimenta-

tion.

There was a practice at one time, prevalent among manufac-

turers of rails, by which the makers guaranteed to replace all

worn or broken rails that had to be renewed within a certain

period (usually five years). Ho such guarantees are now given,

and at most the makers agree to replace broken rails which show

actual flaws. It is generally recognized that methods of

manufacture should be left largely to the discretion of the

maker, the rails being carefully inspected and tested on behalf

of the purchaser. Under present conditions, there is practically

no control of the manufacture; the mills very generally decline •

to make rails fill filling the requirements of the railways, but

furnish those made according to the specifications adopted by

the manufacturers.

In the future, the writer expects more control of the

manufacture of rails by the maker, but more severe tests on the

part of the purchaser.

GROWTH STEEL INDUSTRY

In June 1876 there were ten rail mills in operation and an

eleventh nearly ready to start. At that time one Bessemer

Company had already become bankrupt; and two of these companies

and their works have absolutely gone out of existance. In fact,

but five of these are now making rails. Since 1876, in addition
 26

to those already mentioned, 18 oorporations have erected mills

to roll standard weight steel rails, 14 from steel of their own

manufacture and four from purchased blooms. Seven of the

steel producers are now making rails, two are on other products,

and the remaining nine have gone out of existance, so that there

are now in the United States 10 corporations running 13 rail

mills. The above refer only to the mills that are rolling rails

of 60 pound or more to the yard. Three of the companies are

controlled by one corporation, three others by another, and

two others by still another; thus leaving two single and independ

ent concerns. In addition, the building of another mill of

large capacity to roll basic open hearth rails, together with

the required blast furnaces, steel furnaces, and town, has just

been completed.

In June 1876, the Bessemer plant of the north Chicago

Rolling Mill Co., built from Halley's plans, then in charge of

Robert Forsyth, held the record for a month's production of in-

gots at 6457 gross tons. The production in the whole United

States in 1876 was 469,639 gross tons of Bessemer ingots, from

which 368,299 gross tons of rails were made.

There was a constant increase in output until 1887, when

the production of ingots reached 2,936,035 tons, and of rails

2,101,904 tons. It was not until 1899 that the production

again passed the 2 , 000, Ou^. -ton point. In that year there were

turned out 7,586,354 tons of Bessemer, 2,947,316 tons of open

hearth ingots and 2,270,585 tons of rails all of Bessemer steel.

Following that year, there was a continued increase in the out-

put of open-hearth steel, while that of Bessemer remained more

 27

nearly constant. In 1905, 10,919,272 gross tons of Bessemer

steel, 8,444,836 gross tons of open hearth ingots, and 3,375,611

gross tons of rails were made, 183,264 tons of which were of

basio open hearth steel.

The north Chicago Rolling Hill Company, which held the

monthly record for product of Bessemer ingots in 1876, built in

1882, an entirely new Bessemer and rail plant at South Chicago,

some fifteen miles away from its old one. There have been

changes in the management and additions to the plant since then,

but fundamentally, the converting works and mills are the same,

and their record production is 91,424 gross tons of ingots and

71,424 gross tons of rails in a month. All of these rails were

rolled on one rail mill.

In 1886, the rail mill of the Edgar Thompson works had been

doing very commendable work, but it was being pressed in output

by other mill3. A new mill was started in 1S88 and all promises
for it have been much more than fulfilled. Its greatest pro-

duction of rails for a month is 61,033 gross tons. There are

three sets of rolls in this mill, which are plaoed in a tandem

formation. The converting works have made 102,740 gross tons

of ingots in a month, all of which were reduced to blooms on

one three- high blooming mill. After the starting of the new

mill the original Edgar Thompson rail mill remained idle for

several years. It was then remodeled and used for the product-
ion of rails which were under 60 pounds per yard in weight.

This mill has been in constant operation ever since.

In this connection, it would be well to give some statistics

in regard to the production of pig iron. In 1876 the production

 28

of pig iron of all kinds in the United States was 1,868,961 gross

tons. In 1905, it was 22,992,380 gross tons. The production

of Be3seraer pig iron was separated statistically from other pig

iron until 1887. In that year it was 2,875,462 gross ton?;

aid in 1905 it was 12,279,462 gross tons. The production of

basic pig iron was first ascertained in 1886 when it was 336,403
gross tons. In 1905, it was 4,105,179 gross tons, charcoal
basic pig iron not being considered in either case.

The writer does not look for a more marked increase in

the output of pig iron and steel in the future. In fact it

seems to have reached its limit. The sources of iron ore are

becoming exhausted and new sources are rare. IJ^ny of the new

ore deposits contain sulphur and the ore is mostly iron pyrites

which is very costly to purify.

The steel industry is of vast importance to us because of

its employment of so many people. The Illinois Steel Company

of Chicago, which is a comparatively small steel industry, employs

more than 8,000 men. But, owing to the recent panic, the work-

ing force was reduced by several thousand.

Host of the steel concerns in this country are owned and

operated by one of the larger companies, the Carnegie or Cambria

steel corporation. We are led to think that perhaps almost

all are under the control of the steel trust. Hence this should

be of much interest to the public. Of course, these combines

have their advantages as well as their disadvantages but in

the opinion of the writer they should be under the direct con-

trol of the people.

 -

conclusion

In conclusion, the author ventures to say that for many

years to come, the standard rail for the United States and for

most of the civilized world will he about such a rail as we are

now using. The section will not differ greatly from that which

has already been decided upon; that is, the Du dley- Hawks -Bint
'
orican Society of Civil Engineers section. The average weight

will very likely be more tnan it is now, and the steel harder.

The writer should not oe surprised to see 100 pounds per yard

common, 80 pounds the average weight, and 0.60 percent the aver-

age of carbon content. This rail may be 60 feet long, and

possibly 90. By making it 60 feet long, we eliminate half of

the rail joints - the most troublesome item of a railroad track.

In fact, the Pennsylvania Railroad, the Norfolk and Western,

and some other roads are today putting in many 60 -foot rails.

In all that has been said in the preceding articles, the

subject has been merely touched upon. There is much more to

be said regarding steel rails and their development but the

writer ha3 aimed to give a clear and brief account of the history

down to the present time.

The writer agrees with T

.:r. prout when he says, "If the

young engineer will candidly and courageously, modestly and

without pedantry, study the scrap heap, I think we can predict

for him a reasonable measure of success. Whether or not he

succeeds in getting money or distinction, that way of meeting

the problems of life will secure the respect and esteem of the

best men about him, which, after all is the highest success
 .

success that can be hoped for."

BIBLIOGRAPHY
Trat man's Railway Track and Track Work.

Camps' Uotes on Track.

Webb's Railroad Construction

History of Bessemer Process.

Magazines and periodicals.

Scientific American Supplement 832, Dec. IE, 1891.

Engineering Hews Dec. IE, .1891.

" Oct. 27, 189E.
" Jan. IE, 1893.

" Aug. 31, 1893.

" Oct. IE, 1893.
11
Nov. 2 , 1893.
" April E, 189S.
" Aug. 17 & 31, 1893.

" Feb. 27, 189E.

" " Dec. 1, 189E.

" April 19, 1894.

" Aug. 23, 1890.

" " Nov. 7, 188 5.

" Vol. 1, 1898.

" " " 1, 189 6.

1 & 2, 1893.

1, 1894.
" 2, 1899.

2, 1900.
 ,

Railroad Gazette April 1, 189E.

N
it
Aug. 17, 1894.
M * 11
July 27, 1894.
It It
Aug . 16, 1896.
n
!i
Oct. 18, 1890.
M H
Nov . 6 1891.
11 11
Aug. 10, 1885.
VI it
Sept. 7, 1888.
m -
ii
July 1 3, 1891.
M 11
1 ftp n

11
11
Feb. 28, 1890.
11 11
Aug. 17, 1894.

Street Railway Review, November 189 5.

Journal Assn. Engineering Sooiety Vol. 15, Ho. 4.

Street Railway Journal, January 189 5.

Journal Assn. Engineering Society, Feburary 1893.

Street Railway Review, Fov. 1, 1892.

Trans. American Society of Civil Engineers, June 1893.

11
" " " " " Vol. III.

April 1889 , Vol.

Vcl. 10, 1881.

Vol. 3.

Iron Age, Oct. 1, 1885.

Railway World, Oct. II, 1890.

Engineering Magazine, July 1897.

Aug. 189 7.

Iron Age, Jan. 2, 1896.

Railway Engineering, Hoveriber 1894.

Street Railway Journal, January 189 5.

n