(ransportation
Library
TE
258
22
Kottay
SANDBERG
ON THE
MANUFACTURE AND WEAR OF RAILS.

***

:
!

}
1
1

Mud Humble Bed Wayn
THE
Parkienk
MANUFACTURE AND WEAR

OF
RAILS.
TRANSPE
UNIVERSITY OF MICHIGAN
APR 1 5 1927
PORTATION LIBRAR
BY
CHRISTER PETER SANDBERG, Assoc. INST
WITH AN ABSTRACT OF THE DISCUSSION UPON THE PAPER.
EDITED BY
JAMES FORREST, Assoc. INST. C.E.,
SECRETARY.
By permission of the Council.
Excerpt Minutes of Proceedings of The Institution of Civil Engineers.
Vol. xxvii. Session 1867-8.
LONDON: PRINTED BY WILLIAM CLOWES AND SONS,
STAMFORD STREET AND CHARING CROSS.
1869.
[The right of Pullication and of Translation is reserved.]
ADVERTISEMENT.
THE Institution is not, as a body, responsible for the facts and
opinions advanced in the following pages.
Transportation
Library
TE
isto
258
.522
THE MANUFACTURE AND WEAR OF RAILS.
THE INSTITUTION OF CIVIL ENGINEERS.
March 3, 1868.
CHARLES HUTTON GREGORY, President,
in the Chair.
The following Candidates were balloted for and duly elected:-
WILLIAM DOUGLASS, Lieut.-Colonel JOHN PITT KENNEDY, RICHARD
LAYBOURNE, WILLIAM HENRY EDWARD NAPIER, and JOSIAH
TIMMIS SMITH, as Members; JAMES BINFIELD BIRD, WILLIAM
GAMMON, THOMAS BLAND GARLAND, WILLIAM HUNT, JAMES TRUB-
SHAW JOHNSON, Lieut.-Colonel WILLIAM LAWTIE MORRISON, R.E.,
PATRICK OGILVIE, ROBERT SABINE, JAMES SHAW, SAMUEL SWAR-
BRICK, and ALBERT VICKERS, as Associates.
It was announced that the following Candidates, having been
duly recommended, had been admitted by the Council, under the pro-
visions of Sect. IV. of the Bye-laws, as Students of the Institution:
BENJAMIN HALL BLYTH, M.A., ALFRED GEORGE BROOKES, RICHARD
ERNEST BROUNGER, WILLIAM AUGUSTUS KENNEDY GOSTLING,
ARTHUR HENRY HEATH, CHARLES ASSHETON WHATELY POWNALL,
PERCY FREDERICK TARBUTT, and FREDERICK TOPLIS.
No. 1,196." The Manufacture and Wear of Rails." By CHRISTER
PETER SANDBERG, Assoc. Inst. C.E.
IN these times, when communication is carried on chiefly by
means of railways, it becomes important to determine the best mode
of manufacturing railway bars, so as to obtain the greatest amount
of wear with the least possible cost. As this question is one of
increasing interest, the Author has thought that it might be pro-
fitable to communicate to the members of The Institution of Civil
Engineers, the experience which he has gained, during the last nine
or ten years, whilst engaged in superintending the supply of rails
for the Governments of Sweden, Norway, and Denmark.
The Paper will treat; first, of the best method of manufacturing
rails out of common iron, and of their capacity to resist wear;
secondly, of the disposal of the iron rails when they are worn out;
and thirdly, of the conditions under which either iron or steel, or a
combination of both, may be the most economical material for the
construction of rails.
The mode of manufacturing rails for the Swedish Railways, as
carried out in Wales, between the years 1856 and 1860, consisted
The discussion upon this Paper occupied portions of four evenings, but an
abstract of the whole is given consecutively.
THE MANUFACTURE AND WEAR OF RAILS.
5
10
in hammering the pile for the top slab after the first welding heat,
and in rolling it after the second heat. It was supposed that ham-
mering would produce a superior weld and a harder wearing surface
than could be obtained by rolling alone. This method of ham-
mering was, however, gradually superseded by that of rolling at
other works in Wales, during the period referred to. As the
operation of hammering the slab after the first welding heat
entailed an additional charge of 20s. per ton, it became the duty
of the Swedish Government to determine, by practical trials,
whether the value of the finished rail was correspondingly increased.
With this view, several experimental rails were rolled, and arrange-
ments were made for putting them down on some of the English
lines, in such situations as would expose them to severe wear.
The experiments further aimed at discovering, if possible, how long
the rails manufactured at the Cwm Avon Works, in South Wales,
and imported into Sweden, would resist the traffic in that country.
Five different kinds of 'piles' were employed; twenty rails, of
a flange section, 4 inches deep, and weighing 62 lbs. per yard,
being rolled of each particular sort.
The mode of manufacture was as follows (Plate 14):
The rails subsequently noticed in this Paper as those marked T
were made from a pile formed of No. 2, or welded iron, for the top
and bottom, the rest of the pile being of No. 1 puddled bar iron.
The top slab and the squares next to it were made from a ham-
mered bloom of ordinary puddled bars, and the remainder was
filled in at the middle with crop ends from top slabs, and with
other pieces of No. 2 iron.
The rails marked Y were made from a pile of the same compo-
sition as that of the T rails, with the sole difference that the pile
for the top slab consisted of puddled bars, without any welded iron
pieces or crop ends being introduced in the middle of the pile.
The pile for the rails marked H was composed of a top slab
made from puddled bars, hammered after the first heat, and
rolled after the second heat, similar to the rails marked Y, the iron
for the flange consisting of four pieces instead of eight.
The pile for the rails marked E was exactly similar to that for
the rails marked H, excepting only that the pile for the top slabs
was rolled after the first heat, as well as after the second heat. This
difference in the mode of manufacture was adopted in order to
discover whether the process of hammering when applied to this
common iron improved the rail to a corresponding extent. Instead
of No. 2 iron, puddled bars were chiefly used for the squares near
the slab, and for the foot of the rail.
The pile for the rails marked N consisted of puddled bars,
without any top slab.
All the piles passed through the rolling successfully, with the
THE MANUFACTURE AND WEAR OF RAILS.
exception of the N rails, some of which showed cracks, owing to
the inferior quality of the puddled bar.
The London and North Western Railway Company, being in-
terested in the solution of this problem, allowed experiments to be
made on their line, so far as the wear of these experimental rails
was concerned. The experiments were carried out at Camden
Town Station, where the rails could be sooner and more thoroughly
tried than elsewhere; first, on account of the enormous traffic
which obtains at that spot, secondly, from the constant shunting,
and, thirdly, owing to the grinding action of the engine wheels in
starting the trains. The result of these experiments is shown in a
series of tables, drawn up by Mr. H. Woodhouse, Superintendent
of Permanent Way. (Vide Appendix, page 334.)
The following Table, deduced from Mr. Woodhouse's report, shows
the number of tons passed over each kind of experimental rail before
it was crushed, and also before the rails were taken out.
Mark of Rail.
Crushed.
Worn out.
Tons.
Tons.
TYHE Z
T
3,680,000
5,060,000
4,140,000
5,290,000
H
3,220,000
5,0€0,000
6,900,000
8,970,000
N
3,220,000
5,520,000
Another Table, calculated from the preceding one, shows how
long the rails will last, supposing them to be passed over by three
thousand trains yearly, each train being composed of an engine
weighing 30 tons, and of twenty wagons of 10 tons each, or a
gross load of 230 tons.
From these Tables it was ascertained, that the five different de-
scriptions of rails were on the average crushed in six years, and
worn out in nine years, thus¹-

T
Y
H
E
N
Crushed.
Worn
out.
Crushed.
Worn
out.
Crushed.
Worn
out.
Crushed.
Worn
out.
Crushed.
Worn
out.
Years.
Years.
Years. Years.
Years. Years.
LO
5
7
6
7
5
7
Years.
10
Years.
13
Years.
Years.
5
8
Reference may here be made to Mr. Woodhouse's Report in the Appendix
(page 18), showing how the wear of these rails was tested, and how the figures
giving these average results were obtained,
THE MANUFACTURE AND WEAR OF RAILS.
7
As the object of these experiments was chiefly to ascertain the
difference between a rolled and a hammered slab, both made from
inferior iron (E representing the former, and H the latter), these
two kinds of rails were placed on the same line so as to compare
their relative merits; and the result shows the E rail, with the
rolled slab, to be superior at each of the eight places where the
rails were tested.
The conclusion thus arrived at is, that hammering after the first
welding heat, for this particular kind of iron, does not improve the
enduring quality of the rails; but that rolling, which is the simplest
mode of manufacture, has also the material advantage of being the
best. These trials at the same time establish the fact, that it is
not the wear or the diminished sectional area caused by abrasion,
which produces the unsatisfactory results in the endurance of iron
rails, but that it is the lamination caused by imperfect welding,
and discovered by the repeated blows of the wheels in travelling at
high speed. This explains the great difference sometimes exhi-
bited in the wear of rails made in exactly the same way, but the
welding in the one case being perfect, whilst in the other it has
been very imperfect.
The results obtained at the Camden Town Station, however, are
not applicable to the circumstances and conditions of the wear of
rails under ordinary traffic; but rather to exceptional situations,
where the wear is occasioned principally by the frequent use of the
breaks, and by continual shunting, in a much higher degree than
at any other point of the line. These results may also be attributed
in part to the great weight of the locomotives, in proportion to the
weight of this particular section of rail.
Rails of similar quality of iron to those marked E, have been
tried on the Great Northern Railway, and have lasted during the
passage of about sixty-five thousand trains, of a total aggregate
weight of thirteen millions of tons, one-fourth part of this traffic
being at a speed of about 40 English miles per hour, and the
remaining three-fourths at 15 miles per hour.
These experiments on the Great Northern Railway confirm the
rule laid down in Mr. R. Price Williams's Paper, "On the Main-
tenance of Permanent Way," viz., that the endurance of rails
may be measured by the product of the speed and the passing
weight. The trial-rails on the Great Northern Railway may thus
be said to have borne two hundred and seventy-six millions of
tons at a speed of 1 mile per hour. On the other hand, the
Author's experiments at Camden Town, being made under un-
usual conditions of wear, do not bear out this rule; the endu-
¹ Vide Minutes of Proceedings Inst. Č.E., vol. xxv., p. 353.
C 2
8
THE MANUFACTURE AND WEAR OF RAILS.
rance of these rails being represented by only one hundred and
twenty millions of tons at a speed of 1 mile per hour.
Another method of arriving at an opinion as to the endurance
of these rails may be found in the renewals which have been
already made on the Swedish State lines where these rails have been
laid down. The present experience extends over a period of nine
years on portions of the most severely-worked lines, namely, at
Gottenburg and Malmö; and also during a period of six years
over the whole system. The renewals on the main line have been
in an increasing proportion, being in one case 30 or 40 per cent.
higher than in the preceding year; and a mean calculation gives
the probable result that, taking the last renewals of the rails laid
down on the construction of the lines at about eighteen years from
the commencement, the average life of the rails may be taken at
fifteen years.
The weight which would pass over the rails during these fifteen
years, judging from the reports contained in the annual accounts
of the Swedish Government Railway Administration as to the
traffic-returns of all the State lines during the years 1862 to 1865,
may be assumed to be a yearly increase, in the gross load, of
only 15 per cent. per mile, after the year 1865. The yearly in-
crease, however, of transported goods, on the line nearest Gotten-
burg since 1862, has amounted to 30 per cent., and near Malmö
to 18 per cent. Further, supposing the same proportion to exist
between the gross and the net loads for the year 1865, it may
be taken at about six millions of tons-a quantity which agrees
with the results obtained on the English lines, giving for the life of
the best of the three first sorts of rails, the same average life as
that of the E rails. This apparently tends to confirm the correct-
ness of the theory, that the life of rails may be measured by the
product of the weight and the speed.
The rails used on the Swedish lines are mostly of the E or rolled
class before mentioned. Those tried on the Great Northern Rail-
way were also of that kind of manufacture, but of a heavier section.
The speed at which the load was carried over the experimental
rails on the Great Northern Railway was much higher than on
the Swedish lines, being in the proportion of 20 miles on the Great
Northern, to 16 miles average speed on the Swedish railways.
The conclusions which the Author has arrived at are, that no
rule can be laid down for the manufacture of rails which will be
applicable to every manufacturing district; but that in the case
of Welsh iron, to which he has more particularly referred, it has
been proved that the best method of manufacturing the rail is
that now most commonly practised, viz., rolling the iron into bars,
piling them, and then repeatedly rolling the pile to the finished rail,
without hammering. The Author assumes that the prejudicial
THE MANUFACTURE AND WEAR OF RAILS.
9
result from hammering is owing to the amount of sulphur present
in the Welsh iron. Where the iron contains more phosphorus,
and less sulphur, as, for instance, in the Cleveland, Belgian, and
French iron-districts, hammering has proved beneficial, and top
slabs have been made direct from puddled balls, without the inter-
mediate process of piling-this being, in fact, the method gene-
rally adopted in those places, and being found to answer best.
These experiments seem to indicate that two hundred and
twenty millions of tons may be carried over rails, of the sec-
tion and make referred to, at a speed of 1 mile per hour; so
that any Railway Company, knowing the load which passes
annually over their line and the speed of the trains, may, by multi-
plying the one into the other, and dividing two hundred and
twenty millions by this product, ascertain the life of iron rails in
years.
With respect to the disposal of the rails when worn out, and to
the possibility of re-rolling old rails with advantage by Railway
Companies located at a distance from rail-making districts, the
Author thinks that for railways near the seat of rail manufacture,
their best course will be to continue to dispose of the old rails to
the rail mills; but for other countries, situated like Sweden,
Russia, and America, it becomes important to ascertain whether it
would not be more advantageous to re-roll them on the spot.
The increasing traffic, and the augmented speed of trains, to-
gether with the heavier engines now found desirable, have all a
tendency to induce more severe wear, and to render necessary more
frequent renewals of the rails. These renewals are executed with
more and more difficulty, as the greater number of trains limits
the time at disposal, whilst the stoppage of the traffic thus occa-
sioned often results in accidents. These considerations, together
with the recurring expenses of renewals, have conduced to the
employment of steel, which, as a more durable material for rails,
is already much used. This material has been employed either
for the entire rail, or for the upper portion only, the steel top in
the latter case being welded on to an iron rail.
1
In the case of rails with steel-tops, either the head has been
entirely formed of steel, or the upper surface only has been covered
with a thinner or thicker top slab, giving, say inch of thickness
in the rail. The top slab has been differently applied to the rail
according to the different methods used for forming the pile; so
that it has either lain flat over the wearing part of the rail head,
or has spanned the mass of iron with its exterior edges, thus
fixing the steel without reference to the weld;-it has, in fact,
been partly mechanically fastened to it. The mode of making the
pile at several places where old rails have been used instead of
10
THE MANUFACTURE AND WEAR OF RAILS.
puddled bars, is shown in Plate 14, Figs. A, B, and C.
The
pile used at Dowlais for steel-headed rails made for the Swedish
Government railways in September last, is shown in Fig. B; the
one used at Hörde, in Germany, in Fig. A; and that employed on
the London and North Western Railway, in Fig. C.
The use of steel rails has much increased in England, and many
Railway Companies are introducing them as rapidly as their finan-
cial circumstances will allow.
The existence of a cheap raw material, in the shape of old rails,
and the process of re-rolling them together with Bessemer steel-
tops, affords an opening for carrying out the manufacture of rails
profitably for Railway Companies far removed from the seat of
manufacture, even when exposed to foreign competition. If such
re-rolling were carried out in Sweden, instead of selling the worn-
out iron rails in England, the expense of freight to and fro would
be saved; and this may be reckoned at £2 per ton. The necessary
cost of coal would not increase the price of rails more than about
10s. per ton for re-rolling rails in Sweden.
The question is thus reduced to an inquiry, whether the increased
cost of manufacture due to the smallness of the quantity to be
made, would amount to, or exceed, the remaining part of the differ-
ence of the freight, viz., £1. 10s. per ton. If so, it would then be
useless to attempt the establishment of a manufacture of rails,
either by private capitalists or by the Government in connection
with other workshops for the repair and renewal of railway mate-
rials of different kinds, such as engines, axles, wheels, &c.
An estimate has been made, the particulars of which will be
found in the Appendix (page 20), of the cost of manufacturing rails
in Sweden, composed of Swedish Bessemer steel for the head, No. 2
iron for the flange, or foot, and with the remainder of the pile com-
posed of old iron rails. The rails are of the Vignoles section,
weighing 66 lbs. per yard.
From these calculations it is shown:
1st. That according to Plan A, with the whole head of Bessemer
steel, the cost per ton on an annual make of 10,000 tons is
£8. 12s. 1d.; and, 2ndly, according to Plan B, with the top or
wearing surface of the rail only of Bessemer steel, the cost per ton
on an annual make of 6,000 tons is £8. 6s. Ed.
It therefore follows that the difference between these amounts
and £10, the lowest price at which such rails can at present be im-
ported into Sweden from England, viz., £1. 7s. 11d. by Plan A,
and £1. 13s. 4d. by Plan B, respectively represents the net profit
to be derived from the transfer of re-rolling operations to Sweden.
In other words, providing this calculation is not too low, this
represents in the first case 16 per cent., and in the second case
about 20 per cent., of the whole cost of production.
THE MANUFACTURE AND WEAR OF RAILS.
11
In England, the London and North Western and the Great
Western Railway Companies are re-rolling the worn-out rails at
their own workshops, where the repairs of other railway materials
are also executed, and at both these places Bessemer steel has
been used for the head of the rails. The other English Railway
Companies are selling their worn-out rails, as there is sufficient com-
petition to prevent them being sold below their value.
There is only one railway company in France which has works
for re-rolling the worn-out rails; and in Belgium and Prussia
there are none, the rails being sold to the private rail-mills in
the respective countries, and being used in connection with other
iron for rail-manufacture.
In Austria, the State railways are carrying out the re-rolling
of rails at their own workshops at Grätz, while the making of
axles and tyres is performed at the State workshop at Neuberg.
In Russia, the rolling of the rails is executed at a workshop near
St. Petersburg, altered for the purpose by a private company. The
side irons and top slabs are usually of English manufacture, but
sometimes the top slabs are made at home from scrap iron. English
fuel is used. The Petersburg and Moscow Railway receive in ex-
change for their old rails three-quarters of the weight in new rails,
and pay a certain sum, according to the specification, for each weight
of rails received. By the other great Russian railway the old rails
were at first sent to England; but for some time a private iron-
work near St. Petersburg, called Ugareff, has carried on the re-
rolling, on the following conditions:-The railway company finds
one ton of old rails, and receives one ton of new rails, and pays
£7. 14s. per ton, but has then a guarantee for five years.
Having cited the case of Swedish railways, to show how far the
process of re-rolling old rails may be profitably carried out, similar
considerations may be applied to other railways in like circum-
stances, as in the British colonies and in the countries round the
Mediterranean. The special conditions being different in each in-
stance, it becomes difficult, if not impossible, to give an example
that would suit every case. The principle laid down may, how-
ever, be useful as a guide as to what is to be done with the rails
when worn out.
Passing to the third subject proposed for discussion in this
Paper, the Author will endeavour to determine under what condi-
tions it becomes most economical to employ either iron or steel, or
a combination of these two materials, in the construction of rails.
In the following calculation it is assumed that, under a very
heavy traffic, common iron rails will last five years, steel-headed
rails fifteen years, and solid steel rails thirty years; that the
iron rails cost £7 per ton, steel-headed rails £10 per ton, and
12
THE MANUFACTURE AND WEAR OF RAILS.
solid steel rails £15 per ton; and that the old steel-headed and iron
rails are valued at £4 per ton, and the old solid steel at £8 per ton.
With a rail section of 84 lbs. per yard, 250 tons of rails will be
required for one mile of double line, and the cost of laying the rails
may be estimated at £1 per ton. The following example on iron
rails lasting five years, will serve to explain the way in which the
subsequent Annuity Tables have been calculated:¹
250 tons, at £7 per ton
Cost of laying down
£. s. d.
1,750 0 0
250 0
£2,000 0 0
Which sum, at the end of five years, at 5 per cent., compound in-
terest, becomes £2,552. The difference between this sum (viz.,
£2,552) and the value of the old rails (250 tons, at £4 per ton
£1,000) is £1,552. On referring to the Appendix (p. 22), it
will be found, by the use of Table No. 2, that the annuity required
to recoup this latter sum in five years is £280.
ANNUITY TABLE No. 1.
For one English mile of double line, interest being reckoned at 5 per cent., and
steel-headed rails being calculated to last three times and solid steel rails six
times as long as ordinary iron rails-
The Annuity would be for
Where Iron
Rails last
Iron Rails.
Steel-headed Rails. Solid Steel Ralls.
Years.
£.
£.
£.
2
587
395
325
3
417
307
271
4
332
247
245
5
280
218
230
10
179
163
205
15
134
148
405
20
130
140
200
It may be objected that the prices quoted for solid steel rails in the
foregoing calculations are too high. Rails of this kind have lately
been sold in some districts as low as £12 per ton," but for the very
best quality of rails, such as those on which the experiments were
made, the price was £15 per ton at the time they were manu-
factured. Table No. 2 has, therefore, been calculated for the
different kinds and periods at the following prices, viz., iron rails
1
¹ It should here be remarked that the Author uses the term "annuity " to denote
that amount of capital which, put out annually at compound interest at a certain
rate, would in a given time accumulate to a given amount.
2 March, 1868.
THE MANUFACTURE AND WEAR OF RAILS.
13
at £6, steel-headed rails at £9, and solid steel rails at £12 per
ton, crediting the old iron and steel-headed rails at £3 per ton,
and the solid steel rails at £5 per ton.
ANNUITY TABLE NO. 2.
The Annuity would be for
Where Iron
Rails last
Iron Rails.
Steel-headed Rails. Solid Steel Rails.
Years.
£.
£.
£.
2
574
382
288
3
404
283
233

4
319
234
230
5
268
206
174
10
166
14J
168
15
133
136
163
20
117
126
150
This Table shows that solid steel rails are the cheapest up to ten
years' wear of iron rails; that steel-headed rails are cheapest for
between ten and twenty years; and that iron rails are cheapest
when they last twenty years or more.
The amount of traffic must, therefore, decide which material it is
the most economical to use for the maintenance of the permanent
way. For all railways where ordinary iron rails are worn out in
five years, or in a shorter time, solid steel rails are the most econo-
mical, even at the price quoted in Table No. 1; but where ordi-
nary iron rails last over ten and up to fifteen years, steel-headed
rails would be the cheapest; iron rails in these cases being clearly
proved to be the most expensive, although the cheapest where they
last from fifteen to twenty years and upwards.
As these calculations are founded on the short experience gained
up to the present time, in reference to the relative endurance of
the different kinds of rails, a still longer trial is desirable.
The foregoing Tables refer to rails of the Vignoles section. Table
No. 3 has been calculated for the ordinary double-headed rail, ac-
cording to the prices stated, the considerations being the same as in
Table No. 2, except that the chairs have been taken into account.
Allowance has been made for 140 tons of new chairs per mile, at
£5 per ton, credit being given for the value of the old chairs at
£2.10s. per ton. It may be observed that steel-headed rails are
here estimated to last four times, and solid steel rails eight times,
as long as ordinary iron rails, that is, making allowance for the use
of both faces of each rail,
14
THE MANUFACTURE AND WEAR OF RAILS.
ANNUITY TABLE No. 3.

The Annuity would be for
Solid
Steel-headed rails. Solid Steel Rails.
Where Iron
Rails last
Iron Rails.
Years.
£.
£.
£.
23 HO
780
379
296
551
291
249
4
436
244
228
5
366
233
217
10
229
177
199
15
183
166
20
163
162
: :
This Table indicates that the iron rails are in no instance the
cheapest; but that, on the contrary, when iron rails last only up
to five years, solid steel rails have the advantage; and when the iron
rails have a longer duration, then steel-headed rails are the most
economical.
The experience of the durability of the steel-headed rail and
of the solid steel rail may not agree with individual cases of
failure, where, in consequence of defective welding, the steel head
has come off, or where the solid steel rail has broken. At the
same time, it must be admitted that the process of steel making,
and of welding the steel slab to the rail, is, at present, in its in-
fancy, so that great progress may yet be expected. The principle
ought not to be condemned on account of individual failures.
It is to be hoped that Railway Companies having a heavy
traffic will give due attention to the relative merits of the different
sorts of rails, and submit them to trials on a large scale; and, on
the other hand, that the steel works will strenuously endeavour to
manufacture solid steel, as well as steel-headed rails, of the best kind
for the purpose, so that this important question may soon be de-
cided.
Before concluding, another subject must be taken into considera-
tion, viz., the manner in which the safety of the permanent way
constructed of the three different kinds of rail is affected by high
speeds, severe climate, &c.
This subject seems of late to have engaged the attention of the rail-
way world, and has been discussed, not only in England, but on the
Continent. The Swedish Government, having undertaken the
construction of railways, appointed a Committee, composed of
many eminent men, to consider the subject. This Committee found
it necessary to make experiments with different materials from
England and Germany as well as from Sweden; and, after five years'
consideration and study, the report has just been published by
THE MANUFACTURE AND WEAR OF RAILS.
15
Professor Styffe, the Director of the Royal Technological Institute
at Stockholm.
From this report, it appears that the extensibility and tensile
strength of different materials are influenced by the amount of
carbon which they contain. (Vide Plate 14.)
The EXTENSIBILITY influenced by CARBON.
Description of Materials.
Carbon
per Cent.
Extension
per Cent.
Swedish Bessemer steel and Uchatius steel
1.85 to 1·0
0.3 to 0.9
For cast steel.
•
•
•
0.69,, 0.61
1.2
2.1
""
Bessemer steel or iron
0.42,, 0.33
1·9
4.0
> >
Spec. Grav.
Iron from lake ores (rich in phosphorus)
Iron from Dudley (rich in slag and phosphorus)
Iron from Middlesbro'-on-Tees.
Puddled iron from Sweden and Low Moor
Swedish iron made in the "Lancashire hearth'
0.8
3.4
""
7.5
2.5
4.2
7.65
3.4
5.9
7.77 to 7.80
7.81
6.1
9.5
"
7.3 7.8
""
The TENSILE STRENGTH influenced by CAREON.
Description of Materials.
Breaking Load
in lbs.
per Square Inch.
Carbon
per Cent.
Swedish charcoal puddled iron
0.8
113,381
0.7
84.265
Bessemer steel
0.8
90,921
0·55
$6,941
> >
0.50
71,090
,,
Bessemer iron
0.20
,,
•
48,102
puddled iron
0.70
$3,441
0.70
$3.716
0.6
73,492
0.6
82.344
0.5
78,432
0.7
S6,049
These Tables show that the hardest material has the greatest
tensile strength, but that it has the least ductility; while, on the
other hand, a softer material shows a greater amount of extension.
In the diagram graphically illustrating these results (Plate 14),
the percentage of carbon and of phosphorus is stated in nearly all
cases. The limit for the amount of carbon seems to be for the Besse-
mer material 1.2 to 1.5 per cent. With a larger amount, the tensile
strength, as well as the extensibility, has been found to decrease.
When the amount of carbon does not exceed 0.4 per cent., and the
material is not worked at too low a heat, the elongation seems to be
16 per cent., or the same as for puddled iron from the same pig iron ;
16
THE MANUFACTURE AND WEAR OF RAILS.
and, as such Bessemer material is not only much stronger, but also
more solid or homogeneous than the puddled material, it deserves
a decided preference for all railway purposes. The few cases of
failures of rails by breaking may be accounted for as the result of
too hard a material, not perfectly manufactured, having been made
at the earlier period of the introduction of the process.
The
experience which has now been gained should certainly prevent
any recurrence of such exceptional cases.
Experiments on the tensile strength of iron and steel under the
influence of extreme temperatures (as at 212 deg. and -40 deg.
Fahr.), have led Mr. Styffe to the conclusion (contrary to the general
belief), that the tensile strength is greater during cold than under
ordinary temperatures; that is, iron or steel is stronger in winter
than in summer. The reason why more breakages occur in winter
than in summer is asserted to be due to the extreme cold rendering
the ballast and sleepers rigid; and, therefore, elasticity given to the
rolling stock, in any way, favourably affects the resistance of the
rails.
If, however, the supports have the same elasticity in summer as
in winter, as would be the case, for instance, with a granite rock,
Mr. Styffe asserts that the same rails, either of iron or of steel, can
resist a heavier blow from a falling ball at an extremely low
temperature than on a hot summer's day. Although the experi-
ments have been conducted with the utmost care and skill that
science and money can afford, it seems to the Author desirable that
this theory should be examined on a larger scale than Mr. Styffe
has had an opportunity of doing before it can be relied upon.'
At a meeting of Engineers at Stockholm in March, 1867, it
was decided that Bessemer steel rails, made from charcoal pig
iron, might, without risk, be used 10 per cent. lighter than the
English iron rails; and in Austria this decision has already been
acted upon with success by the Engineer-in-Chief, Wöhler.
It must, however, be observed, that the raw material used in
both cases is charcoal pig iron of a superior quality as compared
with that used in England for making Bessemer rails, which may
be seen from the following analyses made by two eminent chemists.
1 Since this Paper was read, the Author has had the opportunity of conducting
some experiments on iron rails, in Sweden, by exposing them at different temper-
atures to the impact of a falling ball; and the results of these experiments show
that, on an average, the strength of a rail in winter is not more than one-fourth of
the strength exhibited by the same bar in summer. The Author has given
an account of these investigations in an Appendix to the following work:-"The
Elasticity, Extensibility, and Tensile Strength of Iron and Steel."
By Knut
Styffe, translated from the Swedish, with an original Appendix, by Christer P.
Sandberg; with a Preface by John Percy, M.D., F.R.S. London, 1869.
THE MANUFACTURE AND WEAR OF RAILS.
17
Bessemer Swedish Pig Iron,
Fagersta Works,
Analysed by Kohlberg.
Graphite
Combined carbon
Silicon
Manganese
Sulphur
•
Phosphorus
English Bessemer bæmatite Pig Iron,
Analysed by John Percy.
Per Cont.
2.733
Carbon
2.138
Silicon
0.641
Manganese
2.926
Sulphur
0.015
Phosphorus
0.026
Per Cent.
2.993
3.080
0.079
0.021
0.021
These analyses show that the great difference between the two
kinds of iron is the larger percentage of silicon in the English,
and of manganese in the Swedish pig iron; thus explaining why
the one gives a better result than the other, although worked
entirely without the addition of spiegeleisen.
If there be only 0.6 per cent. of carbon in the solid steel, and
0.3 per cent. in the steel for the steel head, the safety ought to be
the same for all the three kinds; and this would not influence the
former calculations as to which is the best and most economical
material for rails.
Having watched the development of the Bessemer process in
England, as well as on the Continent, the Author believes that
in that process a good and pure raw material has a greater advan-
tage over an inferior one than in other processes; and that a superior
product cannot be obtained from an inferior raw material by that
method any more than by others. Indeed, a purer raw material
is needed for the Bessemer process, as the phosphorus is not elimi-
nated by that method of treatment as it is by puddling. In hav-
ing mentioned Swedish material, as an example, it must not be
supposed that it is wished to advocate the use of Swedish iron
in this country, but simply to draw attention to the better mate-
rial; as equally good charcoal iron can be supplied from Canada
and India, both English colonies. It may also be remarked, that
the Author's endeavour has been to arrive at the truth irrespec-
tive of prejudice, and that he has no wish to be deemed an advo-
cate for one kind of rail more than another.
The Paper is illustrated by a series of Diagrams, from which
Plates 14 and 14A have been compiled.
[APPENDIX.
18
THE MANUFACTURE AND WEAR OF RAILS.
TABLE NO. 1.
SHOWING the NUMBER of TONS passed over the EXPERIMENTAL RAILS before they were Crushed and Worn Out.
H
N
APPENDIX.
RESULT of the WEAR of the EXPERIMENTAL RAILS at CAMDEN TOWN STATION,
according to Mr. WOODHOUSE'S Report.'
T
Y
E
Crushed.
Worn out.
Crushed.
Worn out.
Crushed.
Worn out.
Crushed.
Worn out.
Crushed.
Worn out.
Tons.
Tons.
Tons.
Tons.
Tons.
Tons.
Tons.
Tons.
Tons.
Tons.
1,380,000 1,610,000
2,070,000 2,070,000
920,000 3,450,000 920,000 1,380,000 2,300,000 3,450,000 1,610,000 2,530,000|
920,000, 2,300,000 1,610,000 2,300,000 4,370,000 4,370,000 2,530,000 2,760,000
230,000 2,530,000 2,070,000 2,530,000 1,840,000 2,990,000 460,000 4,370,000 2,760,000 3,220,000
2,530,000 2,530,000 1,840,000 3,680,000 1,610,000 3,220,000 2,530,000 5,750,000 2,760,000 3,680,000
460,000 3,220,000, 3,910,000 4,140,000 1,810,000 4,140,000 4,830,000 6,210,000 3,450,000 4,140,000
1,610,000 3,450,000 3,680,000 4,140,000 2,760,000 4,370,000 6,210,000 7,130,000 2,070,000 4,370,000
4,370,000 5,290,000 3,450,000 4,140,000 3,150,000 4,370,000 5,290,000 7,590,000 2,300,000 4,830,000
690,000 5,750,000 2,990,000 4,370,000 3,450,000 4,600,000 8,280,000 8,510,000 3,450,000 5,290,000
4,830,000 6,440,000 3,450,000 4,600,000 3,680,000 4,830,000 6,900,000 8,740,000 4,140,000 5,290,000
6,210,000 7,130,000 3,220,000, 5,750,000 3,220,000 5,060,000 7,130,000 9,430,000 6,210,000 7,360,000
6,210,000 8,510,000, 4,370,000 5,980,000 3,680,000 5,290,000 8,740,000 10,580,000 5,750,000 9,820,000
11,040,000 11,960,000 7,820,000 7,820,000; 4,600,000 5,520,000 7,590,000 10,580,000 4,600,000 6,210,000
5,290,000 8,280,000 1,840,000| 5,750,000 11,040,000 11,960,000 2,990,000 9,430,000
7,820,000 10,120,000 3,680,000 5,750,000 14,030,000 14,950,000 8,510,000 9,960,000
11,010,000 11,960,000, 4,600,000 6,210,000 14,030,000 14,950,000
2,300,000 7,820,000 14,030,000 14,950,000|
7,360,000 10,580,000,
AVERAGE OF THE ABOVE.
Cruslied.
Worn out.
Crushed. Worn out.
Crushed.
Worn out.
Crushed.
Worn out.
Crushed.
Worn out.
4,140,000
5,290,000 3,220,000
6,900,000 8,970,000 8,220,000
000
3,680,000 5,060,000 4,140,000 5,290,000 3,220,000 5,060,000 6,900,000 8,970,000 3,220,000 5,520,000



number of tons that passed over the road; they must be
divided by 2 to find the number of tons that passed over cach rail.
1 The figures given in this table show the
SWEDISH
RAIL SECTION.
62 lbs. per yard,
Piles T & Y
Pile for the top Slab For
Files H&E,
Bails TYA & E.
REFERENCES:
Iron №2, or Wehled Iron,
Půddle Ears.
Bessemer Steel,
HAMY
:
IRON
AND STEEL RAILS.
SWEDISH RAIL
SECTION.
144102
137240
· 130378
123516
116654
Bolled Swedi
Cast Steel: (
atius Pr
•Process) Carbou
go, Phosphorus
Hawnword Swedish Bessenter Stat, (turben 1.147%. Plespre
Phosphorus 0.018 %
Plan A.
66 lbs. per yard,
109792
Hammered Swedish Bessemer Steel (arbou 1.05
BESSEMER STEEL SLABS.
-pu
102930
Hammer
Hann
unered
swedish. Bessermer
Bessemer, Steel
Swedish Bessemer stru (Carbin 1.68
L
BESSEMER STEEL.
Plan B.
66 lbs. per yard,
96068
89206
English Poids per English Square
82344
75482
68620
61758
54896
48034
41172
34310
27+48
1
-+
H
+
1
TA
+
1.
Middling Hart Sxedish PuddledSteel,
Kimpp's Cast Steel, bàn
wered, Wurbon
Murben 0.64 Je, Phosphorus 0:03 2'3
Soft Swedish Puddled Steel
Puddled Steel
(Carbon 0.3300)
Soft Swedish Puddled Steel.
wedish Bessemer from, hamurred
Soft Swedish Puddled
cd Steel
H-
H
Swedishy Luke Ori Grow Fear
8Cc
Phosphionis 0.26 fe
4
5
6
7
8
9
Bessemer and Cast Steel.
Puddled Material,
10
11
#
#
#
¦
PLATE
14.






Swedish Bessemer From, ralled, (Carbon 0.38 %, Phosphorus 0.023 %.


English PrádledFron from Lowmoor
Carbon-0.21 20, Phosphorus 0.068 jo.
Swedish Iron, refined, Carbon 0.06%,
Phosphorus 0.022 0.
Swedish PuddledIron.

12
13
14.
15
16
17
18
19
20
21
Elongation in perlent.
EXPERIMENTS AS TO TENSILE STRENGTH OF CAST STEEL, BESSEMER STEEL, AND PUDDLED MATERIAL,
MADE BY PROFESSOR K. STYFFE.
Minutes of Froceedings of The Institution of Civil Engineers, Vol: XXVII. Session 1867–68.
Pile N
Plan C
84 lbs per yard.
PILES FOR IRON RAILS.
Scate for Files, 2 inches – 1 foot.
PILES FOR STEEL HEADED RAILS.
Do. for Rail Sections, 4 inches. I fùot.
C.P. SANDBERG, DELT
KELL, BROS LITERS CASTLE SE HOLBORN.
Railway to carry coal in and ashes out
1
Heating Furnaces.
Steam Engines
!
1
(Steam EnJING
Railway for carrying coal in and ashes ont.
1
C.P SANDBERG, DELT
Scale: 18 Inch - 1 Foot.
Minutes of Proceedings of The Institution of Civil Engineers Vol XXVII Session 1867 68
KELL BROS LITHRO
Rail Shear,
for cutting old Rails
100)
Steam Engine.
Saw for cutting the hot rails.
Rail Mill
Strong Shears,
for cutting rails, top slabs &c.
PLAN OF RAIL MILL FOR REROLLING OLD
MANUFACTURE AND
WEAR OF RAILS.
RAILS
WITH
BESSEMER
Grinding machine
for Rails.
Modes
cha
Rai
Steam Engine
Straightening
Pr'ESSES.
Railway for sending of New Rails.
Punching
machine.
Inspection
Benches
for
New Rails.
J
PLATE 144
STEEL TOPS.
!


To face p. 19.]
THE MANUFACTURE AND WEAR OF RAILS.
REPORT showing how the WEAR of the RAILS has been Tested, and how the Figures in the preceding Tables have been
ascertained.
London and North Western Railway. Southern Division.
Engineering Department.
Statement on Mr. Sandberg's experimental Rails, laid down at Camden Yard, from November 9, 1863, to February 8, 1865.
Place No. 1.-One length of Rails under Passenger Foot Bridge. 77 Trains pass over these Rails daily.
1st Rail.
The one side of the line.
E Laid down
Crushed
Taken out
9/11/63.
9/9/61
E
21/1/65
5,382,000 tons.
7,590,000
SOUTH
-21 £t-
NORTH
The other side of the line.
1st Rail.
H
Laid down
9/11/63.
Crushed
Taken out
6/2/64
13/5/64
1,564,000 tons.
3,220,000
2nd Rail.
E
Laid down
Crushed
Taken out
13/5/64.
31/12/64
21/1/65
4,094,000 tons.
4,370,000
""
Place No. 2.-Three lengths of Rails from signal towards Chalk Farm Bridge. 115 Trains pass over these Rails daily.
Signal
1st Rail.
N Laid down
Crushed
Taken out
9/11/63.
6/2/64
12/5/64
2,346,000 tons.
4,830,000
1st Rail.
T Laid down
9/11/63.
Crushed
Taken out
31/12/64 11,040,000 tons.
4/2/65 11,960,000
T Laid down
12/5/64.
Crushed
Taken out
31/12/64 6,141,000 tons.
4/2/65 7,130,000
1st Rail.
2nd Rail.
N Laid down
Crushed
Taken out
9/11/63.
20/4/64 4,370,000 tons.
30/6/64 6,210,000
N
"
2nd Rail.
1st Rail.
N Laid down
Crushed
30/6/64.
6/10/64
N
Y Laid down
9/11/63.
Taken out
2/11/64
2,852,000 tons.
3,220,000
Crushed
Taken out
31/12/64 11,040,000 tons.
4/2/65
11,960,000
3rd Rail.
N Laid down
2/11/64.
N
Crushed
31/12/64 1,541,000 tons.
Taken out
4/2/65 2,530,000
1st Rail.
E Laid down
Crushed
Taken out
9/11/63.
31/12/64 11,040,000 tons.
3/2/65
11,960,000
CHALK
E
སྲབ་བསམ་
ARI
1st Rail.
H Laid down
Crushed
Taken out
9/11/63.
21/12/63 3,772,000 tons.
12/5/64 4,830,000
19
2nd Rail.
E Laid down
13/5/64.
Crushed
Taken out
31/12/64
3/2/65
6,141,000 tons.
7,130,000
""
(Signed) HENRY WOODHOUSE.


THE MANUFACTURE AND WEAR OF RAILS.
19
TABLE NO. 2.
SHOWING the LIFE of the EXPERIMENTAL RAILS expressed in TIME.

Number of
Trains con-
sisting of an
Engine of
30 tons and
twenty
Wagons of
10 tons each.
Life in
Years
at ten
Trains
per
Day.
T
Y
H
E
N
Total
Number of
Number
of Rails.
Number
of Rails.
Number
of Rails.
Number
of Rails.
Number
of Rails.
Rails.
Crushing.
Worn Out.
Before
Crushing.
Worn Out.
3,000
6,000
9,000
12,000
15,000
18,000
21,000
1234567
BINH
3
1
1
5
2
1
2 2
24,000
8
•121 4BQ
152∞ ∞ ∞
222
42
4211
1
2
3
1
4
2
2
1
2
1
2
•242¬
11
2
1
1
3
212NN
9
9
6
7
7
10
2
6
27,000
9
1
1
1 2
2
5
7
30,000 10
1
1
1
2
•
33,000 11
1
1
2
2
2
3
36,000 12
2
2
1 1
3
3
39,000
13
1
1 2
42,000
14
21
1
2
4
2
3
45,000
15
1
1
48,000
16
1
1
3
2
51,000
17
1
1
1
3
54,000 18
60,000 21
66,000
3
3
3
Co
3
AVERAGE LIFE IN YEARS.

T
H
E
N
Crushed.
Worn
Out.
Crushed
Worn
Out.
Crushed.
Worn
Out.
Crushed.
Worn
Crushed.
| Out.
Worn
Out.
ما
5
7
6
7
LO
5
7
10
13
LOO
5
8
THE MANUFACTURE AND WEAR OF RAILS.
20
COST of Building a ROLLING MILL, for the Manufacture of 10,000 Tons of Rails
per Annum.
Rolling Mill, train complete, with a twenty-ton fly-wheel, three
boilers, pipes, shafts, cranes, &c.
Blooming-machinery, rolling-train (patent, so that no labourer is
needed), rail-sawing-machine, with steam-engine, punching-
machine, straightening-machines, planing-machines and lathes;
all with separate steam-engines.
Building of galvanized iron, according to plan (Plate 14a)
Railways and weighing-machines.
Masonry for the boilers.
Ditto for welding-furnaces
Foundation, culverts, &c.
Office buildings, warehouses, and ground.
•
£.
s. d.
2,335 0 0
2,240 0 0
1,300 0 0
640 0 0
810 0 0
1,226 0 0
380 0 0 0
1,240 0 0
£10,171 0 0
The cost of the machinery is less by £940 where water-power is used instead of
steam-power; provided the cost of the foundations, the premises, and the channel
to conduct the water, remain the same.
ESTIMATE of the Cost of MANUFACTURING RAILS in SWEDEN, rolled from worn
Rails, using Swedish Bessemer Steel for the rail-head, and No. 2 iron for the foot.
(Vignoles section, 66 lbs. per yard.)
(Plan A, Plate 14.)
Rails with the whole head of Bessemer Steel.
The quantity manufactured is 33 tons per day, or 10,000 tons per year.
For the manufacture of 1 ton of Rails, there are consumed :-
Bessemer steel,
936 lbs. at £10 per ton
Side iron and cramps, 260
8
Old rails,
1,644
3
""
""
2,840 lbs.
From this pile is obtained-
79 per cent. of 2,840 lbs.
8
13
""
""
=
2,240 lbs. = 1 ton of rails.
224 = 2 cwt. crop ends.
376
""
""
th of a ton waste.
2,840 lbs.
Coals, 16 cwt. at 13s. per ton
Wages
Cost of management and repairs of the furnaces
Rent of building and interest on fixed capital, £10,200, and on floating
capital, £3,000, at £12 per cent. (6 per cent. for the wear of ma-
chinery and unexpected accidents)
Deduct value of crop ends at £3 per ton
Net cost per ton.
£. s. d.
4 3 7
0 18 7
2 4 1
£7 6 3
0 10 5
0 16 3
0 2 0
03 2
£8 18 1
060
£8 12 1
THE MANUFACTURE AND WEAR OF RAILS.
21
£. 8. d.
If the quantity manufactured does not amount to more than 6,000 tons
per annum, the cost is increased by 2s. per ton. On the other hand,
where water-power can be used without increasing the cost of con-
struction of the works, and without raising the price of the raw
material, the cost of manufacture is diminished by 2s. per ton, or to. £8 10 1
The form of rail-pile (plan A) is used at Hörde Iron Works in Westphalia, for
the supply of steel-headed rails to several German railways.
(Plan B,¹ Plate 14.)
Rails with Steel slab overlapping, at least half-inch thick in the head of the rail.
The quantity manufactured, 6,000 tons per year, equal to about 20 tons per day.
For the manufacture of 1 ton of Rails, weighing 66 lbs. per yard,
there are required :-
Bessemer steel,
766 lbs. at £10 per ton
·
Side iron for flange,
233
8
Cramp-iron for flange sides,
28
>>
Old rails,
1,813
3
19
2,840 lbs.
From this pile are obtained :-
Rails
•
Waste and crop ends
Coals, 16 cwt. at 13s. per ton
Wages.
£. s. d.
3 9 3
0 18 8
2 8 4
£6 16 3
2,240 lbs.
600
2,840 lbs.
0 10 9
0 18 4
0 2 6
0 4 10
£8 12 8
060
£8 6 8
Cost of management and repairs of the furnaces
Rent of building and interest on the fixed and floating capital of
£12,000, at 12 per cent., of which 6 per cent. is for wear and tear
Deduct for crop ends
Net cost per ton.
CALCULATION OF ANNUITIES.
FOR COMPOUND INTEREST.
1
If a quantity A increases
th every year for p years, it will, at the end of
22
22 + 1
that time, amount to A
n
Table No. 1 gives the values of (~2 + 1 ) p
N
for the different values of Ρ when
x = 20, that is when the rate of interest is 5 per cent.
FOR THE ANNUITY TABLES.
1
―
N
tb,
If x be the quantity, which if set aside at the end of each year increases
the annuity will at the end of p years amount to a given quantity B: then
B
n
(n + 1) (" + 1) P -
ጎ
n.
This plan has been used at Dowlats for the supply of steel-headed rails for the Swedish Govern-
ment Railways.
D
22
THE MANUFACTURE AND WEAR OF RAILS.
the values of (n + 1) ("
n + 1
1) p-
1
N
The Tables give the values of (n + 1)
-n for the different values
of p when n = 20, that is when the rate of interest is 5 per cent.
The annexed Tables have been found by logarithms. The first shows the com-
pound interest after a number of years, and the second the annual sum which is
required to be set aside, at compound interest, to amount to a given sum at the
end of a certain period.
To use No. 1, multiply the given sum by the factor in the second column op-
posite the number of years, and the product shows what the capital would have
risen to in the time given at 5 per cent. compound interest.
To use Table No. 2, divide the given sum by the divisor in the second column,
opposite the number of years, and the quotient is the annuity required to recoup
the given sum after a certain period.

TABLE NO. 1.
TABLE No. 2.
Number
of
Years.
Capital of 1 rises by
Compound Interest
at 5 per cent. to
Number
At 5 per cent. Interest,
of
Years.
the Annuity 1 will
recoup a Capital of
1.000
123HLO
1.05
1.102
1.158
4
1.216
5
1.276
6
1.340
123HLO CO
2.05
3.152
4
4.310
5
5.526
6
6.802
7
1.407
7
8.142
8
1.477
8
9.549
9
1.551
9
11.025
10
1.629
10
12.77
12
1.795
12
15.917
15
2.079
15
21.779
18
2.406
18
28.132
20
2.653
20
33.0649
24
3.225
24
44.501
25
3.386
25
47.727
30
4.322
30
66.438
35
5.516
35
90.319
40
7.040
40
120.8008
45
8.985
45
159.701
50
11.47
50
209.341
60
18.68
60
353.59
70
30.43
70
588.528
80
49.56
80
971.221
90
80.73
90
1594·606
100
131.5
100
2610.04
150
1508.0
150
200
17293.0
200
30139.486
345833.2
For Example.
£.
8. d.
One Swedish mile of rails (single line), 700 tons at £10 per ton
Cost of laying down
Required to find the amount which the above (£7,400) would
come to after 45 years at 5 per cent. compound interest.
7,000 0 0
400 0 0
£7,400 0 0
THE MANUFACTURE AND WEAR OF RAILS.
23
£7,400 x 8.985
Deduct for the old rails at £3 per ton
£. 8. d.
66,489 0 0
2,100 0 0
£64,389 0 0
To find the annuity which it would be required to set aside, at 5 per cent. com-
pound interest, to amount to £64,389 in 45 years.
£64,389 divided by 159-701 gives £403·3; and £403 6 8 is therefore the
annuity required.
The annuity may also be found by dividing the present capital, minus the
sale price, by the number shown in Table 2, and adding the simple interest on
the capital, thus:-
7,400 - 2,100
159.7
add 5 per cent. on 7,400
Annuity
5,300
33.3;
159.7
370⚫
£403.3
To find the time in which a given annuity will recoup a certain capital.
Divide the difference between the cost of new and the price of old rails by the
difference between the given annuity and the simple interest on the total cost.
The quotient stands in Table 2 opposite the number of years. For example:-
In how many years will the outlay for a mile of solid steel rails, on 250 tons at
£15 per ton + £250 for laying them down, be recouped by an annuity of £230?
250 tons at £15 =
£3,750
250
£4,000
less the sale at £8 per ton
ton.
2,000
£2,000
Simple interest at 5 per cent. on £4,000 is £200. The divisor accordingly is
230-200 = 30.
2,000
30
66.67, nearly the number in Table 2, opposite the year 30.
[Mr. SANDBERG
D 2
24
THE MANUFACTURE AND WEAR OF RAILS.
Mr. SANDBERG said, that when this subject was brought before
the Institution two years ago by Mr. R. Price Williams, few Rail-
way Companies had adopted steel rails; but that now nearly all
lines used them to some extent. He thought conclusions had
hitherto been formed without sufficient experience; but he hoped
the statements of permanent-way engineers would supplement the
deficiencies of his Paper. Continental railways, as yet, afforded
very little experience; and it would be for England to show what
had to be done in this important engineering question. On the
Swedish lines steel rails were being adopted, though the traffic was
comparatively light.
4
A simple way of ascertaining that the steel employed was
sufficiently tough for rails, was to bore a hole in the bottom plate
of the ingot-mould, 2 inches in diameter at the top, 12 inch in
diameter at the bottom, and two inches deep; the piece projecting
from the lower end of the ingot was broken off when cold, and
having been heated to a red heat, was flattened into a cake or plate
such as he then exhibited. When cold it was
When cold it was doubled up under a
hammer, and its toughness under this treatment decided its suita-
bility as a steel for railway purposes. Afterwards it was sent to
the laboratory, and the carbon determined by the Eggertz process.
The percentage of carbon and its registered No. were then marked
on each piece for further reference. If under the bending-test
the cake broke short, the ingot from which it was taken could be
used for making springs, &c., which required harder material.
He had been furnished with some particulars respecting the use
of steel rails on the London and North Western Railway, from
which it appeared that steel rails were laid down at Camden Town
in May, 1862, in seven different places, and one pair of rails put
down in January, 1863. They were laid so as to form one side of
a single line of rails, the opposite side being composed of ordinary
iron rails laid down to test their comparative wearing qualities.
The effect of wear upon the ordinary iron rails was shown in the
following statement drawn up by Mr. Woodhouse, the Engineer of
the Permanent Way:-
"The 1st face of 5 steel rails was wearing as
compared with the 2nd face of 5 iron rails.
""
""
12
""
+020003 LOH H
4
3rd
4 >>
5
4th
5
>>
11
""
5th
12
""
>>
29
>>
6
59
""
"1
""
9th
""
5
10th
""
""
19
1
11th
""
""
"
1 (a.)
""
6th
7th
8th
19
""
19
6
5
2
3
5
""
19
>>
1
""
""
>>
15th in August, 1865, when
""
the opposite steel rail was removed. ·
1 (b.) was wearing the 23rd face in March, 1867, when the
opposite steel rail was accidentally broken, and
was taken out.
THE MANUFACTURE AND WEAR OF RAILS.
25
CC
(a.) The steel rail removed in August, 1865, was taken up to
enable Mr. Bessemer to read a Paper on it at the meeting of the
British Association.
(6
(b.) The steel rail removed in March, 1867, was not broken by
fair wear and tear, but in consequence of some vehicles getting off
the road and causing damage.
"At the time of the removal of the above-mentioned steel rails
they were wearing thin, and were bruised on the outer edge. The
latter of the two rails was reduced in height by about
inch."
5
1
ths of an
They
All the forty-eight steel rails remaining in the road were still in
good condition, and none of them had yet been turned.
were still wearing satisfactorily.
If the number of worn-out iron rails in each case cited in
the previous table were summed up, it would be found to amount
to a total of three hundred and fifteen, while the number of steel
rails wearing on the opposite side, but in good condition, amounted
to fifty. So far, therefore, as the experiments had proceeded, each
steel rail was found to have worn out, on an average, six iron rails;
and on these data the annuity tables had been calculated.
Mr. EDWARD WILLIAMS said there seemed to him some danger
of false conclusions as to the cause of the superiority of rails made
by the Bessemer process. These rails would wear well, not so
much because they were made of steel, as that they were rolled
from solid ingots.
Rails, however carefully made by the old process, being built up,
were, of course, more likely to laminate than those which were
made from ingots which had no layers in them. He was con-
vinced that the question of good or bad iron rails was simply a
question of welding, and that the quality of the iron was compara-
tively of little or no importance, except so far as one description
of iron welded more easily than another, and that the kind of iron
that welded most easily would make the best built-up rail, because
there was the least chance of imperfect welds. A rail made from
an ingot would be better than a built-up rail, but whether, price
being taken into account, it might not, in many cases, be econo-
mical to use piled rails, was another question. He was inclined to
think it would be so.
With regard to putting a steel top on an iron rail, it was open to
the objection that steel could not, with certainty, be welded on to
iron, and there would be danger of the steel head coming off
under heavy work.
He thought that what were usually called the best iron rails-
rails for which an extra price was paid, and which were carefully
watched during their manufacture,—were really the worst rails, for
the reason that the iron composing them had to be worked over and
26
THE MANUFACTURE AND WEAR OF RAILS.
over again, and after each working it welded with greater difficulty
than before. No kind of iron welded so easily as puddled bars, and
the nearer the wrought iron was to cast iron, the better rail it
would make. If it were possible to make a rail direct from a
puddled ball, it would probably make the best iron rail that could
be produced, but, up to the present time, it had not been found
possible. The system adopted in the North of England was, he
believed, the best for producing really serviceable iron rails. A
puddled ball, a rough, spongy sphere of about 15 inches to 18 inches
in diameter, was put under a hammer and received a few blows to
flatten it. Another ball was put on the top of it, and it was served
in the same way; the mass was then hammered into a rectangular
slab about 10 inches, or 11 inches wide, and about 2 inches thick.
Two or three such flat slabs were then put together in a furnace,
raised to a welding heat, rehammered, and rolled into the slab to
form the head of the rail to be produced. This system gave just as
much work as was necessary to insure soundness of the slab with-
out overworking it. He had no doubt that this mode of working
was much better than roughing down from piles composed of thin
puddled bars. Indeed, long experience had proved to him that,
whatever system was adopted, it depended upon the capacity of
the iron for welding, and the giving it a proper heat, whether or
not the rail produced would wear a long or a short time.
He thought the system of supplying exact specifications, and watch-
ing the manufacturers of rails, had not succeeded; and he would
suggest one which, he believed, would produce much better rails
than had been turned out lately, and at a cheaper rather than at
a higher price. He would advise that Railway Companies should
only apply for tenders to rail-makers of good reputation, that the
onus as regarded the durability of the rails supplied should be left
upon such makers, and that the Railway Companies should make
no secret of the results; in fact, that there should be a periodical
return of each maker's rails worn out, and the amount of work they
had borne. The more public such returns were made, the better it
would be for those who supplied really good rails, and the system
could scarcely fail to be a better protection to Railway Companies
than anything they now possessed. Of course, the prices paid for
the rails should be given as well.
Sir CHARLES Fox thought the value of steel rails was underrated,
inasmuch as it was generally considered as a question of steel versus
iron. He did not say that the material of steel was not better for
the manufacture of rails than iron, as it undoubtedly was so, but he
thought that was not a fair exponent of the difference of the value of
the two. An iron rail when manufactured, even in the best way, was
little more than a bundle of rods; and the top slab might be com-
pared to a piece of hoop-iron lying on the top of an anvil (the body of
THE MANUFACTURE AND WEAR OF RAILS.
27
the rail), when a heavy locomotive passing over it acted as a sledge
hammer. If a locomotive wheel were permitted to run over a penny-
piece, by its crushing weight it increased its length; but as that
would not be the case with the top portion of a rail, because it was
penned in lengthwise, it necessarily spread sideways, and became
laminated. A steel rail was rolled from one solid ingot, not
obtained in the ordinary way of casting ingots by pouring the
material from the lip, but from a stopple hole at the bottom of the
ladle, by which means no scum would get into the ingot, and it was
consequently homogeneous and perfect. It was then rolled into a
solid rail having no tendency to laminate, and that was one, and, in
his opinion, the main reason why a steel rail was so much more
durable than an iron one.
Having had for some years the charge of a long length of per-
manent way, he had carefully examined many defective rails with
a magnifying glass, to ascertain the cause of their failure, and he
found, in nineteen cases out of twenty, that the defects arose from
lamination of the iron resulting from defective welding. It was,
therefore, important to have a better, harder, and more homoge-
neous material for rails. But he thought that with harder rails
engineers would be tempted to increase the load to be placed upon
them; and that they had got into a somewhat similar dilemma to
the one to which the late Mr. Robert Stephenson used to refer
when he said, "It is of little use making improvements in locomo-
tives, because the result is to induce engineers to make still steeper
gradients."
Steeper gradients had, however, become a necessity; and, there-
fore, it was no doubt important to be able to pass heavier engines
over such gradients, but in doing so care should be taken in using
only such a load as would not be beyond the limit of elasticity of
the material of which the rails were made; for if every time a loco-
motive wheel passed over a rail it sank into it to some small extent
(and it undoubtedly did so), and if that sinking were not entirely
recovered when the wheel passed on, he need not say the destruc-
tion of the rail would not be far distant.
Mr. T. E. HARRISON, Vice-President, agreed to some extent
with the remarks made by the previous speaker; and representing
as he did a Railway Company in the North of England, the chief
part of whose rails were purchased from the manufacturers of the
district, he had minute observations taken of the wear of those
rails, and he might state as the result, that on between 700 and
800 miles of permanent way of the North Eastern Railway the
average duration of the last complete set of rails was found to be
about fifteen years and a half; and some which were laid down in
1834 were still in use. There were rails made in Wales which
had been in use in some cases more than twenty years which were
28
THE MANUFACTURE AND WEAR OF RAILS.
10
6
still in excellent condition. In other cases rails had to be taken
up after being down only a comparatively short time. On the
occasion of the examination of a number of rails taken from con-
siderable lengths of line, which had been down for a long time,
Mr. Harrison had advised that a careful analysis should be made;
and it had been found that the rails which lasted the longest, con-
tainedths to ths per cent. of phosphorus, and many of the
rails which exhibited the least signs of wearing were such as
would be generally considered bad iron. Many years ago large
quantities of rails had been purchased in Wales under peculiar
circumstances; they were called Hudson's' rails; and objections
were raised against their being used, because they were manufac-
tured under a contract, without a specification, and it must be
admitted that the fracture was that of very inferior iron. They
were, however, laid down on a part of the line where the traffic
was very heavy; and after fifteen years' wear, many of those rails
were still in good condition. This result was in accordance with
the views of Mr. Williams and of Mr. Vaughan, of Middlesboro',
that the specifications of Civil Engineers showed that they did not
thoroughly understand the best mode of manufacturing durable
rails. He accorded with that opinion; for when he compared the
rails made on a more expensive specification, with a view of getting
a more fibrous quality of iron, with Welsh or with North Country
rails made without a specification, he found they almost invariably
lasted longer than the rails made in accordance with the most elabo-
rate specification. After consultation with two of the Directors of the
North Eastern Railway, it was determined to collect samples of all
those rails which had been longest down, and to place them side by
side with samples of the most unsuccessful, and to have an analysis
made of the whole; and then they determined to send to all the
manufacturers in the locality, and to ask them to supply each a
thousand tons of rails, without giving any specification. They were
now in course of manufacture; and the prices at which the manu-
facturers had contracted to deliver them were below what they
would have charged for rails made according to the specification
previously used. It was intended to lay down alternate rails from
each of the nine manufacturers; and they would be laid where
there was the largest amount of wear. If they wore well it was
assumed that the manufacturer would be prepared to supply similar
rails for the future. The effect of this plan had been to put the
manufacturers of the North a little upon their mettle. One sample
of rails on the table had been down for twenty-four years: others
had been laid in places where eighty trains a day had passed over
them during fifteen years. Then there were two samples from
Messrs. Bolckow and Vaughan, made without specification, which
contained 0.5 per cent. of phosphorus. They were laid near the
THE MANUFACTURE AND WEAR OF RAILS.
29
Milford Junction, and had not been turned. Many thousands
had been taken from the main line and had been laid down on
some of the branches, where they would last for the next twenty
years. Seeing what could be done with iron rails lasting any-
thing like the period mentioned, the use of steel became unnecessary
as a matter of economy. At the same time, on portions of the
North Eastern line where there were stations and a great deal of
shunting, the rails would not last more than two or three years,
and in those cases steel rails had been laid down. A difficulty
with regard to the iron rails arose from the test to which they
should be submitted. No doubt iron which was best for wear would
not stand a heavy blow, and where the rails were submitted to the
test of a heavy falling weight they would be found deficient. The
difficulty lay in making iron rails of sufficient hardness, and at the
same time of sufficient strength for use. He could not refer to a
single instance of the breaking of ordinary rails when the line was
kept in good order; and the question arose, whether the test of a
heavy ball should be applied; if the road was in good order he
doubted whether any blow ought to come upon the rail and even
that could be met by putting in a few additional sleepers and
shortening the bearings. As a question of economy, with the cost
of the additional sleepers and chairs, an iron rail so treated would
be cheaper than one of steel. He was an advocate for using the
double-headed rail. The North Eastern system embraced a large
extent of branches used for coal traffic, on which the partially
worn rails were used. There were many miles of those lines on
which a new rail had never been laid since the rails were first put
down. These branches were maintained by rails culled from the
main line, and some of them were used without being turned.
Though a rail might be unfit for the main line, it would serve for
a less important part of the system; and under such circum-
stances the double-headed rail was useful. He was not prepared
to say that in all cases it was desirable to use it, because the
hardest iron might be applied to one head of the rail for wear
and a different quality to the bottom web; and a rail thus obtained
would stand a heavy falling test; and if the rails were not going
to be turned it was a matter of less importance. With regard to
the use of steel, he thought if the wear he had mentioned could
be got out of the iron rails it would be some time before steel
would be used to a large extent; but if the size of the locomotives
and the speed of the trains were further increased, steel rails might
become a matter of necessity upon the main lines. The rail-makers
in the North could make as good rails as they did formerly, by
going back to their old mode of manufacture; and it would then
be found that the ordinary rail would regain a position which, he
believed, to a large extent it had for some time lost.
30
THE MANUFACTURE AND WEAR OF RAILS.
Mr. W. R. INNES HOPKINS, being one of the Northern iron-
masters referred to, did not intend to make the rails that he pro-
posed to have tried on the North Eastern Railway according to the
formula of any Engineer. He was quite certain that, as Engineers
had increased their tests, they had augmented the strength of rails
at the expense of their wearing power, and it was a matter of
experience that a strong fibrous rail would not stand heavy traffic
nearly so well as one with a hard crystalline head. In paying
great attention to the wearing power of double-headed rails, they
were frequently rendered brittle, and became liable to be broken,
but experience showed that very far below the standard now used
for testing would give a sufficiently strong double-headed rail to
bear the heaviest traffic. Rails that had been laid down for many
years, in the North of England, and had stood the traffic well, on
being tested, were found to break at one-fourth of the weight
frequently prescribed, whereas rails that had become laminated,
and had quickly worn out, did not break under the operation of
testing. What was technically called 'strong iron,' did not weld
readily, and as the weldable property was diminished, the
homogeneous character and wearing power of the rail was destroyed.
On most foreign lines the Vignoles form of rail, with a broad
lower flange, had been adopted; this gave a greater opportunity for
combining wearing power with strength than any double-headed
section, because hard iron could be used for the head, and great
strength and power of resistance to testing could be obtained in the
web and the flange. He considered that the shape of many double-
headed rails was far from good; they were too quickly cut in under
the head; and many failures were caused by the side of the head
breaking off, from not being sufficiently supported beneath. The
section of the rail used by the Dutch Government showed a pear-
shaped head, which he believed was as good a form as could be de-
vised. There was some trouble at first in 'fishing' this rail, but
the Dutch engineers used an arched fish, which appeared, to a great
extent, to overcome the difficulty.
Mr. FOWLER, Past-President, considered it rather remarkable that
Mr. Harrison should have obtained better rails without a specifica-
tion than when he had given specific instructions; and he could
only account for it by supposing that the specification had been
drawn for some other rail-making district than that in which the
rails were made, for it was impossible to suppose that he would do
otherwise than give such directions as would lead to the manu-
facture of good rails. Mr. Williams stated that the rails should
be hard and crystalline, rather than fibrous. No doubt these were
desirable qualities, and it was equally so that a rail should not be
brittle. He could not agree in the opinion that specifications
were useless, and that engineers did not know what they wanted.
THE MANUFACTURE AND WEAR OF RAILS.
31
On the contrary, he thought that Mr. Hopkins was mistaken, and
that engineers did know what they wanted in a rail. Mr. Hopkins
might suppose that, because the engineer tested against the possi-
bility of brittleness by a falling weight, therefore he desired to have
great ductility: that was a wrong conclusion; but it was important
to be assured that the rail was not so hard as to be brittle; and
that was the object of engineers in some of their tests. Mr. Harri-
son's plan of getting nine manufacturers to lay down their rails.
under the same circumstances would afford an excellent test, and,
although without a specification, no doubt he would get good rails
laid down; but then it was desirable that he should ascertain exactly
the process adopted in the manufacture of those rails, and, after
making careful tests, he should analyse the results, otherwise, if the
manufacturers were allowed to lay down nine different kinds of
rails, and no observations were taken, he would be in no better posi-
tion than he was at first. In his opinion it was not desirable in the
interests of railway proprietors or of the public, that engineers should
abdicate their functions; but he would advise them to draw specifi-
cations, and to take pains to ascertain the process of manufacture of
the iron, and to confer with the manufacturers. In former times he
had got rails without a specification, paying an extra price for quality,
but his expectations were not realized, and he got worse rails for a
higher price.
It would be admitted that no rule could be laid down for the
manufacture of rails which would be applicable to all localities.
The question must be considered chiefly with regard to the mix-
ture of iron to obtain a certain result. He believed steel would
for almost all purposes eventually supersede the use of iron. But,
as in the case of all new applications, caution was necessary.
There could be no doubt, however, that in certain circumstances,
such, for instance, as on short lines for very heavy traffic, the use of
steel was most desirable. With regard to its extended use on main
lines, each case must be dealt with separately, taking into account
the quantity and quality of the traffic, and the varying price of the
material. The price of steel rails was falling; and the difference
between good iron rails and good steel rails would have a tendency
to diminish. At the same time he hoped it would never so far
diminish as to lead to the use of an inferior quality of steel rails. He
would advise engineers to adopt the plan of having their steel rails
carefully tested. The difficulties which led to the manufacture of
steel that was too hard were, he believed, being gradually overcome,
and existed now in a trifling degree only. The Tables that had
been produced were remarkable, but he did not think that the
Institution possessed at present sufficient data to induce it to
regard them as satisfactory. That good steel rails would last much
longer than iron rails no one could doubt; but he had no confidence
321
THE MANUFACTURE AND WEAR OF RAILS.
in a calculation which assumed that steel rails would last from
thirty years to eighty years, though Tables, such as those exhibited
by the Author, even when the life of steel rails was to some extent
assumed, were no doubt useful. There were advantages in the
use of steel rails which had not yet been alluded to. Taking the
case of a main trunk line, such as the Great Northern, Mr. Price
Williams had stated that the life of rails on that railway was eight
years.¹ There was no great object in having sleepers which lasted
longer than the rails, because few engineers would, on a main trunk
line, put new rails on second-hand or half worn-out sleepers. But
if steel rails were used lasting over a period of from sixteen years to
twenty-four years, it became immediately an important object to make
the durability of the sleeper as great as possible, because, in addi-
tion to the economy that would result from such an arrangement,
there was a material element of convenience in not having the
permanent way disturbed more frequently than was absolutely
necessary. If, instead of renewing the permanent way every eight
years, it became necessary to do so only every sixteen years, a great
advantage would be gained. It was well known that when one-
eighth of a main trunk line was always in the course of renewal, it
considerably interfered with the safe working of the line. Simulta-
neously with the use of steel rails it would be necessary to introduce
a superior quality of timber instead of what was called sleeper timber,
which consisted of the tops of trees of an inferior quality. It was
probable also that the form of the rail would be changed, and that,
instead of the double-headed rail with fastenings composed of many
parts, engineers would adopt the Vignoles form; this might be made
with steel, with the bottom so wide that there would be no difficulty
in having the base as large as that of the largest chairs. His own
experience was in favour of a base 63 inches in width. He consi-
dered chairs objectionable; but with a good Vignoles section of
rails, a bed of great simplicity, greater strength, and more security
would be obtained. That was one amongst the many advan-
tages arising from the use of steel.
Then, again, the question of turning would no longer be consi-
dered of importance when the material lasted thirty years. He
did not approve of the compound rail with a steel head and an
iron bottom, as the welding was sufficiently difficult whether in
iron or in steel alone, but that difficulty would be greatly increased
by the attempt to make a compound rail of steel and iron. In some
instances the use of such rails had been satisfactory, but in others
very unsatisfactory; and he would advise Engineers to be content,
either with good iron rails, or with good steel rails, and not to
attempt a combination.
Vide Minutes of Proceedings Inst. C.E., vol. xxv., p. 353.
THE MANUFACTURE AND WEAR OF RAILS.
33
Mr. ISAAC LOWTHIAN BELL admitted that complete success had
not hitherto attended the joint deliberations of engineers and iron-
manufacturers.
Although steel in some other form than that generally used
up to the present time had been alluded to, he apprehended that,
as no practical plan had been hitherto devised and adopted for
the manufacture of steel rails upon a large scale, it would hardly
be prudent to consider any other than Bessemer steel in the pre-
sent discussion. Any comparison of economy of course resolved
itself into a question of cost and durability. Nothing commended
itself more strongly to the mind of the manufacturer than the
production of the Bessemer steel, particularly as it was occasionally
practised in France, where pig-iron was run from the blast furnace
into the converter, and the rail was rolled direct from the bloom,
without passing under the hammer. No doubt that was a most
ingenious process, and one which might be brought into universal
application, but for other difficulties connected with the manufac-
ture of Bessemer steel, which did not obtain in the manufacture
of iron. In this last-named operation, the process of puddling
separated in a great measure the phosphorus from the iron; whereas
it was a peculiarity of the Bessemer process that the phosphorus was
condensed in the steel. In selecting iron for the Bessemer process,
it was important to consider the composition and the difference
in the properties of the various kinds of iron to be dealt with.
From his own analyses, and from the Tables of composition given
in Dr. Percy's "Metallurgy," it might be assumed that in Cleveland
pig-iron there was an average of about 11 per cent. of phosphorus ;
in North Staffordshire, about 1; in Derbyshire, 1; in Bowling,
51; in South Wales, 49; in South Staffordshire, 41; in the
Forest of Dean, 137; in Weardale, 10: and it was only in
hæmatite iron that the phosphorus fell so low as 047 per cent. He
believed that Mr. Bessemer himself held that, by his process, he
could not satisfactorily deal with iron containing above 015 per
cent. of phosphorus. The quantity of iron produced in the
British Isles, fit for making Bessemer steel, was therefore extremely
limited, because the total quantity of hæmatite iron was only about
one-fifth of the entire yield of the kingdom. Bessemer hæmatite
pig-iron, in consequence of the demand and of the limited make, was
worth 80s. per ton, whereas Cleveland pig-iron was worth only about
42s. 6d. per ton. The important question then arose-what other
supplies of iron were available for the manufacture of Bessemer
steel? The total supply of pig iron in Europe was about
eight and a quarter millions of tons. In France, where the
manufacture of Bessemer steel was looked upon as one of great
importance, there was scarcely a ton of ore fit for the manufacture
of iron suitable for the process. At Lavoulte, a small district
·
34
THE MANUFACTURE AND WEAR OF RAILS.
on the banks of the river Rhone, a small portion of suitable iron
ore was obtained; but by far the greater part was made from
ore procured from Mokta, in Algeria, which raised the cost of the
pig iron to something like £4 per ton. In Belgium, in West-
phalia, and in Silesia, there did not appear to be any iron ore fit
for the Bessemer converter. Norway and Sweden produced a
limited quantity; but not all of that was sufficiently free from
phosphorus. He had brought over from Norway a sample of
ore, and he had found that the pig-iron smelted from it contained
one-half per cent. of phosphorus. The Rhenish provinces raised
about five hundred thousand tons of iron ore per annum pure
enough for Bessemer iron. Styria and Carinthia also produced
a limited supply; but he believed that a yield of two hundred and
fifty thousand tons a year was all that could be depended upon from
that locality. He had no knowledge of the resources of Russia;
but looking at the distance of the ironworks from the shipping
ports, he imagined little help could be expected from that quarter.
There were no doubt other ores, as in Biscay, but so far as pig-
iron was concerned, the result was that, out of the eight and a
quarter millions of tons of pig-iron produced in Europe, there
was only about half a million of tons suitable for making Bessemer
steel.
Then the comparative cost of the manufacture of steel and of iron
rails was an important question, not only as it was at present,
but what it might be in the future. He thought that rails might
be made from pig-iron as cheaply by the one process as by the
other, so far as the mere cost of the operation for wages, fuel,
and so forth was concerned, and independently of the price of
the pig-iron employed. The loss, although considerable in the
Bessemer process, took place at a period of the manufacture when
the least amount of work had been done; and that was an ad-
vantage. He was aware that his opinion of the cost of manu-
facture was not admitted by the makers of the Bessemer rails.
One explanation of the cost might be, that the number of re-
jected rails manufactured in England was far in excess of the
number rejected abroad. He had been assured by a manufacturer
in Styria, that he was in the habit of rolling rails for a week at a
time without producing a single waste rail; but whether this pro-
ceeded from the fact that a superior kind of pig-viz., iron from
pure ores, and smelted with charcoal-was used, or from greater
fastidiousness on the part of the engineers in this country, he was
unable to say.
He had long entertained a doubt as to the amount of confidence.
to be placed in some of the specifications prepared by engineers;
and he still adhered to that opinion, notwithstanding the remarks
that had been made. He had some time ago suggested to Mr. Harri-
THE MANUFACTURE AND WEAR OF RAILS.
35
son, when the Directors of the North Eastern Railway were consi-
dering the question of using a rail of steel costing as much as £12
per ton, whether some improvement might not be introduced in the
manufacture of iron rails. The subject was of great importance to
Railway Companies, to British iron-masters, and to the country
generally, and was one which could only be properly investi-
gated by long and patient observation of those facts which
experience in the manufacture, as well as in the use of the rail,
could elicit; and hence both iron makers and engineers should co-
operate in any attempt to produce a good iron rail; and those who
might succeed in accomplishing this would be entitled to be con-
sidered as national benefactors, by maintaining the demand for that
description of metal produced in such large quantities in this
country. Not long ago a rail had been taken up, which had been
down for fifteen years on that portion of the North Eastern
Railway on which the traffic was the largest. From previous
observation, he had ventured to say that it was a bad specimen of
iron, so far as purity and freedom from phosphorus was concerned.
An analysis of the rail had been made with the following result:-
Silica
Graphite
.08
⚫07
Manganese
*36
Sulphur
•11
Phosphorus.
• 504
Iron.
98.11
99.234
Thus showing that it contained almost the amount of phosphorus
stated by Mr. Karsten to be the extreme limit. Other rails were
taken up which had been in use since 1855. The mean of two
analyses of these, made by Mr. Marreco, of Newcastle, was as
follows:
Phosphorus. Silicon.
Carbon.
Sulphur.
One specimen contained per cent.
.513
⚫079
Trace.
⚫090
• 307
•195
0
•109
99
•161
•168
0
Not estimated.
.223
A rail just received, and apparently of serviceable quality, con-
tained 681 per cent. of phosphorus.
As the same kind of pig-iron might on puddling yield a product
varying in quality according to the treatment it received, he had
no doubt that rail-makers might produce a hard or a soft kind of
iron; in proof of which he laid two samples of iron on the
table, puddled from the same pig, one bright and crystalline,
and the other fibrous and tough. In France this was actually
done in practice, the puddler being paid different prices for the
several qualities of iron. Manufacturers in England dared not

36
THE MANUFACTURE AND WEAR OF RAILS.
encounter the risk of making iron which, though strong enough
for rails, might still not resist the severe test-blow to which
Engineers subjected it; indeed when this power and species of
resistance was provided for, a sufficient quantity of No. 2 bar used
in the piles, and no mill cinders admitted into the blast-furnace,
it seemed to him that a rail specification was considered complete.
Of such importance was the presence of phosphorus considered
in France, that when it was found that railway bars were getting
too fibrous, mill cinders, or ores containing phosphorus, were intro-
duced, with the view of correcting what was regarded as a dis-
qualification in that kind of iron; so that it might turn out even-
tually that this hitherto maligned substance might be considered
indispensable in a good rail.
With regard to manufacturing steel by some other process, he
was not without hope from what he had witnessed, and from careful
study, that steel would, by the introduction of new processes, be
eventually produced from iron containing phosphorus in amount far
exceeding the limits laid down by Mr. Bessemer. Such a discovery,
he need not say, would be a valuable acquisition to the manu-
facturer, to the railway engineer, and to the nation.
Mr. JAMES DEAS stated that the North British Railway Com-
pany had used solid steel rails for six years, having laid down up-
wards of 1,100 tons; 500 tons upon main lines, and fully 600 tons
in switches and crossings. About one thousand sets of switches
and crossings had been put down, and not one of them had yet
worn down. Several sets of switches had been in the road for six
years, while the switches of iron rails that preceded them had
worn out in from six to nine months. Three years ago he had laid
250 tons of steel-headed rails, weighing 75 lbs. to the yard, in
24-feet lengths fish-jointed; they were the first rails of the kind
which were made by the Dowlais Iron Company. Out of upwards
of nine hundred rails, the steel had separated from the iron in
only twenty-five of them, and these rails had been turned and had
been left in the road with the iron heads uppermost. The steel
was laid flat on the top of the rail, and not turned over the edges,
as had been done in the Swedish Government rails designed by
Mr. Sandberg.
32
He had laid down alternately, at the same time, nine solid steel
rails and nine iron rails, on the outside of a 12-chain curve. The
original depth of the rails was 5 inches. When they were
examined a few days ago, the solid steel and the steel-headed rails
had worn inch, whilst the iron rails had worn inch. The
inside of the steel-headed rails was, however, worn much more
than the inside of the solid steel rails; and this was probably
accounted for by the greatest pressure of the flanges occurring at
the junction of the iron and steel.
32
32
THE MANUFACTURE AND WEAR OF RAILS.
37
Mr. W. MENELAUS said, that to make a rail of wrought iron which
would endure satisfactorily, under the conditions not unusual on at
least portions of most of the great lines of railway, was commercially
impossible. There were inherent faults, first in the quality of the
metal itself, and next in the mode of manufacture, which rendered
it, in his opinion, impossible to produce a rail of wrought iron
which would stand even a reasonable time under the trying con-
ditions to which some portions of the lines must of necessity be
exposed. No doubt it was possible to produce perfect homogeneity
and sound welding in wrought iron, as for instance in "best-best
tyres; but this could only be done at a great cost, and, beyond all
doubt, Bessemer metal could be produced cheaper than this high-
class wrought iron. Even granting that wrought iron of great
excellence could be produced in large quantities, which he much
doubted, even the best wrought iron ever made was far inferior to
Bessemer metal, at least for rails. In adopting steel for Bessemer
rails there was one great security; bad steel, in the sense of impure
steel, could not be made into rails. He frankly admitted that by
an error of judgment, or by attempting to make the metal hard
and lasting, an overdose of carbon might be given to it, and the
rails would in consequence be brittle; but if this mistake were
avoided, he held that bad steel could not be made into rails at all.
In fact, as a matter of economy, the purer and higher the quality
of the pig iron used, the cheaper must be the Bessemer steel rails
produced. As far as his experience went, nothing was so bad
and so wasteful as to attempt to use pig iron of low quality for
making Bessemer metal: if silicon was present in excess there was
a bad yield in the converter; and the presence of undue quantities
of sulphur and phosphorus caused a great number of waste rails. He
had found it best and cheapest to buy the highest quality of pig
iron only. His experience of Bessemer metal, however, had been
confined entirely to the manufacture of rails, but he had no doubt
all steel makers would bear him out in these statements. He
believed the only charge brought against the quality of steel rails
was their brittleness, as it was called, and he supposed that rails
made of Bessemer metal had broken in regular work, even when
produced by manufacturers who from their long experience in
the trade actually claimed pre-eminence. His opinion was that
these hard rails had been made, not designedly, as a matter of
economy, but simply by accident, or through an error in judg-
ment; mistakes were less likely to happen now that considerable
experience had been gained in working the metal. Excellent
iron rails could be made if railway engineers would but treat
ironmasters with confidence and would pay a fair price; but when
a high class rail was required he was of opinion that, as far
as was known at present, there was no material to be com-
E
38
THE MANUFACTURE AND WEAR OF RAILS.
pared to Bessemer metal for excellence of quality and moderate
cost.
Now as to the union of the two metals in the shape of what
were called steel-headed rails. Very unsatisfactory results followed
the first attempts to make rails of that description; but no doubt
other makers had, like himself, benefited by experience, and now a
perfect junction could be insured between the iron in the body and
the steel on the heads of rails. He first used plain slabs, and he
had made, some years ago, by this method, a quantity of rails for
the Edinburgh and Glasgow Railway. Mr. Deas reported that
the rails had, so far, stood well, and he still expressed confidence
in rails of this description. These rails were made with plain
slabs for the tops, and the union of the two metals depended
entirely on the weld. After a while he adopted a system for
making steel heads, which had been long used in South Wales for
iron; that was, giving a mechanical grip to the top slab in the
rail-head. The slab of iron was rolled in the form of channel iron,
with a rib more or less deep on each side. In the finished rail
the steel was thus not only welded to the iron, but it had a good
mechanical grip upon it. This plan of rolling the steel slabs in
the channel form was a great improvement. All he had arrived
at, however, was to get a good grip of the head: but Mr. Sandberg
having the management at the Dowlais Works of the manufacture
of a quantity of steel-headed rails for the Swedish Government,
suggested that the horns or ribs of the channel-formed steel slabs
should be deepened, so as to insure a complete envelope of steel for
the head; and it was found that, in the rail rolled from the slabs
designed by Mr. Sandberg, the steel actually came down into the
stem of the rail, and made an excellent sound job.
It was by no means a simple matter to roll steel and iron
together, as the two metals required to be worked at different tem-
peratures; but after a little practice the men acquired sufficient
skill in the management of the pile in the furnace to secure an
excellent junction between the two metals, and in his opinion
steel-headed rails, properly made, would be found to stand almost,
if not quite, as well as rails of solid Bessemer metal, though
of course they would not be reversible, having only one table of
steel.
Mr. JAMES SHAW felt from his experience at the Stockton Iron
Works, that he could confirm a great deal that had been said
regarding iron rails by Mr. Bell, whose remarks should be studied
by all Engineers.
Many specifications had come to him from Railway Engineers;
and though some of them had exhibited great practical knowledge,
others had been so deficient that he had been utterly at a loss how
to make the rails. Much had been said about the difference
THE MANUFACTURE AND WEAR OF RAILS.
39
between iron and steel, but he thought that a vast deal depended
upon price, upon the judgment of Engineers, and the advice which
they gave to Railway Directors. When a manufacturer quoted a
price he believed that, as a rule, the lowest tender would be
accepted. He knew that the Railway Company desired to get the
best apparent rail at the lowest cost, and he might be disposed to
believe that some of the conditions were unnecessary. The North
Eastern Railway Company had taken the right course in asking
a manufacturer to tender. They had said to him, "We will take
you into our confidence, although our Engineers have issued excel-
lent specifications, because you know far better how to manufac-
ture rails than we can be expected to do." This example had been
followed by a Foreign Railway Company, and the results in both
cases were very satisfactory.
The question was one of great importance, inasmuch as it con-
cerned the safety of the lives of the travelling public. He would
ask whether it was creditable that there should be issued some one
hundred and fifty different specifications. In France that was not
the case, and even in America a uniform specification was adopted.
In regard to the manufacture of Bessemer rails a good deal
appeared to have been taken upon assumption. As experience of
steel rails extended back only some five or six years, it was, he
submitted, somewhat rash to assume that steel rails would of neces-
sity last sixty years or one hundred years. Steel rails had certainly
the advantage of being more homogeneous, and were therefore less
liable to lamination, yet hardly sufficient regard appeared to be
paid to other and very important considerations relating to extra
cost and the possibility of having bad steel rails as frequently as
bad iron rails: he therefore could not implicitly receive the statistics
which had been referred to.
Mr. VIGNOLES, Vice-President, thought, from the statements
that had been made by the manufacturers and by all who were
best acquainted with the subject, it might be concluded that if a rail
from a homogeneous material could be produced the result would
be better than if it were produced from a pile. Engineers knew per-
fectly well what they wanted, although, perhaps, not being practical
manufacturers of the material, they might not be able to dictate
the best way of arriving at a given result. Besides, they were con-
trolled to a great extent in this country by Directors as regarded
expense, and abroad particularly by the assumption that foreign
Engineers and managers knew a great deal more than the English
Engineers did. Those who had the misfortune of contending with
people half informed on technical subjects, which was a dangerous
thing, would know how difficult it was to satisfy the requisitions
of the foreign Engineers, or of those who controlled them. Within
the last few days he had received a communication from the officials
E 2
40
THE MANUFACTURE AND WEAR OF RAILS.
of one of the most important Governments in Europe refusing to
allow an eminent manufacturer in England to replace rails made.
by him which had been condemned on unreasonable grounds. In
order to resist serious impact, one quality of iron was required,
while a different quality was called for where rails had to sustain
a load, and those two perfectly dissimilar requisitions were expected
to be combined in the same rail. A rail sufficiently homogeneous
would combine those qualities to a certain extent. No doubt a steel
rail would wear better than one of iron, and a steel-headed rail
would wear better than an iron rail; but, in the latter case, the
difficulty lay in combining the two. He admitted that the best
iron was that which would weld the best; but it had been said that
the best iron could be obtained from the puddle ball; and this had
not hitherto been supposed possible. He would be willing to adopt
the idea of accepting a guarantee from a manufacturer as to the
quality of the metal; but he would like to know the price at which
the rails were to be produced, and also the mode, and, as far as
possible, the result of the wear and tear.
Mr. DAVID FORBES observed, that in all cases homogeneous
metal was to be preferred to that which was composed of more or
less hard and soft particles; and therefore it became in the first
place a question as to how such homogeneous metal could be
obtained. Unless iron could be brought into a state of complete
fusion, it would not be practicable to get a material having in all
parts the same mechanical structure, or the same mechanical
composition. As fused iron had not yet been found available,
recourse must be had to a compound which was somewhat more
fusible and easily manufactured, as, for example, cast steel, which
could be produced by the Bessemer, the Siemens, or any other
good process. The question became then one of price, and in that
respect the Bessemer process at present had the advantage. No
doubt good steel could be produced from bad materials; though in
the present state of metallurgy not so as to be remunerative. It
was quite possible to take the worst cast iron, and to oxidize so
much of it as to leave a residue of good steel; but he considered
the Bessemer rails had suffered in the estimation of Engineers
from the attempts that had been made to use a material which was
bad and unfitted for the purpose. He had made some experiments
which had been suggested by the opinions of Mr. Siemens, that
instead of using a steel rail an iron one containing phosphorus
might be employed. The results showed that such a rail would be
an extremely dangerous one; since, although at times very strong,
such rails were liable to be very 'cold short.' It was found ex-
tremely difficult to get the phosphorus so uniformly disseminated
throughout the mass as to produce in all parts the same hardness,
or the same non-liability to break. The main point was to insist
THE MANUFACTURE AND WEAR OF RAILS.
41
upon the employment of a good raw material for the production of
the Bessemer steel.
Mr. H. CONYBEARE was of opinion that the section of a rail
materially affected its duration, and that this was especially the
case as regarded the design of the head and of the upper surface or
wheel-track. It was obvious that each portion of the rail had its
special and definite functions and conditions to fulfil, and a careful
consideration of the character of these should supply rules that
would regulate with certainty the contours and the dispositions
and scantling of the metal in every portion of the section, that
would best conduce to the fulfilment of such conditions.
Some years ago, having occasion to design the rails for Irish and
Welsh railways, he had collected, with a view to determine the
best form of rail, all the sections he could obtain from the Welsh
iron-works, and the London metal-brokers; and he compared and
collated these with a view to ascertain the extreme range of the
variations in practice, and the limits of economy and practica-
bility in manufacture. He also calculated the strain to which each
portion of the section was subjected, and considered the conditions
it had to fulfil, and the extent to which these conditions were com-
plied with in each example. Having done so, he endeavoured to
establish definite rules which should fix absolutely the contour and
the proportionate scantling of every portion of the rail section.
These he had since carried out in practice, and the results arrived
at were exhibited in the following diagrams and Tables. It would
be seen that the diagrams, though also applicable either to the T
rail or the double-headed rail, referred more especially to what was
known as the Vignoles section. Everywhere except in England
the double-headed section and the chair road had long been obso-
lete, and he thought it was much to be regretted that it was not so
in this country. It was obviously essential to the perfection of any
system of permanent way, that the jointing of the rails to each
other, and the mode of attaching the rail to the sleeper, should be
of such a description as to ensure absolute uniformity of gauge and
of cant, and such as to admit of no jar, shake, or movement, which
might loosen the fastenings, and thereby induce hammering and
concussion, which must conduce to the disintegration of the bearing
surface of the rail, and to the injury of the rolling stock; and
judged by this canon, a Vignoles rail, directly fastened by fang-bolts
into beds planed by machinery, in rectangular sleepers, was as supe-
rior to a road in which chairs intervened between the rail and the
sleeper, as the best road of the latter description was to the old
stone sleepered and unfished chair road. As regarded the weight
of the rails, it was possible to make them too heavy; for modern
rails failed not through their deficient strength as girders, but
through the deficient homogeneity and hardness of texture in the
42
THE MANUFACTURE AND WEAR OF RAILS.
bearing surface presented to the rails, which induced the lami-
nation and hammering out of such bearing surface under the con-
centrated pressure and violent impact of the wheels.
It was
essential that the rails should possess an amount of vertical stiffness
sufficient to confine their deflection under the passing load within
unobjectionable limits. But any weight or vertical stiffness in
excess of what would suffice to fulfil this condition was absolutely
injurious; for as a certain amount of elasticity was essential to the
durability of the rails, an excess of rigidity invariably tended to
destroy the rail surface, by acting as an unyielding anvil placed
beneath it, between which and the hammering action of the driving-
wheels the bearing surface of the rail was rapidly beaten out into
laminæ, and so destroyed; and the heavier the anvil the more
rapid was the process of destruction. If extra weight and speed
of traffic were to be provided for, improved quality in the material
was required rather than increased quantity,-a harder surface,
not heavier metal.
4
Thus the rails most remarkable for their longevity had been also
most remarkable for their abnormal weakness and deffectibility as
girders; for example, a length of 25 miles or 30 miles of the ori-
ginal permanent way of the South Western Railway consisted of a
75 lb. rail, laid on transverse sleepers 5 feet apart from centre to
centre. The strength of such a rail as a girder was therefore only
equal to that of a 571 lb. rail laid on 3-feet bearings; and as,
according to Barlow's experience, the deflection of the same rail,
when supported on 5-feet bearings, was to its deflection when the
supports were 3 feet apart, as 64 to 24, the flexibility of this rail
was threefold that of a 75 lb. rail on 3-feet bearings. Here was,
therefore, a permanent way abnormally weak and flexible, its strength
being only equal to that of a 574 lb. rail on 3-feet bearings, and
its flexibility at least four times greater than that of a 75 lb. rail
on 3-feet bearings. When he last obtained particulars of this
permanent way, these rails had been laid down for eighteen.
years, and were still in perfect order. He believed that more was
lost than gained by increasing the weight of even a double-headed
rail above 70 lbs. per yard, which appeared to him to be equal to
the heaviest traffic; and with cross-sleepers equidistant from centre
to centre, and with a section of equal weight per yard, a Vignoles
road was stronger and stiffer than a chair road; for the bearing
of the rail on a chair was only 3 inches, which reduced the clear
bearing only from 36 inches to 33 inches, the square of which was
1,089 but the bearing of the Vignoles rail on machine-planed
beds was from 9 inches to 10 inches, according to the width of the
rectangular sleepers used, which reduced the clear bearing to
27 inches or 26 inches, the squares of which were respectively
729 and 676; and the strength being inversely as the square with
THE MANUFACTURE AND WEAR OF RAILS.
43
the sleepers 3 feet apart from centre to centre, it followed that the
strength as a girder of a Vignoles rail would be to that of a double-
headed rail of equal weight in a chair road, (the sleepers being
the same distance apart from centre to centre,) in the ratio of
1,089 to 729, or of 1,089 to 676, or nearly as 3 to 2. He there-
fore considered that the weight of a Vignoles rail should never
exceed 70 lbs. per yard, nor be less than 65 lbs.: the minimum
would be 62 lbs., if dog-fastenings were substituted for through
fang-bolt fastenings; but such a substitution was, in his opinion,
much to be deprecated.
The section of a rail admitted of three primary subdivisions-the
head, the vertical web, and the foot. The same division applied to
the double-headed section, merely substituting a second head for
the foot. Thus, the principles on which the rail-section, shown in
Fig. 2, p. 47, as outlined, applied equally to the Vignoles and the
double-headed sections.
The width of the head ranged in practice from 23 inches to
21 inches. The former was till lately the rule in England;
but
the flat-footed rails, which had almost superseded every other form
of section on the Continent and in America, and which were
rapidly gaining ground in England, were usually only 21 inches in
width of head. In Mr. Brunel's latest specification for a 75 lb.
double-headed rail for the Eastern Bengal Railway, the width was
only 2.3 inches; on the Grand Russian Railway it was 2-3 inches;
on the Cape Railway and Swedish Government lines the same; and
the rail of the Kustendjie Railway, of the Royal Swedish, and many
others, only 2.25 inches. He considered 2-25 inches the proper
width for lines with light and medium traffic, and 2-3 inches the
maximum for lines with the heaviest traffic. In fact, it could be
demonstrated that nothing was gained by increasing the width
above 2.3 inches. If, indeed, the wheel-treads of the rails were
horizontal planes, and the peripheries of the wheels truly cylin-
drical, so that the contact or bearing of the wheel with and on
the rail were conterminous with the width of the latter, there would
be a manifest advantage in making such bearing surface as wide as
possible. But in practice, the contact of the wheel and rail at any
one moment was merely that of a cone laid transversely on a
cylinder, and such contact, instead of extending the whole width of
the rail, would, were the materials unyielding and non-elastic, be a
mathematical point without length or breadth; and, after allowing
for elasticity, the actual contact at any one moment had been stated
by Mr. Brunel never to exceed th of an inch in width. He
thought this an under-estimate; but, however that might be,
it was obvious that the width of actual contact at any one moment
could never amount to any considerable fraction of the width of the
rail, and therefore that nothing could be added to the width of
44
THE MANUFACTURE AND WEAR OF RAILS.
actual contact, that could in any way diminish the stress of the
wheel on the wheel-tread, by increasing the width of the latter
beyond the limits he had laid down. By so increasing the width of
the rail-head, the overhang or false bearing of the shoulder of the
rail over the vertical web that supported it was increased, and
thereby, pro tanto, the cross strains to which the material of the rail
was subjected. The contour of the wheel-tread was obviously the
point of paramount importance. It was by the lamination at this
surface that the rail perished. It was the weakest link in the
chain, and one which, from the peculiar and exceptional con-
ditions of the case, could not be strengthened adequately.
In ordinary cases of engineering, when a structure of given ma-
terial was designed for the purpose of carrying a load, the amount
of such load was first ascertained: secondly, the ultimate strength
of the material; and finally, an ample margin was allowed for
safety. For example, the ultimate strength of wrought iron being
25 tons per square inch, the load on a wrought-iron girder-bridge
was not allowed to exceed 5 tons per square inch. But it was,
unfortunately, impracticable to apply this rule to the apportioning
of the bearing surface of the rails to the load on the locomotive
driving-wheels. The case was a wholly exceptional one in engi-
neering; for not even the smallest margin for safety was practi-
cable. It was impossible to avoid loading the bearing surface to an
extent much greater per square inch than the greatest that any such
material would be subjected to, if it could possibly be helped; for the
point of contact of the wheels with the rail, supposing the material
of both to be unyielding and non-elastic, would be not a surface at
all, but merely a mathematical point, and the load on the driving-
wheels of a locomotive was often 11 tons, the whole of which might
at any moment be thrown by a jerk on a single rail.
The only circumstance that could save the bearing surface from
destruction under such an emergency was the elasticity of the
wheel and the rail, which allowed the periphery of the former to
flatten at its contact with the latter, and the rail surface to lap
round the contact of the wheel, so as to convert what would
otherwise be a bearing point into a bearing surface; and it was
desirable to encourage this action as much as possible, by giving
the rail as much elasticity and as flat a bearing surface as was com-
patible with the observance of other essential conditions.
In the case of a railway with a central guide-rail, it was obvious
that the wheels of the rolling-stock should be perfectly cylindrical,
and the wheel-treads of the rails horizontal planes. In such a case
the pressure of the wheel-tyre would be distributed over the entire
width of the rail-a condition more favourable to the durability
of both tyres and rails than when the pressure on the wheels
was concentrated on a scarcely appreciable fraction of the width
THE MANUFACTURE AND WEAR OF RAILS.
45
of the rail-as must be the case when the wheel-tread of the
latter had for its contour a circular arc. It should be attempted
in ordinary practice to approach these conditions as nearly as the
coning of the wheels, coupled with the clearance or play of ths of
an inch allowed between the flanges and the inner surface of the
rails, and also the allowance for the inaccuracy in the gauge and in
the angle of the wheel-cant, due to imperfections in the laying and
working of a chair road, would permit, by making the radius with
which the profile of the wheel-tread was struck as large, and conse-
quently the arc as nearly approximate to a plane surface, as the
conditions of the case allowed.
As the wheels were coned to an inclination corresponding to the
cant of the rails, such coning would perfectly accord with a flat-
headed rail, provided the relative position of wheel and the rail were
always kept the same by a central guide-rail; but the play allowed.
between the flanges and the rails, coupled with this coning to a
small extent, and the inaccuracy of gauge and cant incidental to
the laying of a chair road, in a far greater measure caused an
appreciable amount of angular divergence between the profile of a
perfectly flat wheel-tread and the rectilinear profile of the wheel-
tyres. Sometimes the latter would bear on the inner edge and
sometimes on the outer edge of the former. It was to obviate this
that the wheel-treads had been given a rounded contour. The
principle was correct; but the amount of such rounding could be.
demonstrated to be almost inappreciable, and was generally in-
juriously overdone. The proper amount of rounding to be given
to the upper surface of the rail-viz., the radius with which the
circular arc forming the profile of the wheel-tread should be
struck-admitted of accurate determination. The angle of the are
should be twice that of the maximum angle of divergence between
the profile of the cone and the wheel-tread, supposing the latter
to be flat.
The amount of the portion of this angle of divergence which was
due to the coning of the wheels, and the clearance or play allowed
between the wheel flanges and the inner surfaces of the rails, might
be thus determined. As the amount of such clearance only allowed
the wheels to deviate inch on each side of their centrical posi-
tion; and, as the inclination of the side of the cone of the wheel
line was only th, the utmost limits of such deviation could only
raise one end of the axis th of an inch, and depress the opposite end
to the same extent; and, taking the length of the axis in round
numbers as 60 inches, this would only be equivalent to an incli-
nation of th of an inch in 60 inches, i.e., of 1 in 1,200, and the
angle of deviation of the profile of the wheel from that of the rail
head due to this cause would be, of course, the same
that was
1 in 1,200 equal to an angle of 0° 1' 51". To this had to be added
1
20
40
46
THE MANUFACTURE AND WEAR OF RAILS.
the angular variation in the cant and gauge of the rail due to the in-
accuracies incidental to the laying and wearing of a chair road.
The bed of the chair on the sleeper was 10 inches, and the bed was
cut to a template; and if the limit of inaccuracy in the cutting and
wearing of the bed were assumed to be about 4th of an inch at one
end or the other, this would be equivalent to a variation of, or
equal to an angle of 1° 51′ 1″, and this, added to the angle of varia-
tion due to the coning of the wheel and clearance, would make the
total angle of divergence A B C and A' B' C' on Fig. 1 equal to
1° 27′ 51″.
Fig. 1.

A
C
B2
ཅན་གག་ཅ་
་ ་ ་
A!
B
C!
}
1
83
Diagram demonstrating that when B'B = 1½ inch, and the head of the Rail is, therefore, 24 inches
wide, a radius of 40 inches will give ample curvature for the wheel-tread.
The angles DB'B and DB B' are each equal to 880 32 9". The angles CBA and C'B'A' are thus
each equal to 1° 27′ 51″, and B'DB measures 2° 55′ 42″.
In this Figure, B' B represented the bearing surface of a rail,
say 2 inches wide, and A B and A' B' the extreme positions of the
wheel tyre on such a rail, supposing it flat topped; these positions
being inclined to the rail surface at an angle of 1° 27′ 51″. To
determine the radius with which the profile of the bearing surface
THE MANUFACTURE AND WEAR OF RAILS.
47
of the rail must be described, so as to prevent the weight being
ever thrown exclusively on either edge of the rail, and so that each
portion of the surface might take its turn equally with the other in
bearing the weight, draw through B and B' the extreme points of
contact of the wheels and rail surfaces, the lines B D and B' D in a
direction perpendicular to the lines A B and A' B', then the point
of intersection of B D and B' D would be the centre from which
the circular arc forming the profile of the wheel-tread should be
struck; for it was clear that the profile of the wheel contour would
be tangential to this arc in every possible position of the wheels.
The angles E B' D and E B D being obviously similar angles,
therefore B' D B was equal to twice the angle A B C, or 2° 55′ 42″,
and, as the side B' B of the triangle B' D B was equal to 2 inches,
and the opposite angle was equal to 2° 55′ 42″, the length of the sides
BD and D B was 54 inches, which was, therefore, the radius with
which the are forming the profile of the wheel-tread should be
struck. If B' B was equal to only 13 inch, this radius would be
equal to 40-75 inches. This allowed for the inaccuracy of laying
and wearing incidental to a chair road, but with a Vignoles rail a
much flatter one would suffice. He was, therefore, warranted
in adopting 40 inches as the proper radius of the profile of the
wheel-tread in all cases.
Fig. 2.

E
B
C
H
K
Construction of Rail Section. Half full size.
The wheel tread has a radius of 40 inches.
The vertical web is profiled with a radius of 12 inches.
The upper and lower fish planes are at an angle of 60° to each other.
The radius with which the shoulders of the wheel-tread were
rounded off was at first only of an inch: this was found in-
↑
48
THE MANUFACTURE AND WEAR OF RAILS.
sufficient, and the present practice ranged from a radius of 45 inch
to 7 inch. On the Grand Russian, and in the section adopted by
Mr. Samuel, the radius was only 45 inch. In the rail chosen
by Mr. Fowler for the Great Southern and Western Railway of
Ireland, and also in the Royal Swedish, the radius was 5 inch; in
the Great Northern of France, 55 inch; and in the Eastern Bengal
Railway, 6 inch.
The radius of the shoulder should not be less than 5 inch, nor
more than 6 inch; and, in his opinion, the former dimension
answered all necessary conditions, and anything more than that was
injurious, as detracting from the width of the wheel-tread.
The flank of the shoulder should be vertical, as the proper
function of the metal disposed in this portion of the section was
literally to keep the shoulder up to the wheel, and it was obvious
that the best disposition of material for this object was to carry
down the flanks of the shoulder vertically till they met the upper
fishing planes. This disposition of material had also the advantage
of providing a better bearing for the fish-plates.
At its lower termination the vertical plane of each shoulder flank
was met by the surface intended to receive the upper bearings
of the fish-plate.
It was indispensable that the upper and lower bearing surfaces.
of the fish-plate and rail should fulfil two conditions; first, that
they should be plane surfaces; and, secondly, that they should be
equally inclined to the axis of the fish-bolt, which should be placed
centrically between the upper and the lower bearing surfaces.
The necessity of the fulfilment of these conditions appeared to be
obvious. If the surfaces were curvilinear, the principle of the
wedge was excluded, and the slightest inaccuracy in the manufac-
ture would prevent the fish from obtaining a fair and continuous
bearing on the rail. It was equally obvious that the bolt which
pressed these fish-plates into apposition must be centrical, otherwise
the pressure upon the upper and the lower bearing surfaces would be
unequal, and the fish to that extent defective. The same objection
applied to inclining the upper and the lower bearing surfaces of the
fish at a different angle; the pressure on the upper and the lower
bearing of the fish would in that case manifestly differ in the same
ratio as the sines of the angles of their upper and lower bearing
surfaces, and the effect would be the same as if the angles being
equal, the distance between the upper fishing-plane and the centre
of the fish-bolt, and that between such centre and the lower fish-
plane, differed in the same ratio.
The functions which the vertical web had to fulfil were simply to
maintain the head and the foot of the rail in the same relative posi-
tion with itself and each other, under any lateral or vertical strains
to which they might be subjected in the working of the line. In
THE MANUFACTURE AND WEAR OF RAILS.
49
modern practice the centre thickness of this vertical web ranged
from 55 inch, which was little more than the minimum to which it
could be rolled without unduly increasing the cost of manufacture,
to 7 inch. A favourite dimension for the centre thickness of a rib
that increased in thickness towards its junctions with the head and
foot of the rail was 6 inch; but 65 inch, or 7 inch were common
dimensions when the section of the vertical web was parallel
throughout from head to foot. He had adopted in these Tables
(pp. 51-2) in every case a top and bottom thickness of 7 inch, and
a thickness at the centre of 6 inch, the profiles of the sides of the
vertical web being circular, and drawn through these three points,
i. e., with a chord, equal to the depth of the web, and a versed sine
equal to 05 inch, i. e., half the difference between the top and
bottom width 7 inch, and the centre width 6 inch.
•
Horizontal lines drawn through the intersection of these arcs,
with the prolongation of the fishing-planes, formed the demarcation
between the head and vertical web and the vertical web and foot;
therefore any one of the forty-one vertical webs in the series given
in the Tables (pp. 51-2) might be joined to any of the thirty
heads or to any one of the twenty-six feet, and thus about thirty-
two thousand varieties of rail section might be formed from such
combinations. The re-entering angle between these arcs and
planes was filled in by an arc of 3 inch radius.
The fixing of the angle, at which the fish-planes converged at
60°, was the result of a compromise. Manufacturers preferred a
larger angle, so as to give the head the pear-shaped section of the
Erie rail: but this increased the foot of the rail, and rendered it
difficult to form a good fish, whereas with an angle of 60° as
strong a fish might be formed as had ever been made. The only
effect on the fish of diminishing the angle from 60° to 50° was that
in the latter case an equally strong fish might be obtained with less
scantling in the fish-plates and fish-bolts. But no such increase of
scantlings would be necessary with the insistent fish.
The usual depth of the double-headed rail was 5 inches, but, as
already shown, the strength of a Vignoles rail, laid on sleepers
3 feet apart from centre to centre, exceeded that of a double-headed
rail of equal section and depth laid on sleepers the same distance
apart nearly in the ratio of 3 to 2; and, therefore, to ensure the
same strength and stiffness, the depth of the Vignoles rail might
be, as compared with that of the double-headed rail, in the ratio of
2 to 3. In practice the depth had hitherto been in excess of this,
for convenience in fishing only. The Royal Swedish rail and the
Grand Russian rail had each a depth of 45 inches; the Moray-
shire, the Kustendjie, the Great Southern and Western of Ireland,
and the Swedish Government rails, had each a depth of 41 inches,
while the rails for the Cape were 4 inches deep, and his own
50
THE MANUFACTURE AND WEAR OF RAILS.
65 lb. section, laid on the West Cork, the Cork and Kinsale, the
Cork and Macroom, and the Cork and Limerick Direct Railways,
had a depth of 4.3 inches, though for heavy traffic, he made the
depth 4.55 inches.
•
He now came to the foot of the rail. The dimensions of the
trapezium, C D F E, (Fig. 2, p. 47) which formed the upper portion
of the foot, and the sides of which, F D and C E, constituted the
lower fish planes, should be the same for all weights of rail. Its
top width, CD, was 7 inch, its bottom width 2.1 inches, and its
depth 45 inch. From C and D the profiles of the upper surface
of the foot extended at an inclination downwards of 1 in 10, till
they attained the full width intended to be given to the foot, and
from these terminations vertical lines-G IH K—were dropped
375 inch in depth, for the thickness of the toe, which was fixed
by manufacturing considerations, and was the same in all cases, and
a horizontal line drawn through I to K completed the profile of the
foot.
The width of the foot should be at least sufficient to admit of
through fastening, consisting of bolts passing through the sleepers,
fixing into fang-nuts abutting on its lower surface. He thought
the practice of nailing the rail to the sleeper by exaggerated brads,
called dog-bolts, could not be too strongly deprecated. He had
found a width of 43 inches sufficient for fulfilling these conditions ;
and for the heaviest traffic, it appeared to him that a width of
5 inches was ample. If this width were exceeded, the difficulty of
rolling, and consequently the cost of the rail, would be greatly
increased.
He was aware that there were advocates for making the foot, or
lower flange, very wide in proportion to the depth of the rail, in
order to obtain a greater amount of lateral resistance: but regarding
a problem of this sort, it might be said solvitur ambulando. The
Vignoles section had long been in general use, both in Europe and
America, and on both continents, after experience gained over a
period of ten or fifteen years' working, and on a length of over
10,000 miles of railway, a width of 4 inches had been found suffi-
cient, and was still adhered to. And if a base of 4 inches was thus
sufficient with only dog-bolt fastenings, a fortiori, must a width of
4 inches be ample with such superior attachment to the sleepers
as was effected by through fastenings, such as fang-bolts.
He exhibited a diagram containing sections of thirty heads of
rails, varying in top width from 2.25 inches to 2.5 inches, the
bottom width of all being 7 inch, and in depth of shoulder from
•9 inch to 1.0 inch, the total depth ranging from 1.34 inch to
1.521 inch, and the weight increasing from 25.76 lbs. to 32:39 lbs. per
yard. On the same diagram were also shown forty-one vertical webs,
warying in depth from 1.9 inch to 2.9 inches, the width at the top
THE MANUFACTURE AND WEAR OF RAILS.
51
and bottom being 7 inch, and the thickness at the centre 6 inch,
the weight varying from 12.2 lbs. to 18-20 lbs. per yard. There
were also shown twenty-six sections of feet, varying in depth from
898 inch to 1.023 inch, and in width from 4 inches to 6 inches,
and in weight from 24:06 lbs. to 39-43 lbs. per yard. Every
one of these heads would fit on to every one of the vertical webs,
HEADS varying in Width from 2.25 inches
to 2.5 inches, and in Depth from 1·340
inch to 1.521 inch, and in Weight from
25.76 lbs. to 32 39 lbs.
VERTICAL WEBS, varying in
Depth from 1.9 inch to
2.9 inches, and in Weight
from 12.2 lbs. to 18 2 lbs.
•
Width
Depth
No.
in
Inches.
in
Inches.
Depth to
Bottom of
Shoulder.
Weight
Depth
Weight
in
Lbs.
No.
in
in
Inches.
Lbs.
123+
2.25
1.340
.9
25.76
1
1.9
12.2
1.374
⚫925
26.32
2
1.925
12.35
1.396
.95
26.88
3
1.95
12.5
4
1.421
·975
27.41
4
1.975
12.65
>>
1.446
1.0
28.01
5
2.0
12.8
6
2.30
1.361
⚫9
26.55
6
2.025
12.95
7
1.366
•925
27.12
7
2.05
13.1
8
1.411
.95
27.70
S
2.075
13.25
9
1.436
⚫975
28.27
9
2.1
13.4
10
1.461
1.0
28.85
10
2.125
13.55
11
2.35
1.376
.9
27.33
11
2.15
13.7
12
1.401
•925
27.92
12
2.175
13.85
13
1.426
•95
28.51
13
2.2
14.0
14
1.451
•975
29.10
14
2.225
14.15
15
1.476
1.0
29.69
15
2.25
14.3
16
2.40
1.391
.9
28.40
16
2.275
14.45
17
1.416
.925
29.00
17
2.3
14.6
18
1.441
.95
29.60
18
2.325
14.75
19
1.466
⚫975 30.20
19
2.35
14.9
20
1.491
1.0
30.80
20
2.375
15.05
21
2.45
1.406
.9
29.20
21
3.1
15.20
22
1.431
.925
29.81
22
2.425
15.35
23
1.456
.95
30.42
23
2.45
15.5
24
1.481
•975 31.03
24
2.475
15.65
25
1.506
1.0
31.65
25
2.5
15.8
26
2.50
1.421
⚫9
29.89
26
2.525
15.95
27
1.446
.925
30.52
27
2.55
16.1
28
1.471
⚫95
31.14
28
2.575
16.25
30
29
1.496
⚫975 31.77
29
2.6
16.4
30
1.521
1.0
32.39
30
2.625
16.55
31
2.65
16.7
32
2.675
16 SS
CONSTANTS.
33
2.7
17.0
+
In all cases, the width of the top of the foot, of the
bottom of the bead, and of the top and bottom of the
vertical web, was 7 inch, and the profile of the latter
was struck with a radius of 12 inches.
The fishing
planes were inclined to the transverse axis of the rail
section at an angle of 30°; the thickness of the metal at
the toe of the foot was 375 inch, from which its upper
surface rose at an inclination of 1 in 10. The radius of
the wheel-tread was always 40 inches, and that of the
shoulder 5 inch.
34
2.725
17.15
35
2.75
17.3
36
2.775
17.45
37
2.8
17.6
38
2.825
17.75
39
2.85
17.9
40
2.875
18.05
41
2.9
18.20
52
THE MANUFACTURE AND WEAR OF RAILS.
BOTTOM FLANGES or FEET, varying in Breadth from 4.0 inches to 6.5 inches, and
in Depth from 898 inch to 1·023 inch, and in Weight from 24 06 lbs. to 39 43 lbs.
·
Weight
No.
Breadth
in
Depth
Weight
Breadth
Depth
in
in
No.
in
in
in
Inches.
Inches.
Lbs.
Inches.
Inches.
Lbs.
1
4.0
-·898
24.06
14
5.3
-·963
31.95
2
4.1
-·803
24.64
15
5.4
-·968
32.58
3
4.2
-·908
25.22
16
5.5
-·973
33.22
4
4.3
-913
25.81
17
5.6
-.978
33.86
5
4.4
-·918
26.40
18
5.7
-·983
34.51
6
4.5
-·923
27.00
19
5.8
-·988
35.16
7
4.6
-·928
27.60
20
5.9
-·993
35.81
8
4.7
-·933
28.21
21
6.0
-.998
36.47
9
4.8
-·938
28.82
22
6.1
1.103
37.13
10
4.9
-·943
29.44
23
6.2
1.008
37.80
11
5.0
-·948
30.06
24
6.3
1.013
38.47
12
5.1
-·953
30.69
25
6.4
1.018
39.15
13
5.2
-·958
31.32
26
6.5
1.023
39.43
the number of combinations of the thirty heads and the forty-one
vertical webs was therefore equal to 30 × 41, or 1,230; and every
one of these 1,230 combinations of heads and vertical webs would
fit on every one of the 26 sections of feet, making the total number
of combinations 31,980; the rails so compounded varying in width
of foot from 4 inches to 6 inches; in depth from 4.1 inches to
5.4 inches, and weighing from 62 lbs. per yard to 90 lbs. per
yard. The last weight was due to the combination of the heaviest
head, the highest vertical web, and the widest foot.
He did not advocate all these combinations or any extreme
weight. He only mentioned the number of possible combinations
to show that the Tables covered the widest conceivable range of
practice. Neither did he advocate, for light or medium traffic, a
greater width of head than 2 inches, a greater width of foot
than 4 inches, a greater depth than 4.3 inches, or a greater weight
per yard than 65 lbs.; or for the heaviest traffic, a greater weight
than 70 lbs. per yard, a greater width of head than 2.3 inches, a
greater depth than 4.55 inches, or a greater width of foot than
5 inches.
With these Tables, a T square, and the little template called a
"siderodograph" (Fig. 3), any one of these thirty-two thousand
rail-sections might be accurately delineated in two minutes. Say
an Engineer wanted a rail of the Vignoles section that would
comply with the following conditions:-weight from 65 lbs. to
66 lbs. per yard; head to have a width of 24 inches, and a weight
of 25 or 26 lbs., foot to have the minimum width compatible with
through fastenings, 4 inches, the rest of the weight to be made
up by the vertical web; he would look in the Table and find that
No. 1 head and No. 6 foot would answer these conditions, and
THE MANUFACTURE AND WEAR OF RAILS.
53
would weigh between them 52.76 lbs., leaving for the vertical web
12·74 lbs. if the rail were to be 65 lbs. to the yard, and 13-24 lbs.
Fig. 3.

Siderodograph, or Iron Road Delineator. Full size.
The dotted outline No. 1, shows the application of the template to the lower fish planes, and the
**
"
>>
No. 2,
No. 3,
No. 4,
junction of the same with the vertical web.
31
""
to the toe and lower flange of rail.
to the wheel-tread shoulder and the flange of the rail head.
to the upper fish plane, and its junction with the vertical
web.
The concave circular arc at the foot of the template is the profile of the vertical web.
F
54
THE MANUFACTURE AND WEAR OF RAILS.
for a 66 lb. rail; and on looking at the Table, he would find that
vertical web No. 5 would make the rail 65.56 lbs., No. 6
65.71 lbs., No. 7 = 65.86 lbs., and No. 8
the last, the weight and height would be-
66.01 lbs.: adopting

Weight in
Height in
Lbs.
Inches.
Head No. 1
25.76
1.340
•
Vertical Web No. 8
13.25
2.075
Foot No. 6
27.00
⚫923
Total
66.01
4.338
•
He then with a T square would draw on the board four hori-
zontal lines at the vertical intervals in the second column for the
base of the rail, the top of the foot, the under side of the head, and
the top of the same, and a vertical line for the axis of the rail;
and from this axis he would prick off on each side on the two
middle horizontal lines 35 inch for the top of the foot and the
bottom of the head, which would each equal 7 inch, and other
vertical lines for the vertical boundaries of head and foot. On the
accompanying diagram (Fig. 4) three lines of construction were
Fig. 4.

shown as full lines, and the outline of the rail as obtained by the
application of the template to such lines of construction was shown
by dotted lines. The dotted lines on the full size representation of
the siderodograph (Fig. 3) explained the application of the instru-
ment to the various parts of the rail section, and it could be
adjusted as desired to the lines of construction (Fig. 4) by moving
THE MANUFACTURE AND WEAR OF RAILS.
55
the T square up and down, and sliding the template along it like
a set square.
As regarded fastenings, the only description of joint now em-
ployed was the fish-joint; and of this there were two varieties-the
suspended and the insistent fish. The former had to perform two
functions, the latter only one. The suspended fish had first to act
as a girder, by giving vertical strength to the joint between the two
rails; the fishes required, therefore, to be as long and deep as pos-
sible, to be fastened by four bolts, and to have the bearing edges as
nearly horizontal as the fulfilment of other essential conditions.
permitted, in order to keep the ends of the two rails in accurate
apposition, so that the joints might offer no obstacle to the smooth
moving of the wheels.
The insistent fish was not required to add to the vertical strength
of the joint, such joint being supported by the joint sleepers: the
only function it had to fulfil was to maintain the extremities of the
conterminous rails in accurate apposition. For this purpose a
much shorter and shallower fish than would be required in suspen-
sion answered every purpose, and there might be only three bolts
instead of four, as the centre bolt, acting immediately at the junction
of the rails, had then a greater power of holding them in accurate
apposition than it could exert in any other position. Moreover,
there was no occasion to sacrifice the section of the rails to the
perfection of carrying out the principle of fishing, as the angle of
the bearing surface of the fish was comparatively immaterial. The
suspended fish was manifestly defective as a mechanical arrange-
ment, when compared with the supported fish, as the object to be
kept in view was to approximate as closely as possible to the con-
ditions of a bar of uniform strength. Now it was obvious that a
suspended fish-joint could never offer nearly the same resistance to
the load as an unjointed rail of the same bearing length; it there-
fore became necessary greatly to diminish the length of the bearing
whenever a suspended joint occurred, and this inequality of bearing
length was continued throughout the length of the rail, the
bearing gradually diminishing to the joint; and as the resistance
was as the square of the bearing, this inequality was manifestly
incompatible with any near approximation to the conditions of a bar
of uniform strength.
He submitted that Figs. 5, 6, and 7, showed what was the best
arrangements for the fished joints and fastenings for the Vignoles
rail. It would be seen that the fish was an insistent fish, and that
it was a double fish, horizontal as well as vertical; for the wide
horizontal under surface of the rail foot was kept in close juxta-
position with the wrought-iron bed-plate which received the ends
of the rails, extending the full width of the rectangular joint-
sleeper, a width of 10 inches on lines of light traffic, which, if
F 2
56
THE MANUFACTURE AND WEAR OF RAILS.
Fig. 5.

Transverse Section at Joint Sleeper.
The horizontal fish is shown unshaded.
Fig. 6.

Transverse Section at Intermediate Sleeper.
Fig. 7.

A TMT TJ
Side Elevation of Fish.
The upper horizontal fish is shown above the bed-plate.
Scale for Figs. 5, 6, and 7:-2 inches = 1 foot.
THE MANUFACTURE AND WEAR OF RAILS.
57
thought desirable, might be increased to 12 inches on lines of the
heaviest traffic. And the ends of these rails were fished to this
bed-plate by two horizontal fish-plates with three through fas-
tenings to each centre bolt, being placed at the most effective point,
viz., the junction of the rails.
4
These bolts, three on each side, passed through the sleepers,
and were screwed underneath it to a bar inch by 2 inches,
acting as a common nut to the three bolts, thus making the rails
practically a continuous bar of uniform resistance, and the sleeper
one mass, so to speak, with the rails. Joints of this kind were
quite imperceptible when running over the line, and the rails
approximated far more nearly to the conditions of a continuous bar
of uniform resistance than in the case of a suspended fish.
There might be an advantage in using Parsons' patent bolts for
the fish and fang bolts. He had never tried them himself; but
their principle was that of equalizing, by fluting the section and
strength of the shank to that of the screw-tapped extremity of the
bolt, so as to obtain uniform strength and elasticity; this was un-
doubtedly an improvement.
He always specified for fish-bolts with their inner surface con-
cave, and the corner well preserved instead of being rounded off;
by this means the bearing took its grip at the greatest possible
distance from the centre of the bolt, and greater leverage was thus
obtained to counteract the tendency of the nut to untwist. He
had never known such nuts to work loose.
One of the greatest advantages possessed by the Vignoles road,
where the rails were fixed directly to the sleepers on machine-
planed beds, by means of through screw fastenings, was the abso-
lute uniformity of gauge and cant of rails which was thereby insured,
and which rendered practicable a much flatter wheel-tread, and a
much narrower clearance than could be attained on a chair line;
and these advantages greatly conduced to smooth running, and the
preservation of rail surface and rolling stock.
The introduction of the Vignoles rail in this country had been
greatly retarded by the defective manner in which it was at first,
and was still sometimes laid down, viz., on half-round sleepers, and
merely secured by spikes, sometimes without even a nick at the
end of the rails to prevent them from travelling. No line of
Vignoles rails should be passed unless the sleepers were rectangular;
for the bearing of their section on the sleeper should be at least
equal to that of a chair on its sleeper, and that amounted to
40 square inches. This was somewhat exceeded on a rectangular
sleeper of 9 inches wide, which, with a width of 4.5 inches for
the rail foot, gave a bearing of 45 inches for each point of support;
but when a Vignoles rail was laid on a half-round sleeper, scarcely
one-half of the requisite bearing could be obtained, and that only
58
THE MANUFACTURE AND WEAR OF RAILS.
on sap-wood. Another advantage of this system was that with
3-feet bearings, the clear bearing between the supports was dimi-
nished, as compared with a double-headed rail, from 33 inches to
26 inches, and the stiffness of the rail thereby increased inversely as
the squares, or in the ratio of nearly 3 to 2.
He saw no reason why a steel-headed rail, of the Vignoles
section, dovetailed in the manner described by Mr. Menelaus,
should not last as long as one entirely of steel at double the
cost.
He thought it would be admitted that puddled bars made the most
enduring rails; but it was impossible to force iron of this description,
with rollers such as were usually employed, into the slender toes of
a wide-footed rail. He would submit that this difficulty might be
overcome by employing a third roller set transversely to the other
two, for rolling the under surface of the foot, an expedient very
generally adopted on the Continent in rolling wrought-iron girders
of similar section.
Mr. G. B. BRUCE said, with regard to the specifications for rails,
he was not prepared to throw the blame upon or to give the credit.
to Boards of Directors or foreign governments, but was rather dis-
posed to take the responsibility upon himself in cases in which he
was concerned. So far from the iron-masters having any reason
to complain of the specifications of engineers, he thought that they
had learnt a good deal from those specifications. The mode of
manufacture which Mr. Williams had so graphically described and
so strongly recommended was precisely the same which many engi-
neers had long given in their specifications, and which the great
bulk of iron-masters had been driven to adopt, by the engineers so
requiring the puddle ball to be hammered in the way described.
No doubt, the great secret of a good rail was that it should be
homogeneous; but that was not the sole condition, otherwise it
would only be necessary to fall back upon the old cast-iron rail,
which was homogeneous but brittle.
He was ready to admit that in the matter of rails, as well as
in many other things, tests were often carried to an injurious
extent; but when judiciously applied, they were very valuable.
When specifications provided that a rail should not break under
a certain falling weight, and yet not bend more than a given.
quantity, these conditions secured the possession of the two
requisites, hardness and tenacity, so necessary for a good wearing
and safe rail.
He was disposed to think that steel rails were not altogether
perfect. Out of one lot of 200 tons which he had used recently, five
of the rails broke, literally on being thrown out of the wagon; and
in the present process of manufacture, individual steel rails were
THE MANUFACTURE AND WEAR OF RAILS.
59
as likely to be defective as iron rails, and could not be implicitly
trusted. He had tested some flat-bottomed steel rails, punched in
the lower flange, and had observed that when the holes were not
near the part on which the weight of 161 cwt. fell, the rail did
not break, though the fall was 20 feet. But if the same rail
was so placed that the holes were in the middle between the points
of support, the rail broke at a blow with a 5-feet fall. The holes
ought to be drilled and not punched in every case when dealing
with steel rails.
He
Mr. R. P. BRERETON dissented altogether from the view that
good iron rails could not be made if pains were taken; or that it
had become indispensable to seek for other materials, such as steel,
except for special occasions. There was as yet no sufficient expe-
rience of the effect of hard steel rails upon the tyres and rolling
stock-by far the most expensive part of the property belonging to
Railway Companies. With regard to the deterioration in the manu-
facture of rails when these had to be turned out quickly and in
large quantities, it was not to be expected that they should at the
same time be made good. There was no desire to produce a lasting
article; and when Railway Companies undertook to make their own
rails, they did not make them better than the ironmasters.
had taken the pains to ascertain what had become of about
24,000 tons of rails made in Wales in 1849 for the South Wales
Railway, on which there was a large mineral traffic. It was 100
miles in length and was opened in 1850. In the first nine years,
out of the 24,000 tons there were only about 1,080 tons lost, or
4 per cent. of the original quantity; and of the remainder
there were at the present time 12,000 tons in use, and wearing
out at the rate of about 5 per cent. per annum of the original
quantity. This was principally at stations and inclines; and
switches and crossings of the same iron were still in use.
life of these rails had already averaged fifteen years, and at the
rate now wearing out would ultimately average seventeen.
If iron-masters could make such rails years ago they could
make them now; but an inducement was required which was
not offered.
The
Mr. T. E. M. MARSH stated that the rails alluded to by
Mr. Brereton were made from cold-blast iron, and a great part of
it was worked into refined metal. The manufacture was well
cared for. Such good rails were not now generally made. A
great demand for common rails had been met by cheaper and
inferior material and processes. The price had been fixed upon
a sliding-scale based on the price of common bar iron, which
was somewhat lower than the price of rails, the price of the
latter being then about 20 per cent. or 25 per cent. higher
60
THE MANUFACTURE AND WEAR OF RAILS.
than that of bar iron at certain works, because it was supposed
that rails were more expensive to produce. To this was added
from 15s. to 25s. per ton extra. Thus, assuming bars to have
been at £6 per ton, the cost of the rails would have been from
£8 to £9 per ton. The rail was made from puddled bars, one
of the principal advantages of that process being that the result of
the yield of the puddlers might be tested every day, and at any
moment. In that way a good welding quality, and a strong, solid
rail, could be obtained. The rails referred to were not "cold
short." Puddle balls might be badly made, and in fact fre-
quently did inclose chilled cinder and other impurities, which
would cause the iron to be crushed off the rail in large
pieces.
In some districts it would probably be found better to use puddle
balls direct, and not puddle bars; but the quality of the materials,
processes, and appliances of the manufactures must be judged of
by the engineer.
Mr. Marsh then referred to some samples of rails he had placed
on the table:—
No. 1. Rail and piece of slab, made for Mr. Brunel by Messrs.
Thorneycroft, in the year 1858; also small experimental
lots, made in the year 1848; the top slab being formed
out of a single bloom of charcoal iron. The rails were
laid on the Great Western Railway, and were doing good
service now, where from three hundred to four hundred
engines were passing over them daily.
No. 2. Similar to No. 1, made by Mr. A. Hill. The top slab
consisted of Bowling iron: used in the year 1858 and
subsequently, on lines in charge of Mr. Brunel, for points
and crossings, with very good results.
Nos. 3, 4, and 11. Specimens of puddled steel rail, made for
railways in charge of Mr. Brunel, illustrating apparently
good fracture and uncertain welds; and samples of steel
tops separated by wear. These rails were generally
unsatisfactory.
Nos. 5 to 9. Specimens of rails made between the years 1846
and 1850 and subsequently, for Mr. Brunel, being of the
class referred to by Mr. Brereton, as having been laid on
the South Wales division of the Great Western Railway,
between Grange Court and Swansea, a distance of
95 miles, where they were subject to the excessively
heavy wear of the inclines in the Neath and Swansea
district.
THE MANUFACTURE AND WEAR OF RAILS.
61
The duration of the bridge rails, of which 24,000 tons were laid,
was shown in the following statement :—
From opening of line in 1850 to June 30, 1859
June
"
""
July
>>
""
""
""
1859 to July
1863
·
1863 to February 1864
February 1864 to July
1864
July 1864 to December 1864
December 1864 to July 1865
July 1865 to December 1865
December 1865 to July 1866 .
July 1866 to December 1866
December 1866 to July
•
Worn out
1,081 tons.
•
4,598
763
""
494
>>
781
942
""
820
"
888
658
1867
563
•
""
557
July 1867 to December 1867
12,148 tons.
The same quality was also laid on the Great Western Railway,
the Wilts and Somerset, exclusive of the Bath branch, the Berks
and Hants, on part of the Oxford and Birmingham, and on heavily
worked parts of the South Devon Railway, and generally, with few
exceptions, on other lines forming part of the Great Western
system.
The merits of the various qualities, and the best mode of ob-
taining good rails, were continuously and rigorously tested by
Mr. Brunel, and the best results were invariably obtained from the
ironmaster, who did not object to, but invited and approved rigid
inspection, and desired to have the benefit of Mr. Brunel's valuable
experience, and to satisfy him that all details were carried out to
insure good results.
In June, 1853, Mr. Brunel addressed a letter to twelve of the
principal ironmasters in these terms:—
"You are probably aware that it is a very general complaint amongst En-
gineers and Railway Companies that they cannot now obtain with any certainty
rails of a good quality.
"There is every reason to believe that a little more care and skill in the
selection of the quality of metal, at all events for a portion of the pile forming
the rail, and somewhat more care bestowed in the manufacture, would secure
a very superior rail to those now generally manufactured, without any very
great increase of expense. I am desirous of obtaining, to lay before the
Directors of the Great Western Railway Company, proposals for the manu-
facture and supply of about 4,000 tons of such superior quality to be used as
an experiment upon a line of railway where there will be considerable traffic.
"I enclose a specification of the rail required; but I should prefer leaving
the choice of means to those whose practical knowledge must be much greater
than mine, and I should feel satisfied if I am assured by a respectable manu-
facturer that he will undertake to use his best endeavours to meet the ob-
jections raised, and to make a really good rail. I assume, although this is a
time at which you are all very independent of orders, yet that you would be
glad to make the attempt to regain the character which I assure you English
rails are fast losing, and if you are disposed to make a trial with 1,000 or 2,000
tons, and will state the mode of manufacture you would propose, and the
price, I shall feel obliged. The delivery should be immediate, or at ail events
carly."
62
THE MANUFACTURE AND WEAR OF RAILS.
Extract from Specification, referred to in Mr. Brunel's letter,
for Bridge Rails for Great Western Railway, June, 1853.
"One-third in quantity of the whole pile from which the rail is
to be rolled is to consist of a slab of a single piece, forming the
whole top of the pile, and so placed as to form the upper portion of
the rail when finished. This slab shall be made of No. 2 iron,
hammered, or of such other defined quality as may be mentioned in
the tender and agreed upon.
"The rest of the pile shall be formed of whole lengths of bars,
and no rail-ends, or awkward shaped pieces of scrap shall be used."
Mr. J. STRAPP had not much experience of the relative values
of steel and iron rails on the South Western Railway. Some steel
rails had been laid down four years, and during that time about
10,000,000 tons had passed over them without their showing any
perceptible signs of wear. At the same time iron rails had been down
in the same district eight years, and probably from 19,000,000 tons
to 20,000,000 tons had passed over them. He was not prepared
to say that iron rails should be replaced by steel rails where they
lasted so long. On the contrary, he thought that where iron rails
would last over six years, steel rails, at their present price, could not
be substituted with economy. He produced two specimens of rails,
one of which had been down twenty-five years, on the Gosport
branch, about half a mile from a station, over which not more than
4,000,000 tons had passed; the other had been down eight years
in Battersea Fields, and over this 17,000,000 tons had passed; the
average speed of the trains, in both cases, being about thirty miles
per hour. In calculating these weights he had, of course, taken
one-half the gross weight of the trains passing over the one line as
that due to duty performed by one rail. Time, therefore, was not
always a guide to the true worth of rails.
Mr. PETER ASHCROFT produced some specimens of rails which
had been used on the South Eastern Railway. Some of them had
been down twenty-four years. One, an old Dover rail, put down by
Sir William Cubitt, had been passed over by one hundred and five
thousand trains, as nearly as could be estimated, equal to
13,000,000 tons. This was on the down line, near the Dover
Station, so that every train had to pass over it. Another, of which
the maker was unknown, was from the Rochester and Gravesend
line, and had been down since the year 1845. It was only a
66 lb. rail, and the quality was excellent; there was a large
quantity of similar rails still down on that line. Another specimen
which had been down since 1865 on the Charing Cross line be-
tween Cannon Street and London Bridge, was known to have had a
hundred and seventy thousand trains passed over it, amounting to
21,000,000 tons. Only one side of the rail had as yet been used
THE MANUFACTURE AND WEAR OF RAILS.
63
during the two years and eight months which had elapsed since
it had been laid down.
He thought the South Eastern Railway Company had bene-
fited very much by leaving, to a great extent, the specification and
manufacture of the rails to the manufacturers, subject to a satis-
factory guarantee."
Mr. MARSH inquired whether the guarantee had worked satis-
factorily. It was a question of what rails had gone back to the
manufacturer.
Mr. ASHCROFT replied that one manufacturer had supplied
25,000 tons of rails, and the failures had not been more than
1 per cent. On the Charing Cross line, where the traffic was very
heavy, one hundred and ninety-three trains a day passing over one
line of rails, the failures were as much as 10 per cent. Care was
always taken to obtain a sufficient bond of indemnity for a term of
not less than seven years; and for five years on the Charing Cross
line.
Mr. W. MILLS said that a large number of steel rails had been
laid on the London, Chatham and Dover Railway. On a length of
six miles, originally laid with iron, but which had been relaid with
steel rails, he had reckoned the number of trains which passed over
three different parts, and had found that in one case 8,500,000 tons
had been carried, in another case 8,750,000 tons, and in a third
9,375,000 tons; the rails showed no appreciable sign of wear.
Some steel rails, for stock rails for switches, had been laid down at
Stewart's Lane, where formerly new iron rails were laid on an
average every three-and-a-half months. Those rails had been
down three years and three months, and were now in good order,
and would probably last another year or more. From this latter
instance it appeared, that the power of resisting wear and tear in
steel rails amounted to fourteen or fifteen times that of iron rails.
Mr. JOSEPH TAYLOR thought it would be a misfortune if the im-
pression arose that the interests of manufacturers and engineers
were antagonistic. Their interests must be identical; the only
1 Extract from the South Eastern Railway Company's Specification for Rails.
"The directors will not object to receive tenders for the 10,000, 5,000, and
1,400 tons of rails respectively, manufactured according to the ironmaster's own
specification, provided he can satisfy them of the sufficiency of the material to be
used and the mode of manufacture, and enter into sureties for a guarantee of not
less than seven years. In any case it is to be distinctly understood that this
specification is intended to convey and mean, that when any of the rails are once
laid down and being run over, any perceptible lamination, crushing, splitting, or
breaking taking place within the period of the seven years (not being fair wear
and tear), such rail or rails shall be subject to a total rejection without being
turned over, and must be cast aside and replaced by other sound rails at the
contractor's cost. Those rails laid down at stations and within the distance signals
are excluded from the obligatious of this clause, except they are found unsoundly
manufactured."
64
THE MANUFACTURE AND WEAR OF RAILS.
difference being that the engineer wanted the best rail for the least
money, and the manufacturer wished to make the best rail be
could, and to get the best price for it.
There was room for wider difference of opinion, as to the best
means of obtaining the best rails, and the question seemed to be
narrowed into that of specifications or no specifications. He was
not prepared to say that engineers ought to throw away specifica-
tions altogether, and put themselves entirely into the hands of the
ironmaster; but it was possible that specifications had been pushed
too far. He believed they were framed, in the first place,
from the best possible information to be obtained; but he thought
engineers shut themselves out from improvement and discovery in
the trade, when they adhered for some years to the same speci-
fication. There had been some inconsistency in the way manu-
facturers had been compelled to work. It was only reasonable
that, if a manufacturer was required to make a rail to produce
certain results, he should be at liberty to make it according to
the best of his judgment. Specifications had apparently been
directed too much to the production of a high quality of bar-
iron, to the neglect of the wearing quality of the metal. The
heavy falling test, and the heavy suspending test, tended to pro-
duce bar-iron of high quality, but that iron was deficient in hard-
ness and wearing quality. A little hardness might be admitted
into the manufacture of rails with great advantage; and yet the
rails should be sufficiently tough to resist any impact in the course
of the ordinary traffic. Considering the comparative scarcity of
steel-producing ore and the high price of steel rails, he thought
that, for some time at least, the bulk of the supply of rails must be
from iron; and so long as a thoroughly good iron rail could be
made for from £6. 10s. to £7. per ton, to last fourteen or sixteen
years, of which there was no doubt, he did not consider that the
days of the ironmaster were numbered.
3
16
Mr. W. ADAMS stated that, in the month of October, 1862, 20
tons of Bessemer steel rails were laid down on the North London
Railway, over which two hundred and fifty-four thousand two
hundred and fifty trains, estimated to weigh 38,137,500 tons, had
passed. He had lately measured them and found that they had
worn down about ths of an inch. He thought that each head
might wear down about 4th of an inch, and this gave the means of
estimating the durability of the rails. Between thirteen thousand
and fourteen thousand rails had since been laid, and he was in-
formed that there had not been a single failure in one of them.
Some crossings leading into Dalston Junction had been laid between
three years and four years, and the rails were still in good working
order.
Mr. ZERAH COLBURN had received a letter from the general
THE MANUFACTURE AND WEAR OF RAILS.
65
manager of the Erie Railroad, enclosing a report to the Chairman
of the line, in which he stated that during the month of January
upwards of one thousand rails had been broken on that line, and also
that he believed the month of February would show a still worse
result. Another correspondent had informed Mr. Colburn that
within his knowledge, on the Hudson River Railroad, a double line,
144 miles in length, in one day, one hundred and thirteen rails
had been found broken, during what was called a 'cold snap.' He
had received a letter from a friend in Eastern Canada, three or
four weeks ago, where possibly the temperature was lower than on
the lines mentioned, and the mercury at the time the letter was
written was 41° Fahrenheit below zero. At Toronto there had
been unusual cold, and the mercury had ranged from 13° to 17°
below zero.
He had no reason to doubt the statements as to the
breakages of the rails.
Rails in the United States were usually of the Vignoles section,
and weighed 64 lbs. to 74 lbs. to the yard. The Manager of the
Erie Railroad had asked his directors to authorise him to complete
the renewals for this year, to the extent of 25,000 tons, in steel,
and not one steel rail had broken, although their weight was only
56 lbs. to the yard, and some steel rails had been down for the last
five years. He might add that the Hudson River Railroad was
then being entirely relaid with steel rails.
Mr. W. BRAGGE said, in reference to Mr. Colburn's statement as
to the serious disaster on the Erie Railway, for such he called the
great destruction of rails, the engineer had stated that the month
of February would show a greater destruction of iron rails; but
he went on to state that the only portion of the line on which
trains could run with safety and speed was that portion laid with
steel rails, and that no fault could be found with the 10 miles of
track laid with the Bessemer steel rails; of these only one had
broken during the winter, and no lamination and very little wear
was perceptible. It should be borne in mind by those who
were using steel rails, that iron rail-makers had the experience
of thirty years or forty years: whereas steel rail-makers had been
really at work only since the end of the year 1861. Upon him first
devolved the duty of offering steel rails for sale; at that time the
steel rail was looked upon with doubt and distrust, and only by
offering rails to make into points and crossings, on trial, to be paid
for when they had earned their value, was it possible to sell them
in England. He claimed for the makers of steel rails, that they
had availed themselves of every branch of scientific knowledge
which could be brought to bear upon their work; and he contended
that the progress of seven years had not been unsatisfactory.
Mr. R. PRICE WILLIAMS observed that on the North London and
the London, Chatham, and Dover Railways the life of iron as compared
66
THE MANUFACTURE AND WEAR OF RAILS.
with steel rails was certainly of the briefest, while on the other hand
the iron rails on the South Eastern Railway appeared to have shown
an amount of endurance almost as remarkable as the highly phos-
phorized iron rails on the North Eastern Railway.
The consideration of the use of steel instead of iron rails was, after
all, an engineering and economical question, which a more extended
experience of the superior endurance of steel could alone determine.
As far, however, as he could gather, there could be no question as to
the preference of steel rails on main lines exposed to heavy traffic
and high speeds, where the maximum life of iron rails did not exceed
eight years.
The accuracy of the data on which the calculations of the
Author's table were based had been called in question; but he had
carefully gone through them with Mr. Sandberg, and had found
them perfectly reliable. Objection had also been made to the
extended period taken for the life of steel rails; he apprehended
that the Author had only assumed such as affording the only avail-
able means of comparison with iron rails of a proportionably long
life, such as might be anticipated where they were exposed to the
small wearing action of light traffic. He thought, however, that
twenty years' life for steel rails was quite sufficient for all purposes.
of comparison, as it was evident that where iron rails had a life of
even fourteen years, it was cheaper, at the present prices of steel
rails, to use that material.
He was unable to comprehend why Mr. Bell, at the outset
of his interesting account of the iron rail manufacture in the
Cleveland district, should lay so much stress upon the superior
advantages afforded by the process of puddling, in getting rid of the
phosphorus, when it appeared from his subsequent remarks that the
presence of this substance in certain proportions was essential to
the production of a sound and durable iron rail. That an excess
of phosphorus did conduce to the hardness of iron rails there could
be no question; but he thought most engineers who had experience
of cold short iron would confirm the opinion, that such additional
hardness was obtained at the greatly increased risk of sudden and
perhaps disastrous breakage of iron rails.
He was disposed to think that the efforts alluded to, as being
made to produce sound iron rails out of highly phosphorized
Cleveland iron, would prove unsuccessful; if, however, any practi-
cable means could be devised for getting rid of the phosphorus, and
of producing good steel out of Cleveland iron, the discoverer would
undoubtedly confer a great benefit, not only upon that district but
upon the country generally.
He had prepared a tabular statement with the view of showing,
that when compared with the present and future advantages to be
derived from the substitution of steel for iron rails, the difference in
first cost was not of the importance that might at first be imagined.
THE MANUFACTURE AND WEAR OF RAILS.
67
Annuity, £0·5215795.
TABULAR STATEMENT, showing how an ANNUTTY of 108. 51d. per Ton will Recoup the difference between STEEL and IRON RAILS
(£6. 10s. per ton) in a period of 20 Years.
DR.
Cash advanced at 5 per cent. Interest, £6. 10s.
Principal
Recouped.
Interest.
CR.
To 1 Year's Instalment
2
3
6829 OR A CO N
""
""
.2167295
""
""
4
⚫2275585
13
""
⚫2389400
>>
""
•2508870
£.
0.1963130
•2064010
£.
0.3252665
• 3151785
⚫3048500
• 2940210
.2826395
⚫2706925
13
""
7
•2634255
• 2581540
135
อ
•2766010
.2449785
""
2435
• 2904330
• 2311465
1234ROD∞☹
To 1 Year's Annuity
19
""
"
""
5
6
""
་
""
>>
7
""
8
""
9
""
""
10
• 3049540
• 2166255
10
""
•
""
""
11
⚫3202030
• 2013765
11
""
"
""
12
.3362125
•1853670
12
""
""
""
""
13
⚫3530280
• 1685515
13
""
יי
""
14
⚫3706755
• 1509040
14
""
""
,,
15
.3892135
•1323660
15
""
""
19
""
16
⚫4086745
•1129050
16
""
17
.4291040
⚫0924755
17
""
""
"9
""
18
.4505605
• 0710190
18
''
""
19
.4730830
• 0484965
19
"
""
20
•4967430
⚫0248365
20
322
21
£.
0.5215795
⚫5215795
• 5215795
• 5215795
• 5215795
⚫5215795
•5215795
• 5215795
⚫5215795
⚫5215795
• 5215795
.5215795
• 5215795
• 5215795
•5215795
•5215795
• 5215795
•5215795
• 5215795
"
""
"}
,,
""
6.4997400
3.9318500
⚫5215795
£10.4315900
Principal
Interest
£6.4997400
3.9318500
10.4315900
As per Contra.
Annuity required for 1 milo single line, £62. 11s. 9d.
(Iron rails and steel rails, 75 lbs. per yard.)
68
THE MANUFACTURE AND WEAR OF RAILS.
It would be seen that by dealing with the difference in cost
between iron and steel (£6. 10s. per ton), as a deferred annuity, the
small annual payment of 10s. per ton, during a period of twenty
years, would recoup the principal with interest at 5 per cent. during
that period. He might mention that on the main line of the Great
Northern, where the average life of an iron rail was only eight years,
the cost of renewals of the rails alone amounted on an average to £1
per ton per annum. It was evident, therefore, that under such cir-
cumstances the adoption of this principle of deferred annuities would
ensure a saving of at least 50 per cent.
Mr. BIDDER, Past-President, had never in the course of his
practice pretended to dictate to the manufacturers how iron should
be treated; but he had always endeavoured to obtain the best
quality. Engineers were often overruled by their masters, the
Directors; and there was apparently only one species of test they
were capable of appreciating, and that was the lowest price.
Whether as regarded locomotives, or steam-boats, or the manufacture
of rails, he should consider it simple affectation to assume that he,
as an Engineer, had such a knowledge of details that he could
dictate the mode in which they should be constructed or produced.
He knew one instance of the position he desired to lay down, where
in the case of some steam-ships, the result was that when a sum of
about £120,000 had been spent in six steamers, five had never
been used at all; and the only one that was used did not make a
single voyage without loss. As regarded iron rails, every manufac-
turer, urged by his own particular interest, directed his attention to
the mode of manufacture most suitable to his own locality and the
material he had to deal with. Now wherever Mr. Bidder was per-
mitted to have his way, he specified what he wanted, the test, the
power of the metal, and its other qualities; and then he said to
the manufacturer that was what he required, and those were the
tests which he should apply; and he then advised the acceptance
of the tender after judging of the character of the men who
tendered. He necessarily inferred that they were men satisfied with
their own mode of manufacture, and that he might rely upon
them to make the iron in the way they stated. By that means he
had a much better surety than if he had tried to force upon them
a mode of manufacture to which they had not been accustomed.
With regard to rails, all views as to the peculiar class of metal
suitable for rails had been, from the first initiation of railways,
subject to violent anomalies. He remembered a time when every-
body assumed that only cast iron was fitted for rails, when all at
once it turned out that wrought-iron rails were much more
durable. It must be borne in mind that, on a railway, no rail
could be broken by pressure. Take a bearing 3 feet long, and an
engine 30 tons in weight, rolling over it at a speed of 40 miles an
THE MANUFACTURE AND WEAR OF RAILS.
69
hour. It was well known that in 3 feet a rail would bend more
than of an inch before it would break; but if there were no rail
at all, at 40 miles an hour the wheel of an engine would not deflect
the 4th of an inch from gravity; therefore, if a rail deflected more
than th of an inch, no further pressure from the engine could
occur. But what had to be dealt with in rails was impact and
rubbing; the latter being the most serious question. He did not
know how a test for the grinding of a locomotive wheel could be
applied, unless by means of a huge grindstone; and that would not
give the exact effect.
Mr. Bessemer had been a great benefactor to railways and
metallurgy in general: he had given an impulse, the effect of
which could hardly at present be foreseen; but he believed the
result would be to produce eventually a cheap and good rail. He
did not care so much about its tensile powers, but rather its
capacity to sustain the grinding impact of a locomotive engine,
running at the rate of 40 to 50 miles an hour.
At the present day the popular cry was to close the capital
account of companies. But if it were proved beyond doubt that a
rail which cost £14 per ton was the one to be adopted in lieu of
the present rail at £7 per ton, who was to provide the extra £6 or
£7 per ton? On railways which were not in a condition to pay any
dividend at all, if the capital account were closed, and rails were
introduced which would eventually produce great economy, he was
afraid the result would be equivalent to confiscation; for suppose
some 10,000 tons of rails, at an extra cost of £70,000, were laid
down, and this sum were to be taken out of the preference divi-
dends, that would prevent their receiving any dividend for the time
being; and that he would call confiscation.
Mr. T. A. ROCHUSSEN stated that the Author of the Paper had
mentioned the mode of welding first brought into operation at
Hörde, in Prussia. The Plan C (Plate 14) was first adopted in the
year 1851, when the coal and mineral traffic had increased to such
an extent, that the ordinary iron rail did not suffice to resist the
increased weight. The puddling of fine-grained iron led to the
puddling of steel, which not only was applied to rails, but it gradually
became a staple manufacture of the country. About two years
previously he had produced two rails manufactured according to
that process; and only two or three days ago he had an oppor-
tunity of inspecting some of the same rails at Oberhausen, where
they had borne a traffic of 42,000,000 tons. The rails weighed
56 lbs. to the yard, yet only one side was worn; the other was
untouched, and these rails would probably not be turned for another
six months. That welding process had been carried on now for more
than seventeen years, and was still largely in operation; compound
rails were used to a greater extent than both solid steel rails and
G
70
THE MANUFACTURE AND WEAR OF RAILS.
solid iron rails together. The difficulty of welding was greater with
Bessemer steel than with puddled steel; and it had been found
necessary to increase the weight of steel in the slab, so as to throw the
weld into the web, as represented by Plan A (Plate 14). This
process was only applied to the Vignoles section of rail; since there
would be no saving in having the reversible opposite head in steel,
leaving simply the small piece of iron between. It was very diffi-
cult to get the iron web from the steel head in the old rails.
He thought that the fracture of American rails, which had been
mentioned, was due in a great measure to their unfavourable sec-
tion, which was too shallow.
The Hörde works, in Prussia, had gone further than intro-
ducing the weld of steel and iron. They had for a number of years
made different systems of iron angle-bars, with steel heads sus-
pended between them. One system, which he exhibited, had been
used for some time on the Brunswick Railway, and was a most
advantageous mode of supporting a limited amount of steel by a
larger web of iron.
Mr. BENJAMIN BAKER observed that, in testing an iron rail
against one of steel, of lighter section, but of equal stiffness, care
should be taken that they were representative rails of equal average
quality; so that a hard steel rail might not be tested against a
soft iron one. He thought engineers had ample data to enable
them to predict what the durability of such rails would be. The
maximum stress on the rails could readily be ascertained, since
it would be equal and opposite to that on the springs of the engine,
the varying deflections and corresponding pressures of which might
be noted in the ordinary working of the engine, at different rates of
speed, over various portions of the line; the stress being known,
the strain and deflection might be accurately computed. The
comparative stiffness of similar sections of steel and iron rails was
known to average as 4 to 3, and the strength as 5 to 3. It fol-
lowed from this that the stiffness of Mr. Fowler's 84-lbs. section, if
rolled in iron, would be 'equalled by a steel rail of similar but
reduced section, weighing 72 lbs. ; and the strength would be
equalled by a 69-lbs. steel rail.
8
With the section of rail adopted by Mr. Fowler, at 5 feet bear-
ings, the solid rails bore a deflection on an average of 10 inches;
those with drilled holes, only 2 inches; while those with punched
holes snapped with less than ths of an inch deflection. Those
rails were manufactured by the Dowlais Company from tough steel,
of exceptionally uniform quality, the tensile strength varying very
little in any experiment from 35 tons per square inch; and the
breaking weight applied at the centre of the 5-feet bearings averaged
29 tons for the solid rails, 193 tons for the drilled rails, and only
13 tons for the punched rails. To break those rails in a testing
4
THE MANUFACTURE AND WEAR OF RAILS.
71
machine, at the rate of sixty per minute, would require 52 effective
horse-power for the solid rails, and less than one horse for the
punched ones; and the knocking about which the respective rails.
would sustain, without injury in ordinary working, would be nearly
proportional to those amounts. The ultimate deflection of the
punched rails was identical with that of cast-iron rails of similar
section; hence, in one sense of the word, they were equally brittle.
A curious fact, bearing upon the position of holes in the flanges of
rails, was worth noting. In the rails above mentioned, the two
holes at each sleeper were placed 4 inches apart, measured on the
centre line of the rail, and on alternate sides of the web, so that
the loss of section was that due to one hole only. Now theoretical
considerations indicated that the resistance of the rail would be in-
creased by making another hole opposite the existing one, notwith-
standing the double loss of section. It seemed rather an odd way
of strengthening a rail, to punch a hole 14 inch diameter through
its weakest part; but after making every correction, theory indi-
cated a gain of 10 per cent., and this result was corroborated by
direct experiment. It was found upon trial that, whilst with one
punched hole the rails broke under 13 tons' pressure, with two
holes, placed opposite each other, of similar size, they sustained 141
tons.
2
Captain H. W. TYLER had been accustomed for many years past
to hear constant complaints against the manufacture of rails.
It had been said that the manufacturers had lost the art of making
rails-that the introduction of the hot-blast had enabled them to
work up inferior ores-that it was impossible to obtain either
good material or good workmanship-and that, in fine, good
durable rails, such as were formerly supplied, were not to be got
at any price. It was also said that rails were wearing out in this
country in six months, in situations where they formerly lasted for
years; and that those sent out of this country for foreign con-
sumption were worse than those manufactured for home consump-
tion. He had quite recently read a letter, from the other side of
the Atlantic, from a gentleman of great experience, who did not
believe that English manufacturers could now make durable rails,
and who thought that there was no remedy but the employment of
steel. Last summer he saw steel rails being laid down on lines in
the United States, which cost, including transport and a very heavy
duty, about £25 per ton. Only a few weeks ago, he had inquired
into an accident in the Penge tunnel of the London, Chatham, and
Dover Railway, which was occasioned by the failure of a rail, cer-
tainly very badly made, the top of which had fallen away in six
pieces, and he saw many near it failing in a similar manner.
During a recent inspection of the Grand Trunk Railway of Canada,
he found that rails laid down within the last few years had been
G 2
72
THE MANUFACTURE AND WEAR OF RAILS.
failing to an extraordinary extent, and that portions of the line
relaid with new rails were obliged to be again renewed in from
three to five years, while the older rails were still necessarily kept
in the track.
K
Now he had been carefully considering this question, both in
England and in America; and particularly with reference to
Canada—where, in consequence of the severity of the climate,
rails of good quality were still more required than in England.
The road was there in a perfectly rigid condition, for, say five
months in the year, and the thermometer sometimes
40°,
with
the frost penetrating 5 feet into the ground. But under those
circumstances the durability of the rails was very various; some of
them lasting six times as long as others, under the same descrip-
tion of traffic. He found this was not a question of price; for
strangely enough the cheaper rails had lasted longer than those for
which a higher price had been paid, in some cases at the sugges-
tion of manufacturers, to ensure greater durability. The failures
appeared to be irrespective of the specification, and of tests applied.
The principal defect was undoubtedly imperfect welding; and those
rails which had worn best were generally those which showed on
their fractured sections coarse crystals, and what would be called
inferior quality. He did not believe that all the defects of manu-
facture which had been witnessed or experienced were due to
faulty specifications. In truth, the engineers had generally con-
sulted the manufacturers in drawing up those specifications. The
difficulty had been to obtain durable rails under any specification.
And when the number of specifications was complained of, for he
had heard a manufacturer saying the other day that he possessed
copies of two hundred and fifty different specifications, it must be
remembered that a different specification was necessary in the case
of each manufacturer, and of each section of rail, and was allowable
in the case of each engineer.
In regard to tests, he believed that harm had been done by re-
quiring manufacturers to make rails subject to the test of too heavy
a falling weight. He had heard rails spoken of as being extra
Such rails
good because they would stand 17 cwt. falling 25 feet.
were no doubt defective in other essential qualities.
In answering the question how to get a durable rail, it must be
first determined what was required in a rail. A rail had often.
enough been compared to a continuous girder: but it was much
more; for while it was required, like a girder, to resist compression
principally in its upper, and tension chiefly in its lower chord, it
was further required while deflecting very slightly, to wear
gradually away under traffic on its upper flange, and to resist such
wear and tear to the utmost. He by no means agreed in compar-
ing the action of the wheel and the rail to a hammer and anvil,
THE MANUFACTURE AND WEAR OF RAILS.
73
because when the rolling weights inflicted a blow, they did so with
the intervention of springs.
When the rails were shallow, or
wanting in depth, the blows became more serious, and failures by
crushing at the ends, even when the joints were fished, were more
apt to occur; but when the section was deep, and deflection was
immaterial, the fish was more perfect, and a rolling, crushing
action was the principal effect to be encountered.
In regard to the best form of rail, whatever the reason might
be, it was certain that the double-headed section of rail was mostly
adopted in England: while on the continent of Europe and America,
such a section was only exceptionally seen. There were two prin-
cipal reasons for the adoption of the double-headed section: rails of
that section had a broader base by means of chairs, and they might be
turned so as to wear out four instead of two edges. But, on the
other hand, there were strong reasons against their adoption. They
were raised higher over the sleepers, having the chairs below them:
an amount of metal was put into the chairs which might with
greater advantage be put into the rails: they could not, as a
matter of manufacture, be made equally durable and serviceable in
both heads: and when turned, they were more or less defective,
and unserviceable.
With the Vignoles rail, on the other hand, the engineer and the
manufacturer might combine to make a hard durable head without
fear of a rail that would break in the track, because they could at
the same time use fibrous material in the lower flange and with
this section all risk of loose keys was avoided, and there was also
the advantage of fewer parts.
It was a prevalent idea that the Vignoles rail was not a good
form of rail for traffic at high speed, but he could not, himself, see
the objection. He believed, on the contrary, that when deep in
section, with a good broad base, and laid on square cross-sleepers
at a proper distance apart, it formed the safest permanent way that
could be constructed for any speed.
In regard to quality, that to some extent depended upon the
form. As he had already said, a harder and more brittle head could
be used with the Vignoles than with the double-headed section.
Any rail, whatever might be the quality which would not crush
from bad welding, would last for a considerable time. He had
an aversion to steel-topped rails, knowing that the steel and
the iron welded at different heats, and believing that under the
most careful system of manufacture, a large proportion of steel-
topped rails must, sooner or later, fail from the separation of the
steel from the iron. He had seen numerous cases, in different
situations, where the steel had come off bodily, and he conceived
that even the steel tops, which were originally well united, must
be liable, in consequence of being rolled out under the traffic,
74
THE MANUFACTURE AND WEAR OF RAILS.
sooner or later, to separate. He therefore thought that the state-
ment in the Table was deceptive. The general question of the
relative advantages of steel, as compared with iron, granting that
the manufacturers would still make good iron, was simply one of
price. Every one must admit the advantage of a homogeneous rail,
the structure of which rendered lamination impossible; but it was
not every Railway Company that could afford to pay a high price
for steel rails for other than exceptional parts of their line.
He believed that, while guarantees were good in their way, and
particularly in England, when rails could easily be returned to the
makers, there was one method which would be better in the end than
specifications or tests, and that was the unreserved publication of the
results. If all engineers would combine to collect and record the
price, the wear and tear, and the durability of each sample of rails
from the various makers, and to publish their records annually with
the reports of the companies, the manufacturers would be subjected
to a wholesome competition with each other, which could not but
be attended with the best effects.
Mr. J. A. LONGRIDGE said that the first malleable iron rails
were made by his father at the Bedlington Iron Works. There
was a strong contest in those days as to the comparative merits
of malleable-iron rails and cast-iron rails; and the principal
objection against the malleable-iron rails was that they laminated
and exfoliated. Engineers were then in favour of cast-iron rails
in preference to wrought-iron, and among them was George
Stephenson. The first wrought-iron rails were laid on a private
railway for conveying coals to the Bedlington Iron Works; and
to George Stephenson's credit it must be said, that when he saw
that although they did bend they recovered their form again, he
at once recommended their adoption, instead of cast-iron rails,
although he had a pecuniary interest in the latter. The manu-
facturers could still make as good, or better malleable-iron rails
than were produced at that time, or when the London and Bir-
mingham Railway was in course of construction; and if the manu-
facturer was not bound by absurd tests on the one side, and by a
specification which might suit one iron but would not suit another,
and the sort of rail that was wanted was clearly stated, an excellent
rail could now be produced for a fair price. A weight of 3 cwt. fall-
ing from a height of 20 feet on the rails was equal to an accumulated
shock on the rails of 3 tons. Supposing the weight on a driving-
wheel of a locomotive engine to be 7 tons, and that at a joint in the
permanent way one rail stood of an inch above the adjoining one,
then the velocity with which that would fall would accumulate a
shock on the rail of 4th of a ton, instead of 3 tons, as in the case
of the falling weight. He thought, therefore, that such a falling
weight was an absurd test. He did not deny that the dead-weight
4
THE MANUFACTURE AND WEAR OF RAILS.
75
test would give a certain result; but he was sure that almost any
rail would amply support whatever weight might come on it as a
dead weight. He did not agree with Captain Tyler that the form
of the Vignoles rail was advantageous, because crystalline iron could
be put where it was in compression and fibrous iron where it was
in tension. That was very true if a rail was supported simply
between two joints, but inasmuch as a rail was a continuous girder,
in a certain position of the load the top of the rail was in tension
and the bottom of the rail in compression; therefore if fibrous
iron was required in one part it was required also in another. Yet
rails were all so amply strong, that it was not a question of the
strength of a rail, but of its durability. Iron rails should not
be condemned because they wore out, but an attempt should be
made to reform the rolling stock. The enormous weight put on
the driving-wheels crushed the rails; and he thought conical
wheels fixed to the axles should be discarded, as being very
injurious to the rails: the better system would be to have every
wheel revolving independently. Mr. Cross had tried this plan
on the St. Helen's line with perfect success.
He did not ap-
prove of the conical wheel, because, even if the wheels were
revolving independently, one portion of the wheel-tyre must be
sliding. Take the ordinary cone of the wheel, and the ordi-
nary width of its bearing on the rail as only of an inch: then
with a train of 100 tons moving at the rate of 20 miles an hour,
there would be a grinding action equal to 13rd horse power; and
if the train moved at the rate of 60 miles an hour, it would be equal
to a grinding action of 5 horse power. This must have a preju-
dicial effect on the wear of the rails. If they were simply subjected
to a rolling motion, they would last for an indefinite period;
and he was sure it was mainly due to the conical form of the
wheels that the grinding and wearing motion took place: and this
was increased by putting gravel and sand on the rails when they
were at all greasy. He thought that steel tyres, which were now
being adopted, working on steel rails lessened the adhesion, and
The formula for this was--
Horse-power
When 20 =
V
""
>"
1
31
ƒ
1
=
N
wo v b
33,000 f.n D.
total weight of train in lbs.
1
velocity of train in feet per minute.
breadth of surface of tyres in contact with the rail.
coefficient of friction.
degree of conicity of tyres.
D= diameter of wheel in feet.
76
THE MANUFACTURE AND WEAR OF RAILS.
that the weight on the driving-wheels would have to be increased.
Therefore he maintained that wrought-iron rails should not be dis-
carded. The proper course to pursue was to reform the rolling
stock, to distribute the weight on a greater number of driving-
wheels, to have wheels revolving independently, and to abolish the
conical wheel.
Captain H. W. TYLER observed that Mr. Longridge had stated
that it was necessary to provide for tensile strain in the top flange
as well as the bottom flange of a rail. Now, in any case in which
a weight was applied on each side of a sleeper, the top flange of
the rail no doubt would be more or less in tension; but in reality
and in practice the distance between the engine-wheels, and espe-
cially the driving-wheels, was at least 5 or 6 feet, while it was
commonly more; whereas, the distance between the sleepers was
only 21 feet or 3 feet; therefore there could never be a weight
supported on both sides of any sleeper at the same time. When
the rails were sufficiently deep, and when the sleepers were
sufficiently near together, the deflection of a rail under passing
loads was imperceptible; and any strains which could be produced
in tension on the upper flanges of the rails over the sleepers were
practically immaterial, and were not to be compared to the strains.
in tension on the lower flanges between the sleepers.
Mr. JOHN BOYD thought that if engineers abolished tests they
would not get rails as good as those now obtained; although the
falling-weight test was perhaps excessive at the present time.
He did not think the mode of manufacturing rails had been
sufficiently considered. It was certain, that by a proper mixture
of iron in the puddling furnace a better rail could be obtained
than by making use of any one quality of iron; for instance,
by a mixture of Cumberland 'red short' hæmatite iron with
Welsh iron, and also with Cleveland iron, both of which were 'cold
short.' Sir Charles Fox once told him that he saw rails from
England being delivered in America, and several of them broke
when they were thrown down. He had formerly known the
Scotch foreman who was superintending the delivery, and he said to
him, "What a pity it is your Railway Companies don't give £1
a ton more, and get better rails than these." The reply was,
Weel, Sir Charles, they are just as guid as the bonds we gie for
them." Before giving a judgment on the wear of the rails, it
should be known how these rails which only lasted three or four
years were paid for. If engineers did not specify the mode of
manufacture, they should at least specify the description of iron
of which the pile should be formed; and he would suggest that
the top and bottom slab of the whole length and breadth of the
pile should be formed of No. 2 hammered all mine-refined' iron,
and the intermediate slabs of the whole length of the pile laid

(C
(
THE MANUFACTURE AND WEAR OF RAILS.
77
so as to break joint, of puddled bars, and that the quantity of
cinder in them should never exceed one-sixth of the whole burden
on the furnace, and that no crop ends should be used, and the
thickness of the slabs be so arranged that all parts of the pile
should be reheated simultaneously.
He was of opinion that moderate, and particularly falling tests,
should not be discontinued; and that care should be taken that the
bulk of all the rails really were equal to the quality of the tested
rails; as it was notorious that rails said to have stood the test
stipulated for by the East Indian Railway Company's Engineer at
the maker's works had not borne the same tests at Rotherhithe ;
and in his opinion the reputation of engineers and the safety
of the public required that tests should not altogether be
abolished.
He submitted the analysis of West Cumberland hæmatite Bes-
semer pig iron, which was extensively used by the makers of
Bessemer steel throughout the kingdom for rails and other pur-
poses. He advocated the use of a mild steel rail, as having most
adhesion.
"REPORT UPON THE WEST CUMBERLAND BESSEMER' PIG IRON, BY
DR. NOAD, F.R.S.
"DEAR SIRS,
(C
Laboratory, Kinnerton Street, Dec. 10th, 1867.
"I now have the pleasure of reporting on the last sample of Bessemer
Iron which you sent me for analysis. I consider it to be first-rate iron; the
quantity of silicon being by no means excessive, and sulphur and phosphorus
wholly unimportant.
I am, dear Sirs,
Yours very faithfully,
(Signed) HENRY M. NOAD.
"Messrs. West Cumberland Hæmatite Iron Co., Workington."
ANALYSIS.
per cent.
Graphite
3.850
Silicon
1.728
Sulphur
0.030
Phosphorus
0.014
Manganese
0.072
Titanic acid.
Iron
0.152
•
94.124
100.000
"REPORT UPON THE SAME, BY DR. TOSH.
"Chemical Laboratory, Maryport, Dec. 9th, 1867.
"The specimen of pig iron received on the 27th ult. has been found to have
the following composition:-
78
THE MANUFACTURE AND WEAR OF RAILS.
Carbon
as graphite
combined
Silicon
Sulphur
Phosphorus
Titanium
Manganese
Iron
•
•
•
•
•
·
•
per cent.
3.989)
• 496) 4.485
1.767
⚫008
⚫025
trace.
•
•219
93.724
100.228
"From the above results I have no hesitation in saying that this iron is of
excellent quality.
"Considered with regard to its adaptability for Bessemer Steel making, it
is in every respect one of the best irons which I have examined.
(Signed)
EDMUND G. Tosн, Ph.D.
"Messrs. West Cumberland Hæmatite Iron Co., Workington.”
4
Mr. F. D. BANISTER gave some details of the wear of rails on
the Brighton Railway, and illustrated his remarks by a tabular
statement. On the first 123 miles of the line from the London
end towards Brighton, which carried the South Eastern and the
Brighton main and local traffic, a length of 7 miles being a quad-
ruple, and the remainder a double line, the total expenditure in
rails for ten years, from 1856 to 1866, had been £20,865, or
£159. 13s. 10d. per mile per annum; giving a per-centage on the
original cost of £12.11, and showing a life of 8-10 years. In the
suburban district, from Victoria to Norwood, on the West End and
Crystal Palace line, the expenditure per mile per annum was £77;
the total expenditure in ten years, £8,097; the per-centage of these
on the original cost per mile per annum, £9.31, and the life of the
rails 10.73 years. On the remainder of the main line, from Red Hill to
Brighton, the expenditure per mile per annum was £72. 7s. 4d. ; in
ten years, £21,712; giving a per-centage of wear of £8.53 per mile
per annum, and a life of 11.71 years. On the coast lines, Brighton
to St. Leonards, and Brighton to Portsmouth, the results were
nearly the same-expenditure, £40 per mile per annum; total ex-
penditure, £15,000; and a per-centage of £4.94 per mile per
annum on the original cost. There were some curious results on
local lines. On the line between Wimbledon and Croydon the
expenditure in rails in ten years had been £11. 6s. 8d. per mile; total
expenditure, £511. On the coast line, from Lewes to Newhaven,
which had been in use for twenty years, the expenditure for the
last ten years had been only 14s. 10d. per mile per annum for
rails. That showed that those rails were remarkably hard; they
had worn very well, but more broken rails had been found there than
on any other part of the system. The general result of the ten years'
wear was an expenditure for rails of £78. 4s. 11d. per mile per annum,
LONDON, BRIGHTON, AND SOUTH COAST RAILWAY.
AVERAGE COST per MILE per ANNUM, and other Details respecting the MAINTENANCE of WAY and WORKS, for TEN YEARS ending 31st December,
1866, including RENEWALS both of WAY and BRIDGES, but exclusive of STATIONS and SIGNALS. Materials Net.
Original
Cost of Amount
Rails
In Ten
Length.
ex-
Years'
From.
To.
Wages.
RAILS.
Sleepers. Chairs, &c.
Renewal
of
£758.
£7 58.,
pended
Life of
Wear of
Total.
less Cr.
in Rails
Rails In
Bridges.
for Old
in Ten
Years.
Miles.
Chains.
£3 12 6.
Years.
Net Cost
Net.
Rails,
per cent.
on Net
Cost.
Tonnage
per
Anuum.
Remarks.
£3 12 6.
12 65
B. Arms,
Junc.
S. Nest.
£. s. d.
374 18
£. s. a.
5,159 13 10
•
£. s. d.
55 6 2
£. s: d.
49 7 7 8
£. s. d.
11 0 0
£. s. d.
650 0 6 1
£.
17,100
£.
20,865
8.19
12.20
2,096,640
22 23
29 60
Red Hill
Brighton
172 3 4 72 7 4
59 12 8
51 14 8
355 18 0
25,436
21,712
53 Brighton
St. Leonard's 121 16 4 42 11
8
43 13 3
38 18 9
56 9 8
303 9 8
27,926
13,844
11.71
20.17 4.95
8.53
898,560
318,240
Seven miles quadruple,
the remainder double
line. Average speed
35 miles per hour.
Double line. Speed 35
miles per hour.
Ditto. Ditto 30 ditto.
41 43 Ditto
9 27 Croydon
Portcreek.
105 1 10 38 0 0 6
50 14 3
38 10 0
44 9 6
276 16 1
35,514
15,966
22.21 4.49
Epsom
143 2 2 79 15
7
56 6 8
31 13 4
310 17 9
7,983
7,169
11.13
8.98
10 13 Norwood
Victoria
332 5 8 77 0 11
31
•
8 7
17 8 7
458 3 9
6,088
8,097
10.73 9.31
555,440
861,120 Ditto. Ditto 30 ditto.
3,781,920 Ditto. Ditto 30 ditto.
Ditto. Ditto 30 ditto.
•
Average 208 4 7 78 4 11
19 10 3
37 18 10
18 13 2
392 11 10
14.03 8.07
1,418,653
6 4 Croydon
9 9 Keymer
Wimbledon 67 0 0
8 10 0
11 6 8
Lewes
5
47
Lewes
Newhaven
157 2 2
103 9 9 1
9
8 10
43 13 4
0 14 10
37 16 4
10 10 10
6 13 4
18 4 5 102 2 0
79 5 5
93 10 0
2,586
511
411,840
330 10 9
7,791
858
336,960
231 16
6
4,777
48
299,520

80
THE MANUFACTURE AND WEAR OF RAILS.
and a per-centage on the original cost of £8.07, the average mainte-
nance of works over that portion was about £392. 11s. 10d. per mile;
so that the per-centage for wear of rails upon the £392. 11s. 10d.
was nearly 20, and the average life had been fourteen years. He might
mention that between the Bridge over the Thames and the Victoria
Station the rails originally laid by Mr. Fowler were of iron; they
lasted two years, and had been twice renewed, but they had now
been replaced by rails of steel. On a portion of the line near
Bricklayers' Arms Station, in the summer of 1864 he had laid down
400 yards of steel rails, manufactured by Sir John Brown and Co.
Over that portion on the down line the whole of the South Eastern
and Brighton traffic had passed, being 16,500,000 tons, in three
years and eight months. Those rails had been carefully measured,
and he found that the wear had been a little more than 1th of an
inch. The iron rails adjoining were worn out in about sixteen
months, with a traffic of about 6,000,000 tons, the speed in both
cases being 30 miles an hour.
Mr. W. BRIDGES ADAMS observed that in the question of rail manu-
facture, the engineer was the designer of the form, strength, and
quality required, if he were a veritable constructive engineer, while
the manufacturer was the producer of what would sell and make most
profit. The ironmaster preferred pigs, "the sow and pigs" in cast
iron; next, heavy round bars, then square bars, then the section of a
parallelogram, then the parallelogram channelled to make a rail; but
improved forms he called fancy iron, and eschewed them, if he could,
till a sufficient pressure was put upon him by the fear of losing trade.
The quality prized in Bessemer rails was homogeneity. Whether
they were steel or iron, was not a settled point. Mr. Bessemer himself
did not decide whether they were steel or iron. Probably they were
steel, from the fact of their breaking uncertainly, and, as shown by a
sample on the table, the same bar breaking within a foot of the upper
edge, where it was turned down under the engine-wheel as though it
were a piece of lead. A good rough test for steel would be hard-
ening and tempering a portion for a spring. Steel must be used
either hardened and tempered equally, or annealed throughout. If
neither of these processes were resorted to, it would be irregularly
hard and soft, and would infallibly break. This was probably the
reason why the maxim had grown up-" keep them soft." Bessemer
rails in their best form were homogeneous iron, and would of course rub
away under heavy friction, like any other homogeneous iron equally
soft; but they would not laminate, as iron-not being homogeneous
was liable to do. Iron rails were made by what was called piling, i. e.
iron in short lengths of differing quality, covered over with scale,
everywhere except at the ends, where they were not intended to be
united. These piles were full of hollow scale-cells, to which the air
THE MANUFACTURE AND WEAR OF RAILS.
81
got access while in the furnace, and when rolled out they were a
mere series of fibres and ribands, separated by scale, in straight
lines; and if farther rolled out, they would still be straight fibres,
and not gnarled in curves like the grain of oak or elm, or like tinplate
or sinuous iron, or like double thin steel, piled and curved and re-
piled, or Damascus sword blades, or the original puddle bloom from
the furnace. It seemed desirable with single-headed rails to obtain
a hard granular top and a fibrous bottom to the rail, and it was
possible to obtain blooms of those opposite qualities at the pleasure
of the puddler. There was a process of uniting two such blooms
without any scale between them, and therefore of rendering them
homogeneous. They required to be flattened under the hammer
into oblong slabs, with good surfaces and squared edges, and then to
plane the two surfaces which were to be united to a close fit.
Placed together, and heated in the furnace to a welding heat, they
would roll out homogeneously. But when the best rails were pro-
duced, the question arose as to the most judicious mode of applying
them in line. A sample rail of the bridge form from the South
Wales line, after being down for many years, appeared as good as
ever. But it had been laid on a continuous longitudinal timber, and
was boarded on the top, to make the iron cross the fibres of the
wood; for it had been found in practice that with the rails parallel
to the fibres the longitudinal fogs split. The rail was perfectly
granular, and had it been used with intermediate chairs, it would
probably have broken. Chairs formed anvils to rails, and notched
them below while they were bruised above by the wheels. Neither
had fish-joints yet been perfected. Some elasticity must be given,
not positive play, but a vibratory elasticity, and a true joint con-
nection.
Increased speeds had to be provided for, multiplied three-fold for
7 and 8 tons' load on each driving-wheel, instead of two, and for
increased length of wheel-base growing from 7 feet and 8 feet to
16 feet and 30 feet, and constantly increasing curves on the lines of
rails. In short, long rigid parallelograms were being driven round
very sharp curves, the wheels at a tangent with the rails. What
the amount of grinding was might be gathered from the reversal of
a double-headed rail. It at first presented a series of square-edged
notches,inch deep: at the end of a week the edges were taken off;
in another week the metals became small curves, then large curves,
and gradually the rail became a level plane. If the wheels rolled
instead of sliding, they would simply roll over the curved surfaces ;
but as they slid, they became a planing-machine, and cut off the
projecting metal. To get rid of the friction, it was necessary first
to radiate the axles so that the wheels would always be in parallel
planes to the rails, whether on straight lines or on curves, or double
82
THE MANUFACTURE AND WEAR OF RAILS.
reverse curves. Then to form the carriages with curved spring-
ends instead of side-buffer rods, to use short openings between the
vehicles, closing the trains to lessen resistance, and prevent the
vehicles riding on each other's backs in collision, to use continuous
breaks, self-acting by gravitation, like steelyard levers-the normal
condition to be pressing against the wheels, lifted by guard or
driver, and instantaneously in action on inclines, in case of a
coupling breaking. When these things were done, the engine-
power might be reduced, and the weight of the engines or the load
of the train might be increased. Tank engines were used, because
they could run with either end forward. Tender engines, if properly
coupled, making one bending machine of engine and tender, would
do the same. At present, in order to obviate a vicious design,
the driver screwed up the engine and tender rigidly together; and
a long stiff body was thus formed, which was apt to get off the line,
and which indeed would not keep on at all but for its weight. The
question of light engines and vehicles was important; but when the
surplus resistance was removed the trains might be lightened. Light
trains should be used for light branch traffic; but with traffic un-
limited, the loads on the engine driving-wheels should be what the
rails and lines would bear without being crushed, and the train-
capacity should be in conformity. If tank-engines were used for
long journeys, the load of water, which was a diminishing load,
should not be placed near the driving-wheels, but as far away as
possible. Any tender-engines could be altered to run tender fore-
most as steadily as a tank-engine; and two engines could be
efficiently coupled together to work as twins, with water-tanks for
short distances, without injuring lines or rails.'
Mr. E. A. COWPER had prepared a Table showing roughly the
different modes of manufacturing steel and iron, simply for quick
comparison of the number of operations in each of the processes.
In the old practice the iron used always to be refined before being
puddled, but that process was now generally omitted, from econo-
mical motives, though he was afraid in some cases it was but false
economy, and that the frequent very short life of a rail was partly
due to it.
According to the old process for making steel there was of course
the cementing, the melting in pots, the hammering the ingots, and
the rolling, in addition to the refining and puddling. The melting
of steel in quantities of about 40 lbs. each in separate pots was a
most expensive process, the wear and tear of the furnace on such
small quantities was very great, and it was therefore impossible
to make cheap steel rails on that plan, though it was admirable
1 Vide "Railway Practice and Railway Possibilities," by W. Bridges Adams,
London, 1868.
THE MANUFACTURE AND WEAR OF RAILS.
83
STEEL.
IRON.
for tool-steel and other expensive kinds. He had put Lowmoor
as the name of a process, although it was practised at other places.
TABLE showing the Number of Operations in each of the Various Processes for Making
IRON and STEEL.

Siemens.
Old
Modern
Practice. Practice.
Lowmoor.
Marshall. Krupp.
Bessemer.
Scrap.
Ore.
Smelting. Smelting. Smelting. Smelting.
Refining.
Refining.
Smelting. Smelting. Melting
Melting
in Air
Furnace.
in open
Hearth.
ing.
ing
Balls.
and
Selecting.
Puddling. Puddling. Puddling Puddling.
H. I.
Hammer-Hammer-Hammer-Hammer-
ing ing or
Balls. Squeezing Breaking
Balls.
Blowing
in
Converter. Puddling.
Hammer- Hammer-
ing
Ingots.
ing
Balls.
Rolling
P. B.
Rolling
P. B.
Piling. Piling.
Piling.
Hammer-
Rolling
P. B.
Cleaning.
Piling.
Rolling
P. B.
ing.
Rolling. Rolling. Rolling. Rolling.
Rolling.
Cement-
ing.
Melting
in Pots.
Melting
Cast and
Wrought
Iron in
Pots.
Hammer- Hammer-
Smelting.
Puddling.
Rolling.
Harden-
ing.
Breaking
and
Selecting.
Melting
in Pots.
ing
ing
Ingots. Ingots.
Hammer-
ing
Ingots.
Rolling. Rolling.
Rolling.
Same as above.
Same as above.
Melting
in open
Hearth.
Examin- Examin-
ing.
ing.
Hammer- Hammer-
ing
ing
Ingots.
Ingots.
Rolling.
Rolling.
Same as above.
84
THE MANUFACTURE AND WEAR OF RAILS.
The pigs were first taken and puddled to produce hard iron-iron
with about 15 per cent. of carbon in it; the ball was then ham-
mered thoroughly into a thick flat cake; it was not rolled at the
same heat, but it was thrown on one side to cool. It was then
broken up and selected, and he believed that the process of ex-
amining each piece and selecting the iron was one of the causes
why it had won such an excellent name for tyres and for other
things where iron was required to be hard. By these means it
was known that every part was hard; there was no question of
one puddler bringing out a soft ball and another a hard one, and
the iron being all mixed or 'piebald.'
Then came what was commonly called the 'Marshall' process.
In this there was the smelting, the refining, the puddling, the
hammering the balls, and the rolling the puddle-bars; then the
cleaning of the iron, which was a species of milling,' and con-
sisted in putting the small pieces of iron, cut into short lengths,
into a cylinder, with a quantity of sand and water, and turning
them round and round until they were thoroughly cleansed from
all black scale: afterwards this bright iron was piled and rolled,
and blacks' in the iron were thus avoided.
C
In Krupp's process of making steel the metal was first puddled,
then rolled into bars about 0.9 inch or 1.0 inch square, and thrown
into cold water, which of course hardened the steel from end to
end. It was then broken up into short lengths, so that every piece
of 3 cubic inches was examined at both ends, and it was examined
and selected according to quality. The pieces were then put,
like the cemented steel, into pots, melted, cast into ingots, ham-
mered, and rolled. A sort of military discipline was exercised in
bringing the men with the melted steel in the pots up to the bath,
or pond of metal, by the sound of a bugle; and sometimes more
than a hundred pots full of steel were got together for one cast.
Krupp, however, now used Bessemer steel largely; and he had
many Bessemer converters at work, although he did not exhibit
any steel at the Paris Exhibition by the name of Bessemer steel.
Much depended upon the manganese and other ingredients put into
the pots with the steel in this as well as in other places.
In the Bessemer process there was the smelting the pigs, the
melting, the blowing, the casting, the hammering, and the rolling.
The blowing was the chief process, and by it either wrought iron
or steel of various degrees of hardness could be produced. The
use of a certain quantity of spiegeleisen was of great importance.
Last on the Table were Mr. Siemens' two processes. In the first
plan for making steel, from puddled bars or scrap iron and cast
iron, the wrought iron, having of course been smelted, puddled,
hammered, and rolled, was simply melted down in a bath of melted
cast iron, in the hearth of a furnace having an intense heat, but no
THE MANUFACTURE AND WEAR OF RAILS.
85
oxidizing flame, and the ingots were cast from this large bath of
melted metal.
6
In Mr. Siemens' second plan the iron ore was brought into a
state of sponge of iron' by the heat to which it was subjected in
a non-oxidizing atmosphere, or one in which some carbon was
present, though not enough to convert it into cast iron: the iron
was not in any way exposed to a 'cutting flame' or burnt,' but
the drops of metal were absorbed, or, so to speak, dissolved out of
the sponge of iron' by the small quantity of melted cast iron at
the bottom of the hearth. There was, therefore, in this process
only the melting, examining, the hammering into ingots, and the
rolling; thus by either the first or second process there was ob-
tained a perfectly homogeneous cast steel, or wrought iron, without
any blacks or bad welds caused by piling or lapping up of cinder
in the ball, as in puddling.
He thought Mr. Bessemer was quite right when he, some years
ago, said that no perfect metal could be produced without its being
melted; and now there was a furnace that produced such an intense
heat, that tons of steel or wrought iron could be as perfectly melted
as the little drops of steel of 40 lbs. weight were in the steel pot-
furnace of former days, and yet without any burning of the iron or
steel. He thought each of the processes named in the table de-
pended greatly for its success on some one point being attended to.
Thus, in the old process, 'refining' was of great use in preparing
pure 'plate-metal' before puddling; while the practice of break-
ing and selecting' in the Lowmoor and Krupp processes was
admirable. The old plan of melting in pots was, and still con-
tinued to be, a very expensive process; whereas the Bessemer plan not
only made steel and wrought iron very direct from cast iron, but
gave the power of casting large ingots cheaply, though without the
means of examining the metal in the process. In Mr. Siemens'
plan there was certainly the advantage of being able to examine
and test the metal under operation, and either to increase or dimi-
nish the quantity of carbon, so as to make 'high' or 'low' steel
or wrought iron at pleasure; and from the casting of such large
quantities of metal at a time, and the economy of fuel in the fur-
nace, excellent metal could be turned out very cheaply.
Mr. F. STILEMAN stated, through the Secretary, that the total
annual cost for maintenance and renewal of the permanent way of
railways was above 9 per cent. of their gross receipt, and in the
year 1866 amounted to £3,466,668. This expenditure showed an
increase of £507,709, or 21.6 per cent. over the expenditure of the
year 1863, the increased mileage for the same period being
13-22 per cent.
The estimated tonnage of rails laid in main lines was 2,438,520
tons: taking the life of iron at ten years, it would require 243,852
H
86
THE MANUFACTURE AND WEAR OF RAILS.
tons, at a cost of £7 per ton, equal to £1,706,964 to be expended
annually; if renewed with steel rails, lasting double the time,
121,926 tons would be required at £11 per ton, equal to £1,341,186,
showing a saving of £366,778 per annum, or nearly 1 per cent. of
the gross railway receipts.
Mr. BRUNLEES observed, through the Secretary, that the rails
on main lines were, generally speaking, more deficient in bearing
surface than in quality. Their tops were too narrow for the
weight of engines now employed, hence they became laminated,
and were crushed long before they were worn out. If rails were
made with a top double the present width, or of the same width
as the tyres of the wheels, the crushing power of the wheel thus
diffused would be reduced to one-half, consequently the life of a
rail would be doubled. To make such a rail, a weight of about
30 lbs. per yard would have to be added to the 80 lb. rail; but
the additional cost would only amount to one-third of the ad-
ditional cost of laying down steel rails at 80 lbs. per yard. He
had, therefore, no hesitation in saying that the saving which a
wider topped rail would effect in the wear of tyres would alone
very shortly pay for the additional weight of metal; whereas, if
steel rails with only the ordinary width of head were used, the wear
of tyres would be considerably accelerated. To maintain a good
line, and a fair top on the rails, more lateral strength was required,
and this would be gained in the wider rails. There would also be
a considerable saving in maintenance, and that sometimes oscillating,
sometimes trembling motion, due to the weakness of the road and
the wear of the tyres, would be completely done away with. No.
doubt there were special places where steel rails might be used with
advantage, but, generally speaking, iron rails of good quality and
of greater section, combined with a judicious distribution of the
weight on the locomotive wheels, would meet a greater number of
present desiderata than could be supplied by the use of steel.
Mr. JOSEPH MITCHELL expressed his opinion, through the
President, that the flat-bottomed form of rail might be the best, if
of steel, laid irrespective of expense, on a continuous bearing: but
he doubted the superiority of that form of rail when laid on cross-
sleepers, for it thus became, in fact, a girder between two points
3 feet apart.
It was inferior as such to the double-headed rail, as it had less
depth, being for a rail weighing 75 lbs. per yard about 41 inches
in depth; whereas the double-headed rail was generally 5 inches
deep, and was formed like a girder, with its effective strength on
the upper and lower sides; the cross bearing also was only 5 inches,
as compared with the bearing of the chair, which was 10 inches.
If the flat-bottomed rail was made higher than 4 inches, it
was apt to deflect with the weight of the trains. No doubt

THE MANUFACTURE AND WEAR OF RAILS.
87
it was the cheapest, but not much, the saving not exceeding £70
per mile.
Although the double-headed rail, with its numerous fastenings,
still required improvement, yet, in practice, it had been found to be
the best, or it would not have been continuously relaid, and other-
wise so universally adopted by all the principal engineers through-
out the kingdom up to this time.
wear
Again, good ballast was an important element in securing a
sound and cheap permanent way. If a line was not well ballasted
and the ballast of good material, the rails would never
equally, however superior might be the quality of the iron. If
the ballast was soft or clayey, the pressure of the trains would
make the rails sink in wet weather, and thus create irregularities.
which, with the repeated concussion of the engines, tended much to
destroy and damage them. Now, on a large portion of the railways
he had made in Scotland, extending over 280 miles, not only
was there good gravel ballast, but the embankments and cuttings
were mostly gravel. The consequence was that the motion of the
trains was so smooth and easy as to be a matter of general remark.
The rails on the first section opened, which were double-headed,
had been down thirteen years, and he had no doubt they would last
six years longer.
He thought sufficient data had not been obtained to afford a
satisfactory conclusion on this subject, and he suggested that perma-
nent-
-way Engineers might be asked to divide the lines under their
charge into sections of 10, 20, or 30 miles, or such other distances.
as were subject to similar traffic; and to report on the nature of
the ballast, the size of the sleepers, the form and weight of the
rails-and, if possible, the makers' names, the average number of
trains, and the average tonnage passing over those sections per
annum ; also the quantity and cost of ballast per mile, the cost of
wages per mile, and the number of sleepers and rails renewed in
each year.
By a return of this nature, an accurate estimate of the cost of the
maintenance of permanent way under every variety of circumstances,
might be obtained, and thus proprietors of railways might know,
by comparing the maintenance of one railway with another, whether
those they were interested in were maintained in the condition and
at the expense they ought to be.
Mr. J. M. HEPPEL suggested, through the President, that rails
should be tested directly for durability. Suppose, for example, a
pair of rails 10 feet long to be laid down, and a pair of wheels.
carrying 10 tons to be kept, by a small fixed engine with a crank
motion, continually rolling backwards and forwards over them
till they exhibited symptoms of failure. It seemed to him that
this would nearly represent the actual condition under which the

88
THE MANUFACTURE AND WEAR OF RAILS.
wear and tear took place; and at any rate, as a comparative test
between different specimens, would give very useful information.
In this way in about a month an amount of tonnage might be
passed over equal to that by which the best rails were worn out.
No doubt the tendency of such a test would be to stimulate the
production of hardness at the expense of ductility; and to counter-
act this, some test for the latter, such as a blow from a falling
weight, would have to be added.
He thought that if manufacturers were requested to tender on
their own specifications, but guaranteeing a certain resistance to
impact, and a certain tonnage passed over at a given speed, in the
way just mentioned, all their special experience would be utilized,
and there would be a considerable amount of security for the
result.
He was aware that some difficulty would arise in submitting to
test a sufficient number of specimens without undue expenditure of
time and money. Still it might be managed; and if the results to
be expected were as valuable as they appeared to him likely to be,
a considerable outlay in obtaining them would be well repaid.
Mr. W. M. NEILSON suggested, through the President, that rails
should not be tested at the works of the makers, but at a public
testing-machine, under responsible management, to which rails
might be sent for undergoing the test process. Experience would
give those who managed this machine facilities for discovering
the different qualities of rail submitted to them. Suppose four rails
to be tested together, by four wheels hooped with ordinary tyres,
suspended by a lever pendulum frame, the wheels to be loaded
to the required weight, and so carried by a spring as to give
such a shock to the rails as might be found desirable. The
lower end of the lever, by touching projections in its motion, would
cause the side oscillating motion: and the amount of adhesion might
be measured by buffer springs at the ends of each of the rails. A
simple attachment to a small steam engine would give the requisite
motion to the machine, which would, of course, be kept constantly
at work; the number of oscillations and pressures on the buffer
springs being recorded by a tell-tale.
The motion of the wheels round the axis might be regulated
by springs or weights, suspended by chains leading from pulleys
over the centre of the pendulum frame; but the reverse motion
of the wheels might not be a too severe element in the test.
Mr. SANDBERG said the discussion had been such as to render any
reply on his part almost unnecessary. However, he should like to
direct attention. first to the steel-headed rails, and next to the
Annuity Tables.
His object in alluding to the steel-headed
rails arose from the difficulty which existed in foreign countries in
usefully employing the worn-out iron rails. In countries round the
THE MANUFACTURE AND WEAR OF RAILS.
89
Baltic and the Mediterranean, they had been re-rolled with iron
slabs, but of late Bessemer steel had been applied for the heads. He
never expected to obtain perfect homogeneity, nor to get the same
wear out of the steel-headed as out of the solid steel rails; and,
therefore, in the Annuity Tables the former were represented as last-
ing only half as long as the latter. If the worn-out iron rails could.
be converted into solid steel rails by Mr. Siemens' plan, it would
certainly be a great advantage for railways, not only in this country,
but in the colonies and indeed wherever far removed from the seat
of manufacture. These steel rails would, of course, have to be equal
to the Bessemer rails in quality and price; and if Mr. Siemens could
succeed in making them as cheap and of equally good quality,
would not for a moment continue to defend steel-headed rails.

he
As to the Annuity Tables, some objection had been made to the
eighty years taken as the life of solid steel rails. According to his
view, compound interest was an important feature in arriving at a
just comparison between the values of the three descriptions of rail,
and he had no other means of expressing their relative capacities of
enduring wear than by giving their life in years. He thought that
sufficient experience had been adduced in this discussion from different
railways to prove that solid steel rails lasted on an average six times
as long as common iron rails, though he admitted that iron rails
could be made of a quality far superior to that usually employed;
but in such a case they might cost more than steel rails; and there-
fore in the comparison he had made he had taken common iron
rails, without extra price, or any guarantee, or extra specification.
There had been fifty instances mentioned on the London and North
Western Railway, where solid steel rails had lasted six times as long
as iron rails; instances of the same kind had been stated on the
London, Chatham, and Dover Railway; and on the North London,
in about forty instances, as far as their experience had extended up
to the present time, steel rails had lasted four or five times as long
as iron rails. All these steel rails were still in use, and in good
condition. As regarded steel-headed rails, he had received a
favourable statement from the North British Railway; and that
day he had heard from Mr. Ashcroft that some of those rails made
for the Swedish Government referred to in the Paper (Plan B,
Plate 14), and described in the discussion by Mr. Menelaus, had been
laid a few months on the South Eastern Railway at London Bridge
Station. Mr. Ashcroft had given a very favourable report concern-
ing them. However, more experience was needed to prove that
they would last three times as long as common iron rails.'
1 Mr. Ashcroft has stated in a letter to the Author, dated May 18th, 1869, that
trains computed to have weighed upwards of fifteen millions of tous had passed over
the steel-headed rails between March, 1868, and May, 1869. These rails were laid
in the London Bridge yard of the South Eastern Railway, and he had been sur-
90
THE MANUFACTURE AND WEAR OF RAILS.
Mr. Bell had expressed a fear as to the supply of pig iron for the
future make of Bessemer rails. The same opinion was entertained
by the Swedes five or six years ago, when the Bessemer process was
introduced. They had actually tried to prove that nothing could be
done without Swedish ore. However, very little of the Swedish
pig-iron which had come over to this country had been employed for
Bessemer steel rails; and there was no doubt that the English
hæmatite would be sufficient for some time to come.
With regard to the statement that the pig iron made from Nor-
wegian ore at Newcastle contained 0-5 per cent. of phosphorus, Mr.
Bell had admitted that the statement applied to ore from a lately-
opened mine, and that it was not applicable to the great bulk of
Norwegian ore. His own opinion was, that if the English hæmatite
ore should not prove plentiful enough, the ores from Sweden, Nor-
way, Bilboa, Elba, and America, were sufficient to provide for any
future want of pig iron for the Bessemer process.
As to the tests of iron rails, very few Railway Companies had
adopted any more severe than those Governments with which he
was connected that had arisen from the severe climate in the Scan-
dinavian countries, the ground being often frozen 3 or 4 feet deep,
and offering in winter little or no elasticity. The speed had to be
reduced considerably in the winter, to save both the road and the
rolling stock. Although it was well known that greater wear could
be obtained from rails containing a large per-centage of phosphorus,
such as those from the Cleveland district; yet the greater strength,
and consequent safety, of the Welsh iron had been taken into con-
sideration, and most of the supply for Scandinavia had been pro-
cured from the Welsh districts. There had been no breakages of
rails on the Swedish lines, but the Russians had during the severe
frost of 1866 a great number of rails broken; the number amount-
ing to two hundred a week during the most severe part of the winter.
These Russian rails had been made in this country, nearly of the
same section (Vignoles') and weight as the Swedish rails; but
the test for the rails destined for Russia was only half as severe as
that for the rails for Sweden, the latter being a 15-cwt. ball falling
7 feet on the rail, supported on bearings 4 feet apart.
The statement that steel rails in Canada were not injuriously
affected by cold so much as iron rails, was what might have been
expected. Steel was less subject to the influence of cold than iron,
prised to find that they had resisted so perfectly, (there being no apparent failure
whatever in them,) when it was considered that they weighed only 66 lbs. per yard.
Opposite to these rails there had been laid some iron rails, weighing 82 lbs. per
var, manufactured by an eminent firm; the latter exhibited evident signs of
carly failure.
It should be noted that all these rails were laid on a curve and on
a gradient of 1 in 100, and where the breaks of every train were applied to stop
at the platform.

THE MANUFACTURE AND WEAR OF RAILS.
91
especially when phosphorus was present in the iron. But steel rails
had broken when the weather was temperate, in ordinary wear, and
even in unloading from the trucks, as had been remarked by several
speakers. These rails had been examined, and in the generality of
cases they had been found to contain too much carbon. Still there
had been several cases in which steel rails had been broken, without
being too hard, or containing too much carbon, but from excess of
silicon. Formerly it was believed that the silicon was eliminated
in the Bessemer process before the carbon was carried off, forming
a cinder, which acted on the carbon in the same way as in the
puddling furnace. There was now reason to believe that, where
there was much silicon present, and very little carbon, the latter
disappeared before the silicon, and some silicon was left which
rendered the steel brittle. Steel rails had been broken on the
Belgian State Railway, and, when they were analysed, there was
found to be no excess of carbon, but so large an amount of silicon
as 0.6 per cent. In another instance as much silicon as carbon had
been found in the metal. These facts had been confirmed by
Director P. Tunner, of Vienna, who had communicated the fact to
Dr. Percy in a letter, from which he had the privilege of reading
the following extract :—
"At the iron-works at Neuberg, in Styria, the Bessemer metal is carefully
assorted after every cast, and is separated into seven series, or different degrees
of hardness. The brands to be marked are ascertained on the one hand
mechanically, by welding, working, and hardening, by the tensile strength and
elasticity; and, on the other hand, chemically, by the Eggertz carburation
process it is only when the same result is obtained by these two methods
that the numbers denoting the degree of hardness are marked. A very great
difference has been ascertained between the results of the mechanical and those
of the chemical methods, in the quality of some Bessemer steel, which was
made about eight days ago from pig-iron made by charcoal with th its
weight of coke. By the first method No. 3 was indicated, that is to say, a
very hard steel difficult to weld, but otherwise of good quality. By the other
test it proved to be No. 6, that is to say, it had a hardness next to that of
wrought iron. In consequence of this very remarkable difference a chemical
analysis became necessary, which showed that the steel in question contained
only 0.3 per cent. of carbon, but 10 per cent. of silicon. It therefore appeared
that silicon might occur in steel as a substitute for carbon, and it is particu-
larly remarkable, that it was found in a hard steel. This fact further proves
that it is not always true that by the Bessemer process the silicon is first
eliminated, and then the carbon. That some proportion of silicon may exist
in steel has been long known; and Schafheutl maintains that some silicon is a
necessary constituent of steel. But that with 10 per cent. of silicon and
only 0.3 per cent. of carbon, it is possible to obtain a hard steel, seems to be
a new observation worthy of attention."
This confirmed what had been lately found in this country,
namely, that the silicon and carbon disappeared simultaneously, and
not, as was formerly believed, the silicon first and the carbon after-
wards. Therefore, as an excess of silicon had the same influence on
the hardness of the steel as an excess of carbon had, and as the



92
THE MANUFACTURE AND WEAR OF RAILS.
amount of the former was more difficult to regulate in the process
of steel-converting than the latter, which could be judged of by the
flame, he thought it might be concluded that, until further expe-
rience was gained, Bessemer rails would require to be as carefully
tested as those made from iron. In concluding his remarks, Mr.
Sandberg took the opportunity of thanking those gentlemen who had
taken part in this discussion, for the valuable information which
they had brought forward. That information formed a most useful
contribution to the knowledge of the manufacture and wear of rails,
and would doubtless tend greatly to the benefit of both the manu-
facturer and the consumer.
Mr. GREGORY, President, said it must be pleasing to the Members
of the Institution to find light thrown upon professional subjects,
not only by different minds but by different nationalities. The last
subject under consideration was a Paper by an American Member.
The Paper that had now been read was from a Scandinavian Mem-
ber of the Institution, and he was sure all who had heard it would
agree that it was one which merited the thanks of the Institution.
Note by the Author.
Since the foregoing Paper was read, the price of steel rails has
been reduced beyond expectation. At the present time (June,
1869) the cost of solid steel rails, at one of the principal works in
this country where both kinds of rail are made, is £10 per ton as
against £7 for iron rails. As Mr. Bessemer's patent will expire
next year, the reduction of the royalty now paid, will gradually
have the effect of bringing the price of steel rails nearer to that
of iron rails; but it seems doubtful whether any sudden reduction
of price will ensue. Under these circumstances there can be no
doubt as to the expediency of employing steel rails, even on rail-
ways where the traffic is light; but, of course, the heavier the traffic
the greater will be the economy of substituting steel for iron.
During the last twenty years the price of iron rails has been
gradually reduced to one-third of their original cost, but this
reduction has unfortunately been accompanied by the production of
an inferior quality of rail-a subject alluded to by several speakers
in the discussion. It is to be feared that to some extent a similar
effect may possibly follow the anticipated reduction in the price of
Bessemer rails; and it will therefore become more than ever
necessary, for the public safety, that the consumer should carefully
examine the quality of the material which he receives.

LONDON: PRIN
UNIVERSITY OF MICHIGAN
T AND CHARING CROSS.
3 9015 02234 4710

ENGW
TF
258
.52a
A 772,181

・
H
ま
​料
​d
A
き
​*
す
​ま
​ず
​大
​海
​焼
​小
​身
​なかれみ
​・
・
な
​15