strain which test pieces from the same
charge, and frequently from the same
plate or bar, withstand. Although this
circumstance has not hitherto been re-
marked upon, it has been admitted and
allowed for in practice, for the Admir-
alty, the Board of Trade, and Lloyd's
rules all allow a margin of 4 tons, be-
tween 26 and 30 tons, in the tensile
strain of boiler plates, per square inch of
sectional area, while for ships' plates,
Lloyd's rules also allow the same margin
of 4 tons between 28 and 32 tons, which
represents a range varying to the extent
of one- seventh on the minimum strength
insisted upon. In his paper, read at the
Institute of Naval Architects in March
of last year, Mr. Parker said: "This
range of about 4 tons in the tensile
strength of a plate of homogeneous
metal like mild steel is very unsatisfac-
tory." This is the only comment which
appears to have been made on this inter-
esting subject.
In explanation of some of the failures
in steel which have hitherto been classed
as mysterious — ^no attempts at explain-
j ing them having been made — the author
I would suggest the possibility that in cer-
tain cases the gaseous blowholes in an
ingot may have sorted or arranged them-
selves in a series, thus forming a line of
weakness in the plate or bar which has
failed along that line when subjected to
a strain much below that which test-
pieces from the same plate or bar would
withstand. The failure of the boiler
plates described in Mr. Parker's paper
previously referred to, read at the Insti-
tution of Naval Architects, might be thus
explained; and Mr. CuthiU's remarks
during the discussion which followed,
seem to confirm that conclusion, for he
said : " Some of the tests since made of
the plate corroborate the original, while
others are considerably higher, averaging
about 2 tons over it. No explanation
can be given as to this, other than that
the original was taken from the scrap
edges, while the higher ones and better
in extension, have been taken from the
plate." Other failures in steel have oc-
curred while the metal was being worked
up into ships or boilers, which may also
be explained by the existence of blow-
holes, forming a line of weakness in the
plates, but not necessarily a straight line.
Digitized by
Google
VAN nostrand's engineebing magazine.
When the plates were riveted at the ends
of the line of blowholes, the tensile
strength would not be much affected;
but when riveted at ^he sides, crossways,
there would be a certain stress produced
which would lead to breaking. These
failures are appwently complicated by
the circumstance that they occurred
when the men were away from their
work, either during the dumer hour or
at night, but on consideration it will be
seen that this circumstance will materi-
ally assist in arriving at an explanation,
for the rivets, when put in, were in a
heated condition, and heated the plate
to some extent, thereby causing a certain
amount of expansion, which was suffi-
cient to allow for the stress ; but on the
men ceasing work, cooling took place,
stress was put upon the plate, and it
snapped across in the line of blowholes,
but not necessarily through any rivet-
holes.
Speaking of the fractures in the steel
plates supplied for the boilers of the
Eussian yacht Livadia, in their report
dated February 16, 1881, Mr. W. Parker
and Mr. J. T. Milton say : "It was also
suggested that the brittleness might
have been induced by the absorption of
gases.'* It is to be regretted that this
failure was not inquired into from the
point of view here indicated, as such an
inquiry could not have failed to elicit
valuable information. At some works
agitating the molten metal in the ladle
before pouring into the moulds is prac-
tised for the removal of gases from Bes-
semer steel. To our past-president. Sir
Henry Bessemer, is due the honor of
having first suggested that means ; and
in the year 1881 Mr. W. Allen brought
before the meeting of this institute a
mechanical agitator which he had used
with advantage. In the discussion which
followed the reading of Mr. Allen's
paper, the late Sir William Siemens
spoke of some trials which he had made
for poling molten steel in the l^dle, in
order to remove the gases it contained ;
but he was evidently aware that all these
means were expedients, designed to miti-
gate a defect in steel manufacture which
could not at that time be removed, for he
added a remark I shall quote at length
later on as bearing pointedly on the sub-
ject under consideration. It may be
added that at a works where the manu-
facture of both Bessemer and open-
hearth steel is carried on, the latter in
furnaces heated by radiation from flame,
an agitator is used for Bessemer steel,
but is not required for open-hearth
steel, which lies quite still in the moulds
without having recourse to that means.
It has been supposed that the gases
forming blowholes in steel are bubbles of
carbonic oxide formed by the reaction of
iron oxide on the carbon in the mangan-
iferous iron, added to the molten steel
before it is poured into the moulds. This
explanation, however, is open to the ob-
jection that manganiferous iron, being
specifically heavier than steel, sinks im-
mediately to the bottom of the molten
metal to which it is added, being only
momentarily in contact with the slag on
the surface ; and, further, that, as already
pointed out^ we find gaseous blowholes,
or seedy boil, in glass, in the manufac-
ture of which the ad^tion of mangan-
iferous iron, or any material containing
carbon, at the end of the operation is
not practised. If the gaseous blowholes
found in open-hearth steel are bubbles of
carbonic oxide, hydrogen or air, these
gases may have been obtained from those
used for melting ; and this conclusion is
supported by the fact that blowholes or
seedy boil are found in fflass and steel
melted by contact of name, and are
absent when melting without contact of
flame is adopted. Diagrams are given,
representing, respectively, longitudinal
and transverse sections of a plate-glass
melting furnace heated by contact of
flame — such furnaces as were constructed
by Messrs. Sir W. and F. Siemens about
the year 1862. In this furnace the flame
enters the heating chamber at the level
of the siege, and plays around and on
the top of the pots, so as to obtain as
much contact as possible between the
flame and the materials to be heated;
after doing its work in the heating-cham-
ber the flame leaves at the end opposite
that at which it may happen to enter at
the time. In the manufacture of plate-
glass in such furnaces, the pots are
charged at intervals with the mixture
called batch, being filled in the first in-
stance, and refilled each time the ma-
terials are melted into glass, and thus
diminish considerably in bulk. When
the last charge has been melted, the
greatest heat possible is put on for abou^
Digitized by
Google
BLOWHOLES IN OPEN-HEARTH STEEL.
29
an hour in order to heat the furnace and
. its contents thoroughly, after which gas
is entirely, or nearly entirely, cut off, and
the furnace closed, for about four hours,
to allow the metal to fine. The^fining of
metal in this manner is for the purpose
of removing seedy boil, or blowholes,
the gases K)rming them rising to the
surface during this operation. The fin-
ing operation will be completed in about
an hour, but as plate-glass is required in
a plastic condition, flame is kept out of
the furnace for some hours longer before
the metal is ready to be cast. The pots
themselves, and their contents, are re-
moved for that purpose, if large plates
are to be produced; or the metal is
ladled out of them when it is only re-
quired to produce small articles, such as
tiles, ships' Hghts, &c.
The manufacture of window-glass is
carried on in the same manner as plate-
glass, so far as iiegards charging, melting
and fining ; but as this glass is worked
hotter than plate-glass, less time is
allowed for cooUng before blowing is
commenced. Some plate-glass furnaces
hold as many as twenty-four and thirty
pots, called cisterns, each containing
about 12 cwt. of metal, so that each
found in the larger furnace will weigh
about 18 tons. Window-glass furnaces
hold from eight to ten pots, each con-
taining about 3 tons of glass ; in bottle-
glass furnaces, the pots employed are of
smaller capacil^ than those used for win-
dow-glass, each holding about 1^^ ton.
In order to remove seedy boil from glass
melted by contact of flame, recourse
must be had to the method of fining de-
scribed, that is, the metal must be kept
tmder the influence of heat without con-
tact of flame. The operation will be
more or less complete according to the
temperature maintained in the furnace
after the flame has been cut off, and the
time which can be allowed for the glass
to settle. In the manufacture of common
bottles, cheapness of production rather
than good quality of glass, having, until
recently, been the object aimed at, the
glass made, as shown by the bottles ex-
Lbited, contained much seedy boil. In
order to reduce the cost of production,
and at the same time improve the quality,
Mr. Frederick Siemens some years since
invented the continuous glass melting
process, which is adopted by all the
leading bottle-makers in England and on
the Continent, and is now being largely
introduced for making window-glass.
Since the introduction of this process,
bottle-glass furnaces have been enorm-
ously increased both in their holding and
working capacity. Up to 1882 the
largest furnace of this kind was one of
those built in Englaiid, the tank measure
ing 42 feet long by 16 feet vnde, and
holding about 150 tons of metal ; but in
combination with his new method of
heating by radiation, Mr. ^Siemens, at
Dresden, has been able to augment con-
siderably the capacity of these furnaces,
and a photograph is placed before the
meeting of a circular furnace at his works
measuring about 40 feet diameter, and
holding about 230 tons of glass in fusion.
Diagrams are also given showing, re-
spectively, longitudinal and transverse
sections of a tank glass furnace heated
by contact of flame, these diagrams
being taken from the drawings of a fur-
nace similar to that just referred to as
holding 150 tons of glass. The flame is
made to strike on to the metal in
fusion, with the result that the glass
made contained some seedy boil ; but the
glass produced in this furnace being bet-
ter than that made in the furnaces which
is superseded, no notice was taken of the
circumstance at that time. As bottle-
glass-blowers commence work each week
on Monday morning at six o'clock, and
leave off work about the same time on
Saturday morning, that day is available
for cleaning the flues, during which
operation the glass in the tank is partly
refined, as already explained, further fin-
ing occurring on Sunday, when the fur-
nace, not receiving any addition of fresh
batch, gas and air have to be partly cut
off, to reduce the heat, and under these
conditions the flame would not come into
contact with the metal in the tank. Thus
a certain amount of refining was always
practised in making glass by the con-
tinuous melting process, and, when ne-
cessary, the tank could be treated as a
huge pot, all the heat possible being ap-
plied for some time, after which, the
furnace doors and openings being closed,
the flame would be reduced for a few
hours by shutting off nearly all the gas
and air. On one occasion the author
dealt with a large tank furnace in that
manner, and, as was expected, the metal
Digitized by
Google
80
VAN nostrand's ekgineebing magazine.
was found afterwards to be of excellent
quality and free from seedy boil, or
blowholes. By the adoption of Mr.
Siemens' new method of heating furnaces
by radiation, the loss of time and fuel
consequent on fining when pot furnaces
are employed is avoided, and seedy boil
has disappeared in glass made by his
continuous melting process. The pres-
ence of seedy boil m glass is therefore
attributable to the contect of flame with
materials in fusion, and is avoided when
glass is melted by radiation from flame.
The same result is obtained when glass
is melted in closed pots ; this is another
but rather expensive way of avoiding
contact of flame vnth materials in fusion,
which in glass-making can only be adopt-
ed for the best qualities of glass, table
glass for instance. Samples of glass
melted under these conditions are ex-
hibited, and will be found to be quite
free from seedy boil.
A good form of furnace for melting
glass continuously is shown by two dia-
grams, one being a longitudinal section,
the other a sectional plan* The gas and
air enter into combustion at a certain
distance above the metal in the tank, and
the roof and walls are so arranged that
the flame does not touch them, but
sweeps round the heating chamber, in
horse-shoe form, to the exit on the same
side of the chamber as the admission
ports. For the manufacture of glass
this form of furnace offers many advan-
tages, as, while providing a long run for
flame, thus promoting perfect combus-
tion, it gives a large extent of surface at
which the workmen may be placed for
gathering and working. This same form
of furnace has also been adopted with
advantage for melting steel on the open-
hearth ; or long furnaces, holding 40 to
50 tons of steel are used, in which case
they are made fish-bellied in plan, so as
to allow space for lateral expansion, as
well as a long run for the flame. In Mr.
Siemens' furnaces at Dresden, before re-
ferred to, the run of flame from the inlet
to the outlet ports is nearly 100 feet, and
the greatest uniformity of heat is there-
by secured, combined with economy in
working. It has been shown that seedy
boil or gaseous blowholes are invariably
found in glass when made in furnaces in
which contact of the flame with the ma-
terials in fusion occurs, and it has been
shown also that these defects may be re-
moved by fining, or may be altogether
prevented by melting glass without con-
tact of flame — that is, in closed pots, or
by radiant heat. The same remarks
hold good with regard to steel. What
is wanted at the end of the operation is
that the fused metal should lie quiescent
for a certain length of time under the
influence of intense surface heat without
contact of flame. In glass making, seedy
boil may be obviated in two ways — either
by fining, as has been explained, or by
melting without contact of flame, but for
making steel on the open-hearth the latr
ter alternative is alone practicable, for
the reason that it is required to be
tapped as hot as possible, whereas glass
is worked in a semi-plastic condition.
The desirability of maintaining steel
in a quiescent state for some time before
pouring was recognized by the late Sir
William Siemens, who, after describing
the trials he had made of poling steel in
the ladle, before referred to, made the
following prescient remarks: "I have
observed that considerable benefit accrues
to the steel if, after it has been poured
into the ladle, it is allowed to remain
there under the protecting covering of
slag to prevent decrease of temperature."
The objection to this mode of proceed-
ing is that, at this portion of the process,
time is valuable; but I feel sure that
practical benefit will accrue it the steel
can be kept in a quiescent fluid state for
a quarter of an hoar before pouring. Of
course, it would not do to leave steel for
a quarter of an hour in the ladle ; it
would partly set, or become plastic if
that were done ; and Sir William Siemens
referred to that circumstance when
speaking of time being valuable at that
part of the operation. In 1881 all fur-
naces were heated by contact of flame,
but his genius enabled Sir William
Siemens to forecast the requirements of
steel manufacture to render the operation
perfect. Mr. Frederick Siemens had not
then invented his new method of heating
by radiation, and had Sir William i:^ie-
mens been spared to us a few years
longer he would have been pleased to
find in the application of this invention
the solution of the remaining difficulty
in steel manufacture.
There is another point to which, be-
fore concluding, I wish to draw your at-
Digitized by
Google
BLOWHOLES IK OPRN-IIEARTH STEEL.
31
tentioD, at least briefly, viz., the analogy
which exists between glass and the slag
covering steel in open-hearth furnaces.
If the slag were deprived of most of its
iron, it wonld become clear glass, hence
the method of flning applied to glass-
making will also apply to slag in steel-
making. In the early days of the open-
hearth steel manufacture, glass was pro-
posed as a protective covering for steel
to prevent flame from penetrating to the
molten met£d beneath, and the slag float-
ing on steel is perhaps even now looked
upon as affording such protection. The
behavior of glass when melted by contact
of flame will not, however, as has been
shown, sustain that opinion. In bringing
to bear upon the subject of steel-making
the results of experience obtained in an-
other manufacture, members will under-
stand that the author is dealing with
figures worthy of their attention, the
quantities of glass treated in continuous
glass-melting furnaces varying between
80 and 230 tons at a time, and so far as
the dimensions of these furnaces are
concerned, it need only be said that they
much exceed anything as yet employed
in steel -making. This will become evi-
dent when we consider that the tank fur-
nace described, which is one of many
built in England, would hold 375 tons of
steel, and the furnace described as built
at Dresden — Mr. Siemens having several
furnaces of the same size in operation —
would hold 575 tons of steel, the den-
sities of steel and glass being taken as 7
and 2.8 respectively. There can be no
doubt that by manufacturing open-hearth
steel free from gaseous blowholes the
metal produced will be much stronger
and more reliable than that made by con-
tact with flame, and the result will be a
greater confidence in its use. The merit
and utility of Mr. Siemens' new method
of heating by radiation, by which this
may be effected are enhanced by the con-
sideration that the production of steel of
superior quality by its means is attended
with considerable saving in cost of con-
struction, in wear and tear of furnaces,
and in waste of metal, the yield of sound
bars or plates from a given weight of in-
gots being considerably more than was
attainable with steel melted by contact
with flame.
Mr. Aitken said he had given consider-
able attention to the subject of gases in
' metals, and had spent many thousands
I of pounds in experimenting, but had ob-
tained no commercial results beyond a
'slight increase of knowledge. It oc-
curred to him that the difficulty resulting
from gases in steel might be dealt with
by mechanical means. No doubt, as the
paper pointed out, if iron and steel could
be dealt with in the same way as glass was
dealt with in Mr. Siemens* furnaces, and
kept hot for a long time, the gas in the
metal would to a large extent be got rid
of, and a sounder ingot would be ob-
tained. Moreover, these metals were
very much more expensive than they
would be if any practical method of free-
ing gases from metals by mechanical
means could be adopted. Mr. Aitken
then referred to a diagram in which he
suggested one of the modes by which
this object could be attained. They had
now taken up the subject again, because
he thought the time had now arrived
when the subject was rife for discussion.
Seven years a^o he took out a patent for
a mechanical means of freeing gases
from metals, but he had been able to do
nothing with it, because it seemed that
the question was not ripe. The diagram
represented a ladle filled with molten
metal. The exhaust chamber was lower-
ed from the top and the nozzle projected
into the upper part of the metal. The
vacuum was then applied, and four tons
of metal were drawn up into the vacuum
chamber. When Sir Henry Bessemer
made some experiments he took a pot of
molten steel from the Bessemer converter
and poured it into a crucible, and he put
that under a vacuum. As a result, wnen
the vacuum was applied the metal
swelled and almost exploded, and there
was only one third of it left in the pot.
Therefore when the metal was drawn by
the vacuum into the vacuum chamber he
expected that a very large expansion of
metal would take place. When one
charge was done the air would be ad-
mitted again, and the metal would fall
back into the ladle. It was found that
metal treated in that way was of some-
what higher specific gravity. Having
lost a certain amount of gas the metal
was more dense, and the gases in escap-
ing had also absorbed a certain amount
of heat The denser metal would be
found to dettle at the bottom, and by
successive changes it would be possible
Digitized by
Google
32
VAN NOSTBAND'S ENGINEERING MAGAZINE.
to free the whole of the steel in the ladle
from the gases. Then the question arose,
What would be the effect of that t His
own impression was that steel would be
very much strengthened by the extrac-
tion of the gases, and if this plan were
adapted to the Bessemer process he
thought the use of spiegeleisen might be
done away with.
Mr. Walker regretted that a man like
Mr. Head, with his extensive experience
of steel, should have put before the in-
stitute the words, puzzling and mys-
terious, in reference to steel. There was
no such thing as mystery and no puzzle
about it at alL When there was any-
thing wrong with the steel it was be-
cause it had been badly manufactured
and badly heated. Men accustomed to
use both steel and iron knew that there
were ten times as many wasters in iron
as in steel If an iron plate had a hole
in it that you could put your knife in it
was sent back and nothing was , said
about it, but if the same thing occurred
with steel, it would be said there was a
mystery as to the cause of the failure.
Just as the workmen increased in skill
would the number of waster plates de-
crease. There was no comparison at all
between the qualities of steel and iron,
because shapes could be moulded in the
former tnetd which could not be made in
iron at all. The case of the plate re-
ferred to by Mr. Head was not, he
thought, due to blowholes, but to want
of work. It was not well made. He
himself paid particular attention to the
matter, and went over to Oammell's at
the time. There was no mystery about
it, but it was as plain as a pikesta.ff how
the defect came about. With reference
to the apparatus which Sir Henry Besse-
mer had suggested for fusing the metal,
its use was quite the exception, and
there were very few people who did that.
As good steel as he had ever seen had
been made in the Bessemer furnace.
Mr. Windsor Richards said he had had
a good deal of experience during the
last year in making soft steel plates, and
had experienced a good deal of difficulty
and trouble from two causes, one being
the blowholes, and the second and great-
est, the irregular distribution of carbon
in steel. At Eston they operated on
large ingots weighing about 4 tons, 6
feet long, 36 inches wide, and 66 inches
thick. Mr. Head had very carefully
checked the chemist's analysis, and the
carbon in the steel was found to vary
from 0.10 to 0.16. The great trouble
when ingots had to be rolled was to keep
within the hmits required by Lloyd's —
about 4 tons. Very many things had been
tried to get over this serious difficulty,
the mechimical agitator, as suggested by
Sir Henry Bessemer, and poling the
molten lead with a very long pole, as
suggested by Sir William Siemens, the
latter being more comical than effectuaL
They also allowed the metal to rest in
the convertors for half an hour before
pouring, but that did not do, and none
of the things they tried were of any
avaiL At k^t they tried what had been
done at other places, though he thought
not for the same reason. Mr. Biley, of
the Steel Company of Scotland, poured
the metal from one ladle to another, and
he thought if he could, after putting the
ferro-manganese into the ladle, pour the
metal from one ladle to another, he
would get rid of the gas, and get a more
equal distribution of the carbon. He
was glad to say that this method had
been very effectual indeed. He would
suggest to those experiencing a difficulty
to try the process in operation at his own