Non-technical chats on iron and steel, and their application to modern industry

by La Verne W. Spring

Bushelled Iron

After heating a mixture of thin sheet or other soft steel and wrought iron scrap to a welding heat in a reverberatory furnace, it may be balled and put through the squeezer, muck rolls, etc., as was the true wrought iron made by puddling. This manipulation of scrap is known as the bushelling process and the product as “bushelled iron” or “scrap bar.” By using scrap bar for the outer layers and old wrought iron bars cut to length for the interior, box piles are made, which, heated and rolled, make very good material, though not as good as puddled iron. Certain grades on the market, e. g., common bar iron, contain more or less of this bushelled iron, but the better grades, refined wrought iron, double refined wrought iron, engine and stay-bolt iron, are usually the pure puddled product.

As with the latter, repeated shearing, piling, and rolling improve the quality.

Bushelled iron is largely the result of an endeavor to reduce the cost of production of wrought iron. The material has a legitimate place and considerable of it is used.

CHAPTER VII
CEMENTATION AND CRUCIBLE STEELS

In the early days practically the only steels recognized—certainly the only ones desired—were of the high carbon or hardening variety. These were required for the manufacture of swords and other implements of war, for tools, etc., most of which had to have hard and sharp cutting edges.

When softer and less brittle metal was desired, wrought iron was available, but in all probability high carbon steel was the material most largely used.

Having but the two iron alloys and these of very different properties, it was not difficult to distinguish between them. A piece of metal could be heated to redness and plunged into cold water. If it became glass hard when cooled in this way it was thereby proved to be steel; if still soft, it was iron.

But the problem is not so simple to-day. Medium, mild and yet softer steels, and other alloys which have steel characteristics have appeared and are used in immense quantities. Their advent introduced considerable complication.

It will be well, therefore, before taking up our subject, “Cementation and Crucible Steel,” and the several steels which are to follow, to make sure that we all understand, as well as we may, what is “steel” as defined to-day, what are the best known varieties, and what are their characteristics?

For a rough classification it is safe for us to divide the steel world into four general divisions as follows:

1. The harder, high carbon steels used for tools, dies, etc.

2. The mild and medium steels of which wire, rod, bar, plate, pipe and structural shapes for bridges, ships and “sky scrapers” are made.

Hardening a Piece of Tool Steel. Ready to Quench

3. Alloy steels, to which some metal such as nickel, manganese or chromium gives definite properties and the name.

4. Those other modern materials which are known as “self-hardening” and “high-speed steels.”

Bowknot Made from Piece of Steel Pipe

The two classes last named are not simple iron-carbon alloys and their properties are less directly derived from and do not so plainly depend upon carbon. Metallurgically, then, they are not steels in the exact former sense of the word; but as they do require carbon—though perhaps in lesser amount, are made by regular steel processes, have most of the characteristics of steel and are used for the same general purposes, they are undoubtedly entitled to the appellation “steel.” However, to distinguish, they are usually termed “alloy steels.”

We are just now concerned only with the steels of classes one and two—the carbon steels. As explained in a previous chapter, these are alloys of iron with not more than 2 per cent of carbon.

High Carbon Tool Steel (1.25 Per Cent C) as Cast
(Magnification 70 diameters)

Low Carbon Tool Steel (.50 Per Cent C) Annealed
(Magnification 70 diameters)

Low Carbon Tool Steel (.50 Per Cent C) as Cast
(Magnification 70 diameters)

Medium Carbon Tool Steel (.86 Per Cent C) as Cast
(Magnification 400 diameters)

Carbon is the element the presence of which confers upon iron the ability to harden when cooled suddenly from a cherry-red heat, as by quenching in water or oil. If the steel contains less than four-tenths of one per cent of carbon it has little or no hardening power under this treatment; but steel with six-tenths of one per cent or more of the element, has the wonderful property of being slightly malleable when in the annealed state, but extremely hard and brittle after this sudden cooling—leads a dual life, so to speak.

Low Carbon Tool Steel (.50 Per Cent C) Hardened
(Magnification 100 diameters)

At any time, hardened steel may be returned to its former condition of softness by the well known process of annealing, wherein it is reheated to the same cherry-red heat and slowly cooled.

At the will of the blacksmith or metal worker alternate hardening and softening may be repeated a great many times without apparent deterioration.

Various degrees of hardness also, may be obtained according to (1), the percentage of carbon in the steel, and (2), the completeness and suddenness of the cooling.

As considerable brittleness and internal strain in the metal necessarily follow hardening, the hardness is usually “tempered” or “let down” by a careful reheating to a much lower temperature, usually 425 to 550 degrees Fahrenheit. From this temperature a second quenching “fastens” the temper at whatever of the original hardness the steel retains at the temperature chosen by the smith for the second quenching. Much of the brittleness is in this way relieved. The smith calls it “toughening” the steel. Tools so treated are much less liable to break.

The steels that will harden (we will call them “carbon tool steels”), range ordinarily from the .60 per cent carbon variety, used for hammers, cold chisels, etc., to those containing 1.50 per cent of carbon which are selected for razors, scalpels, and other tools requiring high temper. Each one of these many grades is susceptible of a wide variety of temper in the hands of a capable man, who must select his steel and give to it the most desirable temper for the work for which the tool is designed.

Mild Steel Pipe (.10 Per Cent C)
(Magnification 70 diameters)

Blacksmiths and other tool makers become extremely proficient in judging steels and the proper temperature at which each should be hardened and “drawn” (tempered). They judge temperatures solely by the color of the steel when heated. Every five or ten degree change imparts a slightly different shade as the steel grows hotter in the forge fire or cooler when about to quench.

Observation of a good blacksmith at work and a few minutes’ conversation with him about his “art” will give one greater knowledge and appreciation of the carbon tool steels than volumes of writings concerning them. Along with it will come more respect for the skill of these clever men whose handiwork is never exhibited in salons and about whom the world hears little, though indebted to them for a great measure of its civilization and prosperity.

High Carbon Tool Steel Is Extremely Brittle When Hardened and Has Very Little Malleability When Annealed

What and how much would be possible without machines and proper tools?

Quarter-inch Mild Steel Plate with Double Fold. Folded Cold Without Slightest Crack

About sixty years ago steels of much lower carbon content appeared. They have been made softer and softer until we have what we now know as the “mild” steels and even the almost or practically carbonless material which we called “open-hearth iron” or “ingot iron” in a former chapter. These have not the hardening property but they possess softness, ductility and freedom from brittleness which the higher carbon steels always lack. For such real evidences of our Twentieth Century civilization as the great bridges, ships, buildings, etc., they are indispensable, for they are easily cut, bent and otherwise worked into shape, and they combine pliability with sufficient strength for the service intended. Such steels are desirable, for when overloaded they bend before they break, thus giving warning of the danger.

These mild and medium steels are of immense importance industrially. Of the 31,000,000 tons of steel made in the United States during 1912 probably 99 per cent was of the soft and medium varieties.

It has been said that “the exception proves the rule.” Cementation steel is the exception to the rule which we gave in Chapter VI that steel is always melted during its manufacture.

If a thin piece of bar iron be packed in powdered charcoal and heated at low red heat for some time, the metal, after cooling, will be found to have acquired the hardening property. In other words by absorption of carbon it will have become steel with all of the characteristics of that material. Neither the iron nor the carbon by which it was surrounded have melted, yet in some way carbon has penetrated into the iron and if the heating has been sufficiently long, carbon will be found at the center of the bar. But always there will be more carbon in the outer layers of the bar than in those farther inside, i. e., it will be found in diminishing amounts as we approach the center.

Shelby Seamless Steel Tubing Crushed Endwise

Just how and when the cementation process for making steel, to be now described, was discovered is not known. It may have been the result of the non-uniform working of the larger blast furnaces which were developing in Continental Europe during the Thirteenth century. From the German “natural steel” which was probably the steely product too rich in carbon for the wrought iron which they intended to make and much too poor in carbon to be the fluid cast iron which with the growing height and heat of the blast furnace they later did make, may have come the idea. More likely, a piece of thin wrought iron was accidentally left imbedded in glowing charcoal until it had absorbed some carbon.

A Crucible Melting Room. “Melting Holes” Are Beneath the Square Covers on the Floor at the Left. Note the New Crucibles Drying on the Shelves, and the Ingot Molds at the Right

The first mention of cementation steel appears to have been by an Italian metallurgist, Vannuccio Biringuccio, who, in 1540, described the making of steel by heating billets of soft iron for a long time in molten cast iron. The modern method, the heating of wrought iron in powdered charcoal, was certainly known in the sixteenth century and this method of cementation has been practiced in France, England, Belgium and Germany since the seventeenth century.

A Sheffield (England) Cementation Furnace

Reaumur, the Frenchman, whose process of making cast iron soft by annealing bears his name and is still used in Europe, was the first to study and understand to any extent the cementation process. Publication, about 1722, of his complete directions for cementing iron gave great impetus to the manufacture of steel by this process. Fate, however, was unkind and his own nation, France, by reason of her small production of suitable iron for the work, was unable to profit greatly through his discoveries. Sweden, England and Germany were benefited to a much greater extent.

During the early years many were the secret and wonderful mixtures and compounds offered for this work, but of them all carbon in some form was the only necessary element.

Finely divided or powdered charcoal or bone dust has been mostly used.

Huntsman Crucible Furnace—Original Type

One Type of Oil-Fired Crucible Furnace

Sheffield, England, steel makers, have been very successful in the manufacture of cementation steel. Their usual method is to pack flat strips of best Swedish Walloon iron in charcoal in rectangular stone boxes about four feet wide, three feet high and fourteen feet long. Alternate layers of small-sized charcoal and thin iron bars are piled in these boxes until they are filled, the bars not being allowed to touch one another. When full, top slabs are luted on to the boxes to make them airtight.

Fire is kindled in the firebox below and the heat gradually raised until furnace and boxes are cherry-red in color. This heat is maintained for seven to eleven or more days, depending upon the hardness desired, i.e., the amount of carbon they desire absorbed. The furnace is closed and allowed to cool slowly, which requires another seven or more days.

Upon unpacking the furnace the bars are found to be brittle and of a steely fracture instead of the soft malleable material which was put in. They have become high carbon steel.

Expert workmen are able to judge very closely the hardness of the steel by looking at the fracture and they sort the bars in this way, piling bars of similar hardness together.

Huntsman Coke-Fired Crucible Furnace—Modern Type

Bars thus made show many blisters on the surface and the steel became known as “blister steel” on this account. The reason for these blisters was not discovered until along about 1864, when the well-known English metallurgist, Percy, proved that the blisters were caused by the chemical action of carbon on the slag contained in the wrought iron. The gases formed produced the blistering of the bar. That this is the explanation is proved by the fact that bars of mild steel or iron without slag do not blister.

Blister bars heated to a forging heat and drawn out under the hammer or rolled into bar steel are known as “spring steel”, or “plated bars.”

As in wrought iron manufacture, a cutting to length, repiling, heating, welding and again drawing down by hammering or rolling produces much more homogeneous and reliable steel. Piled and reworked steel of this sort became known as “shear” steel because blades of shears for cropping woolen cloth were always made in this way.

Many of us will recognize in the cementation process an extended “case hardening.” Case hardening is very largely resorted to by iron and steel workers, who in a few hours can give a hardened and long-wearing thin outer layer of steel to a piece of iron or soft steel after it has been forged or machined into the desired shape.

Siemen’s Gas-Fired Crucible Furnace—Regenerative System
One pair of Checker-work Chambers, k. h., is being heated by the hot outgoing flame and waste gases while the other pair is heating incoming gas and air. They are worked alternately.

This shear steel was largely made and was quite satisfactory, until, as described before, Huntsman, a Sheffield clock maker, conceived the idea of melting together in a pot or crucible blister bars or bars of shear steel. This he did to equalize the carbon content and give uniformity of product which had never been attainable through the cementation process alone.

From that date (1740) to this the crucible process has undergone only minor alterations and to-day it produces the highest grades of steel which we have. Practically all of the high grade tool steels are produced by this process.

Nor has Huntsman’s form of furnace been greatly changed, as the illustrations prove. Though gas and oil as well as coal are, in many cases, used as the fuel, the general design of the furnace has remained the same.

For a century crucibles were made from clay molded to form, slowly dried and very carefully burned. Usually each steel maker made his own crucibles. They could be used but three times, becoming so thin and tender after use for three batches of steel that they were not safe for a fourth. Graphite crucibles are now very largely used. They withstand the severe heat much better and can be used five or six times. The expense item for either clay or graphite crucibles is a large one.

The Stalwart Melters

After filling with small pieces of blister or shear steel the crucibles are entirely surrounded by coal or coke in the furnace pit. The fire is so regulated that the steel is not too quickly melted. Fresh coal or coke must be put in around the crucibles two or even three times.

When he thinks the steel should be molten, the expert attendant known as the “melter” quickly removes the tight fitting cover of the crucible and with an iron rod determines whether any unmelted pieces remain.

After complete melting the steel must be “killed,” else it will boil up in the mold upon pouring and leave a spongy or insufficiently solid “ingot” or block of steel. This “killing” of steel is a rather peculiar phenomenon. It is accomplished by allowing the steel to remain quiet in the furnace for another half hour or so. Undoubtedly the quieting is the result of the escape of the gases or impurities which are contained in the charge, and absorption of the chemical element, silicon, from the walls of the crucible.

We have met this element, silicon, before in our metallurgical journey and we will likely meet it several times again. To the metallurgist it is secondary in importance only to carbon.

Pulling the Crucible

When the steel has been properly melted and killed it is ready to pour. An assistant lifts the cover from the melting hole, the “puller-out” seizes the crucible just below the bulge with circular tongs and pulls it from the coke which surrounds it. The slag is skimmed off the top and the steel poured into iron molds forming small “ingots,” usually from 2 to 4 inches square and two feet or more long.

Every part of the process, even the pouring, must be done with extreme skill and care or the product suffers.

After liberation from their molds, the ingots are heated and either rolled or hammered down to the sizes desired for tools, etc.

As stated before, crucible steel necessarily is an expensive material both on account of high labor and crucible costs. For this reason, many have resorted to the process used in the very small way mentioned for the manufacture of Wootz steel—the melting of wrought iron bar or soft steel in a crucible with carbon.

In the Wootz process chopped wood and green leaves were used. Nowadays charcoal is substituted or there is added the proper amount of cast iron to give the desired amount of carbon to the wrought iron or soft steel charged. During the melting the iron takes up the charcoal and alloys with it.

“Teeming” or Pouring into Ingots. The Ingots Later Are Forged or Rolled into Bars from Which the Tools Are Made

Proper amounts of silicon, manganese, and other beneficial materials are also charged, which become either part of the alloy itself or have a cleaning or fluxing action upon it.

Steels made in this way are practically, though perhaps not quite, as good as steels made by melting together the properly selected cementation bars. The method has come to be very generally used on account of its directness and because it eliminates the long and expensive preliminary cementation process.

When Bessemer and open-hearth steels made their appearance in the market an attempt was made to use them instead of wrought iron as the base for high grade crucible steels. Though seemingly pure enough, apparently purer even than wrought iron, these metals were not able to compete with wrought iron for this purpose. For some reason, not yet satisfactorily explained, these new materials which are made in 15, 35 and 50–ton batches, when used as a base, do not give as high quality tool steel as puddled wrought iron, which is slowly and laboriously made in 500–pound lots. Considerable of these materials are utilized but it is for a somewhat lower grade of crucible steel.

For many years mild steels for castings have been quite largely made by the crucible process. They are among the best but the crucible and labor costs are usually too great to allow crucible steel castings to compete in present markets.

CHAPTER VIII
BESSEMER STEEL

The “Manufacture of Malleable Iron and Steel without Fuel” was the startling title of a scientific paper read in 1856 before the British Association for the Advancement of Science. This was the announcement to the world of Henry Bessemer’s invention of the process for making iron and steel which led to the greatest commercial development the world has seen.

To those of us who have had little or no experience along manufacturing lines the announcement seems strange enough, but metallurgists, engineers and manufacturers who know how serious is the matter of fuel bills realize at once how revolutionary the claim of Bessemer must have seemed to men of those days.

As occurs with so many new things the idea was scoffed at; Bessemer’s scheme was one purporting to give “something for nothing” and—well, it could not be.

It was ridiculous!

And why should it not have seemed strange when we consider that up to that time fuel had been required in all metallurgical processes. In the old Catalan furnace and the types that preceded it, in the Finery Fire, the Walloon and the several other refining furnaces fuel had to be provided without stint. The lowest proportion that seventy years of experiment and practice had brought about in Cort’s puddling process was one ton of coal per ton of iron, while the blast furnace required at the least four-fifths of a ton of coke for each ton of pig iron produced.

Kelly’s First Tilting Converter

Whether Bessemer, an Englishman of French descent, or William Kelly, an American of Irish descent, of Eddyville, Ky., first conceived the idea of the “pneumatic” process is a moot question. Considerable evidence substantiates the claim that the latter first hit upon the scheme and during the ten years between 1846 and 1856 had considerable success with its development. Perhaps Bessemer had heard of Kelly’s experiments. There is no proof that he did. Whether he did or not, the fact remains that he quite independently and very fully developed the process in England, and with great business sagacity and energy made it the success that it is.

As fortune has withheld from Kelly and from this country credit which was deserved, it is desirable to tell briefly the part which he had in the development of this process that with a single furnace converts pig iron into steel at the rate of a thousand tons in 24 hours and first made mild steel available as a building material.

In 1846 Kelly, with a brother, bought the Suwanee Iron Works, near Eddyville, Ky. After about a year they encountered the same difficulty that charcoal iron manufacturers usually have encountered—the failure of the supply of fuel. This difficulty Kelly, a better inventor than business man, apparently had not foreseen. His business was threatened unless some other way of refining his iron was found.

Crucible with Which Bessemer’s First Experiments Were Conducted

One day while watching the operation of his Finery Fire he noticed that the blast of air from the tuyère made the molten iron where it impinged very much whiter and apparently hotter than the rest. Like other iron makers, he had always supposed that a blast of cold air chilled molten iron.

It appears that Kelly was not long in surmising the truth. In a few days he had rigged up a crude apparatus and made soft iron from which a horseshoe and a horseshoe nail were fashioned by a blacksmith.

Fixed Converter of 1856 with Six Tuyères About the Sides

Being conservatives, Kelly’s customers were not slow in informing him that they did not want iron made by anything other than the “good old process” and he was obliged to accede to their demands or lose their trade.

Like Galileo, however, he had not really surrendered. In the woods near by he built and experimented with seven successive “converters,” as the furnaces are called in which Bessemer steel is made.

Upon learning that Bessemer of England had been granted a United States patent (1856), Kelly came before the patent office and proved that he had several years before used the same process. The priority of his invention was acknowledged, and a patent was granted to him also (1857).

Bottom Blowing Tilting Converter

Financial troubles and finally bankruptcy handicapped him. However, the Cambria Steel Co., of Johnstown, Pa., became interested and let him experiment with his process at the company’s plant. Here in 1857 he built his first “tilting” converter. His first public demonstration resulted in failure and ridicule, but a few days later he was successful. Steel makers bought interests in his patent, which at its expiration in 1870 was renewed by the United States Patent Office, while renewal of Bessemer’s patent was refused.

In 1858 Bessemer Erected His First Converter of the Form Generally Used To-Day

The Kelly Pneumatic Process Company, which was organized to operate under Kelly’s patents, built a converter at an iron works at Wyandotte, Michigan. Here the first pneumatic process steel ever made in this country in other than an experimental way was “blown” in 1864.

Meanwhile Alexander L. Holley, an American engineer, had obtained for another American company the right to manufacture steel here under Bessemer’s patents. He built a plant at Troy, New York, which began making steel in 1865.

Even the Detachable Bottom—to Facilitate Repairs—Was Thought of and Patented by Bessemer—1863

It was soon decided to merge the interests of the two companies and in 1866 this was done, the process thereafter being known as the Bessemer Process. During the early years of the process here Holley became very well known. As consulting engineer he designed practically all of the Bessemer plants which were built during the first ten or fifteen years.

To the majority of the people of the United States to-day Kelly and his parallel part in the great invention are practically unknown, and thus not only he but the United States is without credit which should be ours.

Fortunately Kelly did not entirely fail to profit financially as so many times is the case with inventors. He received a total of about $500,000. Bessemer’s return from his process is said to have approximated $10,000,000 and he was knighted by the British sovereign.

More intimate details regarding Kelly and his work may be found in Munsey’s Magazine for April, 1906, where H. Casson gives information which he received direct from several of the men who knew and worked with Kelly.

While apparently not the originator of the process, Bessemer is without any doubt entitled to most of the credit he received. There is no proof that he had heard of Kelly’s experiments when he began his own or that he was aided by Kelly’s discoveries. He worked out the details of the process independently, as had Kelly, and it was Bessemer who put it on a commercial basis.

As has occurred with other new processes Bessemer’s first licensees were not particularly successful. When those who had bought the right to use his process had failed in their efforts to use it, and become discouraged as most of them did, he quietly bought back their rights and went ahead with his development of the process. Perhaps no man ever exhibited more perseverance in continuing experiments and development under very discouraging conditions than did Henry Bessemer. He had faith.

Sectional View of a Modern Converter Showing Air Duct and Tuyères

He had a genius for invention and was thorough in his experimental work. Practically no type of converter has since been brought out that he did not think of and try, and the process has been modified in but one or two important particulars in the years that have passed.

The essential part of the Bessemer process is the blowing of air through molten cast iron to remove the metalloids by which cast iron differs from steel and wrought iron, as has been explained before.

This being the essential point, and at first thought the lack of fuel seeming so peculiar, we must describe what happens during the Bessemer “blow.”

Pouring Charge of Molten Pig Iron into Converter

Technically speaking, the metalloids are “oxidized.” Oxidation is the chemical uniting of oxygen, generally from the air, which has 21 per cent of this element, with another element or material such as iron, silicon, carbon, wood, coal, etc. If the oxidation is slow as in the “rusting” of iron, the resulting heat dissipates as fast as it is generated and the change is hardly noticeable. If, however, the reaction occurs rapidly and with vigor enough, we say that the material “burns.” The latter sort of oxidation is what we call “combustion.”

The affinity between the metalloids and oxygen has been noted by us before, but in those cases most of the oxygen came from a different source.

In the wrought iron process most of it was furnished by the iron ore or scale which was stirred into the metal, or by the slag which covered the “bath.” In the Bessemer, or as it was first known in America, “Kelly’s air blowing process,” the oxygen of the air blown through the molten metal directly oxidizes or burns out the carbon, silicon, and manganese. The extremely rapid oxidation of these furnishes the heat.

The iron, then, furnishes its own fuel and no outside combustible is needed.

How can this be?

In every ton of molten cast iron there are approximately 70 pounds of carbon, 25 pounds of silicon, and 15 pounds of manganese or a total of about 2000 pounds of these metalloids in the fifteen-ton charge of molten metal which goes into the ordinary steel plant converter.

We know that if burned in a furnace this ton of high grade fuel would generate much heat. Burned inside of the mass of molten metal it generates exactly that same amount of heat and the heat is applied with such rapidity, directness and efficiency that the molten iron which had a temperature of 2300° F., say, when charged, in nine or ten minutes has become steel with a temperature of about 3000° F. simply through this rapid oxidation of its 4 to 6 per cent of metalloids.

How the blast under 15 to 30 pounds per square inch is applied through little nozzles in the bottom of the modern “converter” and the several types of vessels with which Bessemer experimented in the course of his investigations are shown in the illustrations.

Nor is it necessary that the air be blown through the metal. Air blown upon its surface accomplishes practically the same purpose, and in many of the steel foundries of to-day smaller converters of this “surface-blown” type are used for producing steel for castings. The large steel plants, however, use the larger “bottom-blown” converter. Two or three of these vessels, working with proper metal from the “mixer,” produce an immense tonnage of steel each 24 hours.

The “mixer” is quite necessary. It is a large vessel or furnace holding and keeping hot from 75 to 300 or more tons of metal from the blast furnace. It mixes and equalizes irons of various compositions, so that the converters have the advantage of uniform and hot metal with which to work.

In addition it is made to perform a “refining” service. By mixing into the metal a quantity of manganese, considerable of the sulphur present (a deleterious substance) is removed.

The fifteen or twenty minute blowing of 15 tons of metal in the big egg-shaped converters of a steel plant presents a spectacle which, when once observed, will never be forgotten.

One sees a little “dinky” engine come shooting into the converter building with its ladle of molten iron from the “mixer.” With America’s time saving routine not a single minute is lost while emptying the metal into the converter, now in a horizontal position. Almost before the ladle is out of the way, the converter swings to the upright position with the blast already on, for otherwise the metal would flow into the tuyère holes at the bottom.

Comparison of Ingots
A. From Four Pot Crucible Furnace: Each Heat 400 Pounds in 4 Hours or 100 Pounds an Hour, Each Heat Pours 4 Ingots 3 × 3 × 36″. B. From Fifteen Ton Bessemer Converter: 30,000 Pounds in 20 Minutes or 90,000 Pounds an Hour, Each Heat Pours 6 Ingots 19 × 20 × 62″. C. From Fifty Ton Open-Hearth Furnace, 100,000 Pounds in 8 Hours or 12,500 Pounds an Hour, Each Heat Pours 6 Ingots 24 × 32 × 72″.

Reddish-brown smoke and a shower of sparks come from the converter. These gradually develop into a flame.

The blast shows considerable partiality in selecting for its first attention the metalloids silicon and manganese, in preference to the iron itself or any other of the metalloids present. After from three to five minutes half of the silicon and manganese have been burned out. If the temperature of the metal and other conditions have become right the carbon then begins to burn. This gives a change in the nature of the flame which becomes large and of a dazzling whiteness.

The metal is hot—very hot—so much so that pieces of cold steel often must be dropped in to cool it somewhat. This is known as “scrapping” the charge.

An experienced blower can judge through every period of the operation of the condition of his metal and just how things are progressing.

After some minutes the flame begins to waver and later “drops”; i.e., there is scarcely a flame at all. This signal, which is very definite to an experienced man, cannot be lightly disregarded. Oxygen has affinity for iron as well as for the metalloids and it is only because of its greater love for silicon, manganese and carbon that it has thus far largely neglected the iron. With the metalloids mentioned out of the way, as they are when the drop occurs, the iron will begin to burn. Were the “blowing” continued we would shortly have no iron left, but in its place a mass of iron oxide and slag.

Thus we see that during the first minutes of the blow, more than one-half of the silicon and manganese are burned. The remainder of these and all of the carbon are removed in the subsequent five or six minutes. At the end of this short blowing period we have practically pure iron.

Two Converters in Operation and a Third Pouring

The metal is not yet in condition to pour well, however, largely because of the dissolved air and gases which it holds. Something akin to the “killing” of the steel which we observed in the crucible process must be accomplished or ingots from it will be spongy. And, having practically no carbon, it is not yet “steel.”

Bessemer, knowing that the finished steel should contain carbon, tried to stop the blow long enough before the drop of the flame to leave exactly the desired amount of this element. He found this difficult to do and therefore uncertain. It was found to be far better to blow until the drop of the flame and then put back sufficient carbon to give the proper composition.

An English metallurgist named Mushet discovered that addition of manganese ridded the metal of injurious gases and oxides and what is known as “red-shortness.” After a period of difficulty without it Bessemer acknowledged the necessity of manganese and adopted its use. It had before this been used in crucible steel.

Upon turning down the converter at the drop of the flame, the blast is turned off and a smaller ladle is run in on a track above. This brings a molten mixture of irons, usually known as “spiegel” or “spiegeleisen” which contains just enough carbon, manganese and silicon to give to the whole of the molten metal in the converter the metalloids needed to make of it steel of the composition desired. This addition also accomplishes the “deoxidation” of the metal. By deoxidation we mean that the iron is relieved of the oxygen and gases which have remained as a result of the blast. This is necessary in order to give proper fluidity for pouring and the best physical properties to the finished steel.

After “recarburization,” as this addition of manganese-silicon-carbon metal is called, the steel and slag are quickly poured out into a ladle waiting below from which the steel is “teemed” (i.e., poured), through a “nozzle” or hole in the bottom into ingot molds arranged on trucks on the railroad track which runs through the building.

Teeming the Finished Steel into the Ingot Molds

When the molds have been filled and a strong crust develops on the steel the cars are pulled to the “stripper” where the molds are removed, leaving the white-hot ingots standing on the cars.

The ingots mentioned in the chapter on Cementation and Crucible Steel were usually small enough that one pot of 100 pounds of metal filled the mold. A four pot furnace therefore produced 400 pounds. Now for the first time, we are talking in tonnage figures. Instead of a batch of steel making four 3″ × 3″ × 36″ ingots of 100 pounds each, the ordinary “heat” of Bessemer steel from the 15–ton converter gives six or seven ingots about 18″ × 20″ × 60″ in size. Each of these weighs about two tons. The total is 30,000 pounds.

From the stripper the ingots go to the gas-fired soaking pits where the molten interiors of the ingots gradually solidify by cooling while the outer crusts are reheated. After equalizing the temperatures of exteriors and interiors in this way, the ingots are white-hot again and ready for rolling.

Molds Being Stripped from Ingots

The purpose for which the steel is intended, of course, determines the shapes and sizes into which the ingots are rolled. For rails they are rolled down directly, each ingot making about six rails, of thirty-three-foot length. For most other purposes the ingots are rolled in the slabbing mill into billets or slabs which are of intermediate shapes and sizes which are reheated and further rolled down into axles, bars, shapes, wire or other products.

Meanwhile the converter which we saw emptied has not been idle. The American steel engineer has genius for mechanical efficiency and all parts of a great steel plant are so co-ordinated that enormous quantities of material can be handled with not a moment lost between trips. Almost before the ladle of steel had swung away from the converter’s mouth, any remaining slag was dumped from the converter by further tipping, the vessel returned to receiving position and the ladle car, back again from the mixer, poured in the next charge.

Thus blow after blow is made without loss of time.

Repairs are allowed to take no longer than is absolutely necessary. When the lining around the tuyères gets too badly cut by the action of the air and metal the bottom is removed, another one is quickly substituted and the steel making goes on.

Blowers, ladlemen, cranemen, pourers, patchers, vesselmen, sample boys and the other workmen are relieved by their “partners” at the end of each shift, each man of necessity working until relieved—twelve, twenty-four, or even thirty-six hours, for there must be no delay. So day and night, through the entire week from Monday morning at six, when they begin, until the next Sunday morning at six, when the plant shuts down for a brief spell, the converters go on turning out three heats per hour or four to five hundred per week each.

It has been mentioned that most of Bessemer’s first licensees failed with the new process. The reasons for this were various, but one in particular was the attempt of many to use metal of high phosphorus content. Bessemer soon discovered that no phosphorus was removed during the “blow” and that, as phosphorus in quantity over one-tenth of one per cent was detrimental to steel, it was necessary to use raw material which had little of this element.

This could be done, but it barred many pig irons otherwise good. Fortunately Swedish and many English irons had low phosphorus. Germany’s vast beds of high phosphorus ores, however, were useless for the purpose.

For twenty years this situation existed, during which time many metallurgists endeavored to make the process applicable to irons which contained high phosphorus. After long study and many experiments the problem was solved by Sidney Thomas, an English metallurgist. With a cousin, Percy Gilchrist, he made hundreds of blows with a toy converter holding only eight pounds of iron.

Bessemer’s linings had been of sand, clay and other earths which are known chemically as “acid” materials. By using “basic” materials such as limestone, dolomite, etc., for the converter lining and additions of limestone or burnt lime to the charge before and during the blow to make and keep the slag “basic,” Thomas was able to make the phosphorus burn after the carbon had been removed. Therefore, a three or four minute “after blow” following the “drop” of the carbon flame took out the phosphorus,—again, with generation of heat.

So there are two varieties of the process—the acid Bessemer and the basic Bessemer, but the former, only, is used in this country as we have few high phosphorus ores. The analogous open-hearth processes, which are next to be described, are both used in this country with the basic open-hearth greatly in the lead.

However, the basic Bessemer process of Thomas and Gilchrist is credited with making Germany’s great industrial development possible.

Year	Table No. 1 Materials Used for Rails[[6]]			Table No. 2 Total Steel Made by Processes[[6]]
	Wrought Iron	Bessemer Steel	Open-Hearth Steel	Bessemer Steel	Open-Hearth Steel	Crucible Steel
1849	21,710
1850	39,360
1855	124,000
1860	183,000					No Data
1865	318,000
1867	410,000	2,280		2,679
1868	445,970	6,450		7,589
1869	521,370	8,620		10,714	893
1870	523,000	30,360		37,500	1,339
1875	448,000↘	260,000		335,000	8,080	35,180
1880	441,000	↘852,000	12,160	1,074,000	110,850	64,660
1885	13,000	959,000	4,280	1,515,000	133,000	57,600
1890	14,000	1,868,000	3,590	3,689,000	513,000	71,200
1895	5,810	1,300,000	700	4,909,000	1,137,000	68,700
1900	695	2,384,000	1,330	6,685,000	3,398,000	100,500
1905	318	3,192,000	183,000	10,941,000	8,971,000	102,200
1906	15	3,391,000	186,000	12,276,000	10,980,000	127,500
1907	925	3,380,000	253,000	11,668,000↘	11,550,000	131,000
1908	71	1,349,000	572,000	6,117,000	↘7,837,000	63,600
1909		1,767,000	1,257,000	9,331,000	14,494,000	107,400
1910	230	1,884,000↘	1,751,000	9,413,000	16,505,000	122,300
1912		1,100,000	↘2,105,000	10,328,000	20,780,000	121,500
1913		818,000	2,528,000	9,546,000	21,600,000	121,200
1914		324,000	1,526,000	6,221,000	17,175,000	89,900
1915		327,000	1,775,000	8,287,000	23,679,000	113,800
1916		440,000	2,270,000	11,059,000	31,415,000	129,700

[6]. In United States—long tons of 2,240 pounds.

The well-known “Thomas Slag” which is in demand as a fertilizer on account of its phosphorus content is the by-product of the basic-lined converter.

An idea of what the invention of the Bessemer process meant to railroad development alone may be gained by studying for a moment Table No. 1. Wrought iron was our first material for rails, but, being very soft, it did not give long service. But a short time was required for Bessemer steel to displace it for rails when steel became available. The greater uniformity, strength and hardness of the alloy gave such excellent wearing properties that few rails of iron were laid after the year 1880.

During recent years rails have been made of greater and greater strength and hardness to keep pace with the fast increasing weight, speed and frequency of railroad trains, steel being susceptible to much modification of properties.

Now it appears that Bessemer steel is giving way to other products which show even superior properties.

What happened in the railroad world to a great extent has happened elsewhere, as the figures of Table No. 2 show. They are a barometer which indicates what has been our industrial development and our advance in civilization.

CHAPTER IX
THE OPEN-HEARTH PROCESS

Bessemer’s was a wonderful process, but the time seemed to be ripe for great development along metallurgical lines, and the method of converting pig iron into steel which he devised soon had a competitor which was destined eventually to take the lead in steel production. Many years passed before the tonnage turned out annually by Bessemer’s process was equaled by that of the new, but as shown in the last chapter the Siemens-Martin or open-hearth process in 1907 produced the greater tonnage. It has since retained its lead and probably will continue to do so.

As far back as 1845 John Marshall Heath took out a patent for a process for making steel patterned after the old puddling process. In a way he may therefore be said to have devised or forecast the open-hearth process, but because of the great obstacles that had to be surmounted in getting a furnace that would fulfill the requirements he was unable to carry out his scheme. You will remember that in the puddling furnace the purified metal became pasty because of its high melting point. Because of the great heat required it was not until the invention of the regenerative system by C. W. Siemens in 1860 that the open-hearth process was possible. Siemen’s furnace was the first one that could keep the iron molten. It was in Birmingham, England, that the first successful open-hearth furnace was used.

alt='GAS PRODUCER FURNACE AND REGENERATIVE SYSTEM'
Early Type of Gas Producer, Regenerators, and Open-Hearth Furnace. Course Taken by Air, Gas, and Products of Combustion Are Plainly Shown, as Are the Valves That Reverse Direction of Flow

While not as speedy nor as prolific a producer as the Bessemer process and far less spectacular, the open-hearth has several advantages.

The acid Bessemer was always handicapped because pig iron with less than 0.1 per cent of phosphorus was necessary. The majority of ores carry more than this amount. The basic Bessemer requires pig iron containing not less than 2 per cent of phosphorus. The vast quantities of material which contain percentages of phosphorus between these limits are useless as far as the Bessemer process is concerned.

To be successfully used the pig iron must be further limited as to composition. It must have sufficient silicon, manganese, and carbon to give the heat required for Bessemerizing, as the burning of these metalloids has to be depended upon for the conversion to steel and to give proper fluidity to the finished alloy.

Then, too, the large amount of air forced through to a certain extent “over-oxidizes” the bath and some of the gases are mechanically retained by the steel no matter how complete the deoxidation. There also is loss of metal due to unavoidable “spitting,” for the rapid streams of air mechanically carry some metal and slag with the flame out of the vessel.

On the other hand, for the open-hearth process can be used pig iron of widely varied character and composition and, further, large percentages of low-priced steel scrap can be utilized in the charges; as no air is blown through the metal and little comes in contact with it, the conversion takes place quietly and smoothly and with much less loss by oxidation, the yield of steel usually being from 90 to 97 per cent of the metal charged as against 83 to 87 per cent which is the yield by the Bessemer process; besides giving less over-oxidation and gases in the metal, the slowness of the conversion is an advantage, as control is very easy, and, when desired, samples for test may be taken. From his tests the melter can be quite certain when he taps out the steel that it is of the composition desired.

Section Through Typical Stationary Open-Hearth Furnace, Showing Construction of Furnace, Lining, Bath and Air and Gas Ports

The melting in an open-hearth furnace is done largely by indirect or radiated heat, and it is not intended that the flame shall impinge too directly upon the surface of the bath.

Boxes of Steel Scrap and Electric Charging Machine in Front of Charging Doors at Rear of Furnaces

Except during the melting down of the pig iron and other materials charged in the furnace, the flame and air take little part in the actual elimination of the metalloids. Their main function is to furnish the heat necessary. Being used so indirectly—mostly by radiation from the roof and walls—very great heat must be used and much would be wasted if special precautions were not taken to save it. The bath must be kept hot enough to remain molten after purification of the metal, which we were unable to accomplish in the wrought iron puddling furnace.

Under each end of the rectangular furnace are two chambers built up with checker-work of fire brick. These sets are in duplicate and each has one chamber for air and one for gas.

Charging Machine with Box of Scrap Half Way into Furnace

Thus an open-hearth furnace will be seen to occupy a sort of hollow square, the furnace proper forming one side, the regenerative chambers two sides, with the chimney and flues the remaining side. “Reversing” valves force the incoming gas and air to travel each through its respective hot regenerating chamber up through the ports and into the furnace where they unite and burn with a very hot flame. The hot gases leave through similar ports in the other end of the furnace and on their way to the chimney heat the checker-work in the regenerative chamber. Every fifteen or twenty minutes the valves are reversed and the direction of flow is changed. In this way the incoming gas and air are preheated and in the furnace burn with a very much hotter flame than would cold gas with cold air. No blast is required, the draft caused by the chimney being sufficient.

For protection of the roof from the great heat developed and the metal of the bath from too great oxidation, the air ports usually are located above the gas ports. The streams of air, while protecting the roof from the flame, at the same time are prevented from directly impinging upon and too strongly oxidizing the metal of the bath.

The diagrammatic sketches given show roughly a furnace, regenerative chambers, ports, etc.

Charging “Hot” Metal

The original intention was to melt pig iron and reduce it; i. e., burn out the silicon, manganese, and carbon by action of the flame and addition of iron ore. This was the process worked out by Siemens in England. In France, P. and E. Martin altered the method by diluting molten pig iron in the Siemens furnace by melting and dissolving in it steel scrap. It was soon found that a combination of the two methods was better than either one alone and the open-hearth process acquired its name—the Siemens-Martin—in this way.

In the United States about 20,000,000 tons of steel are made annually by the basic open-hearth process while only 1,100,000 tons are produced by the acid open-hearth process.

The two processes are practically the same except that by the basic process the phosphorus as well as the silicon, manganese, and carbon are reduced or eliminated. In order to take out the phosphorus, additions of lime (i. e., calcium oxide or calcium carbonate) are made just as occurred with the basic Bessemer process.

Should we use lime in a furnace having an acid lining, much of the lime, which is a “base,” would react with the “acid” (silica) bricks of the lining, and, becoming neutralized, would not do its work. So, as in the basic Bessemer process, we here have to use either “basic” or “neutral” lining.

The material generally used is burnt magnesium carbonate which is known as “magnesite.” Dolomite, which is a combination of the carbonates of calcium and magnesium, is sometimes used. Chrome bricks, the usual neutral material, are rather too expensive for extensive use. The best magnesite comes from Austria and is usually not very cheap. As acid materials (those of silica or clay) are cheaper and mechanically stronger, a compromise is ofttimes effected by using basic materials for the furnace bottom and acid bricks for the sidewalls and roof. A few rows of chrome bricks may be put in to form a neutral dividing line just at and above the edge of the bath where the action of the slag is the most severe. It also serves to keep the basic and acid materials apart and from reacting with each other.

At the commencement of charging, limestone or sometimes burnt lime is shoveled in upon the bottom or “hearth” of the white-hot furnace.

When cold metal is charged, the pigs of iron are conveyed into the furnace by the melter and his helpers by means of long handled flat iron tools called “peels.” This is followed by charging some or all of the scrap or iron which is to be made a part of the charge.

Even in the smaller 15 or 25–ton furnaces hand charging takes a great deal of time, sometimes as much as six or eight hours, and the labor cost as well as the heat loss is therefore excessive. Modern machine charging which requires not more than an hour is therefore highly desirable.

Row of Open-Hearth Furnaces Showing Pit or Tapping Side

During the melting down of the pig iron with the scrap that has been charged, the air and flame burn out about half of the silicon and manganese of the metal. To remove the remainder of these and the carbon of the charge, additions are made from time to time of sufficient ore to keep the bath “boiling.” This phenomenon results from the giving off of carbon monoxide gas formed from the oxygen of the iron ore and the carbon of the metal, just as happened when the puddler in the manufacture of wrought iron used iron ore in his furnace. The covering of slag which forms and protects the bath from the flame undoubtedly transfers oxygen from the furnace gases to the bath and this helps to burn out the carbon.

The lime charged unites with the phosphorus of the iron and takes it into the slag which covers the bath. If necessary, further additions of lime may be made from time to time during the melting and the “working down” (elimination of the metalloids) of the charge. As long as the slag is kept basic it retains the phosphorus, but should it turn acid the iron of the bath would take the phosphorus back again.

These reactions are all chemical, just as much so as are the burning of wood and coal and the thousands of reactions which are brought about in chemical laboratories.

Additions of ore are made from time to time and the bath rabbled. Samples are taken now and then with a long handled iron spoon or ladle and these are poured into molds to form small bars of steel, which, after quenching, are broken.

Open-Hearth Furnace “Tapping”

The melter has become very proficient in judging the composition of the metal of the bath from the fracture of these broken test pieces. By means of the samples taken he watches the elimination of the metalloids. When he thinks the reactions have progressed far enough he takes a last sample which is rushed to the chemist who makes a hurried “control” analysis for carbon and phosphorus, the metal being held in the furnace meanwhile. If the results of this analysis show the bath to have the desired composition the steel is poured. If the reactions have not been complete, the chemist’s report shows that the carbon and perhaps the phosphorus are still too high, in which case the charge must be still further worked down.

Some melters are able to make fairly uniform and satisfactory steel without a chemist, but for best results a chemical laboratory is desirable.

When ready to tap, the big ladle is suspended from a crane under the spout of the furnace. With a tapping bar the plug of clay is removed from the tap hole and the molten steel gushes out into the ladle. The slag which has covered the bath is the last to drain out. Many times this will overflow the ladle, making a beautiful cascade as it pours over the sides all around to the floor beneath. Especially at night is this a glorious sight.

Teeming the Steel into Ingot Molds

Recarburization is not done to the same extent as it is in the Bessemer process. As the open-hearth elimination of carbon is slower and under so much better control, the furnace usually is tapped when the carbon has been reduced to the percentage desired in the finished steel. When it is necessary to add carbon it is done sometimes by adding pig iron to the bath and sometimes by throwing a weighed amount of coal or coke in the ladle as the steel is going in. Molten iron and steel have strong appetites for carbon and dissolve it very readily. Ferro-manganese is used to prevent red-shortness and to deoxidize the metal. This also is usually put into the ladle as too much loss would occur were it added in the furnace.

While the furnace is again being charged through the charging doors at the rear, the steel is teemed through the nozzle of the big ladle into the waiting ingot molds. These go to the stripper, to the soaking pits, and then to the rolls of the blooming mills just as did the Bessemer ingots.

At the “Stripper”

In the acid-lined furnace no attempt is made to reduce the phosphorus. It would be futile. Therefore the materials charged must be very low in phosphorus and sulphur. No lime additions are made, the flame simply melts down the pig iron and scrap, the iron oxide later is added from time to time to keep up the boil until the test bars show that the carbon as well as the silicon and manganese have been eliminated as fully as is desired. The metal is then tapped as described above.

Three or more hours are usually required to melt down cold charges. The elimination of the remainder of the silicon, manganese, and the carbon requires about four or five hours more. So for each heat the open-hearth furnace requires from eight to twelve hours, depending largely upon the speed of charging and melting.

Of late years the difficulties attending the use of molten metal from the blast furnace in place of cold pig iron have been largely surmounted. The use of uniform metal from the “mixer,” which was described in the article on the Bessemer process, has aided the open-hearth process also. Of course, when molten metal is added none of its silicon and manganese is reduced by the flame as occurred with the cold metals during the melting down, so the molten metal charged is usually low in these elements to compensate. By use of “hot” (molten) metal the time necessary to produce a “heat” of steel is considerably shortened.

The first and perhaps the majority of furnaces yet building are “stationary.” Some have found it advantageous to construct furnaces that can be tipped to pour the metal into the ladle. Such are known as “tilting” furnaces. One furnace designer has even gone so far in a smaller type used for steel castings as to make the furnace removable, thus doing away with a ladle entirely. The big crane simply lifts the whole furnace out from between the housings which contain the ports. It is taken bodily to the molds which are poured directly.

Open-hearth furnaces have been built of larger and larger capacity. A great many fifty-ton furnaces have been built and furnaces which produce eighty or more tons at a heat are now not uncommon.

Furnaces of the Talbot type are built for as much as 200 and even 300 tons of metal, but from these only part of the finished steel is tapped at a time, the remainder being left to help work down the additions of new material which is added to replace the steel tapped out.

The rolling mill industry is so intimately connected with and dependent upon the steel-making methods and equipment that each is designed with reference to the other.

Bessemer steel has been largely used for the manufacture of rails, rod, wire, pipe, merchant bar, etc., while open-hearth steel has gone into plate, boiler tubes, structural shapes, billets for axles, etc. Recently it is being used for rails and very many of the products which were formerly made from Bessemer steel.

Lower Half of a “Battery” of Modern Gas Producers

It should not be inferred from this that Bessemer steel is no longer in demand or that it is not good steel. As you will notice from the table given in the last chapter, the production of Bessemer steel has not declined appreciably, if at all. The fact is that open-hearth steel production has been increasing at a great rate, while the production of Bessemer has remained stationary. With the growing scarcity of ores suitable for pig iron for Bessemerizing, the open-hearth process is becoming able to compete with the Bessemer process in the matter of cost. For some purposes the steel is considered to be a little more desirable, but, as is the case with many good things, the pendulum swings too far and there is no doubt that open-hearth steel is often demanded and used for purposes for which Bessemer steel would be just as good and perhaps better.

For many years it has been said that the Bessemer process is “doomed.” This, of course, was because of the scarcity of low phosphorus ores. Just how “doomed” it is, it is perhaps impossible to say. Certainly it is still a very live process and the combining of processes, such as “duplexing,” will probably prolong its life.

By the “duplex” process, molten blast furnace iron from the mixer is “desiliconized” in the Bessemer converter. Before too much of the carbon has been burned, the metal is transferred to a basic open-hearth furnace where the remainder of the carbon and most of the phosphorus is removed. By this method the advantages of the open-hearth and much of the speed of the Bessemer process are combined. The output of the open-hearth furnace is thus greatly increased.

To-day all kinds of combinations of Bessemer, open-hearth, and electric furnace are being projected and it is difficult and likely impossible for any one to predict the future of any of the processes.

Lest the metallurgical facts scattered through several chapters escape, let us summarize a little. Roughly speaking, the capabilities of and materials required for the processes are as follows—the chemical symbols for silicon, manganese, carbon, phosphorus, and sulphur being used for brevity:

Process	Refining Capability	Material Required
Crucible Process.	Removes no metalloids, but simply remelts.	Very low Si. P. S. and C.
Acid Bessemer Process.	Takes out Si. Mn. and C.	Very low P. and S. (under 0.1%).
Basic Bessemer Process.	Takes out Si. Mn. C. P. and some S.	Very high P. (2% and over).
Acid Open-Hearth Process.	Takes out Si. Mn. C.	Very low P. and S. (under 0.1% of each).
Basic Open-Hearth Process.	Takes out Si. Mn. C. P. and some S.	Wider Variety.
Electric Furnace Process.	Takes out Si. Mn. C. P. and S.	Wider Variety.

In further explanation of the competition in quality of Bessemer and of open-hearth steels it should be understood that in both the acid Bessemer and the acid open-hearth furnaces we get out in quality just what we put in. While for some purposes phosphorus and sulphur of 0.1 per cent is allowable, for other purposes they should not be over 0.025 or 0.03 per cent. To produce steel of the latter high quality, material containing slightly less than this of sulphur and phosphorus must be charged, and these are usually much higher in price than are pig iron and scrap containing greater percentages of these metalloids.

Charging Floor of the Battery of Gas Producers Showing Rocking Arms for Gradual Feeding of the Coal

Where materials of 2.5 to 3 per cent of phosphorus are obtainable, as, generally speaking, they are not in this country, the basic Bessemer should make as low phosphorus steel as does the basic open-hearth.

The great advantages of the basic open-hearth process, then, are that for it can be utilized a much wider variety of raw materials than is possible with the acid open-hearth or either of the Bessemer processes, and, particularly that here, at least, the proper materials are readily available.

The fuels used vary, of course, according to what is most available, considering quantity, quality, and price. Natural gas has been a favorite fuel, as also has oil. But in many localities natural gas never was available and in others which were thus blessed, the supply has been exhausted. By-product coke oven gas and tar are being experimented with with some success.

Largely because of the great size of the open-hearth furnace solid fuel, such as coal which can be used for puddling furnaces, is not adaptable.

As far back as 1839 attempts were made to gasify coal by burning it to ash and utilizing the gaseous products for industrial purposes. These attempts succeeded and the process has been brought to quite a high state of development. There are to-day a large number of efficient types of “gas producers” which furnish gas for general industrial use and it is with this “producer gas” that a great deal of the steel nowadays is made.

While endeavoring to leave out of these articles most of the chemistry and as much of the technical detail as is consistent with clearness, the chemistry of combustion and the “gas producer” is so interesting that it will be well to explain that carbon (coal, coke, wood, etc.) can burn either in one or two stages. Nearly every one has noticed the blue flame with which coal burns in the parlor coal heater or in other furnaces where little draft is used and most of us remember that the gas which is given off from such a fire has asphyxiated many who were unfortunate enough to be sleeping in a closed room, when through insufficient chimney draft or a leaky stove some of the unburnt gas filled the room.

This gas, which is carbon monoxide, is labeled CO in books on chemistry. It is the result of burning the coal with insufficient air. Chemically it is explained by the second of the chemical “equations” which follow. The third equation explains the second stage of the burning which would occur were further air or oxygen admitted to the upper part of the furnace.

The usual one-stage combustion with plenty of air:

1. C (carbon) + 2O (oxygen) burns to CO₂ (carbon dioxide). Non-poisonous.

The two-stage combustion with insufficient air:

2. C + O burns to CO (carbon monoxide). Poisonous.

3. CO + O burns to CO₂. Non-poisonous.

Carbon monoxide asphyxiates by forming a chemical compound with the hæmoglobin of the blood, which therefore is prevented from supplying the body with the oxygen that is required for the sustenance of life.

Carbon dioxide is no such poisonous product, as may be inferred when we remember that it is the gas with which our carbonated waters are charged and which is so commonly served with ice cream in ice cream soda.

Now in a gas producer, by maintaining a sufficiently thick bed of glowing coal and admitting only such amounts of air as will produce mainly carbon monoxide gas, a product of high burning value is obtained. A kilogram (2.2 pounds) of carbon in burning from C to CO generates only 2450 calories or heat units, whereas its complete burning to CO₂ would give 8080 calories. So by conducting the carbon monoxide gas—the product of the first stage of the combustion—through brick-lined pipes to the furnace, and in the latter by addition of air allowing it to burn to CO₂, the greater amount of heat (i. e., 8080 minus 2450 or 5630 calories) is evolved in the furnace. Of course, some of this theoretical two-thirds which is in this way made available at the furnace is lost because a little CO₂ is formed, and always the nitrogen of the air used greatly dilutes the gas. But there are gains, notably the great heat which is carried over by the hot gas from the glowing bed of coal and that from the water-gas which is formed from steam used in the producer. So, all in all, the gas generated in a “battery” of gas producers, all of which discharge into one large main or header to maintain gas of average composition, is quite a satisfactory fuel.

CHAPTER X
CAST IRON

From the preceding chapters we now know pretty well the place which cast iron occupies in the iron family. In the chapters which have succeeded the one in which we discussed the blast furnace and pig iron, every one of the products except crucible steel has been produced through some “refining” operation which greatly changed the composition, structure and properties of the product. Cast iron is not the result of a refining operation in this sense of the word. It is produced through simple mixing of pig irons of various compositions, usually with some admixture of iron castings of similar composition which have outlived their usefulness in the industrial world and have been returned as scrap to be remelted.

When we say that cast iron is not produced through a refining operation, it must not be inferred that no change in composition occurs during the remelting. There is some change, notably a loss through oxidation from the air blast of a little of the silicon and manganese. Aside from this there usually will be absorption of enough, or sometimes more than enough, carbon from the coke used in melting to make up for the carbon which is oxidized. Usually some sulphur also is taken up from the fuel. There is, however, no such actual or intended alteration of composition through burning out of the metalloids as is necessary for the production of wrought iron and steel.

But from this we must not assume that the manufacture of cast iron for chilled rolls, car wheels, machine parts, valves and fittings, etc., is an easy proposition. As we will soon see, accurate regulation is required of metal for proper depths of “chill” for rolls, car wheels and castings which must have high resistance to wear. Too, the metal for valves and fittings and other more or less complicated castings for high steam, air, ammonia, water, etc., must be uniform, of close grain, strong, yet soft enough to machine easily at the extremely high speeds which modern efficient tools and methods demand. The production of the best metal for such work requires the use of properly selected materials, judicious mixing, and clever operation of the cupola furnace, that the molten metal delivered to the foundry for the pouring of the molds may be hot and fluid and of the right composition for the particular work in hand.

Sampling Cars of High Silicon Pig Iron

It is always interesting and instructive to follow the materials through their course from the “raw” state to finished products, and, therefore, we are going to take you on a little trip from the receiving yard of a firm making cast iron goods where we see the cars of pig iron just in from the blast furnace and where the materials are sampled and held pending analysis, to the laboratory where the samples are analyzed, then to the storage bins where the materials are unloaded, and, later, with the weighed charges, to the cupolas which convert them into molten cast iron of the proper composition and quality for high grade castings.

Sampling Other Pig Irons
Pig irons of lower silicon content cannot be broken easily with a sledge but usually are thrown from a height across an iron block.

Twenty years ago it looked as if the iron foundry would be one of the last strongholds of “rule-of-thumb” to give way to scientific methods. It does not look so to-day, though there are many foundries which yet buy and use their pig iron on the basis of fracture; i.e., the foundryman guesses by judging of the color and closeness of grain and other characteristics of fresh fractures of the pig irons how suitable they are for his purpose and in what proportion to mix them. A skillful man can get fair results in this way only so long as he uses the small number of brands of pig iron with which he is perfectly familiar, and even then there must be but little fluctuation in composition of the irons used and he must be allowed considerable latitude in the quality of the iron which he produces.

Success by this method is even more difficult now than it was ten years ago, for the advent of many new blast furnaces and their greater variety of products have made this rule-of-thumb mixing a much more uncertain matter than it formerly was. Machine-made pigs, which are so generally on the market now, give fractures which tell little regarding their compositions.

Drilling the Samples
No oil or other lubricant is allowable and the drillings are taken up with a magnet that no sand or other impurity may get into the sample for
analysis.

While some foundries still attempt to accomplish this difficult and sometimes impossible feat, the majority are now applying more scientific methods to their manufacture of cast iron.

Though the eye cannot tell surely from the fracture the composition or quality of the iron which is used in making up the charges, chemical analysis does definitely give this information. Therefore, every car of pig iron purchased by this firm is sampled and analyzed, the composition of all other materials used in its mixtures is determined, and, irrespective of fracture, which may or may not tell the truth regarding their composition, the raw materials are charged with respect only to their actual content of the metalloids. The resulting molten iron each day is analyzed to confirm the correctness of the mixture and to furnish analysis of the “sprues” which next day are to be used as a part of the day’s charge. Physical test bars, too, are cast each hour or so, and the tensile, transverse strengths, hardness, shrinkage, etc. are accurately determined in testing machines and recorded. In this way absolutely nothing is left to chance or to guess work, and, as you may surmise, any slight deviation from the composition desired is shown at once and the mixture immediately changed to the extent necessary to bring the iron back to normal. It is surprising within what narrow limits of variation compositions and physical properties can be held, with furnace operations continually under such surveillance.

Weighing Out Portions for Analysis
The finely divided mixed drillings are shaken from a thin-bladed spatula on to the balance pan. Drillings are added or taken off until the long needle attached to the beam of the balance swings over an equal number of divisions on each side of the center mark of the white scale in the middle. Accuracy is 1/453,000 of an avoirdupois pound; this is approximately the weight of the lead of a “pencil mark” one inch long.

As the basis for its cast iron, many thousands of tons of pig iron are each year used direct from the blast furnaces. The raw materials come in railroad cars or by boat. The inspector who represents the metallurgical department enters each car and inspects the materials, taking from each a representative sample for analysis. In the case of pig iron this will be from four to eight half pigs, it having been found by experience that these represent very well the contents of the car. So each car of material is held without unloading until it has been determined by inspection and analysis that it is fully up to the specifications upon which the iron was purchased.

Arriving at the laboratory, the half pigs from each car are drilled, equal amounts of the drillings being taken and mixed in an envelope which bears the name of the brand of iron, the number of the car, the date, etc. The sample pigs from each car are treated in this same way, each car being treated individually.

The envelopes containing the drillings then go to the chemists. Frequently samples from fifteen or twenty cars of pig iron, with as many other samples of various derivation, are being analyzed at the same time for the four or six different constituents which it is necessary for the metallurgists to know and control in order that a highly satisfactory product may result. Though a hundred different determinations may be in progress at the same time, spelling “chaos” in the mind of one not entirely familiar with the details of the work, it will be interesting to single out and explain briefly how the samples are analyzed.

Closer View of the Weighing

In such analytical work everything is based upon weight; i.e., constituents are determined and reported in percentages by weight. In chemical laboratory work everywhere the metric system is used, the cumbersome English system of weights and measures being practically impossible. Thus, the metric system is the international scientific standard. The unit taken is the gram, which is equivalent to ¹⁄₄₅₃ part of an avoirdupois pound. One gram of pig iron drillings is such an amount as could be held on an ordinary ten-cent piece.

Working with such small amounts of the sample, exactness and skill are extremely necessary. The balances used are necessarily very delicate—just as delicate as were the scales upon which the jeweler weighed your diamonds—you remember, of course. On these balances we can weigh an inch-long mark made by an ordinary lead pencil.

Dissolving in Acids
This is done under a hood that the irritating fumes given off may be kept from the room.

As the results of the analysis have to be known inside of three or four hours that the cars may be quickly unloaded in order to avoid demurrage, which is the penalty for holding cars longer than the allowable time, separate portions of each sample are weighed out for determination of the silicon, manganese, sulphur, phosphorus, graphitic carbon, and combined carbon. These are necessary in order to determine that the iron is up to the quality specified in the purchase contract and also to provide for its most efficient use in the manufacture of iron castings.

The exactly weighed portions are put into clean, numbered beakers, which are small pieces of high grade glassware that will stand sudden changes of heat and cold. Some of these portions are dissolved in nitric acid, some in hydrochloric acid, others in combinations of acids. In each case the drillings go into solution in the acids, and after various treatments of boiling, evaporating, filtering, etc., well known to those of the chemical profession, the desired results are obtained. In some cases it is by actually weighing a constituent which has been filtered out and burned to ash of a constant known composition, in others it is by comparison of color with standards of known composition, and sometimes it is by other means.

Filtering Silicons
After evaporating the excess acid, baking dry, cooling, and redissolving in weaker acid, the silicon compound formed may be filtered out. The iron and other soluble constituents, now in solution, pass through the filter, which is of pure, porous, unglazed paper.

In all of this analytical work the chemist must take care to lose not one drop of the solution or one grain of the ash from the burned “precipitate,” as the “filtered out” constituent is called.

The pig iron is always bought upon guarantee that it will contain a certain percentage of silicon—the element which in cast iron is known as a “softener.” But this is not the only thing necessary in the iron that is purchased. It must also show proper specified quantities of manganese, phosphorus and carbon, which also are very desirable elements in iron castings, and as little of that undesirable element, sulphur, as possible. Therefore they pay in proportion to the content of silicon, manganese, phosphorus, and carbon—and penalize the seller for sulphur.

“Burning off” the Silicons
The paper and contents, in a little crucible, are placed in a red-hot muffle furnace. The paper is such pure cellulose that it leaves no weighable ash. That which remains after burning is silicon oxide, which is a perfectly white, fine sand. This is very carefully weighed. (Ordinary sand is silicon oxide usually slightly colored with iron.)

The laboratory holds copies of the contracts upon which these materials were bought. If, upon comparison, the analysis obtained complies with the terms of the contract, an O. K. unloading slip is made out and the receiving department is given directions into what raw-material bin in the receiver building it shall be unloaded. If not fully up to the standard called for in the contract, the purchasing department is notified and the car is either rejected or accepted upon some proper terms of adjustment if it can be used without detriment to the product in which it is to be utilized.

Cars of coke, limestone, fluorspar, etc., are inspected, analyzed and treated in the same way, so that nothing is left to guess work. The compositions as determined by the laboratory serve not only as the basis for acceptance or rejection, but the analyses of accepted materials are forwarded at once to the metallurgists, who from them figure the mixtures to be used in the cupolas.

Having great stocks of analyzed raw materials in the labeled bins in the receiver building, the metallurgists who supervise the mixing and melting of the iron determine by mathematical calculation just what irons and how much of each must be taken to give molten iron of the best composition and properties for the castings.

Titrating the Sulphur
Sulphur is evolved from the drillings as a gas (hydrogen sulphide). This is absorbed in a solution of chloride of zinc. The amount of sulphur is measured by slowly running in from a burette a solution of iodine of very accurately known strength. The iodine unites with the sulphur compound as long as any of the latter remains, but the first drop added thereafter turns blue the whole solution because of the reaction of the excess of iodine with starch paste that has been added previously as an indicator. Accuracy is about .005 per cent of sulphur.

The total iron materials charged must have a definite amount of silicon, of manganese, of phosphorus and of carbon. For a 4,000–pound charge for soft cast iron, for instance, the total silicon in the materials which make up the charge must be somewhere near 118 pounds, the manganese and the phosphorus about 30 pounds each. The usual losses of these materials through oxidation are known, of course, and sufficient excess has been allowed that the desired final composition will result.

Titrating Phosphorus
A yellow precipitate containing the phosphorus is filtered out on filter paper. It is redissolved in alkali and titrated with a standard solution of nitric acid, similarly to the sulphur. The solution in the flask turns pink with the first drop added after the phosphorus has been measured.

On several scales which are regularly inspected and kept carefully adjusted, the weighers weigh out the prescribed quantities of the raw materials. “Buggies” holding nearly one ton each are loaded in turn with coke, with proper amounts of pig iron, cast iron scrap, sprues from the foundry castings of the preceding day, and a proper weight of limestone flux. Each charge of two tons of iron requires four buggies for its transportation from the raw-material bins and scales to the cupolas.

The old-time way was for laborers to dump the charges into the cupola and spread the materials by hand, but in modern foundries better ways have been provided. Here a charging machine operated by compressed air lifts into the furnace, one by one, the buggies of coke, and of the other materials, whence, after the dumping of their contents, the buggies are returned to the receiver building to be filled again.

Thus many charges per hour pass through the yawning charging doors of the cupolas, being dumped in fast enough to maintain the level approximately even with the bottom of the charging door.

Reading the Carbons
The higher the combined carbon the darker the nitric acid solution of the iron or steel. The solution is diluted with water or weak acid until the color matches that of a “known” sample or standard. Accurately graduated comparison tubes are used.

In starting, a wood fire has been made on the sand bottom of the cupola. This is covered with coke in such a way and in such amount that, when ready for charging of the metal, a column of glowing coke extends to a distance of one foot or two above the tuyères. Upon this “bed” in alternating layers are piled the weighed amounts of pig iron, sprue, scrap and limestone as described above. Following each charge is a layer of coke sufficient in amount to replace the “bed” coke which is burned away in melting the iron charge, thus maintaining the top of the bed of coke throughout the day at approximately the same height.

Weighing the Graphitic Carbon
The graphite is filtered out on an asbestos pad in a perforated platinum crucible. After drying until all moisture is gone it is weighed, ignited, and weighed again. The loss of weight equals the weight of the graphite of the sample.

Ever since 7 o’clock A.M., when the twelve to sixteen ounces of blast pressure was put on, the charges have been descending gradually from the charging door. Encountering the intense heat in the “melting zone” at the top of the bed of coke a little above the tuyères, the iron melts and trickles down through the three to five-foot bed of glowing coke on to the sand cupola bottom or hearth where it accumulates. The tapper, with his iron bar and “bod stick” with its little ball of moist fire clay, alternately opens and plugs the tap hole at the bottom of the furnace as occasion requires, but throughout the day of ten or more hours there is almost constantly a full stream of iron flowing from the spout. The big “bull ladle” which receives it, in turn gives it up to smaller or “shank” ladles, in which it is conveyed along trolleys to still smaller ladles from which it is poured into the sand molds to form the castings.

Another Close-up View

As in the blast furnace, limestone is added as flux to make liquid and dispose of sand, dirt, scale, etc., which are detrimental. The liquid slag formed from union of limestone with these impurities floats upon the molten iron in the cupola hearth, as it is less than half as heavy as the iron itself. It flows almost continuously from a higher hole called the “slag hole,” in the rear of the furnace and just beneath the tuyères. The slag has little value except as material for filling purposes, etc. So-called “slag wool” can be made by blowing air through it. Sometimes the blast from the cupola blows it in such a way that this pure white “wool” is formed and blows out of the slag hole of the cupola. About Christmas time some of the workmen take quantities of it home for decoration and for fireproof whiskers for “Santa Claus.”

These operations go on continuously throughout the day, each cupola making the particular grade of cast iron or “semi-steel” which is best adapted to the particular castings to be poured, size, shape and purpose of

Determining Carbon by Direct Combustion
In an electric furnace with pure oxygen passing over them, the drillings burn as would splinters of wood. From the gas given off the total carbon is determined with great accuracy.

Weighing Pig Iron for the Cupola
From three to six varieties of pig iron are used in each charge.

The Receiver Building
Here pig irons of different compositions are separately kept in numbered and labeled bins. The magnet which is used for unloading and handling the iron may be seen. The grab bucket for sand is at the right.

Weighing the Charge of Coke
Close weighing is necessary, for a variation of ten or fifteen pounds may affect the running of the cupola.

the castings being the three main determining factors.

Sectional View of a Cupola

When nearing the end of the day, charging ceases, the charging door is closed and the last charge, gradually descending, melts and flows into the bull ladle about an hour later. As soon as the bull ladle is emptied it is run out of the way, the cupola is drained of all iron and considerable slag through the tap hole, and the bottom doors of the cupola are dropped by pulling from under them the “props” which have held them in position. The great mass of bed coke follows with a great burst of heat, light and flame. This is quickly subdued with a stream of water and removed. When cool, the slag and accumulations are chipped from the cupola lining, burned areas are patched with bricks and stiff, fire-resisting mud, the bottom doors are raised and fastened in position, and the eight-inch sand bottom is packed in ready for the next day’s run. After building a fire and getting a good “bed” of glowing coke, the cupola is ready again for the charging of iron.

A Cupola in Operation
The stream of iron from the cupola spout flows into a “bull ladle” and from that into “shank-ladles.” The bull ladle serves as a reservoir and mixer.

Castings of Cast Iron as They Come from the Molds
The “sprues” have not been removed.

CHAPTER XI
CAST IRON (Continued)

Unlike the modern Bessemer, open-hearth, and other steel products which are reworked, i.e., rolled, forged, etc., cast iron is a comparatively old alloy dating back over several centuries. It cannot be rolled, forged, or otherwise reshaped, so its final form must be given to it at once by pouring or “casting” the molten metal into a mold. Its castings serve exceedingly well in the hundreds of places for which they are adapted. They are comparatively cheap, can be readily duplicated in small or large quantities, and those from the softer grades of cast iron may be machined easily.

These cast iron alloys have only from one-third to one-half the strength of steel or wrought iron and are, comparatively speaking, very brittle. Where resistance to severe shock must be withstood they should not be used. Also, some varieties have a “habit” of growing larger upon repeated heating and cooling. This “permanent growth” is particularly noticeable when the alternate heating and cooling is at red heat or over. Pieces of cast iron have been made to gain 15 per cent in linear dimensions, and it is quite common knowledge among machinists that a piece of cast iron which is slightly too small can be permanently expanded by heat.

Nevertheless, the cast irons have large and legitimate fields in which they are very serviceable. From most of their important present uses they are not likely soon to be displaced.

Because of Their Large Size Molds for Very Large Castings Have to Be Made on the Floor of the Foundry, or Partly in Pits

Cast irons in considerable variety of compositions and physical properties are available, as was indicated by alloys Nos. 14 to 19 which were given in the table on page [83], part of which is here reproduced. In alloys 3 to 13 the carbon exerts the great influence on the physical properties, and this is true also of the cast irons. But all of the latter have a total carbon content of more than 2½ per cent, and, under certain conditions, some of the carbon assumes a different form from that which we encountered in the steels. This modified form is “graphite,” well known to us as a flaky, black, greasy-feeling material, which is soft and very fragile. Graphite in the iron alloy naturally weakens it and, as it is itself such a good lubricant, it makes cast iron machine easily if sufficient amount is present.

A Few of the Cast Irons
	Silicon, Per Cent	Graphite, Per Cent	Combined Carbon, Per Cent	Total Carbon, Per Cent
No. 14. White Cast Iron	.70	.10	2.65	2.75	Very Hard
No. 15. Annealed Malleable Iron	.70	2.70	.05	2.75	Machinable
No. 16. Cast Iron for Chilled Castings	1.00	1.00	2.00	3.00	Very Hard
No. 17. Semi-steel	1.75	2.80	.40	3.20	Machinable
No. 18. Gray Cast Iron	2.00	3.10	.30	3.40	Machinable
No. 19. Soft Gray Cast Iron	2.50	3.30	.15	3.45	Machinable

Now, the above must be understood as being typical compositions only. There are, of course, irons of all intermediate compositions, also, and while the total, graphitic and combined carbons, typically, are about as indicated, there may be wide variation.

To illustrate what a variety of chemical and physical properties may be produced, let us assume that the total carbon in a certain cast iron is 3.25 per cent. If this carbon is all in the chemically combined form (i. e., combined with the iron to form the very hard compound which is known to the metallographist as “cementite”) the fracture will be white and the alloy extremely hard. If none of this carbon is combined, but all is in the form of graphite flakes throughout the alloy, the fracture will be “gray” and the alloy soft and machinable. It is possible to produce either of these two conditions or practically any intermediate stage; i.e., we can almost at will split up the 3.25 per cent of carbon into varying percentages of graphitic and combined carbon—the total always equaling 3.25 per cent.

No. 30. Very Soft Cast Iron. Note Large Graphite Flakes

No. 31. Medium Hard Cast Iron
The “combined carbon” is in the roundish, dark parts. It is the “combined carbon” that increases the strength of cast iron and steel.

The “precipitation” of graphite which is necessary for softness is brought about mainly through the influence of silicon, which we before termed the “softener.” Other conditions being equal, the higher the silicon (if not above 4 per cent), the higher will be the graphite and the lower the combined carbon; and vice versa, the lower the silicon the lower will be the graphite and the higher the combined carbon. It is mainly due to the “combined carbon” which is left after precipitation of the graphite that the alloy has greater strength, hardness, and closer grain. So, just as the steels are stronger and harder as the carbon increases (in steel all the carbon is combined), so, other conditions being equal, the strengths and hardnesses of the cast irons, within usual limits, increase as the combined carbon increases.

No. 92d. Semi-Steel. A Closer Grained and Yet Stronger Cast Iron

No. 33e. Mottled Cast Iron
So-called because it is a mixture of white and gray iron.

Just here it is interesting to remember that from the standpoint of metallography cast irons are simply steels in which there is what we might call an impurity or an adulterant, graphite crystals. It will be seen at once that could these graphite crystals be removed from the cast irons shown in photomicrographs No. 74, No. 92d, No. 30 and No. 31, we would have alloys quite similar in appearance to the steels shown in photomicrographs No. 3b and No. 22c which appeared on pages [77] and [78].[^[7]]

[7]. Magnification 70 diameters.

No. 7. White Cast Iron

So the softer cast irons which are used for valves and fittings, machine parts, radiators, hollowware, etc., have high silicon. Parts that do not have to be machined can be of “harder” iron; i. e., made of iron having lower silicon content.

Manipulation of the silicon content is not the only method by which the hardness of cast iron can be influenced. Graphite can “precipitate” (i.e., separate throughout the casting) only if sufficient time is given it to do so. That is, the cooling of the casting after pouring must be sufficiently slow. In a sand mold the iron remains molten for a time, and after solidification it cools slowly enough that the greater portion of the carbon separates as graphite. Therefore, castings of proper composition made in sand molds are soft and machinable.

If the iron is poured into a mold the surfaces of which are made of iron, the molten metal upon entering becomes solid almost as soon as it takes the form of the mold, and it cools with great rapidity. Under such conditions the carbon of the alloy is denied the time necessary to change into the graphitic form and the casting has a white fracture and is so hard that it cannot be machined.

Section of Chilled Car Wheel
Showing white iron rim.

There are many purposes for which the alloy should have extreme hardness and the great resistance to wear which accompanies such hardness. The wearing faces of gears, brake shoes, rolls, and car wheels, for instance, must be hard. For such products, white cast iron, the extremely hard condition of the alloy, just referred to, is utilized. Such castings are usually produced with a white cast iron face, but with a gray iron interior, gray iron being less brittle and less likely to break under shock or strain. A car wheel, for instance, has approximately an inch in depth of white iron on the surface which lies next to the rail on which it runs.

Chilled Cast Iron
The white edge resulted from the more rapid cooling against an iron chill, as did the white rim of the wheel shown above.

Such are known as “chilled castings.” Molds for them are usually made of sand, with pieces of iron (called “chills”) imbedded where white iron is to be produced. The molten iron next to the sand surfaces cools in the usual way and is gray and soft, while that which lies next to the “chill” is white and extremely hard. The “depth” of the chilled layer can be increased or diminished according to the thickness of the iron “chill” used, its temperature, and by the composition and temperature of the molten cast iron with which the mold is poured.

The sulphur and total carbon of the molten cast iron also have considerable influence on the depth of “chill.”

Two-Part Molding Flasks

There is a cast iron alloy which is familiarly known as “semi-steel.” It is simply a high grade and stronger “gray” iron and must be classed as a cast iron, as our table on page [180] shows. While it could undoubtedly be made from materials which are commonly used for cast iron, it is practically always produced by charging with these a certain amount of steel scrap to bring about the lower silicon, phosphorus and total carbon desired.

A Hollow Cylinder
The casting which we are about to make.

Because steel has a higher tensile strength than has cast iron, many have inferred that it was the steel addition which made semi-steel stronger than the ordinary soft cast iron alloy. The rather unfortunate name, “semi-steel,” apparently was given because of the steel used and the intermediate strength which the resulting product possessed.

However, during the melting down of the charge the steel scrap becomes molten and its constituents merge with those of the other iron materials charged. We get out of the cupola, then, a mixture which, disregarding the losses and gains due to the air of the blast, the fuel, etc., is an average of the materials charged. We, therefore, no longer have any steel, but a cast iron which has a somewhat lower silicon, phosphorus and carbon than the softer cast irons. The greater strength of the alloy is due to its composition and only indirectly to the fact that steel was used in its production. The physical properties of the steel charged have been entirely obliterated in the melting process.

Split Pattern of Wood, Surface-Coated with Shellac Varnish

This view that semi-steel only indirectly gets its increase in strength from the steel charged is confirmed by its structural appearance under the microscope, as was shown in numbers 74 and 92d which were given on page [79], and the photomicrographs given here, and by its extreme brittleness under hammer blows. Under such shock it is but little more resistant than cast iron.

Core That Makes Hole in Casting

This weakness under “shock” was shown by tests from which the table which follows was compiled. Bars one inch square and thirteen inches long laid on supports exactly twelve inches apart, were struck at the center by a twenty-five pound weight. It took seven blows to break the cast iron bar, the semi-steel bar required eleven, while cast steel withstood ninety-two blows. Even this does not adequately express the great resistance of the cast steel (another alloy not yet discussed), for the height of the “drop” was being increased one inch with every blow, and the cast steel bar, on account of its bending, had to be regularly turned. The total foot-pounds exerted by the blows are given in the table which follows:

Drag, or Bottom Half of Mold, after Pattern Is Withdrawn

Drag with Core in Place and Cope, or Top Half of Mold Ready to Close

Transparent Mold, Showing Relative Positions of Core, Casting, Sprue, Etc.

Alloy	Tensile Strength	Number of Blows	Total Foot Pounds
Cast Iron	23,400	7	102
Semi-Steel	35,050	11	206
Malleable Cast Iron[[8]]	37,140	22	1,580
Cast Steel[[8]]	72,120	92	10,112

[8]. On account of bending, the malleable iron and the steel bars had to be turned several times.

Semi-steel is a very close-grained alloy of ten or twelve thousand pounds per square inch greater strength than cast iron. It is a most satisfactory material for medium and larger sized castings for which cast iron formerly was used.

Cast Iron Tee, Cock Plugs, and Return Bends, with Sprues and Gates Attached

Did we say that cast iron was very brittle? So it is, comparatively speaking. But just as the chemist will tell you that there are no substances which are absolutely insoluble, just so does cast iron appear to be extremely brittle only when compared with the iron alloys of considerably less brittleness.

The three pictures of the hooped and twisted casting illustrate how unwise it might be to speak with absoluteness. A few years ago the casting illustrated on pages [190] and [192] was brought to this country by a visitor to Europe, with an expression of regret that cast iron produced in this country did not have such qualities of elasticity as had cast iron made abroad. Whereupon, without any change whatever in his iron mixture or cupola practice, the superintendent of a well-known foundry made castings which were exact duplicates of that submitted. The three photographs shown further on (Figs. A, B, C) were of one of these castings. The ability to bend without breaking is, of course, largely due to the shape.

Plugs and Wheels as They Come from the Mold
Occasionally as many as 200 castings can be made in one mold.

Type of Molding Machine

As a matter of fact, such castings were not at all new in this country, having been furnished by American foundries for electrical work for many years. Cast iron springs, piston rings, and many other articles of cast iron are regularly made, which show such elastic quality.

We have said considerable about “castings.” In general we know what castings are, but in the minds of some there may be a little uncertainty as to the manner in which they are produced.

Another of the Many Kinds of Molding Machines

There are few lines of human endeavor which require greater judgment and skill than does the making of molds for castings. Sound judgment based on long experience, knowledge of conditions under which the work immediately in hand must be done, observation, and accurate, deductive reasoning as to the causes of failure are absolutely necessary for success.

Casting, Which Because of Its Length and Small Cross Sections, Requires Very Fluid Cast Iron. (Fig. A)

In general, molding may be said to be done in “pit” or on the “floor” for large work, on the “bench” for smaller work, or by “machine.” Pit, floor, and bench molding are applicable for production of castings of all sizes and descriptions and this general type which we might term “hand molding,” is the form that has been practiced longest. Molding machines are more or less recent inventions which have enabled certain standard shapes and sizes of castings which are in sufficient demand to be produced in great numbers by unskilled workmen and therefore at less cost than is possible by the older hand method.

Each design for casting may be said to demand individual treatment, and the molder must select that method out of the many which alone, perhaps, can be successful. The subject is such a broad one that little will be here attempted further than to give by description and illustration the predominant points of the making of molds and castings. A simple, typical case of bench molding will be taken, that the relation of pattern, mold, core and casting may be clear.

The molding sands used are usually natural sands which contain greater or lesser amounts of clay, which, when moist, acts as a “binder” of the grains of sand. When used without drying, the mold is said to be a “green sand” mold; if dried, a “dry” or “baked” mold, as the case may be. The majority are “green sand” molds.

For the usual casting of which only a few or several duplicates are wanted, the “split” pattern is generally the most convenient.

The two halves of the mold, the “cope” (top) and the “drag” (bottom), are separately made in the two parts of the “flask” or molding box by “ramming” properly selected and “tempered” (moistened, mixed, and sieved) sand over the halves of the pattern. Of these, the drag is made first over the lower half of the separable pattern placed flat side down upon a bottom board. After “ramming,” i.e., packing the sand, just hard enough but not too hard, this half mold is reversed and the top half of the pattern placed upon the lower half, now at the upper face of the drag and flat side up. A little “parting powder” or fine, dry sand is sprinkled over the fresh surface of the half mold so that the upper half, next to be made, will not stick to the lower half, but can be lifted off at the proper time.

It May Be Bent Double Readily Without Breaking. (Fig. B)

The cope half of the two-part “flask” is now put on, filled and rammed with sand as was the drag. Any extra sand is scraped off with a straight edge and at the proper place a hole is cut with the “sprue cutter” straight down through the cope to the “parting.” More commonly, perhaps, this “sprue” hole is made by withdrawing a “sprue” stick (of wood) about which sand had been packed during the making of the cope. It is through this hole that the molten metal will be poured into the mold.

Because of Its Ability to Withstand Bending and 180–Degree Twists It Is Often Jocularly Referred to as the “Rubber Casting.” (Fig. C)

Lifting the cope or top half, it is turned upside down, and, after cutting in the drag the “runner” or “gate” connecting the “sprue” hole with the casting, the halves of the pattern are carefully drawn that the sand may not be disturbed. Now in the cavity left in the drag, to make the hole in the casting, is hung the baked “core” of sand, held together by flour or rosin or a “drying” oil. The cope is carefully replaced upon the drag, thus “closing” the mold.

As will be noted from the drawings, there is left between the core and the mold a space all around, which will be filled by the metal of the casting when poured. Therefore the surface of the core shapes the inside, and the mold itself the outside and ends of the casting.

The molten metal, entering through the vertical “sprue” hole, flows along the “runner” and into the mold through the more or less constricted entrance called the “gate.” The gases formed during pouring and the air with which the mold was filled are driven out through the porous bodies of sand of the mold and core. Had the mold been rammed too hard the gases could not escape through the sand and an imperfect casting would result.

The poured mold is allowed to stand until the metal has solidified and cooled sufficiently, when the casting is “shaken out.” The sand is returned to the molder to be used again. The sprue and runner are broken from the casting, which, after cleaning by “tumbling” with others in a revolving mill, or otherwise, goes to the machining and assembling shops.

Some form of the above general method is everywhere used for the production of all kinds of castings, except for those which can be made by machine at a lower cost for molds.

This kind of molding, which we have termed “hand work,” requires expert molders and is too slow and expensive for the hundreds of standard shapes and sizes of castings which are in great and constant demand. The latter are made on cleverly devised molding machines working with compressed air or by hand power applied through a lever. The pattern is attached to the machine, set and very accurately adjusted by a skilled mechanic. Thereafter the sand is rammed, the runner formed and the pattern drawn by the machine itself, all of these very critical movements being therefore rapidly and unerringly duplicatable any desired number of times by unskilled labor, which has but to put on the parts of the “flasks,” feed in the sand, set the cores, close and remove the mold, and begin the next.

Sometimes there is but one, but for the smaller sizes there are often ten or twenty, and, occasionally, as many as two hundred pieces or castings in a single mold.

CHAPTER XII
MALLEABLE CAST IRON

It almost goes without saying that the capacity to withstand distortion without breaking was the meaning of and the reason for the use of the term “malleable.” But wrought iron is malleable as also is mild steel, and, in Europe fifty years ago (though in general not now) by the term “malleable iron” was meant and understood what we know as wrought iron. You will remember that Bessemer’s paper announcing his great process was entitled “The Manufacture of Malleable Iron and Steel without Fuel.” The first reference was to wrought iron. Bessemer did not succeed in making this by his process but his success in the manufacture of steel was immense. Therefore, while in ordinary conversation such definiteness is not necessary, perhaps, and not usual here, to be safe one should say “malleable cast iron,” and not simply “malleable iron,” for by the latter, many Europeans still understand wrought iron.

Like “Topsy” of “Uncle Tom’s Cabin” fame, the various members of the iron family “just growed.” Therefore a strictly logical classification and nomenclature is hardly to be expected.

It was mentioned in a former article that a process for making malleable the brittle white cast iron was discovered, or at least described, by Reaumur, a Frenchman, about 1722. It is likely that his discovery or acquaintance with it came about through his extended experiments with cementation steel.

Malleable Cast Iron Bars

Malleable Iron Castings

The publicity which Reaumur voluntarily gave to his researches forms a notable exception to the customs of those days when it was the usual thing for manufacturers jealously to guard all trade secrets. These were handed down from father to son or to others of close interest in the business. So aside from Reaumur’s announcements concerning malleable iron, few details of its manufacture came to light during the eighteenth century. Even during the hundred years which have just passed there have been few lines in which greater secrecy has been maintained both in Europe and America. During the last thirty years, only, has real scientific work been done to make known the reactions which occur during annealing and the real causes of the malleability.

The father of malleable iron in this country was Seth Boyden, of Newark, New Jersey, a very ingenious man who well deserves the monument erected in his honor by the citizens of the city, which is pardonably proud of him.

Test Sprues, Showing White, Slightly Mottled, Medium Mottled, and Gray Fractures

Boyden apparently had no knowledge of the existence in Europe of the malleableizing process, but after noticing that a piece of formerly brittle cast iron had become rather malleable, apparently through the action of heat, he set about making experiments to produce a malleable material which could be produced more cheaply than wrought iron. By melting in a forge pieces of pig iron and then annealing in a small furnace in his kitchen fireplace the bars which he cast from the melt, he had worked out by 1826 a process that produced cast iron which was malleable. In 1831 he started a foundry and made a thousand or more different articles for which there was demand, and, from this beginning, an immense industry developed in this country.

Test Bars with One Edge Cast against a “Chill”
The composition of the mixture is regulated according to the depth of the chill as well as by appearance of test sprues.

We must not forget that the malleable cast iron as produced in this country is an entirely separate and distinct thing from the European malleable iron, as will be shown later. So our immense industry is our own and not a copied one.

It is only certain members of the cast iron family that can be made malleable by proper heat treatment. Alloys No. 14 and No. 15 represent one of these alloys before and after the annealing process. While No. 14 was given as a typical analysis for white cast iron for malleableizing it must be understood that compositions can vary considerably without detriment from that given.

There is one thing, however, which is absolutely necessary and that is that all or practically all of the carbon of the alloy must be in the combined form previous to the annealing process. This means that the alloy shall be white cast iron and have no free graphite, for any graphite flakes will remain through and after the annealing process and weaken the alloy just as it weakens gray cast iron.

Sketch of a Coal-Fired Air Furnace

For producing this white cast iron two processes are in general use—the “cupola” and the “air furnace.” The latter predominates.

Operation of the cupola for malleable iron requires great skill and very close attention to detail, for, to malleableize easily and with the best results, the composition of the alloy must be regulated within narrow limits, very much narrower than for gray cast iron. However, this is entirely possible and cupolas are operated continuously for malleable cast iron for ten or more hours with very slight fluctuation.

In general, operation is very similar to that described for cast iron except that the composition of the charge is necessarily different, much lower silicon being required, and more coke has to be used for the melting.

Most malleable iron castings are made in sand molds, and, as stated, the iron poured must be of such composition and temperature that the castings so made will be white of fracture. It is possible to get a quick indication of the condition of the iron for pouring by making test pieces, every one in the same way, which, after cooling and breaking, will show by fracture the approximate composition of the metal. According to these test pieces, called “sprues,” which, at times, may be cast as often as every five or ten minutes, the mixture is regulated to produce a uniform product.

Firing an Air Furnace

To illustrate: The fracture of a round sprue, or test piece, always ⅞ inch in diameter, when poured in the sand, cooled there to low red heat, quenched in water and then broken, should be white with only a few flecks of dark constituent. A gray iron fracture indicates too high silicon content and such iron is usually termed “low” iron. Castings of medium or heavy section, which, therefore, cool slowly in the sand, if poured of too high silicon, i.e., “low” iron, might precipitate a little graphite during cooling, even though thinner-sectioned castings which cool so much more rapidly would come white from the same iron.

While iron giving nearly white sprues is necessary for particularly large castings, to make sure that the usual run of malleable castings will come white in the sand requires very slightly mottled test sprues.

Test blocks also, with one side cast against an iron “chill” are poured to determine the depth of chilling, and test bars of various shapes are regularly made, to test after annealing, for tensile strength, torsion and other physical properties.

“Air furnaces” are much like longer puddling furnaces. They vary in capacity from ten to forty-five tons while occasionally small ones of as little as three or five tons capacity are met with.

Taking Off the Slag

The usual fuel is soft coal. The long flame passes from the grate at one end over the bridge wall and is deflected by the roof down upon the bath beneath. A chimney at the outgoing end furnishes draft. The furnace bed is usually of brick upon which is fritted (slightly fused) a mixture of sand with a little lime. In order to facilitate charging of the materials to be melted the roof is usually removable in parts, called “bungs.” These have frame work of iron which hold in place the fire bricks that come in contact with the flame. During charging these bungs are lifted off one at a time, and the iron materials are dumped through the openings. Small doors in the sides just above the bath allow “rabbling” or mixing of the charge and skimming of the slag which forms, and one or more spouts lined with fire bricks and clay provide for tapping out the metal when it is ready to pour.

Tapping

Unlike puddling furnace and open-hearth no burning out of the silicon, manganese and carbon is desired, though, of course, some occurs and has to be allowed for in calculating the mixture. The intention is simply to melt together with the least possible loss a mixture of such materials as will give the average final composition which long experience has shown to give the proper qualities to the finished product.

Charges usually are of certain percentages of pig iron with not too much phosphorus, sprues from previous melts, more or less good malleable iron scrap and small amounts of steel scrap. These are melted down as quickly as possible. Occasionally the slag which accumulates is skimmed off and, after rabbling, test plugs are poured from the fractures of which the composition of the iron is judged.

When the silicon content is deemed proper or has been adjusted through longer action of the flame if too high or addition of more silicon in the form of a high silicon alloy if too low, the iron is tapped, provided it is hot enough.

Malleable iron is largely used for very small castings. These require very hot and fluid metal. So, even if it is of proper composition, the metal must be held in the furnace until it is of a high enough temperature to pour properly. Through prolonged and strong heating the iron may easily become oxidized or “burnt” and much skill is necessary for proper operation of the furnace.

After tapping, the iron must be got into the molds with the least possible delay.

As has been mentioned with former processes the melting of iron in contact with coke or coal results in more or less contamination with sulphur. For this reason cupola malleable has considerably higher sulphur than has malleable cast iron made in the air furnace. Cupola and air furnace each has certain advantages and certain disadvantages. While strength and elongation are somewhat greater in air furnace than in cupola malleable, both anneal well and give materials which are satisfactory for the purposes for which they are intended.

Cupola metal has an advantage in that the temperature and the composition can be closely maintained the same throughout the heat, perhaps more so than with the air furnace. With the latter the metal at the top of the bath is hotter than that underneath, and, through action of the flame and air, silicon is somewhat lowered before all of the heat can be poured, especially with air furnaces of large size. The metal can easily be “burnt” unless extreme care is taken. In the cupola we can get very hot iron continuously so that it is unnecessary to prolong the heat with the danger of burning that occurs with the air furnace.

Air furnace iron anneals rather more readily than does the product of the cupola, and the strength and malleability are usually greater. The former requires a temperature of about 1350° F., while the latter must have 1500° F., a difference of about 150° F. Whether this results alone from the somewhat higher sulphur of cupola malleable is not definitely known, but it is probable that, also, the slightly higher total carbon gives the iron-carbon chemical compound a tendency to persist more strongly.

The open-hearth furnace is sometimes used for making malleable cast iron. It melts much more quickly than does the air furnace which requires from three and one-half to nine hours per heat, depending upon the size. The quality of the product which the open-hearth furnace produces is of the best, but on account of the continuous operation necessary, this type of furnace is not largely used. Malleable iron has also been made in the Bessemer converter, and, occasionally, in the crucible furnace, but in this country the practice is not at all common. In Germany a great deal of malleable iron is made in the crucible furnace.

When dumped from the molds the castings are extremely brittle. The sprues are knocked off and the castings go to the “tumbling” mills where they are tumbled, either with the sprues, with hard iron (white iron) shot or star-shaped pieces of iron which quickly clean the sand from them and give smooth, clean surfaces.

At the chipping and sorting benches any remaining pieces of gates and other excrescences are removed while the castings are being handpicked and sorted. White iron, because of its brittleness, breaks easily and small protruding parts can more readily and cheaply be removed before annealing than after the castings have been thereby toughened.

Annealing Furnaces, Showing Sets of Pots or “Saggers” in Which the Castings Are Annealed

Having the cleaned castings of white iron of proper composition, malleability is given to them by the heat treatment known as “annealing.” Through the influence of a cherry-red heat continued over a sufficiently long period the iron and carbon, which in the white iron are chemically combined, gradually become divorced, and, after complete annealing, the casting will be found to consist of free iron in which are imbedded throughout very small particles of coke-like carbon. Castings that before this heat treatment were so brittle that they broke into many pieces under a blow and so hard that they might scratch glass, are now found to be capable of withstanding considerable distortion without fracture and so softened that a needle may scratch them.

Not only is a proper temperature necessary for best annealing, but, as stated, a sufficient time must be allowed for the separation of the carbon and iron. The separation requires many hours and the cooling from the annealing temperature must be slow in order that the carbon and iron may not again unite, as they certainly would do were the castings chilled in water or otherwise cooled too fast.

Most manufacturers produce tonnage enough that castings have to be annealed in quantity. As iron at red or higher heat wastes very rapidly on account of scaling (in fact would take fire in air if hot enough), and the quality of castings deteriorates somewhat if even a small amount of air is allowed to come in contact with them during the annealing process, they are generally protected by enclosing in iron containers on which tops are luted and cracks filled with stiff, fire-resisting mud which keeps out the air.

These iron drums are of suitable size and shape that they can be stacked one upon the other to a height of from four to six feet. The stacks or “sets” of “saggers” as they are called, are run into large brick-lined retort chambers which are heated either with coal from a grate, by powdered coal, oil, producer gas or other fuel. The larger the furnace and the greater the tonnage of iron which must be heated, the longer will be the time necessary for bringing the furnace and castings up to the cherry-red heat which is necessary for the annealing. Therefore, the larger furnaces require a somewhat longer time than those of smaller size, though the time required for the annealing of the castings themselves is no longer. In this country with ordinary-sized furnaces the usual time for the annealing operation is approximately one week. This includes the heating of the furnace and castings and the cooling to a black heat again.

No. 109. Photomicrograph of White Cast Iron

Some manufacturers, however, anneal without pots but they aim to have the castings protected from the flame and air.

No. 110. Photomicrograph of No. 109 after Several Hours of Annealing

Handling devices have been designed which facilitate loading and unloading the furnaces. With these the many sets of pots or saggers which the furnace holds can be very quickly charged or removed.

After shaking the castings out of the cooled pots, the dark coating is removed from them by “tumbling” with iron shot, pieces of leather or other polishing material in tumbling mills after which they are ready for any machining which may be necessary.

Photomicrographs Nos. 109, 110, 113 and 35 which show the samples at 75 diameters magnification show the course of the annealing process. No. 109 was taken from an unannealed casting. No. 110 was of the same iron after approximately thirty hours in the furnace. No. 113 shows that after about forty-five hours nearly all of the iron-carbon chemical compound has been broken down into black patches of free carbon, surrounded by the white areas of pure iron. After about sixty hours all of the iron and carbon have been divorced and the annealing operation is complete, as is shown by photomicrograph No. 35.

No. 113. Photomicrograph of No. 109 When Nearly Annealed

No. 35. Completely Annealed Malleable Cast Iron

While heat alone effects the divorcing of the carbon and iron, which is the essential part of the annealing process, in the greater number of cases aid is given by what may be termed chemical means. Reaumur, who about 1722 discovered the annealing process, used iron oxide for the purpose. The white iron was packed in iron ore or mill scale. At the high temperatures employed the oxygen of the ore in some way not yet definitely known, gradually removed the greater amount of the carbon from the casting. It has always been a scientific conundrum how a solid, iron oxide, surrounding another solid, a piece of white iron, could remove from the latter its carbon when neither of them melts nor mingles with the other. Whether some of the oxygen from the ore penetrates the iron and burns out its carbon or whether the carbon of the casting itself migrates is not yet definitely settled. Certain it is that the carbon is gradually removed from the casting, from the surface first and with increasing length of time from greater depths.

No. 390b. Photomicrograph Showing Decarbonized Outer Layer
The photomicrograph also shows that this casting was not fully annealed.

In European practice malleable iron castings are still malleableized in this way, i. e., by burning out the carbon. The castings are made as thin as possible and the annealing in “packing” (iron ore or mill scale) is continued for from one to two weeks. At the expiration of this time the castings have a white, steely fracture which is entirely unlike the fractures of malleable iron castings which are made in this country. Photomicrographs of such malleable iron show few or none of the black spots which No. 35 exhibits, and analyses of castings annealed in this way give very low results for carbon.

Malleable Cast Iron in Which Practically All of the Carbon Has Been Removed by Reaumur Process Annealing

While in this country the Reaumur process of annealing is not followed, a “packing” of ore or scale is generally used. Some use an inert packing such as sand, and as first mentioned, some use no packing at all. Really, one of the main purposes of the “packing” as now used is the prevention of warping of the castings in the pots while annealing. The annealing temperature is not so high as in Europe nor is the annealing continued so long, but when packing is used for the shorter time only, some surface carbon is removed and the carbon throughout the center portions of the castings is precipitated in the coke-like form which is known as “temper carbon” to distinguish it from graphite which is shown in photomicrograph No. 35. To the eye, then, fractures of such castings show black centers and white rims. They are known as “black heart” castings and these form the bulk of the malleable cast iron made in this country.

Fracture of Black Heart Iron
Note the white rims and black coke-like interiors. The majority of American malleable iron is of this “black heart” variety.

We may say, then, that there are in general three varieties of malleable cast iron: the “all black” which is annealed without “packing,” the “black heart,” annealed in “packing” and the most common kind in this country, and in Europe, but very rarely here, the “whiteheart” from which practically all of the carbon has been burned during the “anneal.”

Malleable Cast Iron Swivels of Which Parts No. 2 Are Cast Tightly Around No. 1 and Loosened Only upon Annealing.

Comparison of photomicrographs No. 35 and No. 30 given on page [181], will show at once one of the reasons for the much greater malleability of malleable cast iron. While the total carbon present is very nearly the same in the two irons, the difference in physical form causes great difference in the malleability of the two. In the gray cast iron, No. 30, the carbon is crystalline and in the form of long brittle flakes which cut through and separate the grains of iron. Thus “planes of cleavage” are formed which make the alloy unable to resist severe shock and cause it to be anything but malleable. It is not so with annealed malleable cast iron. Here the carbon is in the form of small pellets which are imbedded among the grains of pure iron, the malleability of which is not seriously impaired largely because of the continuity of the “pure iron” structure. A second reason for the ability of malleable cast iron to withstand shock is that in the burning out of the carbon of the outer portions of the casting very small cavities are left. These allow the surface to become considerably deformed and battered under successive shocks without great strain on the casting itself.

Nothing has been said so far concerning one trait of all of the irons and indeed of most metals and alloys which are used for casting purposes. This is the tendency to “shrink” during the solidification and cooling of the metal of the casting. On account of the freezing of the outer portions of the casting before the metal of the inside, there must result certain hollow places or cavities after the inside metal has cooled unless some channel is kept open through which fluid metal can pass inside to keep cavities from forming. We will not here go into the matter of shrinkage with its great worry to the molder nor the ingenuity and strategy through which he produces castings without shrinkage cavities. One of the methods taken to overcome the trouble will be explained in the chapter on Cast Steel which is to follow.

There is, however, another type of shrinkage—that exhibited by the contraction of the entire piece of metal as it gradually cools after solidification. This presents a rather curious and interesting case.

It is well known among founders and pattern makers, that gray cast iron shrinks during cooling about ⅛ inch per foot, white iron ¼ inch per foot and cast steel ⁵⁄₁₆ inch per foot. That is, a bar cast exactly one foot long will be found when cold to be ⅛ inch short if of gray cast iron, ¼ inch short if of white cast iron and ⁵⁄₁₆ inch short if of cast steel. The patterns have to be made larger than the castings desired to allow for this shrinkage.

But, during annealing, white cast iron loses one-half of its ¼ inch per foot shrinkage and the resulting malleable cast iron is found to have a net shrinkage of but ⅛ inch per foot which is the same as that of gray iron.

It appears that the precipitation of the temper carbon expands the bar throughout to practically the same dimensions which it would have had if flake graphite had been allowed to precipitate through slow cooling, as is the case with gray cast iron.

This is cleverly taken advantage of by manufacturers of swivels of malleable iron, such as those shown. The inner portions are separately cast first and thoroughly cleaned after which they are imbedded in another mold. The outer portions are then cast around them, shrinking so tightly upon the inner portions that they cannot be turned at all. However, upon annealing they loosen enough that they can readily be turned yet remain tight enough that they cannot be separated.

Malleable iron from which the carbon has not been removed can be hardened and given a steely fracture by sudden cooling from a red heat even if it has previously been annealed. Decarbonized malleable iron, also, can readily be recarbonized by the cementation process. These characteristics are often taken advantage of for the manufacture of tools from malleable iron. Hammers, wood working chisels, gears, etc., are quite largely made. Where they are sold at a cheaper price than the better steel tools and without misrepresentation, there can be little objection, but sometimes they pass for steel.

Ofttimes malleable iron castings are made in what are known as “permanent molds” of iron. They are really “chilled castings.” Annealing of these is accomplished in the regular way. Such castings have very smooth and beautiful surfaces but as the iron molds have high first cost they can be used only for castings for which the sales warrant the expense.

While much less malleable than is wrought iron or mild steel, annealed malleable cast iron has considerable malleability. It will resist great shock and can be severely battered and bent without breaking. It has about 75 per cent or more of the tensile strength of mild steel and because of the cheapness of its castings the malleable iron industry has developed wonderfully. About a million tons of this product are produced here each year.

Naturally malleable iron castings are used where a material with properties intermediate between cast iron and steel will suffice. Such are castings for railroad cars, for reapers, binders, and other agricultural machines, pipe fittings, and the cheaper grades of tools.

CHAPTER XIII
CAST STEEL

We have seen how primitive man hunted and fought with no implements and weapons better than clubs, bows and arrows, and stone hatchets, and how his wife cracked and ground the corn between flat rocks or in mortars of stone. In the succeeding “Bronze Age” we found ornaments, idols and tools being made of copper or the copper alloy, bronze. It was only after the next great advance that we found man utilizing iron for his purposes of civilization. This metal, which with us is so common, was in those days very expensive, so much so that it could be used only for purposes of war and as the gifts of kings.

But the world was traveling fast and it was not long before the iron-carbon alloy, steel, was produced. Even so, many hundreds of years elapsed before the present wonderful age was ushered in through the great inventions of Henry Bessemer and the Siemens brothers. And while fine steels for swords and tools have had an incalculably great influence upon the development of the human race, it was the mammoth production of Bessemer and open-hearth steel which permitted its general use as a material for construction of ships, bridges, buildings, and for railroads, that made this the “Age of Steel.”

Speaking in terms of the power house, it is also the “Age of Cast Steel.” Twenty-five years ago the manufacturer and power house man were quite content with their “saturated” steam temperatures and pressures. With cast iron valves and fittings their plants were well equipped.

Steel Castings Showing the Risers on the Flanges
Castings for use on steam, ammonia, water lines, etc., must be of very close-grained metal and require much larger risers than castings for less exacting service.

But the world did not stand still. It became known that by heating the steam out of contact with the water in the boiler it lost the moisture which it carried and became dry and then could be charged with as much additional heat as it was desired to give to it. This “superheated” steam, of course, would do more work and it had also certain other advantages which the old-fashioned “saturated” steam had not.

But while cast iron fittings gave satisfactory service up to temperatures, say, around 450° F., they faltered when forced to work under the new conditions which meant decidedly higher temperatures and pressures. And, too, the repeated heatings and coolings which were often necessary, disclosed a disadvantage previously unknown—a so-called “permanent growth” of the cast iron which was attended by loss of strength, and altogether it was soon found out that when superheated steam was to be used, higher types of materials were advisable than those which had been used under old conditions.

Steel Castings in the Annealing Oven

Superheated steam has rapidly come into general use. Some of the new locomotives and most of the modern power plants are now built for as much as 200° superheat, i.e., a total temperature of approximately 600° Fahrenheit.

Valves and fittings of cast steel not only are the articles “de luxe” for such service but they have come to be considered the necessary articles and their advantages have only fairly begun to be appreciated.

Though our most august scientific societies are proposing and debating upon systems of classification which shall include and satisfactorily define all of our ferrous metals, a satisfactory one has not yet been evolved, and, considering the intricacy of our ferrous metallurgy and the discoveries which are being made almost daily, the outlook for a strictly logical classification is not yet flattering.

With “Cast Steel” our metallurgical nomenclature is again faulty. Before what we now call the “steel casting” was known, crucible steel was poured into ingots, “forged” into tools just as it now is and often went under the name “Cast Steel” to distinguish it from the contemporaneous material, wrought iron. So to-day we buy many tools and implements which bear the name cast steel, which we know to have been forged in bringing them into their final shape.

But it is not these which we mean by the term, cast steel, but rather those steel products which get their final form by being “cast” from a fluid condition into a mold. These are what are rapidly coming to be understood when the term “cast steel” is used.

Satisfactory metal for steel castings may be made in any of three or four types of furnaces, but, as was suggested before, the making of molds for castings is a fine art, as is the preparation of the metal which is to go into them. Further, the making of that special class of castings which are to withstand water, steam or air pressure is a very different thing from the making of steel castings for other purposes, and this is too often forgotten.

For the former are necessary particularly close-grained castings, free from flaws or spongy spots. Under the great pressures applied such defects would certainly allow leakage.

Flanges and Fittings of Cast Steel

Whatever the method of production of steel for castings the metal is poured into molds to receive its final shape. Because of the intensely high heat of the steel only sands of great refractoriness (resistance to heat) can be used as material for the mold. White silica sand is such a material and is generally used, mixed with enough clay and molasses-water to give it “bond.” While molds for some steel castings are made in “green” (i.e., undried or unbaked) sand, baked molds are preferred for fine finish and surest results. After the making of the molds in the usual way they are sprayed with very finely powdered white sand or quartz mixed with a little molasses-water. They are then thoroughly dried in an oven.

Cast steel shrinks during cooling even more than malleable iron and the pattern and mold must be made to allow for this. Upon the freezing of the surfaces of the casting with consequent attainment of rigidity, the interiors, which freeze last, may have cavities unless means for avoiding them is provided. For this purpose heavier pieces, which later can be cut off, are cast upon such parts of the casting as tend to have “shrink holes.” These may be likened to receptacles filled with fluid metal, which being larger than the parts of the castings which they “feed,” hold excess metal in fluid condition until the casting itself has become solid throughout. Such are usually called “risers” or, in Europe, “lost heads,” and the molten metal in them flows down into the interior of the casting and fills the shrink holes which are forming. Not only must the risers be large enough that the metal in them is the last to solidify but they must be built high enough above the casting that sufficient pressure is exerted on the steel entering the shrinking parts to make its entry sure.

Grain of Steel Castings as They Come from the Mold

Grain of Steel Castings after
Annealing

(Magnification 60 diameters)

Baked molds, of course, are comparatively rigid. As the risers which stand on top of the flanges and other high parts of castings aid in resisting the natural shortening of long castings during and after “setting” of the metal, there is great liability that the still red-hot casting will crack somewhere along its length. It is therefore necessary to loosen with bars the sand of the mold as soon as the metal of the casting has set, particularly between the risers, and to break out the sand of the core inside, around which the shrinking metal might crack were the sand left in its hard packed condition.

After the casting is shaken out from the mold, it is cleaned and the risers cut off either by sawing or with the more modern oxy-acetylene torch flame.

Other Typical Steel Castings

Steel castings should be annealed in order to “refine,” i.e., make finer the grain of metal and to equalize “strains” which are set up in the castings during cooling. Coarse grain and internal strains tend to make the castings brittle. No such extended annealing, however, is necessary as is the case with malleable cast iron, for no divorcing of carbon from the iron with separation of free carbon is possible. The castings are carefully heated to a temperature of about 1600° or 1700° Fahrenheit and allowed to cool slowly.

After annealing, they are cleaned and excrescences removed by chipping, after which the castings are tapped, drilled or otherwise machined according to the purposes for which they are intended.

Cast Steel Valves, Steam Separator, and Direct-return Trap for Use with Superheated Steam

While more costly in manufacture and installation than are those of cast iron, valves, fittings and other cast steel products are, so far as we now know, practically permanent. Their notable shock resisting quality is well shown in the following table which is reprinted from page [188].

Pouring Steel into Molds from a Bottom-pour Ladle

It is to be noted that while malleable cast iron far surpasses “semi-steel” in this property, though their tensile strengths are ordinarily somewhere near the same, cast steel, in turn, offers more than six times the resistance of the malleable iron to shock and has nearly double its tensile strength. It is this great strength and resistance to shock, heat and pressure, with freedom from “permanent growth” under alternate heatings and coolings that make cast steel such a valuable material for the many purposes for which castings are to-day employed. Millions of steel castings annually find varied application.

Alloy	Tensile Strength	Number of Blows	Total Foot Pounds
Cast Iron	23,400	7	102
Semi-steel	35,050	11	206
Malleable Cast Iron	37,140	22	1,580
Cast Steel	72,120	92	10,112

In modern power houses and other commercial steam and hydraulic installations particularly, steel castings have come to be the materials usually specified and approximately the only ones which satisfactorily serve under the severe conditions of to-day.

Pouring from a Lip-pour Ladle

Undoubtedly the first steel castings were poured from crucible steel, though we must remember that the crucible is a melting and not a refining furnace. This was only natural. In the crucible the metal can be made very hot and fluid, and if of proper composition and properly “killed” crucible steel makes very fine castings. Crucible steel castings, however, are not in as fortunate a position as are other products of this high grade material. Tool steels ordinarily bring high enough price that there remains a profit to the manufacturer though his manufacturing cost is necessarily high. In the steel casting line, however, there is much keener competition and crucible steel has had considerable difficulty in maintaining its place. It seems to be a matter of price alone.

Open-hearth steel is very largely used for steel castings, more than two-thirds of all made in this country being of this material. About one half of these are poured from basic open-hearth metal, and the other half from acid metal. It is generally considered that the product of the acid-lined furnace is a little freer from over-oxidation.

Tapping Side of Two-ton Oil-burning Open-hearth Furnace for Steel Castings

Open-hearth steel cannot generally be as hot and fluid as are the steels made in other types of furnaces. For this reason as well as because of the larger size of the usual open-hearth furnace, small castings are not generally poured from this steel. It is for steel castings of considerable size and where there are sufficient orders to warrant a steady and large output that the open-hearth has its place. True, smaller open-hearths are now built, some of them of only two or three tons capacity, but, in general, the standard open-hearth for steel castings is of fifteen tons or more capacity and of the style of the open-hearth furnaces which were described in Chapter IX.

In their proper sphere they are highly satisfactory, but they are “inelastic” in that they must be run continuously day and night and should not be allowed to cool until extensive repairs are imperative.

It was mentioned that in the open-hearth process the furnace is always hotter than the metal which it contains and that the heat which can be put into the steel is limited by the ability of the refractories of which the roof and side walls are made to withstand melting. In the Bessemer process the metal is hotter than the furnace because the heat is generated by combustion of certain of the metalloids contained in the metal itself. As metal for castings must be very hot and fluid the Bessemer process is very satisfactory for the making of steel for castings.

A 30–ton Basic Open-Hearth Furnace Tapping
The overflow from the ladle into the pit is slag.

It has, also, the advantage of “elasticity.” The supply of metal is practically continuous and one furnace can make from one to eighteen or even more heats on day turn only and be shut down for the night turn or longer and then started again without such loss as would result from the shutting down of an open-hearth furnace with regenerators.

For the making of metal for steel castings, very small-sized Bessemer converters are used which make from one to three tons of metal per blow. Some converters of as little as one-half ton capacity are being used. While some are of the “bottom-blown” type already described, the majority are what are called “surface-blown” or “side-blown.” In these, from four to eight round tuyères, about one and one-half inches in diameter each, pierce the brick or ganister lining just above the surface of the bath. They slope downward a little toward the bath so that when the converter is tipped to its upright or blowing position the air blast will strike the adjacent edge of the metal and blow across its surface. This three or four pounds per square inch of air blast keeps the metal in circulation, meanwhile burning out its silicon, manganese, and carbon, just as it does in the larger bottom-blown converters. Surface-blown give hotter metal than do bottom-blown converters and very fine steel castings are made from their metal. For these converters, which are practically all acid-lined (i.e., with silica or clay brick or ganister), metal low in phosphorus and sulphur is regularly drawn from a cupola specially run for the purpose.

Small Side-Blown Converter Making Steel for Castings

The remaining recognized type of furnace for steel for castings is the comparatively new electric furnace.

Commercial melting of metals by the electric current has been sought for half a century. In 1879 the first furnaces of promise were patented by Sir William Siemens, one of the Siemens brothers who became so well known through their great work with the open-hearth furnace, the gas producer and many other things metallurgical. While Siemens melted as much as twenty-two pounds of iron per hour in his furnace, the cost of the electric current at that time was so high as to be practically prohibitive for the manufacture of steel in competition with the open-hearth, Bessemer and crucible processes.

Drawing of Side-Blown Converter in Blowing Position, Showing Edge of Metal Even with Row of Tuyères

Little of great moment in the electric furnace line developed during the nineteen years which followed. Then, in 1898, Stassano in Rome, Italy, constructed a furnace in which three carbons gave an electric arc above the surface of the bath. About the same time, Heroult, a Frenchman, was developing the electric furnace which to-day has become so well known in this country, and which bears his name. Other well known furnaces of the arc type are the Gronwall-Dixon, the Snyder, the Girod and the Rennerfelt.

In general, electric furnaces have more or less round steel shells with shallow brick, magnesite or sand-lined hearths, and sidewalls and removable roofs of brick. Heat, of course, is furnished electrically. In most of them long carbon electrodes are lowered through holes in the roof until the lower ends strike an arc with the metal on the hearth. The number of carbons may be from one to four or more depending upon the style and size of the furnace and the manner in which electrical connections are made. All of the furnaces mentioned have been used for the production of steel for castings and the Heroult and Girod are in use in larger sizes for electric steel for rails and miscellaneous products. The steel is first cast into ingot molds and is later rolled down into bars, rods, etc.

A Gronwall-Dixon 5–Ton Electric Furnace Tapping

First Experimental Arc Electric Furnaces Patented by Sir William Siemens in 1879

All of the above use carbon electrodes and are known as “arc” furnaces. There is a distinctly different type of furnace which, also, is in use in commercial sizes. This is the “induction” furnace. In this, what is known to electricians as a secondary current is “induced” in the bath itself and heats the metal. Of this type the Kjellin and the Rochling-Rodenhauser are the best known in this country. While they are in use in the larger sizes for production of steel for ingots, these two furnaces do not seem to have been used to any extent for metal for steel castings.

The details of construction of the furnaces which are used for metal for castings are more or less different, but they are not of particular interest to us. The working of all is similar and a general description should suffice.

Drawing of the Snyder Electric Furnace

Small Snyder Electric Furnace Tapping

Whether starting with furnace cold or hot, materials in molten or in the more usual “cold” form are charged on the shallow hearth of the furnace. The charging doors are closed, the current is turned on and the carbon electrodes are lowered until an arc is struck between the upper electrodes and the metal on the hearth, which in some way is made to connect with the negative electrodes. In one or two of the types mentioned the arc plays between the carbons, all of which are above the bath.

At first there are great fluctuations in the current intensity because of the uneven surface presented by scrap steel on the hearth. In a short time, however, the current steadies. The intense heat of the arcs soon brings cold steel to a molten condition.

Occasional attention from the attendant is necessary to see that the melting is even and that any outlying pieces of steel are pushed to the center where they must melt.

Sketch of the Heroult Three-Phase Electric Furnace
There are three electrodes, all of them above the bath. Only two show here.

In the basic-lined furnaces lime is usually charged with the cold steel. With the iron oxide which is added from time to time this forms a highly oxidizing slag, which, after it takes the phosphorus from the metal, is skimmed off.

As you will remember, the other processes stop at this point, little further refining being possible. In the electric furnace, however, the sulphur, also, can be reduced to almost any desired amount by use of a further addition of lime, and greater heat. Not only can the sulphur be reduced to very small percentages but the over-oxidized bath can be brought to neutral condition and the green or black slag made white with return of its manganese and iron to the bath. This is accomplished by addition of small amounts of powdered coke or coal. The whole process is under very accurate control.

With a practically white slag, which is the signal that the deoxidation of the bath is complete, and the sulphur reduced, the steel is ready to pour provided it is hot enough. Tests of this are usually made either by pouring a little of the steel from a small ladle and observing its fluidity or by observing the quickness with which the end of an iron bar is melted off when plunged into the metal in the furnace.

The Heroult Electric Furnace

Of all the metallurgical steel furnaces, the electric furnace is the most susceptible of accurate control. With the heat applied directly to the metal in the cleanest way possible, i. e., without the admission of coal ash or gas or air of the blast, the atmosphere in the tightly closed electric furnace can be made “oxidizing,” “neutral,” or “reducing” at will. The metal can be held in the furnace and additions made, samples taken, and the operations conducted with regulation and certainty.

This newly devised metallurgical apparatus is coming to be largely used in the production of tool steels. While it has not displaced the crucible method for the production of steels of the very highest qualities, it has proceeded far enough in this direction in the very limited number of years since its introduction, that it is certain that the crucible, even for tool steels, is to have a keen competitor. Tool steels in considerable variety are to-day being quite satisfactorily made in the electric furnace and it is not at all unlikely that steels of the very highest grade will shortly be produced by this method.

CHAPTER XIV
THE ALLOY STEELS

We have learned that steel, fundamentally, is an alloy of iron with carbon, i.e., carbon is the characteristic element. We are now to note what often seem to be exceptions to this rule. While in reality steel is just this iron-carbon alloy, there are alloys known as steels to which such strong characteristics are given by elements other than carbon, that carbon seems not to be the defining constituent at all. Indeed, in some of these, the carbon content may be small enough that, judging from our experience with the carbon steel series, we would not expect any such physical properties as some of these alloy steels show.

You remember that in olden days they distinguished between wrought iron and steel by quenching the piece in water from a cherry-red heat. If the piece was hardened and made brittle, by this treatment, it was thereby proved to be “steel.” Also, it is generally known that by annealing a piece of hardened steel, which usually means holding at a cherry-red heat for a time and then cooling slowly, it is made soft.

Then what shall we say concerning a certain one of these new alloys, Hadfield’s manganese steel, which is made very much less brittle and a little softer by quenching, but which refuses absolutely to soften under annealing treatment—in other words, is almost the opposite of what we know as steel in these chief defining traits? The nickel steel which contains 15 per cent of nickel, also, exhibits just these characteristics, being softened by quenching but not by annealing.

Again, while iron, the carbon steels, and even the magnetic oxide of iron which contains only about 72 per cent of the metal, are strongly magnetic, manganese steel which has 85 per cent of iron is so non-magnetic that it is sometimes used in place of brass or bronze where an entirely non-magnetic material is required. The nickel steels with 24 per cent or more of nickel are also non-magnetic though both constituent metals, alone, are strongly attracted by the magnet.

These are some of the things which make a logical classification of the iron family so difficult. Though derived from the steels which we knew and made from the same materials with the exception that a greater amount of one constituent, manganese, is added, or perhaps, in other cases, another element or two, the resulting alloys have markedly different and often contradictory properties.

However, we must not be led astray. In all probability carbon is still the necessary constituent, but much less of it is needed to produce results when the other elements are present. There is no doubt, however, that in “manganese steel” or in “nickel,” “chrome,” “tungsten,” “silicon,” “vanadium,” “titanium,” and other alloy steels, the added element or elements exert very strong modifying influences, and sometimes obscure the influence of the carbon.

In the first place, we better at once dispose of certain of these steels by terming the added element a “scavenger” only. Such usually are “titanium” and “aluminum” steels. These are generally ordinary carbon steels in which a very small amount of titanium or aluminum has been used to rid the alloy of certain gaseous or other deleterious elements. Upon analysis, steels so treated often show no trace of the element which has been added to do the work, all of it having passed into the slag, carrying with it the obnoxious substances, which, had they remained would have injured the quality of the steel. Manganese and silicon which were spoken of in the discussion of the Bessemer process as deoxidizing the metal, also exert just this same influence, though there is usually added of these enough that a certain percentage remains in the finished steel. Vanadium and titanium have a particular affinity for oxygen and nitrogen, and aluminum for oxygen. By chemically combining with these gases in the metal, and through possible other influence, they help to produce sound steel having very good physical properties. Vanadium, however, is much more than a “scavenger” as will be seen later on.