cord and securely held. In practice, they are placed close to each indicator, thus making it possible to use a very short string to that instrument. This not only reduces the stretch, but it makes it handy for the operator. He can hook and unhook the string without leaving his position, and when string is unhooked, it hangs a few inches only below the indicator. This keeps it off the floor, where it is very apt to get when long strings are unhooked. The endless cord is composed of picture-cord wire, of medium size, excepting where it passes around the end pulleys, where best linen cord is used, such as is sold by indicator manufacturers. The other angles being very slight, there is no danger of the wire breaking, as it would if allowed to pass around the end pulleys 2 and 7. The right and left shown at 13 is for the purpose of bringing the endless cord up to the required tension. When drawn taut there is practically no stretch to this wire; and it will be seen that although

cord is used at each terminus of the endless string, any stretch that may occur to it does not in the least affect the motion to the indicators, as the cord is on the return portion of the endless string, and is not subjected to the strain caused by pulling out the drums against the full tension of the springs. The clamps, 12, into which the indicator strings are hooked, are all on the wire which starts from the pantograph pin 3, and the cord which passes around the end pulley 7 at high-pressure cylinder, is connected at a point 14 beyond clamp which operates indicator at that end. Thus it will be seen that the strain comes entirely on the wire, and the stretch is practically done away with. The writer has seen cards taken where four indicators were in operation at once, and the pencil allowed to trace the cards a hundred or more times. The result was merely a heavy line all around the card, excepting, of course, the expansion line, which must necessarily vary with changes in point of cut-off. When it is remembered that it requires a pull of from 10 to 12 pounds to operate the drum of an indicator which is

adjusted for ordinary speeds, it is apparent that this is a very good showing, as four indicators were used, thus making a total strain of forty to fifty pounds. The whole affair is rigid, has large bearing surfaces, and is oiled throughout by means of stationary oil holes. The pin 3 at the pantograph is provided with a yoke 15, which can be slipped off, thus enabling the endless cord to be stopped without interfering with the running of the pantograph. The small board 16 shown at top of column A serves as a writing-table, on which cards can be kept and data marked on them. An elastic band serves to keep cards from being blown off. One apparatus is made to suit both the smallest and largest engines in use, and is designed to pack into small boxes, and weighs complete, boxes and all, about 20 lbs. It has been in use about six or eight months, and is well liked by the several parties that are using it. Being nickel plated, and the woodwork made of mahogany, it presents a handsome appearance. The inventor has applied for letters patent.

DISCUSSION.

The President, H. R. Towne.-I presume it is a novel feature to have a device by which the four ends of a compound engine can be indicated simultaneously, which I take it is the point of the device described.

Mr. F. H. Ball.-I would like to ask the author of the paper at what speeds he has used this apparatus. I was always of the opinion that this pantograph was not suited to high speed. I thought perhaps this was adapted only to moderate speeds, say, not over 100 revolutions.

Mr. Jacobi.-Mr. Ball is quite right in regard to the speed at which the pantograph can be used, as I have never used it at a greater speed than he mentions.

While I have never had occasion to use this device on highspeed engines, I see no reason why it should not give good results.

It is not absolutely necessary that the pantograph be used for reducing the motion, as the device can be made to work equally well if operated by a lever or pendulum, as commonly used in connection with high-speed engines.

CCCXL.

THE PIPING OF STEEL INGOTS.

BY THOMAS S. CRANE, NEWARK, N. J.

(Member of the Society.)

THE interesting discussion at the last meeting relative to cracks arising in the hardening of steel articles suggested to the writer that it was probable that many of the cracks arose from a cause not referred to at all.

The cast-steel ingots ordinarily made have a serious defect termed a pipe, at the upper end, which is shown in the annexed Fig. 138, photographed from an ingot cut open to expose the pipe.

No subsequent working of the ingot serves wholly to unite the walls of this pipe, as they become quickly oxidized by exposure to the air, and no reheating suffices to make them weld thoroughly together.

Every bar of cast steel formed from such an ingot, therefore, has a defect at one end, which it is common to remove by breaking off the bar in successive sections, until careful inspection shows that the defective part is removed.

Inspection is not, however, capable always of preventing the steel from passing into the market with a portion of such defect, and the defective spot in the steel may, in the subsequent working, find its position at the top, bottom, side, or interior of a tap, gauge, reamer, or other tool, and develop a crack of some inexplicable character when the steel is hardened.

Strenuous efforts have been made, and by many different modes, to prevent the piping of cast-steel ingots, but it is only recently that a simple apparatus has been perfected for practically accomplishing this object, and it is reasonable to suppose that the use of ingots formed entirely free from piping will, in many cases, prevent the cracking of the steel when hardened.

Before referring to this apparatus, I will mention briefly the inost modern means heretofore used.

The "Sweet" process consists in putting powdered charcoal upon the top of the ingot when poured, to prevent its upper end from oxidation, and, by its combustion, to maintain the fluidity of the steel, and thus assist in filling the pipe as it forms. The entire effect is very slight.

The compression process used by Whitworth to form sound steel ingots has never been wholly successful, as it operated to consolidate the exterior of the casting without permitting the free discharge of the gases from its interior; and while it has operated to prevent the formation of a pipe or local depression, it has been liable to produce a spongy or porous casting. Various modifications of White worth's plan have been devised.

S. T. Williams has devised a compression process for making sound circular ingots for saw plates.

The comparatively thin and flat form of such ingots permits the sides to be bent or crushed inward while the interior of the ingot is still at a welding heat, and this effects the desired purpose much better than in a square ingot, where the compression of the sides would tend to induce cracks, as the metal, when first crystallized, is not very tenacious.

4

The "Billings" process for compressing steel ingots was intended to apply the pressure instantly when the casting was formed, but operated only to lock the gases within the ingot.

In experiments tried by William R. Hinsdale, at the Jersey City Steel Works, in the year 1884, it was found that a pressure of 300 pounds per square inch, operating upon a 24-inch piston, and concentrated upon the end of a 3-inch-square ingot, merely produced an ingot containing innumerable globules of gas.

FIG. 138.

If the pressure was deferred until the ingot slightly hardened, a pipe would form in the ingot at the upper end, and would remain permanently, as the hardening would 'prevent the pressure from operating effectively.

The "Billings" and "Hinsdale" process provided a reservoir at the top of the mould, and a movable plunger within the mould, by

which the steel was drawn downward to make an ingot, which would be fed, during the shrinkage period, by the residue remaining in the reservoir. This process is not, therefore, convenient except for the casting of large ingots.

Mr. Hinsdale also experimented at the Jersey City Steel Works

with a pressure of 60,000 pounds per square inch upon the metal, applied to the end of a 3-inch-square ingot,-nearly double the pressure ever applied by Sir Joseph Whitworth.

The result was the shortening of the ingot from 25 to 22 inches in length, and perfect solidity except that the pipe appeared in the same form, a flaw, as it ordinarily displays itself at the piped end of a forged bar.