melted, due to an accident in the pipe line, and in five hours after starting the fire again the heat was remelted, hot, and saved by the use of ferrosilicon and ferromanganese. This would have been impossible in a coal-fired furnace. We commonly melted 15 to 17-ton heats, charged in a cold furnace, in 42 to 5 hours, and a 12 to 15-ton afternoon heat charged in a hot furnace in 32 to 4 hours. As our coal is unloaded from the car at a point from which the fireman throws it into the furnace, we could show no economy in handling coal. You can readily estimate from conditions in your plant and determine whether you could economize in this respect, as well as in the disposition of cinders and ashes.

I am unable to furnish data on the comparative cost of repairs as it is our custom to repair each furnace every week, or after 12 heats, in order to have each furnace repaired in regular rotation, since we have six furnaces in one of our foundries. But we know positively that the oil-fired furnaces were in much better condition on repair day than were the coalfired, and they undoubtedly would have run several more heats safely.

One-Man Furnace Gang

We had but one man in the oil furnace gang. He very comfortably took care of the skimming and loading the slag on the car, as well as the tapping. We have but one spout on most of our furnaces; but on a few of them there are two spouts. Our furnacemen were so pleased with the oil furnaces that there was a bitter rivalry among them to work on them in preference to the coal-fired. Under present labor conditions, this feature is one whose value in dollars and cents cannot be estimated.

I regret that owing to lack of time, due to additional duties, I have taken on for the past few months, I have been unable to embody in this paper a sketch of the principal dimensions of my furnaces, but will try to describe them so that they may be readily understood. Our furnaces are 21 feet 6 inches long between front and back bridge walls, and 6 feet wide. Our bungs have a 9-inch spring to the arch. We made

no change in the back bridge wall. In getting the lines of the furnace we started at the skimming doors for a level. After having torn down the front bridge wall to a point 2 inches. above the level of the skimming door, we filled up the old fire box with brick bats placed on top of the grates. Our combustion chamber occupied the space over the former fire box and extended to a point 10 feet from the front bridge wall, the opening at the firing end extending to the end plate of the furnace, and at this end the bottom of the combustion chamber was about 24 inches above the level of the new front bridge wall, or about 26 inches above the level of the skimming door. The bottom of the combustion chamber was made of silica sand and sloped in a straight line from the firing end to the front bridge wall. The opening at the firing end was 16 inches wide and 20 inches high, into which was inserted the blast pipe which at the opening was 13 inches square, and upon top of which the burner was placed. A 14-inch diameter blast pipe furnished the air. It was controlled by a slide valve. In addition we used our regular top blast, through the regulation of which we secured air necessary for complete combustion. The side walls. of the combustion chamber were 30 inches above the front bridge wall and ran in a straight line to a point 20 inches above the bottom of the chamber at the firing end. In other words, the combustion chamber was the width of the furnace, 6 feet at the bridge wall, and 30 inches high on the sides, with an additional height of 9 inches resulting from the arch spring in the center of the bungs; it gradually narrowed to a width of 16 inches, and a height of 20 inches on the side walls, with an additional height of the spring of the arch in the center of the bungs. The side walls were 12 inches above the back bridge wall, gradually sloping in a straight line between these points. We made no change in the lines of the furnace between back bridge wall and stack. The oil was pumped through a registering meter to the burner by a direct connected centrifugal pump, at a pressure of 12 to 14 pounds after being heated to 100 or 110 degrees Fahr. It was atomized by compressed air, or superheated steam at a pressure of 25 to 35 pounds.

We learned that we could regulate the blast, both direct and top, more effectively from the appearance of the gas than by any fixed rule, and also learned that much less air was used than in burning coal.

Accompanying this paper are given data on the construction and operation of furnaces in three different plants. The table follows:

I take this opportunity to express my thanks to the Iowa Malleable Iron Co. for information and co-operation given me during my early experiments and for data given. I also thank the Jewell Steel & Malleable Co. for data given. I sincerely hope and believe that in the near future there will be evolved a plan by which fuel oil will be economically used in air furnace practice.

ciable

None
Consider-

None

Less

Much less

ably less

The Refining of Cupola Malleable

Iron in the Electric Furnace

By A. W. MERRICK, Schenectady, N. Y.

The American or blackheart malleable cast iron produced in this country is usually melted in air (or reverberatory), open-hearth or cupola furnaces. Either of these types of furnaces offer certain inherent advantages and disadvantages, but as it is with the last named that we are to deal in this paper, the discussion will be limited to this one method of melting.

Its advantages are briefly as follows: Low initial investment; low cost of operation, upkeep and repairs; intermittant or continuous operation which gives a great flexibility of capacity; high melting ratio of iron to fuel; and ability to melt high percentages of cast iron and steel scrap without the consequent lowering of the carbon content which would necessarily follow in either of the other two types of furnaces. From the foregoing it will be seen that the cupola will produce molten cast-iron at the spout cheaper than any other type of furnace, and this fact is universally conceded by all authorities.

There are, however, certain disadvantages in cupola melting as applied to cast iron in general and malleable cast iron in particular. First, the sulphur absorbed from fuel is higher with this process than either of the others. Secondly, it is not possible to produce an iron of a low carbon content, which with malleable iron limits one to the production of castings of very light section. This is due to the fact that the melting stock, being in intimate contact with the incandescent carbon of the fuel, absorbs carbon so readily, that with the constant silicon aimed for in malleable iron, a practically saturated condition is reached so that with other conditions equal, the carbon is never very far either above or below a fixed point. This holds good even where the percentage of steel in the charge is varied considerably. Thirdly, cupola iron, especially on long heats, is liable to quite a variation in temperature and