a pile of slag and many lost castings. I would like to see the bureau of standards take the clay dug in the backyard and show the foundrymen just what they are up against. If some homely examples were given they would begin to think.

MR. HARRINGTON.-Mr. Staley has pointed out the necessity of using a plastic clay in a facing sand where clay is used to replace a portion of the new molding sand. We want to keep the sand as open as possible, and if we are not particular in obtaining a plastic clay, we are going to close up the voids in the sand. There are plenty of clays in New Jersey that are plastic and yet entirely refractory enough for the average gray iron facing work, and this test-either malachite green or crystal violet-will give a very fair test of the clays and their value as a binder for use in facing sand. The founder gains nothing by the use of the cheaper and less plastic clays, as by so doing, because of the fact that for the same binding strength much more of the less plastic clay must be used, he closes up his facing or heats to the point where scabbed castings are inevitable.

When things were booming and there was not enough labor to do the work the night laborer who was mixing the daubing for the day used beach sand instead of fire sand by mistake. When we were using the fire sand, we frequently complained that while it was entirely refractory it did not stay on the walls as well as it should. When, by mistake the beach sand was put in, we noticed that the daubing actually vitrified on the bricks and remained there four or five times as long and gave a sufficiently refractory material so that it really was better to use the beach sand.

MR. STALEY. In general the clays which are used by the crucible manufacturers and which were formerly imported from Germany belong in the class of plastic, vitrifiable, refractory clays that is, they vitrify at comparatively low temperatures, do not melt until very high temperatures are reached, and are clays which have high bonding power. It has been found that there are clays in this country which can be mixed together and produce mixtures which will take the place of the German clays but we have no one clay which answers.

Refractory Cements

By W. S. QUIGLEY, New York

The title I have chosen for this brief paper is purpose, selected to bring out a point with reference to high temperature cements on which there is some misapprehension. A refractory cement is not of itself just a refractory but bears its title from the fact that it is used as a binder for refractories. There is probably not a commodity used today by manufacturers in general that is as little understood or as much overrated as are high temperature cements.

In foundry circles there has been a great deal of educational work done during the past two or three years on the use of refractory cements, and while their successful application is well understood by many foundrymen there remains much information yet to be uncovered on the whole subject of their many and varied uses. Their use sprang up rapidly during the war owing to the necessity for using a material for bonding refractories which, by increasing the life of linings, would prevent shutdowns and by the use of which repairs to furnaces could be quickly made.

Bonding materials may be divided into four classes.

First we have fire clay, the primary function of which is to compensate for the inequalities of the bricks or shapes as a pliable refractory filler, and with which unduly thick joints generally are made. It has no binding strength of itself unless subjected to a vitrifying temperature. Furthermore, inasmuch as heat is required for vitrification in order to obtain a bond, it is obvious that only a surface bond is obtained, as the required heat for vitrification will not penetrate the entire thickness of the wall. The result obtained may be likened to a vitrified shell with a weak structural backing. Furthermore, the shrinking of the fire clay due to its own moisture and the combined water used in mixing it, causes a separation between the bricks or shapes in walls or arches which frequently results in bulging walls or collapse of the whole structure.

Secondly we have coarse grades of mixed materials or

so-called cements which also have no binding quality, depending upon heat or vitrification for a bond. Materials of this class must be made to bind or fuse at approximately the same temperature at which the furnaces are run, or no bond is effected. In other words, a cement which is good for a low temperature annealing furnace is not good for a high temperature forging furnace, or vice versa. Such mixtures are subject to the same surface bonding result as fire clay.

Thirdly, some cements in order to hold the component parts together depend upon a fibrous structure which shrinks and eventually loses its binding value as soon as subjected to any considerable degree of heat. Such cements must lose their effectiveness as the temperature increases.

Finally, to be universal in its application, a cement should air set and not depend upon heat for creating the bond in order to form a union throughout the structure. In process of manufacture it should be passed through a fine mesh sieve so as not to contain coarse particles which would tend to create voids between the brick. It should not shrink when subjected to heat. It should be a material which can be used as a binder with crushed fire brick, old crucibles, fire sand or fire clay, ganisters, and for making rammed-in linings, and doing repair work, hot or cold. And furthermore, its composition should be such that it can be used in neat form for making small hot patches and repairs.

The principal essential of a refractory cement is that it should have at various temperatures, the same co-efficient of expansion, as nearly as possible, as the materials with which it is used for bonding. Refractories themselves differ in expansion co-efficient, as in the case of fire-clay brick and silica. brick, where the former has 0.075-inch and the latter 0.175inch per foot at 2200 degrees Fahr.* Yet a cement with a co-efficient which lies between these two could be advantageously used with both refractories.

The difference in the cost between fire clay and high temperature cements must be justified by the difference in the results obtained.

Chemical and Metallurgical Engineering, Vol. 21, No. 3, page 153.

the Northern Pacific

By ARTHUR F. BRAID, New York

When the 526-foot army transport NORTHERN PACIFIC ran aground in a dense fog off Fire island, Long island, N. Y., on Jan. 2, 1919, she sustained numerous injuries which required extensive repairs in the Brooklyn navy yard. The interior machinery was badly was badly disabled, and the turbines were thrown out of alignment. A great many plates were badly battered, and had to be replaced. In addition to this, the problem arose as to what should be done with the huge cast steel sternpost, for just above the uppermost gudgeon the casting was cracked through, the cross-section of the break forming roughly a triangle, each side of which was about 2 feet in length. A view of the broken stern-post is shown in Fig. 1.

Break Was High on Casting

The high position of the sternpost break, because of the unusual circumstances which caused it, was unique in the history of stern injuries, most of which have occurred much lower down on the stern post, or on the stern shoe near its junction with the ship. The welding of this fracture also offered a special problem because the crack occurred in a hollow part of the casting, just above a solid heavy portion. The wall of the casting at the break averaged about 3 inches in thickness. The high location of the break left the huge lower portion of the casting to be supported solely by a comparatively slender horizontal stern shoe, which must have borne a terrific strain during the ship's rough treatment when she was grounded. To alleviate this strain, after arrival in dry dock, the stern shoe, which now supported most of the stern-post, was blocked up from underneath, and a wooden beam inserted in the propeller space, acting as a brace against the stern-post.

In handling this injury a mechanical repair was considered out of the question, and there remained only the alternative of making a thermit weld, or of purchasing an enormously expensive new casting and installing it at a cost probably

exceeding $50,000. Although the fracture was larger than that of any marine weld ever before made, the fundamental difficulties were not greater than those of many thermit repairs of considerably larger size which are constantly being made in the steel mills. This process, therefore, was finally chosen.

To those who are unfamiliar with the process, it may be explained that thermit is a mixture of aluminum and iron oxide. This mixture is ignited by means of special ignition