quantity of metal was sufficient to permit fair skimming and at 3 hours and 30 minutes a considerable amount of slag of good appearance was skimmed off. At 4 hours and 10 minutes a test bar showed 0.73 per cent silicon and 0.968 per cent sulphur, while a second bar poured at tapping out 4 hours and 40 minutes from starting showed as follows:

A test bar poured from a ladle taken off during tapping out showed 0.68 per cent silicon, 0.081 per cent sulphur and a tensile strength of 35,600 pounds. Another analysis from pig, taken in same way, showed as follows from a different chemist:

Analysis of iron from hand fired furnaces in the same plant showed as follows:

As a result of this test, it was conclusively shown that all of the air should be admitted through the burner and arrangements were then made to set up a burner of sufficient capacity for this purpose. At the time of sending this paper to press, reports on additional heats have not become available. It is hoped that further data will be available later.

Another Installation

A 30-inch burner already described has been installed on a furnace, built for a capacity of 10 to 15 tons in a prominent foundry in Erie, Pa. Acting on the idea of having

the initial combustion occur on the front end of the charge, the combustion chamber was made only about 2 feet in length from the noze of burner to the front bridge wall. For reasons noted, this proved to be unsatisfactory and it was considered desirable to lengthen the combustion chamber about 7 feet, thus completing combustion by the time the gases reach the bridge wall so that the extremely hot gases come in contact with the charge.

This lengthening of the furnace, together with other minor changes and adjustments, as well as insufficient mill capacity to supply coal for full time operation, and the fact that the conveying system from the mill to furnace hopper has not been installed, caused delay in placing the furnace in regular operation. It is contemplated to put this unit in daily service as soon as circumstances permit. Unfortunately sufficient heats had not been made, at the time of completing this paper, to give an idea of the melting ratio. The burner has demonstrated great flexibility in operation and can be regulated at will to secure any character of flame desired. The operating staff is well satisfied that the conditions since altering the furnace are such as to give satisfactory results. It is hoped that further data will be available from additional runs to be made with this equipment later.

Malleable Iron Annealing

In malleable iron annealing, the burner described in this paper performs functions equally as satisfactorily as in the air melting furnace. For greatest economy in the use of pulverized coal, the ovens should be of comparatively large. size. It then becomes a problem of uniform heat distribution and control so as not to burn the pots. With some types of pulverized coal burners it is found that the heat is thrown in too great degree toward the front end of the ovens. At times the poor heat distribution with high velocity burners results in burning top pots near the burner, while the bottom ones on the same stools may not be annealed. Where a cutting flame cuts down the brick of the firebox, the destruction of pots near the burner may be aggravated from slag deposition and fluxing. At other times, if the coal

happens to be a little damp or not ground as fine as usual, some of it may be deposited on the floor and blanket the bottom pots near the burner.

For reasons mentioned, the complete mixture delivered by the burner under consideration, avoiding a cutting flame, largely overcomes the slag difficulties and at the same time. prevents an undue part of the hot gasses being driven to the front of the oven. The low feeding velocity gives a certain time element for the gases to expand and normally fill all parts of the oven, especially near the burner. These burners should with proper damper regulation of the oven produce more uniformly annealed castings than heretofore has been the practice. The first of these units on a pot oven is now being installed at Buffalo, N. Y., and it is hoped to have additional data later.

While no commercial installation has yet been made on the muffle type of annealing oven, a number of test runs were made on such an oven handling railway castings. In a plant having 26 ovens of a capacity about 25 tons each, the practice with hand firing was to secure one pound of castings annealed per pound of coal. An improved form of oven was constructed and on a specially conducted test showed a ratio of 2.2 to 1. A 6-inch powdered coal burner on the same oven showed a ratio of 5.2 to 1 and the test record of flue temperature (presently to be mentioned) indicated that the length of this oven could be increased 6 feet without using any more coal, in which case the ratio would be 7.25 to 1. The difference between 1 pound of castings and 7.25 pounds per pound of coal is shown graphically in Fig. 8. On this basis 14 ovens will do the work of 12 and there is a capital saving of $24,000 besides space released for several molding floors, etc.

Temperature readings in the flue, bottom of muffle, top of muffle and in the flame way at end of oven opposite burner, were taken hourly by a thermocouple which had been calibrated. The readings are shown on the chart, Fig. 6, and it will be observed they are not entirely consistent, due in part to condition of the pyrometer and also to changes made in rate of feeding, setting of stack damper, admission of air

at burner, etc. Curves have been drawn to indicate the average of the readings, the ultimate temperature desired not being as high as with pot ovens. By a more rapid rate of firing the temperature could have readily been brought up faster from the twentieth to thirty-fifth hours and the total time of firing might have been shortened two to three hours. The total time of firing, however, was 49.5 hours as compared to the usual hand firing period of 90 to 110 hours on the same ovens. The special hand-fired test showing ratio of 2.2 to 1 was made in 72 hours firing time. The total coal consumption on this run was 9650 pounds and 25 tons of castings were annealed, making the ratio 5.18 to 1. The castings were of excellent quality.

Conclusion

The author feels gratified in being able to record the progress so far made with the air melting furnace. In so far as he has been informed, the installation at the plant at Erie is the first commercial unit to be put in service on an air melting furnace. The 30-inch quadruplex burner there used, with its nozzle delivery of a stream 6 by 60 inches is believed to be the largest single pulverized fuel burner ever built. This installation therefore should constitute an interesting landmark in the development of improved firing methods for the malleable iron foundry.