by making the retorts extra large (Fig. 21). The latter method is of course used where it is impossible because of local conditions to feed the coal in at both sides of the furnace.

"Fig. 20 gives a cross section of one of the boilers at the new plant of the Buffalo General Electric Co. at Black Rock, just outside of Buffalo, N. Y. Each 1140-hp. boiler is equipped with two 15-retort Riley stokers set back to back. This gives a total grate area of 417.8 sq. ft. and makes the ratio between grate area and water heating surface 1:27.3. This ratio is probably the highest which has ever been used and it means that the coal is fed through the retorts at an extremely low rate, thus resulting in high combustion efficiency. The volume of coal in the furnace is also very large. In this case it actually figures out that there I would be about 20 tons of coal and coke in the furnace at all times.

Extra Large Retorts

"To apply to boilers where a stoker can be installed only at one end and where high ratings are required, we have designed a

high capacity stoker with a retort area nearly 50 percent in excess of present standards, providing for the burning of much larger volumes of fuel and permitting greatly increased ratings. The additional retort capacity furnished by this stoker likewise permits of extra high ratings being maintained continuously with high efficiencies, and extra high capacity during peak loads.

"We hope that central stations will help to establish the use of larger grate areas in boiler settings, so that a still further development may be made along this line.

Maintenance of Brick Work

"With larger stokers and higher ratings, the maintenance of brick work becomes increasingly difficult, so much so that it appears to be one of the limiting factors in operation. In the section of the large stoker shown (Fig. 20), we call attention to the modification we have made in the conventional boiler setting by the removal of the bridge-wall. We believe that the space shown for the storage of excess ashes and clinkers not only does away with bridge-wall troubles but also makes it possible to burn out and then dump refuse satisfactorily without complicated ash removal devices. The stoker itself does not form a deep pocket with the consequent possibility of trouble if a hot fire is carelessly allowed to reach the bottom of the pocket.

"Another development in this line is the admission of air to the fuel bed through openings in the side walls. You will note that we have indicated on the setting (Fig. 20) that the ends of the bricks opposite the fuel do not touch each other. The mortar is omitted so that this space between the ends of the bricks forms a connection between the furnace chamber and a narrow chamber in the side wall. This chamber in the side wall communicates at its bottom end with the stoker wind-box and thus forms a means of supplying air from the air chamber through the side walls of the furnace. These openings between bricks extend for a distance of about 14 inches above the grate level. This ventilated side. wall construction has been developed at the South Boston plant of the Boston Elevated Railway Company.

"Another means of admitting air through the side wall consists of grooves in the bricks so arranged that they come opposite openings in cast iron boxes which are set inside the side

wall. The cast iron box forms a narrow air chamber which is connected to the stoker air chamber by large size pipes. These pipes are also connected to steam so that the pipes, boxes and openings through the bricks may be blown out by steam pressure. This construction, while somewhat more expensive, is perhaps more convenient and may be made equally effective by using enough grooves or holes in the brickwork to connect with corresponding holes in the cast iron boxes. This construction has been used by the Stone & Webster Co., in installing the Buffalo General Electric Co. furnaces.

"We are recommending this ventilated side wall construction to our clients after careful observation of the results obtained. We find that it preserves the brickwork and prevents the adherence of clinkers. The theory, underlying this seems to be that the brick work is cooled by the air while at the same time combustion is induced on the surface exposed to the fuel. Combustion at this point seems to result in ashes rather than clinkers and keeps the temperature high enough so that there is no adherence of clinkers to the brick. No objectionable effect is noted in the fuel bed; on the contrary, it appears to improve combustion and increases the percentage of CO2.

"It seems to us that there are great possibilities for development along this line and we shall be glad to cooperate with any central station members who are interested.

Service

"We believe that such a product as ours is more dependent on engineering skill, not only in design but in installation and operation, than almost any other kind of power plant accessory. With this in mind, we have greatly increased our facilities for rendering service to our clients not only before but after purchase. We have frequently been of material assistance in laying out, plants and adapting settings so as to make the operation of our stokers as satisfactory as possible. Our sales force is in charge of experienced engineers who are interested primarily in successful performance rather than volume of sales. They are well informed on all the developments in this branch of the industry and this is the basis of their ability to place our stokers only where they can do themselves justice.

"After the stokers are installed, we follow them up by a system of inspection service which enables us to keep in touch with all our users."

REFRACTORY MATERIAL

Developments during the past year are principally in the line of standardization of methods of testing clay brick, although increasing efficiencies and overloads by the use of higher combustion temperatures, have continued to put increasing strain upon the refractory materials used in fire brick settings.

Central stations are installing boilers of over 1200 horsepower requiring largely increased combustion chambers and furnaces. The proportional increase in fire brick surfaces and. higher heats causes increasingly greater expansion and contraction strains and consequently more frequent failures in both arch and side walls. Various methods have been tried to strengthen the flat walls with but little success.

Buttresses built inside of the combustion chamber were burned off in a very short time. Sheet iron linings were also tried, but were melted down as soon as the fire brick inside had been burned away or partly destroyed by slag.

The present state of the industry for furnace linings for these large high efficiency, high duty boilers is not altogether satisfactory and further developments are badly needed. Owing to the various chemical, physical and mechanical operations to which refractory fire brick is subjected, there are certain

characteristics which it should have to give a satisfactory life. Variation in the composition of the clay, the method of manufacture and the heat treatment of the brick all have their effect on the final product.

It is necessary, therefore, in order to obtain the best results to make a few tests, if not upon samples of each shipment at least upon each new grade or make of fire brick received, to determine the reliability of the grade.

At the June, 1916, meeting of the American Society for Testing Materials a paper entitled "Practical Methods for Testing Refractory Fire Brick" was presented by C. E. Nesbitt and M. L. Bell, in which several tests for refractory fire clay brick were suggested. The methods described are primarily intended for testing brick about an iron or steel plant, but they could be modified to fit the requirements for brick in other industries. These tests were made to cover 'impact," compression," "abrasion," "spalling," "slagging," spalling," "slagging," "softening," " softening," "expansion and contraction" and "uniformity of size."

The modification of the tests for refractory fire brick for central station boilers would consist of the elimination of the impact and spalling tests, as the brick settings of boilers are not subjected to either the impact of ores or the rapid changes in temperature which are met with in blast furnaces.

Abrasion Test

The abrasion test is to be made by heating the brick to 1360 deg. C. and then pressing it at a uniform pressure for a uniform time against a carborundum wheel. In this test there are possible variables, such as unequal cooling of the brick, variation in the wheels, etc. The abrasion of the brick in the furnace walls is usually due to slag action or barring off the slag, and it is suggested that some recognition of the effect of slag upon the surface of the brick should be made in this test.

Slagging Test

This test is considered both the most important and the most unreliable of any of these physical tests. Most important in connection with boiler settings, as the principal failure of the refractory brick is due to adherence and penetration of slag and subsequent breakage; most unreliable, on account of variations in