til the rakes have reached the back of the retort and all the coke has been drawn out. The rakes are then brought back to their original position on the machine and this is moved to the next set of retorts to be drawn and the whole series of operations repeated. (Trustees.)

5. Is it possible to have a single automatic overflow to remove both tar and liquor from the hydraulic main and at the same time maintain a liquor seal of constant depth on the dip pipes? Illustrate the reasoning by which you arrive at your answer by a sketch showing a cross section of the main and of such an overflow.

Ans. Owing to the difference between the specific gravities of tar and ammonical liquor it is not possible to have a single automatic overflow for the removal of both tar and liquor from the hydraulic main and at the same time maintain a seal of constant depth on the dip pipes. Such an overflow must connect to the bottom of the main, since otherwise some tar will be allowed to stay in the main indefinitely and gradually stiffen up until it can only be removed by scrapers. The accompanying cut shows cross sections of a hydraulic main with a single overflow opening out of the bottom and serves to illustrate the description of its action which follows.

The main being filled with liquor at the start the conditions are as shown in figure No. 1, the liquor in the two legs of the syphon formed by the main and overflow standing at the same level, and the dip pipes being under the proper seal. But when the main is in operation tar soon accumulates and settles to the bottom, and then nothing but tar can enter the overflow and pass over to its outside leg, as shown on Figure No. 2. In this outside leg there is then a column of tar only, while the balancing column in the other leg is composed partly of tar and partly of liquor. The liquor being lighter than the tar this column of tar and liquor cannot support a column of tar of equal height, and before the tar in the outside leg of the syphon can rise to the level of the overflow and run off, the liquor in the main must rise above this level by an amount which depends upon the length of the outside leg and the proportion of the height of the inside leg which is occupied by the liquor. The seal on the dip pipes will thus be made greater than it should be, as shown on figure No. 2.

Since as long as there is any tar in the main there will be nothing but tar in the outer leg of the overflow and nothing but tar will

Figure 1.

high to more than balance the tar in the outer leg and drive it out of the overflow. When this happens there will be liquor on both sides of the syphon, and therefore the heights of the two columns

run off from the main, all the liquor that is formed accumulates, until finally as the tar runs off, the main contains, as shown in figure 3, nothing but liquor standing at a level which is sufficiently

Figure 3.

Jar.

Sketches showing how the Seal

varies in an Hydraulic Main

fitted with a Single Overflow for

both Jar and Liquor.

will equalize themselves by the rapid flowing out of the liquor from the main until the level has been brought back to the normal water line, at which it will stand until enough tar accumulates to again cut off the liquor from the overflow and fill the outer leg, causing a repetition of the sequence of events previously described.

It is therefore evident that with such an arrangement of overflow the water line in the main is continually varying and the depth of the seal on the dip pipes is anything but constant. To secure a seal of constant depth and at the same time remove both tar and liquor automatically it is necessary to adopt an arrangement of double overflows, similar to that described in the answer to question No. 3 of the third series for 1899. (Trustees.)

6. Boilers are often rated as being of so many horse-power. What is the meaning of the term "horse-power" when used in this connection? What is the meaning of the same term when applied to steam engines?

Ans. The term "horse-power" as applied to the rating of boilers is used in two different ways; either as an absolute unit, by which to measure the rate at which the boiler will work, or as an approximate measure of the capacity of the boiler, to be applied in buying, selling, or describing it for trade purposes.

The trade or commercial horse-power is determined according to certain arbitrary assumptions from the dimensions of the boiler, the extent of the heating surface being usually employed for this rating on a basis of 12 square feet of such surface for each horsepower. Thus, a boiler having a heating surface of 1,200 square feet would be ordinarily rated as being of 100 horse-power. The practice in this respect is, however, not uniform and it is necessary when comparing different ratings of boiler capacity, especially in passing upon proposals for the supply of new boilers, to know upon what rule each rating is based. A rough-and-ready rule for approximating the commercial horse-power of either tubular or water tube boilers has been given by the Engineering News as follows: Multiply the number of tubes by their length in feet and the product so obtained by the diameter of the tube in inches, then divide the final product by 50, and the quotient so obtained will be the horse-power of the boiler. The commercial horse-power is also sometimes determined from the grate surface 3 of a square foot being allowed per horse-power.

When used in engineering work for an accurate measure of the rate at which a boiler will do work the term "horse-power" has been defined by a committee of the American Society of Mechanical Engineers as meaning "an evaporation of 30 lbs. of water per hour from a feed-water temperature af 100° F. into steam at 70 lbs. gauge pressure, which shall be considered equal to 341⁄2 units of evaporation, that is to 341⁄2 lbs. of water evaporated from a feed-water temperature of 212° F. into steam at the same temperature. A boiler should only be rated at the power which it is capable of developing with easy firing, moderate draft and ordinary fuel while exhibiting good economy, and should be capable of developing at least 3 more than its rated power to meet emergencies."

When applied to steam engines, the term "horse-power" is used to indicate the rate at which mechanical work is done, and denotes the performance of work at the rate of 550 foot pounds per second, or 33,000 foot pounds per minute. A foot pound of work is the work done in raising the weight of 1 lb. through a height of one foot, or of overcoming a resistance of 1 lb. through a distance of 1 ft., or of overcoming any smaller or larger resistance through a correspondingly greater or less space per minute. (Trustees.)

7. What horse-power is required for the actual work of pumping the gas by an exhauster handling 25,000 cubic feet of gas per hour with an inlet pressure of 5" and an outlet pressure of 9"?

Ans. The term "horse-power" as used to indicate the rate at which mechanical work is done denotes the performance of 33,000 foot pounds of work per minute, that is, the raising of a weight of 33,000 lbs. through a height of one foot, or the overcoming of a resistance of 33,000 lbs. through a space of one foot, or of any smaller resistance through a correspondingly greater space, per minute. The horse-power required to pump gas can therefore be calculated by dividing the product of the resistance overcome and the space through which it is overcome in a minute, by 33,000, the resistance being measured in pounds per square foot and the space in feet. The resistance is determined by the net pressure against which the exhauster is working, that is, by the difference between the pressure at the outlet and that at the inlet of the exhauster. The space can be taken as the number of cubic feet of gas pumped in a minute without any reference to the actual

velocity with which the gas passes through the outlet pipe, since with a given outlet pressure the total resistance against which the exhauster is working varies directly as the area of the outlet pipe, while the velocity of the gas, or the space passed through in the unit of time varies, when the same quantity is pumped per minute, inversely as the area of this outlet pipe, and therefore the product of the total resistance and the space passed through will always be equal to the product obtained by multiplying the resistance persquare foot by the number of cubic feet of gas pumped in the unit of time. The pressure is usually given in terms of the height, in inches, of the column of water which it will balance, and to convert this to pounds per square foot it is necessary to multiply it by the weight of a column of water one square foot in area and one inch high. A cubic foot of water weighs 62.5 lbs., so a column of water 12" high exerts a pressure of 62.5 lbs. per square foot, and a column 1" high will exert a pressure of 62.512=5.2 lbs. per square foot.

The horse-power required for the actual work of pumping the gas can therefore be determined by multiplying the number of cubic feet pumped per minute by the product obtained by multiplying the net pressure in inches of water by 5.2 (which gives the pressure in pounds per square foot against which the exhauster is working) and dividing the final product by 33,000. Putting this rule into the shape of a formula we have:

V= number of cubic feet of gas pumped per minute, and H= the difference between the outlet and inlet pressures in inches of water.

and therefore the horse-power required for pumping the gas, without taking into consideration the friction of the exhauster or any other losses of power in the machinery, is 0.558 horse-power.

From the information obtained by a series of tests made on