from the engine that unless it is provided with an excellent governor it will speed up to a very great rate and may run away. This fact that an increase of resistance diminishes the power required at different speeds is not considered in the tables given; consequently these powers are somewhat in excess of those actually required. The excess of power would depend upon friction and other resistances; consequently no allowance can be made which would be accurate for all conditions.

Dimensions of Horizontal Conduits.-We now come to the question regarding dimensions of horizontal conduits that convey the air from the blower to various parts of the building. There is a great difference of opinion as to the proper velocity of the air through such conduits, and circumstances have a great deal to do with this question. In my opinion the easier you make it for the air to travel the more successful will be the plant. In no plants, in public-buildings, do we advocate a velocity of air that exceeds 15 feet per second, or 900 feet per minute; 600 feet, or even 400 feet, is better, although in an extensive plant the conduits might be so large as to be unsightly and interfere with the convenience of the building. Vertical flues in the walls leading to the various apartments should be so large that the velocity of the air will not exceed 10 feet per second, or 600 feet per minute.

Maximum Velocity of Air.-From an economical and efficient standpoint air should never enter a room through a register, screen, or grille at a velocity exceeding 400 feet per minute (6.6 feet per second). A greater velocity is liable to create such a rapid movement of the air as will stir up the dust in the room and create serious throat affections. Again, air coming in contact with the screen at a very high velocity will cause a low whirr or whistle often proving very annoying. Better ventilation, or perhaps we should say better circulation of the air, takes place when introduced at a moderate velocity than at a high velocity, because in that case the air enters gently and is distributed by gravitation, due to the cooling of the air in contact with cold walls, and the whole body of air is thus kept in slight motion and the entering air is more evenly distributed. If the air is forced in at high velocity, it creates swift currents

and counter-currents, which will completely prevent the equitable distribution of the fresh air.

Introduction of Air.-My method of introducing air into a room is from a register about 8 feet above the floor, connected with a flue located in an inside wall, and discharging the current of air in the direction of an outside wall. The vent register should be located in the same wall as the fresh-air register, but at the opposite side and in the warmest corner of the room.*

General Remarks.—Architects very often combat such arrangements on the ground of interfering with their plans or of taking up too much room, and very often seriously object to making even the slightest alteration. This often leads to sorry arrangements for heating and ventilating plants, which will probably always continue so long as competiting manufacturers design those to be installed in certain buildings.

It may be said generally that while the method of designing, followed by different manufacturers, may be essentially different from that given here, yet the experience of the writer has shown that the quantities, as computed by various manufacturers when submitting plans in competition for the same building, are essentially the same as those stated here.

156. Systems of Ventilation without Heating. Where large quantities of air are required, especially in seasons when heat is not needed, systems of ventilation may be constructed which are independent from the systems of heating. The circulation of the air through the building may be produced either. by exhausting or rarefying the air in the discharge-ducts, or by delivering fresh air to the rooms under pressure, as described for hot-blast heating.

The air may be rarefied in the discharge-flue by heating. either with steam or hot-water radiators, with an open fireplace or a stove. When circulation is produced by heat, the amount of air moved will depend upon the height of the chimney or discharge-duct and its temperature, and will be essentially as that given in the table on page 45. The air may also

*The above opinion gives the practice of Mr. Still, and is different from that of many engineers. See a full discussion of the matter on pages 44 to 49.

be exhausted from the building by induction, for which may be used a jet of steam, water, or compressed air which is de livered from a nozzle into a convergent pipe of somewhat larger diameter and with both ends open. A very strong draft can be produced in this way, although at the expense of more energy than that required to operate exhaust fans or blowers. The air may also be exhausted by means of a fan located in the main flue. In case any of these means for producing drafts by exhausting or rarefying the air in the discharge-ducts is employed, every precaution that has been mentioned in regard to chimney-tops (page 162) should be observed, otherwise a considerable portion of the force may be required to overcome adverse wind currents.

The general remarks regarding hot-blast heating systems. and also the tables of dimensions apply equally well to this case. The tables on page 52 will be useful in proportioning areas of flues and registers for the discharge of a given amount of air; as an allowance for friction add one inch to each lineal dimension.

The blower system of ventilation has been fully described in connection with the hot-blast system of heating, and tables of capacities of various fans given which are applicable to this case. In this system as well as in the hot-blast system of heating especial care should be taken that the resistances in pipes and flues are as small as can be made, that bends are made with a long radius, and that the reduction in size in passing from one pipe to another is as gradual as possible.

157. Heating with Refrigerating Machines.-The refrig erating machine is virtually a pump which removes heat from a body at one temperature and discharges it at a higher temperature. Reckoned on the basis of heat transmitted, it is a very efficient machine, as it may move from a lower to a higher temperature 10 to 20 times as much heat as the mechanical equivalent of the work performed; in all respects this machine is the converse of the steam-engine. By utilizing the heat which is discharged from a machine of this character in warming a building, and also that in the exhaust steam from the engine working the compressor pump, there is a possible efficiency many times greater than that which can be obtained. by burning the coal directly.

The practical arrangement of such a machine, if using air as the working fluid, would be such as to draw in air from the outside, compress it to such a point that its temperature would be very high, pass it through circulating pipes and radiating surfaces when still under pressure, and discharge into a chamber from which the pressure has been removed, or in the outside air after being cooled. If the exhaust steam could be used for heating, such a system would be very economical, although it would be costly and take up considerable room. An ammonia refrigerating machine might be used, in which case the heat in the compressed ammonia could be removed by water, which would thus become heated and could be circulated for the purpose of warming. The scheme of using the reversed heat-engine or refrigerating machine as a warming machine was pointed out first by Lord Kelvin in 1852,* and although it presents great advantages economically, the writer. has no data showing that it has ever been put to practical use.

158. Cooling of Rooms.-The converse operation of cooling rooms, although at the present not undertaken except in the case of cold-storage plants and warehouses, bids fair to be at some time an industry of considerable importance. Rooms may be artificially cooled by a system constructed similar to that described for hot-blast heating. The coils or radiating surface, however, would need to be replaced by ice or constructed in such a manner that ammonia or some liquid at a very low temperature could be circulated. Over these the air could be driven, its heat would be absorbed, and it could be reduced in temperature to any point desired. In lowering the temperature of the air, a considerable amount of moisture might be precipitated, and some means should be provided for artificially removing it without heating, otherwise the rooms. would be made damp. It may be remarked that ordinary pipe-fittings cannot be used with safety for ammonia circulation, and that special fittings are manufactured for this purpose.

* Proc. of the Phil. Soc. of Glasgow, Vol. III, p. 269.

CHAPTER XIV.

HEATING WITH ELECTRICITY.

159. Equivalents of Electrical and Heat Energy.Electrical energy can all be transformed into heat, and as there are certain advantages pertaining to its ready distribution, it is likely to come into more and more extended use for heating, especially where the cost is not of prime importance. The value of mechanical and electrical units has been given on page 5, from which it will be seen that one watt for one hour, which is the ordinary commercial unit for electricity, is equal to 3.41 heat-units; for one minute it is 1/60 and for one second it is 1/3600 this amount. Electricity is usually sold on the basis of 1000 watt-hours as a unit of measurement, the watts being the product obtained by multiplying the amount of current estimated in ampères by the pressure or intensity estimated in volts; on this basis 1000 watt-hours is the equivalent of 3410 heat-units. We have considered in Chapter III the amount of heat required per hour for the purpose of warming. This amount divided by 3410 will give the equivalent value in kilowatt-hours which would need to be supplied for the required amount of heat.

160. Expense of Heating by Electricity.-The expense of electric heating must in every case be very great, unless the electricity can be supplied at an exceedingly low price. Much data exists regarding the cost of electrical energy when it is obtained from steam-power. Estimated t on the basis of

*One thousand watts is called a kilowatt.

The mechanical energy in one horse-power is equivalent to 0.707 B. T. U. per second or 2545 per hour. One pound of pure carbon will give off 14,500 heat-units by combustion, which if all utilized would produce 5.7 horse-power