approach to a one-pipe system of hot-water distribution is obtained by connecting both the flow and return branches of the radiators to the same main, substantially as shown in Fig. 1. The main is of unusually large diameter, acting as a reservoir, so that the movement of the water through it is comparatively slow. The risers are tapped into the top of the main, and the returns are connected into the side or bottom, as shown in Fig. 2, so that they deliver the cooled water into the lower part of the main. In the high-pressure, or closed, system the temperature of the water may be from 212° to 400°, or even more. When there is no objection to high temperatures and the accompanying risks, it may be used instead of the low-pressure system. It FIG. 2 requires strong boilers and radiators, the pressure at 250° being about 121 lb. per sq. in. by the gauge, but the apparatus is much smaller than that used for low-pressure heating. The range of temperature at the radiators is great, sometimes amounting to 150°, or more; consequently, the pipes may be made quite small. Usually, however, the range is about 50° or 60°, giving a motive force about 3 times as great as in lowpressure apparatus. The low-pressure, or open, system operates with a maximum temperature of 210° or 2120, and the range of temperature is usually 20o. The area of heating surfaces must, therefore, be quite large. The motive force is so small that the size of the distributing pipes and mains of extensive heating systems becomes inconveniently large, the large pipes adding greatly to the cost of the apparatus. In a one-pipe system of hot-water heating, the main is usually carried around the basement walls exactly as for steam heating, the return connection being made near the boiler. Equalized System.-The distinguishing feature of the equalized hot-water heating system is that the water is compelled to travel exactly the same distance in going to and from any radiator on a given floor. Thus, a radiator situated close to the riser will have but little advantage over one on the same level situated a long distance away. Resistance of Circuits.-The resistance in a bend made with a common elbow, the ends of the pipes being left square, is about equal to the frictional resistance of a piece of straight pipe having a length equal to 100 times its diameter. If the ends of the pipes are reamed, the resistance may be reduced to 70 or even to 60 diameters. With a long bend having a radius of 5 diameters, the resistance falls to about 10 diameters. A plain Toffers about the same resistance as an elbow, and a return bend from 1 to 2 times as much. The gain made by reaming the ends of the pipe is much less in pipes of large diameter than in pipes of small diameter. To the measured or actual length of any circuit should be added a length of pipe having a resistance equivalent to that of all the fittings. For example, in a circuit having an actual length of 300 ft., there are 8 elbows and 12 T's. Allowing 70 diameters for each elbow and T, the length to be added to the actual length of circuit to represent the resistance of the fittings, would be (8+12) X 70 1,400 times the diameter of the pipe. In the case of a 4" pipe this would equal 467 ft., making the computed length 300 + 467 = 767 ft. The actual length of a circuit is always understood to be the actual distance traveled by the water in going from and returning to the boiler. When the water flows through pipe coils, the actual distance traveled must be ascertained and included in the estimate of length of the circuit, and full allowance must be made for each return bend. It is found by experience that the ordinary flow and return connections from a radiator to the risers or mains having an aggregate length of about 10 ft., and including 6 ordinary elbows or their equivalents, will present about the same resistance to the flow of water as a plain, straight pipe 100 ft. long. Therefore, in computing the friction in a circuit, about 100 ft. should be added to the actual length for each ordinary radiator connection. For general purposes it is assumed that the water will cool 20° in passing through the radiators, and will thus emit 20 B. T. U. per lb., or 166.7 B. T. U. per Winchester gal. (231 cu. in.). PROPORTIONS OF HOT-WATER HEATING APPARATUS. The sizes of pipes should be governed by the amount of radiating surface to be supplied, the height of the radiators above the boiler, and the number of changes in the direction of the several currents. It is considered safe practice to allow from 50 to 100 sq. ft. of direct radiation for each square inch of cross-section in the pipe. If the pipes are short, straight, and high, 1 to 100 would be allowed; if long, crooked, or low, 1 to 50 or more, according to the conditions. In proportioning single-main, or one-pipe, systems satisfactory results will be obtained when the diameter of the flow and return mains is not less than .16 times the square root of the radiation supplied. To determine the amount of radiating surface necessary to easily warm a building in all kinds of weather, and to proportion it so that each room will have the required temperature at the same time, are very important points, requiring careful consideration. No rules that will be applicable to all cases can be given, but for ordinary buildings having the average wall and glass exposures, the following table is found, in practice, to give good results, when used with judgment. |