Then came the inventor again with a new device, a steam-boiler, pipes to distribute the steam, and radiators to give off the heat in the steam to the room. Here at last was a method of heating which would supply warmth in the cold parts of the house, even under the windows, through which the chilliest air penetrated. But the sizes of the radiators were calculated to heat the house to 70 degrees when it was zero outside, although the average winter day was much warmer than this. In this way the occupants of the house were cooked with an excess of heat during moderate weather, for there was no way to regulate the amount of heat given off from the radiator; it either was filled with steam, giving off its maximum quantity of heat, or else it was empty and cold.

To meet this difficulty presented by the steam-heated radiator, the hot-water system was developed. Instead of distributing heat with the medium of steam which under low pressure was fixed at one temperature, heat was circulated by hot water from the central boiler. The temperature of this water could be regulated for mild weather by lowering the fire. However, since the hottest water was cooler than steam, it required larger radiators and more piping, so that the initial cost of a hot-water plant was more than that of a steam system.

Simplified diagram of
Vapor-vacuum system

In order to overcome the disadvantages of the inflexible steam-radiator, inventors finally developed the so-called “vapor-vacuum” system of steam-heating. In this equipment the air was driven from the entire length of pipes and from the radiators by the pressure of the rising steam from the boiler, and forced through a special ejector which closed when the steam came in contact with it, preventing the return of air into the interior. Thus when the pipes and radiators were filled with steam (there being no air left), no pressure was set up to resist the circulation of the water vapor, and when the hot steam condensed in a radiator to a thimbleful of water, more steam was drawn in to take its place, for no air could enter the pipes. In this way the quantity of steam delivered to the radiators could be regulated by a special valve with a varying number of ports, and by turning the valve to a certain position enough steam would be permitted to enter the radiator to keep it half full, or by shifting the valve to another point enough steam would enter to fill the radiator to three-quarters of its capacity. In fact, the requisite amount of steam could be admitted to the radiator to balance the speed of condensation and retain whatever level of steam in it was desirable. Thus the steam system became at once a flexible system of heating, and could meet the changing requirements of the weather.

Hot water radiator
heated by steam

A further development of the hot-water system then came about. In this device the radiators were made to contain water, but the heat was circulated through the pipes by means of steam. This steam was poured over the surface of the water in the radiator and transferred its heat to it. According to the quantity of steam poured over the water, the latter could be heated to various temperatures. Of course the water in the radiator was the medium for distributing the heat outward from the radiator itself.

Still another improvement was made upon the hot-water system by introducing the principle of the closed expansion tank. In the ordinary system the water is allowed to expand at the top through an expansion tank, so that the actual pressure on the water of the system is atmospheric. Under this pressure the temperature of the water cannot be raised to more than 212 degrees Fahrenheit, for beyond this it boils and changes to steam. However, in the closed-tank system a so-called heat-generator is added on the line leading to the expansion tank, which, by means of a column of mercury, is capable of adding 10 pounds more pressure than the atmosphere to the water in the system, and thus raising the boiling-point to about 240 degrees. This generator is so designed, however, that, although it adds this greater pressure to the water, yet the natural expansion of the water in the system is permitted through it in case of emergency. By permitting the raising of the temperature of the water, the size of radiators can be cut down 50 per cent, which, of course, reduces the quantity of water needed and permits a quicker heating of the system when the fire is started. Thus a saving of fuel is accomplished and the disadvantage of the ordinary hot-water system is eliminated; namely, the long time required to get hot water in the radiators after the fire is started in the morning from its banked condition of the previous night.