This section is from the book "The Gardener V1", by William Thomson. Also available from Amazon: The New Organic Grower: A Master's Manual of Tools and Techniques for the Home and Market Gardener.
Although hot water circulating in pipes has been adopted and recognised for many years past as the best system of warming plant-houses, yet there exists at the present time, among hot-water engineers and gardeners, considerable difference of opinion regarding the best shape or form of boiler to employ in which to heat the water in the first instance. And there is no doubt that among the different shapes of boilers in use at present, the form of some is better calculated to answer the purpose in view than that of others. It is not, however, our intention in this paper to discuss the merits or demerits of any particular kind or pattern of boiler, but to direct attention to one or two matters in connection with the fitting up of a heating apparatus that both engineers and gardeners are agreed upon as being essential to a rapid circulation of the water in the pipes, whatever shape the boiler may be. We allude to the practice of sinking the boiler below the level of both the flow and return pipes, and giving the flows a continuous ascent from the top of the boiler to their furthest points of extension in the building or buildings to be heated.
This method of fixing up a hot-water apparatus has been so long adopted, and attended with such an amount of success, that the soundness of the practice may to some appear beyond dispute. Notwithstanding, we do not hesitate to say that the circulation of the water in the pipes will be as rapid with the bottom of the boiler one foot below the level of the return pipes as it would be supposing the boiler was sunk several feet deeper. And instead of a continuous ascent of the flow pipes throughout their whole length being necessary, or in any way contributing to the rapidity of the circulation, this way of fixing them tends to retard the process. If we succeed in showing that a continuous rise in the flows hinders rather than accelerates circulation, the argument in favour of placing the boiler so much below the general body of the pipes will disappear, and the expense consequent upon excavating, draining, and building a deep stokehole, will in many cases not need to be incurred. Under certain circumstances, however, a deep stokehole is a necessity, as, for instance, when the pipes in passing from the boiler to the buildings to be heated have to cross under outside paths. In this case, as well as in some others that could be mentioned, a deep stokehole is unavoidable.
Before stating the grounds on which we say that a continuous rise in the flow-pipes is a hindrance rather than otherwise to the circulation of the water, let us inquire the reason why circulation takes place under any method of fitting up the apparatus. The primary reason why water circulates or moves in the pipes is in consequence of the minute and separate particles of which it consists being unable to impart heat the one to the other. If the particles of water were capable of transmitting heat to each other in the same way as the particles of which solid bodies are composed are, it would be impossible, either through the application of heat to the boiler or by any method of fitting up the apparatus, for circulation to take place in the pipes. We would also here observe that in consequence of the inability of the particles of which water consists to communicate heat to one another, every particle of the body of water contained in the apparatus at the time of lighting the fire must come in contact with the point on which the latter acts, and again in contact with a colder point in the apparatus, before it can contribute to the general warmth of the structure to be heated.
The immediate reason, however, why the water circulates in the pipes is because of the inequality of the specific gravity of the particles at different points of the apparatus, the inequality being caused by the application of heat to a particular point, while the water is being cooled at all the other points.
Perhaps our young readers will better understand what is here meant if we say that when a fire is lighted in the furnace beneath the boiler, the material of which the latter is made transmits the heat generated by the combustion of the fuel to the particles of the water in direct contact with the inner surface of the metal, thereby raising their temperature, in consequence of which they expand and become of less specific gravity, or lighter in proportion to their size than the colder particles above them. Here the law of gravitation as it applies to liquids comes into operation, and in obedience thereto the heated or lighter particles ascend, their place being instantly occupied by descending colder or heavier particles; and this ascending of the lighter and descending of the colder particles is what is called circulation of the water, and goes on until the whole body of water contained in the apparatus becomes of the same temperature, - an occurrence not likely to take place so long as the fire is kept burning, unless the boiler power is very much in excess of the work it has to perform.
We will now endeavour to show why the rise in the flow hinders rather than accelerates the circulation of the water in the pipes. A hot - water apparatus fitted up on correct principles will have its highest point of action occupied by the hottest, and its lowest by the coldest, volume of water contained in the apparatus, the intervening points being occupied by volumes gradually descending in the scale of temperature, as they recede from the former towards the latter point. Now, when the flow-pipes gradually ascend from the highest point of the boiler, throughout the structures to be heated, it is impossible for the hottest volume of water contained in the apparatus to occupy the highest point of the latter, when heat is applied to the boiler. The reason for this is plain: as soon as the fire acts on the boiler, the particles of water in contact with its inner surface bound upwards, continuing to do so until they come in,contact with the inner surface of the upper side of the flow-pipes. Here the particles immediately part with a portion of their heat, and consequently become of greater specific gravity than they were at the time of starting on their upward course, and would now commence to descend towards the point from which they started, but that they are still lighter than the particles composing the body of cold water contained in the flows at the time of setting the fire agoing: hence the partially cooled particles proceed along between the colder body of water and the inner surface of the upper side of the flows, continuing to do so until the cold water has found its way to the boilers in a contrary direction.
Thus, to begin with, we have two bodies of water of different degrees of temperature moving in opposite directions in the same pipe, and at the same time. This fact hot-water engineers admit, but tell us that as soon as all the water in the flow-pipes becomes of equal temperature the process will cease; and no doubt it would, providing it were possible for all the water in the flows to become of an equal temperature. This, however, cannot occur in an apparatus of any great extent so long as combustion takes place below the boiler; and for this reason the water, as it travels from the hottest towards the coldest point of the apparatus, is continually parting with its heat, and the coldest particles in the volume of water at any given point of the apparatus will occupy the lowest place; consequently the coldest particles, throughout the length of the pipes, will rest on the inner surface of their under side, and as the inner surface of the under side of the flows descend in the direction of the top of the boiler, the coldest or heaviest particles of the water contained in them will roll or gravitate down the inclined plane, just as the coldest or heaviest particles did at the time of starting the fire under the boiler.
On these grounds we say that a continuous rise in the flow-pipes is a hindrance rather than otherwise to the circulation of the water. Therefore, the flow-pipes of a properly adjusted hot-water apparatus should be carried to the highest point of action in the structures to be heated as soon after they leave the boiler as the general arrangements of the structures and the position of the boiler in relation thereto will admit, and from this point the pipes should gradually descend until they connect with the boiler again at its lowest point. Thus the heated water, when it leaves the boiler, cannot return when it becomes colder except by the legitimate route of travelling through the whole length of the pipes, and entering at the lowest point of action of the apparatus. J. Hammond.