This section is from the book "Modern Buildings, Their Planning, Construction And Equipment Vol3", by G. A. T. Middleton. Also available from Amazon: Modern Buildings.
Before describing the different designs or "systems" of apparatus erected for furnishing a supply of hot water at baths and other taps, it is necessary to point out and discuss one particular feature, the consideration of which is greatly neglected ; yet it is the one essential feature of an apparatus expressly designed to yield hot water at taps. The average fitter devotes all his thoughts to obtaining a circulation, apparently not realising that this is the easiest thing to get, if certain simple rules are observed ; whereas greater care is needed in proportioning the apparatus to the expected demand for hot water ; and, most importantly, designing its details to ensure that the hot water, little or much, can all be withdrawn from the taps - any tap - without cold water being mixed with it. These works are always erected in such a manner that, on withdrawing hot water from a tap, there is a simultaneous inflow of cold water into the apparatus to make good the loss, and skill is needed to make the connections so that, as the cold water enters, it pushes, so to speak, the hot water before it, and does not mix freely. The outflow of hot water and inflow of cold are very closely related actions, for one depends on the other. When a hot tap is opened it is literally correct to say that the cold water immediately seeks to reach that tap ; therefore the engineer must arrange that if cold water does issue from a hot tap there must be a positive feeling or knowledge that all the hot water has issued before it. In some works, quite a number in fact, the inflowing cold water not only mingles with the hot, but sometimes reaches and flows from a tap when the application of the hand plainly shows that there still remains hot water in the tank of the apparatus. This is a bad result very easy of attainment, if the subject is not properly studied.
To investigate the movements that occur in a domestic hot-water apparatus when a tap is opened, a glass model should be employed, and fortunately this is almost as easy to prepare as those illustrated in Figs. 32 and 33. Let it be clearly understood that the actions and movements occasioned by an open tap are totally different, and brought about by quite different causes, from those of the circulation, which is due to convection. There is no relationship whatever, but fortunately one very rarely interferes with the other except at the moment of occurrence. It is quite peculiar to see, in a model, first the swift but very smooth movement of the circulation, then, on opening a tap, a violent rushing of water, usually in quite a different direction, and finally the resumption of the circulation, without an instant's hesitation, immediately the tap is closed.
Fig. 34 illustrates a model that may be constructed for this purpose. The boiler, pipes, joints, and source of heat are provided in the manner described in connection with Fig. 32, while for the central hot-water reservoir a lamp chimney, short but of good diameter, is used, the ends being closed with corks, waxed over with sealing wax if necessary. The cold-water cistern need not be of glass - a preserve tin will do.
In this model it will be seen that there is a pair of pipes both above and below the hot-water tank. In this respect the apparatus resembles what is known as the cylinder system in its best form, but the idea of providing the two sets of pipes - a circulation below and above the tank, and bearing the trade names of the primary and secondary circulations - is to enable the student to experiment with and learn practically all the actions that occur in an apparatus whatever its system or kind. The two draw-off branches, one below and one above, are provided with this end in view. The cold supply service, unlike that in an apparatus intended for warming buildings, is best not connected into a return pipe. It is customary to let it enter the bottom of the hot-water tank, and in an apparatus erected in a house there is an elbow or tee put on the cold-supply connection, inside the tank, to prevent the inflowing cold water "boring up" through the bulk of hot water above. This undesirable action can be tested in the model. The taps, it may be explained, can be pipe-light cocks with their noses notched underneath and then bent down.