The oldest form of apparatus we have for domestic hot-water supply is that known as the Tank System, and on this account it is commonly condemned as being out of date. Engineers who practise this work largely, however, quite frequently adopt this system when the conditions make it appear desirable to do so, and its erection does not by any means indicate ignorance of the work or of the latest ideas. The chief fault found with it is that, the tank being above the tap branches, it becomes possible to empty the apparatus should a shortage of water occur, and then a certain element of danger may appear. There is little to fear, however, in places well served from a Water Company's mains, but in country houses having the water supply dependent on a man's attention at a pump the tank system is well avoided. The more modern Cylinder System has not this fault, but it has another which the tank system does not possess, which will be explained later. A still more modern apparatus, called the Combined Cylinder-Tank System, has the advantages of both and the faults of neither, though it costs a little more.

It may be said that the names of these appliances are derived from the hot-water reservoirs - those which are square or rectangular being called Tanks, while the others are of cylindrical form and are called Cylinders.

In Fig. 35 are given the customary details of an apparatus on the Tank System. At the base of the apparatus is the boiler, to which the flow-pipe is connected at the extreme top, while the return terminates at a lower point. The reason for this difference in levels of the ends of the two pipes is to ensure the circulation always occurring in one desired direction, up the flow and down the return. The flow-pipe should always enter the top plate of the boiler, as shown, and it is important that it shall not project through inside the boiler in the least degree. The 3/8-inch plate of the boiler is not a very thick substance in which to screw a pipe soundly without the pipe projecting through on the other side of the plate, but this difficulty is got over by the use of a back-nut, as shown in Fig. 36 (the return pipe is not shown). The fault that arises with a flow-pipe which is screwed through the plate, so as to project inside, is shown by Fig. 37. It will be seen that the boiler does not fill with water, there having necessarily to be some air imprisoned in the top when the apparatus is charged. Fitters have been known to remedy this by loosening the manlid, after the apparatus is charged, sufficiently to let the confined air be driven out, after which the lid would be tightened and the difficulty assumed to be ended. There remains the fact, though, that every fresh volume of water when it is heated discharges or releases a certain amount of air, and consequently the air collection at the top of a boiler, connected as in Fig. 37, may be considered as practically permanent. The confined air does not interfere with the circulation to a noticeable extent, but is a prolific cause of noises, which are quite alarming when a good fire is burning and the water very hot.

Domestic Hot Water Apparatus The Tank System 66

Fig. 35.

The return pipe is commonly connected through the top of the boiler, as in Fig. 38, the short internal dip-pipe reaching a little more than half the way down the boiler, as shown. (The flow connection does not appear.) Whenever preferred, the return may be connected to the side of the boiler, either with a bend as in Fig. 39, or coming horizontally through the chimney jamb.

Domestic Hot Water Apparatus The Tank System 67

Fig. 36.

From the boiler the two pipes may take any convenient course to reach the tank, but at all points they must be made to ascend little or much and not descend or even be quite horizontal anywhere. The least rise given to these pipes is 1 inch in 10 feet, and although horizontal pipes may be spoken of, and may be referred to in this book, the pipes should never be quite horizontal, even for a short distance. The purpose of the rise is not to cause or even favour the circulation, so far as inducing the circulatory movement is concerned. The idea that a rising pipe is essential to the action of convection is very widespread, but a brief investigation with a model apparatus will show that, provided the total height which the pipes extend to remains the same, it makes little difference whether the horizontal lengths be quite flat or are given the proper rise. The necessity of giving the rise lies in the fact that provision must be made for the unimpeded escape of air from all parts - air which is being constantly released from the water in this description of apparatus. It is not to be inferred that the circulation will be a success if the pipes do not ascend from the boiler at all, but that the customary rise given to hot-water pipes, where they have to run in a horizontal direction, is a necessary one to admit of air getting away, and this only.

Domestic Hot Water Apparatus The Tank System 68

Fig. 37.

Domestic Hot Water Apparatus The Tank System 69

Fig. 38.

Domestic Hot Water Apparatus The Tank System 70

Fig. 39.

At their upper extremities the pipes enter the hot-water tank, and here again they end at two different levels, though for a totally different purpose to that aimed at in the boiler. In this case also an erroneous idea prevails, that by ending the pipes at different levels some advantage, or essential help, is afforded to the circulation ; but it will be found that nothing of the kind occurs, the circulation being all that could be desired.

always, with the pipes ending level. Had the circulation only to be considered, the pipes would best be both at the bottom of the tank, but in order to obtain hot water at a tap the pipe drawn from must not come from a point where the coldest water is. The bottom part of the tank nearly always contains cold - at any rate, the coolest water, and immediately any tap is opened there is a simultaneous entry of cold water to the bottom of the tank ; therefore the engineer must terminate his flow-pipe away from here and end it where the hottest water is, and where it can all be drawn from, little or much, without cold water mingling with it. On this account the flow-pipe terminates at a high point in the tank, delivering its hot water there as it ascends from the boiler, and drawing the hottest water from there when a tap makes a demand on it.