Before describing the different " systems," or methods of carrying out steam-heating works it is necessary that some particulars be given as to the generation of steam and its action when generated. In the first place, the heating engineer must entertain no idea that the piping of steam works has any relationship to that of a hot-water apparatus. Both may consist of lines of pipes, and in most cases there has to be a kind of circulation (of steam) established ; but beyond this, to associate hot-water practices with those of steam work must only end in confusion and failure. This is mentioned as the idea that the two bear a resemblance to each other is not a rare one.

With many - the inexperienced - there prevails an impression that steam heating is very simple, owing to the fact that steam can be made to readily pass through pipes, and in this respect the impression may be a correct one. With any method of heating by steam, however, there are two difficulties to be overcome, and it might almost be said that the skill required by the engineer lies in dealing with these. One, the chief one, lies in disposing of the condensed water ; the other in defeating the ill effects that air in the apparatus can produce. In affording heat by steam the chief result, other than obtaining heat, is the condensation of the steam into water ; for only by this means can the heat-distributing surfaces - pipes or radiators - in any modern form of apparatus be kept supplied with new steam. A radiator, for instance, when doing its work is perpetually losing heat, with a consequent return to water of some of the steam within it. This causes a state of vacuum (partial) to occur, which is as quickly filled by new steam flowing in. It is a comparatively easy matter to get this action of steam supply to occur, but before it will happen all air must be driven out of the apparatus, and the resulting condensed water must not only be disposed of, but it must be prevented from interfering with the steam supply.

Another natural phenomenon that the steam-heating engineer relies on is the fact that water has its volume increased 1640 times (varying somewhat according to pressure) when converted to steam. This enables him to fill an extensive system of pipes, etc., with steam without causing a serious drop of the water-line in his boiler.

Still another natural and highlyimportantphenomenon is the latent heat of steam. The term " latent," meaning hidden, is a correct one, for although a pound of water at 212° will still register a temperature of 212° when converted to steam, it will have taken up no less than 966 units of heat in the change. It only takes 2t2 units to heat a pound of water from 0° (theoretically) to 212°, so that when this weight of water is converted to steam it has absorbed sufficient heat to raise about five times its weight from, say, freezing to boiling. In other words, a pound of steam, at 212°, carries 966 units, plus 212 which it received as water to reach boiling-point, making together a total of 1178 units. Of these, as will be understood, it has 966 to give out from the heat-distributing radiators, or pipes, before it assumes the form of water again.

As already stated, the condensed water can give considerable trouble if it is not properly disposed of, either by retarding the flow of steam, blocking the steam from its work, or coming in conflict with it. On this account the covering of steam mains to prevent loss of heat - chiefly to prevent the occurrence of water in them - is very important ; and it is desirable also from the fact that cooled steam means steam and heat completely lost. With hot water mains a little loss of heat, wasteful as it may be, may scarcely show at the radiators, but with steam it means the disappearance of a certain proportion of the heating power.

Although, for the purpose of explanation, the heat of boiling water has been given at 2120, as being its easily recognised temperature, this is seldom, if ever, the precise heat that the water and steam have in these works. Low-pressure steam generally registers 3 to 5 lbs. on the gauge, and these pressures would be accompanied by temperatures of from 222° to 227°. These latter figures would be accompanied by 958 and 954 as the number of units of latent heat, but the variation from the number accompanying 212° is so small as to be negligible in affording an explanation.

A remaining detail that may be mentioned here relates to the practical use of the steam in the apparatus. When a radiator, or coil of pipes, is connected on the one-pipe system, the branch, as will be learned directly, is a single pipe, and this has the customary stop valve where it joins the radiator. With the two-pipe system each radiator has a pair of branch pipes and each pipe has a stop valve, so that every radiator with this system has a valve at each end. The reason for thus putting two valves - and the same reason applies with the single valve in the one-pipe branch - is that when it is required to shut off steam from a radiator it must be completely cut off and leave no communication with the apparatus whatever. If, for instance, the steam supply branch were closed and the condense return left open, then steam would back up this return if it could, failing which water would be carried up and cause much trouble.

Another peculiarity related to the valves is that, whether the system be one-pipe or two-pipe, the valves at the radiators cannot be regulated; that is, they cannot be made to control the supply of steam and the heat of the radiator. Valves in this work have to be fully open or tightly closed, and this presents what is a fault in certain cases, in that the temperature of a room cannot be regulated with anything like the degree of nicety that is possible with hot water. This difficulty is aggravated by there being such a small possible range of temperature in the steam itself. If the working pressure is 3 lbs., then, however the fire is tended, only a variation of about ten degrees can be had in the steam, whereas in hot water it is customary to let its heat be from 120° to 190°, a variation of 70°, according to the weather. This difficulty may appear to be greater than it really is, for in large interiors, and most of the places where steam heat is employed, a sufficient regulation of temperature is easily obtained by closing or opening the valves of a certain number of radiators. This would not be possible in a room heated by a single radiator, but in places where each room or space has several radiators or coils of pipe one or more can be shut off independently.

The reason that steam radiator valves cannot be partially opened or closed is this. Taking first a one-pipe connection, this, as stated, is a single-pipe branch with a valve at the radiator. Through this pipe and valve steam has to pass to the radiator in one direction, while the water of condensation has come back through them the other way. These two contrary movements will readily come into conflict, with bad results, if they are allowed to, and it is only prevented by having the pipe and valve of a recognised sufficient size. Thus if a 1 1/4-inchpipe is considered correct, then the valve must be 1 1/4 inch also, and it must have a clear open way through it of this size. If the valve is partially closed, with the idea of regulating the steam supply and the heat of the radiator, all the conditions of using too small a valve are immediately obtained. On this account the valve must always be wide open or tightly closed.

With a radiator connected on the two-pipe system, having a branch at each end, as the valve at the steam supply end of the radiator conveys steam only, it might be thought that its regulation was quite possible, there being no direct troubles associated with controlling the flow of steam through a pipe. No return water comes back through this valve, the one at the other end being provided expressly to convey this away. It will be found, however, that on partially closing the steam supply valve the balance of pressure between the two branches will be disturbed, and the result will be a rise of water up the return branch. This, it might be thought, could be overcome by regulating the return valve to the same extent as the steam valve, but it is not possible to do so with sufficient exactness. A twin valve has been made to be operated by one handle, so that, it would be thought, the opening or closing could not fail to be equal in both valves, but this did not act so successfully as was anticipated.

To the best of belief, this difficulty with the valves, the inability to regulate them, has never been overcome in English practice, but quite recently a system of work practised in France and Belgium, in which valve and temperature regulation is all that could be wished for, has become known in this country, and there is considerable prospect of its gaining ground. This system will be found described a little farther on.

Description of Low-Pressure Steam-Heating Works which return the condense water to the Boiler, known as Gravity Systems practice the steam mains are always arranged as a circuit - as a flow and return, it might be said ; so that the condense water from radiators and pipes flow naturally, by gravity, back to the boiler. This is a detail that has greatly advanced the use of steam as a heating agent, as, by having a constant water-line in the boiler, a skilled attendant can be dispensed with. This desirable result has been perfected, too, by the introduction of the automatic damper regulator, as this automatically controls combustion, generation of steam, and steam pressure. This will be found explained later.

When steam heating was first seriously attempted the practice was to run all pipes singly and with a rise up from the boiler, so that the steam passed to its work and the water of condensation trickled back within the same pipe at all points. This was a true one-pipe system, but it was soon found to be of great advantage to have the steam and condensed water travel together in the same direction, as far as possible, lessening the risk of their coming into conflict while, with their contents flowing agreeably together, admitting of smaller pipes being used.

In all works it is now the rule to arrange that the steam and water travel in the same direction, whenever possible, as otherwise there is some risk of poor results, and it is then always necessary to use larger pipes.