steam motor-cars, steam fire-engines, and boats has been devised by Clarkson and Capel. In this burner the oil is vaporised, and its vapour, in admixture with air, issues round the lip of a mushroom valve, where it burns with a large flaring flame of great heating power. With this burner an evaporation from and at 212° F. of nearly 11 lbs. of water per pound of oil is stated to have been obtained.

A pressure system which has been extensively fitted in the mercantile marine is the Wallsend-Howden system. The atomiser is shown in section in fig. 343. One of the main features is the four obliquely-bored orifices which ensures a whirling spray, the oil particles being thrown out in a well-spreading

FIG. 344.-WALLSEND-HOWDEN SYSTEM FOR
NATURAL DRaught.

FIG. 345.-WALLSEND-HOWDEN SYSTEM FOR
FORCED DRAUGHT.

cone.

Another feature is the simplicity of the mounting which enables the atomiser to be quickly removed and replaced by a fresh one.

When working with natural draught or closed stokehold system of forced draught, the atomisers are mounted on an extension fitted to the front of the furnace. Each spraying nozzle projects through a baffle plate into an airtrunk having lateral openings at its outer end (fig. 344). This air-trunk projects concentrically with a second air-trunk carried by the furnace front, and the annular space between the inner and outer air-trunks is fitted with deflectors constructed to give the air passing through the annulus a spiral motion. Efficient mixing of air with the oil spray is thus assured. The admission of air through the second or outer air-trunk keeps the outer casing cool, and reduces radiation from this part to a minimum.

When working with forced draught on Howden's Heated Air System each

sprayer nozzle projects through the furnace door into a conical air distributor (fig. 345). This distributor is fitted with deflectors constructed to give the air passing through a spiral motion.

The compact arrangement of pressure-pump and oil-heater for sprayers operating on the pressure system is well illustrated in fig. 346, which shows a pumping-heating unit of the Wallsend-Howden system. Connection is made. from the flanged pipe at the base plate with a duplex suction oil strainer (not

shown), the oil is then forced by a Simplex oil fuel pump to the heater, which is carried by suitable brackets, the oil being discharged through one of a pair of discharge strainers (the four hand-wheels controlling the valves of these being shown in the upper centre of the illustration). A loaded relief valve between the suction and pressure sides of such pumping gear controls the maximum pressure at which the oil is ejected from the sprayers.

In a discussion of the subject of the combustion of oil fuel, more particularly for steam-raising, Prof. J. S. S. Brame 1 says: It cannot be too strongly

1 Journ. Inst. Pet. Techs., iii, 194 (1917).

emphasised that no particular type or pattern of atomiser possesses superlative superiority; that success in burning oil fuel depends but very little on the atomiser, providing the design is good in certain general particulars, but mainly on the general design of the whole oil-burning system.

"Another point, and one on which the U.S. Board laid great emphasis, is the tendency to install too few atomisers-a number of atomisers give a far more uniform heat, which is of great importance in the case of water-tube boilers, where the proper circulation of water will not take place if there is undue heating in certain parts. A number of atomisers reduces greatly the blowpipe-like action so frequently found."

Discussing the relative merits of steam, air, and pressure atomisation in relation to the combustion process, the same writer says:-" Air is the natural atomising agent for perfect combustion. The very action of spraying ensures that each oil-vesicle is surrounded by oxygen ready to carry on the process. In the case of pressure atomising, provision must be made to introduce air most thoroughly throughout the cone of oil-spray, and this air should by preference be heated. Taking into consideration the effect of steam on the combustion process when this is the atomising agent, we find that various extravagant claims have been made for the action of steam.

"Steam actually displaces air which would naturally be present in the combustion spaces, and unless it can be shown, as some claim, that it indirectly. assists in combustion of the oil, it is certainly an undesirable atomising agent as far as the actual combustion process is concerned.

"It is well known that the interaction between steam and hot carbon (the water-gas reaction) is markedly endothermic; if it occurs, it can only produce a lowering of the flame temperature in the zone of combustion; it would certainly result in the formation of carbon monoxide and hydrogen, and these gases would burn further on. The net result would be an extension of the flame, with no practical gain, for if the carbon is set free, and is at a sufficiently high temperature to react with steam, there is not the slightest fear that its combustion by oxygen directly to carbon dioxide, with higher thermal intensity, would not have taken place with air atomisation." The same writer was unable to trace any proof of the supposed direct action of oil with steam likely to occur in the furnace which would assist combustion, and concluded that "for the practical application of oil fuel, steam atomisation has advantages, but from the combustion point of view the balance is against steam. There can be no question as to the soundness of design of steam-atomisers in which a good proportion of air is drawn in and intermingled with the steam and oil spray."

Whatever system of atomising may be adopted, the construction of the furnace is of the highest importance, and it is especially needful that there should be a combustion-chamber of ample size, so that the combustion may reach the stage of the conversion of the carbon into carbon monoxide before the flame comes into contact with cooling surfaces. It is also essential that there should not be an excess of air supplied to the furnace, for whereas a little smoke may only mean the loss of 1 per cent. of the heat-giving power of the fuel, an excess of air may easily cause ten times this diminution in heating effect.

An instructive paper on "Unnecessary Losses in Firing Fuel Oil and an Automatic System for Eliminating Them," by Mr. C. R. Weymouth, was read before the American Society of Mechanical Engineers in December 1908. The author points out that the customary method of procedure in the use of steam atomisers is to clamp the furnace dampers in a fixed position, to give little

attention to the steam supply, and to regulate by hand the supply of oil to the burners so as to maintain the desired steam pressure, with the result that there is an excess of air for combustion, particularly at the lighter loads. In some cases, where the engineers in charge are more enlightened as to the principles of combustion, and an attempt is made to work with a reduced air-supply, there is, on the other hand, at times an excessive production of smoke. The author states that with Californian oil from the Bakersfield district, which has a calorific value of 18,600 B.Th.U. (allowing for the presence of 1 per cent. of moisture), about 14 lbs. of air are required per pound of oil, and he shows then an ideal boiler efficiency of 84-2 per cent., obtained with a 10 per cent. excess air-supply, which is reduced to 67-09 per cent. with a 200 per cent. excess, the difference corresponding with a possible saving of over 20-32 per cent. in the fuel used. He further states that various tests have shown that the steam consumption ranges from 0.14 lb. to over 0.5 lb. of steam per pound of oil, the average value of good performances being about 0.3 lb. of steam per lb. of oil. These statements point to the importance of analysis of the flue gases in the installation of an oil-burning system, for in the use of oil fuel it is comparatively easy to secure the combustion of the fuel to CO2 with little excess of air. The automatic system advocated by Mr. Weymouth consists in controlling the oil supply by means of a "bleeder" valve, which increases or diminishes the pressure in the oil main and thus varies the rate of supply of the fuel. This valve is controlled by the steam-pressure in the boiler, and the variation in pressure in the oil main is the secondary means of controlling the supply of steam for atomising purposes and of air for combustion.

2

The steam-boiler is admittedly an imperfect and wasteful device for the conversion of the energy of fuel into power, and it is not surprising that inventors should have striven to find some means of eliminating its defects.

To Herr Brünler belongs the credit of being the first to practically demonstrate the possibility of steam-raising by the combustion of liquid fuel directly in contact with the water. An installation, which has been in operation in Bremen, consists of a cylindrical boiler, or water-container, provided with a number (corresponding to the number of burners employed) of smaller vessels or pockets, each of which can be isolated by a suitable valve, and can be drained by means of a draincock. Each of these pockets contains an oilburner, as shown in fig. 347, with a pilot burner, by means of which the main burner can at starting be sufficiently heated to effect complete combustion of the oil. When the main burners are well alight, water is introduced into the previously drained pockets from the boiler by opening the valves, and the circulation of heated water commences, the products of combustion, together with the steam generated, being used as a source of power in a steam-engine. The flames can be observed through glazed sight-holes provided for the purpose. An experimental investigation of the Brünler system carried out by Professor Pictet demonstrated that there was no unburnt fuel in the products of combustion, and that 3 H.P. could be produced with the expenditure of about 2 lbs. of oil, quite 6 lbs. of good coal being needed to give the same result with the ordinary method of steam generation, even when one of the latest types of water-tube boilers is used. Allowing for the difference in thermal efficiency between oil and coal, the superior economy of the Brünler system is therefore very considerable, but there are several disadvantages in its employment. In the first place, it is evident that the air needed for the combustion of the fuel must be supplied at a pressure greater than that of the steam in the container. This necessitates the employment of powerful and costly aircompressors which absorb a large amount of the power available. Then,

again, it is requisite, in order to obtain the maximum efficiency, to highly superheat the steam and products of combustion, which cause difficulty in the lubrication of the engine. Finally, it is obvious that, owing to the presence of large quantities of carbon dioxide and nitrogen as products of combustion, it is practically impossible to run the engine with a vacuum, as the air-pump

would need to be of such large capacity that no increase of economy would result (the volumes of steam and gaseous products in the mixture supplied to the engine are in the proportion of 7 to 3). This renders it impossible to use the system for steam-generation for turbines, as it is well known that the turbine is not an efficient machine unless worked with a high vacuum. An additional difficulty which has been experienced arises from the circumstance that a considerable quantity of nitric acid is formed as one of the products of