[mbllanhy] efficiencies OF MULTIPLE-EXPANSION ENGINES 129 



Tlie coiidensino- water is supplied direct from the city mains, and 

 after passing- through the condenser, it falls into either of two other tanks 

 each upon a wei<;-liino- machine, and capable of holding 2.500 lbs. As 

 with the hot well discharging tanks, the water, after it has been weighed, 

 is alloAved to escape by means of cocks in the bottom of the tanks. 



Two water brakes of the Froude-Eeynolds type are fitted, one at 

 each end, to the engine. A full description of this type of brake is given 

 in the Proceedings of the Institute of Civil Engineers, vol. xcix., page 

 168, so that it is not necessary to describe them here. It may however 

 be mentioned that they are so designed as to give a constant torque 

 whilst the engine is working. Too much cannot be said of the excellent 

 working of these water brakes. Whilst the engine is running they 

 would regulate the amount of water in themselves so as to always remain 

 balanced even though the engine speed might be made to vary from 90 

 to 200 revolutions per minute. If it had not been desired to keep the 

 speed of the engine as uniform as possible, to attain which the jockey 

 weight on one of the brakes was altered throughout the trial, it would 

 not have been necessar}' to attend to them at all during the trial, so 

 easily will they regulate themselves as to be always balanced. 



The alteration of the jockey weight caused an increase or decrease in 

 the torque, and so counteracted any tendency there might be of the 

 engine to vary its speed, owing to slight changes in the boiler pressure 

 and condenser vacuum. In some of the trials a rope brake was used in 

 addition to the water brakes. 



Gauges are fitted to the steam chests of all the cylinders except to 

 that of No. 1, where it was not thought necessary to place one, as there 

 was fitted a gauge to the steam pipe only a short distance away. A 

 vacuum gauge is situated on the exhaust ])ipe, whilst to the condenser 

 are fitted another vacuum gauge and a niercur}' column. Fittings are 

 provided to place thermometers in the steam pipe just before ]S"o. 1 

 cylinder, in the pipes before cylinders 2 and 4 and also in the exhaust 

 pipe. Thermometers are also fitted in the hotwell discharge pipes, and 

 in the pipes for the circulation of water before reaching and after leavin»- 

 the condenser. 



Method op Conducting the Trials. 



The usual time occupied by each trial was about one hour ; and 

 although many people object to tests of so short a duration, yet it is 

 hoped that with the precautions taken the results obtained are exceed- 

 ingly reliable. It is open to doubt whether vigilance can be maintained 

 throughout a long trial to the same extent as for one lasting only about 

 one hour. With regard to the observers whose duty it is to measure the 

 amount of feed water or hotwell discharge, the errors they are most 

 likely to make are either that of omitting to close the cocks of the 



Ser, III.. 1896. 9. 



