172 Proceedings, &c. t for 1884. 



14 feet 2 inch x 4 feet 10 inch, T 5 g -inch thick; end plate, f-inch. 

 No stay pieces were used. The angle irons used at the ends were 

 of very poor quality, 2J inch x 2| inch x J inch, having a square 

 instead of a rounded internal angle. Boiler pressure was usually 

 50 lbs. per square inch. Fracture took place by the back end of 

 the boiler separating completely from the shell, which shot out 

 like a rocket across the street. The cause of the explosion seems 

 to have been the bulging of the ends ; this split the angle irons at 

 one end completely, and then came the explosion. 



Passing on from this particular case, Professor Kernot gave 

 some information on the subject of the design of boilers. 



In all boiler questions there are three important questions of 

 strength — 



1. As against bursting ; 



2. As against collapsing ; 



3. As against ends failing ; 



of which the last is by far the most difficult to deal with. 



The chief authorities on the subject are Wilson on Boilers, The 

 Board of Trade, and the classical experiment of the Manchester 

 Steam Users, described in Engineering of May, 1876. 



Wilson says, a f-inch plate for working pressure of 50 lbs. per 

 square inch should be supported at 8J-inch intervals. Board of 

 Trade says at 10-inch intervals. The Richmond boiler was 2 3 J- 

 inches in one place, so that one could never have called it a safe 

 boiler. 



Professor Kernot mentioned that it has been his experience that 

 explosions occur not so much from hidden causes as from very 

 simple ones. The explanation invariably is exceedingly bad 

 design or culpable negligence on the part of the man in charge. 



The question of factors of safety was raised. In answer to this, 

 Professor Kernot gave the instructions of the Board of Trade : — 

 " For boilers, in accordance with their design, the working pressure 

 must not exceed one- fifth of the calculated bursting pressure, and 

 the testing pressure must be two-fifths the calculated bursting 

 pressure." 



In testing a second-hand boiler, Mr. M'Lean stated that he 

 simply tests it to one-third more than the working pressure, so as 

 not to overstrain it. 



Locomotive boilers are worked to 130 lbs., and tested to 180 lbs. 



Mr. M'Lean described a new form of diaphragm pressure gauge, 

 which gives very accurate results. 



The meetings then closed for the year, to recommence after the 

 recess. 



