PROCEEDINGS OF THE POLYTECHNIC ASSOCIATION. 613 



Mr. Sewell said that his father had early planned an improvement on this 

 mode, by introducing a positive motion for opening and closing the valve 

 by which the steam escaped. The escape of a cylinder full of steam is 

 almost instantaneous, and there is some difficulty in obtaining any practical 

 advantage by letting a portion of the steam escape into the air. 



Corrosion of Boilers. 



The Secretary then directed attention to the effect of surface condensers 

 on steam boilers, by reading the report of a late discussion before the Me- 

 chanical Engineers' Society, England, on this subject. Mr. James Jack, of 

 Liverpool, had undertaken to investigate the causes which produced the 

 rapid destruction of boilers, and gave an account of experiments made 

 on the boilers of a steamship using a surface condenser. An examination 

 after the first voyage, during which only distilled water had been used for 

 feeding the boilers, showed the following efiects, which were increased in 

 every subsequent voyage, until the practice was adopted of feeding with, 

 say, from one-sixth to one-tenth of salt water. First above and below the 

 water line, the surface of the plates, tubes and rivets, were covered with 

 a deposit resembling hydrated oxide of iron, which, when the water was 

 evaporated, was in a state of fine, impalpable, brownish colored powder. 

 This deposit was thickest above the water line, sometimes averaging three- 

 quarters of inch thick. When the boilers were emptied, a thick, slimy 

 deposit adhered all over the inside, an analysis of which showed that it 

 consisted of 



Oxide of iron 77.50 



Moisture 19.75 



Grease 0.85 



Sulphate of lime 0.80 



Oxide of copper 0.60 



Traces of alumina, chlorides of sodium and magnesium. . . . 



Loss 0.50 



Total 100.00 



Secondly : Underneath this deposit the plates and tubes were found to be 

 eaten into, indented, or " pitted," The indentations varied in diameter 

 from the smallest speck to five-eighths of an inch, and in depth from the 

 merest impression to the entire thickness of the plates or tubes- and 

 although they were found all over the boilers, they were most numerous 

 just over the nre plates. So destructive was this pitting in boilers using 

 the same water over and over again, that in one instance the tubes of new- 

 boilers were actually eaten through at the end of two or three voya"-es 

 extending over only a few months altogether, and it became necessary to 

 put in new tubes, and to use a portion of salt water for feed, to keep ^p an 

 incrustation, so that the boilers should not be acted upon. If the iron of 

 the boilers had all been of one make, it would naturally have been conclu- 

 ded that the pitting was due to the quality of the iron; but the iron had 

 been obtained from different makers, and was of the best quality. The 

 presence in the boiler of a soft metal-^uch as copper from the condenser 

 tubes — it was considered would induce such galvanic action as might affect 



