Feb. 3, 1888.] 



SCIENTIFIC NEWS. 



109 



THE BURSTING OF LEADEN PIPES 

 BY FROST. 



ON questioning plumbers as to the cause and time of 

 pipes bursting during frost, we find the general im- 

 pression to be that the pipe only burst after a thaw had 

 set in, and was caused by the ice in the pipe melting, and 

 the pressure thus becoming too great for the strength of 

 the metal. Messrs. Bruce and Thomson, Glasgow, have 

 for the past three winters been making experiments on 

 the subject, which they have communicated to the Philo- 

 sophical Society of Glasgow. They have been able to 

 refute the time-honoured idea. 



But before proceeding to give the results of their in- 

 quiry, it will be necessary to mention a few preliminary 

 considerations in reference to water and lead. Pure 

 water freezes at 32° Fahr., but it possesses the property 

 of remaining liquid at a temperature as low as —11° 

 Fahr., provided that it be kept still, and that sufficient 

 pressure is maintained to prevent its expansion. If 

 water be enclosed at 39° Fahr. in a cast-steel tube, thick 

 enough to be regarded as inextensible, it will remain 

 liquid for days at —11° Fahr., as shown by the mobility 

 of a steel ball enclosed in the tube. But when this tube is 

 opened, and the water allowed to expand, solidification 

 takes place. At this point of 39° Fahr. water reaches 

 its greatest density, and it expands with the decrease and 

 increase of the temperature from this point. Water in 

 the act of freezing expands by fully 10 per cent. ; this is 

 an important fact, one cubic foot expanding into 176 

 cubic inches more. Besides, the compressibility of water 

 is very slight, one atmosphere (about i5lbs. on the square 

 inch) reducing its volume to the extent of only one-twenty 

 thousandth. 



Of the lead used in commerce the English is the purest, 

 and the Spanish the most impure, the former containing 

 99I per cent, of lead, and the latter 95I — a very hard 

 lead. The presence of antimony, zinc, copper, and tin 

 affects the ductility and malleability very much. Lead 

 has a very low tenacity ; its tensile strength varies from 

 l,4oolbs. per square inch to 3,30olbs. Impurities reduce 

 specific gravity, and have the tendency to make the lead 

 harder, while increasing the tensile strength generally, 

 which in pipes is rather an advantage than otherwise, if 

 not carried to excess. 



Let us now consider a case of bursting of the pipes by 

 frost. A pipe is led in from the street, taking a sharp 

 turn inside the wall in its course to the floors above. A 

 column of ice forms in this pipe, and owing to the bend 

 the ice cannot move. Practically a plug has been formed, 

 so that the water in the pipe higher up between the ice 

 and the tap is subjected to considerable pressure, which 

 increases as the congelation continues in the pipe, until 

 the pressure becomes so great that the strength of the 

 metal is overcome, and a rent takes place. At the 

 moment of release, by the pipe giving way, the remain- 

 ing quantity of water immediately freezes ; and until the 

 temperature rises above the freezing point it gives no 

 inconvenience as regards flooding the premises. In fact, 

 the pipe may be rent for days before being detected 

 after the rise of temperature, called thawing. 



Nowconsidering the facts regarding water previously 

 mentioned, this is just what might be expected ; for it is 

 self-evident that if water at its greatest density be en- 

 closed in a lead pipe and securely plugged at both ends, 

 and the ends be placed in a freezing mixture, so that ice 

 is formed there, then, as the formation of ice proceeds 



from each end, the water in the middle will increase in 

 pressure, with the result that the pipe is bulged out in 

 the centre, and ultimately bursts. But the question 

 arises. At what pressure do these pipes burst ? On 

 account of the difference of opinion on the tensile strength 

 of lead, no accurate answer can be got by considering 

 that property. Pipes commonly used by water com- 

 panies are not selected according to their strength or 

 quality, but according to their weight per linear yard — 

 pipes I inch diameter being ylbs. per linear yard ; | inch, 

 lolbs. ; I inch, 141b. ; i:^ inch, iSlbs. ; and i| inch, 24lbs. 

 The bursting pressures per square inch of these pipes 

 are successively i,82olbs., i,232lbs., ijOgilbs., i,0361bs., 

 and 8i2lbs., although the thickness of the lead in the 

 last pipe is one-tenth more than in the first pipe. From 

 this it is evident that the strength of the pipes is by no 

 means on an equality, for the smallest pipe stands more 

 than twice the amount of pressure that the largest pipe 

 can bear before bursting. This may be accounted for by the 

 fact .that the smallest pipes are not made of so pure lead 

 as the largest, the tensile strength of the tin mixed with 

 it being double that of lead, and of the solder mixed with 

 it being about four times that of lead. 



Mr. Jardine has found, after a series of interesting 

 experiments, that a i|-inch pipe i-inch thick stood 1,000 

 feet head of water without alteration; at 1,200 feet head 

 it began to swell ; enlarged to i| inch diameter at 1,400 

 feet head it burst, giving a bursting pressure of 6o61bs. and 

 2,6 1 1 lbs. of tensile strength. On the other hand, a 2-inch 

 pipe f-inch thick stood 800 feet head without alteration ; 

 but at 1,000 feet head it burst, giving 433lbs. bursting 

 pressure and 2,41 2lbs. tensile strength. Box, in his 

 " Strength of Materials," gives the safe working pressure, 

 without any shocks, at one-tenth of the bursting pressure, 

 and, with shocks, at one-twentieth of the strain. 



Now what causes the shocks so often heard in the 

 water-pipes ? The sudden closing of a tap. This well- 

 known knocking or hammering sound is scarcely avoid- 

 able in the ordinary water supply service of towns ; but 

 if a screw-down tap or a ball-cock is used, the flow of 

 water is stopped gradually, and the noise is removed. 

 It may be the case that this is the reason why the smaller 

 pipes are made so much stronger in proportion ; but it 

 would not hold good in the case of a number of small 

 pipes and short lengths taken from a large pipe. In this 

 case the large pipe would be too weak to stand the 

 repetition of a sudden shock of two or three taps being 

 closed suddenly at the same time. Another reason for 

 the smaller pipes being so much heavier may be on 

 account of the numerous bends to which they are sub- 

 jected when being fitted up in houses, and which, in 

 the case of the larger sizes, are not of so frequent 

 occurrence. 



When the bursting orifices of different pipes are care- 

 fully examined, a marked difference is observable 

 between the forms when the pipes are burst by water 

 without frost and with frost. Pipes burst by water 

 pressure, without the intervention of frost, have a break 

 showing a long, narrow slit, just sufficient to ease off" the 

 pressure ; while those burst by the action of frost have 

 the lips of the orifice forced out to varying widths. This 

 is caused by the solid ice being forced out by the expan- 

 sion of the water at the moment of freezing. In this 

 way the character of the break might be used as a test 

 for the quality of the metal of which the pipe is made. 



The question might be asked as to what length of pipe 

 would require to be frozen to produce sufficient pressure 



