46 



THE CIVIL ENGINEER AND ARCHITECTS JOURNAL. 



[Fbb. 



consequence of imperfections in their action or condition, do not properly 

 perform their duty. Therefore, in so far as similar safety apparatus are used 

 for both high and low-pressure boilers, they must he liable to similar inter- 

 ruptions in their working. Esperience has shown this very often, and it 

 has been found that even the vertical open-mouthed feed-pipes of low-pres- 

 BDre boilers, wliieh act as escape-pipes when the boiler pressure is too great, 

 (these are wanting in marine engines,) are not always secure. If then an 

 overfilling of the boiler with steam is equally possible in both high and low- 

 pressure engines, both are liable to danger from this source ; as the strength 

 of the metal is adapted to ihe working pressure, and therefore the proper 

 dasticity for which the vessel is constructed must he exceeded wlien such an 

 occurrence happens. But there is an advantage on the side of the high- 

 pressure engine, for the elasticity must be increased in a much higher ratio 

 than with the low-pressure engine, before it overcomes tlie pressure at which 

 the boiler is proved (usually three times the working elasticity); and there- 

 fore a much longer time will elapse before absolute danger arises. For ex- 

 ample, in a boiler working at eight atmospheres, it will take a much greater 

 lapse of time for the pressure to rise to 24-atmospheres, than it would to 

 reach 12 lb. per square inch in a boiler working at -1 lb. ; and these would be 

 the points at which danger may be supposed to arise in the respective cases. 

 This gives a key to the experience of late times, that as great a proportionate 

 number of low as of high-pressure boilers have exploded, as well in England 

 as in America and France ; and that among the latest instances, the accidents 

 with the former have reached an alarming extent." 



This extract brings us to the first allegation, that it takes a longer time to 

 overcharge a high-pressure than a lower pressure boiler. The general truth 

 is not stated with sufficient precision; it may be explained by the following 

 example ; by referring to the best tables for the relative volumes of steam 

 at different pressures, we find that any given quantity of water will produce 

 249 times its bulk of steam of eight atmospheres' pressure, and 1C69 times 

 its bulk of steam of one atmosphere. Consequently, to fill a boiler with the 

 latter or low-pressure steam, rather less than seven times as much water 

 would be required as would be necessary for filling it with the former or 

 high-pressure steam. 



Now, it has been ascertained that the pressure in the boiler has no influ- 

 ence on the rate of vaporization — that is, with a fire of given intensity and a 

 fire-box of given dimensions, the same number of pounds of water will be 

 converted into steam in a given time, whether the boiler-pressure be one 

 atmosphere or eight atmospheres. Coupling this consideration with that in 

 the preceding paragraph, we arrive at the conclusion that the overcharging 

 a boiler which can only resist a pressure of one atmosphere, takes about one- 

 seventh of the time required for overcharging a boiler which will bear a 

 pressure of eight atmospheres. In order to the accuracy of this conclusion, 

 it is requisite however to suppose nothing altered but the boiler pressure, 

 and that the capacity of the boiler, the intensity of the fire, and the dimen- 

 sions of tlie fire grate, are in all cases the same. With tliis proviso (which is 

 not stated by Dr. Alban), we may establish the general conclusion, that a 

 low-pressure boiler is overcharged in a shorter time than a high-pressure 

 boiler. 



Our author then proceeds to consider the causes of explosion, and details 

 the various hypotheses which have been suggested, such as that of the gene- 

 ration of an explosive gas from the decomposition of the water — the genera- 

 tion of hydro-electricity — and the sudden conversion of water into steam by 

 coming in contact with overheated parts of the boiler. The latter of these 

 hypotheses is by far the most probable ; but there is one important point of 

 agreement in them all, namely, that the ultimate or inducing cause of an 

 explosion is the sinking of the water too low in the boiler, and the conse. 

 quent over-heating of the metal. To this point, therefore, attention must be 

 confined, when the safety of high-pressure and low-pressure boilers is com- 

 pared. We have simply to ascertain which of the two is most liable to be 

 overheated. One of the principal causes of this evil is 



" Too great an accumulation, either general or partial, of scale or earthy 

 sediment in the boiler. These substances being bad conductors of heat, 

 prevent, when in large quantities, the proper distribution of caloric to the 

 vi-atcr, or at least injuriously retard its transmission. Tlie heat of the metal 

 then increases to too great an extent, and may frequently rise to incande- 

 scence. Sometimes it happens that the layers of deposit arrange themselves 

 in such wise as to leave interstices to which the water cannot penetrate : 

 now if any of the adjacent portions become cracked, the water will suddenly 

 find its way upon the hot metal, and will cause a local explosion, thereby 

 loosening the scale not only from the part previously affected, but for a con- 

 siderable distance round, and consequently increasing the contact of the 

 water with the heated metal. This produces a rumbling commotion in the 

 water, which, if the incandescent spot be large, may be in the highest degree 

 injurious to the structure of the boiler. The steam thus suddenly formed 

 angmenis the pressure, and hence again increased danger may ensue, parti- 

 cularly as the spot overheated will have been rendered more susceptible of 

 dnmage. It has often been remarked that explosions were immediately pre- 

 ceded by the rumbling noise alluded to above. The high-pressure engine 



has in this respect also an advantage over the low-pressure, in that the sedi- 

 ment, when the elasticity is great, seldom attaches itself firmly to the sides 

 of the boiler, but collects in a loose state, and is easily removed." 



The comparison proceeds in a fair manner as follows : — 



" Boilers which are fitted with imperfect water gauges or feed apparatus, 

 arc particularly liable to the evils of a partial exposure of the fire surface, 

 and unfortunately these defects are but too common, particularly with high- 

 pressure engines. The same liability to danger is also incurred where inter- 

 nal fire.tubes are inserted, or where the water space is too flat and confined, 

 and is exposed in an injudicious manner to the flues. When tubes are in- 

 troduced, they seldom lie deep enough under the water level, and are therB- 

 fore soon left uncovered by an accidental slight depression of the latter; and 

 if the water chambers are too confined, the water viill be often driven out 

 during violent ebullition. Marine and locomotive boilers are particularly 

 liable to this, A steam boat boiler which burst at Hull (an account of the 

 accident, with a description of the appearance of the boiler after the explD- 

 sion, will be found in the ' Civil Engineer and Architect's Journal,' August, 

 1838, p. 283) furnishes an example of such an improper make. Both imper- 

 fections were united in its construction, and the collapsed fire -tubes showed 

 that the metal of these parts had been overheated in consequence of the 

 water being driven out of the too contracted surrounding chambers, and that 

 by such overheating the parts were weakened, and at last suddenly gave way 

 to the pressure. It is much to be regretted that marine boilers are usually 

 subject to the evil of too confined and too shallow a water space ; because 

 the ship's motion renders them particularly liable to the exposure of the 

 fire-tubes : the use of sails increases the mischief, for when the ship has lain 

 over on one side for some time, her righting or careening will throw the 

 water back upon any portions of the metal that may have become over-heat- 

 ed, and thus danger may ensue in proportion to the length of time the parts 

 have been exposed and the degree of exposure. Hence we find the majority 

 of explosions occur on board steam boats, and proportionately but few on 

 shore. 



Now since all marine boilers, as well for low as high-pressure, are liable, 

 if injudiciously constructed, to similar dangers of the kind we have named 

 above, no conclusion to the prejudice of high-pressure engines can be drawn 

 from such accidents. Indeed of late years a general comparison has been tn 

 favour of the high-pressure system.* One reason why low-pressure boilers 

 must, under the evils above-mentioned, be less secure than high pressure, is 

 that in the former the ebullition is much more violent, and the water thereby 

 more liable to be expelled, whereas under a great elasticity the bubbles of 

 steam generated take a smaller volume, the ebullition goes on more quietly, 

 and therefore the danger is lessened. 



The common chest form of low-pressure boilers with straight sides tends 

 to increase the liability to the exposure of parts heated by the fire, especially 

 if furnished with internal flues, as is generally the case wiih marine boilers. 

 The large flat surfaces easily bulge out by an increased pressure within, and 

 the consequent augmentation of cubical content causes a sinking of tlie 

 water surface ; after which the restoration of the elasticity to its original de- 

 gree may throw back the water over the spots it formerly left, and thus the 

 source of danger is at hand." 



The last mentioned evil is not enlarged upon in a manner corresponding 

 to its importance. It may be demonstrated that boilers with flat sides ace 

 subject to much greater strain than those which are curvilinear in every part 

 If the boiler be of the form known in geometry as a solid of revolution with- 

 out flat ends (that is, if every section perpendicular to its axis be a circle), 

 the elastic pressure within will not tend to bulge it. We think it may be 

 shown that, in this case, the tension of the metal is direct or tangential, ami 

 that there are no transverse strains, analogous to those of a deflected beam 

 or girder. But where ^a< surfaces are exposed to the action of steam, 

 there is a tendency to make them belly out, like the sails of a ship. 

 In this case, the metal is subject to transverse strains, and in consequence of 

 the tendency to bending, will be subject to forces of both extension and com- 

 pression (like a deflected beam) ; these forces greatly exceeding those arising 

 from direct tension in the solid of revolution. It will be seen therefore that 

 boilers with fiat sides have a great disadvantage — in addition, it must be ob- 

 served, to the weakness at the angles, — from the imperfect connection of the 

 plates. 



Another cause of danger in low-pressure engines, which Dr. Alban insists 

 upon, is their great size. 



" The greater the content of a boiler, the greater surface it must oflfer to 

 the pressure of the steam, and the greater danger it must be subject to. This 

 truth is so self-evident, that it is incomprebeusilile how it should be so uni- 

 versally neglected. The size of many boilers at present in use is truly as- 

 tounding. I have not unfrequently seen them as large as 5 or 6 feet in dia- 



« Viile ' Echo du Monllc savant,' No. 24, p. 178. Up to the year 1634, only twenty ex- 

 plosions had occurred in America with hij^h-prossure engines, while thirty-two had hap- 

 pened with low pressure ; and it Is well known how common the high-pressure engine Is 

 in tliat country, particularly in the Western States. At a later date, the proprietors of 

 steam boats in North America have stated, in a memorial to Congress, that since the 

 mure general introduction of high-pressure steam, the number of accidents has not ooiy 

 not increased, hut tjecome lessened in nu eitraordioary degree. 



