Imágenes de páginas

Table of Gases absorbed by Water

Amount of Gas absorbed by 1,000 c.c. Water under a Pressure of oue


Name of Gas

[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]


Note.-The above are a summary from various experiments by Bunsen, also A. Winkler, · Zeitschr. f. phys. Chemie,' 1892, vol. ix. p. 171 ; and Bohn and Bock, • Wied. Ann.,' 1891, vol. xlviii. p. 319. The volumes in this table are for the respective temperatures.

By forcing gases into water, its temperature is slightly raised, the liberated heat being apparently equal to the latent heat of evaporation. This is certainly true for steam, and seems to be nearly true for other vapours, such as carbonic acid, whose latent heat is 134 calories (Muir, pp. 294 and 996). The speed with which oxygen is absorbed by water is according to M. Lodin (* Comp. Rend.,' 1880, vol. xci. p. 217) 0:000036 lb. per sq. Ib. per hour at 64° to 68° F. ; at 212° F. the rate is ten times faster. Lime water absorbs it at the rate of 0.000047 lb. and 8 times faster when hot.

Boiling Phenomena.—If one had never seen water boiling, and if one were asked to describe what one would expect to occur if water of a boiling temperature were placed in contact with a hotter plate, one would be tempted to say that, at first, innumerable small bubbles would form all over the surface, like dampness on a window pane; that these bubbles, like the water drops on the glass, would increase in size, and would join together, and when an individual had acquired a sufficiently large size, it would break away and rise to the surface, the space thus left bare being at once re-covered with minute bubbles. It is, however, well known that water does not boil like this. When only little heat is applied to the plate one may have to wait for long periods between the appearances of bubbles, and when they do show they grow so suddenly and tear themselves away so quickly that the process cannot be watched. It will, however, be noticed that, on its upward journey, each bubble increases in size, especially at first. This cannot be due to the decreasing head of water over the bubble, for it would have to rise 30 ft. before its diameter is increased to 30%.

The increased size cannot be due to the inertia of the water; for assuming that the bubble is formed with such rapidity as to impart an appreciable velocity to the mass of water above it

, and that in coming to rest the water exerts a suction on the bubble and thus increases its size, then clearly after the first increase the bubble would



decrease again. This does not seem to be the case; besides a suction which would double the volume of the bubble would have to be about 7 lbs. per square inch, and for however short a period this reduced pressure might exist, it would cause the whole body of the nearly boiling water to burst into steam. Only one further explanation suggests itself, and that is retarded ebullition, which means that the water is superheated.

Retarded Ebullition has for generations done service as explanation for boiler explosions (L. E. Fletcher, ' Experiments on Redhot Furnaces,' Manchester ; A. Witz, ‘Comp. Rend.,' 1892. vol. cxiv. p. 41), but since these have been carefully enquired into the theory has never been confirmed. It is however well known that, if water be carefully heated in a very smooth vessel, the temperature may be raised several degrees above 212° F. before the water bursts into steam. Chemists are sometimes troubled by retarded ebullition, which now and then may eject the contents of a test tube. No serious amount of superheating is required for this phenomenon; for if ebullition takes place when only F. superheat has been reached, then the volume of steam evolved is about equal to the volume of water. Of course if this sudden ebullition takes place in a closed vessel hardly any bubbles would be seen, for the increase of pressure due to half a degree of temperature is only one-sixth of a pound, and this slight increase would at once check the sudden ebullition. Boiler explosions from this cause are therefore impossible. If cold water be heated in a glass vessel over a flame and properly illuminated, one can see the change of density of the locally heated water. If boiling water be gently heated, the irregular density is not very marked until boiling actually commences, then it will be seen that the heating surface is covered with a layer of what appears to be less dense water, and in all probability this is superheated water, though judging by the small bubbles which are occasionally evolved from it, the amount of superheat can hardly be more than 15.° F. If now a bubble be formed on the heating surface, the surrounding superheated water gives up its steam, and rapidly enlarges the bubble, which then rises to the surface. The current which its passage produces draws more superheated water to the bubble's birthplace and along its path. Every succeeding bubble will therefore most likely start at the same point and travel along a path which has now become a channel for superheated water, and it is this water which gives up its steam to the bubbles and causes them visibly to increase in size as they rise to the surface.

After a time the superheated water will be distributed throughout the mass of hot water, and if small solid particles are floating about, they may form the starting points for a series of bubbles. A somewhat similar condition of things undoubtedly exists in effervescent waters which have just been poured into a glass. These are supersaturated with carbonic acid gas, but steam is also a gas, and superheated water is water supersaturated with steam. There is, however, this marked difference, that steam has a much greater latent heat of evaporation than carbonic acid, and its cooling effect is greater, but this is in a way balanced by its greater volume.

Erratic Ebullition. If this view be accepted, it is easy to under


stand why the generation of steam is a jerky performance, notably in water-tube boilers. Suppose that in a bottom tube of a Belleville element, of say 100 ft. total length, a steam bubble be suddenly formed ; it will have to set the whole water column in motion; to arrest it a suction force would have to be applied in the lower column, which means a reduction of pressure. In the early days of this boiler, this action caused much trouble, for instead of setting the upper 100 ft. column of water in motion, it was easier for the steam to drive the water back into the downcast tube, and a backward circulation was sometimes set up. In modern Belleville boilers, non-return valves are fitted at the entrance to the lowest tubes, and the jerkiness of the steam generation is now an essential advantage to this boiler, for the suction which follows each outburst of steam draws fresh water into the lower tube and thus materially assists circulation.

Circulation in Water-tube Boilers.—Elaborate and instructive experiments have been made on this subject by Professor W. H. Watkins, ' N.A.,' 1896, vol. xxxvii. p. 267, who constructed a large number of models, using glass tubes instead of metal tubes and introducing oak sawdust into the water to show the circulation. It was then found that, especially in the large tube boilers, like the Belleville, the movements of the steam bubbles were no indication as to the movement of the water; that, in fact, the water remained fairly stationary while the bubbles moved through it. It could frequently be noticed that little heaps of sawdust collected in the more horizontal tubes just below the points where the flames were striking them, even although much steam was being generated a little further on just over the flames. Occasionally in the Belleville boiler model these heaps would be swept forward, due no doubt to a momentary suction assisted by the action of the non-return valve. This stagnation was not so marked in other large-tube boilers, and seemed to be a rare occurrence in the Babcock and Wilcox model boiler, but the finding of loose scale in the lower tubes of full-sized ones shows that their circulation is not as powerful as one would imagine. Certainly this boiler offers very much less resistance to water circulation than the Belleville boiler, but it suffers from the disadvantage that the water which passes through the lower tubes has overcome more resistance than that which passes through the upper ones.

In small-tubed boilers the sawdust and the bubbles travel together, and while steam is being generated a most violent circulation is set up. The motive force is the difference of the pressure of the column of mixed steam and water moving upward as compared with the pressure of the solid water in the downcast tubes. This motive force can of course never exceed a head of water corresponding to the height of the boiler, and a theoretical limit to the steaming power of small-tubed boilers is therefore fixed.

The limit is a high one, possibly exceeding an evaporation of 200 or 300 lbs. of water per square foot per hour with one-inch tubes in boilers of ordinary dimensions, but this consideration shows that attempts to reduce both the diameters of water tubes without increasing the heights of the boilers may lead to failures.

Circulation Reversed. -- Another important matter revealed by experiments like the above is that the currents in water tubes can be reversed, that the bubbles instead of rising can be made to descend, and then to rise in the upcast tube. This condition can be easily reproduced if in a U tube the downcast is first heated until much steam is generated there, then if the flame be moved from this tube to the downcast, there will be no reversal of current, the bubbles will be seen to form near the flame in the old downcast, travel downwards and then up the old upcast tube, which is of course not being heated. On rare occasions a similar result is produced in experimental boilers, and is then doubtless due to the already mentioned jerky generation of steam. In some glass models, notably in that of the Sterling boiler, one can notice that when the flame is just giving the proper amount of heat the generation of steam in the front rows of tubes may be so sudden that the bubbles are shot both upwards and downwards. In water-tube boilers of the Yarrow type having no regular downcast tubes, it is probable that reversals of current often take place in those tubes not in immediate contact with the fire. This is perhaps no disadvantage.

Circulation Power of Bubbles. The following simple considerations may be of some assistance in arriving at a correct view of the part which steam bubbles play in producing circulation. If a beaker with water be placed on a spring weighing machine, and a sphere of some heavy material be raised or lowered with a velocity of v feet per second, the balance will indicate an upward or downward pressure

22.02 of about pounds, where d is the diameter of the sphere

400 measured in inches. If the sphere be made to travel upwards with such a velocity that the pressure is exactly equal to the weight of water displaced by the sphere, then the maximum velocity is found with which bubbles of the diameter d can rise towards the surface.

v = 2:7 v d". Bubbles of s in. diameter could move upwards with a velocity of 0:9 ft. per second, the resistance encountered would be 10000 lb., equal to a head of 11; in. If eight such bubbles combine, forming one large one of ; in. diameter, its upward velocity would be 1.9 ft. per second, and its resistance equal to a head of 3 in. water (C. H. R. Sankey, 'C. E.,' vol. cxli. p. 133, finds 1 ft. for inch bubbles as a limit; larger bubbles are probably too irregular in shape). The total relative resistance offered to the eight small, or to the one large bubble is as 1 to 8, but the total force exerted is equal to the total buoyancy and is of course the same. It is probably the upward pressure of the bubbles which is the prime cause of circulation, and from the above it is clear that in boilers with free circulation the formation of large bubbles should be encouraged, and these should not be broken up by tubes, but should have a clear channel above them (see fig. 89). It must, however, not be forgotten that the circulation velocity is much reduced by the total resistance, and in marine boilers one generally wishes to get a rapid circulation amongst the tubes, and therefore places these over the furnaces, as shown in fig. 90. In this arrangement the bubbles are likely to press towards the sides and thus interfere with the down current.

In water-tube boilers the movement of bubbles is based on some


[ocr errors]
[ocr errors]


[ocr errors]

what more complicated conditions. Assume that the bubbles in a vertical water tube are moving upwards, each bubble having a mean length 1, and separated from its neighbour by a water space of the length L. The diameter d of the bubble is not quite as large as D, the bore of the tube, as could be seen if a glass tube and coloured water be used. The buoyancy of the bubble is y.l.d.? , where y is

4' the density of water. The sum of the pressure against its upper side and suction on its lower side is p= y.l. or l.

head of pressure. This pressure, p, imparts a downward velocity, v, to the film of water round the bubble: v feet per second = 5.71 (approximately). The work performed by the bubble in its upward course at a velocity V is V.y.l.d.?, and this is equal to the frictional work done by the

4' water on the walls of the tube: F= 0 (v3/ + w"). . d, where o is the coefficient of friction, and w is the upward velocity of the water above the bubble which is found as follows :

(V – w).do. = v (D? – d)

4 Combining these equations we get a cubic one which expresses the velocity V in terms of the ratio D : d or vice versú; then finding the maximum value for V, the various other values can all be found approximately

v1.(22.d.1 - L)

: Here V is expressed in feet per second and the dimensions in inches. D. v and w are found from previous equations. If the rising bubbles have to produce a downward current in other tubes, and if any of the uptake tubes are not vertical, the problem grows a little more complicated. Whatever deductions

may be drawn from these formulæ, or from similar considerations as to the action of bubbles, as to the least height of boilers, smallest diameters of tubes and greatest inclination, experiments must be undertaken before any definite decision is arrived at. Only recently a case was mentioned to the author where a four-inch waterpipe swelled locally to about nine inches diameter before it burst, apparently without a cause. Here the material could certainly not be blamed, nor was scale or grease found, and the only other explanation which suggests itself is want of circulation.

Moist Steam.--It has been clearly demonstrated that any moisture carried over with the steam very materially reduces the efficiency of an engine, especially by increasing the initial condensation and reevaporation; it is therefore desirable to fit steam traps to the main steam pipes whenever these are long and likely to cause condensation. On trials the moisture of steam should be measured ; this should be done by a separator calorimeter, Carpenter's being a very serviceable one.

Superheated Steam.-The specific heat of superheated steam'under constant pressure varies from 0:6 at 290° to 0:8 at 408° F., and should

"J. H. Grindley, “ Manch. L. and Ph.' 1901, vol. xlv. No. 3.

[ocr errors]



« AnteriorContinuar »