Imágenes de páginas
PDF
EPUB

not be confounded with the co-efficient 0-305 in Regnault's formula for the total heat of evaporation, for it occupies a position between the two specific heats under constant pressure and under constant volume. After having expended much heat on the production of saturated steam, only a little more is required to superheat it. If this is done, all moisture is of course removed, and if sufficient heat has been added by this means, the condensation which would otherwise take place in the engine, due to work done, will not take place. It is very desirable to attain this object, but repeated attempts of a sustained nature both at sea and on land have borne no lasting fruit, the mechanical difficulties seeming more than to balance the economic results, which may be stated to be from 10 to 15 per cent. The chief troubles have been cutting of the high-pressure cylinders and valve faces, occasional burning of the superheater tubes and of the engine's cylinder lagging, and then there is of course the danger of melting the brazed seams in the copper pipes. These troubles are in the author's opinion due chiefly to the absence of suitable bye passes for the funnel gases. It should be possible to divert these from the superheater when the. engines are not working, or when the steam temperature has grown too high.

[ocr errors]

Feed Heaters. A study of properties of steam (see tables, p. 54) shows that by far the greatest amount of heat is spent in converting water into steam, only about one fifth of this heat being used for raising the temperature. All the heat of evaporation is carried away by the engine condenser water, unless before the steam enters the condenser a part is abstracted for feed-heating purposes. would, of course, be useless to employ the exhaust steam from the low-pressure cylinder, for this is nearly as cool as the condenser and the feed; it is therefore customary to take the steam for the heaters from the valve chest of the low-pressure or intermediate cylinder. If the engines are of the triple expansion type, the abstracted steam has then done respectively two-thirds or one-third of its share of work, but the expenditure of heat is only about one-tenth of what it would have been if this steam had been passed on to the condenser. The saving is therefore considerable. The steam for these heaters should be drawn from the bottoms of the valve chests, so as to carry with it any condensed water.

2

See 'C. E.,' 1891, vol. cviii. p. 474.

[ocr errors]

2 See J. A. Normand and others, N. A.,' 1895, vol. xxxvi. p. 35.

67

CHAPTER III

CORROSION

EXPERIMENTS and papers on this subject are fairly numerous, and before discussing the various theories it will be advantageous to mention them. R. Mallet, 'Brit. Assoc.,' 1838, vol. viii. p. 253; 1840, vol. x. p. 221; 1843, vol. xiii. p. 1; and 'N. A.,' 1872, vol. xiii. p. 90. -In these experiments and papers the question of boiler corrosion is hardly touched upon, but galvanic action and related subjects are thoroughly discussed. Crace Calvert, 'Manch. L. Ph.,' 1871, vol. x. p. 99, shows that air in water causes corrosion.

'Parliamentary Reports,' Admiralty Committee on Corrosion in Boilers, appointed in 1874, three reports, 1874, 1878, and 1880, c. 2662. -These experiments were very exhaustive; they aimed at ascertaining whether there was a difference between the behaviour of iron and steel, whether the lubricants in the engines affected boiler corrosion, and at determining the influence of zinc, of galvanic action, of various fluids, and of air in water.

D. Phillips, C. E.,' 1881, vol. lxv. p. 73, discusses the above reports, and comes to the conclusion that iron does not corrode as fast as steel. W. Parker, I. and S. I.,' 1881, p. 39, like the Boiler Committee, exposed various materials in actual boilers, but isolated each plate. The results do not show a great difference between iron and steel. D. Phillips, 'Marine E.,' 1890-91, adds further experiments in support of the abovementioned views.

[ocr errors]

Other experiments on corrosion will be found in the following papers:-M. Lodin, Comp. Rend.,' 1870, vol. lxx. p. 321. A. Mercier, An. Mines,' 1879, 7th ser. vol. xv. p. 234, gives experiments on the influence of fatty matter on corrosion of iron and steel. M. Lodin, 'Comp. Rend.,' 1880, vol. xci. p. 217, experiments on corrosion of various wires in hot fluids to which vegetable matter has been added. M. B. Jamieson, C. E.,' 1881, vol. lxv. p. 323. J. Norris, N. A.,' 1882, vol. xxiii. p. 151, determined the influence on corrosion of the air contained in water. L. Gruner, An. Mines,' 1883, 8th ser. vol. iii. p. 5, relative corrosion of 28 materials under 4 different conditions. This paper contains an appendix by Bustein, which shows that exposures of the samples for 112 days inside a boiler, and also in a boiler flue, affected the mechanical properties (see p. 150). J. R. Fothergill, 'M. E.,' 1884, p. 339; F. Marshall, p. 344. Professor Lewes, 'N. A.,' 1887, vol. xxviii. p. 247, influence on corrosion of the air contained in water, and general views on corrosion, chiefly in sea water. A. C. Brown, I. and S. I.,' 1888, ii. p. 129. Professor Lewes,

'N. A.,' 1889, vol. xxx. p. 340. T. Andrews, C. E.,' 1884, vol. lxxvii. p. 323, and 1885, vol. lxxxii. p. 281, gives experiments on corrosion in sea water. A. Wagner, Bayerisches Industrie u. Gewerbe Blatt,' 1875, p. 102, various theories on the chemistry of corrosion, and gives the results of his experiments on the action of various salts in hot and cold water. Bursteyn, Mitt. Pola,' 1879, vol. xii. p. 503, experiments on the influence of high pressure and fatty matter. H. Schnyder ('Berg.- H.-Z.,' vol. xxvii. p. 212), experiments on the behaviour of zinc under various conditions. J. Dewrance, C. E.,' 1900, vol. cxli. p. 107, deals with air in boilers, and the discussion refers chiefly to corrosion in H.M.S.'s water-tube boilers.

The main object of all these experiments is to ascertain the true causes of corrosion, and to discover means for preventing it; and since steel has been substituted for iron the question of its relative liability to corrode has repeatedly come to the front. No engineer with extended experience will hesitate to admit that steel does behave worse than iron. Fortunately this view has not only not led them to return to iron, with its numerous bad properties, but it has driven them to adopt preventives which have now reduced corrosion in both iron and steel boilers to very small proportions. These preventives are

I. The substitution of mineral lubricants for animal or vegetable fats or oils.

II. The use of fresh or even distilled water wherever obtainable instead of sea water.

III. The removal of air from the feed water.

IV. The use of zinc.

V. The use of alkalies.

In order to understand how these practices have been arrived at it will be necessary to discuss the various theories of corrosion.

I. and V. Action of Engine Lubricants. All vegetable and animal fats or oils behave chemically as if they were salts, consisting of a base, and an acid. In fats and oils the base is glycerine, and the acids have numerous names. Fats and oils, like some other neutral compounds-for instance, the carbonates of lime and magnesia and the sulphate of iron-are split up into acids and bases when heated. The temperature at which this splitting up takes place with fats is a little above 212° F. Thus stearine, which is the same substance as stearic acid, is manufactured by raising tallow and water to the proper temperature. Naturally the same thing can happen in a boiler, and the liberated fatty acids are now capable of corroding the same amount of iron as an equal weight of sulphuric or other mineral acid. The resultant compound is ferric soap, which makes up the greater part of the filthy greasy substance to be met with in some boilers. By bringing the fatty acids in contact with other substances, other soaps are formed. Thus with potash we get soft soap, with soda hard soap; with lime we get the very much harder soap called putty, while with the oxides of lead and zinc medical ointments are the products.

Whereas glycerine is one of the weakest bases, lime, potash, and soda are the very strongest, and whereas the one is incapable of

retaining the fatty acids at a boiling temperature, the others will only part with them in presence of a stronger acid. Carbonic acid is one of the weaker ones, and therefore it is possible for the acids in some oils and fats to supplant it, if brought into contact with carbonate of soda. Some of them are not strong enough to do this direct, but must first attack the iron, forming ferric soap. In such a case the soda does not prevent corrosion, while in the previous one the liberated carbonic acid might do just as much harm as the fat. When using organic lubricants it is, therefore, better to add caustic soda or lime instead of the carbonate. Mineral oils are not double compounds, and, like water, consist of only two elements-viz. carbon and hydrogen and these are harmless. To add soda where these are in use would be useless.

II. The Use of Fresh Water.-Since the introduction of surface condensers sea water has gone out of use, and now that evaporators are largely used the make up water is generally distilled. Nevertheless, sea water does leak past the condenser tube ends, and some of its salts are decidedly injurious.

Chloride of Magnesia.-Professor Lewes states that if sea water is distilled while in contact with iron it gives off hydrochloric gas when the volume of the water has been reduced to one-fifth, and it is only too probable that before it is produced in sufficient quantities to escape it must have been attacking the iron. He also mentions that magnesic chloride and calcic carbonate (lime) react on each other, and are converted into calcic chloride and magnesic oxide, the carbonic acid escaping at a boiling temperature. When exposed to the influence of the air magnesic oxide absorbs carbonic acid, so that on re-filling a boiler which has been opened for some time, and heating it, this acid gas is given off again. He looks on magnesium salts as being decidedly injurious to the life of a boiler; but, as all sea water contains them, this should, if possible, never be admitted.

The Boiler Committee tube No. 21 contained chloride of magnesia and attacked iron and steel very severely.

A. Wagner shows that if water contains chloride of magnesia, but no air, it commences to attack iron at a temperature of about 212° F., while the following chlorides will only attack it in presence of air :they are arranged in the order of their corrosive power-ammonium, sodium, potassium, barium, calcium.

The experience which is accumulating with the Manchester Steam Users' Association does not confirm these conclusions; at any rate, pure chloride of magnesia and water will not attack iron either cold or at a boiling temperature even if oxygen be forced into the water, and as far as can be ascertained the results mentioned above are due to absence of precautions for the rigid exclusion of even a trace of carbonic acid, such as distilling caustic soda solution in vacuo. Such a condition cannot of course be attained in a boiler, and for practical purposes the advice to exclude chlorides of magnesia is a good one. Chloride of Lime acts similarly to chloride of magnesia. Sulphate of Lime is the chief scale-forming salt, but it has not yet been asserted to have a corrosive influence.

Common Salt, if pure, seems to do no harm to iron, but even the best commercial salt usually contains some chloride of magnesia.

Free Acids, which are sometimes proposed for boiler-scale solvents, should never be used, nor should the feed water ever be taken from a river near a chemical factory. Occasionally waste acids are discharged there, which may do a serious amount of injury. This was mentioned by Mr. Hallett (M. E.,' 1884, p 350).

Whenever there is any doubt as to the harmlessness of fluids or salts intended to be put into a boiler, a simple test is to boil them, and then to place a thoroughly clean knife-blade into them. Should any rust be formed, should the water be discoloured, or should copper deposit itself on the blade, then the substance ought not to be used. If certain free acids are present the above test will give no warning, but a few drops of dissolved yellow prussiate of potash or of tannic acid should be added. If iron has been dissolved, a light bluish precipitate is at once formed by the first test, which slowly turns dark blue. The other fluid produces ink.

Copper Salts seem to be an unavoidable, though only a minute, constituent of all feed water, as is proved by the green scale in all old boilers. For some unaccountable reason it does not deposit itself uni

0 FIG. 92

formly over the inside of the boiler. Its presence is denied by many engineers, therefore little information could be obtained as to the points where it is most generally found. The author's experience is that patches of green scale will be found on the zinc slabs, and also near them on the iron. This is true even when the zinc t is suspended from the steam-space stays (fig. 92). It will be found that the tubes marked sometimes contain thick patches of greenish-grey scale. Small quantities are also found near the water lines of boilers, particularly at both end plates. Larger quantities are found at the front end plates, between the nests of tubes, and on the lower part of the shell plates. The furnace bottoms usually contain the greatest number of patches, but it is difficult to detect them, as these parts are discoloured by grease. The furnace crowns and combustion chambers are nearly always quite free from them, from which it would appear that it is chiefly the non-heating surfaces of a boiler which get covered.

If the copper would only distribute itself uniformly, it would act, as a protective scale, but as it is deposited on small areas, these may become sources of danger by producing galvanic currents, at any rate

« AnteriorContinuar »