November, 1920 away at the ends of the tubes (not infrequently accompanied by wearees ferrules) culminating in perforation at from 1 to 3 inches from the end. In many cases where the greatest intensity of the deterioration occurs at the upper inside surface of the tube, the presence of air has been held to be the chief contributory fac- tor, and under such circumstances the provision of air-releasing devices at the highest point of the circulating water system has proved of marked benefit. However, with all such mechanical precautions, the evacuation (before entry to the tubes) of air which is necessarily in suspension in the circulating water is at the best only partial. This is more especially the case in high-speed shallow draft vessels, and as will be evident from an examination of the accompanying illustration, the sharp edges both of ferrule and tube, assisted to some extent by thé possibility that the tube and ferrule may not always lie per- fectly concentricaily, tend to break the flow of water and to fonm pockets of an unstable nature where air may be reasonably expected to accumulate, Dangers from Abrasion This type of defect has been par- 'ticularly noticeable in vessels serving in shallow waters containing a con- siderable proportion of solid matter, such as sand, chalk, etc., and_ this, together with the comparatively high- water velocity through the tubes, points to the influence of erosion, or the abrasive action of gritty particles, as being of considerable importance. The presence of such gritty matter cannot of course be avoided on serv- ice, and it appears that investigations with regard to the elimination of this type of defect must take the lines of endeavoring to obtain a: metal whose surface is better able to withstand this abrasive action. Whereas in the cases of tube-end deterioration the number of tubes af- fected in any condenser (apart from those which actually fail) may be considerable, cases of local pitting are rarely evident in more than one ofr two tubes in the condenser concerned, and when once a failure has occurred and the tube plugged or replaced, little "further trouble is experienced. In 'these cases perforation occurs at iso- lated Spots, and the failure has been erally considered to be due to er local defect, such as a spill or a." Elimination of defects of ' a Is entirely a question ly a, ae of the 'mechanical- ees ube, which desirable result only be attained by the drawing do : wa of a mechanically perfect cast THE MARINE REVIEW shell. The practice of turning and boring each shell, even if regarded as merely for Purposes of exploration of the metal immediately beneath ithe surface skins, has a considerable safe- guarding effect toward the reduction in the number of such surface irregu- larities, 'as may be discovered on sighting, Perforations, the cause of which has _ been directly traced to obstructions, were particularly pre- valent in coal-burning ships where the ashes were discharged beneath the waterline, and' though it is possible to nullify the effect of a vessel's own refuse by suitably placing the ash chutes and by the adoption of oil- ZZ DY KL iSLL LE ALLE SS LOW OF CIRCULATING WATER THROUGH INLET END OF TUBE fuel firing, yet such obstruction of tubes by cinders, etc., may still. be caused by refuse from other vessels. The recorded cases where failure has occurred through the impinge- ment of steam and water directly on a limited area of the outer surface of certain tubes are few, and in gen- eral have been confined to condensers other than those of admiralty design. Such failures should not occur, pro- vided thait orifices are suitably placed and baffled. Why Ends Are Crushed The proportionate number of fail- ures due to crushing is small, and these may be regarded as_ isolated cases. In no case has it been estab- lished that over-tightening of screwed glands has been resorted to, and the fact that crushing sometimes does occur, serves to emphasize the im- portance of the tubes being sufficient- 'ly hard to stand a reasonable margin of pressure due to tightening up above that 'absolutely necessary to secure watertightness, though it may be remarked that crushing is sometimes observed as a natural result of the weakening of tubes which have been attacked internally by corrosion of a general nature. There is some divergence of opin- ion as concerning the most desirable degree of purity of the 70 :29:1 alloy, but investigation has shown that a large proportion of the tubes, which 615 have failed from splitting, have been Manufactured from materials of a lower Standard of purity than that Tequired by the admiralty specification. Apart from the question of purity of constituents, the internal condition of stress in the finished tube must have considerable bearing on the liabil- ity to season cracking and _ splitting. The work done during the final draw (which must be such as to leave the tubes sufficiently hard to stand the test of grometing) of necessity leaves the material of the tube in an inde- terminate condition of internal stress, and anything that can be done which will eliminate this internal stress, while at the same time allowing the tube to retain the requisite hardness, _Will be of great benefit in minimizing defects from this source. The case in which failure has been directly ascribed to the use of unsatis- factory circulating water is excep- tional, but 'it certainly gives weight to a widely held opinion that the con- ditions under which many condensers are worked in the early stages of their life, that is in comparatively stagnant fitting-out basins contami- nated by various kinds of refuse, are conceivably of importance as a pri- mary cause of tube decay. On the occasion referred to, the vessel had teen anchored in waters contaminated' by sewage, and the case is quoted as emphasizing that materials for marine usage should -be able to resist corro- sion under varying conditions which do not allow of the selection of par- ticular materials to suit particular waters. The deleterious effect of vegetable oils has already been de- scribed before this institute, but the writer is not aware of any experi- ments having been carried out to de- termine the effect of oil fuel in regard to the acceleration of corro- sion; though it may be remarked that in one of His Majesty's vessels which reported a considerable number of defective tubes within two months of entering into service, no definite cause could be traced beyond that extended basin trials had been car- ried out in comparatively stagnant water contaminated by a considerable leak from one of the ship's tanks. Study of General Corrosion The proportionate number of re- ported failures due to general corro- sion does not adequately express the importance of this class, these being cases of actual failure only, that is general corrosion culminating in per- foration; whereas the great majority of cases in which general corrosion is present are not recorded, and it is the possibility of failure at an un-