In the Journal of the American Society of Naval Engineers for May, 1910, in an article on the subject of gun erosion, Lieutenant-Commander H. E. Yarnell, U. S. N., advances a theory of the cause of erosion which, though once generally accepted and still frequently put forth by writers on ordnance, is, I believe, erroneous and susceptible of being proved so.
As stated by Mr. Yarnell, this theory is that "gun erosion is caused by the escape of the highly heated gases of combustion through the minute openings between the rotating bands and the bore of the gun before and shortly after the projectile begins to move. After the shell has advanced sufficiently far to enable the copper rotating band to mould itself to the form of the rifling the erosion decreases to a considerable extent."
Now erosion as it occurs to-day in U. S. naval guns is a smooth wearing away of the bore, beginning in rear of the rifling and extending further and further down the bore as the firing is continued. This wear is greatest at the rear, and is about twice as great on the lands as in the grooves, so that its effect is to make the bore slightly conical from the front of the powder chamber forward and to gradually obliterate the rifling. Thus, for example, in a 12-inch 45-caliber gun with 2700 f. s. muzzle velocity, after too rounds of nitro-cellulose powder have been fired, the diameter across the lands at the origin of the rifling will have been increased from 12.00 to about 12’.20, and that across the grooves from 12’.10 to about 12”.22, and this wear will regularly diminish towards the muzzle so that at 10 or 12 calibers forward of the origin of rifling it amounts only to 0”.002 or 0”.003 on the lands and nothing in the grooves.
From the point where the erosion just described ceases, the bore seems to be practically unaffected by continued firing and the rifling remains perfect until within a few calibers of the muzzle, where wear again becomes apparent. This muzzle wear, however, seems to have no relation with ordinary erosion and at any rate is of little or no importance, seldom exceeding a few thousandths of an inch, even in large guns after enough firing to have caused really serious wear at the breech end. Its cause is believed to be the attrition of the powder gases as they escape from the muzzle with greatly accelerated velocity after the exit of the projectile.
Coming back to the erosion proper at the breech end of the gun, the following objections to the theory of its cause advanced by Mr. Yarnell seem to me to be conclusive:
(1) The forward slope of the powder chamber and the bore behind the rammed-home position of the projectile erode, the rate of wear of the bore just in rear of the rotating band being the same as that of the bottoms of the grooves at the beginning of the rifling. This wear cannot be accounted for by escape of gas past the projectile.
(2) In the rammed-home position of the projectile and until it has moved forward enough for its rotating band to be forced into the rifling, the grooves offer comparatively free passages for the escape of gas; moreover there is at all times much closer contact between the copper band and the tops of the lands than there is between it and the bottoms of the grooves. Consequently the wear of the bore would be chiefly, if not wholly, confined to the grooves were it caused by escape of gas past the projectile. As a matter of fact the wear of the tops of the lands is about double that of the bottoms of the grooves.
(3) The pressure of the gas, which is only about 2 tons per square inch when the projectile begins to move, increases rapidly, attains its maximum of 15 to 18 tons when the projectile has moved from 6 to 10 or 12 calibers along the bore, and then decreases slowly. The linear expansion of the bore abreast the rotating band also increases as the projectile moves forward, both on account of the increasing powder pressure and on account of the diminishing thickness of the gun wall, the maximum enlargement of bore probably occurring when the shell is about half way to the muzzle. Moreover the driving edges of the rotating band wear away as the projectile moves. Consequently there should be a very much greater escape of gas past the shell after it has moved a few calibers than when it has merely moved enough to force its band into the grooves, and the greatest erosion, if it were caused by gas escape, would not occur at the beginning of the rifling but further down the bore.
(4) The latest type guns have much greater muzzle velocities than earlier types mainly because of their larger powder chambers that allow the use of larger charges of powder without much increase of maximum pressure. If the pressure curves of such guns are compared, it will be found that from the origin up to the point of maximum they do not differ much, the greater velocity given by the larger charge being mainly due to the more sustained pressure from the point of maximum to the muzzle. The conditions relating to the escape of gas past the projectile before it moves and during the early stages of its motion are, therefore, almost the same in high power and in low power guns of the same caliber using the same sort of powder with about the same maximum pressure. But there is an immense difference in the erosions of such guns; a 12" Mark II gun that has been fired 500 rounds with velocities not exceeding 2100 f. s. is only eroded to about the same extent as a 12" Mark V gun after 50 rounds with 2700 f. s. This is clear proof that erosion depends upon the quantity of gas (number of heat units) behind the projectile as well as upon its quality (pressure and temperature), whereas the amount of gas that escapes past the projectile and its erosive effect can only depend upon pressure and temperature (the time of action being considered constant).
(5) Experiments with gas escaping through vents having shown that copper is worn away more rapidly than gun-steel, the rotating band, past which all the gas that is supposed to erode the bore must escape, should be much more eroded than any part of the bore by a single round. As a matter of fact the rotating bands of fired projectiles show no signs at all of erosion.
If the foregoing objections are valid, as they seem to me to be, it is impossible to reproduce the conditions of gun erosion by the escape of the products of combustion of gun powder through vents. Yet the numerous experiments that have been made in this line have some value since they demonstrate that the higher the temperature of the gases and the lower the melting point of the metal attacked, the more rapidly is the surface softened and worn away. The actions in the two cases are not identical, but they are near enough alike to allow some conclusions to be drawn from one case that will apply to the other.
On the other hand, we must guard against concluding too much. For example, the amount of metal worn away by the gas escaping through a vent can be made to vary materially by altering the method of ignition of the powder, everything else remaining unchanged; while, in the case of a gun, erosion is not appreciably affected by the character of the ignition. And, again, the use of water, paraffin and other substances may greatly reduce the erosion caused by gas escaping through a vent, while in guns they would be ineffective unless used in such quantities as to reduce materially the ballistic results.
Considering erosion as it actually occurs in a gun of any given caliber, it may be shown to depend upon three factors—temperature of combustion, weight of charge and time of action.
(a) Temperature of combustion is generally recognized to be an important factor in gun erosion, and to its higher value in the case of nitro-glycerine powders is properly attributed their greater erosive effect when compared with nitro-cellulose powders. The temperatures of combustion of the Italian ballistite and of the English cordite are from 500° to 600° Cent. greater than that of the pure nitro-cellulose powders used by France and the United States, and how great an effect this difference of temperatures produces is well illustrated by a comparison of the erosions of English and United States naval guns of large caliber. The English 12" Mark VIII gun is of 35 calibers length of bore and its charge of 167 lbs. of cordite imparts a velocity of 2367 f. s. to an 850-lb. shell. Our 12" Mark III gun is a 40 caliber gun and has an 870-lb. shell with 2400 f. s. muzzle velocity, the charge being 237 lbs. of nitro-cellulose powder. The maximum pressures are about the same in each case. In the former gun the "probable maximum wear at 1” from commencement of rifling" is stated in the official Treatise on Service Ordnance, 1904, to be 0”.50 after 80 rounds and 0”.76 after 130 rounds. In the latter gun the wear at the same point is only 0”.13 to 0"15 after 80 rounds and 0”.20 to 0”.22 after 130 rounds. Notwithstanding the cordite charge is 70 lbs. less than the nitro-cellulose charge, its higher temperature of combustion results in nearly four times greater erosion for somewhat less power developed.
It is commonly believed that pressure plays an important part in producing erosion, and experiments have in fact shown that the smaller charge of quick powder (of the same chemical characteristics) causes as much erosion as the larger charge of slow powder required to give the same velocity in a given gun. But the explanation of this is that at the higher pressure more CO2 and less CO is formed than at the lower pressure, so that the temperature of combustion in the former case is greater than in the latter, and the greater quantity of heat from the larger charge is thus compensated for by the higher temperature of the gases of the smaller charge. The exhaustive investigations of Sir Andrew Noble, recently published, indicate that there is an increase of about 100° Cent. in the temperature of combustion of nitrocellulose powders for an increase of maximum pressure from about 13 tons to about 18 tons per square inch. 1
(b) Weight of charge is the factor that, for a given kind of powder and a given caliber of gun, plays the predominant part in producing erosion. It is the need of larger charges to give greater velocities that practically limits the power of modern guns by making the wear prohibitive when the power is increased beyond a certain point. It is easy enough to get 3000 f. s., or even more, from a gun of 45 or 50 calibers length of bore, but the weight of charge required generates such an immense quantity of heat that the resulting erosion, even when nitro-cellulose powder is used, is too dear a price to pay for the gain of power.
That even comparatively small guns as now constructed will not stand a velocity much in excess of 3000 f. s. for many rounds is well illustrated by the competitive test of the Brown and Crozier 6-inch wire-wound guns. The program of tests contemplated 250 rounds or more from each of these guns, which had chambers large enough for charges of about 80 lbs. of nitrocellulose powder, but the Board reported that due to erosion the limit of effectiveness for firing service projectiles had been reached in 88 rounds from one gun. The program was then altered to 98 rounds, the last 10 with excessive charges. The average powder pressure for 75 rounds (all that were fired except 13 with reduced and 10 with excessive charges) was 40,375 lbs. per square
1 For a very important correction of Sir Andrew Noble's results see Obstrvations relatives aux travaux de Sir A. Noble, par le Chef d'Escadron Bourgoin, de l'artillerie Coloniale. Memorial de l'artillerie navale. Part 1909.
inch for one gun and 40,760 for the other. The highest velocity in the first 88 rounds was 3455 f. s. for one and 3533 f. s. for the other gun. In the io rounds with excessive charges the highest velocities were 3740 f. s. in one and 3860 f. s. in the other gun. At the end of the test of 98 rounds both guns were completely worn out by erosion of the bore.2
Another illustration of the effect of weight of charge is given by a comparison between the U. S. Navy 6-inch guns of Marks IV and VI. The former is a 40-caliber gun with a charge of about 19 lbs. and a muzzle velocity of 2100 f. S.; the latter is a 50-caliber gun with a 30-lb. charge and 2600 f. s. muzzle velocity. If we take the average velocity of the projectile while traversing the bore to be six-tenths its final velocity, the travels in the two guns being respectively 17.2 and 20.7 feet, it will be seen that the time of exposure of the bore at the origin of rifling to the action of the gases is almost exactly the same in the two guns, so that the time factor cannot affect the comparison. The maximum service pressures in the two guns are also not far apart (15 tons in the former and 16 in the latter). But the maximum wear of the gun with 3o-lb. charge after 254 rounds is as great as that of the gun with 19-lb. charge after 1198 rounds, besides which the wear extends much further down the bore in the former. (Maximum enlargement 0”.071 in each gun, enlargement 3" forward of origin of rifling 0”.055 in one and only 0”.029 in the other.) It is true that the 40-caliber gun uses a cartridge case with a mouth cup, while the 50-caliber gun does not, but I can see no reason for supposing that this affects erosion, and the greater erosion of the latter gun must, I believe, be attributed to its larger powder charge.
(c) The time of action of the powder gases is the third factor in erosion. Its effect is shown by the greater erosion of long than of short guns using the same powder charge. In the usual case, however, where 5 or even 10 calibers length is added to the bore of a gun to gain 100 or 200 f. s. velocity, the increase of erosion is unimportant because the extra time taken to traverse the bore is very small (on account of the high velocity of the projectile near the muzzle).
2 It will be seen that the statement made in Mr. Yarnell's article already referred to, that these guns fired ma rounds each with an average muzzle velocity of 3600 f. s. goes considerably beyond the facts.
Thus far we have considered the erosion of a gun of fixed caliber, but when guns of different calibers are compared the time of action becomes the predominant factor in erosion, and a new factor, relative surface exposed to the gases, appears.
In similar guns, similarly loaded, as caliber becomes greater, surface area of bore per pound of powder diminishes and time of exposure of that surface to the action of the powder gases increases—the larger the gun, the more heat per unit area of bore and the longer it acts on that area. Hence there is good theoretical ground for supposing that the erosion of similar guns will be proportional to about the square of their calibers, and, in fact, a 12" gun does wear at least four times as fast as a 6" gun of the same relative power.
From all the foregoing we may conclude, I think, that the necessary and sufficient conditions for gun erosion are the intense heating of a thin film of metal at the surface of the bore and the movement of the gases over that surface. The temperature must not only be high enough but must be maintained long enough to bring the surface almost, or quite, to the melting point. The larger the gun the less the relative surface to be heated and the longer the time the heat is applied, hence the greater the erosion. Also, the larger the charge the longer the high temperature is maintained and the more the erosion.
That movement of the gases over the heated surface is a requisite for erosion is shown by the fact that the rear part of the powder chamber is not worn away by repeated firings, nor is the inner surface of a bomb in which powder charges are burned for experimental purposes. Whether this movement of the gases rubs off the heated surface of the bore, or merely acts to increase the rate of transmission of heat from gases to bore, I do not pretend to say.
As to the possibility of preventing erosion, the most important thing to do is to use powder having a low temperature of combustion, and the U. S. Navy smokeless powder has the lowest temperature of any in use. No advantage is likely to be had from the use of special metals for the gun tube, ordinary gun-steel being apparently the most resisting material that it is practicable to employ. If we insist upon high power in guns, we must expect them to wear out with comparative rapidity, and the only apparent remedy is relining after a certain number of rounds.
As regards the extent of erosion, many of the statements made, even by persons in a position to know the facts, convey an erroneous idea because they have been based on experience with guns and projectiles not properly designed for the high velocities now common. With a suitable profile of bore and properly proportioned rotating bands a 12-inch gun of 45 calibers length of bore, such as the U. S. Navy 12-inch Mark V gun, using nitro-cellulose powder, will stand 200 rounds at 2700 or 2800 f. s. velocity without material loss of accuracy or falling off in velocity, and a 6-inch gun of equal relative power will probably stand 1000 rounds. Nor is there any danger that the erosion of the guns of a ship will produce variations in their ballistics that will affect the accuracy of salvo firing.
I have not referred to Vieille's Study of Erosion, a translation of which forms part of Mr. Yarnell's article, for the reason that it seems to me to relate to a sort of erosion that we do not experience in our present guns. I should not presume to oppose my own opinion to that of the distinguished French engineer who originated our smokeless powder and whose knowledge of interior ballistics, based upon profound study and immense experimental research, is beyond compare, but the date of his article (published in 1901), as well as various statements made in it, seem to me to show that he did not have in mind the conditions we now have to deal with. At all events, since he wrote, much experience has been gained that may well have led to a change of opinion on his part.
As bearing on this subject, the following quotation from the English official Treatise on Service Ordnance is interesting:
"Erosion is a general term applied to the effects produced by the action of gas on the bore of a gun. These effects are of two kinds, viz.:
" (1) The gradual enlargement of the bore and smooth wearing away of the surface by the action of the gas in rear of the projectile, which is. called wear and is independent of the sealing of the bore by the projectile.
" (2) The irregular eating away of the surface of the bore in holes and gutters, which is called scoring, and is mainly due to windage and imperfect sealing of the bore.
"These two actions may occur separately or conjointly, and with cordite charges the differences are marked.
"In the latest high velocity B. L. guns, in which the sealing is nearly perfect, wear occurs with very slight or no trace of scoring.
"In R. M. L. guns, where there is considerable windage, the scoring is considerable and the wear practically nil, while in the older type of B. L. guns both wear and scoring appear."
The smooth " wear " described in paragraph (I) of the preceding quotation is the only sort of erosion that occurs at the present day in U. S. Naval guns, and the statement that it is caused " by the action of the gas in rear of the projectile . . . .and is independent of the sealing of the bore by the projectile' cannot, in my opinion, be successfully controverted.