Tests of the resistance of Krupp armor at the end of the 19th century
Some of the earliest information about the testing of Krupp armor at my disposal is a mention of it in an article in Naval Annual magazine for 1897, a description and partial translation of which is contained in “Naval Collection” No. 1 for 1898. Unfortunately, the source provides only an indirect description of these tests in the section that describes the 1896 shooting of the Harveyized armor plate produced by the British Cammel plant, which I cited in the previous article.
Let me remind you that of all the cases of practical testing of armor cemented using the Harvey method known to me, it was Kammel’s 152-mm product that demonstrated the best result. It was fired five times with 100-pound 6-inch Holzer shells, and the armor withstood the impacts of four of them, respectively, the “K” factor is equal to or greater than 2.
Is Harvey better than Krupp?
The description of the tests provides a recalculation of the durability of the Kammel plate into an iron one using the Tresider formula. This formula itself is not given, and it is not of great interest to us. The only important thing is that, according to these calculations, the durability of the 152-mm Kammel plate was equated to 13,45 dm of iron armor, “which is 2,24 times greater than the thickness of the tested plate.”
Further, the author of the article in Naval Annual refers to tests of a 5,75-in (146,05-mm) plate, surface-hardened using the Krupp method, at which 15-cm and 21-cm German shells were fired, “at speeds giving the thickness of the pierced iron armor according to the Tresider formula from 10,2 to 13,6 dm, which is 1,77 and 2,35 thicknesses of the experimental plate.”
Unfortunately, nothing is said about the results of the shelling, but it is concluded that “... the Kammel plate showed greater resistance to penetration than the excellent above-mentioned Krupp plate.”
Of course, there is nothing impossible about this, since during testing of the armor for Poltava, Krupp’s armor showed a “K” slightly lower than Kammel’s armor – 2.
Thus, the publication of the Naval Annual fully confirms the thesis that the best armor plates created using the Harvey method were quite on par and could even exceed the regulatory (minimum) requirements for Krupp armor. Moreover, at least one of the famous armor manufacturers shared the same opinion. The article stated:
“Our Sheffield factories (Brown, Cammel and Vickers) can be congratulated on the fact that they have resolved this important issue in this way. These three plants use nickel to make the armor supplied to our fleet, and they all acquired the right to use the Krupp process.
It is now believed by Kammel that equally good results are achieved by Harvey's processes using nickel; but Vickers and Brown recognize the advantage of the Krupp gas method, which gives a particularly favorable result in the production of thick armor plates, giving them particularly high viscosity.”
It is now believed by Kammel that equally good results are achieved by Harvey's processes using nickel; but Vickers and Brown recognize the advantage of the Krupp gas method, which gives a particularly favorable result in the production of thick armor plates, giving them particularly high viscosity.”
The same “Marine Collection” No. 1 for 1898 briefly describes the tests of the Krupp slab with a thickness of 11,8 inches (299,72 mm), which took place in Meppen in 1895. It is indicated that they fired a Krupp projectile weighing 712,6 pounds or 323,23 kg. This projectile struck the slab at a 9-degree deflection from normal, i.e., at an angle of 81 degrees to the slab surface at a velocity of 1 ft/s (993 m/s). It is stated that: “the depth of the holes was not given, but judging by the buckling on the back surface of the slab and light cracks, it must be recognized that the resistance limit of the slab was almost reached.”
Unfortunately, the source does not indicate the caliber of the projectile that was used to test the slab. But in the Marine Collection No. 1–2 for 1900 there is an article “Tests of armor plates carried out in 1898–1899,” which describes other tests that took place a year later, on June 5, 1896. It states that shells weighing between 712 and 718 pounds were fired from a 305 mm gun. Consequently, it becomes possible to calculate the durability of Krupp armor during the 1895 tests.
If the projectile had pierced the plate with the above initial data, its “K” should have been determined equal to 2. But the projectile still did not penetrate the armor, although it was close to this, accordingly, it is necessary to assume “K” to be no less than 168–2 170. Which, again, is quite comparable with the results of shelling the Krupp plate for Poltava and the standards for its production in Russia.
However, Krupp not only produced his armor, but also sold the technology for its production to everyone who wanted to purchase it, and there were many buyers. Among them, of course, were England and the USA.
For the needs of the Royal Navy
Two Krupp armor plates, manufactured under Krupp license at two different factories, were subject to testing. So, on July 12, 1898, firing took place on a 305-mm Krupp-Brown plate produced by the Atlas Works plant. They fired, again, not too heavy twelve-inch shells weighing 714 pounds, or 323,9 kg (did they collude with the Germans, or what?).
Three rounds were fired at armor impact velocities of 1, 852, and 1 fps, with none penetrating the armor. And this is not at all surprising, because even if the armor was overcome by a projectile with the highest speed of 856 ft/s or 1 m/s, this would indicate a “K” of only 849, while the armor Krupp's resistance was obviously higher.
The second plate produced by Krupp-Kammel formally seemed to have a thickness of 305 mm, but, according to the author of the article “Tests of armor plates produced in 1898–1899,” it was thinner. The fact is that in the British Navy they usually operated not on the thickness, but on the weight of the armor plate, and the author points out: “The thickness of the plate is not shown exactly, but its weight does not exceed 480 pounds per square foot. Taking this weight into account, we see that its thickness should be somewhat less than 12 inches, since in a slab made according to the Krupp method, 1 square foot should weigh 490 pounds. It can be assumed that its thickness was 11,66 inches.”
They fired three 12-inch Holzer shells that weighed between 718,5 and 719,75 pounds or 325,9 and 326,5 kg. That is, the armor plate was tested with relatively light projectiles, and even at relatively low speeds at the moment of impact: 1 ft/s maximum. Accordingly, it is not surprising that the plate was not pierced again - if it had been pierced by a 866-pound projectile at a speed of 718,5 ft/s (1 m/s), this would indicate “K” = 866 568,8. Obviously, the “K” of the Krupp plate should be higher, and there is nothing surprising in the fact that the armor plate was never pierced.
But the question arises: why didn’t the British, during testing, consistently increase the speed of the projectile on the armor, and didn’t achieve a breakdown of the plate, because this is the only way to confidently talk about the limits of its durability?
The answer, apparently, lies in the guns from which the shelling was carried out.
The British very actively developed their naval artillery, and in 1895 a very good artillery system, the 305 mm/35 Mark VIII, was developed and sent into mass production. This gun was installed on Majestic-class battleships and was armed with an 850-pound (385,55 kg) armor-piercing projectile. Subsequently, British battleships and the first dreadnoughts were equipped with armor-piercing shells of precisely this weight (probably different in design, but the weight remained the same) until the advent of 305 mm/50 guns.
However, Krupp’s armor was not fired at by the latest 305 mm/35 guns, as the weight of the shells used clearly indicates. Such shells were fired by 305 mm/25 guns, similar to those installed on the Colossus-class battleships and Collingwood, built in the 1880s.
Battleship Collingwood
“Short-barreled” twelve-inch guns provided, according to the passport, only 1914 feet per second maximum initial speed. It can be assumed that the guns from which the Krupp armor was tested already had some firing, and could not provide projectiles with velocities exceeding 1–856 ft./s on the armor. And the British were quite satisfied that the durability of Krupp’s armor in this case at least corresponded to the best examples of Harvey’s armor plates.
This is probably why Lord Brassey, to whom the article referred, pointed out: “After these experiments, the possibility of producing armor plates in English factories using the Krupp method, both thick and thin, should be considered fully proven.”
Tests in the USA
Tests of American-made Krupp armor are much more informative, because the shelling was carried out from more modern artillery systems than the antediluvian British 305 mm/25. As a result, in both cases, which will be described below, the Americans managed to penetrate the tested armor plates and determine the speed of the projectiles, which was close to the maximum that these plates could withstand.
In both cases, the armor produced by Carnegie was subject to testing; the thickness in the first case was 305 mm, and in the second - 152 mm. Shooting was carried out with shells whose caliber was equal to the thickness of the plate.
Three shells weighing 305 pounds were fired at the 850mm plate. The first, with a speed on the armor of 1 ft./s (833 m/s): “deepened 559 inches and, having settled in the slab, did not create cracks in it.” The second projectile, at a speed of 8,5 ft./s (2 m/s), penetrated the armor, but did so at the limit, as it got stuck in the lining, however, severely damaging it. The third projectile, having a speed on the armor of only 022 ft/s (616 m/s), quite expectedly did not penetrate the armor, penetrating only 1 inches into it.
Undoubtedly, the twelve-inch Carnegie slab showed excellent results. Considering the second, effective, hit as extremely close to the armor's maximum resistance, we get its “K” equal to or slightly below 2.
As for the six-inch Carnegie armor, it was tested on July 13, 1898. The slab was set on a backing of 12" thick oak and two 5/8" iron sheets - alas, it is not noted whether this is the thickness of one sheet or two sheets at once. Four shots were fired at the armor plate with 4 mm Carpenter shells, each weighing 152 pounds (100 kg). But we will consider only the first three, since the fourth shot was fired by a projectile with an armor-piercing tip. A photograph of this armor plate (after shelling) is in the title of this article.
Obviously, the third shot turned out to be extremely close to the maximum resistance of the armor: after all, the lining under the armor itself had negligible resistance. At the same time, the projectile itself was destroyed, that is, for a “clean” penetration, in which the projectile, even at the limit, would overcome the armor as a whole, an even greater speed on the armor would be required. But even so we get a wonderful “K” = 2!
Do we have reasons to mistrust the results presented?
Could it be that American-made Krupp armor turned out to be much better than German?
Quite obviously not. After all, domestic, Russian armor produced using Krupp technology demonstrated quite similar indicators: “K” = 2 for a 335 mm thick plate in relation to 305-inch shells and “K” = 12 for a 2 mm thick plate in relation to 566-inch shells.
Conclusions
In the course of analyzing the durability of armor plates manufactured by the Krupp method in Russia and abroad, attention is drawn to the excessive variation in the durability of Krupp armor in comparison with his predecessor Harvey. “Early” Harvey showed resistance “K” according to de Marr at the level of 1–700, that is, a difference of 1 units. The new method of harveying, which was invented and used by specialists at the Carnegie plant, provided “K” at the level of 950–250, that is, 2 units. But, as we can see from the test results, the durability of Krupp armor ranges from 000–2 or 200 units!
But there is an important nuance here.
The upper limits of Krupp armor resistance are shown by medium-caliber projectiles, that is, 6-dm, while heavy twelve-inch projectiles show a “definition” “K” in the range of 2–150, that is, a very reasonable 2 units. It can be assumed that the abnormally high resistance of Krupp armor against 400-mm caliber shells is explained by some peculiarities of its production, which do not apply to calibers over 250 dm, but, not being an expert, I cannot judge.
On the other hand, Harvey's armor also showed its record performance on medium-caliber projectiles. Is it possible to assume on this basis that the upper limit of durability of the “early” and “forged” Harvey (“K” = 1950 and 2, respectively) is applicable to the assessment of only 200–6 inch artillery systems, and for 9–10 inch projectiles the durability of Harvey will the slabs be slightly lower?
Maybe so, maybe not, unfortunately, I don’t have the knowledge to put forward such a possibility as a hypothesis. Perhaps in the future, having developed and expanded my statistical base, I will be able to make some assumptions in this regard.
In general, the data at my disposal today allows me to evaluate the comparative durability of armor plates made by the Harvey and Krup method in the following proportion:
If we compare the average values for large-caliber projectiles, we find that to ensure equal durability with Krupp armor, one should take the armor plate of the “improved” Harvey about 12% thicker than the Krupp one, and the “early” Harvey one – 37% thicker than the Krupp one.
But here I once again draw the attention of the dear reader that all the above conclusions were obtained empirically, that is, experimentally, based on a relatively small statistical sample of tests. And although they are to a certain extent confirmed by the opinion of specialists of the late 19th and early 20th centuries, you need to understand that their opinions were formed just as empirically - except that they had a larger sample. Accordingly, the results I obtained should be assessed as a hypothesis, but, of course, not as the ultimate truth.
I propose to continue to look for the test results of armor and projectiles, calculate them according to armor penetration formulas and, based on the results obtained, edit and supplement the picture presented in this series of articles. However, in the absence of compelling objections, I believe it is possible to use the data I obtained to model the capabilities of domestic naval artillery during the Russo-Japanese War.
Продолжение следует ...
Information