Metallurgical Study Of Enemy Ordnance

Warship design and construction, terminology, navigation, hydrodynamics, stability, armor schemes, damage control, etc.
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Metallurgical Study Of Enemy Ordnance

Post by Dave Saxton »

I recently ran across a US military study from Feb 1945 untitled simply: Metallurigical Study Of Enemy Ordnance. It examines a wide range of both German and Japanese metalurgy, including shells and armour.

A central theme of the peice is the shortages of strategic materials in Germany. However they do point out that Germans seem to find a way around these problems, and usually maintain the same performance standards, by engineering new materials. This is confirmed in other German research reports from 1943, giving the performance results of molybdenum free steels (The began using vanadium more and changed the, Carbon to Manganese ratio, along with different heat cylcling ...ect..)

The report gives examples of German tank armour composition from different years, and how it evolved in compensation for reduced availablity of Mo and other strategic materials. They make pains to point out that German homogenious armour post 1934 is a Chromium-Molybdenum type armour, not the more common Nickel/Chromium armour type. They felt this was done as integral metalurgical design approach, and not in anticipation of nickel shortages.

"Most German armor whatever it's thickness, or if it's homogenious, or flame -hardened, relies on chromium and molybdenum alloying agents...(a chart comparing earlier Panzer III armour, to later Tiger Tank armour is put here)
..in most German armor, the chromium may run as high as 2 1/2 %, the molybdenum from 1/2 to 3/4%. The most common figures for tank armors are 1.5% Cr and .5% Mo. When the Cr content is down near 1%, silicon may be boosted to 1% or above, but more commonly the Mn is raised to 1% or above. If Vanadium is present, it will be about 0.20%, but it may be absent. Molybdenum is always present in much more than residual amounts. Nickel is only present as a residual (Wh and Ww naval armour contained Ni in more than residual amounts, but it was flip flopped in amounts with the Cr content, compared to more standard armour compositions. DS) In a few cases it may run 0.25, or even 0.40%, but in many more cases, it hovers around .015%, and ofton only a mere trace is present...even in cr/mo armour, nickel at the levels in the analysis of Czech tank armor listed in the first four lines of table3 (may be present.) However, the German homogenious armor, as shown, runs consistantly lower in nickel... According to the water town arsenal report the 1 1/2 inch plates from a PzKw IV tank contained 0.40% C, .30% Ni, 1 1/4% Cr, .17% Mo, no V, and .0015% B. The Mo content has been dramatically reduced from that found in earlier samples. The boron figure may mean the Germans are playing with "needling" ( "intesifiers" or "reaction" alloys), although there was no need for this in this instance, as it has ample hardness for the section, with out intensifiers, All samples evaluated have meant the usual German minimum standards for homogenious armour of 130,000 psi tensile, and 20% elongation. "

The PzKw II armour composition was .40%C, ,25% Si, trace Ni, 1.3% Cr, .50%Mo, .18% Cu and .17% V. The Tiger tank armour contained no V or Cu, but the C was upped to .50%, the Cr upped to 2.55%, and Mo actually increased to .60%.

The Czech tank armour ran ~.50%C, 1% MN, 1.10% Si, .25-.50% Ni, 1.5% Cr, .30% Mo .07% V .18% Cu.

By way of Comparison, Tirpitz's prewar Wh ran ~.28%C, .35% si, .32%Mn, 2.29%Cr, .93% Ni, .48 Mo, .17% Cu, and no V. There are simlarities to the earlier PzKw III armour, but it should have been more ductile, with a welding friendly, less than .30% C, and no V.
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

One of the things that stand out in the 1945 US report, is speculation about the German use of spaced arrays. In about 43 the Germans began to use spaced arrays in most of their armour protection systems, extensively. The American opinions and assumptions of this turn of events is interesting:

"Dr. Ritchie comments on the late use of two tempered plates seperated by 4 to 6 inches by spacers, instead of one thick plate. '(spaced) armor is not liked by many authorities as well as a single, thick plate, although there are significant data to indicate that, while not giving a profound advantage expected by some theorists, it comes pretty close to acting as well as a single plate, other things being equal. But, in order to make other thing equal, a thick single plate must be tempered through (his italics), and to secure through hardening consistancy in heavy plate, alloy must be piled in-whereas thinner plates requires much less alloy for hardening. From the point of of alloy conservation, it makes sense to use a pack of relatively thin plates of lower alloy steel rather than a single heavy plate. ..In view of the doubtful value of mere spacing of plates, it seems likely that the real German purpose in using two plates seperated by spacers was to save alloy, by avoiding a single heavy plate. The stunt of of limiting plate thickness to that which avoids slack quenching of steel, with a limited amount of alloy, and using more than one such plate, instead of a single thick one, looks to be good engineering, and worthy of emulation when alloys need saving, or when the armor gets so thick that slack quenching cannot be avoided, even when the steel contains lot's of alloy'."

What I think, one concept Dr Ritchie may have been alluding to; is the possibility of a steel containing greater amounts of Cr to have microstructure grain size growth, when allowed to remain at high temps for extended periods. This would be un-avoidable in heavy plate, as the outside would cool more rapidly, but the inside may remain too hot for very long periods. The results could be a heavy plate that is more brittle towards the center, with inconsistant microstructure through it's section thickness. Some Cr/Mo plate does indeed change from ~24% elongation (in 2") to ~18% elongation when improperly heat cycled, during welding. We know that the Germans usually restricted Wh to 15cm or less, section thickness.

It is interesting that we find American experts, during WWII, alluding to data supporting the notion that spaced arrays can provide an effective thickness at least equal to that, of a single thick plate of the same total thickness(provided the single plate is properly quenched and tempered).
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

In the Metallurgical Study Of Enemy Ordnance, the American experts seem impressed by the quality of the German AP shells of all sizes and applications. One of the German inovations they seem impressed by, was the construction of welded tip capped AP shells, also with precisely controlled differential heat treatments. In 1940, the Germans began using a new design capped AP. This new APC had a shell tip of different composition welded on to the main body, and over this it used a new heat treated cap with a more blunt shape. The newer APC gave much improved performance during oblique impact:

" Armor-Piercing Shells- A projectile in which hardenability is necessary, and to which very careful heat treatment is applied. The German AP have a mild steel windshield spot welded to a penetrating cap, which in turn is soldered to a projectile body that is very hard at the nose, and much softer at the base. On oblique impact, the penetrating cap is supposed to break through the hardened zone on the armor, before the cap cracks up or flies out of the way. The hard, sharp, nose of the projectile proper digs further into the armor, before it in turn cracks up. The tougher body and base of the projectile stay together and as a flat nosed punch. If the body spalls off, so as to be conical, it would be more likely to slide off than penetrate. To increase the probabilty of spalling on a more or less straight line, the Germans have welded a high carbon tip to a lower carbon base. A hardness survey of the (pre 1940) one piece projectile and it's pentrating cap is shown in figure 4. and the duplex welded on nose projectile in figure 5. (see also figure 6) The way the two tend crack up is shown by the grinding and etching cracks, developed in sectioning and etching them. The two designs are probably both uniformly quenched and differentially tempered. To avoid quenching cracks the the hardened tip and the base should be closely the same and the weld must be perfect.

The introduction of welded APC illustrates the length the Germans will go to achieve a desired result. It seems two be a question in the minds of American ordnance experts whether the welded projectile has a real advantage over a solid one, of proper hardenability, with proper diffrential heat treatment, but the Germans do think it has a advantage. This welding job is know easy matter. The forged tip would have to be held in a special electrode fixture, the rolled base in another, with special precautions to get good electrical contact. for the very high amperage required. The butt welding equipment to apply the pressure and generate the power needed to make the weld would be decidedly expensive. Such equipment has been made available to all the various sizes of projectiles to allow for this fussy extra step.

...A beautiful welding job has been done on the projectiles that are put in service. Anything less than a perfect weld might well cause cracking during quenching.

The Chronology of this welded class of projectile is interesting. Back in the copper driving band era, (and some of the un welded projectiles), and the penetrative cap were of steel containing around .40%C 3.5% to 4% Ni, 1% Cr, and anywhere from .05 to .55% Mo, while others were of high Cr steel. Then in the era of the duplex driving bands , had about .60% C, .80% SI, .90% Mn, 1-1.5% Cr, .30-.50% NI and only small amounts of MO. "

The photos and diagrams of the two APC designs are interesting. The caps are very different in both shape and mass. The brinell hardness mapping of the two shells show the newer design to be much more dfferential, with a more precisely controlled gradiant in hardness change. It would appear that the new design would fair much better in oblique impacts.
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Interesting material. I have seen similar post war investigation reports prepared by the British. Of course one can never really get a feel for the ability of the Germans to maintain performance standards when faced with shortages in strategic materials unless the armor is subjected to ballistic testing. One would also have to look at side-by-side ballistic performance of early German armor steel relative to late war armor steel to determine if there was no real decrease in ballistic protection. Conversely I suppose one could compare ballistic performance of late war German Armor Steel relative to say British or American armor steel.

Does the report contain any ballistic limit testing data?
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

Hi,

I havn't had the opprotunity to spend much time on line recently, so I apologize for not responding to your question earlier.

No, it doesn't give specific mathmatical velocity data on ballistic limits. The report was done by US metalurgy experts, and it reports the typical metalurgy data such as the chemical compositions and the results of mechanical dustrustive tests such as ultimate tensile strength, yield tensile strength, elongation in 2 -inches, reduction of area, brinell hardness, impact toughnesss ...ect...

Their evaluation of Japanese armour and metalurgy is also quite interesting, in light of many modern assumptions though.
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Dave Saxton wrote: and the results of mechanical dustrustive tests such as ultimate tensile strength, yield tensile strength, elongation in 2 -inches, reduction of area, brinell hardness, impact toughnesss...ect...
These are easy enough to conduct side-by-side comparisons relative to early war sample testing. Or we can compare the material properties with British and/or American armor steel of the period. WHat are the material properties for some of these German Armour steels you have been examining?

ultimate tensile strength, and
yield tensile strength,
elongation in 2 -inches,
reduction of area,
brinell hardness,
impact toughnesss.

Or; what is the report title you've been examining? Perhaps I have a copy of the same report here in my references.
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

Hi Marty,

I must offer an apology for my late reply once again. I had started to write a few times but ran out of time. I had wanted to include more detailed data from documents I have recently encountered, but I only have limited time.

The title of the peice is Metalurgical Study Of Enemy Ordnance. It was printed in Feb 1945, but owing to the time lag expected from writing and printing, one would expect the actual materials evaluated would have been no later than early 44. Keep in mind that this appears to mainly be an evaluation of samples recovered from AFV's, and aircraft. Examples of German naval armour, as actually worked into the ships, would have been rather difficult to obtain during the war, or even for sometime following VE day. It is mentioned that they recieved British evaluations of some naval armour from 1940 (Possibly Graf Spee?) that meant the same performance standard they had encountered in the other samples of German homogenious materials, up to that point.

They mention that all samples had a UTS of least 130,000psi, YTS of 80,000psi, reduction of area 70%, elongation of 20%. If the Germans were able to still obtain this perfomance standard consistantly from mid 44 to VE day, is rather doubtful in my opinion. In 1965 the top US Army expert in ballist materials at that time, Col. George Brady (he was also involved in thechnical missions) wrote concerning the war time German materials, in the context of new advances in tank armour materials. He mentioned that the Germans made remarkable progress in the use of manganese and other more common alloying materials, with ingenous use of heat treatments. He also cited the high perfomance standard obtianed by their 18-8 stainless steel welding electrodes through the use of manganese. He mentioned that the Germans themselves considered the late war materials to be strictly erszat, compared to the earlier materials. Nonetheless, in the writings of Dr C White (a member of the technical mission) he mentioned that the late war armour materials still obtained a minimum of 130,000 psi tensile and 20% elongation in destructive tests. White offered as his opinion, that the German materials late in the war were no better, or perhaps not as good, in manfacture quality, compared to allied efforts, based on their rather rough appearance, particularly the cast armours.

What 20% elongation really means compared to other elongation figures is not easily defined. I have done numerous elongation and tensile destructive tests, and it can vary considerably from sample to sample, and it depends on how the measurement is taken. A strap is cut out, usually with a oxy gas torch, and marked; in the center, and at measured distances out from the center. The peice is then bent with a press with the apex of the bend in the center, and the amount of distortion is measued from the center to the various marks. The amount of distortion in the center will be the most, but is a highly unreliable figure. Therefore, the elongation measured at two inches, or 8 inches are normally used. It is customary to use the elongation in two inches for high strength steels like armour plate, and to use the 8 inch figure for mild steels. The figure at 8 inches will be a few % points less than the elongation in two inches. For example, British DW steel elongation tests gave 24% at the apex, 22% in two inches, and 18% in 8 inches. The Germans used slightly different elongation measurements standards.

The Germans had a habit of using the worst case scenario as the official elogation rating for a material, and they used an additional elongation measurement for transverse deformation limits. This was call elongation in "quer". The other was in "lang in 50mm" or the same as elongation in 2". For example, Stahl 52 construction steel was officially rated by the German Navy at 21% elongation, but four samples of St52 from the late 30's, and early 40's gave elongations of:

32.4% lang/31.4% quer,
29.9% lang, 24% quer
30.9% lang, 26.1% quer
28.2% lang, 21.1% quer

Note that the worst case elongation figure was the one that matched the official rating. This particular sample had a UTS of 63kg/mm2, while the ones with about 30% elongation had a UTS ranging from 54kg/mm2 (the 1942 sample) to 61kg/mm2.

What 20% dehung for the homogenious armour may really mean, might be clearer by examining some elongation ratings of experimental German Chromium-Molybdenum high strength steels published in 1931. Two of the materials were submitted by Krupp, and the chemical compositions closely match those tested of the Wh and Ww recovered from Tirpitz in the 50's salvage. The first of these two Krupp materials Called FK CR/MO Stahl, had an elogation of 20% in quer and 23% in lang. The other Krupp material called FK 340-000, had 24% in quer, and 27% in lang. Several other materials from various manafactures were also examined. One had 5.5% Cr and .4% MO, with no nickel. This material contained vanadium too. This material was much like mid-war German tank armour. This material obtained 20% elongation in quer, but when heated to 400* C, it actually declined to 18%. It remained at 18% after being heat cycled to 400* and allowed to normalize.
George Elder
-
Posts: 168
Joined: Mon Oct 18, 2004 4:23 pm

I am impressed...

Post by George Elder »

If you have material that needs to be translated, please advise.

George
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Just a test to see if this image will link to the forum.

Image
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Image
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

About a year or so back I had been looking at ballistic limit velocity as a function of plate hardness & tensile strength. These are a couple of the figures I threw together sometime ago to get a better handle on some of the material properties for a couple of US manufactured armor plates. The data was derived from a post WWII report written by NPG-Dahlgren in 1946. There was no Izod or Charpy test data included with this particular set of test data.

I have always had a mind to sit down and compare physical properties of German, British, and American armor steel relative to ballistic protection, but haven’t as yet had the time or inclination. However I think it would be an interesting exercise.
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Just a side note, but optimized RHA & STS plate hardeness relative to ballistic limit varies as a function of both t/d (e/d) and attack-angle/obliquity. Limit velocity increases as both obliquity and hardness increases -- up to a peak hardness at which point limit velocity will again begin to decrease. If t/d is held constant at say unity, at very low obliquity a softer armor yields a higher limit velocity. Conversely at high obliquity a much harder armor is ideal.
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

I wish I had the computor savy and the gear to post charts and graphs like that! A picture can say it so much better ofton times. It seems that the tensile strength increase of STS for hardness, and/or reduction of ductility, follows about a 1/2 slope linear function. Some of the charts I have seen, give about 7/6 linear slope for carbon steels, and 1/2 slope for alloy steels. My guess is that molybdenum bearing materials like NCA and Wh would follow a similar aproximate intial slope, but may be shifted upward relative to the ductility curves. There's strong evidence that Mo content of about 0.5% dramatcally lowers the transformation temp of notch ductile to notch brittle condition, due to the grain refining aspects. This would indicate a great improvement in toughness, or the strength vs ductilty properties. It seems that Wh was obtaining about 133 kSI ( 93kg/mm2 ) at an elongation % in the mid to lower 20's. The harness at about 125 ksi, was about, IIRC, 240 brinnell. Given the copper content this seems very reasonable, and right in line with expectations. The actual tensile relative to the ductilty varies quite a bit, depending on the condition of the steel. For example, an American high tensile steel with a UTS of 100-120 KSI gives a completely different response, depending on the heat treatment, all with the same composition. Although not graphed, it charts like this:

Normalised from 815*--97 KSI UTS, 56 KSI YTS, 23.5 % elong. W/ coarse grain
88 KSI, UTS, 54 ksi YTS, 29% elong. W/ fine grain size

Normalised from 870*-- 99 ksi, 52 ksi 22% elong, coarse grain size.
89 ksi, 52 ksi, 28% elong fine grain size

Normalized, quenched & tempered from 565*--116 ksi, 81 ksi, 20% elong coarse grain.
110 ksi, 77 ksi, 23% elong, fine grain size.

With the fine grain size condition, the impact toughness is more than 3X the coarse grain size impact toughness rating. (90 ft-lb's vs 15-25 ft-lb's)

Since Wh is a Cr/Mo alloy system instead of a nickel hardened alloy like NCA or STS, or even AOD, it will be very likely to encounter chromium grain size growth if mis-handled, or subjected to temper brittleness, although the use of .47 %Mo should help.

I doubt that Wh also follows a linear function for increase in tensile vs increase in hardness....
Last edited by Dave Saxton on Sat Feb 26, 2005 10:19 pm, edited 1 time in total.
User avatar
Dave Saxton
Supporter
Posts: 3148
Joined: Sat Nov 27, 2004 9:02 pm
Location: Rocky Mountains USA

Post by Dave Saxton »

Wh probably follows an slight expotential function for increase in strength vs reduction in ductilty, instead of a linear function, due to the employment of copper and Ni as micro alloys, in a molybdenum, fine grained alloy matrix. When I first learned that Wh and Ww used ~.17% Cu, I was a bit taken back. I have investigated the possibilty that they may have been doing this to manipulate the tempering temps, and the arrest points in the cooling schedules. Indeed, there is some allusion to this in the primary literature. In 1937, American metalurgists from Carnigie Steel, toured Krupp's electric arc furnace facillities at Essen, and noted that Krupp was tring to manipulate tempering schedules through non traditional micro alloys(noteably boron). However, I recently ran into an exaustive American study of the use copper in fine grain steels from 1945.

The study was done by two Metalurgists at the NBMI, named Mathews and Rosenthal. They had several heats of a carbon/manganese high tensile steel rolled with a nominal UTS of 112,000 psi, at 25% elongation, when at 0% copper. The heats had varing amounts of copper added from 0-2,5%.

M&R expressed the opinion that in amounts of less than .2%, the Cu was probably simply serving an anti-corrosion function, since the amount had to be greater to create a significant copper/carbon precipitation strengthing function. They found that the copper reduced the rate of corrosion by about 50%. However, the impact on the physical properties, while less profound, were not insignificant.

The most notable change was an increase in brinell hardness of 20-40 points, with no reduction in the elongation and reduction of area specs. Your finding, that a subtle increase in hardness improves performance vs oblique impacts is rather interesting in this regard. Actually in the Q&T condition the ductilty specs increased a few % points, but in the normalised condition this boost was not noticed. The no ductilty improvement in normalized condition is collaborated by at least two German studies of ST-52, that could contain up to .3% Cu.

When the Cu % began increasing beyond .2%, the impact on the strength curves began to be profound. The UTS, YTS, and Hardness, began increasing at an expotential rate, with the ductilty declining at about the same rate as your STS graph does with increases in hardness. At 2.5% Cu, the YTS was actually up to about 120 KSI. The elongation had fallen from 25% to 17% at that point with the ratio of YTS to UTS approaching 1:1.1 The effect on the YTS was more profound than on the UTS. This would explain why the Germans remained prudent with the use of Cu, foregoing the potentially great increase in YTS.

The more I learn, in my opinion, Wh (in a fine grained condition) looks to have been a remarkable material for the time. It's ratio of YTS to UTS seems to be about the widest, and at given ductilty rating, it seems to have been about 17% stronger than most the nickel/chromium homogenious materials. That Japanese MNC material may have been a very interesting material too.
marty1
Member
Posts: 219
Joined: Tue Nov 02, 2004 6:31 am
Location: Seattle

Post by marty1 »

Dave:

I wonder if we should ask Javier to move this thread to the Naval Technology portion of the forum?

In the mean time I will try and throw some additional lab data together for US or British Armor steel.

Perhaps -- when you have the opportunity – you would be kind enough to elaborate a bit more on the subtle differences between plane-Jane Class-B armor and German Wotan Harte. I guess my own understanding of Wotan-Harte has always been a little vague. Is it simply a high-hardness RHA? Thanks.

P.S. If you ever need to post any images here let me know. You can always email the image to me, and I would be glad to make sure that it gets posted or linked to this thread.
Post Reply