George Forty Research on PzKpfw IIIH Front Hull Resistance

[rexford]

GERMAN TANKS OF WORLD WAR II, by George Forty, has an interesting passage on page 67 regarding the effective resistance of the PzKpfw IIIH front hull (32mm over 30mm, both face-hardened by many but not all accounts).

"from late 1941 many PzKpfw IIIs had extra face-hardened plates fitted, for example, on to the frontal armor, which defeated the 2-pdr and 37mm except at very short ranges. British tank gunners were complaining that their AP shot just bounced off the enemy tanks."

It is unfortunate that a reference is not given.

Anyway, 2-pdr and 37mm uncapped AP have 100m face-hardened penetration of 62mm and 65mm (vertical target armor), which suggests that 32mm/30mm on PzKpfw IIIH upper front hull was at least as effective as a single 62mm face-hardened plate, if not more so.

The British tests in Cairo during May 1942 suggested that 32mm/30mm resisted like a single 69mm face-hardened plate.

The penetration resistance figures from Forty's passage and the Cairo tests suggest that two face-hardened plates in contact did not lose resistance compared to a single face-hardened plate of the same overall thickness.

In addition, U.S. firing tests against the 30mm/50mm face-hardened plates in contact on the front hull of PzKpfw IV's suggested that the effective resistance was similar or slightly greater than a single 80mm thick face-hardened plate.

If the PzKpfw IIIH had carried 32mm/30mm homogeneous plates in contact, the effective resistance against 2-pdr AP hits would have been about 73% of the overall thickness, or 45mm. The reduction in penetration range due to using face-hardened armor instead of homogeneous would be 700m down to about 100m.

[ November 07, 2003, 10:48 PM: Message edited by: rexford ]

[MrSpkr]

Interesting. Two questions -- what was the time period the dual plates were used (was it mid-1941 on, or what), and will CMAK reflect this new research?

Steve

[ November 08, 2003, 01:07 AM: Message edited by: MrSpkr ]

[kmead]

As I recall, the III had spaced armor added only to the upper hull front on the plate that runs across the driver and radio operators faces. It was a riveted/bolted on construction and was added to most IIIs in the field.

[Mr. Tittles]

major questions are:

1. Are the plates parallel?

2. Do they remain that way under attack?

3. Does the AP penetrator reach its diameter width through the first plate before hitting the second?

4. What effects on the penetrator does the loss of spin have? Does it pitch/yaw/roll?

5. What effects would having melted metal on teh penetrator from the first plate have on the second plate?

[JasonC]

"which defeated the 2-pdr and 37mm ***except at very short ranges.*** British tank gunners were complaining that their AP shot just bounced off the enemy tanks."

Meaning, at very short ranges they did not just bounce off, they went in.

"2-pdr and 37mm uncapped AP have 100m face-hardened penetration of 62mm and 65mm (vertical target armor)"

And 100m certainly qualifies as very short range, so they should on this testimony penetrate that much. Meaning, 30+32 is penetrated by something that is rated to penetrate 62mm vertical. Meaning, there is no increase in effectiveness for layering, as opposed to one face hardened plate of equivalent thickness.

Not exactly what you said. It means the layered is not better than unlayered, not that layered is at least as good.

Which had been in dispute in the past. This testimony (one item obvious) says layered does not improve performance. It might not detract (remains to discuss), but it can't improve, or the statement would not be "except at very short ranges", it would be absolute.

Notice further that this is for vertical armor without side angle. Also, the figures are for 100m, while the report is for "very short range", without specifying how far that means. Given tactical realities in the desert it probably means under 500m - longer than that might have seemed "close" but not "very close", since many engagements took place between 500 and 1000m.

Some reduction in effectiveness for layering would predict penetrations starting somewhere between 100m and 700m - when the side angle is zero. It is quite likely that reduction is not a full 27%. They'd have noticed and reported if 2 pdr and 37mm was still effective out to 700m. But how much more can we infer from this report?

Not a lot, it seems to me. It is a tactical, in practice report, so it presumably reflects a dispersion of side angles. "Very close range" might mean only 100m or it might include 300m. 100m flat says "0 reduction, 0 increase". 300m at varied side angles, perhaps 15-20, says something more like "up to half the maximum possible layering reduction".

Layered plates do not resist better than single, but when hardened and facing uncapped AP, they may well resist somewhat better than suggested e.g. by the naval equation, i.e. not 73% of total effectiveness. Which is, incidentally, a larger figure than I've typically seen for two even plates - I usually see ~.85 estimated for that. Or, it is the thinner plate only that is ~.73, while the thicker is 1.00 - which would give 54mm or .87 effectiveness to 32+30.

Now, 54 lies basically midway between 45 and 62-65, so linearly interpolating it would suggest first vulnerability at around 400m. Which typical tactical side angles might reduce to 300m. It is not obvious that any additional correction is necessary to account for the report.

Uniform failure at 500m, combined with only occasional side angle sensitive success in the 300-400m range, and reliable success only closer still, would be reported as failure at combat ranges and success only at very close range. But so would uniform failure at 250m, with first successes at 100m and reliably ones never or only flat.

Which means the report tells us the proper layering correction term is probably somewhere between 0.87 and 1.00. Which we could have guessed beforehand. I'd use something like .9 or .95 to minimize the likely error, without pretending to know it wasn't 1.00 on the one hand, or .87 on the other.

[ November 08, 2003, 08:00 PM: Message edited by: JasonC ]