The Russians test the Pak 38 and its PzGr 39

Discussions on the vehicles used by the Axis forces. Hosted by Christian Ankerstjerne
critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 15 Jan 2018 18:19

attached below is a graphical plot of data from a soviet investigation into a sample of 134 frontline T34 losses from a NI48 study, covering the period august to sept. 1942.
The data used in this plot selected the 50mm impacts, which represented the quantitatively largest cause of T34 combat losses in this timeframe, though no information is given for range, deflection, type of gun (short 5cm KWK or 5cm KWK L/60 or 5cm PAK) or type of projectile (except for that APCR were rarely ever encountered). It might be mentioned that the data are alo biased for looking only into tanks knocked out in combat, not into those, surviving combat.

As can be seen the protection offered by the T34 highly sloped, high hardness MZ-2 hull armor is very good vs 50mm impacts, and particularely resistent for hits on the glacis. This hands over a tactical edge to the T34 by forcing the antagonist to aim for other parts, such as flanks or turret. Most of the hits accordingly occurred on the flanks of the tank, and of those, roughly half of them penetrated. This number is very good because a substantial fraction of flank hits failed to get through. The sloped upper side hull resisted 50mm impacts much better than the vertical side hull and we can expect that a large fraction of hits was oblique due to various target angle´s to the side hull.

The turret on the other hand did not resist 50mm impacts, except for the rare extreme obliquity ricochet which it jointly came across. Despite beeing a small target area, compared to the hull (particularely glacis, but also side hull), a significant number of tanks were knocked out by 5cm turret hits, indicating a good controll of the aim point. A reason for this occurance can be found in the low resultant netto impact obliquities against turret hits. Unlike the hull sides, the turret is usually trained towards the general direction of the enemy, and thus significantly reduces off deflection angles.
This combined with high hardness MZ-2 generates a condition where MZ-2 works against it´s protection due to it´s ballistic inferiority of this material at low obliquity to RHA when the attacking projectile cannot be broken up (5cm Pzgr39 APC-HE really exploits this situation here).
Another reason was the unsatisfactory use of MZ-2 for cast turret elements. MZ-2 (8-S steel) was specified to be RHA material, not cast material, causing problems with gas bubble inclusions and weakening of cast elements. With RHA, the gas bubbles would be cross rolled out, creating only very small tiny laminations rather than local weakenings but cast material was not mechanically treated. Only with the introduction of 71-L material in the T34/85 was this defect later corrected in 1944.
You do not have the required permissions to view the files attached to this post.

User avatar
Mobius
Member
Posts: 543
Joined: 12 Jan 2005 20:45
Location: Glendale, CA

Re: The Russians test the Pak 38 and its PzGr 39

Post by Mobius » 15 Jan 2018 20:14

50mm graph show terminal velocity at 1000 m. to be around 560 m/s. German tables say it is 591 m/s.

critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 16 Jan 2018 00:41

That´s correct. We don´t know how they derived with their figures, but their assumptions of terminal velocity for the 50mm PAK and KWKW L/60 appear to be too low downrange. Because the vulnerabiliyt diagrams prepared by NII48 are based upon firing trials and these terminal velocity graphs, the possibility exists that the area of vulnerability may have been somehow larger than given.

Peasant
Member
Posts: 249
Joined: 16 Oct 2018 17:21
Location: Italy

Re: The Russians test the Pak 38 and its PzGr 39

Post by Peasant » 03 Jan 2019 23:54

If you need a high quality scan of the said report, I've found it: http://www.mediafire.com/file/tjy4mnydz ... robivn.pdf

critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 05 Jan 2019 17:00

Thanks. One note on the type of penetration. At the sampling timeframe, a significant quantity of capped 5cm Pzgr39 was already appearing on the eastern Front, it seems fairly likely to me that this is reflected in the graph of my previous memo.
The pzgr39 would attempt to penetrate intactly, driving out a plug of armor. This is reflected by the soviets assessment that a large number of holes were "satisfactory" types of penetrations. Keep in mind that "satisfactory penetration" according to period soviet definitions required a hole smaller than 3.0 cal projectile diameter (150mm for a 5cm impact), which is considerably larger than what UK/US or germans defined (2.0cal/d).
Does anybody have the document OCR´d?

Peasant
Member
Posts: 249
Joined: 16 Oct 2018 17:21
Location: Italy

Re: The Russians test the Pak 38 and its PzGr 39

Post by Peasant » 10 Nov 2019 18:57

critical mass wrote:
05 Jan 2019 17:00
Something I've noticed, and can potentially be important to keep in mind when examining the data from this document, is that the results presented were obtained in two distinct batches: the first one at Sverdlovsk, and the second one at Gorohovez's(thank for correction) firing ranges.
The two used distinct batches of plates manufactured for the tests, where even those of the same thickness and type(alloy type and whether nominally tempered to high or medium hardness) show variations in thermal processing applied, which results in different hardness levels.

Image

Image

The results obtained at Sverdlovsk are shown in Table 5 and Picture 11, while those from Gorohovez are the diagrams 12, 13 and 14. The rest of the diagrams in the document simply show this data under different forms, but supposedly hold no original information which can be easily verified using the terminal velocities chart n.15.

A most striking example is the results of firing german 5cm uncapped AP (5cm PzGr.) against 75mm armour of medium hardness. While the plates tempered to 269 - 285 BHN resisted the hits well enough to present a PTP limit at 150m (striking velocity 790m/s circa) only with a perfect 90°strike, the other ones, tempered to the hardness of 302 - 331 failed at 0°with much lower striking velocity of around 640m/s . Since the limit determined was the PTP one, a backspalling can be assumed to be the cause.
Last edited by Peasant on 11 Nov 2019 00:50, edited 2 times in total.

critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 10 Nov 2019 22:16

Sverdlovsk and Gorochovez, if I read correctly.

There is a fair degree of chemical scatter between the high hardness 8s and medium hardness FD-6633.

The 5cm uncapped Pzgr result is shocking. Minimum acceptance requirement vs 5cm pzgr39 (without cap, decapped mass production shot used for testing armor) was 80mm RHA beeing Safe (no hole through or backspall allowed), to 900m/s at 90 deg.
Failure of a 75mm plate at 640-790m/s, Even considering the different thickness would constitute inferior plate quality. This is probably reinforced by the older, more inferior model 5cm Pzgr Gg, used in the soviet trials instead of pzgr39 o.K.
Thus, not only was the plate strength objectively inferior due to the lower limit velocity, it was also exhibiting this inferiority against the low projectile quality of old model service AP.

Yoozername
Member
Posts: 2083
Joined: 25 Apr 2006 15:58
Location: Colorado

Re: The Russians test the Pak 38 and its PzGr 39

Post by Yoozername » 11 Nov 2019 01:31

Has anyone read that the Soviet RHA was only rolled in one direction? That is, not cross-rolled.

Also, I would not expect tungsten carbide ammunition for the 5.0 cm guns to be rare in 1942.

Peasant
Member
Posts: 249
Joined: 16 Oct 2018 17:21
Location: Italy

Re: The Russians test the Pak 38 and its PzGr 39

Post by Peasant » 11 Nov 2019 16:01

Sometimes, in the mass produced lots:
Image
I would assume that the test plates are made with more care.

The german tests were conducted at 100m with the "v0 = 900m/s", which is reasonable to assume is the muzzle velocity. The actual striking velocity would be only about 866m/s. DeMarre K = 2418 vs K = 2380 in the soviet test, both plates 95kg/mm^2 strength circa.

Edit: I have found am error in the document. The graph 11 shows the PTP and PSP limit for 35, 45 and 65mm plates BUT there were no 65mm plates used in the testing, only 60mm ones. I believe they meant to write "60mm", so keep that in mind.

Image

Avalancheon
Member
Posts: 219
Joined: 23 Apr 2017 06:01
Location: Canada

Re: The Russians test the Pak 38 and its PzGr 39

Post by Avalancheon » 12 Nov 2019 19:37

critical mass wrote:
10 Nov 2019 22:16
The 5cm uncapped Pzgr result is shocking. Minimum acceptance requirement vs 5cm pzgr39 (without cap, decapped mass production shot used for testing armor) was 80mm RHA beeing Safe (no hole through or backspall allowed), to 900m/s at 90 deg.
Failure of a 75mm plate at 640-790m/s, Even considering the different thickness would constitute inferior plate quality. This is probably reinforced by the older, more inferior model 5cm Pzgr Gg, used in the soviet trials instead of pzgr39 o.K.
Thus, not only was the plate strength objectively inferior due to the lower limit velocity, it was also exhibiting this inferiority against the low projectile quality of old model service AP.
The test results are surprising in more ways than that. You would expect the harder plates to perform better against uncapped AP, and yet they actually did worse. Those plates must have been improperly tempered.

L/24Stug
Member
Posts: 10
Joined: 27 Jan 2019 11:23
Location: Spain

Re: The Russians test the Pak 38 and its PzGr 39

Post by L/24Stug » 12 Nov 2019 22:00

The test results are surprising in more ways than that. You would expect the harder plates to perform better against uncapped AP, and yet they actually did worse. Those plates must have been improperly tempered.
In fact, Russian harder plates had not much tempering at all.

critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 16 Nov 2019 14:44

I have to correct myselfe. The striking velocity at 100m (V100) is <900m/s for the Pzg.39 o.K. Thanks Peasant for pointing it out correctly, above.

The weight of the Pzgr.39o.K. is testified as 2060g in Miles table. This compares with the 5cm Pzgr.39m.K. = 1900g (notice, that the weight of the capped AP. is given in Miles table WITHOUT the weight of the cap as noted in the first page, and is therefore correct for the 5cm Pzgr.39). From this I gather that the Pzgr.39o.K. is not just a decapped Pzgr.39, as I previously assumed but a rather a "warmed over" 5cm Pzgr Gg (whether or not it at least received the Pzgr.39 heat treatment and welded on tip, as implied by the name, is unknown to me so far). If that is correct, the Pzgr.39 o.K. needs to be fired on the test range at higher than usual velocity from the guns (the V100 was there to determine the striking velocity, which if done like on other cases, was likely recorded by selenoids on the range).

De Marre K= 1890 for the poor 75mm FD-6683 plate (42-s) failing at 640m/s
De Marre K= 2332 for the good 75mm FD-6683 plate (42-s) failing at 790m/s, very approximately
De Marre K= over 2444 as acceptance minimum for german production 80mm RHA vs Pzgr.39 o.K.
(using the De Marre K formulas preferred by ARTKOM)

Thus, as can be seen, the good plate is within 95% of the minimum ballistic resistence of german RHA and thus not bad, while the poor plate has 77% of the minimum ballistic resistence. That beeing said, the good soviet plate would have just missed the minimum acceptable german procument standarts by a small margin (80mm minimum thickness to ensure safety for 5cm pzgr.39o.K.) on account of its ability to fail early against uncapped 5cm AP. The good soviet plate would require 85mm minimum thickness to ensure resistence of 5cm Pzgr.o.K., while the poor plate would require >110mm minimum plate thickness to ensure 5cm Pzgr o.K. safety (though probably little less due to the larger t/d ratio). The large amount of loss of effective safety caused by the poor plate is what made me characterize it as "shocking" above. It is, however, within the range of expectable results for an embrittled plate.

These medium hardness, medium thickness plates (f.e. for KV-1) were tempered (unlike the higher hardness 8-s grade products for the T-34), and my preliminary conclusion is that either the knowledge of, or the controll of tempering was not sufficient to allow consistently high quality products from different manufacturers. Keep in mind that unlike the compared with german data, these are not minimum acceptance tests but rather explorative tests and retests of production plates.

Yoozername
Member
Posts: 2083
Joined: 25 Apr 2006 15:58
Location: Colorado

Re: The Russians test the Pak 38 and its PzGr 39

Post by Yoozername » 20 Nov 2019 21:54

This is from the CIA report on the T34/85 that was manufactured in 1945 and captured in Korea.

https://www.cia.gov/library/readingroom ... 0001-4.pdf
crossrolled.jpg
You do not have the required permissions to view the files attached to this post.

Peasant
Member
Posts: 249
Joined: 16 Oct 2018 17:21
Location: Italy

Re: The Russians test the Pak 38 and its PzGr 39

Post by Peasant » 23 Nov 2019 21:31

critical mass wrote:
28 Sep 2017 12:58
I've got a question you may have the answer to: why the german RHA has such low G(s) limit at high obliquity? Those curves Miles posted, 1 T/D plate needs to be set at 65° to obtain protection from 5cm shot? (K = 1387)
I've seen the data on post war experiments with T33/M77 shot where similar armour(0,86 T/D; 65°; 280 - 320 BHN) under similar conditions was relatively much more effective at resisting based on Army/Protection limits (K around 1600).
https://i.imgur.com/5laE5Ze.jpg

critical mass
Member
Posts: 590
Joined: 13 Jun 2017 14:53
Location: central Europe

Re: The Russians test the Pak 38 and its PzGr 39

Post by critical mass » 25 Nov 2019 08:59

I think the differences are attributable to the different Charakter of the data and their reference in regard to performance.
Miles curves (Sollkurve and Abnahmebschusskurve) are minimum service acceptance references for service acceptance of armor plate. For a plate to be worse than the numbers, then the whole lot will have to be rejected (if failing retest). This by definition means that in general, acceptable plate quality such as procured for AFV armor is same or better than the reference. By how much is unknown.
The US data on the other hand are explorative trials, and record data as is obtained.

The two datasets are therefore not directly comparable.
There are some other aspects involved but I don’t know their actual effect yet. The nose shape at 0.8 t/D and high obliquely is irrelevant, as the projectile will effect base first penetration regularly at such acute angle. But larger cal projectiles require somewhat more energy for obtaining holing and perforation at high obliquity than smaller caliber projectiles. It’s not a big difference but it’s measurable. I have it in datasets for US 8in, 14in and 16in APCBC, in German navy APCBC and German 75mm and 88mm Army APCBC, and I strongly suspect that it’s true for 50mm vs 90mm, too. Perhaps worth investigating further.

Return to “The Ron Klages Panzer & other vehicles Section”