Dispersion data for KwK 42 L/70

Discussions on the vehicles used by the Axis forces. Hosted by Christian Ankerstjerne
Ulater
Member
Posts: 246
Joined: 09 Mar 2015, 20:36
Location: USA

Re: Dispersion data for KwK 42 L/70

#16

Post by Ulater » 07 Jun 2017, 09:41

Found something:

"Higher muzzle velocity also provides a much flatter trajectory (flight path) for projectiles. This was revealed during post-World War II testing of Panther tanks by the American military at Aberdeen Proving Ground, Maryland, in 1946, in which it was shown at 1,000 yards (914 m) that the 75-mm main gun on the Panther tank could place all of its shots within a 12-inch (31-cm) circle. Tests conducted at the same location the prior year on captured Panther tanks had demonstrated that the fired projectiles had such a flat trajectory that the gunner did not even have to change his elevation settings until he began engaging enemy targets at ranges greater than 2,000 yards (1.8 km)."

Panther - Germany's quest for combat dominance, page 59.

Well, its not a complete table, but its something.

critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#17

Post by critical mass » 13 Jun 2017, 17:33

Note that for comparative purposes that total deviation (~100%) in german sources was specified as beeing equal to four times the size of the rectangle where 50% of the shots fell. This rectable specified the distance from edge to edge. Dispersion figures were calculated excluding wild shots.

Soviet dispersion was specified as an oval where the radius from the centre to the horizontal or vertical edge was specified as distance. Ignoring the differences between rectangle and ovaloid shape for a second, it´s 50% dispersion figures need to be multiplied by 2.0 to approach german 50% dispersion figures.


critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#18

Post by critical mass » 16 Aug 2017, 18:09

according to Miles Krogfuss, the 75mm L70 KWK42 had the following dispersion:

"Using German data that I have, here are some of my Pzgr.39 figures to meditate about concerning 50% Breite/Hohe scatter:
100 meters 0/.1,
500 meters .1/.2,
1000 meters .2/.315,
1500 meters .45/.55,
2000 meters .6/.8,
2500 meters .9/.11,
3000 meters 1.1/1.5."[/quote]

viewtopic.php?f=47&t=229856

The 50% zone of dispersion at 1000m is extremely low. A 50% zone for dispersion in height of 0.315m equals a probable error of 15.8cm, a true mean dispersion of 19cm and a standart deviation of 23.3cm.
Even the 50% zone at 2000m is among the lowest figures, which can be encountered. Out of curiousity, is there record of any ww2 service tank gun which was rated more accurate?

User avatar
Mobius
Member
Posts: 645
Joined: 12 Jan 2005, 21:45
Location: Glendale, CA
Contact:

Re: Dispersion data for KwK 42 L/70

#19

Post by Mobius » 16 Aug 2017, 19:11

critical mass wrote:Even the 50% zone at 2000m is among the lowest figures, which can be encountered. Out of curiousity, is there record of any ww2 service tank gun which was rated more accurate?
The US 76mm gun firing the M62 round may rival it. In an average of US and UK data it had horizontal deviation of 0.28m and vertical of 0.225m. (This included pro-rated from 500yd data though.) I don't have long range deviation data.
And even more impressive deviation came from the Russians.
http://tankarchives.blogspot.com/2014/0 ... llcat.html
76mm M1A1 or M1A2
13. The gun has good precision. At a range of 1200 meters, average horizontal deviation is 0.16 meters and vertical deviation is 0.1 meters.
That translates to 50% zone h=0.267m v=0.167m at 1000m.

The 3" M7 didn't do as well in Russian testing. Russian tests as 50% zone at 1000 m: 0.48m horizontal 0.28m vertical.
A comparison of 75mm KwK 42 vs US data 76mm M62 deviation.
Attachments
comparison75vs76.jpg

critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#20

Post by critical mass » 16 Aug 2017, 20:24

Thanks. That´s an outstanding gun accuracy.
Just a minor note, I think You mixed up horizontal with vertical dispersion here.
download/file.php?id=405244&mode=view
A-Б-B
first three letters of russian alphabet (spelled a-b-w[or -v at times])

BA = length (dispersion in range)
BБ = deflection (horizontal dispersion)
BB = height (vertical dispersion)


"na distanzia 1200 metrov BB= 0,16 metra, BБ=0,10 metra.."

http://ic.pics.livejournal.com/yuripash ... iginal.jpg

indicates that again, figures for probable error have been used here, as could be expected in russian sources.
This single shooting thus had a 50% dispersion zone at 1200m = 0.32 m & 0.20m, each for height and deflection, respectively. Almost identic to the PANTHERs dispersion at 1000m (0.315 & 0.20m).
Considering the variances, -a single shooting with new gun vs officially rated performance from firing tables (firetables are referenced against 1/3 worn guns) it´s probable that they both are of equal accuracy.

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#21

Post by Yoozername » 17 Aug 2017, 17:58

Supposedly, the US 76mm HVAP was among the most accurate.

12th ARMY GROUP TESTS AT ISIGNY: 20-21 AUGUST, 1944
c. The 17pdr APCBC and the 76mm HVAP, T4, are both highly accurate ammunitions. In the opinion of the members of the board, two of whom have had considerable experience test firing British and American tank and antitank weapons, the 76mm HVAP, T4 is the most accurate tank or antitank ammunition encountered to date.
The early US 76mm weapons had a problem with the rifling. It was initially 1 in 40 and later changed to 1 in 32. The Soviets even noticed the problem in the early Hellcat TD they were sent. It was still accurate but it did not fly 'true' or it came uncorked at a angle. This would be an issue with a muzzle brake. The M1A2 finally got a muzzle brake, so I would assume the rifling change helped. The later US ammunition was also modified with a better propellant layout to reduce the smoke from firing. If you can't see your tracer, it effects accuracy!

User avatar
Mobius
Member
Posts: 645
Joined: 12 Jan 2005, 21:45
Location: Glendale, CA
Contact:

Re: Dispersion data for KwK 42 L/70

#22

Post by Mobius » 17 Aug 2017, 19:31

Yoozername wrote:Supposedly, the US 76mm HVAP was among the most accurate.
12th ARMY GROUP TESTS AT ISIGNY: 20-21 AUGUST, 1944
c. The 17pdr APCBC and the 76mm HVAP, T4, are both highly accurate ammunitions. In the opinion of the members of the board, two of whom have had considerable experience test firing British and American tank and antitank weapons, the 76mm HVAP, T4 is the most accurate tank or antitank ammunition encountered to date.
Weapon Research Committee: accuracy of anti-tank guns and rigidity of gun barrels", 1948 [PRO piece number WO 195/10134],
Gun............Ammunition.....m.v.(f.p.s.)....Average mean deviation of strike
76mm M1A1..... APC............2600.................0.5'
....................AP/T...........3400.................0.5'
17-pr .............APCBC..........2900................0.8'
76mm AP/T is HVAP. It has the same deviation as the APC but with it's higher velocity it has a flatter trajectory so would have a higher hit percentage when range estimation is included. To find the deviation distance at 1000 yds determine the tangent of the angular degrees then multiply by 1000.
[Tangent (0.5/60)] x 1000 x 0.84535 x 2 = 50% zone deviation of 0.2359 yd.
(Tangent (0.8/60)) X 1000 X 1.6907 = 50% zone deviation of 0.374 yd.

Below are excerpts from reports on 17 pdr accuracy with compressed title pages.
17pdraccuracy1j.jpg
17pdraccuracy1j.jpg (170.06 KiB) Viewed 1962 times
2.44 times the 50% zone is the 90% zone. So to get the 50% zone divide by 2.44.
(Tangent (3.3/60)) X 1000 / 2.44 = 50% zone deviation of 0.393 yd. This is pretty close to the 0.374 yds as found in the WO 95/10134 report.
17pdraccuracy2j.jpg
17pdraccuracy2j.jpg (110.75 KiB) Viewed 1962 times

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#23

Post by Yoozername » 19 Aug 2017, 16:32

76mm AP/T is HVAP. It has the same deviation as the APC but with it's higher velocity it has a flatter trajectory so would have a higher hit percentage when range estimation is included. To find the deviation distance at 1000 yds determine the tangent of the angular degrees then multiply by 1000.
Yes, so it has better accuracy. Accuracy is nothing more than hitting what you are aiming at. Dispersion should not be confused with accuracy.

i suppose if there was a plot of strikes showing a dispersion pattern, and they were numbered in sequence, one would find the strikes actually 'climbing' as the weapon was fired. The 'cold shot', or initial shots, would be lowest. In fact, I read a M60 tank zeroing procedure and they dicated that the weapon was to fire a round first before trying to zero. This is similar to someone using a torque wrench first on an object before calibrating it. unlike the torque wrench, in combat, you might not get that first 'warm-up' shot.

When doing these dispersion studies, repeatability of the weapon tube is critical. Nowadays, it is easy with lasers or CMM like machines. Hopefully, they had some mechanical calibration equipment to determine the repeatability of the lay.

To look at something like the 88mm Firing Tables, and the ranges they have data for, I can't help but think that it would be of little use to a soldier. It might be needed to design the sights or determine the range you could even try to zero at. By that i mean that if the dispersion is too great at, say, a 1000 meters, you are kidding yourself if trying to hit a 2 meter tall target and adjust to a spread onto the target.

critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#24

Post by critical mass » 20 Aug 2017, 11:53

Accuracy as a term is general.
For tank guns and flat trajectories, it can be spread up in different aspects:

A) dispersion -presuming the line of fire is same, grouping of impacts and deviation of it from MPI
B) MPI error (in this plane to a large degree pointing errors) -the distance between target centre and MPI
C) ballistic differences (time of flight to hit target)
D) consistency

In many cases I know, range tables or firing table dispersion is somehow larger than tested dispersion data from new guns (17pdr, 122mm, 88mm).
Therefore test data and FT-data should be treated seperately.

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#25

Post by Yoozername » 20 Aug 2017, 14:50

I am speaking in engineering terms. Specifically using ISO definition (ISO 5725). Image

According to ISO 5725-1,[5] the general term "accuracy" is used to describe the closeness of a measurement to the true value. When the term is applied to sets of measurements of the same measurand, it involves a component of random error and a component of systematic error. In this case trueness is the closeness of the mean of a set of measurement results to the actual (true) value and precision is the closeness of agreement among a set of results.

The dispersion can be expressed as precision. If that precision can be adjusted onto a target or 'zeroed', we have accuracy.

Image Low accuracy, good precision, poor trueness ImageLow accuracy, poor precision, good trueness

critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#26

Post by critical mass » 20 Aug 2017, 19:20

The perfectly symmetrical, normal distribution form shown in the sketch is a presumption, but one, which doesn´t necessarily always holds true and one, which need to be checked for. Usually, the dispersion distribution shows a skewness, caused by an impulse, particularely at low range. Sometimes, bimodal distributions are encountered (caused by rather subtle differences in the exterior ballistics, such as driving bands, or coupling with stabilization of the projectile in flight). Under normal distributions, there are relatively strict relationships between true mean dispersion, standart deviation, 50% zone and 25% zone. However, rarely the sample size is large enough to give reliable data, that´s why FT data are more reliable than single (preselected) tests, which can go one way, or the other.
F.e. the true mean deviation of the 76mm AP M62 and HVAP was 6" at 1000 yd from a limited test. Another, single 1946 Aberdeen accuracy test claimed that 100% of all test shots did fall in a 12" wide diameter circle for the 75mm L/70 KWK42 (i.e. would have required a significantly lower dispersion error than shown by the 76mm AP M62 and HVAP test if the sample is taken to be representative). Both data are better than their guns official FT data, but do not need to represent representative samples.

I suggest to stay in ballistic terminologies. Accuracy is an exercise in understanding various types of error.
MPI errors, dispersion errors and consistency errors define accuracy.

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#27

Post by Yoozername » 20 Aug 2017, 20:03

I am pointing out that accuracy is mainly limited by the ability to zero the weapon. In engineering, which ballistics includes, we talk about uncertainties. If one is to zero a weapon against a 2mx2.5m (hxw) target at 1000 meters. And you want to do that with 3 shot 'groups' (fire three shots and make some adjustment from the results), you better have a very tight 50% or you might just be chasing your tail. The 88mm FlaK weapons had something like 0.7 m hohe at 1000 m? Zeroing at that range might even have over and under shots on a 2 m target.

In engineering terms, the dispersion of the weapon, no matter what the shape, is called a characteristic. That is a fundamental concept.

As mobius pointed out, flat trajectories from high velocity weapons can achieve accuracy 'forgiveness' since the actual flight path will intersect targets either closer or further away mitigating ranging errors. i suppose the fact that 88mm FlaK weapons DID have range finding equipment, they sort of had built-in range error mitigation. I remember discussing this with rexford awhile ago.

Having tall targets, like the sherman and later t-34/85, also helps achieve 'accuracy' as can be imagined. Dispersion is dispersion, it is not accuracy.

critical mass
Member
Posts: 740
Joined: 13 Jun 2017, 15:53
Location: central Europe

Re: Dispersion data for KwK 42 L/70

#28

Post by critical mass » 20 Aug 2017, 20:32

Kind of. But then again, what You call "accuracy" is the general term on the supra level covering more than one item below it´s level. Dispersion error, and MPI error (pointing errors for flat trajectory guns) and ballistic trajectories are just specific aspects of the general term of accuracy.
I have long since proposed to use the uniform standart deviation criterium rather than 50% zones to assess dispersions.

That beeing said, the problem is true in the 2nd part, that is that too large 50% zones may completely cover potential MPI errors. A single shot ranging does not disclose the difference between MPI and pattern error at all. You don´t obtain a mean point of impact but just a point of impact, which has to be judged in relation to target. The tabulated dispersion figures for 50% zones are a key information for the operators here to judge how significant an observed deviation X at distance Y is in regard to what would be tabulated as normal dispersion error. Figures exceeding that can be expected to represent incorrect rangeing and/or aiming errors. The more the observed deviation from the aiming point exceeds the 50% zone figures, the more likely they cannot be caused by mere dispersion.

88mm FLAK weapons had -compared to tank and anti tank guns- fairly sophisticated rangefinders and mechanical computers to establish range and slant ranges and even to calculate 3D firing solutions, a precision not really required vs surface targets at relatively close range.

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#29

Post by Yoozername » 21 Aug 2017, 00:32

Yes, I am taking a Systems and sub-systems approach to this. Horse in front of the cart. And the weakest link is, again, the zeroing procedure.
The tabulated dispersion figures for 50% zones are a key information for the operators here to judge how significant an observed deviation X at distance Y is in regard to what would be tabulated as normal dispersion error. Figures exceeding that can be expected to represent incorrect rangeing and/or aiming errors. The more the observed deviation from the aiming point exceeds the 50% zone figures, the more likely they cannot be caused by mere dispersion.
If by 'operators', you mean the soldiers using it, then you must think they can ascertain the tracer being in the 50% zones?!?!?? And IF they could, why would your logic hold? A firer with perfect range measurement (AND a precisely zeroed weapon) can still expect strikes outside the 50% zones by 3-4 times the area.

The dispersion is a characteristic of the ammunition variance, the tube and elevation/traverse variances, i.e. a sub-system. The inclusion of the sight, soldiers, tracer observation is a closed loop System.
I have long since proposed to use the uniform standart deviation criterium rather than 50% zones to assess dispersions.
So did I.

Yoozername
Member
Posts: 2615
Joined: 25 Apr 2006, 16:58
Location: Colorado

Re: Dispersion data for KwK 42 L/70

#30

Post by Yoozername » 21 Aug 2017, 00:56

88mm FLAK weapons had -compared to tank and anti tank guns- fairly sophisticated rangefinders and mechanical computers to establish range and slant ranges and even to calculate 3D firing solutions, a precision not really required vs surface targets at relatively close range.
They certainly used coincidence range finders

Image

Post Reply

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