You are on page 1of 77

UNCLASSIFIED

AD NUMBER
AD301343

CLASSIFICATION CHANGES
TO:

unclassified

FROM:

confidential

LIMITATION CHANGES
TO:
Approved for public release, distribution
unlimited

FROM:
Distribution authorized to U.S. Gov't.
agencies and their contractors;
Administrative/Operational Use; MAY 1958.
Other requests shall be referred to Army
Ordnance Department, Research and
Development Division, Washington, DC
20310.

AUTHORITY
d/a ltr

28 jan 1981; d/a ltr

28 jan 1981

THIS PAGE IS UNCLASSIFIED

D,'ECLA$SIFICAI 0bN
ANuEDULE
THIS DOCUMENT

is:

CLAU IFIS________
SiJ~t to Serd Iudn*Mk cgkhlt~ of
Ernth1rkIMN
1a
iyd ungered at
2 You Igmgs., HAWEI SI SECEMIEI 31
ftlfm howsit,'"~ Ciotti
ftolst too~ Aleicy

ImXI$~h

DISCLAIMER NOTICE
THIS DOCUMENT IS BEST QUALITY
PRACTICABLE. THE COPY FURNISHED
TO DTIC CONTAINED A SIGNIFICANT
NUMBER OF PAGES WHICH DO NOT
REPRODUCE LEGIBLY.

ui'

~~-I

4W0 I CI

mcul

1416

LTIC
gem"USI

3UM
I I2

,,.a,

UAWm SM
~AA

-49
yeON

-tau"w'

v*i*i

4L)LJ J.'A

~NtL4

FV

c4

SIF1ED.

IL

UNC'A SS' FlED

II'I
['rmed Olf,,vices Technical Information, ,gency
,L

"

ARLINGTON HALL STATION


ARLINGTON , 12 VIRGINIA

NOrICE: W IEN OVERNMENT OR OTHER DRAWINGS, SPECIFICATIONS OR OTHER DATA


WR ff-D
I DR ,n 'Y PURPOSE OTHER THAN IN CONNECTION WITH A DEF1NITELY RELATED
GOVE7tNMEV T P," OCUREMENT OPERATION, THE U. S. GOVERNMENT THEREBY INCURS
NO RESPONIBIJ, YY, NOR ANY OBLIGATION WHATSOEVER; AND THE FACT THAT THE
GO VERNMEVTI A:,Y HAVE FORMULATED, FURNISHED, OR IN ANY WAY SUPPLIED THE
SAID DRAWE IGc,' PECIFICATIONS, OR OTHER DATA 13 NOT TO BE REGARDED BY
IM13LICATION OT., , )THERWISE AS IN ANY MANNER LICENSING THE HOLDER OR ANY OTHER
PE.ASON OR "OOR 1.t ',RATION, OR CONVEYING ANY RIGHTS OR PERMISSION TO MANUFACTURE,
USEA OR SEL!, AN7 PATENTED INVENTION THAT MAY IN ANY WAY BE RELATED THERETO.

CONFIDENTIAL

Final Report

SANANALYTICAL STUDY OF DATA ON ARMOR


\PENETRATION BY TANK-FIRED,
>5KINE

ISAENERGY PROJECTILES

WARNING
fTSUrAL SM1
M TM WAX
NATO
AMUNOF TMB MOM&=*S 1AYS
Too
r V aAUC.,
SCTP W1 A M
, A b.6
TW tVIIAU Or RD5 0
MONSO CMU
TO AN UUArlrt
TOM%, P ANT MOpMl
Mwii

il i

LAW.

1
I

BATTELLE

MEMORIAL

CONFIDENTIAL

INSTITUTE

SLP

' 1958

I
!

.----

---

p
-a,.

.-

-.. --

- .

w~

This document is the property of the United States


Government. It is furnished for the duration of the contract and
shall be returned when no longer required, or upon
recall by ASTIA to the following address:
Armed Services Technical Information Agency, Arlington Hall Station,
Arlington 12, Virginia

NOTICE: THIS DOCUMENT CONTAINS INFORMATION AFFECTING THE


NATIONAL DEFENSE OF THE UNITED STATES WITHIN THE MEANING
OF THE ESPIONAGE LAWS, TITLE 18, U.S.C., SECTIONS 793 and 794.
THE TRANSMISSION OR THE REVELATION OF ITS CONTENTS IN
ANY MANNER TO AN UNAUTHORIZED PERSON IS PROHIBITED BY LAW.

4
-

CONFIDENTIAL
FINAL REPORT

on

AN ANALYTICAL STUDY OF DATA ON ARMOR PENETRATION BY


TANK-FIRED, KINETIC ENERGY PROJECTILES

Ord Project No. TB3-1-Z4B


D/A Project No. 5B0304004

to

ORDNANCE DEPARTMENT
RESEARCH AND DEVFLOPMENT DIVISION
DEPARTMENT OF THE ARMY

31 May 1958

by

J.

C.

Bell and L.

E. Hulbert

Contract No. DA 33-OI0-ORD-1l4,

BATTELLE MEMORIAL N
;5 King Avc-"iie
Columbtus 1, Ohio

ttUTIL

CONFIDENTIAL

CONFIDENTIAL
I

I
TABLE OF CONTENTS
Pa ge

SUMMARY..

BACKGROUND.....................

ASSEMBLING THE DATA.

.. .

...........
...........
.

EVALUATION OF BALLISTIC LIMITS

OBSERVATIONS CONCERNING THIS STUDY


Success of the Methods of Analysis ..
.

.....
..

5
5
5
7
10

13

.....

Probit Analysis for Substantial Tests.................13


Choice of Formula for Fitting Penetration Data..
Maximum Likelihood S:lution for the Ballistic Limit
. .
Precision of the Ballistic Limit ......
.......
Probit Analysis Without Variation of Hardness
Brief Analysis Used for Cases With Few Data
..........
Compilation of Ballistic Limits

Application of the Results

.
.... .

Search for Data .................


Selection of Data to be Tabulated
.
Editing the Test Results
of
Data
Found......
Classes

..

13
14
18
3
24.......
4

25
.

.........

APPENDIX
BIBLIOGRAPHY

...

I
CONFIDENTIAL
ONt

# I %

Al
N

TA

CONFIDENTIAL
AN ANALYTICAL STUDY OF DATA ON A kMOR PENETRATION BY
TANK-FIRED, KINETIC ENERGY PROJECTILES
by
J. C. Beli and L. E. Hulbert

SUMMARY

This report d.rie_.asurveo A rden Proving


data on penetration of
homogeneous armor by kinetic energy projectiles. The projectiles were of the types
that have been or might be fired from tanks, with calibers ranging from 37 mm to
155 mm. The data were collected for individual rounds and pertinent information for
each round was punched on IBM cards, one card to a round. The data processing
began only when all similar rounds fired at similar targets had been assembled. Insofar as possible, ballistic limits for each projectile-target situation were computed by
carefully designed statistical procedures (programmed for the IBM Type 650 Magnetic
Drum Data Processing Machine) with which measures of the precibion of each limit
could be stated. Through a careful and continuous study of the data, sevral ob~ervations were made concerning factors which affect the dependability of results from punetration tests. Finally, the re.sults of the survey, based on o,er .0,000 rL'uds ot , r
50 kinds of projectiles, are compiled into a large table. This table should furnish a
convenient reference for penutration data, and shuuld serve as a step toN-.ard furthcr
correlation and condensation of the data.

CONFIDENTIAL

CONFIDENTIAL
z
BACKGROUND

The conventlonal means tor defeating tank armor has long been projectiles w'hich
penetrate the armor by virtue of their kinetic energy at the instant of impact. Many
studies, mostly empirical but some theoretical, have been made to find how fast a given
projectile must be traveling in order to penetrate a given piece of armor. The theoretical work is difficult, so simplifications must be introduced, and these generally prevent
the results irom having broad applicability. On the other hand, experiments with regular projectiles are expensive, so many ei.rly experiments were conducted with relatively limited test conditions.
One way to get fairly broad empirical results economically is to restrict attention
to small caliber projectiles. A widely quoted formula from the National Physical
Laboratory* was based on experiments with projectiles from 0. Z96 in. to 1. 565 in. in
diameter against plates 5 to 80 mm thick, having Brinell hardness 250 to 4S0 kg/mm 2 .
Later work permitted the formula to be amended to include the influence of obliquity of
the line of fire with respect to the target. As the formula was proposed it predicted
velocities required for intact projectiles to achieve penetrations complete in the Navy
sense, that is, with at least half the weight of the projectile passing completely through
the plate. In order to broader. the usefulness of the formula, means were developed in
Project Thor** for adapting the formula to broken projectiles, and for predicting penetrations complete in the Army sense, tiat is, where light shows through the hole when
the projectile is removed. Still this work referred to small projectiles, ranging in
diameter from 0.30 in. to 37 rmm, and most of the target plates hzd thicknesse. of an
inch or less. Data for the Project Thor correlation were drawn largely from a single
Aberdeen Proving Ground Report***.
The NPL iormula, as quoted in the Project Thor Report, is:

mV&.
L

43.4/

d3

e
3t!
-sec-+9Z9

ll,400

6-

54.000

where
VL

ballistic limit in ft/see

diameter of the projectile core in inches

t . ...
.' tffreC

T1w (u

L , tfj * .

flaubq*| l
s
n

~%. P. 1. fpwru
ruklare,

-1#r

I lnfkaopp,
mp

sa

it.,S

fAfttg

'u mark
i

LI

nartoluArW faw

?rCb01r PvO.r

to a WIv

i ""'w.'a

oraioaf1c 1" fPrti',vwabt

fql i a#~,p
cr ,4#kql- ia
4
Usi
4f$VW I Wkt cowdiSaI.
,4

.klIt4'W
P. 1. I
.ow
l.
.
%a6mw "Igfet*r
.

tIt a

t. 4-., %,
.6

# 'VW

CONFIDENTIAL

it' Is. P"Op' I it

'A.46
1

'..n"
il i

CONFIDENTIAL
3
B = Brinell Hardness Number in kg/mm 2 units
e = thickness of the armor in inches
Bo = 500 - 160 log 1 0 (d

d' = 1. 565 in., the diameter of the British 2-pounder shot


G = angle of attack, measured in degrees off the normal.
The corrections to this formula devised .or Project Thor are shown graphically in the
same report.

Although the studies mentioned were iimited to small caliber projectiles, there
are many data referring to full-scale tests with the kinds of projectiles used by tanks,
that is, from 37 to 120 mm, and some even larger. The Aberdeen Proviag Ground has
conducted many experimental programs with such projectiles for various (though often
limited) motives. Thus, a moderate amount of firing was done in testing early designs
of new projectiles. One substantial program (Project AX 23) tested various steels for
a single projectile, the 90 mm AP, T33E7. A long series of related tests (Project
2864) was designed to show the effect of hardness of the plate in withstanding the shot,
and since plate obliquity was known to be important, variations were introduced in that
too. Fairly recently, a large, more fully organized program (Projects TB4-150M and
TB4-10A) was performed for testing the effects of plate hardness, obliquity, and type
(whether cast or rolled*), using more modern projectiles. In addition, many other experiments were performed for a variety of special reasons.
Since the voluminous results of these full-scale tests were quite scattered, the
Ballistic Research Laboratories undertook to assemble the results into a more concise
form. Two reports were published on this subject. The first** was simply a collection of ballistic limits as reported in the many firing records. The second report*P*
provided an empirical correlation of these limits. The number of ballistic limits considered in the correlation process was 2364, with 1475 for rolled and 889 for cast
armor. The projectiles included I I varieties of AP shot, 9 of APC, and II of HVAP.
There were, of course, many obhoquit_,s and plate thicknesses, chosen to make 4ultable
targets for the projectile; being tested. The results consisted first f a series of
graphs showing how the rntas, red ballistic limits varied with the tid ratio (armor
thlckness/core diameter) for given projectiles, obliquity, and armor type. Second.
a set of predictive curves was fitted in the formn is
aa
^al
PC mk
. w i.
fht artl"ot. Af'e ' v s;d) .A twheutta it vsi,
,"i '1jeI.t r t 11etJ
e
A1#
At,-.I A 11191n OII-l &llll, Ith
rt I'M
1 t l Pelt I lo"*).
i.A ) . An A M lnl) 4. ;),. +Ak
A I V e mftn,# MITlF *w1
911 0l , 1 I , 1 *4k#l 'r,,+, h t *44 C iMl AIV,'O l P U Tl, 4fi . '
Lit " )

"ltw to,"
l~p+'
Wfo,*lrg
ft ,h ^ .t

"T'16eA
4,r 1

40

,
i ..

ltI

,sl,+',,
a i "
r rd+

I ,..i

is. "nmltf IM14,i4


l
l,

.nf
a t

,fjl

alt-+
Nove,**.

IMA,?

i'W&41M~t V ,*"At
W C

'I,

04fim

.ld
A'

,ur. II. J
rm, .,, %

A%,
A L ie'
r

rs

'I A
I

O ll t

i,

'

4f

e Ftsl' tVIS,#

1,44&l4d

iP, r J[IIW ;.., r 1,

~IL I

I 910 O
0I,

4 &of d

$4,j o ,

k m lla

LA,%q'
-*" ,q04p1

1t 00l 6 ! !

CONFIDENTIAL

"+

CONFIDENTIAL
4
Ballistic Limit = K 0 (t/d)n,

where n = 0.85 for AP projectiles, 0.78 for APC, and 0.65 for HVAP; and K., was tabulated for each combination of armor type (rolled or cast), projectile type (AP, APC, or
HVAP) and obliquity. Generally speaking, the correlations were fairly good, but some
substantial discrepancies between predicted and measured values remained.
Brief inspection of the ballistic limits tabulated in the BRL collection shows the
difficulty of choosing a representative ballistic limit for any test condition, because the
tabulated limits for many conditions scatter badly. This is not surprising since the
bulk of the limits were found as average velocitiei of only one complete and one partial
penetration, even though more extensive tests usually produce both complete and partial penetrations over a wide range of velocities. The question arises whether there
might not be some advantage in combining the penetration data from all similar tests in
order to compute an over-all ballistic limit. The resulting limit %houldbe less equivocal, and it may be possible to get confidence limits for it. It should be remembered
that this over-all limit can be based on result4 from many more rounds than contributed
to the limits in the BRL collection, since the rounds discarded in getting individual
lirits
can be retained in getting the over-all limit. In addition, results from other
kinds of tests, such as shot or plate acceptance, may be considered for use, even
though they were never part of a ballistic limit test.

I
I

It is the plan of collecting round -by-round results, instead of individual ballistic


limits, that forms the basis for the present work. It is recognized, of course, that
confusion may result from mechanically combining data from many divergent sources,
possibly having systematic discrepancies. However, any such confusion would seem
to be a legitimate part of the results, provided all the data are equally relevant.

I
I
I
I

CONFIDENTIAL
COA
t

0 ftlofDA 0

Oft

UL

CONFIDENTIAL
5
ASSEMBLING THE DATA

Search for Data

It was known that firing records showing individual rounds were available at several military librar'es, but the most promising pLace to find records for tank-fired
projectiles was the Technical Information Branch at the Aberdeen Proving Ground. A
total of about 6 man-weeks were spent searching the firing records stored there, selecting those which seemed to contain relevant tests. After the pertinent records were
found, the TIB prepared duplicates and sent them to Battelle Memorial Institute. This
large work of reproduction (totalling about 14,000 sheets) proved to be of great assistance later in sorting and editing the data. The volume of records found at the TIB
seemed to be about as many as could be handled in the present program, so little further searching was done except through the Armed Services Technical Information
Agency at Dayton, Ohio, where a few more records were found.
During the search, certain limitations were placed on the kinds of records to be
considered. First, they needed to refer to kinetic-energy projectiles, as ofposec' to
HEAT or HEP or other explosive shots designed for the penetration of armor. This
retained only the AP, APC, HVAP, and discarding sabot varieties of shot. The caliber
of the weapon was required to be between 37 mm and 155 mm. The plate neec'ed to be
homogeneous, so face hardened, spaced, and siliceous cored armors were orritted.
Finally, no armor was accepted unless at least I inch thick.

Selection of Data to be Tabulated

The medium to be used for assembling the data ultimately was to be IBM punched
cards which could store 80 numerical or alphabetic characters. It seemed most dedirable that each test round should be represented by its individual punche!d card, and
that all working information about the round should be on that card. Since firing rec ords come in a great variety of formats, containing various spectra of information,
considerable planning was required to decide N-hat data should be recorded on the cards.
In order to know the origin of each shot card, some notation of reference was
needed. Since the basic record is the Firing Recordp its number was recorded together with the page and round number of the shot. (The AD or other project -eports
almost always show the raw data by appending several Firing Records.) In order to
show associations of the test, a record was made of the AD report number (if any), or
else of the library book number (if any), These reference data made it possible to
check or amend the data by returning to the original firing records. They also helped
reveal accidental duplication of records. (Many firing records occurred in two or
more places in the library,) In addition, a coded notation was added #Ao show the kind
of test involved, whether acceptance or experzmental, whether testi ig shot or plate.

*ATtLL9

MIOA
t

IAL

# 0 i ItIt,

?t

CON FI DENTIAL
*~~

I~

I&

--

*-

CONFIDENTIAL
6
The firing records may list many bits of information about the armor plate, sach
as nominal thickness, actual thickness, type (rolled, cast, wrought), physical properties including surface or cross-sectional hardness, impact tests, chemical composition, heat treatments, manufacturer, heat number, and so forth. Out of all these
attributes it was decided that only the plate type, average Brinell hardness of the surface, and nominal thickness were indispensible. The actual thickness was tc be recorded and used in the analysis if it were available, but the nominal thickness could be
substituted for it. The Charpy or Izod impact test result also was to be recorded as
an alternative or supplement to the hardness in case some need or means were found
for so using it. Chemical composition and heat treatment were thought to be so complicated of description and so intertwined in their significance that it was unprofitable
to include them directly. Along with other data, they could be had by returning to the
original records via the reference punches, if this were ever needed.
Description of the shot in most firing records is implied in the model number.
Therefore it was decided to describe the shot by assigning each model a three-digit
code number. The codes were assigned fairly systematically, so that a person conversant with the system could recognize the shots fairly readily. In records about experimental shots, where many minor variations of shot might be made without changing
the model number, double punching with "x" or "y" was added to distinguish between

separate varieties. This system allowed these shot varieties to


merged, depending on how the computing machines are wired*.
fair number of records list hardness measurements of the shot,
ness at the nose was listed as a bit of stand-by information. No
shot were available broadly throughout the firing records.

It might have been better to record toxe precise obliquities, when

available, on the punched cards, planning to compensate for obliquity roundoff by performing a velocity adjustment, but this was not done, at least partly because of scarcity
of precedent. One other test condition which is reported in connection with climatic
tests is the temperature. This was eliminated in the present study by omitting the cold
weather tests.
Atiother test specification commonly recorded is the weapon used in firing the
shot. Since the rifling of the weapon has been a matter of occasional interest in plate
penetration, presumably because of the effect of yaw, it was decided to keep a record
of the weapon. To this end, each model of weapon was assigned a two-diglt cod., %t
preference was liven to the tube model number in case that differed from the weapi-i
model number.
*For exampe, *he APVO mm M7 was awigrnd Code Sli. Tr code sit ueast the "in
computing machines teognize the ldotel c
as llof AiII, depend
a, .I: .14l.

A t I LL a

" I

MORWAL

10

3
m
I

be distinguished or
In addition, since a
the Rockwell C hardother data about the

Certain test conditions, too, were indispensable for any analysis, namely the
obliquity and the striking velocity. Striking velocities almost always are reported to
the nearest foot per second, so they were taken in that form for the present analysis.
It was intended to adjust those velocities by an appropriate factor (the square root of
the ratio of nominal and actual plate thicknesses) in order to compansate for irregularities in the plate thickness. In this way, the tests could be divided efficiently into only
a few plate-thickness groups. The obliquities, too, needed to be divided into a few
groups, but this w.s accomplished merely by rounding the obliquity to the nearest multiple of 5 degrees. In most of the firing records, the obliquities were already rounded
in this form.

#X wmht a w*uiItel,

I
I
I
I

3
I

-he

stTVt a

CONFIDENTIAL

I
CONFIDENTIAL
7
Editing the Test Results
record. For penThe result of a given test is, of course, a vital part of the test
the penetration is
etration tests, the result is usually described by saying whether
been used. This ampartial or complete, but several definitions of completeness have
attempted, such as in the
biguity is most unfortunate when a synthesis of data is being
present project.

of penetration
In order of severity, the criteria commonly used for completeness
projectile
the
of
weight
are (a) the Navy criterion which demands that at least half the
requires that some
shall pass through the plate; (b) the protective criterion which
behind the target plate
fragment shall perforate a weak screen placed a short distance
the Army criterion
(the screen material and thickness seem to vary somewhat); (c)
plate when the shot is
the
in
hole
the
which requires only that light shall show through
remnoved.

I
I

The firing records for penetration tests shov. evaluations of completeness by at


Firing
least one of these criteria, but no one criterion i& common to all the records.
directly or
records often rate a few rounds for completeness by two or more criteria
to the
according
chosen
implicitly, but it is more common to use just one criterion
results on the
purpose of the test. The firing records alio usually describe several
and
plate, such as extent of bulge, cracking, spalling, and hole size, but the emphasis
of
bits
these
all
thoroughness of description varied widely with the observer. From
penetration
information, it was necessary in the present work to construct a uniform
rating for each test round.
The Army criterion seemed to be most commonly used in the better firing reccould be
ords assembled for this study. Moreover, it appeared that the Army criterion
other descripapplied most satisfactorily when ratings needed to be inferred from the
completeness of
for
tive material. Therefore, it was decided to try to rate every round
recorded
penetration by the Army criterion. Ratings by the other criteria were to be
of the
author
the
by
implied
be
when they were shown explicitly or evidently iltended to
firing record.

3
3
I

5
3

On the basis of experience in reading the records, some rules were developed for
deducing an Army penetration rating. Except when there was clearly some unusual
or
spalling, CP(P)* was taken to imply CP(A). If the only rating shown was PP(N)
2P(P), then the question was referred to the deuct,,tive matter. Any measurable hole
on the rear implied CP(A). Also, when there was mention of a large bulge un the plate
this
with punching started or with a substantial crack (longer than the plate thickness),
was taken to imply CP(A), unless it appeared that the Author was unduly lavish in his
On the other hand, the rating was set at PP(A) if a cracked bulge was
descriptions
only medium or small, or if the crack was very short. This method of assigning Army
ratings by noting bulges and cracks v is checked oftcn when A-my rat-ngs were given in
the firing records, and the method appcared modciAtely dependable. Occasionally,
they
when protective ratings were given alone without any other descriptive matter,
not
was
expedient
were assumed to coincide with Army ratintr, but fo rtunately this
needed often. Rarely, in view of persistent doubt about it, no rating was assigned.
*That is-,

omplcte penttilion C by the prokctive t ntwras (F).


gORN *
M
t
I I LL
iA

A L

I*NST

U
1 1I

CONFIDENTIAL

it

CONFIDENTIAL

i
I

8
One other relevant test result appeared to be the rupture of the projectile. For
example, several test programs have distinguished between ballistic limits for shattered as opposed to unshattered projectiles, and the Project Thor analysis distinguished
between broken and intact projectiles. Therefore, in the present project, a brief recFive classes were recognizedord of projectile rupture was included for each tea.
intact, broken but not shattered, nose shattered, other shatter, and no comment. The
firing record cescriptions varied, so the classification was sometimes difficult, but
plausible guesses could be made.
In order to prepare the data for keypunching onto cards, workheets were prepared on which an editor could assemble whatever information he could gather from any
firing record. The worksheets were arranged in the order that was to be used on the
punched cards. Vertical columns were allowed for the following kinds of information:
Class of Information

Card Columns

Reference Data

I
I
I

Library Number
Book Number or AD Number

1-8

Page Number in Library Book

9-12

13

Kind of Test (Plate Experiment, etc.)

Firing Record
Record Number

14-23

Page Number

24,Z5

Round Number

26,27

1
1
!

Active Identification

Shot (code for caliber, type, model)

31-33

Test Obliquity

34, 35
36

Plate Type

37-40

Nominal Plate Thickness (to 0. 0I in.)

1
1

3
I

I
SA ?t

LL 9

MgMOR1AL

IIUI
1t

CONFIDENTIAL

CONFIDENTIAL
9
Class of Information

Card Columns

Active Data
Actual Piate Thickness (to 0.01 in.)

41-44

Plate Hardness, Brinell

52-54

Striking Velocity

62-65

Penetration Rating (Navy, Protective,


or Army)

+ or - in 68,
69, or 70

Occasional Reference Data


Projectile Rupture Code

71

Weapon Code

7Z,73

Plate Impact, Charpy or Izod

74,75

Shot Hardness, Rockwell C, on Nose

76-78

The worksheets served not only as a place to store data but also as a reminder of
the data that were bting sought. This latter function was really important since the information was often scattered throughout a report, and at times had to be se rched out
of more than one report. Since several people did the work of editing, and they could
not afford individual decisions on the problems which kept arising, the problems too
were jotted onto the worksheets until they could be settled.
Considerable pains were taken to get all the really necessary information about as
nany valid test rounds as possible. Information regarded as necessary was that listed
under Active Identification and Active Data. If entries under those headings could not be
completed, the test was rejected, except for a few that were admitted withuut the
Brinell rating. Another reason often used for rejecting a test was that it was really not
a penetration test. This objection applied to most projectile-through-plate tests, and
tests involving proof projectiles. Still other tests were rejected because they were isolated tests with poorly rated experimental projectiles.
When the firing records reported ballistic limits computed on the rounds reported therein, those limits too were jotted on the margin of the worksheets.
While the editing was performed bf several people, _t was checked in the end by
one person who attempted to insure uniformity as well as accuracy of the work. Then
the worksheets were sent to keypunch operators who punched the data on cards and then
verified the punching. In this way, reasonable accuracy seens to have been achieved
in the punched data, despite its scattered origins. Some mistkes were fomnd subsequently in the punched data, but they have beer, relatively few.

CONFIDENTIAL

8 A Tr Ir f L L. 9

Ml 9 l0 0

I 1 1%L

too*

fi 1

u ti

CONFIDENTIAL
10
The precise number of reports edited does not mean much, because the number
of rounds per report varied from 2 for many acceptance tests up to about 640 for one
armor report (AD 679). Nevertheless, it may give some idea of the editing task to
state that the following approximate number of records were edited:
Approximate Number

Kind of Record
AD Reports

45

Project Reports

10

Firing Records of Substantial Experiments


With Armor

80

With Projectiles

80
1850

Shot Acceptance Tests

750

Plate Acceptance Tests

2800

Total

Classes of Data Found

The purpose of this project was to combine the data from the .nmny reports that
were edited, and to examine them for pattern and consistency. To gage this undertaking, it is helpful to know how many kinds of tests were found reported in the reference records. A risumi of the kinds of tests is given in Table I. It can be seen there
that about 50 principal models of shot were encountered, but that if one counts all the
variations of these shot, there were over three times that many kinds of shot. (The
precise nutmber of distinct shot varieties is hard to know, because some minor variations were merged in the editing process, and other variations probably were not mentioned explicitly in the firing records.) When one counts how many tests were made
with each of these shot varieties, separating the tests into groups according to armor
type (rolled or cast), plate thickness (roundec, to one of the standard values), and
ribliquity (rounded to a multiple of 5 degrees), he finds that over 1000 different kinds of
tests were reported.*
The great diversity of tests can be reduced by observing that not all varieties of
shot need to be analyzed separately. Thus, it appears reasonable to merge results
from firings of the same shot with and without windshield, since it has been observed
often that the windshield does not much affect the shot's ability for plate penetration at
a given striking velocity, Again# shot designers have decided often that there was little

~JIttac.a
we'MAg
in hie

N' vh 114o
-"all put I The ffterial wavtmd tv tAlMW I femed aetwd V"Ike r". Tim elo lit~
A kwf list t , f the weu"dWfd i' epl it ift imp*4
4 *0.
JIM&
l
'it watb a n
5 il l' r 'eas dlertled ofst'td
Malcol That w *diveJ
y4cpwpe
l. '1 ti with the ke,' l1wld I ,
ti

N
AC 1 O L L

0 I 0 0 i 1AL

I It VI a

CONFIDENTIAL

CONFIDENTIAL
11
difference between one or more closely related shot varieties, and usually there are
too few data altogether to dispute that viewpoint. Since large groups of data are to be
preferred for the present analysis, several of these closely related shots were lumped
together when they were employed ir, otherwise equivalent tests. Whenever a serious
analysis of any particular test was planned, then considerable care was exercised to
keep the lumped shots as compatible as possible. However, in some cases (notably the
HVAP, 76 mm "specials") where no close analysis seemed feasible, even dissimilar
of the data. These
shots were lumped together in order to provide a brief rofsumr
processes of lumping shots account for the fact that the number of analyses shown in
Table 1 is substantially less than the number of kinds of tests.
Broadly speaking, each of the various kinds of tests was subjected to one of three
kinds of analysis, chosen according to the quantity and quality of the data. Each of
these analyses was intended to estimate the striking velocity at which half of the rounds
could be expected to penetrate completely. Groups with enough tests (perhaps 25
rounds or more), fairly well distributed with respect to plate hardness, were treated
with a probit analysis, specially planned to account for variations with plate hardness.
Good smaller groups, especially ones involving small ranges of plate hardness, were
treated with an ordinary probit analysis. Groups with only a few tests (less than about
12), or with very irregular patterns, were treated with simple analyses of familiar
A
form, such as taking average velocities of a few partial and complete penetrations.
I.
in
Table
bhown
is
cases
varioui.
to
the
applied
of
analyses
kinds
of
the
breakdown
Details regarding method are given in the next section.
Table I also shows how many rounds of each main variety of shot were accepted
for analysis. Numerous others were not acceptz d for analysis in the end, usually because they were from shot acceptance tests where the rounds were fired deliberately at
velocities considerably above the ballistic limit.

FA MIALk

Nt

JRit

tft

CON FIDENTIAL

CONFIDENTIAL

TABLZ 1. SUMMARiY OF CASES EDITED FOR EACH SHOT

Claso

Prokctlo
Cal,
m
37

AP

57/40
S7
7S

76

90

AVC

10S
I20
ISS
6 in.
37
57
75
76
90

104

VAV

320
is
76

Model
M74
mSo
Taper(s)
M70
MUz
T43
T148
T149
M79
TIZ6E6
(M339)
TI"
M?7
T33
T33E7
T43
TS4EI
TIZ
T116
MIlt
Mk XXVII
MS1
M46
M61
T42
M41
M8
TZ5
T16
TI7
T39
T50
T%EI
T13
TSIZI
T14
TO2?
M9
T4
spe aml
TO2
T644I
umI,ns
"104

shot
Kinds

III%

2Z49
0
67
Z13
1S94
16
134
9
213
261

7
2
7
14
1
1
4
4
1
1
1
2
1
1
1
6
1
1

22
33
32
53
2
32
7
44
14
3
40
is
17
2
43
S9
1
1

70
676
9"
734
23
617
s
412
122
11
1792
IS03
72
6
1011
1240
7

I
13
2
a
3
1

1
32
16
37
a
4
35
4
9
20
44
7
I
is
16

7
310
101
614
20
73
270
1s
119
It9
117
34
AS
sitZ|
632

1
I

I
14
1
1

II
121
149

14

so

lilt

1
1
4
Is
1
1
6
A

Tht
Mo pejts"

its0

13

TOW*

is fee & "pw~

AIt

omitted

Z6
1
10
43
3Z
4
32
12
22
7

wto & Wt*4

LLI

1024
73

39z$
2761

14
5

14
3

937

4
4

57S
26
77

6
It
4

is

3
1

Z34
30
26

1s
16

1409
23S

9
I

11
0

10
7

3
I
1i

4
1

1176*
"1

11i

2
1

13
11
12
I2
1
12
IZ
14
23
22
20
12
49
2
14
a
43
10
2
15
16
14
a
a1
36
1
1
1
S
1
23
4
2
a
as
4
4
1?
4

beeo

0 " a p

w O

Z6
0
IS
40
30
z
1a
I
zz
Z?
I
31
Z4
S3
3!
2
46
i
2
40
51
IS
2
42
51
1
1
1
9

36
4
4
14
4
7
l0
4
7

11

10

I
1

I
1is4
I!3
2

*
*
is

29
Is.0

Total No.

14

Niumber of Amalyse
Probits
Smplo
wjB94H
w~o 04
A aoio

Roumd

1
1
1
2
2
Z
9
3
1

Mill
T30
144
TS)
1112,06

Number of Test.
Test
Round*
Kinds
Used

1
111

Of

I A L

ff

sIf

CONFIDENTIAL

wIf

CONFIDENTIAL

13

If

EVALUATION OF BALLISTIC LIMITS


Probit Analysis for Substantial Tests

Choice of Formula for Fitting Penetration Data


The ordinary method of evaluating a ballistic limit is to fire enough rounds to get

I
y

certain number of partial and complete penetrations within some predetermined velocity
interval, i.nd then to average the velocities of those rounds. This method depends on
the notion that there is a fairly definite velocity below which the shot will not penetrate
the plate completely, and above which it will. However, it is well known that there is
really a zone of velocities in which mixed results can be expected. To put it differently,
the probability of getting a complete penetration is indeed a function of the striking
velocity, but it may be quite different from a step function jumping from 0 to 1.
The great majority of test situations have been the subject of so little firing that
the probability of success (i. e., complete penetration) has hardly begun to be measured
at any particular level of the strikiag velocity. However, among the test situations
where many rounds have bten fired, a common result is that the probability of success,
as a function of velocity, rises slowly at first, then rapidly, and then slowly approaches
unity, thus approximating a cuimulative normal distribution. The slope of the central
portion of the curve vari,s among test situations, and for some cases where shattering
of the shot intervenes, the curve may drop again to some low value before it rises
finally to unity. This matter of "shatter gaps" is one which will be treated separately
later (and then only briefly), since the subject is difficult enough without it. For the
present, attention will be restricted only to cases where the probability curve follows a
cumulative normal distribution fairly well.
To rationalire this shape for the probability curve, one may assume that whene ,r
a shut is fired there is some critical velocity above which the shot %il1 penetrate cun-1pletely, below which it will not, and that these critical velocities are normally distributed for repeated rounds under nominally identical firing conditions. This distribt,,n
of critical velocities may be attributed to minor, unseen variftions in the physical
makeup of the shot and the portion of plate whis-b it strikes. Then it V., . 0 are the
me-in and standard dtviation of these crit,-,d velocities, the probability ot suc.ss with
A
shut at v4locity v il:

V-

Z/" wk.,

where, t

-}!
(v - V1

"h-,% i'. tht cknmultve nornal distribution. The quantity 7 is also the ve:lo. ity .t %hit h
h., ot the rounds would achieve complete penetrition. and is thus eus,-ntAIlly iist a r, fn.m-ent of the trdintry concept of the ballistit limit

I
I

CO"NFIDENTIAL

CONFIDENTIAL
14
It has often been observed, too, that the ballistic limit varies with the hardness
of the plate, and the NPL formula quoted earlier gives an empirical law for that variation. However, that law was derived for small shot and thin plate, and it can hardly be
extrapolated to all cases of interest here since the denominators of two of its terms
(65-& and Bo-B) can vanish in the p ,esent range of interest. Nevertheless, in order to
get an idea of the influence of plate hardness on the ballistic limit, that limit was plotted
as a function of hardness, asing the NPL formula and several combinations of the other
parameters. Most of the curves can hardly be distinguished from parabolas operning
downward, This suggests that a parabolic relation between V and the hardness h is a
reasonable form to use in striving for an empirical correlation of the armor penetration
data. This type of relation was tested further by comparing it with actual penetration
data, and it appeared to serve ;,s well as 4ny other acceptably simple relation. Therefore, it is here assumed further that:
+ bh'2

V= k ,h

where k, a, andb are constant.

Maximum Likelihood Solution for the Ballistic Limit


Suppose now that a group of tests have been performed at velocities v, with
, n), and that each has been rated as yieldi (i = 1, 2, ...
ing a complete or partial penetration. It is required to estimate the parameters k, a,
b, and a which best account for this set of results. The method which will be applied is
a form of probit analysis. * First observe that if our assumptions regarding critical
velocities and ballistic limits are valid, then the probability of success with the ith
round is Pi = p(ti), where t i = (vi-k-ah -bhi 2 )/o. The probability of achieving all the
observed results is:
associated plate hardnesses h

Pln = p
i=l

q,

'-iq,

where )i = I for a success and 0 for a failure.


the logarithm of P, that is
L

1>
t

= l-p,

The likelihood function L is dtined as

In pi + (1-51) In q1 j

the plan for finding k. a. b, itnd ' is to chv.ose them s04 ,it to i.--.iet
1. fr,,im attnong
,ill the values it can have when the velocities v,I hardnesses h,, .tnd rating number%
-ire the ones observed in the actual experiments.

U. 0 1 ric), 1). 1., PnL~i AnA I -i8

I?tw MqW"T
AOwIS

Pis

flc, tt-TWA hk

IA~il41

i t".

espet ia I I

*I~e~i
-j. nq r~*

i' riAtilkI
Rrk-fiti LAS.'40 It, T I
%MA
kAkl~

^A ? It i

LI

igAL

Ii

Vet

%7 A.

~4 *

in.. INI ',+vr *40-,

1 ^ L

I'!

af
ftj

in 4,4,r~
L.

. 1114

(A&44*v

i,

it* jiv.,

16.,

fi,

t#

Ia

CONFIDENTIAL
4p

CONFIDENTIAL
15

a*) is wanted which makes

The solution (k*, a*, b,

,)L

AL

7k

oL

& 7b

-a =

However, since these equations involve several transcendental lunctions of the unknowns, it is not possible to solve them directly. Therefore, it is expedient to begin
with an estimated solution (kl, al, bl, al) and to seek corrections for it. Expanding
the derivatives of L as power series around the estimated solution, there follows:
6L=6L

+ (k-kl)t&2-lN

+ (a-a

) t "2a

+ (b-b I ) t,

(1-

I)(162LN

~\

and similar equations for the other derivatives. Ifthe estimated solution is not too far
from the required solution, then the series truncated at this length are valid there, so
that:
(k* -k 1)k-

+ (a*a,)!(7-a)l

+ b

'bl\k~bl

+ka)
( 'T)kJ

+\k~/
(b*-bl)

or

)l

a 1b\

and there are three other similar equations. These equations are linear in the corrections k*-kl, a*-al, b -bl, and a' -a. Solving for these, and adding them to the first
estimates yields improved estimates of the desired solution. The correction process
can be repeated until the corrections become negligible, and the last estimates are then
the desired solution (k*, a*, b*,

*).

In order to find the corrections k *-kl, and so forth, one needs the second derivatives of L. To get these, note first that if 61 and :-2 are any two of the parameters k, a,
b, or a, then:

L.Pi
.2

i
P(1 P1 )

Pi + '1(I-2p 1)
Px oPi
'
" Piz"
(1-pi)Z

A convenient notation is:


Zi

(v-

met/Z,

k-

ahi - bh1 )/a

Then,
Pi

)tj

Atl

tl

and
d1L

"i

-'4,t

A,'P ,h

CN

A t t 9 L L 9

_'tj

I tt

0O0

# A L

+ ', f -,p,|)

TP IAL

1 00 6t T I

1U t

CONFIDENTIAL

aK

CONFIDENTIAL
161
Now let

u.i=i

Nofe

+
w

Z 1/p 1 for a succ essl, an


and
Zi/qi for a failure J

= u i ti - u i .

Then, recalling that 6i = I for success, 0 for failure, it follows:

i= itl

6Ati

21-

-,

ae 2

Thus the second derivatives of L with respect to the various parameters are:

Lkk

1jwi

Lka= a-21wihi,
Lkb = a

wi h
2

Lka =

(wi t,

Laa = a"

ui)'

w i hi

Lab

a-?-Iwi h t ,i

Lbb

a_

-a"4
'w

hi

Lba = a"

(w i ti

Loa = C

(W t

u) h
Zu

t.).
1I

Letting A = k *-ki, 6a a*al, 6b b *-bl, and ba


must be solved for the corrections is:

(Lkk)l 5k + (Lka)lIa
6)
(1-ka) Ixr#t +(Laa)
I1

(Lkb)I

+ (Lkb), 5b + (Lk) 1 6o
(Lb)xb I
a + (Lab)i

+ (aO)l
(Li 1 I5.j -

+ (Lab)1 5a + (Lbb)I ,b+ (Lb.)1 5a


k

(La -

A t

5 + (Lba) I 6b + (Lao), 6c
LI
lM I M 0 ItI !A IL

It I ILi

a - l, the system of equitiuns lhat


-R

ut)I

-- (Iju

u1I hi

- 0u 1

h"),

I E N T"l0A# L
T VNF I
I

Iu
I

CONFIDENTIAL
17
Solving these equations by hand calculations would be possible but laborious, so a
program was prepared to do the work with an IBM Type 650 Magnetic Drum Data ProcThe plan of the program required first that an initial estimate should
essing Machine.
Then, as the card for each round was fed to the
be made for each of the parameters.
that round made to each of the sums apcontribution
the
compute
would
it
calculator,
pearing in the coefficients of the system of equations. * The only difficulty in this process was in finding the quantities, Zi/pi. For these, a table was stored in the machine
which enabled the machine to find Zi/pi correctly to one whole number and five decimal
The whole operation proved to be efficient since the calculations performed
places.
for each card took about as long as the minimum time to feed a card to the machine.
After the entire group of cards had been used, then an end program proceeded to solve
for the corrections 5k, etc.
In practice, solving for the parameters k, a, b, a proved to be a delicate art.
Unless the first estimated solution was quite good, the iteration procedure might never
converge. Therefore, all the penetration tests analyzed by this technique were first
graphed, using an autorratic plotter, and initial estimates of the parameters were made
from the graphs. Even with this good beginning, some cases failed to converge, presumably because either the normai distribution of critical velocities or the parabolic
variation with hardness was not a good assumption. For nonconvergent cases, the only
recourse was to divide the data into smaller groups, chosen as logically as possible,
and to analyze the subgroups separately. Another rare cause for failure appeared when
a parabola could be fitted to the data so as to separate entirely the complete and partial
penetrations.
This, of course, makes a indeterminate; so for these cases a parabola
was fitted using the fairly arbitrary value of 10 ft/sec for a.
One further refinement that was added stems from the fact that the par-Ameters k
For the sake of clarity of meaning in the parameters.
and a are relatively meaningless.
it is more desirable to express V ab:
V

+b(h-h

o)

This is the same as the former expression provided:

a2

-a

2b

4b

ti the
Since the values of k, a, b. and " which maximize L correspond by these relatumi
values of vo, ho, b, and 3 whzch maximize L, the latter values are 0Asi1V tuand trwyi thqformer. The reason for using the parameters k, a, b, : was merely that the v. Kdird
simpler expressions for the second derivatives of L However, when the pir.,ny-tr re
inaire expressed as v o . ho, b, and ', they become respectivoly the mnxlmumf.ir jtn,.
,ish.s:gballistic limit, the optimumfor prsoxmumr hardnres., a measuare nf srnaitj 1'.
in placr hardness, tnd the itandard aeviatior of critio Al vel(K Itier

,rw t ail *,La -W

4r
i1
,I

tf"'

"

-"

1k Vt

'41,#*U-

*Vq*I,*Jaw
4h
tf

A111

ka

V#'

tI~
Ik twi

i 1atith. *A1211wu,

0' LO
c.

ti i

s~it *iJ'

f~4 1

*t Ifti e"wI

,~ii # *"n

.~

"?e" fein Weft

*if.
rwr.oe -4
I'rIi, *i, -

M1wt

q Wt j#

~l

ft WrO9W Alt - 1* .

CONFIDENTIAL
titOt

FlDET

Iftst
IA

$q

L9 It

4 1i

l~

1$4!

1
CONFIDENTIAL
18
U

Precision of the Ballistic Limit


One of the great difficulties besetting the evaluation of ballistic limits has been
uncertainty regarding the dependability of the limits as they have been computed in the
past. Therefore, when probabilistic evaluations of the limits are being made, as they
are here, some estimate of their precision is highly desirable. To get this estimate of
precision, we can use a proposition from mathematical statistics, which states that the
variances and covariances of the maximum likelihood estimators for the parameters of
a population are the elements of the inverse to the Hessian matrix:

I1-E(

1,

II

whtre E(
) denotes an expected value, and ( i and Gj represent any two of the parameters being estimated. Restrictions on this proposition are that the number of tests n
is sufficiently large, and that to a good approximation*

1E
3L

L\11\
'2

Regarding sample size, Golub and Grubbs gave an illustration with n = 5, so the present
aim to keep n more than or almost Z5 is probably fairly conservative. The a':proxirnate
equality between expected values was deemed plausible by Golub and Grubbs, and seems
to be so here too.
Regardless of which set of parameters is used, the equation (a) for no1,I

is still

valid. In particular, we now regard it z.s applying to the case where 61 and e7 are any
of the parameters vo, ho, b, or a. Since only expected values of the second derivatives
are required now, and since E(6i)

pip it follows from equation (a) that


J90)Piqi
i~ 1

) &1

6v o

ti

The required first derivatives are:


t

Zb(hi-ho)

2t i

-(hi-ho)

1ti

tj

Is

pp

2110

-ib

Using these expressions for the derivatives, and the notation:

Piq,

Tus sta-e.ent of this ptfulption pg.als that 4 Colub md Gubb-, OF. "'It., p. 40 Fut a Mm .wsp,.td,. boot
%tatnunt,#ee Kendall, &I.G., lhe Ad.anced Ihuotiy 4e t
golunn i1, iecti.om 17* 46 md 17.26, Garffm
Vhti,
Ltd,, Londtqi (1 4ti).

8 A T1

LL L

Mi

0@A
! L

#NhOI

t It

CONFIDENTIALI

de t

and I

mim

CONFIDENTIAL
19
it follows that the required Hessian matrix is:

-u_

2
_ Z Ui(hi-h o )

Z Ui(hi-ho

~Ui(hi-ho)

ZUiti

)-

i(hi- h 0 )~4

j-zUi(hi-ho)-ti

ihihot

ZbV
Ui(hih

h o ) ti

,-U~ih)t

U-"

-LT7

i (

-U

>U(hi-h.)'
2b

-U iti

--

jh-.)

Zb
-

4bZV"

OZbjih~o
-

1-zUi(hi-h

Ui(hi-ho)

Uiti?-

4,

The elements of the inverse to this matrix are the variances and covariances of the
maximum likelihood estimators of the parameters vo, h., b, and a in that order.
The estimate of the ballistic limit V is,

of course,

V = v o + b(h-ho)

Since V -s a function of parameters that have been estimated to within known variances
and covariances, it follows* that the variance of thi estimator for V can be found from:

Vo

CY

\io/i~

+(
v0 ,o
hho

ho

b 3 Vo abv

ho,b

=Zo+ 4b 2 (h-ho)Z O~ 4"(-o)

4b(h-ho)

"4b(hho)3

h2

h,

This expression makes it possible to state a variance for the estimate of the ballistzc
over the range of plate hardnesses that appear
limit V as a function of hardness, at last
analyzed.
in the case being

As a conclusion to the computing for tests havri variable plate hardness, a suipplementary program was prepared to find the parameters v , ho, b, 0 and ther variances and covariances.
The results for all the cases analyzed in this way are shown in
Table Z. Values of OV for three levels of hardness covering the range of th# test data
If values %f -O ore desired t other hardness levels,
are incorporated later in Tohle 3

(Jf.,

Hald, A., SW.IM, Al 1he,.

Witt,

L.nmeini

L LL

041

Jki

onh, p, 1ii,

ET A

johm Wiley

1" A i

%of*, t4ew yiwfk (I

CONFIDENTIAL

I a

"

CON FIDENTIAL

93

'1-'

11111

all

a' NN

gs

1. fi
*~

I
0

A
CONFIENTIA

1jj

86

1 N2M 0

51

goE

nos

No

is~*t

at 1B 1

kI
1AL

IN5TfI

il

26111111

CONFIDENTIAL
zl

11J1

'

qm~ilreu,
1
IfsI

II

r~

SOII

~E~U!

a A

II

Io

inl nn.
It-

!II
I

m3pm
muul*
$mta
itH
INltll!!

tii!

00

, ..

I ~Ih*!I.uu
l
IM I 0 RII

P11111 I

CON FIDENTIAL

0m5.
I ag!
I

I i

!I 3 3U

CONFIDENTIAL
i

22

, a, i!" of 9IN
Vai sma-ll asi of

="l

Ill

X1

-A

MI

Ia

IlN af t

a
getl II II

as* ons*"I "Iam a iI an


" ata
-o

Iu

lite

I i A

Il

8 9 a age 4

L 1

4?0 t

10I
A

*ou

S'
W *0

# It I I

CONFIDENTIAL

I t

CONFIDENTIAL
?_3
and covariances shown in Table 2. Three other
the y can be derived from the variances
a,) were computed, of course, but did not seem necessary
co'~variances ('
for- inclusion in the table, since they would be used only for predicting standard errors
cent probability.
of ballistic limits at levels other than 50 per
In most cases, the values of ho shown. in Table 2 confirm the idea that the optimum
hardness of armor plate is in the vicinity of BHN 300, but there are many exceptions. A
common reason for these exceptions is that as the plate hardness increases the projectile tends to shatter more readily, and this raises the ballistic limit. This increase uf
data is not simply statistical chance, for
ballistic limits at the high hardness end of the
the values of ab show that b is generally significantly different from zero, whether b is
negative or positive.
The values a show that the scatterirng of critical velocities in most of the test
of them were above 60 ft/sec. Thus the
situations was substantial. Better than half
chances for a seemingly erratic result with any small number of tests is reasonably
gr eat.
five instance3 in which the parabola fitted
It may be observed too that there were
analysis did not lead to an evaluation of
probit
the
the data with so little overlapping that
a cr of the variances of the estimates of the parameters. One of the cases (with the
IIVAP 90 mm T44) was somewhat degenerate in that the data had only three hardnesses,
but the other cases involved f..ur to six hardnesses. For the latter cases, the parabola
offered a surprisingly convenient fit.

Probit Analysis Without Variation of Hardness


There are many penetration tests that have been performed having so little
variation of plate hardness that the preceding analysis is not justified. In this case, a
probit analysis without variation of the hardness may still be applied, provided a reasonresults include overlapping velocities of
able number of tests weie performed and the
pa rtial and complete penetrations.
An analysis for this case can be had from the broader analysis described earlier,
provided it is assumed that a = b ; 0, so that V = k. This meanp that the maximum
likelihood solution now requires finding only two parameters, Vand a. If initial estimates ;I and o are available, then the equations for the corrections 'V and 5,, are:

where

0 A

T1LL

w0#4 A

t#t,

UT t

CONFIDENTIAL

I
CONFIDENTIAL
24

Lv V

= a

Laa

a:

wt

(witi-

IVI

ui )it

,t

while w i and u i are the same funtions of t i as before, but with t i = (v i - v--)/0. The
variances and covariances in the solutions for V and a3after the iteration process is
completed are the elements of the inverse of the matrix

~I

where U i is as before except for the change in t i . All these results are equivalent to
formulas given by Golub and Grubbs, except for the small differences already noted in
Leo and Lj.
Very little additional programming was needed to adiipt these formulas to machine
calculation, because the only requirement was to drop some of the terms that appe~ired
formerly.
This method of analysis was applied to many cases where the accumulated data
included as many as 12 rounds, but with little varia4.on in the Brinell hardness number,
that is, less than about 50. However, in order to avoid getting unreasonable variances,
it was necescary to exclude several cases that were complicated by shatter gaps, or
systematic discrepancies from other sources, some unknown.
Results of the calculations ignoring variation of hardnesm are not tabulated here .t
a group, because the most significant results, V and c-v, are included later in the comprehensive Table 3.

Brief Analysis Used for Cases With Few Data

t he
-..thods of probit analvsis require a sample reasonably good in both sir.e and
itn 'i urdrr to produce useful results. If the sample is too small the logic bei .. u ne probt inalysis is weak.
If there to no overlApping nf partial and complete pen-tr.tiuns, #ihefnthe soluition beconmes indeterminate. There were many cases of these
... th4e dtt .ssvn
,d f,
th;s pr~~'ect, plus oth.r cases having very Irregular
,
data i),ttt rrs. For thtr . viv,
otme an4lvsis, or at least some method of descrptim.
edv.
In .)rder not 1,- ovt-riix the dat,, simple instead of iphialtited
i4n-vas
ill ,. du 4ir tible.
N,

-l

I
S

I
I
3
3

1-

0 A Tt

t 00l,.aqlllt
0 4 A

gLP0

t 0 t 'U t a

CONFIDENTIAL

CONFIDENTIAL
25
For cases where there were less than about 12 rounds, or where there was no
limit".
overlap, it was decided to use the old averaging technique to define a "ballistic
highest
three
the
and
CP's
lowest
three
This was based, preferably, on six rounds: the
in
velocities
test
only
PP's. If these lay within an interval of 150 ft/sec, and were the
by the
that interval, then their average velocity constituted a good ballistic limit
atudy.
present
the
for
limit
ordinary standards, and it was accepted as the ballistic
an
within
If this six-round average could not be had, then an average of four rounds
int,-rval of 100 ft/sec was accepted, or even two rounds within an interval of 50 ft/sec.
Accepting these as ballistic limits in the present study does not mean that they have any
good or even known degree of accuracy. It means only that these were about as good
limits as could be obtained, considering the state of the data. So long as the origin of
Of
each limit is stated, the varying definitions of the limits should not cause trouble,
course, an extra bit of useful information about these simple ballistic limits is the
spread of test velocities included, so it was decided that that too should be mentioned
for each case.
A moderate number of test conditions have yielded test results that do not follow
the ideal scattering of partial and complete penetrations. Sometimes there is an
identifiable cause for this, such as shattering. In such a case, a split analysis is
sometimes useful, say one for intact or broken shot, and another for shattered shot.
It was decided to use a split analysis if that enabled one to give an instructive description of the test results.
In some cases, the partial and complete penetrations were mixed over a wide
range of velocities, sometimes over practically the whole of a wide range that was
tested. In order to describe this situation, it was decided that a simple statement
would be made describing the mixed zone. Such a statement does not pretend to define a ballistic limit, but it does picture the state of the test results.
For cases where the highest PP was considerably below the lowest CP, It wAs
decided merely to state those two velocities. Finally, if there were no CP's, then a
highest partial penetration (HPP) could be listed; or if there were no PP's, then the
lowest complete penetration (LCP) could be used.
It may be repeated that simple analyses of these sorts do not always give good
evaluations of ballistic limits. Oftentimes they may be poor indeed. Good examples of
this can be found in long sequences. of PP's at successively higher velocities, followed
at last by a lone CP. Such a sequence of tests may allow the computation of a simple
two-round ballistic linat, but it has littLe meaning beyond that of identifying A velocity
which lies in or near the mixed zone,

Compildtion of Ballistic Limits

A comprehensive collection of the results of the analyses performed for the presincie the table preent project is Itive1 in Table 1, at the end of the text of this report.
sents a wide variety ot results, the user should try to keep in mind what toe foiAndati.nl
of the table are.
8A I tE9LkL

I It# 0

06f?ittI
*

CONFIDENTIAL

CONFIDENTIAL
Z6
First it may be said that the objective that was followed in preparing this table
was to describe the results of the penetration tests as instructively and concisely as
The results were to be in the form of ballistic limits, but even the definition
possible.
of ballistic limit needed to vary from case to case, depending on the quantity and quality
of the data, Methods were rLeeded both for selecting the type of limit and for describing
the results of the analysis.
The approximate order of preference for the methods of finding the ballistic
limit was:
(1)

Probit analysis with variation of hardness

(2)

Probit analysis without variation of hardness

(3)

Averaging if 6, 4, or 2 rounds, half being CP and half PP within a


suitably small interval

(4)

Locating the zone of mixed CP's and PP's in the data

(5)

Locating the highest partial penetration (HPP) or lowest complete


penetration (LCP) if the results were all of one kind.

For each test condition, the analysis was begun by trying the highest seemingly feasible
method on this list. If that method failed (say, by lack of convergence of the probit calculation, or by unacceptable scattering of PP's and CP's for simple averaging), then a
lower method was applied. At times it was expedient to split a large group of data into
smaller subgroups in order to make them manageable.
The splitting was done as
plausibly as possible, on the basis of shattering, plate hardness, shot hardness, or even
distinction between references.
Separation according to references (applied twice, both
times to data for the APC .37 mm M51) implies that there were real, but unidentified
systematic differences between separate series of tests.
Except when parabolas were fitted with negligible overlapping of partial and complete penetrations, all the probit -Analyses were completed to the point of finding the
standard error in the estimate of the ballistic limit. The probit analyses may be
identified in Table 4 by the fact that the standard error (SE) is stated. If three ballistic
limits, together with their standard errors, are shown by braces as havng been derived from the same set of rounds (that is, only one entry for the "number of rounds"),
then the analysis was one allowing for parabolic variation of hardness. The levels %,.,
plate hardness cited for these parabolic cases are roughly the extreme and middle hardnesses appearing in the reference data.
Limits obteined by averaging are identified by statements such as 6R(1Z4). which
Would imply that the average was based on six rounds spread over a velcily interval
of 124 ftisec. The remaining types of analyses are identified by self -explanatory
statements such as "Mix 1041 to 3465", or "Z669 HPP".
The probit analyses wre perfc ned by the electronic calculator which, amnonog
other thirgs, adpisted the velut ity of ,-ach round to an equivalent velocity against a plate
of staivdard or "nominal" thikness at ordinj to the formsnijl
* A t I &L L 9

M 9

0 4 1 A L

# of 6

tI T V Ita

CONFIDENTIAL
.

"

*a-

CONFIDENTIAL
27

(nominal thickness1/

adjusted velocity= observed velocity kactual thickness )

This formula is probably fairly accurate for the small corrections that were needed.
Since these corrections were performed round by round, no further corrections for
irregularities of plate thickness are needed in the results of the probit analyses. However, the simpler limits obtained by averaging or location of extreme velocities were
evaluated by hand, and the round-by-round thickness correction for them became too
burdensome.
Therefore, for these simpler analyses, a note is added showing the
average plate thickness of the rounds involved in the calculation, if that avcrage was
known to differ from the nominal plate thickness. These notes are written briefly in the
form like "t 3. 03", which would ir ply that the average plate thickness was 3. 03 in.
Since projectile breakup often influencc ; the probability of complete penetration,
an effort was made to dvscribe briefly the typical breakup in each case. This is contained in the notes by statements such as:
SI

shot generally intact,

SB ~ shot
SS

rally broken but not shattered,

shot generally bhattvred,

or by combinations such as:


SI&B

some shot intact, somet broken

SI&S

breakage ranging from intact to shattered.

Many of the rounds edited and keypunched were omitted finally from the analysis,
almost always because they were fired in acceptance tests at velocities intended to be
significantly above the ballist%c limit. However, the existence of the rounds is noted in
Table 3 by entries under the cases where they would have appeared had they been used
Thus, for the AP 37mm M74 against rolled armor, there appear two entries, "S4ARO'

and "84ZARO", which mean 84 ind 842 iccept.Ance rounds omitted.

Such entries are

sprinkled throughout the table.


When special modifications or lots of shot were usd, notes Are Idded in T.1b.e i
identifying these shots, provided such identification was thouagh to be useful.
A sprci.d
abbreviation was used when the shot hardness was variable. An example of this is
"R16". which me.,ns thAt the hardness ,.as 61 on the Rd-kkell C sctlc
Ihe hardneis

listed is that of the nose, or of the bourrelet if the shot *ere truncAtted
Referentv-s to the firing records in each test situation were not mt luta4d in
r able 1, %ince uth an inclusion threatched to obscure the results
roarever. .4 list ,4
referent r retx-rtP used for this cellec ion of penetration tests ist inluded i the Appendix

CONFIDENTIAL

CFDAtf

l0

0 0A 6

ITItAt

CONFIDENTIAL
28
OBSERVATIONS CONCERNING THIS STUDY

Success of the Methods of Analysis

When this pr( ct was begun, it could not be anticipated what all might result from
putting data from so many sources into one great lump. In order to forestall trouble,
as much information as possible was carried regarding each individual test. In consequence, the results are at least moderately consistent and precise. Because these results come from so many test programs, they are also as unbiased as seems possible
at pre sent.
In the process of performing the analyses, very many sets of data were studied by
persons who wanted to understand those data as much as possible. When the scattering
of the data was bad, reasons for this were sought. Only rarely were there cases where
different firing records gave noticeably different results without a plausible reason. In
only two cases was itthought profitable to split the data according only to the reference,
both times inv.)ving AD 1084, which seemed to refer to either some very good plate or
sonie very poor shot. The fact that separation accoiding to reference was invoked so
seldom seems to imply, on the whole, that combining raw data from many sources was
a reasonably successful venture.
Use of the parabolic form for the influence of plate hardness was also reasonably
successful. There were relatively few cases where the assumption of this form seemed
to deny convergence to the process of probit analysis. When the parabolic form proved
too inflexible, it was most often because a sharp increase in projectile shattering occurred as the plate hardness increased. Another occasional difficulty arose when the
plate hardnesses went above BHN 350, for then the ballistic limit curve sometimes
seemed to descend more rapidly than the parabola.
The assumption of the cumxulative normal distribution for probability of complete
penetration as a function of striking velocity was nioderatoly successful, but it wa3 often
troubled by the occurrence of projectile shattering. A bacA example occurred in the d.ata
for the APO 57 mm M8b fired against 3-inch R H. plate at obliquity of 35 degrees. In
this case 105 rounds were recorded for Brinell hardnesses 295 to 335, having veluciti t
between Z450 and 2800 ft/sec. At no velocity did the probability of a CP seein to rise:
above 40 per cent, and it seemed to fall to zuo at both ends. The shot breikAgVe in this
case ranged from intact to shattered, but no record of the brea'-age was available for
many of the rounds that were fired.
As a foundation for statistical analysis of the penetrat:on tests, the assumption of
the t wnulative normal distribution seems as good as any reasonably simple assumption
that ;an be made, but a highly detailed analysis of multitudinous data (more than ex,st
now) would probably use a more flexible form for this probability function
The standard t, rror%in the *allisti, lnits vary widely, but are typically from 15
to 50 ft/sec. A few are rtiiih larger, mi4 yet more would have been larger if the Analy.
set had not been split in the vario-a wt - , tiready mentioned. Ths precision may prov.
di sappumnting to some people, blit it appelre tv h.- what present data imply. Thr
8 A I 1L L K

M 9 M 0 IM A

IIf

t It ?

CONFIDENTIAL

CONFIDENTIAL
29 and 30
implications regarding the ballistic limits obtained by averaging are that those limits toc
are subject to substantial errors. This is particularly true for the many limits that
were based on only a few rounds, and for those not selected by an over-under method.

Application of the Results


The early hope of the present project was that it would reach a simple, comprehensive correlation of the great mass of penetration data. This hope has not yet been
realized because the great weight of the data has made progress slow. However, the
objective is still there, and substantial progress seems to have been made toward
achieving it. It is now possible to scan in 32 pages (Table 3), information which was
contained at the start in about 15,000 pages of firing records, and the information has
been processed according to fairly reasonable statistical procedures.
If and when further work of correlation is done, it should now be able to rest on
firmer foundations, in that now a large collection of ballistic limits is available having
associated measures of precision. Moreover, since the influence of projectile shattering seems so important, it should help to have at least the sketchy survey of breakage
that is inclided in the table.
In view of many of the entries that can be found in Table 3, it is surprising that
the earlier correlation work of Kilian, based on much the same data, was as successful
as it was. Attention is called particularly to the many models of experimental shot that
were used. Here many instances can be found where specii.l variations of nose design
or metallurgical design seemed to exert real influence on the ballistic limit, even though
these results are rarely precise in the statistical sense. It should be noticed also that
the last-resort analysis of citing simply the mixed &one was used most often for tests
of experimental shot designs. Any over-all correlation of penetration data covering
many designs of shot probably needs to be quite perceptive in its recognition of which
kinds of shot may properly be grouped togother.
In their present form, the results of the present survey of armor penetration data
should be useful in that they provide a ready reference to a large body of information.
Nevertheless, it is to be hoped that the further work of correlation and condensation will
be continued.

N&L1 tNT
IA

AL

I"It

It

CON FIDENTIAL

CONFIDENTIAL
31
TADBI

3.

SUMMARY OF DALLISTIC LIMITS

Part 1. Armor Piercing Projectiles Versus Rolled Homogeneous Armor


Sbot
Cal,
mm

Model
No.

37

M74

-alistic
Plte
BHN.
Nom. t,
in.
kg/mm 2

Nom.
Obl..
deg
0

324-385
265-363
269-337
r230
f300

1.00
1.12
1.25
1.50

L370
235-286
235-293
255

2.00
2.25
2.36

57/40
Tapered Bore
(Cf. 0P5829/1)

1314

SE W9

1685
1832
1880
2060
2091

94
42
3

S1, 84ARO
SI
60-mm armor, t 2.31

27

S&S

5
14
7
5
7
6
6

SB&S, t 2.51
SI&S, t 3.03
SS, 842ARO
SI
SI&S

4
2
3
40
25
2

t 2.51
t 2.51
t 2.51
SB&S
SS
t 2.51

73ARO

3.00
1.50
2.00
1.50
1.50
1.50
2.50
2.50
2.50
1.00
1.25
2.50

251
269
251
341-375
302
269

2346
2423
2620
2430
2720
2662

2R(51)
2R(44)
2R (48)
SE . 7
SE x 6
HPP

20

1.00

--

3.00
4.00
10.00
6.00
3.00
4.00
6.00
3.00
4.00
6.00
3.00

50
55
60

M80

SE w 22
SE v 13
SE z 7
SE a141
SE6

2346
2591
2815
1789
1842
1617
1808
2212

20
25
30
40
45

37

899
1087
1209
1256
1381

f310
347
255-347
266
258
255
255
255

r230
270

10
30

45

55

*A

T?

I L L It

66
39
60

SI
SI
S1

545

$1

--

241
22'3
205
224
241
229
224
241
22)
224
241

18"#
2308
4047
2663
2332
2738
Mix 3041
31A4
3313
Mix 2920
4049

0 N I A L

2R(80)
2R(21)
HPP
HPP
2R(18)
6R(94)
to 3463
2R(36)
6R(88)
to 4022
HF?

4
C
2
I
5
8
12
5
11
10
3

II

CONFIDENTIAL
a1.

-W

mo-

Notes
(AbbreviaUons
explained on p 27)

Limit
Sigmfi.
cance

SE a 13
SE a 53
2R(31)
SE a 165
SE 81
SE w 102J
HPP
6R(179)
LCP
2R(29)
6R(179)
2R(28)
2R(36)

2.50

Number
of
Rounds

BL(A).
ft/sec

*W

0-

#AI?

SI&S

SS
SB&S
SS
SB&S
SB&S
SS
SB&S
Sl&S, w/o PP at 3528
SS
SS

CONFIDENTIAL
32
SUMMARY OF BALLISTIC LIMITS

TALE 3.

Part 1.
Bet
Cal,
57

Model

Nom.
Ob.,

NO.

deg

M70

Armor Iercing Projectiles Versus Rolled Homogeneous Armor


Plate
Nor.

t,

in.

Iallistic Limit
IL(A).
Sigimfi-

BHN.

kg/mm

Number
of

ft/sec

canoe

Rounds

Note

(Abbviaions
explained on p 27)

1.50

271-330

999

SE u 14

46

SI

2.00
S. 25

235-286
235-293
220
280

1248
1403
1454
1549

SE a 9
SE a 11
SE 36
SE a 12

35
37

SI
St

68

SI. 1ARO

340

1470

SE s30

20
310
400
00
280

1664
1817
1524
1964
2384

SE 27
SE a 12
SE a 29
SE w48
SE a 18

350

2154

SE , 27

20

5.00
3.00
4.00

388
217-259
226
262

2860
2867
2204
2992

25

4.00

357-388
388

30

3.00

2.50

3.00

4.00

4. 00
35

3.00

40

2.00
2.50

SI

6R(124)
6R(113)
2R(52)
HPP

11
10
2
6

SS. t 4.02
SI, t 5.09, PP at 2418
SB. 3921ARO
SS, t 4.02

2911
2911

2R(24)
HPP

15
6

SS, t4.01, lone PP


SS.t 4.02

r280
340

2813
2580

400
388
2"0
(340
400
291-300
220
230

2429
2325
3015
2690
2514
1,67
2116
2624

SE a 24
SE.w 171
SE - 23
WP
SE 71
SEa 51}
SE a 35
SE a41
SE a 177
SE a 128

32
3.0
3.W
1.50

160

SS

3S
l

93

SS

21

SO1

i&S

is
5
13

SS
t 3. 12
SI&S

I L L a

400
262
330

2387
267)
2618
1734
2094

360

2303

M IM

I
I

SE()3
a 10 6
SEa41
SE
3.
SE,
Zt(31)
SEa17

10.1

CONFIDENTIAL

a A 1

SI

41

30
45
50

210

0 2I1 A L

11 6 1 1T V

CONFIDENTIAL
33
SUMMARY OF DALJSTIC UMITS

3.

TADBL

Part 1. Armor Piercing Projectiles Versus Rolled Homogeneous Armor

Cal.
mm

Model
NO.

Nom.
Obl.
deg

57

M0

55

Shot

Plate
Nom. t.
in.

60

70
75

M72

I4N.
kg/man

290-291
327-353
273-352
331
289-302
327
273-341

2494
2941
2375
2412
2'785
2939
2907

SE a 38
SE = 60
SE a 67
6R(92)
SE a 17
HPP
SE a 41

14
20
35
10
21
6
35

SB

1.50
2.00
2.25

255
23 -286
223-293

737
924
1048
1120
1149

7
54
37

SI
Sl
Sl

55

SI

1241
1311
1367
1263
1536

2R(28)
SE a 13
SE w7
SE 26
SEu 14
SE - 3s8
2R(24)
SE -1l
SE 14
SE 30
4R(84)

r200

1570

SE 33j

4.00

1260
20

1711
1655

SE
SE

5.00
3.00
4.00
2.00
1.50

217-217
246-262
300
255-258
255

1995
1634
2582
972
790

3.00

298

4.00

--

1.50
1.50
1.50
2.25
2.50
3.00
2.00
2.00
1.50
2.00

2.50

290

2.75

S962
252

r220
3.00
3.50

25
30

35
40
46

50
IS
40

41

Notes
(Abbreviations
explained on p 27 )

Number
of
Rounds

2.00
2.50
1.25
1.50
2.00
2.50
1.25

r230

2u

Ballistic Limit
SisgufiIL(A).
cance
ft/sec

T K L L I

4310
L400
262-289

SS
SI&B
t 1.49. SB&S
SB
SS
Si&B

t 2.73. 6ARO

3
79

S1

t 3.49

41

SI&S. 11ARO

6R(147)
LCP
2R(43)
2t(35)
2t(52)

7
11
3
10
6

S1
SI&B. 2744ARO
S. t 3.96
S1
SINS

20

2R(41)

2595

4R(42)

255
255
255

10s
1013
1210

Lcp
2R(36)
2R(36)

6
7
3
2
4

25
269
300
162
162
M65-266
826-300

092
2116
S35
2001
1916
1847

LOP
H
IP
2R(52)
21t(2)
H"
44(12)
S9-44

N 9 NtA

095

510
25

f t IL

SONS
ss

st
SI

3
5, t 2.51
58, t 3.ul
UI

4
2
1
11
17

IIt

CONFIDENTIAL

s1
ss

tIt

CON FIDENTIALI

34

SUMMARY OF BALLISTIC LIMITS

TABE 3.

Part 1, Anno, Piercing Projectiles Versis Rolled Homogeneous Armor

Ballistic Limit

Plats

Nom.

Shot

Nom. t.
in.

BHN.
kg/mm 2

BL(A).
ft/se

Sigw ficance

4R(56)
4R(85)

Number

Notes

of
Rounds

(Abbreviations
explained on p 27 )

Cal,
mm

model
No.

Obl.
deg

75

T43

20
55

5.00
3.38

226
253

2147
2920

2.00

30
45

3.00
2.00

277-285
285
262-298
233
293
269
269
262
265-285
277-400
262-293
269

Mix 842 to 1191


Mix 1047 to 1648
Mix 1869 to 2215
2R(21)
1406
2R(84)
1518
Mix 1620 to 1800
2R(3)
2136
HPP
2198
Mix 1194 to 1414
Mix 1437 to 1965
1700HPP, 2105LCP
2R(28)
2210

19
12
12
6
4
6
2
2
15
42

300-308
300
308
300
300
302
302
302
300-321
300

2641
2619
2590
2403
2369
2632
2306
2427
2582
2366

6R(132)
6R(102)
4R(144)
2R(34)
6R(58)
41(98)
LCP
IFP
4R(82)
6R(6)

16
10
6
3
9
9

321
291-321

2180
2025

2R(33)
4R(118)

4
9

T148
75
(11 exp. lots ( a ))

55
60

2.50
2. 50
3.00
1.50
2.00
2.50

75

T 1 49 (b)

20

4.00

45

3.00

55

3.00

60

2.00

SD&S. t 5.02
SS

1148. 4 lots. SB
M72 mod, 3 lots. sB
T148, 3 lou, SI&B
T148. 1 lot, SI
M72 mod, 1 lot, SI&B
T148. 1 lot, Sl&B
T148, 2 lots, SI&B
T148, 1 lot, SI
T148. 3 lots, SI
T148. 8 lots, SI&B
M72 mod. 2 lots, SI&B
T148. 3 lots, SI&

7
7

RC61, t 3.97, SB&S


RC55, t 3.96. SB&S

ic6 1 ,
R-61.
Rc55.
RC61
RC61,

R"61. w/o tip

w/tip
t 3.01, SS
t3.01. SS
w/tip

RC61
R",5, SS

10

Rc6l. w/tip
RC61. wyo Up

The great majority of these teus with the T148 dhot were performed under Project TAI-1251. This project used three varieties of
M72 shot with its up cut off. and six vaieties of M72 shot which were (ist 1oftened, then had their Ups cut off, and were rehardened. Theme sx varietAcs, differisug in windshield and heat-treatment, were designated T148. Since the reference report a)s no
variet) *Awed clear supertority, all are lumped together here. included also are two lots used in Project TAI-1301. For more
details, see Report 2 on Project TAI-1251.
(b) The teas with the T149 were part of Project TAI-1254. Shot were used with iockwell C hatdess either 61 or 55. In addition,
thome here marked w/*.p were utuuicated, then had the nps reattached. Those marked w/o Up were trucated, but the taps were
not replaced.

(a)

I
I

I
*AII

LL 9

N MIkMONIAL

1I0 6 T I?

CONFIDENTIAL

CONFIDENTIAL
35
TABE 3. SUMMARY OF BALLISTIC UWITS

Part 1.
Shot
Cal,
mm

Model
No.

Nom.
Obl.,
deg

76

M79

Armor Piercing Projectiles Versus Rolled Homogeneous Armor


Plate

223
229-341

391
1089

2R(36)
SEW 15

3
17

SI, t 2.24
Sl

3.00

251-363

1336

SE x 11

25

S1

18

Sl

SI

SS

1
5
11
8
5

SS,
SI,
S1.
SS,
SI,

10
20

3.00
2.25
3.00

30

2.25

r200
260
f320
212-269

3.00
4.00
2.25
3.00

2.50

1578
SE=64
1640
SEW 0
1762
SE= 48J
Mix 1932 to 2234

311

2577

HPP

363
223
302
363
223

1424
102U
1782
1301
11936

HPP
2R(27)
2R(41)
4R(100)
2R(20)

r230
2.50

55
60

13i1

SE

51

16,14

SE.

40

32

Sl&S

390
302
363
207-302
22,1-233
331
379

17s5
2 65
1 27
2532
1852
2246
2012

SE.63
2R(24)
4R(93)
SE a 34
SE a 37
4R(74)
2R(24)

6
7
14
16
7
7

SS. t
SS, t
SO&S
Si&S
SS. t
SS. t

230
310

2004
2057

SE mall)
SE $Mau

21

331

2356

2.25

269-293

3 0

379

3.00
2. uU

260
306

235

PP's or CP's.
6

SS, t3.05

2U32

4R(56)

SS. t 2.28

2369

4R('5)

SS, t3.05

2976
2178

2R(36)
4t(32)

3
6

SMS
s

2520

SE1.12
SE 10

26

Ss

2B38

2R(3)

so

CONFIDENTIAL

IIAL
O L L90

3.05
3.05

4R(92)

2o2626
2..

3.05
3.03

fS&S. Parabola fitted


w/o overlapping by

SE small

3.00

3.0o

t 3.03
t 2.24
t 3.05" 937AR
t 3.03
t 2.24

310

39163
50

Notes
(Abbreviations
explained on p 27)

2.25
2.50

5. 00

45

Number
of
Rounds

BHN.
kg/mm 2

4.00

40

DalListic Limit
BL(A).
Saigfift/sec
cance

Nom. t,
in.

O a#A
E
. TT 01k I

Ia

CONFIDENTIAL
36
TARXE 3.

SUMMARY OF BALLISTIC UMITS

Part 1. Armor Piercin. Projectiles Versus Rolled Homogeneous Armor


Shot
Cal.
Mm

76

Model
No.

T128E6
(M339)

Nom. t.
in.

3.00

Ballistc Limit
BL(A).
Significance
ft/sec

6R(43)
6R(133)
SE - 34
2R(8)
to 256o
2R(39)
SE - 43
SE - 101
6R(113)
4R(J5)

4.00
5.00
4. OU
5. 00
2. UO
2.50
3.00
4.00

5.00

250

3179

HIP

75

2.00
2.50
3.00
4.90
3.00
2.00
2.50
3. U0
2.00
2.50
2.00

290
291
298
260
302
290-21
291
298
2J 0
291
290

1J34
2266
2718
3206
2506
2481
2851
3138
2830
3112
3253

6R(151)
6R(81)
6R(127)
4R(61)
2R(55)
SE u 15
6R(135)
4R(67)
4R(64)
6R(155)
HP

8
3
7
8
5
15
9
7
9
6
4

30

4.00

--

2302

--

2653

--

2382
2640HPP,
2554
2441
255
255t;
.572

4R(73)
4R(77)
2R(69)
2d22LCP
4R(14)
4R(103)
2R(46)

8
9
8
4
7
I
3
4
6

45

55
60

65

.. UL

298
250
302
23v
261)
2,)8

Notes
(Abbeviauons
explained on p 21)

Number
of
Rounds

1517
1J87
2043
2221
Mix 2088
2443
1358
16J6
251
2832

30

T166(a)

BHN,
kg/mrnm2

298
238
260
26U
260
260
290
2)1
238
25u

20

76

Plate

Nom.
o,! .
deg

15
9
16
8
13
5
10
13
16
7

SI
SB&S
SI&B
SI, t 5.13
SB&S
SB&S. t 5.13
SI
SB&S
SB&S, t 2.93
SB, t 4. 07

2R(611
4R(66)

SB, t 2.04
SB&S. t 2.52
SB, t 2.93
SB. t 4.07
SB
SB&S
So, t 2.52
SB, t 2.93
SO&S, t 2.04
SB&S, t 2.52
SB, t 2.02
E2,
F2,
E2.
El,
E2,
E2,
E3,
E4,
Es,

RC66, W/o WS, SS


R&2. w/o WS, SS
Ra',, w o WS, SS
S5&S, BL(P) - '1446
RC6., wWS
w/o tip or WS, sB
S8

SI&D
bl

(a) The tests with the T166 Are from Project TAI-1301. The T166 shot is essentilall the T128%E (or M334) shot with its tip first
removed thei reattadied b cemenung or welding. The vairotis modificatiuons, El to E5. refer to Ahsht chances in the geometr)
of the igive and up. Modificati i E2. which was the prefened one, also was tested with or without the tip a#dWor windshield.
and at atrious Rockwell G hardneses at the bourrelet.
Pr,ubit araltsex Vpplied to all the T1"

data lot the fist five combinatons of obliquity And plate thickness $ive the following
rc-uit',
,11divel.
, 273, SE & 'Ri L x 2661, SE a 533; B1a 1 l1, SE a 46, 9L . 2543, SE a 34; IL a 233, %sa 61.
Tic
reults illusittet the effrcts of girouping %Ihhd)diver,4cat data.

SA ? 1

LLI

M E M O# AA A LI

1 11 V

CONFIDENTIAL

Ia

CONFIDENTIAL
37
SUMMARY OF BALISTIC LIMITS

TABL 3.

Part 1. Armor Piercing Projectiles Versus Rolled Homogeneous Armor


Model
No.

Obl.,
deg

Nom. t,
in.

76

T166

60

2.00

90

M77

Ballistic Limit

Plate

Nom.

Shot
Cal.
mM

Notes

of
Rounds

(Abbreviations
explaineJ on p 27)

BL(A).
ft/sec

Significa ce

288

1714

LCP

El, SI

289

2027

2R(58)

E2. RC66, w/o WS, Sd

21
289
306
306

190i
2048
2047
1040

6R(152)
2R(51)
LCP
2R(67)

9
4
5
6

E2.
E2,
E4,
ES,

2.50

306

2618

2R(1)

E2. RC62, w/WS. SI&B

3.00

306
306
260

2514
257J
2977

2R(85)
4R(14-)
HPP

7
6
4

E4. SI&B
ES. SB
E2. RC62, w/WS, SI&S

260
295

2)16

HPP

E4, SS

2R(3-1)
HPP

8
5

ES, S8
E2. w/o up or W$, S5

SI

70

2.00

291

2880
2,303

2. 50

252

1281

LCP

1550
1537
1643
1545

LCP
SE 461
SE. 24

3.50
4.00

262-28?
r200
.f260

L320

Number

BHN.
kg/mm 2

SE

40J

w/o tip or WS. SI&B


RC57, w/o WS
SI&B
SI&S

37

SI. 4 shattered

5.00

217-274

Mix 1804 to 2527

8.00

220

3204

HPP

44
2

sI&S
SI. t7.94

10

8.00

220

3430

HPIP

SS, t 7.94

20

2.50

255

1254

Lcp

SI

229AR

3.00

......

5.00
6.00
2. 50
4.00
5.00

220-228
270-275
229-311
207-269
220

35

6.00
2.50

45

2.50
4.00
5 O0

220
255
r230
2130
f350
207-2864
220-224

55

2,,0

252-255

203-1

60

3.
3.38
2.00

2
230-24
2.4

200
242
2144

30

{,

2. 50

to 2886
HPP
SE . 18
Sf325
4R(119)

33
10
20
12
5

29)S

2R('!5)

SS, t 5. to

3220
1489
la3
1843
1775
2614
31-,1

HFP
4R(63)
SE 24
SE 20
NE 29
6*(81)
4R(105)
S
I
4R(24)
it* 34
40(7%I)

3
I

SS
SIMI

38

SSs

10
13
1A
#
43

so
sS. t 4. 41f

NI I

M ID

Ai. , aw&~ hited


%

23

23

s ff

IN

S, t 2.-4I

o o verlap

by W~SWcr,

CONFIDENTIAL
I

S3&S
SS
S1
S6s
So, t 3.15

wnaU

223
tF,

Mix 1096
3146
1234
2334
2703

It V

CON F1 DENTIAL
38
SUMMARY OF BALLISTIC LIMITS

TABLE 3.

Part 1.
Nom.
Obl.
deg

Shot
Cal,

Model
No.

90
T33
(misc. models,
other than E7)

Armor Piercing Projectiles Versus Rolled Homogeneous Armor

Nom. t.
in.

DHN,
k/mm2

Balistic Limit
BL(A).
Signiflft/sec
cane

.50
4.00

229-364
210-300
326
200-311
265
220
220-228

1495
1558
1307
1873
2150
3234
2008

LCP
61(119)
LCP
SE a 14
SE a 114
HPP
SE a 53

1939
2852
3244
1880
2000
3198
2370
2291
26P3
3004
2613

SE v 29
4R(51)
HPP
6R(57)
2R(20)
2R(97)
SE a 50
6R(148)
SE a 11
SF"38
SE a 17

Plate

5.00

6.00
20
30

60

3.38
4.00
3.00

220-280
220-225
220
290
262
225
2g0-302
285
230-263
282-294
2'3-311
259
256
4,3
308
241*-311
34
29 !

192
2302
2418
2135
2371
3026
2733

2R(113)
2R(5)
2R(7)
SE a 41
SE a iO
4R(95)
SE a57

6
S
35
s
6

60

5.00
6.00
6.00
4.00
3.00
4.00
3.00

20

5.00

22
27

"
473
2093

LCP

11

0
20
30
55

91)
T33E7
(Proiect AX23s)

4R(8O)

PP12626 to 2749
";52
t.CP

(')

9
16
5
31
14
2
271
24
9
3
13
10
2
25
7
302
12
56

4.00
5.00
6.00
3.00
3.00
5.00
3.00

40
45

90
T33E7
(not part of Project AX23)

8.00
5.00

Number
of
Rounds

s
20
1

Notes
(Abbraviations
explained on p 27 )

SI
Sl
SI
SI
SI&B
SS
SI&S, acceptance test,
SI&S
SB&S, t 5.11
SS, t6 06
SB. t3.02
SS, one PP at 2073
SS. t 5.08
SB
w/boom mad tail
SS. acceptance tests
SB
SI, t 5.10
SI. t5.96
SI&3B t6.07
77ARO

LA. RC62. t4.99


LR RC62 S
LR RC62 . shatter zap
Lii
SS. t4.99

Pr14c AX!i 'ha' a test of steel for the shot T31E--. with various lots ddffering metallurgically. The Rockwell C hatdnrs of
tie different lots is shown here brielh. For exarmsle, R C62 rtns hadness 62. Other meatllurgical features can %ef.,'id

in Proiect AX23 repotts h notisg the lot numbers. Here LA and I" detot the reference lot. with and without windaicl4
LI to l10 dencte Lots FA-PD-1 to PA-PD-10; and LII d4enotes the lot made from PS-4160 steel.
In P tect AX23,

eers at obliqmuit 20 and 2.N derees v"e ctwled to velocities well above the low ballbik limis, in an

effort to find wloclt' tanges where projectile shatatin


entties hodwn "Xlf fot

might procgce WIal poetatok*A. Thi accom. for the multipie

W ballblc Umis.

0 At It It
L

iL M

0 01 1 A L

11

1 Iy Uira

CONFIDENTIAL

CONFIDENTIAL
39
TABE 3. SUMMARY OF BALLISTIC LIMITS

Part 1. Armor Piercing Projectiles Versus Rolled Homogeneous Armor


Shot
Cal.
a=

Nom.

Model
No.

90
r33E7
(Project AX23. continued)

Obl.
deg
20

G.00

BFIN.
kg/mm 2

25

55

s0

4.00
5.00

3.00

Notes

of
Rounds

(Abbreviations
explained on p 27 )

MIx 2495 to 3155


2472
2R(27)
4R(86)

51
14
6

L.R RC6 2. SB&S, t6.04


LS RC62. SI&S. t6.03
L1, RC61, S][B. t6.02

262-280

Mix 2944 to 3118

16

Ll, R

276-280

2530

l.

shatter PP's

4R(69)

L2. RC 1, SI&B. t6.07

Mix 2982 to 3149

13

2R(72)
to 3118
to 3389
HPP

7
9
27
18

6R(162)

L2, RC6 1. shatter Pp's


L,3. RctO. SI. ,G.02
L6. RC60. shatter PP's
L8. Rc 6 O. SB&S. t6.04
L10, RC60. SS. t6.03
L7, RC58, SI, t6.03

3092

LCP
4R(82)

16
14

L7. RC58. shattered


L3. RC55. SD&S. t6.07

270-280

3128

2R(9,,)

14

262-276

3124

2R(73)

12

L4, RC52. SS. t6.04


One CIA). no CF(P)
LO. RC50, SS. t 6.04
One CPA), no CP(P)

269-276
269-276

2519
- Mix 2934
Mix 2531
3158

269-276

f2554

269-275

12909

277

f2iut.;.00

2R(15)

277
249-300
277

f .P 2311 to 3041
2891
LCP
Mix 1984 to 2584
2825
2R(78)

10
1
28
8

6 2
.
LR, Rc
2
LR. RC6 .
LR. Rce2,
LR', Rc2.
LR, RC62.

277

2855

2R(22)

277

2944

21(16)

2-301

300

2507
2511

SE-a15
6R(100)

29

2510

299

2457

299

2490

41(35)

299

259

61(119)

302

2489

4R('.7)

295

2515

6R(69

302

2512

2R(27)

299
29
299

2449

4R(19)

2518

6R(63)

2441

AS5

2.37

7
IN
10
13
0

L4, RC 5 2 .

S. t4.98

L3. lC29 SS&S,

7
A

Li, RC40 sU&. t3.01


L2. IcRCS
SS
L6. Rc6O. S, t3.02

LI. RC60, ss

2428

294

2439

201

. 13

i"(P)

4R(69)

2,00

Ll. ROl1. SS. t4.98

6R(47)

2"9

9 0l

SS. t4.98

LI. RC- 2 . SMSS


L'R. RC12. SS

4.00

3 10

SS&S, t4.)9
SB&S. t4.99
So, t4.99
SU&S, t3.98

2
11

60(iN)
CR(P2)
2R(34)
00(114)
()

I L L. a

Number

269-280
270
22506

262-270

30

Ballistic Limit
SigiufiBL(A).
ft/sec
cano:e

Plate

Nom. t,
in.

A LY

L10.

L3. RC;SS.
L4, A
M~ Rcsv,
L', RI 42.
12.

1 1 It

CONFIDENTIAL

C'O. SB&S, t2.94

"C5, SS, t2."


111. I C4, SS
L7,

Ic-

SS
S

SS, t 2, 9
56

. SS

Ia'

#.C1

.0

)
,-;
4 Ot, 2_

I "

t3.02

...

CONFIDENTIAL
40
SUMMARY OF BALLISTIC UMITS

TABLE 3.

Part 1.
Shot
Cal,
U=
90

90

Model
No.
T43

T54E1

Armor Piercing Projectiles Versus Rolled Homogeneous Armor


Plate

Nom.
Obl.
deg

Nom. t.
in.

55

0
30

45

55
60

55.00
s
60

t It

Number
of
Round&

Notes
(Abbreviations
explained on p 27

302
210-326

2398
2990

6R(22)
6R(116)

10
13

SS

7.60
4.00
5.00
6.00
7.60
4.00
5.00
6.00
4.00
2.00

258-297
292
245-347
249-296
258
295
245-337
297
297-315
285
r260
320

'3205
2068
2982
3223
3180
2553
3205
3121
2847
2428
26f1
2691

HPP
6R(53)
SF w 61
HPP
HPP
SE a 38
SE - 23
HFP
SE a 3
64(41)
SE a 31
SE w 18

7
10
33
19
4
15
31
5
26
10

SB&S. t 7.65
SI&S. t 4.01
SS&S
SS. t5.99
SS. f7.65
SS
SB&S
SS. t6.02
SS
S,. t2.03

41

SS

L390

2337

SE 20J

s0
340

3182

9w7

3064

SEa 42

41

SS

380

2995

SE

3.00
2.00
3.00

256-306
296
307-381

3034
2860
317.7

SE 52
4R(118)
SE a 27

20
9
25

SS
SS
SS

Mix 30,40 to' 3430


Mix 2951 to 3218

38
20

5l&E

4.00

20-285
248-30u

4.Ov

105
T182
(E2, E3. 4. ES)

Ballistc Limut
UL(A).
Sigmrfft/sec
camoe

3.00
4.00

3.00

65
70

INN.
kg/mm 2

It L

L I

0 9 M 0 N

AL

39J

It

I It U

CONFIDENTIAL

CONFIDENTIAL
41
TABLE 3. SUMMARY OF BALLISTIC LIMITS

Part 1. Armor Piercing Projectiles Versus Rolled Homougeneous Armor


Shot
Model
No.

Cal.
rnta
120

Nom.
Obl.
deg

T116

(hlodsE4, E5)

Ballistic Linit
BL(A).
Significance
ft/sec

10.00

239

3298

6R(103)

E5, SB, tO. 14

7.60

248

2464

2R(21)

E5, SB. t7.55

8.00

240

3344

2R(70)

E5, SB

30

6.00

241

2302

6R(6s)

FS, SB&S, t6.07

45

4.00
5.00

297
291
26 -2X7

1978
2243
2615

2R(42)
6R(142)
SE = 27

7
10
17

7.60

260

3369

HPP

5.00
6.00
4.00
5.00

259
247
262
267-287

2817
3127
2514
2942

R(105)
HPP
4R(55)
SE - 72

9
2
7
16

2.00
3.00
4.00

283
285
297

1752
2353
2460

6R(87)
4R(79)
6R(105)

2.55

230

4R( 58)

6E5.

5.u

293

287n

4R(93)

6E4

2.UO

27 7

2070

LCP

3.uO

293

2564

6R(I11)

4.0(

294

3%t64

2R(57)

t;

255

3106

LCP

60

ii

E4. SB, t4.02


E4. SB
E5. SB

E5. SB,
Er.
E4.
E.S,
E*;.

7.75

SS. t5.09
SB, tS.',W
SS. t3.96
SB

ES. SI&S. t2.02


[4 w/o WS. SS. t3. o1
E4. SB, t4.03

SB, t4.02
w/o WS, 'S
fES. SB, t2.02,

, BL(P) a 2116. 6;R(9b)


1S, SB, t3.04

E4 w/u WS. Ss. t4.u6


E5. SB&S. L4.02,
BL(P) - 3176, 6R(124)
I.,
5 SS, t,.02

70

2.00
3.00

2S5
285

2282

2962

6R(84)
4R(76)

75

2.0

294

2641

2R(lp)

ES. SB, t2.03

3.00

303

3.96

HPP

ES, SB. t2.91

250"2-4

1414

2R(41)

1642

ICP

M12

0
1

.,:0
.'

30

4.(m)

13'.."

:_o

3;,07 11,4

1643

~
4-,

Notes
(Abbieviations
explained on p 27

55

Number
of
Rounds

20

50

155

BHN,
kg/mm2

Plate
Nom. t.
in.

43
r

10

'

U-,

A II L L a
.

0i

I :P

2493

0(OI)

0 aIA
k

3.01

I, t ,-4
SI, t

"4,
a

1
&

TP
C-

,.(Ii;
.,(,24

E4 w/o WS. SS,

NA
4

11

2()

C(
5I,
SM-

1 0. a

t v f a3

CONFIDENTIAL

1.,

CONFIDENTIAL
421

Part 2.

37

Armor Piercing Capped Projectiles Versa Rolled Hcmogeneous Armo r

Model

Nom.
Obl.

Nom, t.

No.

deg

ir.

Shot
Cal.

M51

SUMMARY OF BALLISTIC UMITS

TALE 3.

Plate

1.00

IN,

kg/mm 2

Ballistic Limit
SignfiL(A).

Number
of

Notes
(Abbreviations

ca ce

Rounds

explained on p 27 )

ft/sec

240
320
400
240

199
1211
891
1189

SE a
SE a
SE
SEa

1-3
10
13
2

39

SI, 9ARO

1.12

310

1260

SE a,1

60

SI

1.2f

380
302

1148
1455

SE a 321
4R(93)

SI

1.50

230
300

1564
1630

SE - 141
SE- 7

687

L370

1641

SE w 13

1280

1949
2304
2226
2348

4R(61)
2R(42)
SE-151
SEu9 p

340

2517

2604
1324

SE-111

6R(155)
4R(83)

6
S

250
320

1260
1276

SE a14
SE a 12

f4oo

1062
1377
1501
1634

SE a20
GR(56)
4R(76)
S
6

12
4

1730
1774
1223

SEu12}
SE 16
41(38)

45

240
20

1409
1397

SE ai17
11

SI

1396

SE
1
SE 1

34

400
302-326
241-302

1563
1614

SE a 41
OR(122)

17
13

1
SL. tl.2

r220

1859

1.0

300

1902

BE

43
46

35

1.00
1.25

341-375
302

Ism
1791

40

1.00

320^

17^

400

1779

BE a22)MJ

us
3880-402

21HA
2110

23(31)
0(140)

429

2oo6

(22)

SAO

W
Is"

2.00
2.25

302-306
282

r220
2.50

10

3.00
1.12

20

1.00

281-288
302-321

1.12
1.25

302-321
302
r23 0

1.50

300

L370

25

1.00

30

1.00
1.12
1.25

1.12

OA

L L a

C::

324-385

3:70

0ai

6
3

St 16}

$r -20J
S9

14
1 a 20

AL

SL, t 1. 13I
SI

Si&S
SI&S
Ljo

~e
4)cOM
so

6t

1 1 %1

CONFIDENTIAL

SI, 9ARO

HM 4 321.

Safore

t aS3
.3
*.#7

It

Si&s
SL. t1.13

41
20
3

0 17

8ARO
t 2.36

23

..
l

23

0of

403I 1

ftuu
from
AD 1004,
Mjsl

e wth AV' $8

CONFIDENTIAL
43
TABLE J. SUMMARY OF BALLISTIC LIMITS

Part 2.
Shot
Cal,
mm
37

Model
No.

M51

Armor Plercihg Capped Projectiles Versus Rolled Homogeneous Armor

Nom.
Obl..
deg

Nor. t,
in.

WiN.
kg/mn i 2

allistic Limit
BL(A).
Sigmfift/sec
cance

40

1.25

241-255
311

1853
2060

6R(133)
6R(124)

Plate

230
1.50

57

M86

45

1.50

50

1.00

2186

SE a 21

2311
2653
3033
2794
2457
2649
2875

SE a 19
SE a 21
HPP
2R(73)
SE a41
SEa31
SE a 73

r240

2183

SE a 56

320
L400

2242
2060

SE a 30
SE = 31

2139
2395
2433
2360
2498

SE u 65
SE a 40
SE a 80
4R( 144)
6R(72)

300
370
302-364
375
240
1290
340

r.0

r240

1.12

.310

1.25

f310
(241
'55-269

D8at from AD 679.


58
15
3J

a from AD 686 and

J
}*

SI&B

10
1"r

S1. Low precision 9L


St
SB. t 1.23
SI. t 1.-1
Sl. t1.51

1.25

302

1(49

4R(79)

r240

1292

SE a22
SE a 23

56

L36,1

799

255
'230
1 290
TS(
2t0

1743
14,t4
1710
1702
2112

2R(46)

130
3

193

53 a

3.

210

2417

06

4.00

,
2440.

C9e

1.2.

301109

240
1.0

SE a13
SE a
SIE 23
SElI

21

13-2
1141

242104

[k L

aIN

211

Of # A L

SI

:0
fr

18 SSS

1.541

V a9 " p

4 t-

HNv 4o

(u'l
%Ifor IHN'*

IF a Inf.

Ml&|121

51

St a 1t
M a5
St * Is

W. SMA, 1,,
*i~al

13312l119

400

St

SI. t 1.,.19

r tNe I47

2W('4)

)..3

4v
300

S1

SE a 21

'41

0 A I

66

9
25

C.0

20

341

2R(61)
SE a 59
HPP

2.00

AD 1084. S for BHN


31. SI&S for BHN

86

2812
2770
222

1227

SI for BHN 302.


SI& for SHN 344
S.
Data from AD 1084.
Compare with AD 679

40

11
229-215
266-3"

.390

Si. t 1.27
S1. tl. 23

6
9

1.,0

1.-'0

Notes
(Abbreviations
explained on p 27 )

Number
of
Rounds

SI
. VMS%, a fY$Ytd

20

q smallhL

ftI

4,n

c~
'

F I T W It

CONFIDENTIAL

+q(

CONFIDENTIAL
44
TABLE 3.

SUMMARY OF BALLISTIC LIMITS

Part 2. Armor Piercing Capped Projectiles Versus Rolled Homogeneous Armor


Cal,
MA

Model
No.

Nom.
Ob1.,
deg

Nom. t.
In.

BIN.
kg/mm 2

Balhstic Limit
IL(A),
Significance
ft/sec

57

M86

20

4.00

25

3.00
4.00
1.25

220
357-388
347
388
302

2503
2807
2397
2765
1264

6R(148)
HPP
2R(26)
HPP
2R(20)

1432

SE.
19
SE a 15

Shot

30

Plate

1.50
2.00
2.50

3.00

/400

4.00

22

388

r 261-283
35

3.00

40

1.25

10
10
4

{240

300
f,360
58-277
r220
2J0
360
r 260
330

2.8-321
347-360

1470
1359
1755
1826
2116
2153

SE a 50

2648
2613

SE a 27
SE a 43

255u
2757

HFP
HPP

2732

SE 111

14'77

SE
SE. 23

2S6

24)6

SE

20

2720

IPP

40o

214

2W(I2,)

2.0o,
2.5 o

-85
360

303
-

St a 24

1.25

262-30j2

IOZ4

6 (wsh)

360

41-285

f 1.22U

2161

1',.Q
370

SI

15

SI
SI lor BHN k 311
SB&S kr BHN *341

167

J"[SS

rS&B fo, BHN k 32J


SS for BHN
w 31)
for BHN a 400
-41

SS. t 4.U2

52

l&l
Ni&S. probit dive:r ,
sS, lone CP at i743

10b
26

2128

3ui

31

24

Mix 2501 to 2792


PP2b16 to 2812
sE a 104
6R (G4)
SE = lt
SE .10
SE 12

1,5

2
4

2724
1465
1675
1646
1707

400-408
277-302
r 23j

SI
SD&S, t4.01
SI&B. t3.04
SS, t 4.02
SI

SE a 18
SE a 10
SE a 60
SEa31
SE a 36

2278

Notes
(Abbreviations
explained on p 27

Number
of
Rounds

32
7

S
I

25

s.
-1 Paraboic cib~

17J
SI:

did not cOlw,.


tot DIN 4 27

fNl

29AS

NJ&

10

1I

l
44S, t Z.4

SA
244
Sfor II

IRI

L41

aft

. SI
IVY

a A T T It L k I

0I 9 1 O 0 1 A L

11I

I5
I

CONFIDENTIAL

Nt

14

Ia

13

"

-4

CONFIDENTIAL
*

45
TABL

U.

3. SUMMARY OF BALLISTIC LIMITS

Part 2. Armor Piercing Capped Projectiles Versus Rolled Homogeneous Armor


Cal.
MW

Model
NO.

Nom.
Obl.,
das

57

M86

Shot

Ballstc Limit
StgflBL(A).

Plate

Notes
(AbbvIauo

Rounds

explained on p 27)

DHN,

in.

kg/mm 2

ft/s .

55

1.50
2.00
2.60

269
285-302
353

60

1.25

273-352

2341
2R(48)
PP 2585 to 2712
2705
HPP
2179
SE 28

5
15
5
30

t 1.48
SI. lone CP at 2691
SB. t 2.51
SI

1.50

300

2634

SE 4'1
SE 33

53

SI

360
316-352
277-341

SE 48
2592
2504
SE u 24
PP 2573 to 2710

23
21

SI&B
S1, lone CP at 2664

287
282
232
273
282
262
267-287
224
269

952
1175
1596
1755
1548
1971
958
3175
2097

1
8
3
3
5
3
17
3
1

Sl. t 2.36, one PP


SI. t 3.98
SI, t 3.98
S1. t 2.36
ii, 1489ARO
Sl
SI&S. t 5.12
55. t 2.51

265-265

2047

262
400

2095
1782

HPP
2R(64)
2R(41)
2R(34)
2R(57)
2R(36)
SE4
HPP
HPP
2R(6)
HPP
LC

225
253

2324
2429

4R(50)
HPI

4
2

223-293

24

51

26

SI

202468

75

Number
of

Nom. t,

65
70

1.25
1.25

1.50
2.25
4.00

M61

20
25
34)

45
60

2.25
3.00
1.50
5.00
2.50
1.50
k.00

75

T42

20
56

6.00
3.38

76

M62

2.25
2.0

cance

1348

SE0S

"ro

1420

SZ-29

1 290

1375

S1.21
St a 26

340

47

[100

1603

st- l,

3.00

350
400

4.00

204-326

1330
1204
lol

SE143
sE.31
SK[a)3

3.0

302-346

17 2t

4itl

147

14

21

2
1
SS
SS, t 3.44

26

1-1, 214AW

O21)
71(

Att$LL9

00 I 0

00

1 A@ISL

CI.P at 171

001*

t1

CONFIDENTIAL

t6

CON F1 DENTIAL
46
TABLE 3.

SUMMARY OF BALLISTIC UMITS

Part 2. Armor PierLine Capped Projectiles Versus Rolled Homo6,eeous Armor

Sht
CaI.
26M

Model
No.

76

M62

Nor.
ob. ,
deg
30

Plate
Nom. t.
BN,
in.
k/mWrn 2
2.00
2.25
2.50

3.00
4.00

40

2.00
2.25
3.00

23.5-277
223-2J3
230
2J0
340

1472
150u
1627
16",2
1558

SE 3 16
SE - 22
SE an43to
SE w 26
SE z 38

"220

15J4

SE

310
.410
202-273

1 J5
2285
2333

SE 89
SE
-25
6R(16))

244-302

2674

6R(74)

235-269
223-2133

1861
1772

SE u 48
SE a 59

r220

2146

SE :77~

310

2437
2211

SE 43
SE. 51

1.50

(201438
300

1460

6E
SE

2.00
2.25
2.36

t~701408
235-269
258
255
230
310

1789
2079
1983
2D14
2280

SE 23)
SE 19
6R(')3)
6R(98)
SE 1181
SE 76

2.50

50
55

60

3.00
2.00

2287

SE- 122J

2.25

2. 25

2664

2R(78)

259
248
223-293
258

24
2082
2309

S1
SI

23

SI

a 18

9
28

KPP
2R(61)
SE a 56

14

SI&S

13

Sl

14
13
27

SB&S (;hatter gap)


SI&B
SI

36

SS

52

Sl

19
9
11

SI

47

SI&B

I
3
30

is
Si
SI

255

Max 2437 to 2684

t 2.31

2w'

2592

SI

r=30

2114

SEa31

45

Sl

2062
2640

sita37J
0R(100)

11

sI, t 1."9

1fF

1.50

02200

t~7
=16-V?7
246

2"9

11PP
SE a 67'I

H"

2.50

22*-341

262

2.50
3.00
3.50

262
229
262-20

1310
2R(i13)
Mix 1611 to 1764

4. 00

1.0

r~x174

$1 f 6;

2334

W 0 4

S.

7
7
2

bI
Sl

1071

S.

9w0
274

SA4 f r It L L It

1)90

#4

00

A L

to

t S.07

CONFIDENTIAL

a,

t 2.3]

2.50

00

M82

f.390

13
23

2.36

2.25

90

00

Notes
(Abbreviatons
explained on p 2'7)

Number
of
Rounds

Sigmficance

L4
45

Ballistic Limit
BL(A),
ft/sec

I1

CONFIDENTIAL
47
SUMMARY OF BALLISTIC LIMITS

TABLE 3.

Part 2. Armor Piercing Capped Projectiles Versus Rolled Homogeneous Armor


Shot
Cal,

Model
No.

Nor.
Obl.
deg

90

M82

10

BalUstic Limit
Sgpufh@L(A),
clnce
ft/sc

6.00
8.00
2.5a
3.00

236
209
255
229-285

2592
3115
1418
1612

2R(45)
LCP
2R(36)
6R(95)

11
2
4
9

SI
SI. HPP(P) w3145, t7.94
Sl
t 3.01

4.00
5.00

204-326
220-232

2019
2442
1563
1487

SE a 37
SE a 15
SE small
SE small

28
42

SI
SI&B

L340
1220

1433
1818

280

1778

SE small '
SE a 15
SE w 1"

t340

1744

SE w17

(200
260

1316
2253

SE a 157
SE a 68

t320

2500

SE a 170

r230
30

Notes
(Abbreviations
explained on p 21)

BHN.
kg/am 2

Plaw

20

Number
of
Rounds

Nom. t,
in.

2.50

3.00

4.00

280

Sl.
17

Parabola fitted
with almost no over-

lap by PP's or CP's


48

52

Sl

.5
3

SI
SE, t 6.03

5.00
6.00

200-300
220

2654
3041

SE a 48
2R(62)

35

2.50

255

1720

6R(127)

40

3.00
3.00
4.00

226-Z4
235
223

1854
2050
2711

6R(83)
LCP
4R(142)

12
4
6

t 2. 38
SI&B
SI

45

2.36

255
220

1.0
1J.J

2R(63)
SE a 61

t 2.31

2. b0

290

2075
1936

SE a 35
SE w 45

41

SI&S

2186

SE * 35"

23 )3

SEa21

201

SI&S

3.00

t340
202-204

2357
2617

244-273

2820

HP1P

5.00

220-224

311,4

405)

50

'..25

254

Mix 180J to 2231

56

2.50
3. 30
2.36

264
226-269
255

2386
25.2
2421

2.50

245-2

2607

3.00

269

reS6

3.38
2.26
L60

43-263
246
22t-311

3231
2504
2180

",137)
La
4i(#4)

4.00

60

*A

LLIt

0 It 10 0 0I

1E
10
R(144)

A Li

SI

SI, t 3.37

4
S
7

SI&S, t 3.)

2R(27)
$SE
2R(dI)

4
l6b
3

ss, t 2. 1

S1026

I)

$I

S&S, t 6.02

t 2.31

2
11
3
11

lll lit

CONFIDENTIAL

v ti

WS.t1.46
CP(I4J at 2044
U&S. t L 3

CONFIDENTIAL
48
TABLE 3.

Part 2.

Armor Piercing Capped Projectiles Versus Rolled Homiogeneous Armor

Cal.
mm

Model
NO.

Nor.
Ob.
des

Nor. t.
il.

90

T25
T26
T27
T28

20
20
20
20

90

T39

30
45

ftt

SUMMARY OF BALLISTIC LIMITS

N,
k&/mm 2

Ballistic Limit
IL(A).
Sigrufsft/sec
cance

5.00
5.00
5.00
5.00

220
220
220
220

2314
2R(39)
Mix 2174 to 2322
2421
6R(88)
2366
4R(124)

60

3.00
3.00
5.00
3.00
3.38
2.00

220-223
262
220-223
290-320
263-270
285

Mix 2196
2253
3061
2735
TJ65
2373

55

Plate

90

TSO(a)

46
55

5.00
3.38

220-224
243-260

)0

T50E1

5.Ou
6.00
7.60
6.00
H, 00

25)
256
258-297
248
240
2,31
230-232
240
279

20

3.00
4. O
50
5.00

30

6.00
7.60
6.0o
4.00

40
46

5.00

f~340
29:
27
266
262
2,45
245
2'5-337
247
290-300

55
46

SI&S, 8 special lots


SR&S, 8 special loUt

2176
2385
2845
2562
J280
1731

6R(90)
4R(90)
SE - 21
2R(40)
HP4
6R(103)

8
3
21
1

SI. t 5.10
SI, t 5.J6
SI
SI&D, t 6.07

*154

SEWS

22

2308
530

SEw 1
S E 34

42

SEa S

5%
SS

3141
2W
"140

mt73
*(
w1

24
1
1I
4

s$

33)6

"PP

)o -30
t

It 0 0 0

t 6.02
t 1.47

3"01

311

S1&,
Si&,

s
SS. t 4.43

3114

It L L ak

IV
10
6
3
#

34
6
5N

26)6

"o

A t't

6R(124)
"PP
6R(,S)
6R(120)
6R(161)
4 (46)
HIPP

st 4
H"
20(21

*6-3)4

4.00

326)

Sl

A
F
to 3136

3.00

{441

12

I1
5
41
20

so

3.0

t 5.03
t 5.34
t 5.04
t 5.03

21190
SE a 1.4
Mix 2811 to 3251

56

3.00
2.00

to 2585
6R(74)
SE - 74
SE a 68
SE a 166
4R(62)

SI&,
SI&S,
SB&S,
SI&S,

SI&S, 13 special lots


SS. 1 lot
SI&S, 5 special lots
SS, I lot
SS, 8 special lots
SD, t 2.03, 1 lot
S

4.00
3.00
4.00

fs
70

7
8
8
7

Notes
(Abbreviations
explained on p 2q)

58
13
27
83
45

,t4
3214
31193
*.51
WS4
2 44
312
3114
21l7
MIx 3044

247-31$

Number
of
Rounds

0IA L

411,30

#fastIli

I f utt

CONFIDENTIAL

S1. t 6.02
se
S&S. t 4.94

3.4

so. t

sIu.02
u.5
11

CONFIDENTIAL
49
SUMMARY OF BALLISTIC LIMITS

TABL 3.

Part 2. Armor Piercing Capped Projectiles Versus Rolled Homogeneous Armor


Sbot
Cal.

mm

Model
No.

105
T13
(Mods El. E2. and E3)

Nom.
Obl.
deg

Plate
Nom.r.
in.

30

120

120

T32E1

T14

T14

(bods El. E3)(a)

(a)

4.00
5.00
6.00
3.00

220
---

llistic Limut
BL(A).
Signifft/se.c
cance

Number
of
Rounds

Notes
(Abbreviauons
explained on p 27)

-*

2115
2187
2664
2612

LCP
6R(99)
LCP
2R(19)

1
14
2
3

El;
El.
El.
El;

55

7.00
5.00
6.00
4.00

210
221
243
225

2537
2620
2890
2975

SE a 18
SE a 18
6R( 102)
6R(155)

27
25
9
12

SI&B
SI&S
SI&S
SB&S. t 4. 01

20
30

3.00
8.00

209
208

2482
2719

4R(58)
SE u 2'2

11
23

SI. t 8.06
SI&B

45

6.00

--

2892

SE a 43

46

SB&S

0
20
30

8.00
8.00
6.00
7.60
8.00
4.00
6. 00

2482
2669
24310
2823
2678
3334
2427
3180

2R(41)
2R(40)
2R(36)
4R(77)
LCP
2R(137)
4R(70)
6R(38)

6
8
5
6
4

45

238
239
265
262
210
210
291
252-263

55
105

iIN,
kg/mm 2

30
45

2866

2R(41)

3334

4R(132)

10

ElS,

2999
2151S
Mix 2806
.914
3091

6R(160)
4R(63)
to 3115
2R('26'
HPP
4R(37)

6
6
20
5
2

E3 w/cap; S8
[! w/cap se
3 w/cap; SD, t3.91
E3 w/o cap; So
Cl w/cap; so

256
269
225-297
23)

5.00
0

2.00

ft.
300

I 1
f

2^00

El W/cap; SI. t6.06

Records P464-1 and P411-1".

L L 1

w/cap; SL t7.96
w/cap; SIAB. 17.96
w/cap; St. t 6.02
w/o cap; SI. t7.65
w/cap; SI, t8.12
w/u cap; SS. t 3. 12
w/cap; SB. t4.04
E3 w/c&p. SD

225

5.00
3.00
4. 0

At

E3
E3
E3
E3
El
E1

223-1.1-

55
60

Thuse twts are ttwkc t~prned in Fixwn

"3
11

SB
E2. E3; SI&B
E2
SI&B

0E

M6 9@

A L

I Of 6 11 1 1 to I ft

CONFIDENTIAL

3bth wo0

t3 Wlcap; SS

ap

I
CONFIDENTIAL
(

50
TABLE 3.

SUMMARY OF BALLISTIC LIMITS

High-Velocity Armor Piercing Projectiles Versus Rolled Homogeneous Armor

Part 3.

Plate

Cal.

Model
No.

Nom.
Obl.
deg

75

T27

20

3.00
6.00

--

30

3.00

--

5.00

--

Shot

76

M93

0
30

76
T4
(various mods)

BHN.
kg/mm 2

Nom. t.
in.

6.00
8.00
4.00
6.00

Ballistic Limit
SlgrufiBL(A).
cance
ft/sec

Notes
(Abbreviauons
explained on p 27)

Number
of
Rounds

1888
2694
2303
2526

2R(47)
HPP

6
2

2R(40)

2R(46)

t 5.12

213-225
220 et al.
220-229
223-245

2534
Z375
2265
2343

4R(94)
4R(25)
SE a 108
SE w 19

SI
SB
SB& S

SB

320

Xix 2767 to 3357

8
13
27
16
13
4

40

4.00

320

3276

2R(52)

r10

SE m686

4.00

t2 5u

2,J68

45

1301)

3065
3763

SE a3 8
SE 29

30

6.Ou

220

....

55

3.00

255

33,44

4R(114)

2
6

SB, HPP(P)
SS

0
2)

3.25
3.25

235
23

1564
1775

2R(l))
LCP

6
4

T4 Z17. Ss, t 3.31


T4E17. SS, t 3.31

4. 00

22J

2162

2R($1)

T4

6.00
8.00
3.00

250
223
208
231

2124
2"719
3167
1472

2R(1)
LCP
2R(100)
2R(10)

3.25

2W5

2072

2R(46)

4.00

220-240

.00
6.00
3.25
31.25
3.00

220-240
220-225
250
22U
23,5
235
265

SE 3L
a0
Mhx 2274 to 25t61
Mix 2066 to 2525
2825"P.
34(CP
30)9
2R(86)
24'u
4R(,4)
311,
k,3)
A3 '
HPP
33,
HPP
335wP,
364."P

30

40

s
55

2545

13
5
6
6

3403

T4. Sk&s
T4E17, %r%
T4, t 1.,33
T4
T4E17. Ss, t . 31

T4, s

4'4

T4E17. !S
T4E23, SI&S
T4. S
T4, SS, t 5_17
T417. Ss, tJ. 1
T4EP, S. t 3. 34

31
12
3
6
6
6

74. SS
T4E1 '

2
4

74E17.

I1

k, t a. it

I
I
k

o t w0

tA

lINSbY

?V

ftI

CON FIDENTIALI

CONFIDENTIAL
51
TAKl 3.

High-Velocity Armor Piercing Piojectiles Versus Rolled Hoasogeneous Armor

Pan 3.
Shot
Cal.

76

Model
No.
Special(a)

Nor.
Ohl.,
deo

55

226 et al.
285
220-230
235
220
240-277

Mix 2623 to 3742


4R(115)
2516
Mix 2638 to 3.527
Mix 3103 to 3294
HPP
3775
Mix 3154 to 3927

8.00

30

6.00

45
55

4.00
3.00

220
223
32D
320
255
240
250

3612
2702
3733
3017
2671
3576
256

2R(38)
2R (28)
HPP
2R(28)
2R(38)
2R(15)
2R(56)

6
5
3
5
5
7
S

30

4.00
6. uO

28)
253

26.)
4u56

2R(36)
6R(107)

5
10

60

2.00

0
30

7.00
4.00
6.00
7.00

T29(b)
76
(vaJous models)

T66E3

k1331
76
(various models)
(HVAP-DS)

8.00
4.00
3.00

45
6

2.00
3.00

so

4.%
(a) Tibe to

Ibate sawuned bit~fiv at

248
238-283
248
248
248
248
u48
23
300
285-307
20.5-307
211
213
256

2938
2661
3198
3332
35)2
34 )J*
,4)3
3572
3811
3815
370-4
3588
3500
3110
3"4ho
3 86

261-244

3 O.i

34

4323

--

T29E2
T23E5
T2J
T2'3E2
T29E5
T2J
T2JE5

176ARO

......

246

10 shot dciisn, SI&B


1 shot design. s6&S
14 shot designs. SB
J shot designs. SI&S
3 shot designs. SB&S
14 shot designs, SB&s

48
5
59
13
7
45

8.00
4.00
6.00
4.00
6.00
3.00

0
30

Notes
(Abbreviauons
explained on p 27)

Number
of
Rounds

Ballistic Lirut
GigmfhBL(A).
cance
ft/sec

Plate
BHN.
Nom. t,
kg/rmm 2
in.

45

76

SUMMARY OF BALLISTIC LIMITS

dewrild a Fmtng Xrcqw,4 IPOW'N,

6R(144)
2R(54)
6R(7.)
SE w 41
2R(3)
6R(155)
6R(133)
6R (10 1)
6R(170)
6R(43)
6R(s
6)
SE a 43
SE a 16
8R(IJ3)
SE 24
6R(In)

11
24
I(

M331A2
M331AI
M331A2
.M331
MJ31A1
,%Wlrk)
Lot KNC-E-1(c)
Lot KNC-E-;.c)
Lx KNCE-3r)
M331AI
M331A2
'513,1A1
I1331A."
NM331A*2
M331
M331AI

1t6

N133IA2

3.31A1

7
8
47
1
6C
6
8
b
8
10
61
iA

So?
HIp

P412035, al P43564.

dlbt SWW~AIly 1010Wi to**h T4, 60 with wvetaI nwel ebwse, n pjgrtmett

id matetiahs.

The tern ma-olv*4 14 vatiuct

A br~t ewnnpaft 0( te

*atbt
fh ele, bat*e) ae desmttbl in th tefevente ?-krd
FinalI eof p444 aU F444!'.
f m !
M te
IthIOU WW% Mttall*gc*Aly.
ft 0
1fuMPUsesed PS"W.
(C) t
mW me

A T It I LL I

W a

A I A*

I"

o?

CONFIDENTIAL

CONFIDENTIAL
52
TAKE 3.

Part 3.
Shot
Cal.
am

Model
No

90

M304

SUMMARY Of IAIJSTIC UIMITS

High-Velocity Arnot Piercing Projectiles Versus Rolled Homogeneous Armor

Nom.
Obl.,
deg
0
30

Plate
Nom. t.
in.

lHN.
kg/mm 2

7.60
4.00
5.00

258-304
280
275
338

6.00

8.00

260
1.300
262
297
239

4.00

280

7.60

45

r223

l&llistc Limit
IL(A),
Signhflft/sc
cance

2523
2178
2298
2397
2468
2559
2882
2938
3208
3140

Notes
(Abbreviations
explained on p 27)

Number
of
Bounds

SE w 15
SE = 37
6R(59)
4R(53)
SE - 59

21
21
10
10

SE=41

165

SEE 65J
4R(68)
6R(166)
2R(62)

SI&B
SB
SB
SS, t 4.97
SI&S. 392ARO

10
11
9

SB, t 7.65
SI&B. t 7.47
SI&S. t 7.96

rs3.

30

Mix 3017 to 3208

I
5.00
6.00

55

7.60
3.00

3527
3723
3548
3750
3011

SE = 9
HPP
HPP
HPP
4R(4")

34C3

SE

SE 18
SE 25
SE 46
HP?
6R(80)
6R(64)
2R(17)
6R(71)
df?

61

SS

257-321
290-368
391
257
363
391
319-391

3607
3462
3285
3748
3556
3556
3640
3432
3846

34
14
13
10
10
11
7

SB&S
SS
SS. t 3.98
SS
SS. t 3.05
SS
S3&S

203
212
252

2822
1689ff,
3124

2R(56)
1951CP
HIPP

3
4
3

275-320
249
297
275-297
280

r200
4.00
60

3.00
4.00

65

3.00

70

3.00

94
M304(&)
(w/20 lb care)

0
30
45

14.00
6.00
6.00

90

30

6.00
. 00

251-206
235

Mix 25
4100

s0

3100

302

3373

M332

22
10
8

9
5

to3 3"1
Hf

40
3

*(U)

So
SS
*IAl

()Tbese Noe vil&a U0-lb coe aet described in firog aeco- P41364.

,A ?2SLL I

t4.0S. Byprobit.
OLu2938.SE 201.
SS
S9, lone CP at 3700
SS. t 6.02
SB&S, t 7.54
SB. t 3.01

ASW@1016

,rut

CONFIDENTIAL

CONFIDENTIAL

53

TADUE 3. SUMMARY OF BALLISTIC IMITS

Pat 3.
Cal.
mM

shot

Model
NO.

High-Velocity Armor Piercing Projectiles Versus Rolled Homogeneous Armor

Nor.
Obl.
deg

90
T30
(various models)

20

30
i8.00

45

Nom. t,
in.

Plate

IN.
kg/mn.

allistic Limit
BL(A).
Signfift/sec
cance

8.00
10.00
12.00
6.00
8.00

208-220
220
220
245
220

2481
2733PP.
3298 o
2151P ,
2765

6R(113)
2916CP
2R(53)
2291CP
2R(54)

12
5
4
5
5

T30E16.
T30E16.
T30E16.
T30E16.
T30E16,

10.00
4.00
6.00

205
220
244" 45

3110
1958
2407

2R(61)
2R(41)
SE = 106

5
2
37

T30El6, SB&S
T30EIS, SD&S. t 4.03
T30E16, SB&S. 110ARO

220

2799

2R(108)

4.00

220

2906

203284

J0

T44

T63

T0

I!

SE

3610
369 3

SE
34
SE-39

21

55

6.00
3.38

044
235

3336
3376

.R(601,
2R(93)

5
3

8.00

1260
330

2605
2755

SE small
SE small

3
23

10

8.00

390
339-390
260

2665
2954
2665
3185
3026

SE small)
4R(103)
4R(63)
SE - 91
2R(37)

4
5
23
10

339

36"6

lll'r

8.00
8.00

L390
339

35

8.00

260

337q

GR(156)

40

8.00

26

3645

8.00

260

0
30

14.00
6.00

65

4.00

A 't t

i, k L 9

33

**
226

04 t

SI. t 8.OC
SI&S
SI&S. t 12.19
SB&S, t 6.06
SS

T30 E16. SS

T30E16, SS. t 3.57

34

260
f310

20
15

2R(74)

5.00

Notes
(Abbreviatons
explained on p 217)

Number
of
Rounds

T30E16, SI&S
T30Elo, S,. t C.06
T30Ed. t 3.31
SI&B. Parabola fitt d
wth
w. almost no overlap by PP's & CP'.
SI&S, t 8.06
sI&s. t 7.8
SI&S
SI&S, t 8.04

sl&S, t e of

IV

SI&S, t 8.64

2l (t)

Sl&S, to '.)4

3721

HPP

!US, t 8. 04

3426",
2212

317 CP
2k(.4)

4
4

s&l, t 14.25
SO, t 5.62

3531

4R(6,q)

HVAP-*D..

6'1

0 00 1 A L

Hi'P

#"

'f I
2U

CONFIDENTIAL

ss, t 8.o6

Lrl

,Us

CONFIDENTIAL
54
SUMMARY OF BALLISTIC LIMITS

TABU 3.

Pan 3.
Sftt

High-Velocity Armor Piercing Projectiles Versus Rollud Homogeneous Armor

Cal.

Model
No.

Nor.
Obl..
deg

Nom. t.
in.

90

T137(s)

10

7.00

(various models)
(HVAP-DS)

Plate

30

4.00
6.00
7.00
8.00

45
55

4.00
4.00

DHN.
kg/mm 2

Ballistic Limit
31(A).
Slgusflft/lec
cance

279

2561

279
277
262
262
2A2-254
242
235-249
235
277
248-294

Mix 2595
4"27
3831PP,
Mix 3728
3704PP,
4309PP.
Mix 3130
4637
2304
Mix 3608
3707
3966

269-293
248-300

5.00
60

3.CO
4. On

105

I"

T2994

T36

0
30

14.00
6.00

(w/-6 a"

("Me
aw

9
8
I
5
13
16
2
26
3
8
15

270

3900

12R(110)
SE - 61
2R(1)

285-288
293
285

4317
4260
3624

HPP
HPP
2R(35)

8
1
5

248-294
248-308
270-291

4115
6R(93)
Mix 3817 to 4151
4627
IPP

13
15
6

200
242

3562
231(N

2R(40)
2R(54)

30
45
6

El Mod 3(c)
E3 Mod 3(d)
E9 Mod 3(d)
El(e)
E3 Mod 1, E9 Mlod 2 (d)
El Mod 3(c)
E9 Mod 3(d)
El Mod 2(b)
El Mod 2 (d)
El Mod 3(c)
EO, El Mod 2. E2. 94(b)
El. El Mod 3 (c)
EO. ED Mods 1.2,3d)
E21( )

EO Z2. E4(b)
EP4
E7, I(c)
E0, E2. FA(b)
El M 3 ,odL7. t(c)
E,
ods I Maad
2(d)

4
6

SS
SS

10.00

197

3326

2R(S1)

SS

4.00

27.9

4.00

3474P",

LCP
3621CP

60

225
225

SS, KIP) a 23,


SS

0
30

14.00
6.00
10.00

-205

3305
222W.
262

HPP
234CP
41(47)

3
4
s

4.00

--

291

20(20)

14.00

2-S2

'*2

=(4

30

4.04

2U3

267

~ 2 (7

dedgm of NVAIPs-T, 900


OfE
tpum U soem too"ec.

Tiee ms".
"b
ufw
(4 Tfes muds aleap
(4 Thee msh
,ad e")
(40 Ms
lllillid

6R(152)
to 2800
LCP
4018CP
to 3948
3983CP
4424CP
to 4201
HPP
2R(67)
to 3763

Notes
(Abbreviations
explained on p 27)

45

44
IM lbSl

Number
of
Rounds

-o

i Irt woo wtud a Pat at rissc i-A1iw4.


The "a "A" aw"s we"w tv 69 -minTrm

ra

y bd Am 1. f-W cot ,,,0


l vo,".
had a 7.5 * 7,4&rS tc wi dobte
ei m.Te,
a 4. 006
w1ft &ft*
rome&[ r.w.
Io'm
"Wefto fo.

4dA T

Lk a

! A

It

CONFIDENTIAL

Jraes~ift

2Rj62)

15, t 14.26
*1, t 4.,U
iSi, . 07
"51 tb.vA

*4
d 1am. S"

GTg

CONFIDENTIAL
55
TABLE 3.

SUMMARY OF BALLISTIC LIMITS

Part 4. Armor Piercing Projectiles Versus Cast Homogeneous Armor


Shot

Plate

Cat,
mae

Model
No.

Nor.
Ob.,
deg

Nom. t,
in.

37

M74

1.00

BiHN
kg/mm 2

--

r230

150
1.00

240

1525

SEa 26

60

1.00

320
2.6-326

1622
240J

SE a 24)
SE w 17

2.00

r220

1132

SE - 1:1

260

1187

SE =

'400

1133

SEn 14

124')
1384
1652
1u
15

SE a 64
SE U751
SE- 14
SE u 34

2076

6R(11.,)

2.00

3.00

M70

2.25
3.00

248-283
(1)

{260
L330
(217-236

4.00

J 242-277
301
1f323

20
30

3.00
4.00

272
323
r20
- 70

2134
2.44

520

SI

20
32
1dJ

(Probit analysis tailed


to converge.
L SI&S, t 2,81 to 3.41

60

SI&B

1.56

SI&S

SI

76
20

SI

30:

St

12

SI

J1

lss.)
5h

6R(86)
IP

10
%

M&S, t 3. Iu
SS. t 4.16

217G

2645

SE.21
sE-l
a

37

425

2.00

1l,

2R(25)

.'t6@

sita3o}

V310
264-30t
I-297

2522
244
234

5
2
s103
La !

No.

3M5

st-

al a 0 O I I A

#0 It

s$, t 2.06

30
21
11

CONFIDENTIAL
f f Ot L L t

t 1.41

51
SS

st a

0a

SARO,

2727

2426

II

5,

4
11

,320

2.00

376

SE a 21

3.00
2.00

10

A3,AFo

2.1(4)
GR(84)

35
0

3.0q

Notes
(Abbreviauons
explained on p 27)

45

1.75

280
133o
252
230
270
310
199-232
235-270
288-305

Nurnber
of
Rounds

1225
SE w22~
1289
SE7
1120
SE 26
1422
2R(45)
1559
SE.3
1672
SE6 b
1723
SE w 27
Mix 2049 to 2226
Mix 2004 to 2435
PP2252 to 2J46
E24"
,
1269

1.50

51

Ballistic Limit
BL(A).
Sgm-fi
ft/sec
cance

I tL

sus
It
0

CONFIDENTIAL
56
SUMMARY OF BALLISTIC UMITS

TABLE 3.

Part 4. Armor Piercing Projectiles Versus Cast Homogeneous Armor


Shot
CAI.
MM

model
No.

75

M72

Nor.
Obl..
deg

PlAte
onom.t.
in.

OHN.
kg/mm 2

Ballistic Limat
@L(A).
Sisruhcamce
ft/sec
SE - 65
SE= 7 51

i J1156
, -u1248

3.00

('310

1194

SE. 35

2414

1263

HPP

(i 60

14 6

230

SE- 851

3.25

20

6. 0
4.01)

228
193

1602
1718
2058
1852

4.00

300

M79

30

3.00

25)

1959

2.(92)

76

T128E6
(M339)

20
30

55

4.00
2.00
4.00
2.00
4.00
4.00

251
255
251
255
251
251

2118
1210
2610
1768
2824
3212

6R(122)
2R(15)
6R(81)
2R(21)
2R(43)
6R(100)

3.00

190
240
290
190
24

1064
109.)
1135
1433
1452

-254
183-232
204
206
230
237
237-261

1356
16,)1)
2007
3124
3112
20,1
1983
1845

SE a 66"
St =20
SE =0 0
SE: 65'
SE - 21
SE 40
SE 44
at 25
HI"
HP
41(72)
L(P
21(44)

45

90

M77

4.00

T33

20
30
40
46

5.00
6.00
3.00
1.00
4.00
3.00
3.00

4.00

....

6.00

241-245

2144

4.00
10.00

06
10?

3167
313

,L
00

W4

3130

$1

a048w
2 2?
3641
If-t
170.374

29"
9000
216

Hall
S1141
Sa 1

MIS

S11131

46

&to0
Me
.0
211
4.#)

LAI
k

9S 0

SI

1015
2

SE= 19
SE a 46
HPP
2R(67)

75

Notes
(Abbeviauons
explained on p 27)

Numbe
of
Rounds

190

2
4

SI
SB
SS, t 2.85

11
8
8
7
7
9

SB&S. t 4.02
SI&3. t 2.04
SS. t 4. 02
Si. t 2.04
58&S. t 4.02
SS, t 4.02

61

St. 346ARO

177
23
49
I
1
7
2

St

SI
55
Ss

SI, t 3.$4
r1, t 2. IS
s. t 2. 4

1A L

SI

1A
Hl
HI

1
2

26AKn
S(P)
K
a2102, 20(37)
ss

! I of

14
40
70
21

If

CONFIDENTIAL

1
*S
0

It I

I
I

CONF! DENITIAL

57
SUMMARY OF ALUISTIC UMITS

TADLE 3.

Part 4. Armor Piercing Projectiles Versus Cast Homogeneous Armor


Cal.
am

goet

Model
No.

90
T33E7
(Not part of
Project AX23)

Nom.
Obi..
deg

23
30
45
55
60
90

T54E1

0
30

45

55

Nom. t.
in.

lItN,
kg/m

70

AI 1)

Notes
(Abbeviatons
explained on p 21)

232
235
248
248
232
235
219
233

1701
1610PP.
2090
2136
1866
2761
2671
2746

4R(80)
1731CP
6R(131)
LCP
4R(65)
6R(115)
6R(149)
2R(62)

6
9
8
7
8
6
7
8

7.60
5.00
6.00
7.60
4.00
5.00
6.00
3.00

247
238-258
254-260
245-260
243-282
238-273
246-267
258-266
326
220
288

2459
2553
2876
3289
2487
2732
3232
2410
2232
2810

6R(117)
SE a 15
SE a 31
HPP
SE - 52
SE a 35
SE a 82
SE a 34
SE a 24
SE a 37
6R(41)

10
2v
2)
11
22
20
20
20
10
15
11

S&3. t 7.55
SB&S
SS
SD&S. t 1.&3
SS&S
so
Se
Sl&S

4R(82)
SE 39

I
30

So. t 2.08
SO&S

43

So&$

2616
2263
2675

SI. t 3.97
St. t 5.19. DL(P) a 1766
S1. t 6.01
Sl&S, t 5.99. BL(P) , 2196
S&l. t 3.97
SS. t 5.19
t 3.99
t 3.02

So&$

2.00
3.00

230300

4.00

260

3084

3
St 3

3.0O0
4.00

221-300
2w
238

210 1
33"
3144

%IF
at 30
ff
2lt(l)

0
3
10

S
SS,
fA,

2.00
3.00

222
223

2929
3226

4R(49)
4R(31)

6
10

w, 1.91
ts. 3.0

tie

31"

GR(6)

224

22D

45

Number
of
Rounds

Balistic Limit
L(A).
Signifift/sec
canoe

4.00
5.00
6.00
6.00
4.00
5.00
4.00
3.00

4.00
60

Plate

fO

~ 2990
2M

Stag -

UtVIt

CONFIDENTIAL

4. 20
4.04, lone Cp

S$, t 3,06

CONFIDENTIAL
58
TALE 3.

SUMMARY OF BALWLSTIC UMJTS

Part 4. Armor Piercing Projectilcs Versus Cast Homogeneous Armor


shot
Cal,
mW

Model
No.

120
T116
(Mods M. E5)

Nom.
Obi.
de s
30

6.00
1.60
4.00
5.00
6.00

45

7.60
5.00
6.00
2.00
3.00

55
60

4.00
5.00
65
70

4.00
2.00
3.00
4.00

155

M112

kg/mm 2
248
252
275
280
238-244
249
252
233-274
246
243-244
275
325

2091
2779
2010
2239
2923
3130
3135
2693
3336
1593
36
2322

6R(95)
4R(34)
4R(5)
6R(148)
6R(132)
6R(89)
HPP
SE - 40
HIP
LCP
6R(96)
6R(94)

9
8
5
7
10
8
3
64
7
9
8
9

249

2600

6R(103)

2297

6R(123)

7
9

280

239
284
246
244-251
323
252-274

2854
2993
3277
2008
2604
286

J24

2R(32)
6R(114)
HPF
2t(41)
21t(50)
6R(91)

3314

HPP

3138
2632

4R(78)
4R(60)

f285
253-257

6.00
6.00
10.00
4.00
6.00
8.00
4.00
6.00

192-230
206
197
212-200
2W
183-25
i6-261
185-06

1509
2R(18)
2462
LCP
282
1W
Mix I12 to 1466
1722
4R(97)
1771
SE 129
1172
2R(74)
2690
6R(141)

6.00

185-195

1648

ILLa

M It

15

a At V

It

5
7
5
7
7
15
5
'7
8

2.00

45

& XXVI

WhN.

75

30

-.

Number
of
Rounds

Ballistic Limit
Sigpift@L(A).
cauAc
ft/sec

Plate
Nom. t.
in.

21
4
3
16
46
5
I

6i(162)

A L

W11S061
# U

CONFIDENTIAL

Notes
(AbbreviaUons
explained on p 27)

E,. SB. t 6.01


E4. S9, t 1.62
64w/o WS, SS, t 4.16
E4 w/o WS, SS, t 5.16
4, SS. t S.99
ES, SS. t 6.27
E4. SR. t 7.62
FA, SIB&S
a, SS. t6.26
E5, SIB. t 2. 06
E5. SS. t 3. 13
54 w/o WS. SS. t 3.16

E. SS. t4.18
E,5. SB. t3.90
4. SS.
4 w/o
ES, SS.
E.SO,
S,
54 w/o
E5. SS.
E5 SS.
E5, SS.
ES, s6,

t 4.96
WS. SS, t 5.21
t 4.17
2.07
WS. SS, t 3.12
t 3.12
t 4.17
4. 18
t 1.99

St. only one PP


SI&S
so
S1, only 2 PP. 234AR0
, t 4.90
SI, moly &ce. tels
SLAB. t 3.q4
S&S. t 6.04
5.161. 38Alo

CONFIDENTIAL
59
SUMMARY OF IALLISTIC lIMITS

TANA 3.

Part 5.
bo

Armor Piercing Capped Projertiles Versus Cas Homnogeneous Armor


MuN,
kg/mm 2

lallistic Umit
EL(A).
Siginifft/s
cance

260

Piace

Cal,
I=

Model
NO.

Nom.
Obi.,
deg

Nor. t.
in.

37

M51

1.00

1045

SE w141

f310

995

SE=

L360

948

SE

1684
1817
2032
2364
1167
1042
1368
1369

SE a 56
SE * 32
SE a 19
2R(21)
4R(62)
2R(33)
SE- 55
SEw 28

2273
2436
2624
1903

SE a 57
SE 12
SE a31
LCP

20

1. 50
2. 00
2.25
.400
1.00

30

1.00

38

2.00

200
280
L300

40

1.00

262

248-311
242-302
262-283
269
262-332
364
260
310

2601875

MN

53

SI

11
111
138
70
5
17
3
29

297

Sl
S1
Sl
SI. t 2. 87
Si, t 1.08
t 1.01
St

Si&S

SE : 41
SE 24

48

SI&

SE 44
SEw3l

37

SM

13
11
17
4

51, t 2.
SIAS. t 4.16

1.00

310

55

1.00

360
262-364

1968
1868
2326

3.00
4.00
1.50
4.00

2)-272
323
235-330
323

1649
2278
1224
2121

6R(106)
6R(57)
SE :HPP

1.80

302

1403

LCP

t 1. 55

t 1.41

45

57

Notes
(Abbrviations
explained on p 27)

Number
of
Rounds

30
35

330

1163

LCP

2.00

251

1783

"(M9)

2.50

254
201
27
230

2073
3491
1631
2M

"(11
Stu 4
S|a "11
29043
1"70

1727

51.35

47

1196
IM
:M

$|-45
(
6
3

3.00

45

1.50

50

3.00
2.00

1320
20
330

41

of

SAt

24-M

H,5
2.00
Lee

?SLtkLS

3
a"46
854-3
2f44o1

0 l 0

an
2736
fill

#t A0

cy
10(40
0("t
f

SS. t 4.16

sI&,

t 2.06

12

St. t 2.67

74

"as

42

:.1O
"

11

t1.01

4
1
0

S
IL L 44
5me. t 1.U
o

6 It I It w ta

CONFIDENTIAL

CONFIDENTIAL
60
TAUKE 3.

Part 5.

Bustic Limit
SignafDL(A).
cance
ft/sec

45

3.00
4.00
1.26
1.50
1.00

269
268
262
284
340

1418
1940
972
962
1991

2.00
4.00
3.75
4.00
2.00
3.00
4.00

241-291
314
280
217
188-207
243-307
269
204-245
273-321
220

4.00

250

1092
912
2014
1871PP,
%ix 2047
1243
1720
2286
2699
2379
2339
2734
1790

Model
No.

Nom.
Obl.
deg

'5

M61

0
25

76

M62

Armor Pietcing Capped Projectile* V,, m Cast Homogeneous Armor


NuN,
kg/mm2

Cal.
mm

Shot

SUMMARY OF BALLISTIC UMITS

20
30

36

plate
Nom. t.
in.

280
21

46

2.00

260

t310
2.25
2.50

259-311
259-6

r240-263
6

2.oo

262-285

1672
1590
1696
2291
2129
2150

M82

7
5
1
2
6

6R(87)
LCP
2R(44)
2033CP
to 2628
6R(84)
4R(100)
4R(82)
2R(8)
SE a 24"
SE w 36
SE a 25
SEm6'f''

16
4
3
8
20
17
5
10
7

t 2.08
t 2.08
t 3. 87
SI
Sl

46

SI&S

SE u44
SE v 59
6R(3)
SE w24
6R(126)
4(2)

20
30

237-269
217-200
218

6.00

15-232

6.00
10.00
8.00
3.00

206
1"7
206
235-303
2)

4.00
6.O

2043PF.
1687
2156
2341
3186
3167
3170
109
2115

202067110

2212CP
Sw 13
LCP
6R(146)
"IP

1W
LA;
St w 22
5E 41

1 0D

Sl

13
28
12
6

t 2.30

11

t 2.0J

6
44
4
28
2
2
2
47

t 2.95
'M
St, t 5. 12
,1
$8
St
S&l3

t 2.10)
t 1.94

fi

sit - 08
1118

4 1! AN

He

l At

Probit
analysis
diverged

fSl f 111IN

CONFIDENTIAL
OI L L It

SI. t 2.82
SI. t 3.94
SI&S. t 3.94

S:tS

130=1
Mt

S. t 2.82
Sl
SI. t 1.19
SI
SS, t 1.02

132

(39)

3.00
4.00
o.00

Notes
(Abbreviations
explained on p 27)

2R(37)
2R(48)
HPP
HPP
4R(127)

f3072211
90

Number
of
Rounds

Lt

t i

SPI

20
X 321

CONFIDENTIAL
61
SUMMARY OF 5ALUSTIC UMITS

TAR& 3.

Part 5. Armor Piering Capped ProjectileS Vesus Cas Homogeneou Atmor

Cal.
mM
90

f8ot
model
NO.
M82

late

MOM.
Obl.,
dg

Mora. t.
in.

35
40

2.00
4.00

45

3.00

253
229-301
900
1 280

66

3.25
4.00
3.00

s8
199-280
264

IIN.
kg/mms

90

T39

TSOE1

2
5

228320
2139
SE 80
2189
2(50)
Mix 2583 to 3220
2593
HPP

212

SI. t2.03
S1. t3.91
SI&

3
82
1

S&S
sm a)
t3.02

2216
2328
3129
2744

SE a 14
SE a 15
HPF
6R(104)

61

SI

4
10

S. t4.04
SI. t3.02

2652
SE = 24
2280
SE a 13
2623
SEu63
2943
SE a 21
2479
6R(140)
2652
6R(171)
2928
SES 19
3259
fFP
2M
S1889
3136
4R(147)
2727
U(2)
2168
41(52)
2740
Sf w 21
3213CF, 3447FP
F
R
312
8(60)
Mix 279' to2388
2.
3350
IF
P

24
33
2.%

2.00
3.00
4.00
3.00
2.00
3.00

247-269
238-258
254-260
245-260
238
281
238-241
254-267
249-280
245
288
225
279-303
220-304
262
22
240

Si
SI
iI
Sl
SI&3. t '1.13
S8. t Zo9
Si&
5&S, f'.2
,1
S, t 4.04
S, t3.95
S, t2.01
I'
,2 IP's , 31M
SS, t3.07
,5.t 2.07

30

7.40

246-M

*41

St11

1"

45

6.00

SW-37

344

515

7.60
.00
4.00

2"
2-8
W4-16.

314
i30
2Ml

PP
F
81(123)
*(131)

It

soF, t
Is

81(74)

&i , I 5,.11

.0
4.00

21
2"

3118
3814

(
IVP

:".M
t3,I
4'

30

5.00

260

55
60

4.00
3.00

1270
24
263

0
30

7.60
5.00
6.00
7.60
4.00

55

60

65
70

T140

2109

2R(49)
2R(4)
SE
0

SE- 171

45

120

2346
2691

Notes
(Ablu, vistiom
explained on p 27)

Number
of
Rounds

228

r230

90

Balistic LUwt
SinIpfiIL(A).
cance
ft/sec

6.00
6.00
3.00
4.00

5.00

70

18i0-16

34

(A) Pftbwft~m ia dill VIVM


Ulde MOW fed
WOM Yft PIN" hW600.
TgLs
i ow"
wit.
%HdARISC OwN $am* I*
ADM " MW

10
31
I
21
6
7
11
21
I
i

Wbdf
t~

CON~FI DENTIAL

41i 1

i~

.U

CONFIDENTIAL
62
TANAE 3.

Part 6.
Shot
Cal.

90

Model
eNo.

High-Velotity Armor Piercing Projectiles Versus Cast Hnogeneous Armor

Nom.
Obi.
des

M304

SUMMARY OF BALLISTIC LIMITS

Ballistic Limit

Plate
Nor. t.
in.

IHN.
kg/mm 2

IL(A).
ft/sec

Signiflcace

Number
of
Rounds

!
Notes
(Abbreviations
explained on p 27)

7.60

245-269

2414

SE a 13

32

SI

30

5.00

236-273

2220

SE a 15

22

Se

45

6.00
7.60
4.00

5.00

254-260
245
243
282
236-273
246-267
247
243-280
231-280
243-272
258-266
301
239

2482
2782
2728
2862
3128
3673
3744
2759
3374
2991
3795
3686
3750

SE - 24
6R(95)
6R(138)
2R(3)
SE w 57
SE a 41
HPP
SE - 23
SE w 13
SE . 11
HPP
6R(101)
HPP

21
10
10
10
22
21
4
21
23
20
14
10
4

SB
S1
SS, t 3.94
SS. t 3.85
58 at 236. SS at 273
SS
Se
SO at 243. SS at 280
SB
S5 at 243. SS at 272
SB&S. t 4.05
SS, t 3.93
SB. t 5.16

70

3.00

245-280

3853

HPP

55
60

5.00
6.00
7.60
3.00
4.00
3.00
4.00

90

T30E15

20

10.00

191

2693PP,

3065CP

90

T44

0
25
35
40
45
50

8.00
8.00
8.00
8.00
8.00
8.O0

223
223
223
223
223
223

2513
2816
3248
3445
3660
3672

4R(36)
6R(101)
6R(34)
6R(72)
21(25)
IHP

10
11
10
10
6
3

15Z

T35

30

4.00

201

1467

HPP

fA

f6LL6

hMLa

I A 61

# " 0

[S9 at 245 w/t 3. 08


SS at 280 w/t 3.11
SS
SI, t 8.,)2
SI&S. t 6. 02
SI&S. t 8. 02
SI&B. t 8.02
SI&S. t 8.02
SS, t 8. 02

CONFIDENTIAL

t a

CON FIDENTIAL

APPENDIX

BIBLIOGRAPHY

CON FIDENTIAL

CONFIDENTIAL
A-I
APPENDIX

BIBLIOGRAPHY

Following is a list of the principal references that were edited to get the penetration data that were analyzed.

(Approximately 4,60G rounds)

Projects TB4-150M and TB4-10

Ar 16981
16983
16984
16985
16988
16994
17050

Ar 17094
17131
17146
17154
17158
17163
17221

Ar 17186
17390
17845
18060
18223
18494
19106

Project TT1-5
Ar 17826
17838
17910
18073
18073sup
18107
18489

Ar 18490
18513
18513sup
18553
18553sup
18658
18703

Project AXZ3
Ar 20318

Ar 20730

Project TB3- 1224

0.

P.

.!

li'~~

558
586
679
686
834

Ar 19447
1q476
205e6

Ar 20895
(Approximately 200 rounds)

Ar 214Z1

2864, Effect of Hardness


AD
AD
AD
AD
AD

Ar 18752
19076
19182
19183
19187
19276
19366

(Approximately 530 rounds)

Ar 19945k

Ar 18504

AD
AD
AD
AD
AD

Ai
I}

818
992
1007
1041
1042

Lr
Ir

Lf 9k

(Approximately 3,400 rounds)


AD 1043
AD 1064
AD 1080

M I M 0

A-12581
1258
1U580
1s85
1260b

1 A Lk

1 01 6 T'

A- 12614
Ar 16231

T v

CONFIDENTIAL
r
It

Ar 20802
20Q4 9
21080
21203
21344

(Approximately 1,400 rounds)

Ar 17784
17791
17792
17794
17796
17798
17804

.1

Ar 18065
180 6 5sup
18084
18084sup
19675
19843
20701

I'

CONFIDENTIAL
A-Z
Miscellaneous Test Programs on R. H. Armor
AD 210
AD 369
AD 509

AD 54Z
AD 574
AD 652

AD 689
AD 830
AD 839

Miscellaneous Test Programs on C. H. Armor


AD
AD
AD
AD

517
560
571
587

AD
AD
AD
AD

590
630
658
663

AD
AD
AD
AD

678
685
694
697

Acceptance Tests of Armor Plate

1Z7G
127GI
1Z7K10-5
127L
127MI

i27MZ
127M3
1Z7M4
I 7M7
127M0- 1

Shot Design Projects


O.P. 5870
5757
5758
6135591

AD 844
AD 1014
AD 1033

AD 1084
Ar 13801

(Approximately Z,400 rounds)


AD
AD
AD
AD

836
990
999
101Z

AD
AD
Ar
Ar

1074
1076
15Z44
15Z56

(Approximately 2,400 rounds)

Records From These Library Books


IZ7C
127CZ
IZ7D
127DI
127DZ

(Approximately 1,400 round6)

BC 1141
lZ7Q
1Z7S
BG 114A
BC 1 14G
BG 114E

BG 163B
BG163G
BG174
BG4
G74A

(Approximately Z, 700 roumds,


Project TAI-I ZSI
TAI-1Z54
TAl-1301
TAI- 1302
TAl-1460
TAI-1503

Acceptance Tests of Shot

Some single reports:


ADP 194
P35543
ADP197
P39979
P25184
P41354
P34137
P56080
P34144
(Approximately 1,200 rounds)

Records From These Library Books


AIZ8
BC163D
BG16

BGZ1
BGZZ
tBGZ6

BG28
DAZ I
DA74

DA88
DA99
DA104

Other Firing Records


P53966 to P60803, 21 records on the HVAPDS, 76 mm, M331
PSZ5Z0 to P54821, 7 records on the AP, 90 mm, T33

CIT

DA128
EB41

I
CONFIDENTIAL

A-3 and A-4


Following is a list of the principal references that were edited, but that were not
used in the analysis.

Acceptance Tests of Armor Plate

(Approximately 800 rounds)

Records From These Library Books


127G
127CZ
127D

127DI
127D
IZ17G

1Z7M7
127GI
127M1
127M2

Acceptance Tests of Shot

I 7M4
127M5
127M6

IZ7Q
127S
A1Z8A

(Approximately 11,400 rountds)

Records From These Library Books


AN2
AN4
AN7
AN1Z
BC114B
BC I 14E

A103
AIZ8
AI Z8A

A136
AN
ANI

BC163
BC i 63A
FgC 163B
BC 163C
BG163D
BC 174

DA74
D.A 88
DA99
DA1Z8

07.

C7,4B

G95
C96
DA13
DAZ I

Other Firing Records


P52366 to P55575, 15 records on the AP, 90mm, T33

IFollowing

is a list of references which were not edited, but apparently contained


material that could have been used.

Shot Design Projects

(Approximately 700 rounds)

Project TAI- 1475,


TAI-1602,
TAI-500Z,
TAI-1302,

Report 1 and 6 other firing records


Reports I, Z, 8 and 5 other firing records
Report 3 and F. R. P-60401
Report 13

Acceptance Tests of Shot

(Approximately 800 rounds)

Records From These Library Books


BC-114-B, C, D, and E, 30 records on the AP, 75mm, M7Z and APC, 75mm,
M61, BC163 and BC163A, 27 records on the AP, 75mm, M7Z ind APC, 75mm,
M61

I.

Other Firing Records


P-46161 to P- 62446, 45 records on the AP, 120mm, T116

!
a01tA L

cAT?

'

T I ItUI

~CONFI~DENTIAL

ip

CONFIDENTIAL
DISTRIBUTION LIST

:o. of

1'

Oranization

OrganizaTion

Chief of Ordnance
Department of the Army
Washington 25, D.C.
ATTN: ORDTB - Bat Sec
ORDTW

Chief of Research and Development


Department of the Army
Washington 25, D.C.
ATTN: Dir/Developments Combat
Materiel Division

British Ioint Services Mission


1800 K Street. N. W.
Washington 6, D.C.
A^1 TN: Mr. John Izzard
Reports Officer

Commanding General
Rocket and Guided Missile Agency
Redstone Arsenal
Huntsville, Alabama

Commanding General
Picatinny Arsenal
Dover, New Jersey
ATTN: Samuel Feltman Ammo Labs.

Commanding General
Continental Army Command
Fort Monroe, Virginia
ATTN: Ass't Chief of Staff
for Development and Test

Commandant
Command and General Staff College
Fort Leavenworth, Kansas

President
U. S. Army Armor Board
Fort Knox, Kentucky

President
U. S. Army infantry Board
Fort Benning, Georgia

Commandant
The Armored School
Fort Knox, Kentucky
ATTN: Weapons Dept, Combat
Development Group

Commandant
The lr.fantry School
Fort Benning, Georgia

Canadian Army Staff


2450 Massachusetts Avenue
Washington 8, D.C.

Commandant, Marine Corps


Hq, U. S. Marine Corps
Washington 25, D.C.
Chief of Naval Operations
Department of the Navy
Wvashington 25, D.C.
Al TN: New Development And
Operational Eval Div.
Commandant
(T. S. Marine Corps School
Cuantico, Virginia
ATTN: Dir, Marine Corps
Development Center

No. of

Director
Armed Services Tech Info Agency
Documents Service Center
Knott Building
Dayton 2, Ohio
ATTN: DSC-SD
Director, Weapon Systems
EvaluAtioi Group
Office, Secretary of Defense
Washington 25, D.C.

CONFIDENTIAL

CONFIDENTIAL
DISTRIBUT,"N LIST

(Continued)

No. of
Copies

2,iuza1ion

Commanding Officer
Detroit Arsenal
Centerline, Michigan

Commanding Officer
Watertown Arsenal
Watertowr, Massachusetts
ATTN: OMRO

Commanding General
Frankford Arsenal
Philadelphia, Pennsylvania
ATTN: Fire Control Director
Commandant
Army War College
Carlisle Barracks, Pennsylvania
Commanding General
Ordnance Weapons Command
Rock Island, Illinois
ATTN: OPDOW-GEN
Mr. S. 0. Mulhollan
Commanding Officer
Watervliet Arsenal
Watervliet, New York
ATTN: R&D Division
Mr. Frank John

. r as Hopkins University
Operat:ons Research Office
U. S. Army
7100 Connecticut Avenue
Chevy Chase, Maryland
Washington 25, D.C.

Commanding Officer
Diamornd Ordnance Fuze Laboratories
Washington 25, D.C.
ATTN: ORDTL 06.33

Director
Ballistic Research Laboratories
Aberdeen Proving Ground
Maryland

CONFIDENTIAL