Critical Technology Events in the

Development of the Abrams Tank

Project Hindsight Revisited

Richard Chait, John Lyons, and Duncan Long

Center for Technology and National Security Policy

National Defense University

December 2005
 Form Approved
Report Documentation Page OMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it
does not display a currently valid OMB control number.

1. REPORT DATE 3. DATES COVERED

2. REPORT TYPE
DEC 2005 00-00-2005 to 00-00-2005
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Critical Technology Events in the Development of the Abrams Tank. 5b. GRANT NUMBER
Project Hindsight Revisited
5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

REPORT NUMBER
National Defense University,Center for Technology and National Security
Policy,300 5th Avenue SW, Fort Lesley J. McNair,Washington,DC,20319
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT

NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT

Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES

14. ABSTRACT

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 56
unclassified unclassified unclassified Report (SAR)

Standard Form 298 (Rev. 8-98)

Prescribed by ANSI Std Z39-18
The views expressed in this report are those of the authors and do not reflect the official
policy or position of the National Defense University, the Department of Defense, or the U.S.
Government. All information and sources for this paper were drawn from unclassified
materials.

Richard Chait is a Distinguished Research Professor at the Center for Technology and
National Security Policy (CTNSP), National Defense University. He was previously Chief
Scientist, Army Material Command, and Director, Army Research and Laboratory
Management. Dr. Chait received his Ph.D in Solid State Science from Syracuse University
and a B.S. degree from Rensselaer Polytechnic Institute.

John W. Lyons is a Distinguished Research Professor at CTNSP. He was previously director

of the Army Research Laboratory and director of the National Institute of Standards and
Technology. Dr. Lyons received his Ph.D from Washington University. He holds a B.A. from
Harvard.

Duncan Long is a Research Associate at CTNSP. He holds a Master of International Affairs

degree from the School of International and Public Affairs, Columbia University, and a B.A.
from Stanford University.

Acknowledgments. A project of this magnitude and scope could not have been conducted
without the involvement of many people. Their cooperation and willingness to recount
events that happened many years ago made this paper possible. The Army Science and
Technology (S&T) Executive, Dr. Thomas Killion, who requested this study, had the
foresight to apply a “Hindsight” approach to specific Army systems, which benefits not
only the Army technical and acquisition community but DOD and the other Services, as
well. On-site visits were the result of the coordinating efforts of Clay Miller; David
Hackbarth and Pearl Gendason; James Ratches; and Michael Audino for visits made to
Warren, MI; Aberdeen, MD; Fort Belvoir, VA; and Watervliet, NY (details of the visits
are given on p.10 of the text). Each and every person mentioned in Appendix A played an
important role by providing valuable technical information and commentary as well as
directing us to other contacts. Some, denoted by an asterisk, also reviewed sections in
their area of expertise for accuracy and completeness. Also contributing greatly were the
full text reviews provided by Robert Baer and Peter McVey (both former Abrams
Program Managers), Jerry Chapin, and Wayne Wheelock. Al Sciarretta, working with us
under a contractual arrangement, ably assisted us by providing in-depth document
reviews and research assistance, as well as providing relevant information based on his
field experience while on active duty.

Defense & Technology Papers are published by the National Defense University Center for Technology and
National Security Policy, Fort Lesley J. McNair, Washington, DC. CTNSP publications are available online
at http://www.ndu.edu/ctnsp/publications.html.

ii
Contents

I. Introduction ......................................................................................................1

II. Background .....................................................................................................3

Project Hindsight ............................................................................................................ 3
U.S. Tank Chronology .................................................................................................... 5

III. Study Methodology........................................................................................9

Scope............................................................................................................................... 9
Approach......................................................................................................................... 9

IV. Armament Related CTEs.............................................................................12

Main Gun ...................................................................................................................... 12
Gun Accuracy ............................................................................................................... 14
Penetrators..................................................................................................................... 15
Sabots............................................................................................................................ 18
Propellants..................................................................................................................... 21

V. Armor and Other Survivability Related CTEs.............................................22

Armor............................................................................................................................ 22
Crew Protection ............................................................................................................ 24
System Testing, Modeling, and Analysis ..................................................................... 27

VI. Engine and Drive System CTEs .................................................................29

Engine ........................................................................................................................... 29
Transmission ................................................................................................................. 32
Track and Suspension System ...................................................................................... 34

VII. Vetronics, C4ISR, and Fire-Control CTEs.................................................36

Vetronics ....................................................................................................................... 36
C4ISR............................................................................................................................ 37
Thermal Imagery........................................................................................................... 39
Fire-Control System and Related Sensors .................................................................... 41

VIII. Findings and Concluding Remarks .........................................................45

Findings......................................................................................................................... 45
Concluding Remarks..................................................................................................... 47

Appendix A ........................................................................................................48

Appendix B ........................................................................................................52

iii
I. Introduction

The urge to remain militarily strong has long been a driver of technological advancement.
This interplay between strength and technology, so evident in America’s global military
reach, has for decades prompted U.S. defense planners to engage in technology
forecasting. Analysis of emerging technologies was, and is, vital to making wise defense
investments. Among the preeminent examples of such analysis are the studies undertaken
by Theodore von Karman just after the Second World War. The von Karman reports1
represent an exhaustive review of science and technology related to the military services.
His analysis projected the importance of unmanned aircraft, advanced jet propulsion, all-
weather sensors, and target seeking missiles. A steady stream of other forecasts have
followed, such as “Strategic Technologies for the Army of the 21st Century”2 and “New
World Vistas—Air and Space Power for the 21st Century.”3

While it is important to assess the needs and challenges of the future, understanding past
military technological successes can be equally important to defense science and
technology (S&T) investment and management. To complement the above efforts and the
many other technology forecasts too numerous to mention, this study is the first in a
series that will examine some of the key factors that have led to meaningful technology
generation and ultimate incorporation into the U.S. weapons systems we see in the field
today. Included here are such factors as where the technical work was performed, funding
source(s) for the effort, collaboration between government and non-government
laboratories, and management style. This series of studies will focus only on Army
weapons systems, beginning with the mainstay of the Army’s armor force, the Abrams
tank. Analysis of other Army systems, such as the Apache helicopter and the Javelin and
Stinger missiles, will follow. The results of all studies will be compiled in a wrap-up
report that will focus on the implications of the findings for today’s S&T environment.

We begin the paper by briefly reviewing a project that served as a source of inspiration
for this study: Project Hindsight, a 1969 Defense Department (DOD) report.4 Hindsight
was an in-depth study sponsored by the Director of Defense Research and Engineering
(DDR&E) that provided some insights into the development of approximately 20
weapons systems across the DOD spectrum. Following the review of Hindsight, we
present a short history of U.S. battle tanks as well as a summary of events leading up to
the Army decision to replace the M60 Patton tank with the Abrams tank. This is followed
by a description of the methodology used to gather key data on the development of the

1
Theodore von Karman, Toward New Horizons (Washington, D.C.: United States Army Air Force, 1945).
2
Board on Army Science and Technology, Commission on Engineering and Technical Systems, National
Research Council. Strategic Technologies for the Army of the Twenty-First Century: Technology Forecast
Assessments (Washington, D.C.: National Academy Press, 1993).
3
United States Air Force Scientific Advisory Board, New World Vistas: Air and Space Power for the 21st
Century: Summary Volume (Washington, D.C.: Department of Defense, 1995).
4
Office of the Director of Defense Research and Engineering, Project Hindsight: Final Report
(Washington, D.C.: Department of Defense, 1969).
Abrams. The information is broken out by topic area (armament related subjects; armor
and other survivability related subjects; engine and drive system; vetronics, C4ISR and
fire control) and presented in terms of critical technology events (CTEs). CTEs are ideas,
concepts, models, and analyses, including key technical and managerial decisions, that
have had a major impact on the development of a specific weapons system. CTEs can
occur at any point in the system’s life cycle, from basic research, to advanced
development, to testing and evaluation, to product improvements. CTEs can even relate
to concepts that were developed but ultimately not incorporated into the weapons system.
Also, CTEs can originate anywhere, from in-house laboratories, to private industry, to
academia. The final portion of the paper presents the concluding remarks and findings
based on the CTEs that characterize the Abrams tank’s development.

The CTEs are noted in the left margin throughout the report. They are summarized in
Appendix B. CTEs are numbered only for ease of reference; there is no hierarchical or
chronological significance to their order.

While the link between high-tech weapons systems and battlefield success is often readily
apparent, the geneses of and processes associated with CTEs often are not. CTEs depend
on several important factors, including effective management, adequate funding,
establishment of clear priorities, fostering of proper technical competencies, and
leveraging of the resources of the private sector and academia. It is our hope that this
retrospective look at the Abrams tank can highlight the importance of these factors, and
thus can be of value to current S&T leadership within the Army and DOD as they wrestle
with tight budgets, a changing workforce, and new acquisition strategies.

2
II. Background
In this chapter we highlight some of the objectives and findings of the first Hindsight
report of 1969. We also present a brief chronology of U.S. tank development. Included in
the chronology is a look at the Abrams’ predecessor, the M60 Patton tank, and at the
requirements that were set for the design and development of the Abrams.

Project Hindsight

The study undertaken here is modeled in part on a 1969 report, Project Hindsight.5 In
1965, the DDR&E, Dr. Harold Brown, launched a project to take a retrospective look at
DOD investment in research and development (R&D), evaluate the results, and take stock
of lessons learned. Brown’s overarching objectives for the study were to identify
management factors that were associated with the utilization of the results produced by
the DOD S&T program and to devise a methodology to measure the return on
investment.6 He was motivated in part by the House Committee on Defense
Appropriations, which had questioned the efficiency of management and the overall
payoff for the part of the Research, Development, Testing and Evaluation (RDT&E)
program that pertained to S&T.7

The study was conducted by ad hoc teams of military and civilian in-house personnel.
Some 20 weapons systems were selected for review and a set of subcommittees was
arranged, one for each system. The systems selected for review included air-to-surface,
ballistic, and tactical missiles; a strategic transport aircraft; a howitzer; and an antitank
projectile. Data were gathered by questionnaire and evaluated according to the following
four criteria:8

1. The extent of dependence on recent advances in science or technology.

2. The proportion of any new technology that resulted from DOD financing of
science or technology.
3. The management or environmental factors that appear to correlate with high
utilization of S&T results.
4. A quantitative measure of the return on investment.

The project teams made the following findings with respect to these criteria: 9

1. Markedly improved weapons systems result from skillfully combining a

considerable number of scientific and technological advances (Criterion 1).

5
Ibid.
6
Harold Brown, Letter to the Assistant Secretary of the Army (R&D), the Assistant Secretary of the Navy
(R&D), and the Assistant Secretary of the Air Force (R&D), 6 July 1965 in Project Hindsight: Final
Report (Washington, D.C.: Department of Defense, 1969), 135.
7
Ibid.
8
Ibid., xiii.
9
Ibid, xxi.

3
 2. More than 85 percent of the new science or technology utilized was the result of
DOD-financed programs (Criterion 2).
3. The utilization factor appears insensitive to environmental or management
differences between industry, in-house laboratories, and university-associated
S&T centers (Criterion 3).
4. Most utilized new technological information was generated in the process of
solving problems identified in advanced or engineering development (Criterion
3).
5. Most utilized new fundamental scientific information came from organized
research programs undertaken in response to recognized problems (Criterion 3).
6. Technological inventiveness and the utilization rate are dependent on the
recognition of a need, an educated talent pool, capital resources, and an adequate
communication path to potential users (Criterion 3).
7. Any crude approximation in measuring cost-performance will tend to be delusory
(Criterion 4).

With regard to finding number seven, the study failed to find a satisfactory method for
assessing cost-benefit or cost-performance from S&T work. To illustrate the difficulty
that the study encountered, the report cited the example of the silicon-based integrated
circuit. The circuit, invented during the period under review, revolutionized electronics
and information technology and became a crucial part of virtually every system in the
arsenal; there was no effective way to subdivide the effects on individual S&T programs.

This paper will not attempt to redress this or any other shortcoming of Project Hindsight;
Dr. Brown’s goal of quantifying the payoff of DOD investment in research and
technology is if anything a loftier target today than it was in 1965. The fundamental
purpose of this report, however, closely mirrors that of its predecessor: by examining the
development of select Army systems, and in particular those signal technology events
that propelled these systems to success, we hope to shed light on the factors that lead
defense S&T research to fruition.

In addition to sharing a broad goal with the original Hindsight report, this paper also
takes from it a similar unit of analysis, the CTE. Hindsight evaluations were based on a
concept called a Research and Exploratory Development (RXD) Event. In the original
report, a RXD event has the predominant meaning of an event that “defines a scientific or
engineering activity during a relatively brief period of time that includes the conception
of a new idea and the initial demonstration of its feasibility.”10 There may be one or two
such events in the development of a component or system, or a whole string of such
events. In the case of basic research RXD events, the report distinguishes between
undirected (curiosity driven) and directed (problem driven) work. Lastly, the final
fabrication of the system component or device “may or may not involve an Event
depending on the state of the technological art at the time of fabrication.”11 Please note
that in our paper we use a definition for our signal events, CTEs, that differs from
Hindsight’s RXD event. Most significantly, as noted previously, CTEs can occur at any

10
Ibid., xiv.
11
Ibid.

4
point in the life cycle; we leave open the possibility that CTEs might result from efforts
that have utilized funds other than R&D.

U.S. Tank Chronology

The British Army first introduced the tank in combat in World War I, at the Battle of the
Somme in 1916. Tanks were used with varying degrees of success throughout the
remainder of the conflict. The tank’s place as a major factor in warfare was cemented in
World War II, when German panzer divisions swept across Europe in 1939–1940.
Though the Allies never built a tank as effective as those found in the German ranks, they
countered with numerical superiority. Early in the war the Army deployed the M5A1
Stuart light tank, some 8,800 of which were built and used in Africa and the Pacific
theater. The Sherman medium tank appeared in late 1942. Though it was significantly
overmatched by German heavy tanks, the Sherman was fast and reliable. Over 49,000 of
these tanks were built, and thousands were shipped to American allies. The Sherman was
in service into the 1950s.

As the Soviet Union emerged as the obvious adversary of the future, the Sherman was
succeeded by the M46–M47–M48 Patton series of tanks.12 The M48 Patton was in
service from 1952 to the 1970s. It was replaced as America’s primary tank by the M60
Patton, which incorporated most of the basic design elements of the M48. The M60
weighed 52 tons, could travel 30 miles per hour, and was equipped with a 105mm high-
velocity rifled gun. The armor was cast steel.

In the mid 1960s, the Army began to develop a main battle tank13 to replace the M60.14
This undertaking first took the shape of the MBT70 program, a joint venture with the
Federal Republic of Germany begun in 1963. This partnership was eventually terminated,
and the U.S. program was redesignated the XM803. The XM803 mounted a 152mm
gun/launcher combination capable of firing both conventional tank munitions and
missiles. The project was canceled by the Congress in December 1971 because of high
cost, but served, along with the last of the Patton series, the M60A3, as a significant
technical predecessor to the Abrams series.

The termination of the XM803 program left the M60 as America’s chief tank for the
foreseeable future. Yet, despite incremental improvements to the M60, the Army
remained convinced that it needed a new tank design. Their main motivation lay across

12
It is important to note that heavy tanks like the Patton series were not the only technological contenders
to anchor the U.S. armor force. Lighter combat vehicles, like the M41 Walker Bulldog Light Tank and the
M551 Armored Reconnaissance Airborne Assault Vehicle (ARAAV), also received programmatic support
(John Wiss, email to authors, 13 June 2005. All individuals cited in the footnotes appear in Appendix A,
which lists their Abrams-era organization and their current status.)
13
American tanks were not designated as “Main Battle Tank” until the advent of the MBT70. The M46,
M47, M48, and M60 tanks were designated as “Medium Tanks.”
14
An excellent source for historical information on the Abrams is R.A. Hunnicutt, Abrams: a History of the
American Main Battle Tank Volume 2 (Novato, CA: Presidio Press, 1990). Another useful reference is Rolf
Hilmes, Main Battle Tanks (London: Brassey’s Defence Publishers, 1987).

5
the intra-German border.15 In 1967, the Soviets, already ahead in quantity, fielded the
qualitatively superior T–64 tank. The T–64 had its faults, but it boasted a 115mm gun
(later upgraded to a 125mm) whose munitions could punch through the M60’s thickest
armor.

With the Soviet threat in mind, Congress authorized a new effort to develop a main battle
tank at the same time it canceled the XM803 program (see timeline of Abrams
development at the end of the chapter). The Army set up a task force at the Armor School
at Fort Knox, KY. With help from the Advanced Concepts Branch at Tank-Automotive
Command (TACOM), the task force identified 19 characteristics that a new tank should
possess. It listed them in order of importance, the first five being:16

• crew survivability;
• surveillance and target acquisition performance;
• high probability of hit with first round
• time to acquire and hit a target;
• cross-country mobility.

The task force also issued more specific criteria, such as a 25:1 horsepower (hp) to
weight (ton) ratio and a 46–52 ton gross weight.17 Furthermore, after the complexity and
cost concerns that had contributed to the termination of the MBT70 and XM803
programs, Congress required that unit cost be tightly controlled.18 An initial unit cost
ceiling of $400,000 (in 1972 dollars) was set; this figure rose to $507,790 (also in 1972
dollars) by the time requests for proposals were issued to industry.19 This figure was
$70,000 more than the estimated unit cost for the last M60 series tank, the M60A3, and
$100,000 less than the estimated cost of the cancelled XM803.20

Eight initial design contracts were given, later down-selected to two: General Motors
(GM) and Chrysler. The designs drew on advances made in the MBT70 and XM803
programs (for which GM was the prime U.S. contractor), and in the M60A3 (which was
built by Chrysler). Chrysler and GM also drew on in-house laboratory R&D on
components and design techniques, work that was not tied to any specific vehicle
program. The Chrysler design ultimately won the competition, and Chrysler21 was given
the contract to enter full scale engineering development of what became known as the M1
Abrams main battle tank.22 Production began in 1979. The Abrams went from

15
Orr Kelly, King of the Killing Zone (New York: W.W. Norton & Company, 1989), 19–23.
16
Ibid., 108.
17
The weight limit was later pushed to 57.5 tons.
18
R.A. Hunnicutt, Abrams: a History of the American Main Battle Tank Volume 2 (Novato, CA: Presidio
Press, 1990), 172.
19
Ibid.
20
Ibid.
21
The Chrysler tank division was purchased by General Dynamics in 1982. General Dynamics has
produced all versions of the Abrams from the M1A1 forward.
22
A point must be made about nomenclature. The tank did not become known as the Abrams until 1981,
when it was type-classified as the M1 Abrams (after General Creighton Abrams, who commanded a tank
battalion in World War Two and later became Army chief of staff). The tank’s first official designation was

6
Congressional mandate to the field in 8 years, a journey that often took 15 to 20 years for
other systems. Over 8,800 Abrams main battle tanks have been produced, primarily for
the Army but also for the Marine Corps and foreign nations.23 Abrams also have been
modified to serve as breaching vehicles, bridging vehicles, and mine-clearers. The United
States is buying no newly built Abrams, though portions of the existing fleet receive
periodic upgrades. The Abrams’ production timeline is provided below.

ABRAMS PRODUCTION TIMELINE

1971 The XM803 program is canceled.

1971 Congress authorizes a study at Fort Knox to develop a main battle tank. The
program is eventually designated XM1.
1972 The Fort Knox study team issues a report on proposed characteristics for the new
tank.
1973 Contracts are awarded to General Motors’ Detroit Diesel Allison Division and the
Defense Division of Chrysler Corporation to develop prototypes of the XM1.
1976 Chrysler’s design wins the competition and is selected to become the new main
battle tank.
1980 The first production of the M1 Abrams is completed. The M1 remains in
production until 1985.
1982 Chrysler sells its tank-building division to General Dynamics. All future Abrams
and Abrams upgrades are built by General Dynamics.
1984 The second Abrams model, the Improved Performance M1 (IPM1), is produced.
It remains in production until 1986. The IPM1 was produced to take advantage of
various improvements from the M1A1 program (know as the M1E1 program)
before the full M1A1 was ready for production.
1985 The third Abrams model, the M1A1, is produced. It remains in production until
1993. Among other upgrades, the M1 105mm gun is replaced with a 120mm gun.
1992 The first M1A2 tanks are produced. Existing M1 and M1A1 tanks are also
upgraded to the M1A2 configuration. The great majority of M1A2-configured
Abrams are upgraded versions of M1 and M1A1 tanks, but some are new-
production tanks. The M1A2 includes an independent thermal viewer for the
commander and an Intervehicular Information System (IVIS), among other
upgrades.
1999 The first M1A2 System Enhancement Package (SEP) tanks are delivered to the
Army. The M1A2 SEP has an embedded version of the Force XXI Battle
Command, Brigade and Below (FBCB2) command and control architecture,
improving the appliquéd version found in earlier Abrams.

actually XM815, quickly changed to XM1. The tank was known as XM1 throughout its development, until
it was type-classified. To improve readability, this report sometimes refers to the Abrams and the Abrams
program even in the context of pre–1981 events.
23
GlobalSecurity.org, “M1 Abrams Main Battle Tank.” Available online at:
<http://www.globalsecurity.org/military/systems/ground/m1-intro.htm>, accessed 9 November 2005.

7
In summary, the first production M1 Abrams weighed 60 tons and produced 1500hp,
giving it the 25:1 ratio that the Army required. It had a top speed of 45mph and a cruising
range of 275 miles. Also, per Army requirements, its survivability was much improved
over the M60. It had more effective armor and superior crew protection features. Like the
M60, early versions of the Abrams were equipped with a rifled 105mm gun that could
fire a variety of ammunition. It was soon replaced with a more effective 120mm
smoothbore gun. The Abrams has evolved through several successive upgrades, the latest
being the M1A2 SEP. After detailing the study methodology in the next chapter, chapters
IV–VII identify the CTEs that provided the capabilities that led to the Abrams’ battlefield
successes.

8
III. Study Methodology
Scope

We have chosen to focus this report on those things we deemed to be major technical
developments. The Abrams has hundreds of components that undoubtedly required some
innovation, and this study neither intends to cover every CTE in the course of developing
the tank nor to provide exhaustive technical detail on those CTEs that it does address.
The intent is to concentrate on major technical developments that relate to the Abrams’
core capabilities.

We have divided this report into four major topic areas: armament related subjects; armor
and survivability related subjects; engine and drive system; and vetronics, C4ISR, and
fire control. This separation of topics comes at the acknowledged price of diminished
discussion of integration; systems engineering achievements; and the teaming of in-house
laboratories, contractors, and the program manager (PM). The important integration work
performed by the contractor, working closely with the PM shop and in-house
laboratories, was vital to the final product. This fact is highlighted again in chapter VIII
and will be the subject of additional discussion in a summary paper when the other
reports in this series have been completed.

Approach

This report is based primarily on interviews and correspondence with people who were
directly involved in the development of the Abrams. Given the technical emphasis of the
report, we interviewed and corresponded with many technical professionals. We also
sought out personnel who had been at the PM office and with the contractors. The
objective of these communications was to obtain a picture of how the critical technology
events unfolded.

The interviews covered a broad range of pertinent topics, including the historical
background of the developments in question. The focus of discussion, though, was the
CTEs. We asked interviewees to identify those technology events that they considered
critical to the development of the Abrams; to detail the impact of the CTEs; to indicate
where the work in question was done; who contributed to it; who funded it; the nature of
the funding (e.g., 6.1, 6.2, or 6.3);24 the number of staff involved; and the management
factors that contributed to success.

24
DOD divides Research, Development, Test and Evaluation (RDT&E) spending into seven different
activity categories. Category 6.1 refers to the budget line item for Basic Research; 6.2 is for Applied
Research; 6.3 is for Advanced Technology Development. These three categories are referred to collectively
as S&T.

9
Often, we first interviewed a source and then obtained further information through
follow-on conversations and correspondence. Though we used a questionnaire to guide
some early discussions, we found it more fruitful to let the interviewees approach the
subject in their own way. Almost all of the discussions began with the interviewees
providing highlights of relevant experiences, after which we asked focused questions on
topics not initially covered.

It must be noted that the interviewees and correspondents were asked to relate events that
took place as many as 40 years ago. A few of these individuals are still in government
service but most are retired or active in the private sector. Detailed information was
sometimes unavailable. Precise data on funding levels, for instance, were obtainable only
intermittently. Wherever possible, we consulted multiple individuals on the same subject
and checked their accounts against written sources. When interviewees and
correspondents differed on what constituted a critical technology or who had made
essential contributions, we revisited the issue until we established the most accurate
possible picture of events. As a result, we are confident that we have captured the most
pertinent information related to the major technical events in the development of the
Abrams.

As noted, we made a concerted effort to contact individuals who had played key roles in
the Abrams tank development. Interviews were conducted both in person and by phone.
In total, approximately 60 individuals were contacted, of whom about 50 were able to
provide information in varying degrees. Some of these individuals were also asked to
review selected sections for accuracy and completeness. (Appendix A lists individuals
interviewed and their affiliations at the time of the Abrams program.) The following
discussions were held on-site:

• Aberdeen Proving Ground, MD, 3 November 2004;

• Detroit Arsenal, the Abrams PM office, and General Dynamics, Warren, MI, 22–
24 November 2004;
• Fort Belvoir, VA, 6 and18 January 2005;
• Watervliet Arsenal, Watervliet, NY, 26 July 2005.

Most of the individuals interviewed were employed by one of the following companies,
in-house laboratories, or program management offices involved in the Abrams program:

• Ballistic Research Laboratory (BRL) at Aberdeen Proving Ground (now the

Weapons and Materials Directorate of the Army Research Laboratory at
Aberdeen Proving Ground)
• Armaments Research, Development and Engineering Center at Picatinny Arsenal
• Tank-Automotive Command Research Development and Engineering Center at
Detroit Arsenal
• Benet Laboratories at Watervliet Arsenal (now part of the Armaments Research,
Development and Engineering Center)

10
• Night Vision Laboratory at Fort Belvoir (now part of the Communications and
Electronic Research, Development Engineering Center and known as the Night
Vision and Electronic Sensors Directorate)
• Chrysler Defense in Warren Michigan (now General Dynamics Land Systems)
• Ordnance Laboratory at Frankford Arsenal (now closed, with mission transferred
to the Armaments Research Development and Engineering Center at Picatinny
Arsenal)
• Program Executive Office, Ground Combat Systems (PEO GCS), located at the
Detroit Arsenal.

11
 IV. Armament Related CTEs
We start our discussion of critical technology events with armament, covering those
CTEs associated with the main gun and its ammunition.

Main Gun

When the MBT70 was being designed in the early 1970s, the U.S. Army initially wanted
to equip it with a high performance version of the light-weight M81 152mm rifled cannon
then used for the M551 Sheridan Armored Reconnaissance Airborne Assault Vehicle
(ARAAV). That gun could fire guided missiles as well as high explosive, antitank
(HEAT) rounds. The proposed gun for the MBT70 was increased in length and was
capable of high chamber pressures to enable fin-stabilized kinetic energy ammunition to
complement the missile and the spin-stabilized HEAT round.25 For the Abrams, however,
the notion of combining missiles and projectiles was dropped in favor of the 105mm M68
cannon, a proven weapon then used on most of the M60 Patton series of tanks.26 The
M68 was a 105mm rifled gun with a suite of ammunition that included kinetic energy
ammunition (discussed at length later); spin-stabilized live and training (inert) HEAT
rounds; high explosive plastic rounds; white phosphorus rounds; and anti-personnel
rounds.

The days of the M68 as the main gun were numbered, though. Even at the time it was
selected, the Army wanted a larger gun. Though the likely candidate, a 120mm
smoothbore gun under development in the United States and West Germany, was deemed
CTE 1 not ready at the time, the M1 designers were directed to provide a gun mount and turret
that could handle the larger and heavier weapon when it became available.27 This
forethought paid off in 1981, when the Army elected to adopt the 120mm smooth bore
cannon. The Army purchased the gun design and the know-how and equipment to
manufacture it from West Germany. Watervliet Arsenal installed a rotary forge, worked
out the remaining manufacturing difficulties, and started up production of the gun.
Progress at Watervliet Arsenal was bolstered by long experience with the manufacture of
large gun tubes.

By 1985, the new gun was installed on the M1A1 tank model. It is important to note that
this was a relatively rapid advance. The Army went from purchasing the gun design to
introducing it to the tank fleet in four years. This demonstrated the advantage of the
ongoing partnership with Germany. It also shows the importance of having an

25
Common projectiles are either spin- or fin-stabilized: spin-stabilized projectiles are gyroscopically stable,
while finned projectiles require slow roll to achieve precision. Finned projectiles can be fired from rifled
guns through the use of a special obturator that functions as a slipping clutch, while spin-stabilized
projectiles must be fired from a rifled gun. Fins are typically canted or otherwise trimmed to maintain or
provide roll when launched from a smooth-bore cannon.
26
The sole exception was the M60A2 vehicle, which featured the M81 152mm cannon used for the M551
Sheridan armored vehicle.
27
Hunnicutt, 190.

12
 experienced technical staff at the Benet Weapons Laboratory and the Watervliet Arsenal,
ready and able to transfer into the Army such a complex manufacturing process and make
it work.

Problems with premature gun tube failure in 175mm artillery guns in Vietnam had led to
more stringent requirements for both the materials and the manufacturing processes.28
Work at the Benet Laboratory improved the resistance of large caliber guns like the
CTE 2 120mm to unexpected fast fracture. Using linear elastic fracture mechanics to assess gun
tube failure, researchers developed an understanding of the fatigue and fracture behavior
of the tubes, especially in solving the issues associated with brittle fracture on the 175mm
cannon.29 This work developed a new fracture toughness test specimen from a thick-
walled gun barrel and a new test procedure that made use of the specimen.30 With these
developments, both resistance to fast fracture and the fatigue crack growth rate could be
measured using a cost-effective test methodology.

This design information was used in conjunction with a manufacturing process known as
autofrettage to increase gun tube life. In autofrettage (an existing process applied to large
gun tubes at Watervliet Arsenal), the tube is subjected to internal hydraulic pressure,
which causes plastic deformation up to halfway through the gun tube wall.31 The result is
a significant level of compressive residual stress that remains upon removal of the
hydraulic pressure. The residual compressive stress acts to lower the applied cyclic
tensile stress that occurs during service. Around 1970, Watervliet Arsenal developed
CTE 3 swage autofrettage, in which pressure applied by a mandrel replaces the hydraulic
pressure.32 This process, now used world-wide, was first used on the M68 105mm gun
and subsequently on all 120mm guns.

Additional work at Watervliet Arsenal involved electroplating, a process that had long
been used for decorative purposes and on smaller caliber weapons. Watervliet Arsenal
engineered this technology for depositing increased thickness chrome plating in the bore
of the 90mm gun of the M48 tank, which was fielded in 1953.33 This process, also
utilized for the chrome plating of the Abrams’ 120mm gun, resulted in significantly
increased service life due to decreased erosion caused by hot propellant gases.34

28
John Underwood, telephone interview with authors, 12 October 2004.
29
Ibid. Underwood, emails to authors, 12 January and 20 January, 2005.
30
J.H. Underwood and D.P. Kendall, “Fracture Toughness Testing Using the C-Shaped Specimen,” ASTM
Special Technical Publication 623, 1977.
31
T.E. Davidson et al., “Failure of a 175mm Cannon Tube and the Resolution of the Problem Using an
Autofrettage Design,” Case Studies in Fracture Mechanics, Watervliet Arsenal Report AMMRC MS 77/5,
June 1977.
32
Underwood, email 20 January 2005.
33
Michael Audino, interview with authors, Watervliet, NY, 26 July 2005, and Michael Audino, email to
authors, 30 August 2005.
34
Ibid.

13
 Gun Accuracy

Another major challenge was to improve gun accuracy. As the range of U.S. and Soviet
tank guns improved, it became clear that the battles of the future would be conducted at
much greater distances than the tank-on-tank combat of World War II. Errors in aiming
that were tolerable for close-in combat were unacceptable for long-range firing. A very
slight deviation in any of a great many variables would result in aiming error. Among the
possible sources of aiming error were imperfect sabots; erratic burning of the propellant;
flawed gun tubes, rangefinders, or gun sights; tube droop; wind or heat; and the cant of
the tank.

By the time of the M1 program, the Army had already made strides in confronting these
complex problems. In the 1960s, researchers at Frankford Arsenal put together an error
CTE 4 budget (an analysis of the sources and size of the errors) for tank guns.35 Their report was
the basis for early fire-control systems fabricated at the Arsenal. Fire control (discussed at
greater length in the “Fire Control and Related Sensors” section) was only part of the gun
accuracy equation, however. With rounds ejected from the gun tube at speeds in excess of
1.6 kilometers per second, small flaws in the tube or imperfections in the shell or the
sabot could change the trajectory of the round in a way for which no computer could
correct.

The first step in addressing these problems was to find a way to measure the effects.
Researchers at BRL were able to develop experimental, computational, analytical, and
statistical models that could determine the relative magnitude of each component of
CTE 5 error.36 They were then able to focus their attention on the most important variables and
apply vector analysis to factor together the errors from the several sources.

Armed with a comprehensive suite of tools to assess the critical errors, engineers were
able to address each technical challenge. The critical factor was the recognition that this
was a system with an interdependent collection of error sources. This resulted in the
development of a multi-disciplinary approach that created integrated high-level physics-
based models of the system. Analysis showed that lack of straightness of the gun barrel
was one of the key sources of error. Gravity and differential heating of the gun surface
due to sun exposure caused warping of the barrel. To compensate for the effect of
gravity, a constant factor was applied in the fire-control software. Drawing on a concept
used on the M60’s 105mm gun, General Dynamics designed, fabricated, and supplied a
thermal shroud for the 120mm gun to mitigate the bend in the tube caused by sun
exposure.37 A muzzle reference system (discussed in the Fire Control and Related
Sensors section) enabled the fire-control system to compensate for any remaining bend.
CTE 6 Also, a special machine press was designed at Watervliet Arsenal to address tube

35
Walter Hollis, interview with authors, Arlington, VA, 7 December 2004.
36
Peter Plostins, interview with authors, Aberdeen, MD, 3 November 2004.
37
Larry Rusch, email to authors, 1 June 2005.

14
 straightness and profile after manufacture and to correct tubes brought in from the fleet
for overhaul.38

This important work on gun accuracy continued over two decades. It involved many of
the best people at BRL, Picatinny Arsenal, and Watervliet Arsenal, who collaborated
closely with the PM office. Mission funds were provided in the Army budget in the
amount of $1 to 2 million per year and a matching amount of customer funds came from
the PM.39 Mission funding ended in the mid 1990s. Customer funding continues to this
day. Interviewees who worked in these organizations cited strong and patient
management support as critical to their ability to do the work. Laboratory management
also pressed for collaborations among all the parties.

Armament Enhancement Initiative (AEI): In 1985, AEI was launched in

hopes of gaining a “leap-ahead” advantage over America’s Soviet adversary
and to offset the huge numerical superiority the Soviet Union enjoyed during
the Cold War. The AEI technology program was led by the PM, Tank Main
Armament Systems (PM-TMAS). It was well-resourced, with expenditures
totaling at least $100 million per year. Successes include: a suite of advanced
kinetic energy rounds featuring improved penetrators, advanced propellant
charge technologies, and very lightweight sabots; and a multi-purpose sub-
caliber HEAT projectile with a discarding sabot and a proximity fuse to
address both ground targets and helicopters.a Many of the advances in sabot
technology, penetrator technology, and propulsion technology discussed in
the sections that follow were funded through AEI.

a. Al Horst, email to authors, 4 November 2004.

Penetrators

The Abrams can fire two primary types of munitions: chemical energy (HEAT) rounds
and kinetic energy (KE) long-rod penetrators. The latter rounds, which vary from 10:1 to
30:1 in length-to-diameter ratio,40 received perhaps the greatest attention from the
Abrams program.

In developing the 120mm gun, the Army made the key technical decision that it should
CTE 7 be optimized for long-rod penetrators.41 The bore of the 120mm gun offered a bigger
“engine” than the 105mm; rounds could utilize 50 percent more propellant to produce

38
Plostins.
39
Ibid.
40
Al Horst, Bruce Burns, and Brett Sorenson, interview with authors, Aberdeen, MD, 3 November 2004.
41
As mentioned in the section of the report on sabots, a long-rod penetrator was also developed for the
105mm gun.

15
 higher operating pressure so that more energy could be put into the round.42 Long-rod
penetrators were best suited to take advantage of this additional capacity, as their armor-
piercing ability comes largely from their very high KE—there is no explosive. Flight
stability for a long-rod penetrator comes primarily from the four to six fins at the rear of
the rod and also from some spinning—approximately 20 revolutions per second.43

Figure 1: A long-rod penetrator.44

The first long rods for the Abrams (also discussed in the Sabot section of this report)
were designed for the rifled 105mm gun. The AEI funded a larger research effort to
CTE 8 perfect a long-rod penetrator for the 120mm gun. This program, which involved years of
research in KE penetration mechanics and thousands of sub-scale experiments,45 played

42
Bruce Burns and William Drysdale, email to authors, 19 May 2005.
43
GlobalSecurity.org, “Large Caliber Ammunition—Types of Projectiles.” Available online at:
<http://www.globalsecurity.org/military/systems/munitions/bullets2-types.htm>, accessed 19 July 2005.
44
SPG Media PLC, “APFSDS Ammunition—Armoured [sic] Piercing Fin-Stabilised Discarding Sabot.”
Available online at: <http://www.army-technology.com/contractors/ammunition/apfsds.htm>, accessed 21
July 2005.
45
Randy Coates, email to authors, 13 November 2004.

16
 an important role in the development of one of the mainstays of the Abrams’ armament,
the M829 series of 120mm armor piercing fin-stabilized discarding sabot (APFSDS)
rounds.

A great deal of modeling was done to design the M829 series. For instance, early
CTE 9 modeling work was done at BRL on the structure of the penetrator.46 A central goal of
research on the structure was to craft a long rod with the highest possible ratios of length
to diameter and also with the highest possible density, two qualities that optimize armor-
piercing capability. Since the late 1950s, penetrators had been made primarily of
tungsten. Researchers did their best to respond to improving armor technology by
reworking the composition of the tungsten alloy used in rounds. In 1970, researchers built
a model using finite element analysis to determine the structural integrity of the bullet.
Staff also did systems analysis of interior, exterior, and terminal ballistics for the rods.
Based on this modeling, researchers determined the optimal materials for use in the
round.

Tungsten was replaced by depleted uranium (DU) in the mid 1970s. Researchers had
CTE 10 previously known of DU’s superior ballistic qualities, but it was not until improvements
in adversaries’ armor outstripped their ability to adjust the tungsten alloy that they made
the switch to DU. Whereas tungsten penetrators became blunt as they cleaved through
armor, crushing the tip into a mushroom shape and impeding the rod’s progress, DU has
a self-sharpening property.47 The tip of the DU penetrator shears such that it remains
sharp as it passes through armor. This substantially increases the range of any tank firing
DU rounds, as even with diminished velocity a DU round can defeat a target’s armor. In
the mid–1980s, Battelle Northwest Laboratory suggested a new process to improve the
CTE 11 compressive strength of the DU rod.48 The desired strength was obtained by a thermo-
mechanical process known as high-rate forming, which enabled the use of longer rods or
higher length-to-diameter ratios. The Oak Ridge Y12 plant helped out as well by
supplying tungsten and depleted uranium for the program.

The M829 series has been a valuable contributor to the Abrams’ success: it helped the
Abrams to knock out Soviet-made tanks in Iraq during Operation Desert Storm, often
with a single shot.49 Most of the R&D on the M829 series was done in-house, with some
modifications made by the sole manufacturer, Alliant. Collaboration among BRL,
Picatinny Arsenal, the PM shop, and the contractors was “at an all-time high.”50 The total
funding for the penetrator work from 1984–1988 was on the order of $30 million per
year.51 This includes funding for the associated manufacturing technology program.

46
Al Horst and Bruce Burns, email to authors, 8 November 2004.
47
William S. Andrews, “Depleted Uranium on the Battlefield,” Canadian Military Journal, Spring 2003,
43.
48
Bret Sorenson, interview with authors, Aberdeen, MD, 3 November 2004.
49
Picatinny Arsenal, “Tank Munitions Development.” Available online at:
<http://www.pica.army.mil/PicatinnyPublic/products_services/products12.asp>, accessed 19 July 2005.
50
Bruce Burns, email to authors, 8 November 2004.
51
Al Horst, Bruce Burns, and Brett Sorenson, interview with authors, Aberdeen, MD, 3 November 2004.

17
 The Abrams has continued to benefit from advanced research on its APFSDS
ammunition. The M829 round has been through successive generations; the most
advanced KE round fielded today is the M829A3. Unofficial estimates indicate that the
M829A3 has a maximum effective range of over 3,000 meters and can penetrate over
900mm of rolled homogeneous armor at 2,000 meters.52 The result of work on
penetration mechanics, the M829A3 is designed to mitigate the effect of reactive armor.53

Important Facilities: Research work on penetrators and sabots drew on

unique and vital Army resources. Supercomputers and, later, the Major
Shared Resource Facility (one of several high performance computing
facilities throughout DOD) at ARL-Aberdeen Proving Ground, enabled
CTE 12 ballisticians to run very complex models on penetrator-target interactions
that otherwise would have been impossible. This program, which provided
valuable insights on how to defeat various armors, is described by a
participant as “the most complex and thorough terminal ballistic evaluation”
ever performed at BRL/ARL.b Another facility, Experimental Facility 9 at
Aberdeen Proving Ground, was constructed to handle DU in bullets and
armor. Without this enclosed facility, full-scale firings with DU would have
been difficult if not impossible. It should also be noted that the BRL shops
supplied precisely machined parts and developed new techniques for making
sabots for experimentation.

b. Randy Coates, email to authors, 13 November 2004.

Sabots

Sabots are equally important to the design of a round. To achieve and maintain high
speeds for penetration, modern KE tank ammunition is normally composed of a narrow
long-rod penetrator surrounded by a sabot, which expands the diameter of the round to
the full barrel diameter of the gun. The sabot enables the pressure of the propellant gases
to push on a larger base and produce rapid acceleration of the round (see figure 2 below).
Once outside the barrel, the sabot, which is parasitic mass, falls off, leaving the high-
speed penetrator with a smaller cross-sectional area that reduces aerodynamic drag during
flight.

52
The Armor Site, “M1A1/2 Abrams.” Available online at:
<http://www.fprado.com/armorsite/abrams.htm>, accessed 21 July 2005.
53
Randy Coates, email to authors, 13 November 2004.

18
 Figure 2: Pressure builds behind the sabot and propels the projectile down the
gun tube.54

Sabots are essential to the use of long-rod penetrators, but it was difficult to adjust sabot
technology to pair an APFSDS round with the M1’s 105mm gun. The initial KE rounds
for the rifled 105mm gun did not have fins—they received all the spin they needed from
the rifling of the gun tube. The sabot that encased these munitions, dubbed Armor
Piercing Discarding Sabot (APDS) rounds, was designed with a copper ring that engaged
the rifling grooves. This sabot design was not suitable for longer rod penetrators like the
M829 round, which rely primarily on fins to create stabilizing spin, rather than the rifling
of the gun tube. By fully engaging the rifling of the gun tube, the APDS-type sabot would
impose too many revolutions per second on the APFSDS round and undermine the
stabilizing effect of its fins. In-house technical work was able to resolve this problem by
CTE 13 adapting an earlier design. A slipping rotating band, or “slipping obturator,” was used on
the 105mm HEAT rounds to prevent unwanted spin.55 The slipping obturator makes it
possible for the round to engage the rifling of the barrel while the round itself turns at
fewer revolutions per second than does the obturator. This concept was applied to the
APFSDS’s sabot.56

Another challenge for APFSDS rounds was to have the sabot separate from the round
cleanly after firing (see figure 3 below). Any strikes on or damage to the penetrator or its
fins due to an asymmetric lift off by the sabot could cause erratic flight and significantly
affect the probability of a hit at extended ranges. Comprehensive experimental techniques
developed at BRL to measure all launch accuracy errors were applied to the problem.57 A
CTE 14 tipping ring was designed for the rear of the sabot that pivoted the sabot segments so that
they were in a position to fly cleanly away from the rod.58 R&D work at Picatinny

54
Federation of American Scientists, “M829 120mm APFSDS-T.” Available
online at: <http://www.fas.org/man/dod-101/sys/land/m829a1.htm>, accessed 19 July 2005.
55
Renata Price, email to authors, 6 June 2005.
56
Ibid.
57
Bruce Burns, “An Executive Summary of the M829A2 Sabot Technology Program,” Army Research
Laboratory, ARL–TR–350, February 1994.
58
Bruce Burns and William Drysdale, email to authors, 19 May 2005.

19
 Figure 3: A representation of the final stage of sabot separation from the penetrator.59

Arsenal and BRL also produced a series of designs for the sabot focusing on exact shape
of the scoops or ramps. The final double ramp sabot shape used on the M829 was the
result of computer modeling by BRL.60

In addition to crafting the precise shape of sabots, Army researchers made advances on
their composition. Sabots were first made of steel, then of aluminum and magnesium
alloys, and today are made of composite materials. Staff at BRL used finite element stress
analysis to establish the proper geometric design of optimized, minimum-mass,
CTE 15 aluminum sabots for both the 120mm M829 and M829A1.61 They were configured,
developed, and produced under the AEI program. The technology for the new APFSDS
rounds was successfully transferred to Picatinny Arsenal and to industry. Several
interviewees commented on how effectively this was accomplished.62 A technical staff
member at BRL involved with this work was transferred to the PM office at Picatinny
Arsenal, carrying with him intimate knowledge of the BRL work on the program. He
worked at the PM office for about two years, and then moved to industry. Thus, the
technology was transferred by the individual as well as by written reports and periodic
visits by other staff members. The end result was an effective technology transfer and a
strengthening of the collaboration between BRL, Picatinny Arsenal, and industry. This
approach was devised by the BRL Director.

The next step was the incorporation of composite materials. Although the finite element
CTE 16 methods of the time were incapable of capturing the highly anisotropic nature of the
sabots, especially in predicting failure of these materials, careful experimentation defined
the lay-up of the fibers in the composite. Investigators at BRL—teamed with composite
material specialists at the Lawrence Livermore National Laboratory and with industry—
developed the means to design the architecture and processing for the first composite
sabot used for the M829A2 round.63 The AEI provided for a companion program that

59
GlobalSecurity.org, “120 mm Ammunition.” Available online at:
<http://www.globalsecurity.org/military/systems/munitions/120.htm>, accessed 19 July 2005.
60
Bruce Burns, email to authors, 13 November 2004, and Renata Price, interview with authors, Arlington,
VA, 8 December 2004.
61
Burns, ARL–TR–350.
62
Dick Vitali, phone interview with authors, 8 February 2005 and John Frasier and Renata Price, interview
with authors, Arlington, VA, 8 December 2004.
63
Burns, ARL–TR–350.

20
 established production facilities for the new composite sabot. The next variant, the
M829A3 round, was designed using a new thermoplastic matrix material in place of the
thermoset plastic used in the first versions. The Army-funded Center for Composite
Materials at the University of Delaware contributed a piece of processing equipment that
is in use at the manufacturer’s plant.

Propellants

Another crucial element of a KE tank munition, along with the penetrator and sabot, is
the propellant. Researchers strove to create higher density, more energy, and a proper
burn rate and burn progression.64

Among the most important aspects of developing better propellants was modeling. In the
CTE 17 1970s, a model of propellants known as NOVA (for new computer code) was developed
at the Naval Surface Weapons Center, Indian Head.65 One of the two key researchers on
this project transferred to BRL in 1977 to continue this work on Army problems.
Subsequent efforts at BRL resulted in an improved computer code called the XKTC (for
the express, kinetics, traveling charge version of NOVA). XKTC has been widely
distributed among America’s allies.66 Studies with a scanning electron microscope
revealed the micro-mechanisms of the behavior of the propellant in the M829 and other
120mm rounds.

These studies led BRL investigators to develop a new propellant design that could deliver
both more total charge mass and more efficient transfer of energy to the projectile.
Previously, gun propellant was typically manufactured in one of two ways. In the first
method, short cylinders with perforations provide needed increases in burning surface,
the length of the cylinders being short to allow venting of the perforations without
internal overpressures. The alternative was slotted, single-perforated sticks. The stick
configuration permits good flow of ignition gases in a tightly packed bundle of sticks and
the slot provides a vent for the single perforation. The problem with both designs was that
the burning surface did not increase with time, and pressure dropped off rapidly as the
CTE 18 projectile traveled along the tube. The BRL design incorporated partial transverse cuts at
regular longitudinal distances in the sticks instead of a lengthwise slot, thus allowing
multiple perforations in long, large-diameter sticks that both packed well and provided
highly progressive gas generation rates with time.67 This combination of features
provided the necessary increase in interior ballistic performance.

64
Al Horst and Bruce Burns, email to authors, 8 November 2004.
65
Ibid.
66
Ibid.
67
Al Horst, email to authors, 8 November 2004.

21
 V. Armor and Other Survivability Related CTEs
The Abrams’ survivability has been instrumental to its success. The importance of armor
developments and other improvements cannot be understated because, since the first
major use of tanks in World War II, the lethality of antitank munitions has increased
steadily. Israel’s tank losses in the 1973 Yom Kippur war, in the early days of the
Abrams program, underlined the importance of developing a well-protected tank.68

Armor

The M48 and earlier M60 tanks had a rounded shape and made use of large monolithic
armor castings with relatively few welded joints. To incorporate complex, non-
monolithic armors and increase design and manufacturing options, much of the Abrams
was fabricated with rolled homogeneous steel plates, known as rolled homogeneous
armor (RHA). The shape of the Abrams stemmed from several important concept studies
CTE 19 and analyses that TACOM, in conjunction with input from other in-house laboratories,
had authored in 1972 and 1973.69 Through the use of modeling and simulation at the U.S.
Army Tank-Automotive Research and Development Center (TARDEC), and other
system engineering tools, the designs of the M1 series tanks optimized the vehicle
silhouette, the location and type of armor, the location of the crew and vulnerable
components, and many other survivability factors.70 For example, as shown in table 1, the
front silhouette area of the M1A2 is 12.7 percent smaller than the M60A1. It is beneficial
to survivability to reduce tank vehicle height and present a harder target to the enemy.

Width (in) Height (in)

M60A1 145.00 129.17
M1A2 144.00 113.60

Table 1. Comparison of the dimensions of the M60 and the Abrams.71

While the new hull configuration optimized survivability, it demanded innovations in the
production process and made use of many more weldments than in the M60. Drawing on
extensive welding experience gained during WWII fabrication efforts, in-house engineers
from several laboratories and arsenals established joint design and welding techniques to
CTE 20 be used in fabricating the hull.72 Advances in welding equipment were incorporated into
the process to optimize quality control. R&D funding in the mid-to-late 1970s that

68
Hans Halberstadt and Erik Halberstadt, Abrams Company (Ramsbury, England: Crowood Press, 2000),
16.
69
Hunnicutt, 168.
70
Al Sciarretta, interview with authors, Washington, D.C., 20 May 2005.
71
Federation of American Scientists, “M60 Series Tank (Patton Series).” Available online at:
<http://www.fas.org/man/dod-101/sys/land/m60.htm>, accessed 19 July 2005 and Federation of American
Scientists, “M1 Abrams Main Battle Tank.” Available online at: <http://www.fas.org/man/dod-
101/sys/land/m1.htm>, accessed 19 July 2005.
72
Terry Higgins, telephone interview with authors, 7 February 2005.

22
 amounted to about $3–5 million supported this work.73 The results of these efforts were
transitioned directly to the Abrams production facilities in Michigan and Ohio. Industry
welding and quality control personnel benefited from the direct interface with Army
welding experts.74

Improvements in hull design and manufacture were matched by improvements in armor.

The response to more effective antitank munitions was, for many years, to add thicker
and thicker steel plates. But in the case of the Abrams, the conventional way of
improving armor was not feasible. The specified weight for the XM1 was 58 tons, and
the mobility specifications set the requirements for the power of the engine and drive
train based on this weight.75 For the protection levels desired, RHA alone was not a
feasible solution at a reasonable combat weight; concept studies early in the M1 program
established that lighter complex armors also would be required.76 Thus, the M1 program
started with the intent to use what the Army called “Special Armor.”

This special armor came out of technology developed through exchanges with the United
CTE 21 Kingdom and through U.S. indigenous advances. In U.S.-UK technology exchanges from
1965 to 1969, the U.S. Army was made aware of the British concept known as Chobham
armor. This work complemented parallel work in U.S. Army laboratories.77 During a
hiatus in the exchanges, the Army continued to develop these concepts and, after
demonstration by BRL, the technology was selected for incorporation into the XM1.
With the signing of a Memorandum of Agreement in late 1972, the United States and the
U.K. again engaged in armor technology discussions. When the validation contracts for
the XM1 were signed, a team of Army specialists and representatives of the two
contractors, Chrysler and GM, visited the U.K. to view their armor implementations.
Lessons learned during this visit were incorporated into the contractors’ vehicles.78 These
armor designs and implementation were unique to the M1. Throughout the development
phases, the contractors and BRL refined the armor configurations and implementations to
enhance vehicle survivability.

Details of the Abrams’ armor design and composition are classified, but this much can be
said: instead of using a single material—steel—the Chobham concept uses steel over one
CTE 22 or more layers of different materials, each layer designed to perform a different function
against incoming munitions. The armor is therefore a layered composite. The result is
that one can either have protection equivalent to using only steel at a much reduced
weight, or one can have much more protection at the same weight as a steel-only
configuration. A great deal of research was performed to perfect the design; later
improvements went far beyond the initial ideas.79 Developers made heavy use of
experiments and early computer models to develop ever more complex and effective
composite armor.
73
Ibid.
74
Terry Higgins, telephone interview with authors, 10 February 2005.
75
The weight of the M1A2 is approximately 70 tons.
76
Kelly, 115.
77
Tom Havel, email to authors, 19 May 2005.
78
Hunnicutt, 178.
79
Tom Havel, Walter Rowe, and John Runyan, interview with authors, Aberdeen, MD, 3 November 2004.

23
 Since the decision to incorporate the composite armor concept into the M1, there have
been further important armor advances. To improve protection against ever-increasing
threats, BRL researchers, with funding from the Abrams PM office, developed new
CTE 23 armor concepts, the most notable being one that incorporated DU.80 A team effort
involving BRL, the PM office, the Department of Energy (DOE), and General Dynamics
resulted in this new technology being incorporated into the Abrams’ turret armor.
Selected for its high density and special performance in high-shear fracture, DU makes an
ideal armor component. This upgrade was fielded on the M1A1 and M1A2 models. As
with the basic composite principles, the details of the upgraded armor are classified.

More recent advances have further bolstered the Abrams’ protection. BRL, building on
special armor technology, developed a new side armor concept that significantly
increased protection with minimal weight impact.81 This was handed off to the
manufacturer, General Dynamics, and is in the latest Abrams model, the M1A2SEP.

In the armor-antiarmor race, the Army laboratories have played and continue to play a
dominant role. All the work in the United States in this area has been funded by DOD,
and most of it has been done in-house. These new advances were developed primarily at
BRL, in close collaboration with other Army laboratories, DOE, and General
Dynamics.82 Funding was provided by the Army. BRL assigned roughly 12 technical
staff and technicians plus range crews—about 30 people in total.83 Staff at BRL—
continually seeking the optimal balance of weight, protection, and cost—continue their
work to this day. For instance, contingency armor kits have been developed and continue
to be advanced to provide users with options to further tailor the protection of the vehicle.

Crew Protection

The initial XM1 study team at Fort Knox named crew survivability as the number one
priority for the new main battle tank. A great many of the technological advances on the
Abrams discussed elsewhere in this report contribute to crew protection—greater speed
and agility for less exposure to enemy fire, a lower noise signature, and better armor, to
name a few. But while the developers of the Abrams strove to prevent the tank from
being fired upon, to prevent it from being hit if fired upon, and to stop the incoming
round if hit, they also took into account the possibility of penetration by an enemy round,
the risk of self-started fires, and nuclear, biological, and chemical (NBC) threats.

Improvements in crew compartment design made one of the biggest contributions to crew
protection. Ammunition stowage was a particular concern. The goal was to ensure that, if
armor was compromised, the stowage system was able to limit the damage to just a single
warhead. Ammunition was stowed openly in the turret in the Abrams’ predecessors and
the propellant in the ammunition was such that a hit on a round would produce

80
Ibid.
81
Ibid.
82
GM and Chrysler also made contributions in early design competition.
83
Havel, Rowe, and Runyan interview.

24
 deflagration within the turret. Propellant fires cannot be extinguished by current fire
suppression systems;84 water deluge is the extinguishing technique of choice for a
propellant fire, but systems have not been developed for vehicle use. The rounds for the
Abrams’ 120mm gun use highly flammable nitrocellulose cases (unlike the steel cases
used in the 105mm variant), putting a special premium on compartment design. A
combination of providing a separate compartment for the on-board ammunition,
controlling burning or explosion by propellant design and separation of the rounds, and
safely venting reaction products to the exterior of the vehicle were found to be necessary
for crew survivability.

Staff at BRL designed a separate compartment to stow ammunition.85 The compartment

CTE 24 makes use of automatic doors to isolate the ammunition from the crew while still
allowing the loader to easily access the rounds for firing. There was considerable effort
on perfecting the door seals using monitors and sensors to measure air quality in the crew
compartment during ballistic tests.86 The approach taken was later used in developing the
protocols for ballistic evaluation of full scale testing of vehicles and structures. The
compartment design provided sufficient venting for any explosion by installing blow-out
panels that would direct energy from the blast away from the crew.

Researchers at BRL also addressed ammunition sensitivity and warhead shielding.

Controlling the interaction of stowed high explosive warheads was critically important to
CTE 25 an ammunition compartment concept. To this end, BRL developed less shock and crush
sensitive warheads and included plastic shields between stowed rounds.87 The shields
served to reduce the velocity of the detonation wave to manageable proportions.
Researchers also developed a new test rig in an effort to study the reaction of munitions
CTE 26 to impact.88 The test fixture, a 3 ton pendulum referred to as the “BRL Ballistic
Pendulum,” quantifies the response of ammunition components to impact, thereby
providing guidance for compartment design.

R&D on crew compartment and munitions design enjoyed stable funding. Increased
money for restructuring the crew compartment was obtained from the PM through the
AEI. It must be noted that without the preexisting 6.1 and 6.2 research that had created
broad expertise in propellant technology at BRL, these developments would have come
much more slowly.89 Crew protection efforts benefited from sympathetic and patient
management, a strong support staff, and close collaboration among many participants,
including BRL, the Navy, Picatinny Arsenal (for the manufacturing technology), other
parts of DOD, DOE laboratories, and contractors Aerojet, Honeywell, and AAI.

84
The Halon extinguishment system cannot put out a propellant fire, since the agent relies on depriving the
fire of sufficient oxygen and the propellant carries its own oxidizer. Halon is effective against petroleum-
based fires.
85
Jerry Watson, Gould Gibbons, and Pat Baker, interview with authors, Aberdeen, MD, 3 November 2004.
86
Terry Dean, fax to authors, 1 August 2005.
87
Watson, Gibbons, and Baker interview.
88
Ibid.
89
Ibid.

25
 Another contribution to crew protection was the development of combustible casings for
the tank rounds. In earlier tanks, there was a risk of fires started by the inadvertent
ejection of burning debris from the tank gun breach into the turret, which had additional
combustible casing rounds and spilled hydraulic fluid on the floor. The Abrams program
sought to avoid any such issues. To minimize accumulation of empty casings on the floor
of the tank compartment and to reduce fire hazards, the Army had developed casings that
were supposed to be consumed after firing in the breech, leaving only the metal base
CTE 27 plate to be ejected. When the 120mm gun technology was purchased from Germany, a
design was included for such combustible sidewalls in the casings. However, Army tests
revealed some deficiencies, such as incomplete combustion, low strength, and trouble
with surface coatings.90 A team from Picatinny Arsenal and two contractors—Honeywell
and Armtec Defense Products (now Esterline Armtec)—resolved these difficulties.91

The Abrams’ design also drew on infrared sensor technology to protect the crew. Sensors
in the Abrams’ crew compartment can detect a fire in a few milliseconds and (if it is a
CTE 28 petroleum fire) extinguish it within tenths of a second using Halon agent. A similar
system is installed in the engine compartment with provision for a manual second shot of
Halon if a flare-up occurs.92 This upgraded fire suppression capability grew out of an
R&D effort that was underway at TACOM in the late 1960s as a result of experiences in
Vietnam.93 The TACOM program funded efforts in the private sector and academia that
significantly advanced the state of the art in key areas such as sensor technology and flow
of the suppression agent. Total cost for fire suppression R&D, including testing, was
approximately $1 million per year over a 10-year period.94

Based on analysis of the 1973 Arab-Israeli war, and conscious that the Soviet adversary
was armed with weapons of mass destruction, the Army pursued another area of crew
protection for the Abrams: defense against NBC attack.95 In 1977, Congress requested
that the Army incorporate a system that would provide decontaminated air to the crew
compartments of armored vehicles.96 Existing tanks, like the M60, then relied on a
ventilated facepiece system (VPS) that delivered filtered air to a protective mask. The
Army’s Edgewood Chemical Biological Center (ECBC), which had developed the VPS
technology, was tasked with analyzing the feasibility of a hybrid collective protection
system that would make use of both individual gear and compartment-wide filtration and
overpressure. ECBC research indicated substantial benefits to a hybrid system, and,
among other NBC-related work, the lab determined the necessary airflow rate and
overpressure level.97 In 1980, ECBC awarded a contract to develop Hybrid Collective
Protection Equipment (HCPE). This HCPE effort included the development of the NBC
CTE 29 filter currently being used on the Abrams. This system was not ready in time for the M1,

90
Renata Price, email to authors, 7 July 2005.
91
Ibid.
92
Terry Dean, fax to authors, 1 August 2005.
93
Steve McCormack, phone interview with authors, 1 February 2005.
94
Ibid.
95
James Baker, email to authors, 7 July 2005.
96
Ibid.
97
Ibid.

26
 which relied instead on a VPS-only system, but it has been installed on subsequent
versions of the Abrams.

These technical developments provided the crew with much-improved protection. Should
a round penetrate the turret’s ammunition compartment, it is likely that only one or two
rounds will be set off and the explosion will be vented harmlessly to the outside.
Similarly, measures taken to lessen the risk of fires started by munition casings and
diminish the threat of an NBC attack greatly enhanced crew safety.

System Testing, Modeling, and Analysis

As part of the development process, DOD required full scale testing under controlled
conditions (developmental testing) and then testing under field conditions with soldiers
using the equipment (operational testing). Tests of armaments include firing for accuracy,
lethality, speed, and range. Testing of armor requires live-fire tests in which selected
portions of the complete tank are exposed to the fire of specified munitions.

Researchers did modeling and testing work on the Abrams in response to a full range of
threats, including nuclear blast and hacking of the on-board computers by an adversary.
The most prominent work, though, was done on the most prominent threat: impact from
an antitank munition.

BRL began developing models and computer codes to predict the vulnerability of combat
CTE 30 vehicles (both aircraft and ground vehicles) shortly after WWII. Until 1984, these models
were deterministic, meaning that they could not account for the stochastic nature of the
interaction between an attacking munition and the target vehicle. Thus, for any specific
engagement condition, only one prediction of the outcome of the ballistic interaction
could be made. There was no way to estimate the effect of variability on a projectile’s
ability to defeat armor, the yaw or pitch of an attacking munition at the moment it strikes
the target vehicle, the size and nature of the damage inflicted on internal components, and
myriad other phenomena.

Beginning in 1984, BRL developed computer models to predict the outcome of specific
CTE 31 live-fire test shots.98 When adequate input data are available, these new models account
for the highly stochastic nature of ballistic damage. Consequently, analysts can now
model the variability of results observed in tests or even those experienced in combat by
generating a statistical distribution of possible outcomes of a particular test event or
combat engagement. When data to characterize the variability of the encounter conditions
are not available, the new models are used to provide deterministic estimates of the
system’s vulnerability.99 The models take into account the shot trajectory, location of the
impact on the target (including angle of attack), the physics of the penetration of the
armor, and characteristics of the behind-armor debris (pieces of armor and pieces of the

98
John Beilfuss, interview with authors, Aberdeen, MD, 3 November 2004.
99
John Beilfuss, email to authors, 28 June 2005.

27
penetrator after it has perforated the armor).100 The behind-armor debris are characterized
by the spatial distribution, mass, and velocity of all fragments. The model predicts which
components or crew members will be hit by the behind-armor fragments, resulting
damage or injury, and consequences of that damage or injury in terms of the capability of
the vehicle to perform its intended mission. The models produce maps of target
vulnerability viewed from any direction around or above the vehicle. The models also can
handle indirect fire from a munition that detonates outside the target, when fragments and
blast constitute the threat. Algorithms in these models and input data needed to run them
are based on ballistic and controlled damage tests and experiments conducted at
Aberdeen Proving Ground and elsewhere.101

Over the years, these models have evolved. The predominant model now used at ARL is
called MUVES–S2,102 which is actually a computing environment containing
approximation methods designed for specific attacking munition/target vehicle
combinations, including aircraft and antiair munitions.

Computer modeling yields great cost savings for testing armor and armaments. If
vulnerability studies had to be done solely by field experimentation, performance testing
would involve many rounds, major repairs between shots, and the sacrifice of several
targets, such as tanks, personnel carriers, and helicopters. The cost of each live-fire shot
is substantial: $50,000 to $100,000 is not uncommon.103 With modeling, the number of
live shots is considerably reduced, because they are used only to investigate specific
vulnerability or lethality issues and to validate the analysis model.104 Computer modeling
permits hundreds, if not thousands, of simulations to be run quickly and enables the
investigation of many more engagement conditions than could be considered in even the
most extensive test program.

The computer modeling work on the Abrams was done in-house by the people at ARL
using 6.2 and 6.6 mission funds and customer funds from the Abrams PM.105 The
development of the models was done primarily in the ARL Survivability and Lethality
Analysis Directorate, which receives 6.2 and 6.6 mission funding. At its highest level of
effort, this modeling work on the Abrams, including model development, involved about
10–15 people at ARL.106 The level of effort has dropped off and currently involves only
one or two people at ARL.

100
Behind-armor fragmentation can take place even if a round doesn’t penetrate the exterior armor: the
force of impact on the hull causes the softer metal inside the tank to burst apart, with dire results for the
crew and sensitive instruments and ammunition. Indeed, some antitank weapons are designed to have this
effect.
101
John Beilfuss, email to authors, 28 June 2005.
102
Ibid.
103
Ibid.
104
Nonetheless, the testing and validation of the Abrams required many live fire tests. One participant
recalls that “over eighty ballistic tests have been conducted on full-up and fully operational vehicles. Large
caliber tests on full scale structures and ammunition compartments are in excess of 800.” Terry Dean, fax
to authors, 1 August 2005.
105
This is the case where funds beyond the usual categories are treated as S&T for management purposes.
106
John Beilfuss, interview with authors, Aberdeen, MD, 3 November 2004.

28
VI. Engine and Drive System CTEs
We now turn from armament and armor to other equally important parts of the lethality-
survivability equation. This chapter will discuss vehicle mobility-related CTEs.
Significant increases in engine power, transmission capabilities, and the quality of the
suspension system have given the Abrams greater dash capability than the M60,
especially when maneuvering uphill or across soft soil. This maneuverability reduces
exposure to enemy fire and contributes to the Abrams’ superior survivability (see table 2
for significant survivability-related mobility differences between the Abrams and M60
tanks).

Engine

The Fort Knox study team required that, whatever the weight of the Abrams, the tank
have a 25:1 hp-to-ton ratio.107 As the weight of the M1 grew along with its armor and
armament capability, its designers found themselves increasingly in need of a more
powerful engine.

The fundamental choice was between a diesel engine and a gas turbine engine. There
were pluses and minuses to abandoning the diesel engine that was then standard in
ground vehicles. Mostly because of more efficient fuel consumption, a diesel engine was
better suited for situations characterized by idle time and/or low, steady speeds—the kind
of operation associated with a defensive posture or overwatch role. The turbine engine,
which had a higher initial cost, was considered better for offensive and highly mobile
defensive operations, where wide-open acceleration is needed. Also, the turbine engine
worked better in cold weather, weighed less, and took up less space than the diesel
engine. It could run on multiple types of fuel, and though it had a significant heat
signature, the turbine engine produced less smoke and noise than a diesel engine.

By some measures, the most obvious choice for the Abrams was the AVCR–1360 diesel
engine. GM, the prime contractor on the XM803 program, had used this engine in its
design for that tank. GM was also the prime U.S. contractor on the joint U.S./FRG
MBT70 program, and had incorporated a predecessor of the AVCR–1360, the AVCR
1100, into that design. Built by Continental Motors, the AVCR–1360 (Air-cooled,
Variable Compression Ratio, 1,360 cubic inches) was a highly capable engine. It was a
120º Vee, 12–cylinder, with turbo charging, supercharging, and intercooling, using
variable compression ratio pistons. The version prepared for the XM1 was rated at 1,500
gross horse power (GHP). This leap forward in performance was well over double the
horsepower of its closest predecessor but raised some doubts about its long-term
durability.108 It did, however, have a long track record: development of the AVCR–1100
began back in 1957; by 1965 the engine had accumulated 1,356 hours of test operation,

107
Kelly, 98.
108
John Wiss, email to authors, 13 June 2005.

29
 including 4,425 miles of vehicle test rig testing.109 Further, through their inclusion in two
major (though ultimately unsuccessful) tank programs, both GM and the Army had
considerable developmental experience with the AVCR 1100 and 1360.

Despite the attraction of the diesel, a gas turbine engine was chosen: the AGT1500 gas
CTE 32 turbine engine was part of the Chrysler design that won the XM1 competition. The roots
of the AGT1500 lie in a TACOM effort that predates the Abrams program. Gas turbines
had been investigated as potential combat vehicle propulsion systems ever since their
development and installation in aircraft during the latter portion of World War II. Aircraft
experience had demonstrated improved maintainability compared to piston engines.
CTE 33 About 1960, TACOM engineers initiated an R&D program in gas turbine engine
technology for ground vehicle applications.110 The entire multi-year (mid–1960s to mid–
1980s) turbine engine developmental program was funded to a level of about $300
million.111 This program first investigated the potential of turbine engines, funding the
development of three prototype engines using three different technological approaches.112
After technical analysis, TACOM set forth demanding performance specifications for an
advanced gas turbine engine.113 Initial performance requirements specified that the
engine produce 1,500 GHP on a 125ºF day and demonstrate a best brake specific fuel
consumption of 0.38 (pounds per hp-hour) at 80 percent engine power at that elevated
temperature. Cold-day performance at 2,000 hp was required to assure adequate power
for subsequent upgrades.

The AGT1500 and AVCR1350 programs ensured that GM and Chrysler could choose
between a diesel and a turbine engine as they submitted advanced development and full
scale engineering development proposals and prototypes for the XM1 competition. Based
most likely on the availability of developmental data and analysis, as well as the inherent
advantages noted above, the Chrysler team selected the gas turbine engine as the
propulsion system for its submissions, even though some in the Army favored the diesel
engine option.114 Ultimately, though, both contractors included turbines in their designs.
In July 1976, just before DOD was to award the M1 contract to GM or Chrysler,
Pentagon leaders decided to extend the competition and compel both companies to
submit designs that provided for the use of either a diesel or a turbine engine. There were
several reasons for this delay, among the most important being the preference of some
senior civilian leaders for the turbine.115 DOD announced that Chrysler, which had used
the time extension to significantly lower its design’s projected unit cost, would receive
the M1 contract, and the gas turbine engine became the power plant for the Abrams.

There were, however, several aspects of the turbine engine that still needed to be
addressed. An important concern was air filtration. Designers had to find a way to
provide sufficient air cleaner volume to allow the vehicle to operate for a reasonable
109
Leroy Johnson, email to authors, 10 May 2005.
110
Charles Raffa, written communication to authors, 23 November 2004.
111
Ibid.
112
Wiss.
113
Johnson.
114
Jim Mesko, Abrams in Action (Carrollton, TX: Squadron/Signal Publications, 1989), 5.
115
Kelly, 143-158.

30
 period without cleaning the air filters. The M1 was fielded with a two-stage filter.116 A
CTE 34 pre-cleaner prevented large particles from entering the intake. The rest of the particles
were trapped by a filter called a V-Pack. When the filter became clogged an indicator
light in the tank would alert the crew, who would have to remove the filter and clean it
manually. Improvements were continually sought, culminating in the upgraded M1A2
filtration system. This system, developed and supplied by Donaldson Company,117 uses a
pulse jet air cleaner that removes the need for the crew to manually clean a filter.118
Between $10 and $20 million of government funds was spent in developing the pulse jet
air cleaner.119

There was also a need to address fuel efficiency—an idling turbine engine could consume
over 10 gallons of fuel an hour.120 Several engineering approaches came forward to deal
with the issue. As part of the TACOM test program, the gas turbine engine had a
recuperator, a well-established means of improving fuel consumption by using exhaust
gases to preheat air entering the engine's combustor. Also, there were efforts to explore
the use of an auxiliary power unit (APU) that would provide electrical power when the
vehicle was stationary (e.g., in a defensive posture) for long periods of time, thus
enabling the engine to be shut down to reduce wear and fuel consumption. APUs had
been considered for previous U.S. tanks and had been used by some British tanks.
Evaluations of candidate external diesel and gasoline APU units obtained from industrial
suppliers were conducted.121 The final evaluation included one bustle-mounted diesel
APU, one bustle-mounted gasoline APU, and a fender mounted diesel APU. Based on
this evaluation, a bustle-mounted, diesel APU was selected for production in 1991 in
support of Operation Desert Shield.122 Today most active Army M1A1s have the bustle
mounted APU.123 This APU was suited to peacetime operations, but it was not the
optimal solution for combat because the crew had to get in and out of the tank to turn it
on, shut it off, and fill it with fuel, and because it took up storage space in the bustle and
used different fuel.124 The Army recognizes the problems posed by an external APU and
while awaiting an under-armor APU has added batteries to the M1A2 SEP to run the
vehicle’s electronics without turning on the engine.125
116
Irv Smith, telephone conversation with authors, 29 September 2005.
117
Herbert Dobbs, interview with authors, Arlington, VA, 7 December 2004.
118
Smith.
119
Ibid.
120
GlobalSecurity.org, “M1 Abrams Main Battle Tank.” Available online at:
<http://www.globalsecurity.org/military/systems/ground/m1-specs.htm>, accessed 19 July 2005.
121
The evaluation was done by a combination of PM-Abrams (TACOM), Combat Systems Test Activity
(CSTA) at Aberdeen Proving Ground, ARL’s Harry Diamond Lab (HDL) in Adelphi, MD, and the Army
Materiel Command’s Field Assistance in Science and Technology (FAST). Al Sciarretta, email to authors,
14 July 2005.
122
James J. Chopack et al. Development and Testing of Auxiliary Power Units for the M1A1 Tank System
(Adelphi, MD: U.S. Army Laboratory Command, Harry Diamond Laboratories, 1991), 44.
123
Irv Smith, telephone conversation with authors, 30 November 2005.
124
James J. Chopack et al. Development and Testing of Auxiliary Power Units for the M1A1 Tank System
(Adelphi, MD: U.S. Army Laboratory Command, Harry Diamond Laboratories, 1991), 31–32.
125
Roxana Tiron, “Army Future Force: Abrams Tank Still Far From Retirement,” National Defense
Magazine, October 2004. Available online at:
<http://www.nationaldefensemagazine.org/issues/2004/oct/Abrams%20_Tank.htm>, accessed 30
November 2005.

31
 Transmission

During the same period that the diesel and turbine engine candidates were maturing,
TACOM issued a contract to Allison Transmission, a division of GM, to develop a
CTE 35 hydromechanical transmission. Allison, which had long made transmissions for Army
ground vehicles, was directed to design a transmission that could be used with either of
the two engine candidates. Proceeding under TACOM technical guidance, it created a
hybrid hydrostatic-mechanical transmission, designated the XHM1500 (Cross Drive
HydroMechanical, 1,500 HP). While the final version of this transmission, the –2 series,
functioned well, and the hydrostatics performed very well, the hydraulic control
technology available at the time did not have the flexibility and capability to optimally
control the transmission.126 Furthermore, improved fuel economy—one of the primary
reasons for choosing a hydromechanical transmission—was not shown in development
tests. Allison ultimately recommended that the hydromechanical technology not be
pursued for the XM1 effort.

Based on the Allison recommendation not to pursue the hydromechanical technology

approach, TACOM contracted Allison to re-optimize an existing Allison transmission,
the X700 (Cross Drive, 700 HP), as the X1100 (Cross Drive, 1100 Net HP). These
contracts, totaling about $10 million, covered the redesign, fabrication, and
demonstration/validation of the X1100 transmission for the XM1.127

TACOM laid out several challenging requirements for the X1100.128 First, in case the
CTE 36 XM1 program did not succeed, it had to fit in the M60. Second, it had to be adaptable to
three different engines: the M60’s AVDS1790 diesel engine (again, in case the XM1
program failed), as well as the AGT1500 gas turbine engine, and the AVCR1360 diesel
engine (because the Army did not yet know which engine the XM1 would use). To meet
these multiple requirements, Allison developed a transmission based on the Allison on-
highway commercial transmission design. The new transmission, the X1100 is a 35-
cubic-inch, radial displacement, hydrostatic steering mechanism that remains unique in
the world today.129

The X1100 represented a major departure from the CD850 transmission utilized in the
Patton series tanks.130 It featured an automatic-shifting, 4–speed forward/2–speed reverse
propulsion scheme; the CD850 had a 2–speed forward/1–speed reverse powershift. The
CTE 37 X1100 also incorporated power-assisted service brakes, which allowed maximum-effort
brake stops with minimal brake pedal force; the CD850 required considerable force to
achieve maximum braking. The above features, combined with hydrostatic speed-

126
Johnson.
127
Charles Raffa, email to authors, 31 January 2005.
128
Johnson.
129
Ibid.
130
Ibid.

32
controlled power steering, gave the X1100 performance and drivability never before
available in a U.S. tank.131

Further development of the X1100 conducted as part of the M1 Reliability and

Maintainability/Durability and M1A1 development programs resulted in additional
technology events worthy of note. Initial production X1100 transmissions experienced
limited brake life and occasionally required replacement of the transmission brake plates.
Continued development of the X1100 multi-plate, wet disk brake system has resulted in
an extremely durable, reliable, high-performance brake system. Subsequent studies have
validated the X1100 approach as the safest, most cost-effective system for braking a
track-laying combat vehicle.132

It should be noted that prior to the start of the X1100 development effort, significant
efforts had been expended to gather duty-cycle transmission performance data on a wide
variety of actual test courses. The analog tapes of data from these courses were
subsequently digitized and became the basis for laboratory durability development of the

M60A1 M1A2
Weight, Combat 52.5 tons 68.7 tons
Loaded
Engine 750 hp Continental Textron-Lycoming
AVDS–1790 V–12 AGT–1500 multi-fuel
diesel gas turbine with
recuperator

GHP (max) 750 1,500

HP per ton 13.1 21.8
Maximum level road 30 mph 42 mph (governed)
speed
Acceleration time 0 15 sec 7.2 sec
to 20 mph
Cruising Range 300 miles 298 miles
Fuel tank 375 gal 504 gal
Transmission Detroit Diesel Allison Allison X–1100–3B
CD850 series, 2 forward Automatic hydro-kinetic,
and 1 reverse gears hydrostatic steering, 4
forward and 2 reverse
gears
Drive Rear Rear

Table 2. The engine and power train of the M60 compared to the M1A2.

131
Ibid.
132
Ibid.

33
 X1100 and all subsequent Allison military transmission products. The test regimen
proved to be an extremely valuable tool and time yielded significant schedule and cost
savings during transmission development.133

Track and Suspension System

While the gas turbine engine remains the propulsion system in all the Abrams tank
variants, there were marked changes in some of the other drive system components.
Engineers recognized that with the increased speed and weight of the Abrams, the
suspension system needed attention, especially if the tank was to perform well on rough
terrain.

The original M1 tank suspension system was developed by Chrysler for a 58 ton
vehicle.134 The production M1A1 tank resulted in a Gross Vehicle Weight (GVW) of 65
tons, which increased even further with the introduction of the T–158 replaceable pad
track system and additional survivability enhancements. With the introduction of the
M1A2 upgrades, the GVW for the tank rose to about 70 tons. The suspension system had
to be upgraded to accommodate the additional weight and maintain a high degree of tank
performance in terms of ride, mobility, reliability, and durability.

The suspension system proposed for the Abrams M1A2 is known as the "Improved
Suspension System." This system consists of larger torsion bars, which increase the
CTE 38 suspension spring rate by 25 percent, and larger shock absorbers, which double the
damping capacity.135 Aluminum torsion bar housings were designed for the undamped
suspension stations in an effort to minimize weight increases. The hull interfaces for the
improved suspension system were not changed. The "Improved Suspension System" was
implemented in production on the M1A2 tank and is scheduled for the M1A1 upgrade
tanks.

Manufacturing processes, maintenance procedures, and changes in the design of the seal
package were developed to improve vehicle suspension assembly and maintainability.136
In particular, to produce such a robust torsion bar it was necessary to use steel in heavy
section processed to high strength levels. For this, it was possible to utilize advanced
heat-treating methods available in the private sector. Modest R&D efforts, funded by
TACOM, evaluated heat-treating methods such as induction hardening and other
thermomechanical treatments used in the private sector.137 These processing methods
were successful in achieving the required mechanical properties and resulted in the use of
high-strength steel for the torsion bar.138

133
Ibid.
134
Michael Blaine, email to authors, 26 May 2005.
135
Ibid.
136
Ibid.
137
Ray Cellitti, telephone interview with authors, 31 January 2005, and John Wiss, email to authors, 13
June 2005.
138
Hunnicutt, 212.

34
 Another area that needed attention was the track system. The T–156 track system initially
installed on the M1 had integral rubber track pads that exhibited far less track life than
designers had hoped.139 Under normal operating conditions, the integral rubber pad
required replacement after about 700–800 miles, far short of the 2,000-mile design
goal.140 TACOM engineers knew from prior work on ground vehicles that, despite a
weight penalty, a replaceable track pad would increase track life. In fact, the M60 had
CTE 39 replaceable track pads. TACOM funded efforts in the private sector on track design and
improved rubber compounds that would return the Abrams to this earlier track concept.141
Modest funding was provided for efforts in academia on systems engineering and
analysis. TACOM engineers integrated and managed these efforts, providing oversight of
design reviews and proof-of-principle testing. The result of this re-engineering effort
(approximately 20 man-years in the mid–1980s from concept to testing) was the T–
158 track system, which had replaceable pads that met the initial track-life goals.142

In the mid 1980s, this T–158 track system was optimized for weight reduction to produce
the T–158LL track system, which was installed on the M1A2.143 The T–158LL has the
same form, fit, and function as the T–158, yet is 1,008 pounds lighter in weight. The T–
158LL track system allows removal of the replaceable rubber road pads to expose the
steel tracks; this, though usually prohibited in peacetime on public roads, can improve
traction in most conditions. Steel "ice cleat" can be installed in place of the rubber road
pads to enhance traction over snow and ice.

Among the most recent track system efforts was research performed in academia to
analyze a problem with the drive sprocket on the M1A2.144 The existing sprocket teeth on
CTE 40 the M1A2 Abrams were designed for an optimal fit with zero track tension. However,
during normal operating conditions, the track tensioner loaded the track with as much as
15,000 pounds of tension. This caused the overall length of the track to increase,
adversely affecting the interaction between the sprocket and the end-connectors and
causing a hang-up of the track during its exit from the drive sprocket. This produced
accelerated wear of the components and noise. The result of this effort was a new
sprocket better suited to the high tension, resulting in a 20-percent improvement in track
end connector wear and a reduced acoustic signature.145

An additional, modest, in-house effort at TACOM tackled the unacceptable acoustic

signature of the track.146 Using complex signal analysis, engineers determined that the
source of the noise was the rubbing of the track center guide against the road wheel. They
changed the shape of the center guide and significantly reduced the signature.

139
Blaine.
140
Hunnicutt, 248.
141
Dennis Sweers, telephone interview with authors, 2 February 2005.
142
Ibid.
143
Blaine.
144
Ibid.
145
Ibid.
146
Grant Gerhart, interview with authors, Warren, MI, 24 November 2004.

35
 VII. Vetronics, C4ISR, and Fire-Control CTEs
As a result of the armaments, survivability, and mobility CTEs discussed in chapters IV,
V, and VI, the Abrams was a substantial improvement on the M60 Patton. Equally
important to the tank’s success are its digital electronics, the way it operates as part of an
effective unit on the battlefield, and its ability to find and hit the enemy. This section will
discuss the technology events that contributed to the Abrams’ abilities in the areas of
vetronics; command, control, communication, and computers (C4); intelligence,
surveillance and reconnaissance (ISR); and fire control.

Vetronics

As the M1A1 began to evolve into the M1A2, engineers at the TACOM in-house
laboratory proposed a radical redesign of the internal electronics using digital
architectures.147 The laboratory had gained ground vehicle control architecture experience
from a late-1970s/early-1980s program in electrical power management and control
systems. Drawing on this experience, the new electronics approach for the Abrams
became the subject of an intense R&D program.

TACOM envisioned a significant change to the electronic integration of the Abrams

using a digital architecture approach similar to that used in aircraft. Specifically, this
CTE 41 “vetronics” approach would change the arrangement from a hard-wired point-to-point
configuration to a software-controlled multiplexed data bus approach, thus eliminating a
significant amount of wiring. During the mid–1980s, the TACOM laboratory funded
efforts with the private sector to confirm and demonstrate that an avionics-type
architecture was applicable to the Abrams, could manage the electrical power, and could
be interfaced with digital radio communication.148 Also in this time period, GDLS,
working under a DARPA/TACOM contract as part of an advanced ground vehicle
technology program, generated important algorithms that were applied to vetronics.149
This GDLS work formed the basis of the M1A2’s digital intravehicle architecture.150 It
also demonstrated the digital interfaces that later enabled the Abrams’ inter-vehicle data
transmission through the Intervehicular Information System (IVIS, discussed below).151

The Abrams vetronics architecture links onboard electronic subsystems, such as sensors
and computers, in real time on a 1-megabit-per-second MIL-STD–1553 data bus. An
advantage of this architecture is seen in the Abrams fire-control system. On the M60A1,
mechanical inputs to and outputs from the ballistic computer, such as selecting
ammunition type, were made by turning a hand crank on the computer or through metal
shafts connected to the commander’s coincidence rangefinder (computer input) and the
147
Don Sarna, interview with authors, Warren, MI, 23 November 2004.
148
TACOM, “Vetronics Evolution” summary document and Don Sarna, interview with authors, Warren,
MI, 23 November 2004.
149
Jerry Lane, email to authors, 2 December 2005.
150
Ibid.
151
Ibid.

36
 main gun elevation system (computer output). As compared to the Abrams, this
mechanical input/output was slow and prone to error.152 Lethality and survivability were
also degraded, as the gunner had to take his eyes away from his sight to look at the
computer during ammunition selection.153 On the Abrams, the vetronics architecture
supports almost instantaneous, push-button selection on the gunner’s controls, provides
connections to significantly more inputs from a variety of sensors (range, cant, wind,
etc.), and allows more rapid adjustment of the position of the main gun. Additionally, the
digital vetronics enables the gunner and commander to share capabilities that previously
had been available to only one, e.g., ranging the target, sighting, and firing the main gun.
This same vetronics architecture also supports control of the power train, mobility, other
internal vehicle control systems, and internal/external communications. Where the M1A1
was 90 percent analog, the M1A2 is 90 percent digital.154

C4ISR

While the internal systems and components on the Abrams are integrated with the
vetronics architecture, the linking to other vehicles, foot soldiers, and commanders is
done only with secure data links and voice radios. The vetronics digital architecture
CTE 42 approach meshed nicely with an early-1980s concept put forth by the Army Science
Board that was aimed at tying together the various pieces of information generated during
the course of ground vehicle operations to give tank crews the ability to know where they
are, to see what other tanks see, and to exchange information between vehicles.155 Crews
on the M60 tank had to use hand-and-arm signals or RF voice radios to communicate, a
system that limited command and control capabilities.156 Improved battlefield
communications and battlefield management systems held the promise of multiplying the
Abrams’ already-formidable capabilities, and the Army leadership strongly supported this
concept.

The Army laboratories at TACOM, Picatinny Arsenal, and the U.S. Army
Communications-Electronics Command (CECOM) moved forward in several areas
during the 1980s and early 1990s, funding internal and external efforts to make the vision
of using digital capability to enhance battlefield communication a reality. The concept of
CTE 43 an overarching Battlefield Management System was first manifest in the Intervehicular
Information System (IVIS).157 Communications networking architecture and protocols
were developed by CECOM for IVIS in the late 1980s and early 1990s to allow M1A2
tanks to communicate situational awareness information to each other and to the
Battalion Tactical Operation Center. The communications protocol provided an efficient
way to transmit digital data from a single M1A2 tank over the bandwidth-limited Single

152
Al Sciarretta, email to authors, 16 June 2005.
153
Ibid.
154
Hans Halberstadt and Erik Halberstadt, Abrams Company (Ramsbury, England: Crowood Press, 2000),
19.
155
TACOM, “Combat Vehicle Command and Control” summary document and Don Sarna, interview with
authors, Warren, MI, 23 November 2004.
156
Al Sciarretta, email to authors, 6 February 2005.
157
Sarna.

37
 Channel Ground and Airborne Radio System (SINCGARS) reliably.158 This was the first
time that digital data had been communicated between M1A2 tanks in the forward areas
of a battalion and below.

A key enabler of this battlefield communication concept is the M1A2’s

CTE 44 Position/Navigation (PosNav) system. PosNav is a nuclear-hardened, autonomous
navigation system with a global positioning system (GPS) component.159 It was designed
specifically for the M1A2, though other military systems have adopted it. PosNav
provides real-time position, heading, and attitude information to the tank crew.160

The Abrams cannot rely solely on a satellite-based GPS system, because terrain often
blocks access of the GPS receivers to satellites.161 PosNav overcomes this problem by
coupling GPS with an inertial navigation system. Based on an initial calculation of the
vehicle's location, PosNav uses the rotation of the track to determine the location of the
vehicle. To compensate for track slippage that could affect the accuracy of the inertial
system, a tank commander can use GPS to update PosNav.162 When connected to IVIS
using the protocols described above, the system enables the creation, sharing, and
constant updating of the battlefield picture.

With all of the Abrams tanks on the battlefield linked together with IVIS, commanders
can keep track of the positions of other vehicles, transmit maps, and share information
about the enemy. The ability of a group of tanks to synchronize their fires against a target
or targets is especially important to unit fire control. In the M60A1 units, this was done
through much talking on the radio and much training.163 The M60A1 unit commander
had to estimate the location of each of his tanks, as well as their orientation and fields of
fire, but he could never really “see” what each tank was seeing. The communications and
information-sharing advances of the Abrams significantly improve the unit commander’s
ability to coordinate fires against enemy targets.

During the mid–1990s, C4 capability was further enhanced by providing communication

and radio access for crew members on all models in the Abrams fleet via the Vehicle
CTE 45 Intercommunication System. This system, available for the M1A2 line and retrofitted to
other Abrams models, consisted of the necessary hookup items and headsets. The latter
included the capability to reduce ambient vehicle noise significantly by using an
electronic Active Noise Reduction unit.164 The unit permitted the crew to operate within
the 85-decibel level established by the Surgeon General without ear plugs. As a result,

158
Gary Blohm, email to authors, 9 March 2005.
159
GPS, an innovation that contributes to numerous weapons systems, was based on concepts that
emanated from DOD laboratories. Tim Coffey, interview with authors, 1 November 2005.
160
SPG Media PLC, “Smiths—Heading/Pointing and Land Navigation Systems.” Available online at:
<http://www.army-technology.com/contractors/satellite/smiths/>, accessed 7 July 2005.
161
Al Sciarretta, email to authors, 15 July 2005.
162
U.S. Army FM17–5. Available online at:
<http://www.nuui.com/Sections/Military/Field_Manuals/FM17-15/chp2.htm>, accessed 12 July 2005.
163
Al Sciarretta, email to authors, 16 June 2005.
164
Blohm.

38
 commands given over the intercom are clearly heard and understood the first time, thus
reducing the potential for fratricide and other mishaps.

The idea of extending C4 capabilities across all battlefield elements, including other
ground vehicles, helicopters, and artillery, was expanded upon in Force XXI Battle
CTE 46 Command, Brigade and Below (FBCB2) packages that were developed by CECOM in
the late 1980s and early 1990s.165 These packages were appliquéd into the M1A1 and
embedded into the M1A2SEP. It was advantageous to have FBCB2 fully embedded in
M1A2SEP to fully leverage the tank’s sophisticated digital architecture.166 The software
system design for the SEP upgrade removed IVIS from the baseline M1A2 system
software and provided a software interface to allow the embedding of FBCB2. The
embedded FBCB2 software was used with the upgraded SEP processors and soldier-
machine interfaces, such as displays and input devices.

FBCB2 significantly enhanced situational awareness as well as command and control to

the lowest echelons. Specifically, it provided a seamless flow of battle command
information across a Tactical Internet system by collecting, integrating, and displaying a
common picture of the battlefield. The development of FBCB2 was greatly helped by the
resources available at the Tactical Internet Laboratory, which was conceived and
developed by CECOM to optimize performance through analyses and simulations.167

Thermal Imagery

The digital advances of vetronics, IVIS, and FBCB2 allowed Abrams’ commanders to
operate their 70-ton vehicles with previously unachievable agility and coordination.
Advances in thermal imaging meant that they could use this capability in battle on a 24-
hour basis, a critical component of American military superiority. The efforts of the
Army’s Night Vision Laboratory (now known as the U.S. Army Communications-
Electronics Command’s Night Vision and Electronic Sensors Directorate), in concert
with private industry and in-house applications engineers, such as those at Frankford
Arsenal, have been instrumental in thermal-imaging advances. As a result of their work,
the first versions of thermal-imaging systems for tanks, such as the externally mounted
Far Infrared Thermal Indicator and the first under-armor thermal viewer device (named
the Far Infrared Periscope), were developed in the 1960s. These early thermal devices
were created by a strong combination of in-house research, design, and prototyping with
industry-produced, non-standardized components.

Non-standardized components, however, meant that the Army bore the cost of outfitting
and maintaining a variety of thermal-imaging systems for individual weapons platforms.
CTE 47 A common sensor module approach, pioneered by the Night Vision Laboratory (NVL) in
the early 1970s, reduced the skyrocketing cost of custom imaging systems. This approach

165
Ibid.
166
Clay Miller, email to authors, 27 January 2005.
167
Blohm.

39
 had its beginnings in an in-house study at NVL on the feasibility of such a concept.168
The report studied the community of Army thermal-imaging systems and concluded that
it was possible to introduce a greater degree of compatibility.

As the next step, in 1973 researchers at NVL patented a device called a Universal Viewer
for Far Infrared. The patent described the configuration of all of the major components,
including the detectors, imager optics, electronics, display of light emitting diodes, and
scanning device that became the Army Common Modules.169 Subsequently, NVL funded
primary and secondary sources for each of the modules to enhance competition and
further reduce acquisition cost. These Common Modules were the basis for a whole
generation of thermal imagers that the Army used not only in the Abrams, but also in
missile systems and helicopters.170 As a result of these efforts, the Army and the other
Services were able to achieve parts standardization and significantly reduce costs. The
success of the approach was shown by the fact that it became accepted production
practice by industry and has led to affordable first and second generation Forward
Looking Infrared (FLIR) thermal imaging systems on the Abrams.171 Savings of many
millions of dollars have been documented in the implementation of this sensor
modularization.172

The Abrams benefited from significant advances in performance optimization of thermal

sensors. Researchers at NVL very early on proposed that the ability of an observer to find
military targets in scenes viewed through an electro-optical device was dependant on how
well an observer can resolve bar patterns at varying spatial frequencies.173 This model
became universally known as the Johnson model, after its inventor, who was a scientist at
NVL. A paper published in 1970 in the classified literature, co-authored by an NVL
researcher and an industry researcher, further quantified recognition performance in
CTE 48 terms of performance related to a minimum resolvable temperature difference (MRT).174
A four-bar pattern is imaged by the system, and a human subject attempts to recognize
the four bars on display at different spatial frequencies. A higher spatial frequency and
lower MRT is indicative of a better system/human performance. This measurement
technique was subsequently standardized at NVL and has became an industry standard
for performance evaluation of infrared systems. This approach to system performance
modeling and specification quantification was codified and documented in an Army

168
James Ratches, email to authors, 27 April 2005. The study itself is Patrick J. Daly Report of the 1971 Ad
Hoc Study Group on Commonality of Thermal-Imaging Systems (Unclassified Version), (Fort Belvoir, VA:
U.S. Army Night Vision Laboratory, 1977).
169
Daly, 77.
170
Ibid., i.
171
An excellent summary of the common module success story is found in a report written by Walter
Morrow, Jr., while he was a scientist at NVL. Walter B. Morrow Jr. Common Modules: A Success Story
(Fort Belvoir, VA: Center for Night Vision and Electro-Optics, 1988).
172
Ratches.
173
John Johnson, “Analysis of Image Forming Systems,” Proceedings of the Image Intensifier Symposium,
1959, n.p., 249–273.
174
J.M. Lloyd and R.L. Sendall, “Improved Specifications for Infrared Imaging Systems” Proceedings of
IRIS, 1970, 109–129.

40
 report175 and was adopted by all three services, DOD, NATO, and the entire electro-
optics community. A close working relationship between the in-house researchers at
NVL and private industry has been key to the performance optimization of the thermal-
imaging systems noted above.176

The M1A2 added a commander’s independent thermal imaging system, which also was
fabricated using the common modules approach. This system gave the tank commander a
sighting system completely independent of the gunner, thus allowing the commander and
CTE 49 the gunner to identify and track separate targets simultaneously. This development has
significantly increased the Abrams’ lethality.177

Fire-Control System and Related Sensors

Advances in thermal imaging technology gave the Abrams crew the ability to find targets
in all conditions, be it at night or through smoke and haze. Seeing the enemy is vital, but
it is only the first step: the Abrams must also be able to destroy the target.

Some aspects of hitting a target have been addressed in the Gun Accuracy section.
Another crucial contributor to meeting this challenge is the gun stabilization system. To
improve accuracy of firing while the tank is moving, the Abrams, like most versions of
the M60 Patton tank, is equipped with a gun turret drive stabilization system that isolates
the gun platform from the effects of vehicle pitch, roll, and yaw. The system makes
automatic adjustments to keep the gun trained on the target. This equipment is essential
to the tank’s ability to fire accurately on the move. The M60A1 had such a system, which
was designed and built by Cadillac Gage.178 Subsequent redesign by Chrysler for the
Abrams corrected for deficiencies in the M60A1 system.179 Over rough terrain the
M6OA1 had a stabilized hit probability of approximately 75 percent at 1,500 meters, at
15mph. Under the same conditions, the M1A1 has a hit probability of 95 percent at 2,200
meters at 25mph.180

What really sets the Abrams’ lethality apart from the M60 is the digital fire-control
CTE 50 system. The advances discussed above—improved munitions, higher muzzle velocities,
gun stabilization—combine with the fire-control system to make the Abrams a singularly
formidable threat.

The Abrams’ digital fire-control system can be thought of as combining two primary
elements: the sensors that collect any information that might affect the flight of the round
and the digital ballistic computer that integrates the information and generates an accurate

175
James Ratches, et. al., NVL Static Performance Model for Thermal Viewing Systems U.S. Army
Electronics Command Report 7043, AD–A011212, (Fort Monmouth, N.J.: U.S. Army Electronics
Command, 1973).
176
James Ratches and Paul Travesky, interview with authors, Fort Belvoir, VA, 18 January 2005.
177
Clay Miller, email to authors, 9 June 2005.
178
M60A1 tanks with a stabilization system were called M60A1(OAS) for Add-On Stabilization.
179
Kelly, 195.
180
Dennis W. Beal, Major, USMC, “The M1A1 Tank: Its Role in the Marine Corps.” Available online at:
<http://www.globalsecurity.org/military/library/report/1991/BDW.htm>, accessed 7 July 2005.

41
 firing solution. The sensors include the laser rangefinder, cant sensor, muzzle reference
system, wind sensor, and atmospheric temperature and pressure sensors. Of these, the
laser rangefinder and muzzle reference system in particular deserve further discussion.

The laser rangefinder replaced the dual mirror-based coincidence rangefinder found on
the M60A1.181 The M60’s rangefinder needed constant adjustment and was susceptible to
temperature and moisture. It took a well-trained crew 7–10 seconds to put a round on
target. The Abrams’ rangefinder, on the other hand, reduces the round-on-target time to
2–3 seconds from target acquisition and has a range three times greater than that of the
M60A1.182

The ruby laser, the first successful optical laser, was the basis for the first type-classified,
CTE 51 man-portable rangefinder in the mid-1960s. In the 1970s, Army researchers first
developed laser rangefinders based on Neodymium: Yttrium Aluminum Garnet (Nd:
YAG) lasers. These laser rangefinder units achieved two pulses per second, and with
continued research efforts the output was increased to 10 pulses per second.183 The
advantage of the Nd: YAG system over the ruby laser system was that they could be run
at higher repetition rates and higher output energies and could operate more covertly,
primarily because of thermal stability.184

With all the advantages of the Nd: YAG laser, there was one disadvantage that required
attention: Nd: YAG lasers were not eyesafe because of their operational wave length
(1.06 microns). Over the years, many investigators have worked on eyesafe replacements
to the Nd: YAG laser. Several approaches were considered. One was to pass the original
laser beam through a methanol cell, which caused the wave length to shift.185 Others
included using an optical parametric oscillator, a special crystal, to convert the frequency
of the laser output to the eyesafe region (such as 1.5 microns) and developing new
materials that emit at eye-safe wavelengths.186 These efforts yielded an industry-
CTE 52 developed eyesafe Erbium laser rangefinder, which replaced the Nd: YAG laser on the
Abrams.187

Another important development in the Abrams’ fire-control sensor suite was the addition
of a muzzle reference system (MRS). The concept of an MRS stemmed from the early
CTE 53 1970s, when field inspections indicated that a substantial portion of the U.S. tank fleet
had improperly adjusted guns. In August 1972, a team of fire-control designers from
Frankford Arsenal filed a patent for an MRS that used a flat mirror fixed to the muzzle of
the gun.188 In this design, a beam of light emanates from the gunner’s periscope and

181
Ibid.
182
Ibid.
183
Joseph Lehman, email to authors, 17 March 2005.
184
It should be noted that once the Nd: YAG laser was perfected and was being utilized in weapons
systems, it opened the door for commercial applications in areas such as medicine and metals joining.
185
Maurice Yeager, telephone interview with authors, 16 March 2005.
186
Ratches, email.
187
Eugene Del Coco, email to authors, 29 August 2005 and DefenceJournal.com, “Abrams M1A1/2.”
Available online at <http://www.defencejournal.com/jul99/abrams.htm>, accessed 30 November 2005.
188
Joseph Lehman, email to authors, 1 June 2005.

42
 reflects to the muzzle mirror. This provides compensatory azimuth and elevational error
signals which are algebraically added to azimuth and elevational range signals produced
by a ballistic computer in response to a rangefinder sighted on a target. The MRS
positions a movable reticle correctly in the periscope to enable the gunner to aim the gun
accurately, thereby increasing first-round-hit probability. Later, more advanced MRSs
utilized an automated sensor mounted on the turret wand that measured the deviation of
the muzzle with respect to that fixed point on the turret.

To field a rugged system, Benet Laboratories at Watervliet Arsenal did extensive analysis
associated with the mechanical attachment of the Abrams’ muzzle sensor unit to the gun
tube.189 During the early years of fielding, there were many instances of breakage of
internal optical components, loosening of the muzzle sensor on the tube, and loss of
calibration (alignment of tube and optics). At that time (late 1980s–early 1990s), Benet
technologists discovered a previously unknown dynamic strain wave that travels axially
through the gun tube walls along with high-speed, KE projectiles.190 It was found that
CTE 54 when the projectile approaches the so-called “critical velocity” of the dynamic strain
wave, severe resonances occur that greatly amplify local strains and accelerations in the
tube—enough to cause the MRS damage being observed. General Dynamics used this
information to modify muzzle sensor design to maintain functionality.191 Since that time,
Benet has developed gun tube design techniques that can mitigate these dynamic strain
waves and prevent MRS damage.

Information from the laser rangefinder, the muzzle reference system, and other sensors
are collected and provided to the other primary component of the fire-control system, the
CTE 55 Abrams’ digital ballistic computer. The digital computer aboard the Abrams, which
benefited from the well-documented development of the U.S. computer and electronics
industry, represents a major leap forward from the analog system used by the M60A1.

The work that led to the M60’s automated fire-control systems was done at BRL and
Frankford Arsenal in the 1960s using first mechanical computation methods and then an
analog computational approach.192 One such analog computer, the M16, went into the
M60 tank. It was initially patented in the 1970s by engineers from Frankford Arsenal,193
and later versions were jointly patented by the original inventors and an engineer from
Hughes Aircraft. This fire-control system used a periscope sight with a ballistic reticle, a
coincidence rangefinder, and an analog ballistic computer that required the gunner to
manually select the type of ammunition. The ballistic reticle assisted the gunner in
adjusting fire and leading moving targets. The rangefinder and ammunition selection
compensated for only two of the essential variables involved in putting a round on target.

189
Ibid.
190
Rick Hasenbein, email to authors, 23 May 2005.
191
Lehman, 1 June 2005.
192
Lehman, 17 March 2005.
193
Ira Goldberg et al., U.S. Patent No. 3686478 (Washington, DC: U.S. Patent Trademark Office, 1972);
Paul Marasco, U.S. Patent No. 3733465 (Washington, DC: U.S. Patent Trademark Office, 1973); Paul
Marasco et al., U.S. Patent No. 3743818 (Washington, DC: U.S. Patent Trademark Office, 1973).

43
In the 1970s, engineers at Frankford Arsenal were able to develop a digitized fire
computer with greatly improved speed and accuracy.194 The Abrams’ gunner manually
inputs ammunition type, temperature, and barometric pressure. The computer integrates
the data, takes into account such things as how much the target should be led, based on
the range and speed of the target and the speed of the Abrams, and generates a firing
solution. The software in the computer draws on instrumented experiments, computer
modeling of the system errors, and confirming field firings (as discussed in the System
Testing, Modeling, and Analysis section) to account for the possible errors in the firing
sequence, including range, cant, wind, speed, and gun firing jump. With digital
processing capability to integrate the data, the Abrams is able to swiftly acquire a target
and bring accurate fire to bear.

194
Lehman, 17 March 2005.

44
VIII. Findings and Concluding Remarks

Findings

1. We have identified the following distribution of 55 critical technology events in the

development of the Abrams tank.
• Armaments and Armor CTEs
o Main Gun: 3
o Gun Accuracy: 3
o Penetrators: 5
o [Sidebar on Important Facilities: 1]
o Sabots: 4
o Propellants: 2
o Armor: 5
o Crew Protection: 6
o System Testing, Modeling, and Analysis: 2
• Engine and Drive System CTEs
o Engine: 3
o Transmission: 3
o Track and Suspension System: 3
• Vetronics, C4ISR, and Fire-Control CTEs
o Vetronics: 1
o C4ISR: 5
o Thermal Imaging: 3
o Fire-Control System and Related Sensors: 6

2. The funding for the development of these CTEs came almost entirely from DOD. There
are many aspects of the Abrams that utilize Army-unique technologies, such as armor and
armaments, crew protection, and fire control and its components. These were all funded
by DOD, either in-house or on contracts. Other critical technologies have broader
applicability and have received additional support from industry. Examples are found in
the power train and suspension system, computers, and communications. Some of the
technologies began as strictly military items but later found their way into civilian
systems, like GPS.

3. The Army did not develop the Abrams all on its own: while some CTEs emanated from
the in-house laboratories, others came from industry, academia, and abroad. Here it is
worthwhile to restate what was said in the Introduction: the goal of this paper is to
identify CTEs related to the development of the Abrams tank. The focus on technical
events should in no way diminish the importance of the close team environment that
existed between in-house laboratories, the PM office, and industry to the success of the
Abrams program. The result of this close working relationship was the effective
transitioning of CTEs noted in this paper.

45
4. Systems integration was key to the transitioning process. This function was overseen
by the Abrams PM office and implemented by Chrysler/General Dynamics. Good
teamwork was evident in the system integration process. Though (as was noted in the
Study Methodology chapter) this paper did not detail the integrative role played by these
parties, their work in this regard was vital and it required an intimate knowledge of the
systems and components and their intended use. As was noted in the Sabot section in
chapter IV, the transfer of in-house personnel to team with the PM office was an effective
way of integrating advanced concepts. Throughout its lifecycle, the Abrams development
program reaped the rewards of close collaboration and teamwork among all of the
Abrams program members.

5. We have five findings regarding the in-house laboratories:

a. A staff of highly skilled/experienced engineers and scientists and the right

equipment and facilities to do the work were critically important. These
capabilities enabled the Army laboratories to recognize valuable new ideas,
concepts, and promising technologies and to apply them to the Abrams tank
development. The Abrams’ developers drew heavily on their prior experience.
In many cases, this experience was gained only through long service in military-
specific S&T areas. To cite only one example, the technologists who improved
the resistance of the 120mm gun to brittle fracture were informed by their past
work on other weapons, most notably the 175mm artillery piece.

b. The Army also benefited from prior investments in unique research

equipment and facilities. At Aberdeen Proving Ground, for example, there was
research instrumentation for measuring the behavior of propellants, penetrators,
sabots, and gun tubes under the extraordinary conditions of launch. Similarly,
the Night Vision and Electronic Sensors Laboratory at Fort Belvoir developed
instruments and techniques that were essential to the work of contractors. This
equipment enabled the laboratories to push the state of the art in night vision.

c. Even where the in-house laboratories were not directly responsible for a
CTE, they often played a crucial role. The in-house laboratories often passed
the findings of their R&D to industry for use in technology developments.
When the technology came from other sources, the in-house laboratories
performed an important role in taking new ideas and adapting and perfecting
them to fit into the overall design. Both these functions drew on the in-house
laboratories’ existing body of professional expertise, tools, and facilities. Either
by laying the groundwork for other advances or making existing innovations
useful to Abrams, the in-house laboratories made essential contributions without
which the new ideas would have been dropped.

d. The in-house laboratories were most deeply involved in those CTEs that
pertained to requirements unique to ground combat vehicles. CTEs in these

46
 areas—composite armor, the 120mm gun, improved crew protection, the more
powerful gas turbine engine, the digital command and control applications, and
so on—were at the core of what made the Abrams so much better than its
predecessors.

e. The in-house laboratory management created a supportive and patient

environment for the technical staff. The management was responsive to the
needs of the users. Many of the interviewees noted the role of management in
allowing and encouraging the staff to explore new ideas. In most cases—not
all—whatever funding was needed was supplied. In virtually all cases relations
among the different in-house laboratories, the PM shop, and the contractors
were described as close and productive. This was sometimes accomplished by
moving staff to meet project needs, even if it meant locating them at industry
facilities. Also, the user played an important part in identifying the threats and
stating the needs. The role of Fort Knox exemplifies this.

Concluding Remarks

These findings and conclusions apply to the Abrams tank and may not apply to other
Army weapons systems. While the results of this report have suggested to us some
general recommendations that likely would apply to other Army weapons systems, we
will reserve judgment until we have completed subsequent papers on additional
platforms. Also left until we complete the papers on additional systems will be any
comments related to current matters of interest, such as acquisition strategies and the
technical personnel skill mix.

We would like to emphasize again that this study has not set out to capture every
technical innovation in the development of the Abrams. Nor have we striven to present
the CTEs in exhaustive technical detail. We are confident, however, that we have
captured most of the major technical events pertaining to the tank, and that these events
support the above conclusions.

47
Appendix A
Individuals Contacted for the Abrams Project
Key
Civil Service CSE Academia ACD Active AM
Employee Military
Government GR Military Retired MR Private Sector PSR
Retired Retired
Consultant CST Private Sector Employee PSE Contractor CTR
* denotes that the individual reviewed some or all of the draft document for completeness and
accuracy.
Last Name First Abrams-era Organization Current Status
Name
Audino Michael Watervliet Arsenal (CSE) CSE
*Baer Robert Abrams PM Office (AM) MR
Baker James Edgewood Arsenal (CSE) CSE
Baker Patrick Ballistic Research Laboratory CSE
(CSE)
Bartle David Abrams PM Office (CSE) PSE
*Beilfuss John Harry Diamond Laboratory CSE
(CSE)
*Blaine Michael Tank-Automotive Command CSE
(CSE)
*Blohm Gary Communications-Electronics CSE
Command (CSE)
Bolon Michael Chrysler/General Dynamics PSE
(PSE)
*Burns Bruce Ballistic Research Laboratory CSE
(CSE)
Buser Rudy Night Vision Laboratory GR, CST
(CSE)
Cardine Chris Abrams PM Office (AM) MR, PSE
Cellitti Ray International Harvester (PSE) PSR, CST
Cerrato Louis Frankford Arsenal/Armament CSE
Research, Development and
Engineering Center (CSE)
Coates Randy Ballistic Research Laboratory CSE
(CSE)
*Chapin Jerry Abrams PM Office (CSE) GR, PSE
*Dean Terry Abrams PM Office (CSE) CSE
DeBusscher David Chrysler/General Dynamics PSE
(PSE)

48
Del Coco Eugene Frankford Arsenal/Armament GR, CST
Research, Development and
Engineering Center (CSE)
Dobbs Herbert Tank Automotive Research, GR, CST
Development and Engineering
Center (AM/CSE)
*Drysdale William Ballistic Research Laboratory CSE
(CSE)
Foos Michael Tank-Automotive Command CSE
(CSE)
Frasier John Ballistic Research Laboratory GR, CST
(CSE)
*Furmanski Donald Frankford Arsenal/Armament GR, CST
Research, Development and
Engineering Center (CSE)
Gerhart Grant Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
*Gibbons Gould Ballistic Research Laboratory CSE
(CSE)
Giordano Robert Communication and GR, CST
Electronics Research,
Development and Engineering
Center (CSE)
Goodman Samuel Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
Gow Edward Tank Automotive Research, GR
Development and Engineering
Center (CSE)
Harju William Abrams PM Office (CSE) GR, CST
Hasenbein Richard Frankford Arsenal/Armament CSE
Research, Development and
Engineering Center (CSE)
*Havel Thomas Ballistic Research Laboratory CSE
(CSE)
Havrilla Alan Abrams PM Office (CSE) CSE
*Higgins Terry Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
Hollis Walter Frankford Arsenal (CSE) CSE
Hoogterp Frank Tank Automotive Research, GR
Development and Engineering
Center (CSE)
*Horst Albert Ballistic Research Laboratory GR, CST
(CSE)

49
Johnson Larry Ballistic Research Laboratory GR, CST
(CSE)
*Johnson Leroy General Motors (PSE) PSE
Kuss Paul Ballistic Research Laboratory CSE
(CSE)
Lane Jerry Tank-Automotive Command CSE
(CSE)
*Lehman Joseph Frankford Arsenal/Armament GR, CST
Research, Development and
Engineering Center (CSE)
Lett Phil Chrysler/General Dynamics PSR, CST
(PSE)
May Ingo Ballistic Research Laboratory GR, PSE
(CSE)
McClellan Richard Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
McCormack Steve Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
*McVey Peter Abrams PM Office (AM) MR, PSE
*Miller Clay Abrams PM Office (AM) MR, CTR
Motzenbecker Peter Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
Nietubiez Charles Ballistic Research Laboratory CSE
(CSE)
*Plostins Peter Ballistic Research Laboratory CSE
(CSE)
Price Renata Armament Research, GR, PSE
Development and Engineering
Center (CSE)
*Raffa Charles Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
*Ratches James Night Vision Laboratory CSE
(CSE)
*Rowe John Abrams PM Office (CSE) CSE
Rowe Walter Ballistic Research Laboratory CSE
(CSE)
Runyan John Ballistic Research Laboratory CSE
(CSE)
Rusch Laurence Watervliet Arsenal (CSE) CSE
*Sarna Donald Tank Automotive Research, GR, CST
Development and Engineering
Center (CSE)

50
Sciarretta Albert Armor Unit, U.S. Army (AM) MR, CST
*Smith Irving Abrams PM Office (CSE) CSE
Sorenson Brett Ballistic Research Laboratory CSE
(CSE)
Stevens Randy Abrams PM Office (CSE) GR, CTR
*Sweers Dennis Tank Automotive Research, CSE
Development and Engineering
Center (CSE)
Travetsky Paul Night Vision Laboratory GR, CST
(CSE)
Underwood John Benet Laboratories (CSE) GR, CTR
Vigilante Greg Watervliet Arsenal (CSE) CSE
Vitali Richard Ballistic Research Laboratory GR, CST
(CSE)
Watson Jerry Ballistic Research Laboratory CSE
(CSE)
*Wheelock Wayne Tank Automotive Research, GR
Development and Engineering
Center (CSE)
*Wiss John Tank Automotive Research, MR, ACD
Development and Engineering
Center (AM)
Wynbelt Walter Abrams PM Office (CSE) GR, CST
Yeager Maurice Martin Marietta (PSE) CST

51
Appendix B
Critical Technology Event List
Number CTE Report Section

1 120mm gun decision Main Gun

2 Fracture mechanics application Main Gun
3 Swage autofrettage process Main Gun
4 Error budget Gun Accuracy
5 Statistical models Gun Accuracy
6 Gun tube straightening process Gun Accuracy
7 Long-rod penetrators/120mm gun decision Penetrators
8 Long-rod penetrator development Penetrators
9 Long-rod penetrator modeling Penetrators
10 Depleted uranium LRP decision Penetrators
11 High rate forming DU process Penetrators
12 Penetrator/target interaction analysis [Sidebar on Facilities]
13 Slipping rotating band Sabot
14 Sabot tipping ring and scoop design Sabot
15 Aluminum sabot technology Sabot
16 Composite sabot technology Sabot
17 Propellant modeling and analysis Propellants
18 Propellant design and development Propellants
19 Hull design and analysis Armor
20 Hull joining technology Armor
21 U.S/U.K armor technology exchanges Armor
22 Special armor design Armor
23 DU armor application Armor
24 Ammunition compartment design Crew protection
25 Less sensitive munitions Crew protection
26 Ammunition sensitivity test rig Crew protection
27 Combustible casings Crew protection
28 Fire protection system Crew protection
29 Nuclear, Biological and Chemical protection system Crew protection
30 Predictive computer models for live-fire tests System Testing, Modeling
and Analysis
31 Robust model for live-fire tests System Testing, Modeling
and Analysis
32 Gas turbine engine decision Engine

52
33 Gas turbine engine development Engine
34 Air filtration system Engine
35 Hydromechanical transmission Transmission
36 X1100 transmission requirements and decision Transmission
37 X1100 transmission gears and brakes Transmission
38 Improved suspension system Track and Suspension
System
39 Replaceable track pad Track and Suspension
System
40 Drive sprocket fix Track and Suspension
System
41 Vetronics digital architecture Vetronics
42 Army Science Board concept C4ISR
43 The Intervehicular Information System (IVIS) C4ISR
44 Position/Navigation system C4ISR
45 Intercom system C4ISR
46 Force XXI Battle Command, Brigade and Below C4ISR
47 Common module approach Thermal Imaging
48 Models to predict Minimum Resolvable Temperature Thermal Imaging
49 Commander's Independent Thermal Imaging System Thermal Imaging
50 Digital fire-control system Fire-Control System and
Related Sensors
51 Laser rangefinder Fire-Control System and
Related Sensors
52 Eyesafe laser rangefinder Fire-Control System and
Related Sensors
53 Muzzle reference system Fire-Control System and
Related Sensors
54 Muzzle reference system fix Fire-Control System and
Related Sensors
55 Digital ballistic computer Fire-Control System and
Related Sensors

53