L` (p312NASA Contractor Report 187154
AIAA-91-3451

An Historical Perspective of the
NERVA Nuclear Rocket Engine
Technology Program
W.H. Robbins
Analytical Engineering Corporation
North Olmsted, Ohio

and
H.B. Finger
Consultant
Washington, D. C.
for
Sverdrup Technology, Inc.
NASA Lewis Research Center Group
Brook Park, Ohio

July 1991

Prepared for
Lewis Research Center
Under Contract NAS3-25266

NASA

National Aeronautics and
Space Administration

 AN HISTORICAL PERSPECTIVE OF THE NERVA NUCLEAR ROCKET ENGINE TECHNOLOGY
PROGRAM
W. H. Robbins*
Analytical Engineering Corporation
H. B. Finger
Consultant
Figure 1 - Propulsion System Performance
Potential for a Manned Mars Mission

Abstract
Nuclear rocket research and development was
initiated in the United States in 1955 and is still being
pursued to a limited extent. The major technology
emphasis occurred in the decade of the 1960's and was
primarily associated with the Rover/NERVA Program
where the technology for a nuclear rocket engine
system for space application was developed and
demonstrated. The NERVA (Nuclear Engine for Rocket
Vehicle Application) technology developed twenty
years ago provides a comprehensive and viable
propulsion technology base that can be applied and
will prove to be valuable for application to NASA's
Space Exploration Initiative. Also the unique
organization that evolved to successfully manage this
effort requiring major contributions and involvement
of two government agencies in partnership with
industry is a useful model when a similar development
is initiated for SEI. This paper, which is historical in
scope, provides an overview of the conduct of the
NERVA Engine Program, its organization and
management, development philosophy, the engine
configuration and significant accomplishments.
Introduction
Manned exploration missions to the near planets
planned in NASA's Space Exploration Initiative (SEI)
are particularly difficult. Long term exposure to the
hostile space environment requires the development
and improvement of mission critical technologies
including life support systems, electric power and
propulsion. The space vehicle propulsion system is
particularly important since it has a major influence
on the vehicle configuration including its size, weight,
and perhaps the most important consideration, cost
and trip time required to perform the mission.
Therefore, it is necessary that planetary propulsion
systems be developed with a step increase in
performance and specific impulse over conventional
chemical rockets. The nuclear rocket, with a specific
impulse potential two to three times that of chemical
systems, is an attractive option and perhaps the only
option in the foreseeable future. The reduction in trip
time with increasing specific impulse is illustrated in
Figure 1. Trip time could be cut at least 100 days from
that of advanced chemical systems.

'Member AIAA 

1

10000

c
0

y

ai

a 1000
F
J
L;

U

4)

a
loo
100 

200 

300 

400 

500

Mission Round Trip Time - Days
Nuclear rocket research and technology
development was initiated in the United States in 1955
and is still being pursued to a limited extent. The real
technology emphasis occurred in the decade of the
1960's and was primarily associated with the
Rover/NERVA Program. The Rover program consisted
of both research and technology development directed
to nuclear rockets for space application. The
principal research thrust, conducted by the Los
Alamos National Laboratory, LANL, was directed to the
development of reactor fuel and reactor systems that
would operate with hydrogen at temperatures above
2200°K. The technology development, called NERVA,
the focus of this paper, was directed to nuclear engine
system components and the conduct of ground test
engine system demonstrations. The NERVA
technology, developed twenty years ago and
summarized herein provides a viable technology base
that will prove valuable for application to NASA's
current Space Exploration Initiative. Also the unique
organization that evolved to successfully manage this
effort requiring major contributions and involvement
of two government agencies in partnership with
industry is a useful model when a similar development
is Initiated for SEI. This paper, which is historical in
scope, provides an overview of the conduct of the
NERVA Program, its organization and management,
the development philosophy and the engine
configuration. In addition, the verification approach,
test philosophy and the significant accomplishments
of this highly successful technology program are

 presented. The emphasis of this paper has been
directed toward the engine system development. Two
topics are emphasized - the rationale for establishing
the engine configuration and the "proof of concept"
approach that demonstrated engine operational
feasibility. LANL reactor research, although a major
and enabling activity in the program, is not discussed
in detail.
The Information presented herein is a synopsis of
the authors' experience as overall Program Director
and Project Manager of the Engine System
Development Program through the decade of the
1960's. Fortunately, the NERVA Program was well
documented. Approximately 100,000 reports and
memoranda were generated and many of these
documents were utilized as a "memory check".
References 1 and 2 were particularly helpful and some
of the information from these references is contained
herein.

the nozzle and the engine control system. Facilities for
engine assembly/ disassembly and engine system
tests were also recognized as being essential and a
significant challenge. Therefore, the design and
construction of these facilities was initiated early in
the program.
In addition to the major program thrust, which was
the determination of engine system characteristics
using graphite reactor technology with hydrogen
coolant, a small but significant activity was directed to
the pursuit of higher risk, high- benefit technologies.
Two primary alternative reactor technologies were
funded; refractory metal reactors, and liquid and gas
core reactors.
There was also a continuing systems engineering
activity associated with mission and flight engine
definition which guided technology development.
Organization and Management

Technology Development Philosophy
After several years of nuclear reactor research
conducted by the Los Alamos National Laboratory.
rocket engine development began in 1961 with the
selection of the industrial contractors, Aerojet and
Westinghouse, to develop the NERVA engine. Initially,
a nuclear rocket engine flight test was planned.
However, in 1963 the nuclear rocket program was reviewed and redirected toward an engine technology
program focused on ground tests only. Nevertheless,
space mission plans associated with manned
planetary exploration, unmanned solar system
exploration, and extended lunar exploration continued
to strongly influence definition of propulsion system
requirements and guided ground test engine definition.
The objective of the NERVA Technology Program
was to establish a technology base for nuclear rocket
engine systems to be utilized In the design and
development of propulsion systems for space mission
application. The principal task was associated with
the assessment of real system performance and
operating characteristics in advance of firm mission
requirements. This approach ensured that system
performance would be well understood when mission
objectives and propulsion system requirements were
clearly specified.
Technology priorities were determined at the
outset — safety and reliability requirements took
precedence over weight and performance
considerations. Emphasis was placed on the
development of critical engine components which
significantly affected system interactions and system
characteristics, namely the propellant feed system,

2

The initial NERVA management issue was
associated with the fact that the nature of the program
required the technical capabilities and functions of
two government agencies, NASA and the Atomic
Energy Commission (AEC). Perhaps the most
significant management issue of all was associated
with the fact that it was impractical to separate
reactor development from the development of the nonnuclear engine components because of the
interactions and critical interfaces that existed among
components of a rocket engine. The technical
management solution to this issue became obvious. A
single program/project organization was mandatory
for efficient and effective program implementation.
There were, of course, many institutional and political
problems that required joint agency resolutions such
as the flow of funds, agency roles and responsibilities,
and staffing. The organizational approach that
evolved, however, gave priority to technical
considerations and was structured to maximize the
possibility of technical success.
The NERVA program was organized as a joint
NASA/Atomic Energy Commission (AEC) program and
both agencies provided resources. AEC funded reactor
research and technology development as well as the
development of reactor and fuel test facilities. NASA
funding was directed to non-nuclear component
technology, engine system development, and the
design and construction of engine test facilities.
The NERVA program was managed by the Space
Nuclear Propulsion Office (SNPO). This "Program
Office" was located in Washington D.C. and was staffed
by both NASA and AEC employees. The director of the

  

Figure 2 - Nuclear Rocket Program Organization
NASA

AEC 

—

The Comm,—i = 

\dmtnts rator

Grneral Manager

Space Nuclear Propulsion Office 

Wiston ofeac
Rtor
Devclop n—t 

,dam .

Adrmcea Rcseazch
and Techr»iogy

(SNPO) 

Director - H. S. Finaer
Deputy Director - M. Klein
Albuquerque opa tlon
oMce

C eveiana Eztc
ISNPO-Cl

Loa Alamos 
Lanoratoty 

11 Spice Fhgh 

EFiVA

KPAI Q PHOEBUS

I I 

Extmston 
I 

LIALSON 

I SNPO -Nl

h-UCLF.AR ROCK

TECHNOLOGY

I

Program Direction

------- 

Reeearct
Center

Leads

Cents 

Line Authority

Figure 3 - Space Nuclear Propulsion Ofnce Cleveland Extension (SNPO - C)
CZ Or THE CffiFJ

PROJECT MNAGERS
A

'

ChkL R. W. d<hrwedcr 

PIANO t PROORAHS OFFICE 

J. li Stager

FACILITIES 

W. Ir Emle£
P. E. Re—

ENGINES 

W. H. Robb Los

HFi►LTH t SAFLTT OFFICE 
D. S. Stntih

F. 6, Ross 

ASSISTANT CHILE  F'OR 
TZA=1NICAL OPERATION• 
L- C- Cc—bwtsu 

REACTORS 

R  EISABa= R 9UALSIT 
ASSURANCL GROUP 
J. W. DI. H 

ASSISTANTHIE
CF FOR 
FACMITIEa DESIGN 
Ir D. Wmi.L s 

1 

ASS IDTA.NT C= FOR
ADl@GETRATION
O. L- Ki+ "11

Btwch

A. A. MedeLr
E- W. R.—

!ce[ Yfaanla^ R
Opatsrbos 8rzmh
a H ReMV

office was a NASA employee and his deputy was on the
AEC staff. The balance of the Program Office staff,
approadmately twenty people, were a combination of
AEC and NASA employees. Specific responsibilities of
the Program Office included program and resource
planning and evaluation, the justification and
distribution of program resources, the definition and
control of overall program requirements, monitoring
and reporting of progress and problems to NASA and
AEC management, and the preparation of testimony to
the Congress.
The Nuclear Rocket Program Organization as it
existed In 1961 is shown in Figure 2. As shown in the
chart, the Director of SNPO reported to both agencies.
The NASA reporting chain was through the Associate
Administrator of the Office of Advanced Research and
Technology to the NASA Administrator. In the case of

AEC, the Director reported through the Division of
Reactor Development and the AEC General Manager to
the Atomic Energy Commission.
Three SNPO Field Offices reported directly to the
SNPO Director. The Cleveland extension (SNPO-C)
was a - Project Office" responsible for the management
of NERVA Engine Technology Development. The office
was located at the NASA-Lewis Research Center
(LeRC). The NASA-Le RC staff provided a major portion
of the engineering support throughout the technology
program. The SNPO-C Project Office managed the
activities of the industrial contractors. Aerojet and
Westinghouse. In addition, SNPO-C responsibilities
included the management of the design, construction
and activation of engine assembly/disassembly and
engine test facilities at the Nevada Test Site.

 The 1966 SNPO-C Organization is shown in Figure
3. The Project Office contained all the technical,
administrative and support functions necessary for
the management of the technology development
activity that was conducted by government, industry
and university teams.
This Program/Project organization was unique in
two respects. 'Me Project Office (SNPO-C) was a line
management organization, and it was comprised of a
co-located AEC/NASA staff, that reported
institutionally as well as programmatically to the
Program Office, SNPO. The Program Office had direct
line management authority over all Project functions.
The Project Office, in turn, was responsible for the day
to day project management activities including
technical, procurement, budget, schedule and safety.
With this organizational structure, roles and
responsibilities were clear and the response to all
Project Manager actions, particularly procurement
actions, was greatly accelerated. The project
organization also benefited from its location and
utilized NASA-Lewis engineering support in the
conduct of day to day project management activities.
The Nevada Extension, located at the Nuclear
Rocket Development Station, managed the test site
base support activities and provided on site support
during facility construction, activation and operation.
The Albuquerque Extension provided liaison with
the Los Alamos Laboratory who had the responsibility
for the KIWI and Phoebus research reactor programs.
KIWI and Phoebus reactor research was directed to
advanced concepts to demonstrate reactor operation
at progressively higher temperature, higher power
density and higher power with reduced fuel corrosion.
In addition to the line management organization
described, SNPO retained technical direction authority
over the Los Alamos Laboratory for the KIWI and
Phoebus reactor development activities.
This organization was unique in NASA history in
that it was a single joint Program Office that was
responsible for implementation of policies and
practices of two government agencies. In addition,
field offices, responsible for Project Management and
liaison, were line organizations reporting to the
Program Office institutionally and programmatically.
This organizational approach proved to be
efficient and effective in managing institutional,
programmatic and technical issues. The keys to its
success were associated with the ability to effectively
resolve interagency issues, provide clear leadership
and direction to the Project organization, and maintain

4

the support of the agencies involved, as well as the
Administration, the Congress and the general public.
NERVA Engine Technology Development

The long lead times required for the acquisition of
system and reactor test facilities forced early
establishment of engine size and propulsion system
performance goals. An early NERVA priority task was
flight system definition derived from NASA mission
planning and the analysis of missions which were
then thought to be 10 to 20 years in the future.
Because it was recognized that firm missions would
take years to evolve, the nuclear rocket engine system
and the associated technology thrusts were defined
such that the resulting technology development
activities would, in so far as possible, be applicable to
a variety of space missions.
System Definition

The missions of interest in the early 1960's were
not substantially different than those being
considered today in NASA's Space Exploration
Initiative. Extended lunar exploration and principally
manned Mars expeditions received the most attention.
The propulsion system requirements for the Mars
mission were the most severe. Chemical propulsion
systems that existed or were under development were
not considered practical for manned planetary
application because of the large vehicle weight and the
trip time required to perform the mission. For nuclear
rockets, spacecraft weights between 1.5 and 3 million
pounds were considered necessary and propulsion
stages with specific impulse values in excess of 800
seconds and thrust levels of 200,000 pounds or more
were desired. In contrast, engine thrust requirements
for the lunar missions, because of reduced vehicle
weights, were much lower, below 100.000 pounds.
Because it was necessary to make an early engine
size selection for multiple missions, one option was
propulsion modules with multiple engines with the
number of engines per propulsion module
commensurate with the required mission thrust
requirements. Using this approach, a single flight
baseline engine was defined with a thrust level of
75,000 pounds and 825 seconds of specific impulse. It
is important to note that the reactor power
requirements, somewhat in excess of 1000 MW, were
similar to the design requirements of the KIWI-B
reactor under development by the Los Alamos
National Laboratory.
The flight engine configuration is shown in Figure
4. The engine was 22 feet high from the upper thrust
structure which mates to the hydrogen propellant tank

 there. The flow then cools the shield and enters the
reactor inlet plenum.
Propellant is distributed from the inlet plenum
into several parallel paths. The bulk of the flow enters
the reactor fuel-element cooling passages and is
heated to a high temperature. The remaining
propellant is distributed to flow passages which

PUMP
TURBINE
CONTROLS
REFLECTOR 

PROPELLANT

PRESSURE VESSEL 

TANK

LINES

TANK
VALVE

Figure 4 - NERVA Flight Engine Configuration

4 

to the nozzle exhaust exit. The components shown
from top to bottom consist of the conical upper thrust
structure, spherical bottles which contain actuation
gas, and the gimbal assembly between the upper and
lower thrust structure. The turbomachinery is
mounted in the lower thrust structure. The reactor and
internal shield is contained within the pressure vessel.
The rocket nozzle mounted to the pressure vessel is
cooled by the main propellant flow.
Prooellant Flow Path
The propellant flow path is shown in Figure 5.
During steady-state operation, main propellant flow
begins with liquid hydrogen, under tank
pressurization, passing through the tank shutoff valve
into the pump suction line. This propellent inlet line
contains a gimbal bearing for thrust vector
adjustment. A centrifugal flow pump pressurizes the
propellant. ^ The pressurized propellant enters the
pump discharge line and flows to the nozzle cooling
passages, removing heat transferred to the nozzle from
the main exhaust stream as well as heat generated in
the nozzle and pressure vessel by deposition of
nuclear radiation energy. The coolant leaves the
nozzle as a low temperature low-density fluid and is
split into parallel flows to cool the pressure vessel,

reflector and control drums. Propellant exits from the
reflector region and is directed along the pressure
vessel dome to remove radiation energy deposited

5

TURBINE EXHAUST
^ 

REACTOR

BLEED PORT
NOZZLE

C-90-09790

Figure 5 - Nuclear Rocket Engine Schematic
provide coolant to various reactor structural elements
and to the peripheral region between the hot reactor
core and the regeneratively-cooled reflector. These
various cooling flows merge at the reactor exit and flow
through the nozzle reaching high exhaust velocities
and high specific impulse.
Cycle Selection
The choice of the turbine-drive cycle was one of
the principal system design selections that was
required because the turbine-drive cycle can have a
significant impact on propulsion system performance
(specific impulse) and engine start-up characteristics.
In addition, major component designs and
interactions between components are also affected.
Included among the components are the turbopump,
reactor, and nozzle.
The approach and rationale for the turbine-drive
cycle selection that is discussed herein was typical of
the process that was utilized for decision making. The
NERVA engine utilized a hot bleed cycle wherein hot

 Figure 6 - Turbine Drive Cycle Options
DESCRIPTION  

CYCLE 

REMARKS

Cold Bleed

•  Hydrogen bled from pressure vessel dome for
turbine drive gas

Gas Pressurization

Cx o(hlgh pressure liquid
hydrogen and flow control system
•  Does not require a turbopump

Chemical Gas Generator

Topping Cycle

Hot Bleed

I,w temperature turbine
large hydrogen now rates required
•  Large impulse penalty
• Minimal acceleration margins during startup
•  Large area ratlon nozzle requtred for heat input
Stmph, reliable sy  ste m
Results In high H 2 storage volumes and high tank weights
Impractical for a flight system

H2- O 2Combustion Products drive a
nurbopump system

Independent power source avoids system integration
Issues
Liquid oxygen must be carried - complicates stage design
•  increased weight - decreased reliability precluded selection

•  Uses reactor heat energy to drive the
turbopump
flow utilized to drive the
•  Total propellant
ProPe
turbtne

• Provides highest specific impulse
Heat energy derived from reflector
Compl e operates at low temperature
•  Complicates reactor design
•  Reactor design effected
'  strength requirrments
•  sealing requirements
'  reflector design
•  Discarded for first generation system

Hydrogen from nozzle chamber mixed
with cold hydrogen to a turbine Inlet
temperature compatible with material
capability
The Selected System

Bleed gas flow rate flied by turbine inlet temperature
•  Small 1  n penalty
•  Nozzle and bleed port development can proceed without
system interactions
•  Bleed port Is a significant development

gas is tapped from the nozzle chamber, diluted with
cold hydrogen gas to reduce the temperature to the
level acceptable to the turbine materials, and passed
through the turbine to drive the turbopump. The
turbine exhaust gas was dumped overboard with some
thrust recovery (Figure 5). This cycle was selected
after a thorough evaluation of engineering alternatives.
Alternatives included pressurized gas cycle, a
chemical gas generator system cycle, topping cycle,
and hot and cold bleed cycles. These turbine drive
cycle options are summarized in Figure 6 where a cycle
description and remarks on each cycle, which include
advantages and disadvantages of each alternative, are
listed. The hot bleed cycle was selected primarily
because the bleed cycle components could be
developed and the performance could be verified with
component testing. Complete system testing was not
required to successfully demonstrate the technology.

Components whose performance were key to the
success of the engine system, namely the fuel
elements, reactor support system turbopump and
nozzle received the most emphasis. Test verification
criteria were established at each assembly level and
higher level tests of sub-systems and systems were
initiated only after meeting well-defined test
verification criteria at the prior assembly level. The
approach is shown in Figure 7.
Figure 7 - Design/ Test Verification
Design/ Verification
Design 

Following the completion of the NERVA engine
design a series of design reviews were conducted
which addressed both component and system designs.
The design review process included reviews of the
configuration and supporting analyses and test data
that established material properties, structural
capabilities and operating environment. After
completion of component fabrication, component
tests were initiated to verify that the functional,
environmental and structural capabilities of each
component met or exceeded the design requirements.

6

Structural/ Environmental
Analysts/ Tests

Design Revicwlsl 
Test/ Verification 
Non-Nuclear 
Component 
Functional Tests 

Test/Verification Approach

Simulation

Component/ System Analysis 

^
Component/
Subsystem 
Structural Tests

Cold Flow Test
Vehicle 

I 

I I

Furl ElcmentTests

I Reactor Subsystem
Test

Engine System
Test

Testing of major sub-assemblies followed
successful component testing. Many reactor subsystem development tests were performed by both the
Los Alamos National Laboratory (KIWI and Phoebus
Test Series) and the NERVA contractors. In addition, a
cold flow test of an engine with an unfueled reactor

 Figure 8 - NRX Reactor Assembly and
Flow Path
Inlet Plenum
\Shield
Reflector 
Outlet

Support Plate

Plenum

Pressure 

Vessel 

Lateral

upport

Core 

System

Reflector---b

Noale

Charm 

>44, ;^

 
r
 

Inlet
Plenum
Nozzle

was made to determine the engine system
performance during the initial start transient. This
was a critical test which determined that a "bootstrap"
start, using the stored heat energy of the nozzle and
reflector, was a feasible startup procedure.
Finally, only after successful performance of all
critical sub-assemblies, engine system tests were
performed. This technique proved to be sound for the
NERVA program and is still utilized to a large extent
for space system development and qualification today.
This test verification approach forced a signL'icant
amount of discipline into NERVA technology
development. Each test had criteria for successful
completion that were met prior to proceeding to the
next assembly level.
KIWI/NERVA Reactor

The NERVA reactor, called NRX (Nuclear Reactor
Experimental), selection was based upon the Los
Alamos National Laboratory KIWI-134 research
reactor. The KIWI-I3 reactor concept had evolved as a
result of several years of intensive research and

development at LANL prior to the start of NERVA and it
was logical therefore that the NERVA (NR)Q reactor
should take maximum advantage of the prior KIWI
research and development effort.

7

The NRX reactor assembly included the reactor
core, reflector, control drums and internal shield and
is shown in Figure 8. The reactor core was composed
of graphite fuel elements impregnated with pyrolyic
graphite uranium carbide particles and supported
both axially and laterally. The reflector was made of
beryllium. Twelve rotary control drums with a boral
sheet located in the reflector were utilized for power
control. The reactor was configured so that, when
incorporated into a flight engine system configuration,
the engine would produce approximately 75,000
pounds of thrust and a specific impulse of 825
seconds. The reactor operating temperature was
approximately 2300°K and the initial reactor
operating Lifetime goal was one hour. A Mars mission
could be successfully accomplished with a propulsion
system with this performance. The KIWI/NERVA
reactor research and technology program was clearly
a major key to the success of NERVA and has been well
documented.
Nuclear Reactor Experimental/Engine System Test NRX/EST

KIWI/NERVA reactor research and technology
made great strides in terms of structural design and
the reduction of fuel element corrosion in the early
1960's. On the basis of significant advancements in
reactor and fuel technology, an early test to determine
the feasibility of an engine system was both desirable
and necessary. In addition, many of the experimental
engine test objectives could be achieved years earlier
than planned. It became obvious that the fastest way
to accomplish this major program milestone would be
to combine an engine system demonstration with a
planned reactor test. NRX/EST (Nuclear Reactor
Experimental/ Engine System Test) therefore emerged
as a key element of the NERVA program. The NRX/EST
test article is shown in Figure 9. A turbopump was
added and installed within the enclosure on the test
cart as shown in Figure 10. The arrangement of the
engine components, although physically different from
a conventional engine, was functionally similar to a
flight system. The critical engine components, namely
the reactor, the turbopump, and the nozzle, were also
functionally "flight-like" components.

 'URBINE, POWER CONTROL VALVE

J 

TO

REACTO^LE

INLET

-^
_ _^-fJ^_ PUMP DISCHARGE VALVE

yn

C-91-06729`
7

Figure 9- NRX/ EST Test Configuration

Figure 10 - Engine components within NP-K/ EST
Enclosure

The NRX/EST Test Proiiram
The primary objectives of the NRX/EST test were
as follows:
1. Demonstrate the feasibility of starting and
restarting the engine without an external power
source.
2 Evaluate the control system characteristics
(stability and control mode) during startup,
shutdown, cooldown, and restart for a variety of
initial conditions.
3 Investigate the system stability over a broad

operating range.
4 Investigate the endurance capability of the engine
components, especially the reactor, during
transient and steady state operation with multiple
restarts.
Real issues were associated with engine
performance. The engine start transient was a major
consideration, that is, could the turbopump be
accelerated (bootstrapped) utilizing the latent heat of
the nozzle and reflector? Another concern was the
design performance of the nozzle bleed port which
provided hot drive gas to the turbomachinery.
Operation of the engine control system over a wide
range of conditions also had to be demonstrated, since
the only information, prior to system testing, was
obtained from control system analysis and simulation.
Results from prior reactor tests, however, provided
some confidence that the engine controls would
perform satisfactorily.
The NRX/EST test program was conducted in
February 1966. All test objectives were successfully
g

accomplished. The engine system was started and restarted several times; the engine was demonstrated to
be stable over a wide range of operating conditions;
required temperature ramp rates were shown to be
achievable; and the engine control system
performance proved to be predictable, safe and

reliable during both transient and steady state
operation. In addition, the endurance capability of
engine components was demonstrated both at rated
power, thrust and temperature as well as during
transient operating conditions. NRX/EST operated
nearly two hours of which 28 minutes were at full
power. This exceeded the operating time of prior
reactors by more than a factor of two.
'Die NRX/EST test program was important in that

it was the culmination of a long line of Rover research
and development tasks. The test demonstrated the
feasibility of a nuclear rocket engine for space
application. The NERVA activities following NRX/EST
resulted in significant incremental technology
improvements beyond that embodied in NRX/EST that
were necessary to accomplish prior to proceeding with
flight system development.
Ground Experimental Engine XE

The second nuclear rocket engine test was
conducted following the completion of construction
and activation of engine test stand # 1, ETS-1, in 1969.
In contrast to NRX/EST run three years earlier, the XE
engine configuration (Figure 11) closely simulated a
flight system both physically and functionally. Flight
component designs were utilized selectively, i.e., only
when component characteristics had an important
influence on overall system performance. An example

 was the use of a flight-design turbopump because
mass and inertia effects had an influence on chilldown
time and acceleration characteristics. In so far as
possible, facility type components and subsystems
were used to save cost and time. Examples were many
valves and the pneumatic system that were not
deemed to affect engine system performance. In
addition, an external radiation shield was added to the
configuration to protect engine components. This
eliminated the need to radiation harden many
components utilized in system testing.
si^
 

----

— PROPELLANT SHUT OFF VALVE

TEST STAND ADAPTER

 

UPPER THRUST STRUCTURE—

TURBOPUMP ASSEMBLY
TURBINE BLOCK VALVE
TURBINE POWER CONTROL VALVE

Ai
COOLDOWN LINE--—Jr! i-- 

EXTERNAL SHIELD

Figure 12 - XE Engine Installed in ETS - 1

CONTROL DRUM ACTUATOR

._S ?(''^ 

successfully demonstrated and that no enabling
technology issues remained as a barrier to flight
engine development.

LOWER THRUST STRUCTURE
PRESSURE VESSEL
TURBINE INLET LINE 

r-r

TURBINE EXHAUST LINE 

I 

PUMP DISCHARGE LINE 

- 

str-T DILUENT LINE

It
- 

i 

TURBINE EXHAUST LINE

NOZZLE
C-90-09789

Figure 11 - XE Engine
The test facility, ETS-1, was designed so that the
engine could be operated in a down firing attitude in an
enclosed compartment with reduced atmospheric
pressure (1-PS1A) so that the space environment was
partially simulated. A picture of the XE engine
installed in the ETS-1 facility is shown in Figure 12.
The installation and removal vehicle (EM can be seen
behind the engine. The EfV was capable of remotely
installing and removing the engine from the facility.
XE Test Program

The engine was successfully started without
external power and operated over a range of transient
and steady state conditions including full power. In
fact, in this test series, completely automatic startup
was successfully demonstrated. The engine was
operated safely and as predicted over the wide range of
operating conditions in 'Several control modes
including the range encompassed by the solid lines of
Figure 13. One that was particularly significant was
operation under closed loop pressure and temperature
control. This control mode was planned in flight for
control of thrust and specific impulse.
soot

q^

^l' 

3 

Fuel Tem^crature IJ nIt

_U 4oa

The objectives of the XE Test Program were similar
to those of the NRX/EST. However, many incremental
improvements in the reactor, other engine
components, and the control system were
incorporated into XE as a result of the information
derived from NRX/EST and other prior reactor tests.
These improvements were to be evaluated in terms of
their effect on startup, power operation, off nominal
operation, shutdown, and cooldown. Because this test
was also the first test in a new down-firing test facility,
a major objective of the test series was to demonstrate
the practicality of the facility for flight engine
qualification and acceptance. Clearly, however, the
paramount objective of this test was to demonstrate
that engine system operational feasibility was

9

"rIM Impulse
Thr . 

Design

o. 3000 

c

E 2000

^

F 

_^

coo

1000

100 

200 

300 

400 

Soo 600 

Nozzle chamber Prr—um, PSI

(Related to Thruetl

Figure 13 - Engine Operating Map

700

 The total run time on the XE engine was 115
minutes and included twenty-eight starts to power
operation. The XE test series was significant in that it
confirmed that a nuclear rocket engine was suitable
for space flight application and was able to operate at
a specific impulse twice that of chemical rocket
system. All engine feasibility issues were successfully
addressed, no component or systems issues were
observed, and the development of a flight nuclear
rocket system could proceed with confidence.

The final KIWI reactor, 134E, the seventh and last
reactor in the series, operated successfully at a
temperature of approximately 2000°K for eleven
minutes and was utilized as the basis for the design of
subsequent reactors.
The NRX series of developmental reactor test were
started concurrently with the KIWI-B tests. The NRX
series of reactors continued the advancement of
reactor technology. The NRX Series demonstrated
improved structural integrity, reduced fuel corrosion
rates (longer life) and higher temperature operation, in
excess of 2200°K. NRX-A6, the last in a series,
operated one hour continually. The fuel corrosion rate
was low. The extrapolated reactor lifetime, 2-3 hours,
would be commensurate with today's manned Mars
mission requirement. In parallel with the NRX Test
Series, research reactors (Phoebus, Pewee and Nuclear
Furnace) demonstrated high power density, high
power (4000 MW), and the potential of long life fuel
operating at high temperature.

In addition, the engine test facility was proved to
be practical for qualification and acceptance testing of
nuclear rockets. In today's environment, however, an
exhaust gas scrubber would be required to control
fission product release.
Summary of Rover/ NERVA Technology Base
The Rover/NERVA program conducted through the
decade of the 1960's was a highly successful
technology program. It's goals and objectives were to
demonstrate the feasibility of a nuclear rocket engine
system for space application. This "Proof of Concept"
program was mission oriented and culminated in the
successful demonstration of a ground test engine
system.

In addition to the reactor test performance, engine
systems tests NRX/EST and XE, described previously,
were successful. No engine system feasibility barriers
to flight development were manifested.

Reactor/Engine Technology
The Rover reactor program started when a series
of research reactors, called the KIWI Series, were
designed, built and tested to determine feasibility of a
reactor concept which would operate at high
temperature with liquid hydrogen. The first series of
tests, called the KIWI-A Series, were conducted in
1959 and 1960. The KIWI test series met its objectives
by demonstrating operational feasibility with
hydrogen and at the same time meeting performance
goals of high temperature operation required for space
application.

1959

T
a

d

1960

1961

1962 

1 

1963

NRX -Al

NRX

REACTOR
TEST

I

U
V)
z

&A A
•
KIWA

A3
KIWI

KIWI

B'A

A significant technology thrust was directed to fuel
development. Graphite was chosen as the fuel element
1964
•

1965 

B4A 

PHOEBUS  

1966

1967

NRX -A3
•

•  KI WI B16 
K1T1

Fuel

NR) -A2

ENGINE
TESTS
KIWI

A summary of the major tests conducted in the
Rover/ NERVA program is discussed in Reference 2
and shown in Figure 14. A total of twenty reactor tests
and two engine tests were performed. The operating
history is shown in Figures 15 and 16. These two
figures dramatically illustrate the major investment
that has been made to produce the significant
advancement in nuclear rocket technology. Seventeen
hours of operating time were accumulated with six
hours in excess of 2000°K.

•

•

1969

1970

1971

1972

NRX -A6
NRX -A5

•
NRX/EST
KIWI B4D
• •KIWI TNI'
• KIWI B4E
PHOEBUS IA

• 

1968 

•
O- FCF

XE

PHOEBUS IB
• 
PHOEBUS 2A
•

PEWEE

40

NUCLEAR

PEWEE
NT'-1 

FURNACE

Figure 14 - ChronolopV of Maior Nuclear Reactor Tests
10

•

  

Figure 15 - Operating Time Versus
Coolant Exit Temperature
for the Full Power Reactor Tests

1

s'wc-n2 -aa

0.9
0.8

60 

TFE> 2500 K

40 

TFE 2200 to 2500 K

0.7

C

C

TFE' 2200 K

n' 20 

0.6

O

L

G
C

a5

O

U

c
2
X

Rf

Q4

0.3

ti
0.2

0

0.1
0
1965 

1966 

1967 

1968 

1969

Figure 17 - ROVER/.NERVA Corrosion Rate History
Calendar Ttme (years)

material because of its favorable neutronics
characteristics and potential to operate at high
temperature. However, it was recognized that
hydrogen reacted with carbon which resulted in fuel
corrosion. Therefore, all fuel elements were coated to
prevent high corrosion rates. Niobium carbide and

zirconium carbide proved to be the most effective in
minimizing corrosion. Progress in reducing corrosion
is shown in Figure 17. Improvements in coating
technology reduced the corrosion rate by an order of
magnitude.
Test Facilities
A major key to the success of Rover/NERVA was
the development of test facilities at the Nuclear Rocket
Figure 16 - ROVER/NERVA Reactor Operating Time

1000
900

^ 

The safety record at the Nevada Test Site during
the period 1959 to 1972 was excellent. Personnel
injuries were limited, the most serious was a result of a
hydrogen explosion in which two workers sustained
foot and ear drum injuries. In addition, three reactors
sustained damage during reactor testing. The damage
to two of the reactors was the result of structural
design deficiencies that led to failure caused by flow-

\\\\\\\\\\\\

Cumulative nme ai

0
1964' 1965 ' 1966

Perhaps the most significant facility when
reviewed from todays environmental test requirements
associated with no fission product release was the
nuclear furnace test facilit y. The nuclear furnace was
a test reactor which was capable of evaluating fuel
elements and fuel clusters in a nuclear environment.
The nuclear furnace test facility was operated with an
exhaust gas scrubber which provided complete
containment of all radioactive material. Although the
scrubber had a much smaller capacity than would be
required for testing reactors planned in the future, the
feasibility of scrubbers has been demonstrated in
small scale.
Safety

cumulattvc T1me at loafers
Above Onc Megawatt

200 
100 

Development Station at Jackass Flats, Nevada. Two
reactor test stands were built and utilized for reactor
sub-system tests. The reactor was oriented In an
upward firing position. The engine test facility was a
downward firing facility with an exhaust system that
provided altitude simulation so that space conditions
could be partially duplicated.
Maintenance, assembly, and disassembly
buildings were also utilized for assembling and disassembling engine and reactor systems. The buildings
contained all the necessary equipment for remote
assembly and disassembly of reactors and engines
and were, in themselves, significant technological
advancements over current "state of the art" facilities.

1 1967 ' 1968' 1969' 1970' 1971 ' 1972
11

 induced vibrations. In both cases fuel elements were
ejected from the core. The damage sustained by the
third reactor was caused by a procedural error which
resulted in the exhaustion of all propellant which of
course is the reactor coolant. The core overheated and
extensive damage to the core including the release of
fuel elements resulted. Fortunately, no personnel
injuries were sustained in these reactor mishaps and
the failure mechanisms were understood and
corrected.

same time attenuates the institutional biases of the
government agencies involved.

Ei)iloEiue

2 Experience Gained form the Space Nuclear Rocket
Proiiram (Rover)
Daniel R. Koening
Los Alamos National Laboratory
Los Alamos, New Mexico 87545, LA10062H

Currently, nearly twenty years after the
termination of the Rover/NERVA Program, there is
considerable new interest in space nuclear
propulsion. As part of the on-going activity associated
with NASA's Space Exploration Initiative, nuclear
thermal propulsion has been identified as an enabling
technology for manned missions to Mars. In this light,
as future propulsion system requirements evolve for
planetary exploration, technology effort similar to
NERVA is likely to be reborn in this decade; perhaps
with the name Rip Van Winkle. Fortunately, NERVA
technology is still applicable to our understanding of
current propulsion needs. The demonstrated
Rover/NERVA technology base evolved as a result of a
long and comprehensive program, including a series of
analyses and tests progressing from parts to
components, subsystems and finally engine systems.
The legacy of twenty reactor tests and two full-scale
engine system tests is impressive evidence of the
proven state of available nuclear rocket technology.
Reactor and engine system tests were
conducted at temperature, pressure, power levels, and
durations commensurate with today's propulsion
system requirements. Feasibility of the nuclear rocket
has been clearly established so that future nuclear
propulsion development associated with new space
exploration initiatives can be directed to incremental
performance, reliability and lifetime improvements. In
addition, a model for an effective management
approach involving two government agencies in a
major technology development has been
demonstrated. The real future development challenge
will be associated with engine and reactor system
ground testing in an environmentally acceptable
fashion.
The management of multi-agency programs is a
difficult task. It remains to be seen whether a Space
Exploration Initiative management structure evolves
which maximizes the probability of successfully
addressing the significant technical challenges
associated with nuclear rocket development and at the

12

References
L Nuclear. Thermal, and Electric Rocket Propulsion:
Fundamentals, Systems and Applications
R. W. Buscand
Gordan and Breach Science Publishers
1964

 NASA

Report Documentation Page

National Aeronautics and 
Space Administration
1. Report No-

NASA CR-187154
AIAA -91-3451

2. Government Accession No.

3. Recipient's Catalog No.

4. Title and Subtitle

5. Report Date

An Historical Perspective of the NERVA Nuclear Rocket
Engine Technology Program

July 1991
6. Performing Organization Code

7. Author(s)

8. Performing Organization Report No.

W.H. Robbins and H.B. Finger
10. Work Unit No.

505 -43-11
9. Performing Organization Name and Address
11. Contract or Grant No.

Sverdrup Technology, Inc.
Lewis Research Center Group
2001 Aerospace Parkway
Brook Park, Ohio 44142

NAS3-25266
13. Type of Report and Period Covered

Contractor Report
Final

12. Sponsoring Agency Name and Address

National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135 - 3191

14. Sponsoring Agency Code

15. Supplementary Notes

Project Manager, Thomas J. Miller, Nuclear Propulsion Office, NASA Lewis Research Center, (216) 433 -7102.
W.H. Robbins, Analytical Engineering Corporation, 25111 Country Club Blvd., North Olmsted, Ohio 44070, subcontractor to Sverdrup Technology, Inc. (Subcontract No. 6401-85). H.B. Finger, Consultant. Prepared for the Conference on Advanced Space Exploration Initiative Technologies cosponsored by the AIAA, NASA, and OAI, Cleveland,
Ohio, September 4 - 6, 1991.
16. Abstract

Nuclear rocket research and development was initiated in the United States in 1955 and is still being pursued to a
limited extent. The major technology emphasis occurred in the decade of the 1960's and was primarily associated with
the Rover/NERVA Program where the technology for a nuclear rocket engine system for space application was
developed and demonstrated. The NERVA (Nuclear Engine for Rocket Vehicle Application) technology developed
twenty years ago provides a comprehensive and viable propulsion technology base that can be applied and will prove to
be valuable for application to NASA's Space Exploration Initiative. Also the unique organization that evolved to
successfully manage this effort requiring major contributions and involvement of two government agencies in partnership with industry is a useful model when a similar development is initiated for SEI. This paper, which is historical in
scope, provides an overview of the conduct of the NERVA Engine Program, its organization and management, development philosophy, the engine configuration and significant accomplishments.

17. Key Words (Suggested by Author(s))

18.  Distribution Statement

Nuclear engine for rocket vehicles
Nuclear rocket engines

19. Security Classif.  (of the report)

Unclassified
­. 

I- " I M 

Unclassified - Unlimited
Subject Categories 20 and 91

20.  Security Classif. (of this page)

Unclassified

21. No. of pages

14

*For sale by the National Technical Information Service, Springfield, Virginia 22161

22. Price'

A03