PROJECT ICARUS

Disc

overy

ph

oto

grap

h o

f lcaru

s

(Rep

rinted

from

F. G

. Watso

n, B

et

wee

n t

he

Pla

ne

ts (195

6), p

late 2, p

ermissio

n o

f th

e Ha

rvard

Un

iversity

Pre

ss)

PROJECT ICARUS

MIT Student Project in Systems Engineering

The M IT Press

Cambridge, Massachusetts, and London, England

Copyright © 1968 by

The Massachusetts Institute of Technology

Second printing, 1 97 9

This book is set in IBM Univers Type

by A & B Typesetters, Inc., Concord, N.H., printed and bound by Halliday Lithograph Corporation

in the United States of America

All rights reserved. No part of this book may be reproduced or utilized in any

form or by any means, electronic or mechanical, including photocopying, re­

cording, or by any information storage and retrieval system, without permission

in writing from the publisher.

Library of Congress Cataloging in Publication Data

Main entry under title:

Project Icarus.

Includes bibliographies.

1. Astronautics-Systems engineering. 2. Space vehicles. 3. Hydrogen bomb.

4. Planets, Minor-(156611. Massachusetts Institute of Technology.

TL870.P76 1979 62 9.43 78-31763

ISBN 0-262-63068-0 (paper)

Contents

Foreword

vii

1 Icarus

2 The Mission Plan

12

3 Nuclear Detonation and Interaction

24

4 Launch Systems

43

5 The I carus Spacecraft

52

6 Guidance and Control

70

7 Communications

95

8 Intercept Monitoring Satellite

108

9 Management a'ld Economic Impact

138

10 Mission Evaluation

1 47

Credits

153

Project History 155

Foreword

ICARUS-an asteroid about one mile in diameter-one of

many such rocks whizzing through space, remnants of some

ancient upheaval-weaving a path around the sun that brings

it near earth every nineteen years. Just a boulder a mile

across; yet if it were to strike earth, the result would be a

cataclysm of unimaginable proportions, unleashing an ex­

plosive power equal to half a trillion tons of TNT. If Icarus

plunged into the ocean, say 1000 miles east of Bermuda,

the resulting tidal wave would wash away the resort islands,

swamp most of Florida, and lash Boston-1500 miles away­

with a 200-foot wall of water.

Imagine 4400 megatons of destruction hurtling towards

earth. Orbit, velocity, mass-all these can be calculated. But

how can the asteroid be stopped? By what complex con­

figuration of aeronautic systems can an explosion five times

as powerful as Krakatoa be averted?

This is the task set a carefully chosen team of MIT

engineers: Stop Icarus. As the days tick past and impact

approaches, plans are drawn, discarded, refined. Weapons

Project Icarus viii

must be devised, guidance and control instruments adapted,

communications systems honed to fail-safe accuracy. The

project will strain the manpower and resources of the en­

tire nation, but there is no alternative.

It didn't happen in 1968; Icarus missed earth by 4 million

miles. But in the scale of the solar system, 4 million miles is

uncomfortably close. And in 1987, how near will Icarus'

eccentric orbit carry it? What are the chances of some other,

as yet undiscovered, asteroid-or worse, a random meteor­

making its way straight for earth? The MIT team's plan may

yet be put to the test.

PROJECT ICARUS

1

Icarus

On June 26, 1949, Walter Baade discovered a faint streak on a 1-hour star-field exposure (frontispiece) taken with the 48-inch Schmidt camera at Palomar Observatory. This heaven­ly body, tagged (1566) Icarus (fk' ar as), and sometimes called Baade's body, is classified as an Apollo asteroid, that is, an asteroid whose orbit crosses that of the earth. At the time of its discovery, and again on June 14, 1968, Icarus passed within 4 million miles of the earth. Within the past 35 years, 3 other Apollo asteroids-Apollo, Adonis, and Hermes-have missed the earth by 2 million, 1 million, and 0.5 million miles (just twice the distance to the moon), respectively. The fact that these Apollo asteroids pass relative­ly close to the earth, in terms of interplanetary distances, has generated considerable interest in them around the world, and in some cases even grave concern.

Since the Apollo asteroids are relatively small, typically a mile in diameter, and since they pass the earth so quickly, at speeds of perhaps 20 miles per second, they are visible for at most a few hours at a time, even with the largest of tele-

Icarus 2

scopes. This limited observation time makes it difficult to

obtain sufficient data for calculation of the orbits of the

bodies. Icarus, however, has been observed on enough oc­

casions since its discovery to insure the calculation of its

orbit to within approximately 150 miles. One of the unique

properties of this orbit is the ratio of its period to that of the

earth, 19 to 17, which results in a near miss every 19 years.

It is thus important that future generations keep a wary eye

on Icarus to allow time for preparation should a collision

become imminent.

Of course, a collision with Icarus or with any other Apollo

asteroid in the near future is highly unlikely. Perhaps the only

way in which the present orbit of Icarus can be perturbed

into a collision orbit is by a glancing blow from some other

asteroid as Icarus streaks through the fringes of the asteroid

belt beyond Mars. Such an occurrence is improbable-but not

impossible. Perhaps more frightening is the fact that the dis­

covery of the Apollo asteroids by accident, under the most

favorable conditions for observation, indicates that most of

them whiz by undetected and perilously close to the earth.

It should be re-emphasized at this point that a collision

between an asteroid and the earth is unlikely to occur soon;

Watson, for example, has estimated that such a collision is

unlikely to occur more often than once every 100,OQOyears (1).

But there is evidence that meteoritic impacts have occurred

in recent geological times; the Barringer Crater in Arizona,

4,200 feet across, is believed to have been formed on impact

between 5,000 and 50,000 years ago. Other rimmed circular

depressions like the Richat Structure in Mauritania, 75 miles

across; the Vredefort Ring in South Africa, 85 miles; Mani­

couagan Lake in Canada, 40 miles; Hudson Bay in Canada;

and even the entire Pacific Ocean may have been formed by

the impact of gargantuan projectiles millions of years ago.

Project Icarus

The consequences of a collision with Icarus are unimagin­

able; the repercussions would be felt the world over. In dis­

sipating the energy equivalent of half a trillion tons of TNT,

100 million tons of the earth's crust would be thrust into

3

the atmosphere and would pollute the earth's environment

for years to come. A crater 15 miles in diameter and perhaps

3 to 5 miles deep would mark the impact point, while shock

waves, pressure changes, and thermal disturbances would

cause earthquakes, hurricanes, and heat waves of incalculable

magnitude. Should Icarus plunge into the ocean a thousand

miles east of Bermuda, for example, the resulting tidal wave,

propagating at 400 to 500 miles per hour, would wash away

the resort islands, swamp most of Florida, and lash Boston-

1500 miles away-with a 20D-foot wall of water.

In light of the consequences of a collision with an asteroid

the size of Icarus, the possibility of such a collision, no mat­

ter how remote, cannot go unrecognized. The world must be

prepared, at least with a plan of action, in case it should sud­

denly find itself threatened by what had so recently been

considered a folly. Thus Project Icarus was conceived. Icarus

would collide with the earth in just 70 weeks from the proj­ect's inception-unless, of course, the project team, care­

fully handpicked at the Massachusetts Institute of Tech­

nology, could successfully complete its mission. No funds

or manpower would be spared; the resources of the nation

and of the world were at the disposal of this select group of

scientists and engineers.

As the study progressed, the intricacies of a realizable

solution became more and more evident. But the members

of the Project Icarus team, recognizing the remote possibility

of disaster, pursued a solution with relentless determination.

This determination resulted in what the team felt to be much

Icarus

more than a pure academic study; it resulted in a solution to

a problem perhaps more imminent than anyone realizes, and

the goal of that solution is the most rewarding of all 90als­

the saving of human lives.

The Orbit of Icarus

4

In its present orbit of eccentricity 0.83, Icarus at perihelion

passes twice as close to the sun as does Mercury, while at

aphelion it reaches past the orbit of Mars, almost to the com­

mon asteroid belt, about 2 astronomical units from the sun.

Its orbital plane is inclined 22° to that of the earth such that

Icarus approached our planet in 1968 from above the ecliptic

plane (fig. 1.1) at a speed (relative to the earth) nearly equal

to the earth's orbital speed, approximately 18 mi/sec. The true orbital elements of Icarus' path were altered such that

impact would occur at noon in the mid-Atlantic, about 1000

mi east of Bermuda, on June 19, 1968. Projections of the

EARTH POSmol( DEC 22

EARTH POSiTION MAR 21

1.1

DESCENDING NODE dUN 21

Orbit of Icarus. 1968

\ \ \ \ \ I

\ I \ ,

, • I ',..... \ /'

. APHEUO;--I'

OCT 11. 1961../

Project Icarus

1.2

VERNAL EQUINOX ,.

o 5 10 20 , ,

Ecliptic projection of collision orbit (seen from earth)

SUN VERNAL EQUINOX

I 1)-3e

ICARO!!

� 0-32

0-2' 1 AU

1.3 Ecliptic projection of collision orbit (seen from sun)

5

Icarus

hypothetical collision orbit on the ecliptic plane, as seen by

observers moving with the earth and with the sun, are shown

in figs. 1.2 and 1.3, respectively.

Brightness and Visibility of Icarus

6

The brightness of a planet or asteroid (as seen from the earth)

varies inversely as the square of the distance from the asteroid

to the earth, inversely as the square of the distance from the

asteroid to the sun, and directly as the fraction of its visible

surface which is illuminated by the sun (that is, the phase of

the asteroid). By assuming that Icarus is a perfect sphere, that

exactly one-half of its surface (that is, a hemisphere) is il­

luminated by the sun, and that sunlight is uniformly scattered

from the illuminated surface, one can derive a brightness

curve from the geometry of I carus' orbit and from a bright­

ness measurement at a known time. Figure 1.4 shows

the brightness variation during the 50 days before collision

in terms of visual magnitude, mv' Although at its dimmest

.�".....--.----'---r-"""'T-"""

IS I;

E:"- Iii ;J 15 § II � 1:1 " l� � II ..J IU � !) !!l • ;,.

1.4

��'-�I.��2.��a.�-��.--i� TIM.: Til CULJJSIIlN (DAYS)

Visual magnitude of Icarus (from 50 days to collision I

Project Icarus 7

3 " S 6 7 � 9 10 11 12 TIME TO COLIJSION (III')

I I I I I I I I 100 �OtJ 3110 1110 Sou nuo ;00 �lHl

IlL'TAl<n: t'IlCJ�1 t:AIlTIl (JO�O M J)

1.5 Visual magnitude of Icarus (from 12 hr to collision)

during this period, 18th to 19th magnitude, Icarus can be

detected only,

with the largest telescopes, it begins to

become visible to the naked eye at about 5th or 6th magni­

tude, less than 9 hr before collision (fig. 1.5). If the sun did not obscure the star field behind it, the trace

of the hypothetical orbit of Icarus and of the sun's apparent

path on the celestial sphere would be as shown in fig. 1.6.

From D-l (1 day before collision) until impact (D) Icarus

would appear as a brightening point on the celestial sphere.

Such a view is most easily obtained outside the earth's at­

mosphere, where diffusion of sunlight is minimized. Although

the proximity of lines of sight to Icarus and to the sun from

just outside the earth's atmosphere hampers observation of

the asteroid, nevertheless, observation by a star tracker is possi­

ble with a suitable sun shield (see section on electro-optical

instrumentation, p. 79).

Within the atmosphere, optical telescopes are severely

restricted. Since Icarus remains extremely dim until a few

days before collision, even the largest ground-based tele­

scopes require a relatively dark background, that is, after

Icarus

•

•

•

•

I ,"

I ."

1.6

• • • . . .

•

. .

i .h

. .

Apparent path of Icarus across the heavens (hypothetical orbit)

CAPELLA • . •

• • .. BETl LGEVSE

• I ...

8

sunset, to insure observation. The proximity of Icarus to the sun

causes Icarus to be, at most, low in the sky when the su n is over the

horizon. Its elevation is then reduced approximately 15°/hr after

sunset or before sunrise. Observation periods, if available at all,

are thus very short. Figure 1.7 shows the maximum elevation achieved by Icarus during the last 50 days before impact. "Civil"

twilight is the beginning of "night" for legal purposes, while

"astronomical" twilight is the condition of near-maximum dark­

ness. By waiting for astronomical twilight, a necessary condi-

tion for observation, one must look for I carus near the horizon,

where city lights and the th ickest portion of the atmosphere

hamper viewing and where most large telescopes will not even

operate. Despite the difficult visibility problem, every attempt

to track Icarus during the last few days before collision must be

made in order to pinpoint its trajectory as accurately as possible.

Project Icarus

a

DAY OF OBIERYATIOK

1.7 Elevation of Icarus

''CIVIL" ftlUClIIT

Physical Characteristics of Icarus

Although Icarus is commonly called an "asteroid," its origin

is in fact a subject of considerable debate. If Icarus is an

asteroid, its eccentric orbit can be explained as the result of

9

a perturbation by other heavenly bodies. But according to

Opik (2, 3) those "asteroids" whose orbits are considerably

different from those of the bodies in the asteroid belt may

have a cometary origin. That is, Icarus may in fact be a dead

cometary nucleus consisting of a conglomerate of ice, dust,

and gases. A number of other theories exist concerning the

origin of Icarus, each theory implying a different set of

physical characteristics for the body. The theory of cometary

origin, for instance, suggests a relatively low density, while

a true asteroidal origin implies the density of stone or iron.

To insure a conservative mission plan, that is, a "worst-case

design," for Project Icarus, bounds on the physical character­

istics of the body were determined from existing data and

theories. First, the albedo or reflectivity of Icarus was es-

Icarus 10

tablished between 0.07 and 0.28, approximately that of the

moon and of the large asteroid Vesta, respectively (4). To­

gether with observed brightness measurements, the albedo is

sufficient to indicate the limits of the radius of a spherical

Icarus (5): 130 0 to 2500 ft. The most commonly used value

for albedo yields a most-probable radius of 2100 ft.

According to Whipple, the density of Icarus is between 1.3 g/cm3 and 8. 0 g/cm3 (4). The weight of Icarus is then between

380 megatons (Mt) and 17,000 Mt, although the most prob­

able density, 3.5 g/cm3, together with the 21oo-ft radius,

yields a nominal weight of 4400 Mt.

The rate of rotation of an asteroid and the axis of its rota­

tion can be found approximately by careful analysis of the

shape and variation of its light curve, a tabulation of the

brightness of the asteroid as it varies with time. In general

the period of rotation can be found during 1 long night of

observation, or over a period of 2 or 3 consecutive nights (6). The shortest period of rotation of an asteroid determined

to date is 4 hr, 9 min. The longest period is about 18 hr. Ac­cording to Ahmad, the theoretical limit of rotational period

is 3.3 hr for an asteroid of density 3.5 g/cm3 (7). This limit,

calculated from Jeans (8), is valid for an incompressible fluid,

and thus accounts only for the gravitational attraction of the

fluid particles. If one considers also the tensile strength of

particles in a stony or iron asteroid, the rate of rotation can

increase considerably, perhaps to as high as 1 rpm, without

breaking up the asteroid. For purposes of narrowing the

detection bandwidth of the radar used to track Icarus, it is

most appropriate to consider Icarus' rotational rate as the

fastest that has been observed to date. Telescopic observa­

tions of Icarus as it approaches the earth may afford more

precise information about the true rotational rate.

According to Groereveld and Kuiper, very little can be said

Project Icarus 11

about the shape or axis of rotation of an asteroid until it has

been observed on at least 4 epochs during 2 oppositions, pref­

erably near the stationary points (6). The high eccentricity of

Icarus' orbit makes such observations a near impossibility.

Only the shapes of the largest asteroids can be observed as

more than points with the world's largest telescopes. Such

observations have indicated both irregular, elongated shapes

and nearly perfect disks. One might even postulate a dough­

nut shape for Icarus, similar to that of the iron meteorite of

Tuscon, Arizona. Because of its relatively small size, the shape

of Icarus will remain its most uncertain characteristic.

References 1. Watson, F. G., Between the Planets (Cambridge, MA.: Harvard University Press,

1956), pp. 25-28, plate 2.

2. Opik, E. J., "The Stray Bodies in the Solar System. Part I. Survival of Cometary

Nuclei and the Asteroids." In Advances in Astronomy and Astrophysics, Z. Kopal,

ed. (New York: Academic Press, 1963), vol. 2.

3. Opik, E. J., "The Stray Bodies in the Solar System. Part II. The Cometary Origin

of Meteorites." In Advances in Astronomy and Astrophysics. Z. Kopal, ed. (New

York: Academic Press, 1966), vol. 4.

4. Whipple, F .• Smithsonian Astrophysical Observatory, oral communication,

MIT, Cambridge, MA .• February 14, 1967.

5. Allen, C. W., Astrophysical Quantities, 2nd edition (New York: Oxford Uni­versity Press Inc, 1964), p. 153.

6. Groeneveld, I., and G. P. Kuiper, "Photometric Studies of Asteroids. I," The

Astrophysical Journal 120, July 1954.

7. Ahmad, I. I., "The Light-Curves of Ceres, Hebe, Flora, and Kalliope," The

Astrophysical Journal 120, July 1954.

8. Jeans, J. H., Problems of Cosmogony and Stellar Dynamics (Cambridge, England: University Press, 19 19).

2

The Mission Plan

Mission Possibilities

Various possible methods were considered to meet the Icarus

threat. These included the following:

• a soft landing of rockets on the surface of Icarus which

could be emplaced and utilized to perturb Icarus from its

collision course;

• the detonation of a nuclear explosive charge implanted be­

neath the surface of I carus after a reasonably low-speed land­

ing to break the asteroid into small rubble which would either

miss the earth entirely or burn up on passing through its

atmosphere;

• the disintegration of Icarus with a hydrogen bomb delivered

by an interceptor at a high closing speed; and

• perturbation of the orbit of Icarus by detonation near the

surface of a bomb delivered at a high closing speed.

The last of these involves a complicated interaction process

whose possibilities were recognized only after considerable

study.

The Mission Plan 13

The Effect of Mission Constraints

The various possibilities were not obvious at the outset of the

study; in fact, the mission plan finally selected was developed

in detail only as the various capabilities and limitations were

explored. It is appropriate, therefore, to review the important

constraints and the ways in which they influenced the defini­

tion of a mission plan.

The fact that the problem was posed in the spring of 1967,

instead of a year or perhaps 5 years earlier, had a major in­

fluence on the solution. First, it was too late to perform the

task leisurely, both from the standpoint of orbital mechanics

and from that of hardware development and implementation.

A rendezvous with Icarus either for orbital perturbation or

for disruption should be made at aphelion. There, when the

asteroid is moving slowly, its orbit can be most easily affected,

and rendezvous requires minimum spacecraft propulsion

capability.

However, to make the rendezvous at Icarus' aphelion of

November 1967, the space veh ide wou Id have to have been

launched 8 months earlier, that is, within a few weeks after

the problem was posed. Such a launch time was of course im­

possible. Similarly a fly-by probe, which would have been

launched no later than October 1967 to catch Icarus at its

ascending node in April 1968, was dropped from considera­

tion because of insufficient lead time. Launch dates in 1968

were accepted as feasible only on the basis of adaptation of

existing hardware and by postulating top emergency priority

in all related technical and industrial efforts.

Under this difficult time constraint, a soft landing or even

low-speed impact to deposit rockets or explosives on Icarus

was found to be impossible. A soft landing would require not

only escaping from earth but then reversing direction and

matching Icarus' velocity. The total velocity increment ra-

Project Icarus

quired for this maneuver would be at least 140,000 fps-far

beyond the capability of existing launch vehicles.

14

Preliminary studies were made, therefore, of the various

possibilities of destruction of Icarus or of perturbation of its

orbit by a high-speed intercept with a nuclear explosive during

the last days before collision. Complete disruption, as shown

in chapter 3, would require detonation of a charge equivalent

to 1000 Mt of TNT or more at the surface of Icarus. Even

with our present awesome capabilities, it appeared doubtful

that a bomb so powerful could be designed and manufactured

in time. Nor could the equivalent of such a large device be ob­

tained with a number of smaller ones because the timing of

the detonation of nuclear explosives is so critical that one

might destroy the others. Rendezvous techniques were con­

sidered as a means of achieving a larger total payload, but with

the Saturn V still several months from its initial flight date,

the development of both the hardware and the rendezvous

technique was considered impractical.

The use of multiple vehicles, each carrying independent

payloads, was suggested next. Such a scheme offers the ad­

vantage of improved mission reliability. For these independent

missions, standardization of launch vehicle and payload con­

figurations was necessary because of insufficient time to per­

mit individual variations.

The Saturn V was found to be the only suitable launch

vehicle for the Icarus mission. An accelerated schedule of

booster production and of launch pad construction was postu­

lated under the emergency conditions. Even under these

assumptions, no more than 9 complete Saturn V launch

vehicles, it was estimated, could be produced in the allotted

time. A minimum of 3 flight-test vehicles would be required

to qualify the Saturn V before assigning it to such a mission;

therefore, the number of interceptor missions was limited to

The Mission Plan 15

6. The accelerated schedule of launch pad construction

made possible a launch rate of 1 vehicle every 10 days by

May 1968.

In addition to Saturn V propulsion, an additional space­

craft stage was required to perform guidance maneuvers. After

evaluating several such stages, the Apollo Service Module (SM)

was chosen as the basic vehicle, primarily because of its com­

patibility with the Saturn V. The on-board power supply

being developed for the Apollo Applications Program can

operate in the Service Module for a maximum of 60 days.

Since replacement or major modification of this existing

system was impossible in the time available, the power-supply

lifetime became a constraint on the total flight duration.

High-Altitude Interception

If a bomb no larger than a few hundred megatons were de­

livered to Icarus, the body could not be completely destroyed.

Instead, the possibility exists that Icarus might break into

several large fragments, some of which would still collide with

the earth. The impact of several of these fragments could

cause perhaps as much destruction as the whole of Icarus it­

self since the individual impact energies, although reduced,

would be more widespread. Subsequent interceptors in a multi­

launch series could attempt to reduce such large fragments to

an acceptable size, but the guidance and control problem

would be much more difficult than that involved in a rendez­

vous with the entire asteroid. A "deflection" mission thus

seemed most attractive, since a successful perturbation of

Icarus from its collision course would spare the earth of a"

damage in June 1968. Icarus' nominal collision trajectory as viewed in an earth­

centered nonrotating coordinate system is essentially straight-

Project Icarus

400 rr----,-----r---,-----,

> <I � 200 0:: :.: > III Z < '" "" 100

o�-�-_� __ � _ __J

2.1

o 10 15 �o INTERCEPT ALTITUDE (J06 Mil

Required transverse deflection velocity

line constant-velocity motion from about 0-15 days. Over

16

this short time-interval, for practical purposes, the distance

that Icarus is deflected from its nominal collision course due

to a transverse impulsive velocity increment propagates linear­

ly with time. Figure 2.1 shows the transverse velocity incre­

ment required to deflect Icarus 4000 mi (1 earth radius) from

its nominal collision point as a function of the altitude above

earth of the application of the impulse. Note that only com­

ponents of velocity increment normal to the collision trajecto­

ry deflect Icarus; components parallel to the trajectory

merely change the time of collision. (This reasoning neglects

the effect of solar gravity due to the difference in position of

the earth and Icarus. A more sophisticated treatment of the

first-order orbital perturbations of Icarus indicates the approx­

imate method to be accurate to within 1 percent) .

The next step in the evaluation of high-altitude missions

was to determine an optimum combination of intercept alti­

tude and weight of the bomb within the maximum flight time

of 60 days. This optimum combination would maximize the

The Mission Plan 17

total deflection from the predicted collision point. The deflec­

tion distance is the product of the velocity increment normal

to Icarus' approach path and the time before collision at

which the increment is imparted. It was assumed that the

velocity increment imparted to Icarus was directly propor­

tional to the mass of the bomb used to effect the velocity

increment. Thus a reasonable figure of merit for a deflection

mission is the product of the mass of the bomb and the

altitude at interception.

Figure 2.2 shows intercept altitude as a function of vehicle

mass for the Saturn V I Apollo SM. Capabilities of this vehicle

system are presented in detail in chapter 4. I n deriving fig. 2.2,

the total propulsive capability of the combined stages was

utilized, with a flight duration of 60 days. The intercept alti­

tude was then obtained using a simplified model for orbital

motion (1). The launch configuration was assumed constant

and the bomb weight for each vehicle taken to be the only

variable weight.

25 �--�----�----�--�

2.2

FIXED SPACECR!\FT WEIGHT. 56.000 LB

110 WEIGHT OF SP!\CECRAFT+ BOMB (JOOO LB)

Intercept altitude versus weight at launch

Project Icarus

1.0 r----::r:=---_:::::--r----,

0.63'':-5 --4�O---!-.45:----:5":-O -�55

Wt:rCIIT OF ROMR (lnnn LR)

2.3 Figure of merit versus weight of bomb

18

A trade-off curve (fig. 2.3) was thus derived, and it became

evident that a high-velocity, small-payload mission is desirable,

where "small payload" implies a bomb with a yield of the

order of. 100 Mt. For guidance and control purposes, a high­

altitude interception is difficult because earth-based tracking

and communications systems approach the limit of their

performance capabilities. Most guidance and control informa­

tion must then be obtained and processed on-board the inter­

ceptor. Since, as mentioned earlier, deflecting impulses must

be applied normal to the encounter velocity, the guidance

problem is further complicated. I nstead of the whole of

Icarus as an aiming point, the true aiming point must be the

edge of Icarus. The difficulties involved in achieving this

lateral aiming point are relieved, however, in 2 ways. First,

the penalty for missing the edge by an amount or is only

(or/ml, where R is the radius of a spherical Icarus, and I is

the impulse delivered to Icarus (fig. 2.4) . Thus, an error of

0. 1 R reduces the effective component of the impu Ise by only

The Mission Plan

2.4

DESIRED TRAJECTORY

Penalty for missing edge of Icarus

19

10 percent. Second, the optical sensor on the interceptor can

track only the illuminated portion of Icarus, which happens

to be in a crescent phase.

Low-Altit!Jde Interception

The object of a low-altitude interception is to bombard Icarus

with the heaviest possible payloads to destroy as much of the

asteroid as possible_ Due to the limitations of the launch

facilities, a "salvo" launch of 6 vehicles is not possible; some

of the intercept vehicles must be sent aloft well before the

planned interception. Two schemes for parking these early

vehicles were investigated: periodic orbits and low-energy

escape trajectories.

Low circular orbits (of the order of 500 nm) require the

smallest possible velocity increment, but the kinematic dif­

ficulties of interception for vehicles in these orbits seem pro­

hibitive. To improve the kinematics, it was suggested that

orbits be selected in such a manner that the interception

wou Id occur along a straight line. Had it not been precluded

Project Icarus 20

by fuel boil-off in the S- IVB stage of the launch vehicle, an

ideal solution would have been to remain in a low parking

orbit until a few hours before the desired interception time.

Then the intercept kinematics would have been improved by

a long burn of the S- IVB and spacecraft propulsion systems

to "straighten out" the orbit.

Long-period (highly eccentric) elliptical orbits and slow

escape trajectories seemed more promising than very low

orbits, since interception could be made to occur while the

vehicle was moving in a relatively straight line. These orbits

can also be achieved without incurring boil-off of cryogenic

propellants; thus although they require some 10,000 fps more

than low earth orbits, the possible bomb weight is still nearly

twice that possible for high-altitude missions.

Low-altitude missions also offer better guidance accuracy

than high-altitude missions because of proximity to earth and

lower closing speeds. I n addition, the shorter average flight

duration results in better component reliability. But such

considerations as the net payoff of the various missions, that

2.5 Ascent trajectories for final mission plan

The Mission Plan 21

is, deflection versus fragmentation, as well as the increased

complexity of the rendezvous geometry, precluded a decision

to make all interceptions at a low altitude.

The Final Mission Profile

Ascent Trajectories The final mission plan includes 4 high­

altitude ascent trajectories and 2 low-altitude trajectories. For

each mission the full capability of the launch vehicle is re­

quired. The requirement that interception occur when the

Haystack radar facility in Massachusetts and the Goldstone

facility in California can both "see" Icarus restricts the orbits,

as do the launch plane requirements, which necessitate the

use of 1 of only 2 launch windows per day. The first high­

altitude trajectory represents the maximum attainable altitude

for the given propulsion system and flight time, while the

other trajectories are convenient selections from the set of

possible orbits. A day's delay in the launch of any vehicle

would not be crucial to the successful execution of its mission.

Figure 2.5 and table 2.1 show the 6 orbits and the nominal

launch schedule associated with those orbits, respectively. If

all Saturn V systems operate successfully, 1 vehicle will be

launched onto each trajectory.

Table 2.1 Launch schedule for Icarus interceE!tors

Launch Intercept Intercept Interceptor Date Date Range Number (0-1 (0-1 (10· mil 1 72.9 12.9 20.0

2 58.3 9.9 15.5

3 44.0 6.9 10.8

4 32.7 4.9 7.7 5 5.3 0.9 1.41

6 4.9 0.8 1.25

Project Icarus 22

Redundancy of Launches Although time for refurbishment

of the launch pads would aJlow for 5 high-altitude intercep­

tions, only 4 such attempts were scheduled. In addition, the

intercept altitudes of 3 of these 4 missions were staggered at

altitudes below the maximum attainable altitude to provide

a margin of safety. I f the bomb is more efficient than esti­

mated, these lower-altitude interceptions will still retain their

capability of deflection. The launch schedule for the first 4

flights is so arranged that a failure of 1 of the first 3 vehicles

can be backed-up by the launch of a subsequent vehicle with­

in 12 hr. If a failure did occur, the fourth launch of the series

would be scrubbed. The contributions of launch redundancy

are treated further in chapter 10.

Mission Profile In summary, the nominal mission plan speci­

fies that 6 interceptor vehicles, each carrying a 100-Mt bomb,

will be launched on 6 different trajectories during the period

from 0-72 days to 0-5 days. Each vehicle will be carried to a

100-nm parking orbit by the Saturn V launch vehicle. After

a coast of 1 orbit or less to the proper injection point, the

5-1 VB stage will be restarted, and, together with the ApoJlo

Service Propulsion System, wiJl provide the injection velocity

increment. A coasting period of up to 60 days will follow,

interrupted only by midcourse guidance corrections.

The terminal phase of the mission begins when the optical

Icarus sensor acquires the asteroid 3 hr before rendezvous.

During this terminal period, trajectory corrections are made

to achieve the desired accuracy. The radar system begins to

supply range information approximately 4 min before rendez­

vous, and this information is used in the final correction

maneuvers. At 5 sec before impact, the fuzing radar acquires

Icarus, and the hydrogen bomb is armed. Detonation occurs

within 100 ft of the surface of Icarus, somewhere on the sun-

The Mission Plan

lit edge, and the resulting explosion either fragments Icarus

or deflects it from its collision course.

Reference

23

1. Hollister, W. M., The Mission for a Manned Expedition to Mars, Report TE-4, Experimental Astronomy Laboratory (Cambridge, MA: Massachusetts Institute of Technology, May 1963).

3 Nuclear Detonation and Interaction

Energy Transfer Mechanism

The explosion of the nuclear device in the vacuum of outer space is accompanied by a large release of radiant energy. To qu ote G lasstone (1) :

. . . One important difference between nuclear and con­ventional (or chemical) explosions is the appearance of an appreciable proportion of the energy as thermal radiation in the former case . . . .

At the temperature of a conventional chemical explosion, for example, 5000° K, the radiation energy density is less than 1 erg per cm3 , compared with rough Iy 108 ergs per cm3 for the material energy, Le. , kinetic energy and internal .. . energy .. .. In a nuclear explosion, on the other hand, . . .

some 80 percent of the total energy may be present as radia­tion energy . .. .

Immediately after the explosion time, the temperature of the weapon material is several tens of million degrees, and the pressures are estimated to be many million atmospheres .. .. Within an extremely short time, perhaps a hundredth of a microsecond or so, the weapon residues consist essentially of completely and partially stripped atoms, many of the lat-

Nuclear Detonation and Interaction 25

ter being in excited states, together with the corresponding free electrons. The system then immediately emits electro­magnetic (thermal) radiation, the nature of which is deter­mined by the temperature. Since this is of the order of several tens of million degrees, most of the energy will be in the soft X-ray region.

The primary thermal radiation leaving the exploding weapon is absorbed by the atoms and molecules of the sur­rounding medium. Consequently the medium is heated and the resulting fireball reradiates part of its energy .. .. The remainder of the energy contributes to the shock wave formed in the surrounding medium . ... Ultimately, essentially all the thermal radiation appears as heat, although it may be dissi­pated over a large area. In a dense medium such as earth or water, the degradation and absorption occur within a short distance from the explosion .. ..

There is another mechanism, in addition to the one just described, for the transfer of part of the kinetic energy of the fission fraqments to the surroundinas .... Becausp. of thF! very high pressure within the exploding-weapon, the residue, consisting of fission products and all other weapon materials, moves outward from the center of the explosion at a very high velocity .... After a few microseconds nearly all of the debris is contained in a moderately thin shell of high density called the "hydrodynamic front" .. .. When the hydrodynamic front reaches the ambient medium it acts like a fast-moving piston. Energy is thus transferred to the medium by impulse, and a compression wave, which rapidly becomes a steep­fronted shock wave . .. moves outward.

The energy transport mechanisms are thus radiation and kinetic energy of debris, both of which move outward more or less uniformly in all directions. The amount of this energy deposited on Icarus depends on the solid angle subtended by the target. Assuming Icarus to be spherical, and the explosion effects to be isotropic, fig. 3.1 shows the fractional part inter­cepted as a function of the height of burst above the surface. The importance of a low altitude of burst is clearly apparent. The actual effects of a small elevation of burst above the sur­

face are more closely estimated on pages 30 to 38.

Project Icarus

0.5 <I; !<l 0: <I; ..l 0.4 <I; U ii: "l

0.3 :.: '" '" ..l <I; '"" O. Z � to. 0 ;0: 0.1

� U < � 0

3.1

o 0.1 0.2 0.3 0.4 0,5

RATIO. M/R

Fraction of isotropic emanation intercepted by spherical target

Estimated Destructive Effect

26

Unclassified empirical data reported by Glasstone, Vaile, and

Nordyke, obtained in nuclear tests and high-explosives tests

in Nevada and also in the Pacific Ocean, provide a basis for estimating the damage inflicted on the asteroid as a function

of the size of the nuclear device (2,3, 4) .

When detonated adjacent to or in earth or rock, a megaton­

range explosion produces pressures so high that material strength properties are insignificant in comparison, large

volumetric compression occurs, and hydrodynamic flow re­

sults. The suddenly applied pressure initiates a shock wave

that propagates into the subsurface material as it attenuates.

Expansion occurs at the free surface, and material is conse­

quently ejected, resulting in a crater. The destructive effects

of the pressure wave, evidenced by shattering of the under­

lying rock or earth structure, extend a considerable distance

farther than the "apparent" crater dimensions. The resulting

Nuclear Detonation and Interaction 27

3.2 Typical crater profile

damage profile is shown schematically in fig. 3.2. Here Ra and

� are the apparent crater radius and depth, and H,. is the

depth of the ruptured zone.

For predicting the destructive effects on the asteroid Icarus,

one is concerned particularly with the depth of the ruptured

zone. Glasstone and Vaile indicate that, for very large craters

and near-surface bursts, crater depth is about half the crater

radius (2,3) . Since this ratio probably results in part from

gravitational effects, it is conservative to use it for Icarus. The

negligible gravity also assures that there will be no fall-back and that, even with the sl ightest asteroid 'rotation, most of the shattered material in the ruptured zone will also leave the

crater. Glasstone suggests that the diameter of the ruptured

zone may be estimated at up to 1.5 times that of the apparent

crater (2) . It is assumed here that this same ratio applies in all

directions. Thus

(3.1 )

and

H, ::::: 1.5 Ha. (3.2)

Project Icarus 28

Vaile indicated that the size of the crater may be estimated

by a scaling law of the following type:

Ra = WlIm F('Ac). (3.3)

where W is the weight of the charge in Ib TNT equivalent, and

Ac = (charge depth in feet) /Wl/3. I n this empirical relation,

deduced from experiments with TNT, m takes different values

depending upon the type of ground. The differences between

TNT bursts and nuclear explosions are discussed in detail by

Vaile (3) , and suitable provisions made in the function F(Ac)' Figure 3.3 gives predicted crater size for a surface burst of

either type in parametric form as a function of the soil param­

eter m. Further modification would be necessary if the altitude of burst were appreciable (for example, > 100 ft) .

The methods of Glasstone (1) give somewhat different results,

but Vaile's report is well correlated with empirical data and is more conservative for extrapolation to the multimegaton

range (3) .

Sf-_z -f-� �

1.6

1.4 ci :.l 1.2 f-I>l ::Ii 1.0

;j < 0.8 '" I>l 0.6 N iii 0.4 0: :.l .. 0.2 < 0: U

3.3

fo«' .... ,:,� "

�� " i"

C "

,

�� , «.<0 " fo«' ¢ , ,:,� , "i'� , , it' I

I �� I �<o I ��

,,� v� .. ,,c �

3.0 3.5 4.0

SOIL PARAMETER. m

4.5

Crater-size parameter versus soil parameter (from R. B. Vaile, Jr., "Pacific Craters

and Scaling Laws," Journal of Geophysical Research 10( 1961 1:3413-3438. figs.

10 and 111

Nuclear Detonation and Interaction

o�--�----��--�----�-­I

3.4

10 10 10 TNT EQUIVALENT IIiTI

10

Estimated total crater depth produced by surface burst

29

To be conservative it is assumed that for complete destruc­

tion of Icarus the entire diameter of the asteroid must be

placed equal to the depth of rupture Hr. According to Vaile,

a value of m = 3.6 is recommended for rock and 2.7 for dry

soil (3) . From figure 3.3 one finds for these materials, RalWllm = 0.77 and 0.26, respectively. Then, since H, = 0.75 Ra from

eqs. 3.1 and 3.2, the depth of rupture is shown in fig. 3.4 as a

function of the size of the nuclear device for an assumed

composition of Icarus similar to both rock and dry soil.

Since the asteroid diameter is only approximately deter­

mined between the extremes of 2600 and 5000 ft, and its

composition between that of porous stone and iron, it is

clear that possible bomb requirements for total destruction

extend over an enormous range. While the minimum possible requirements indicate the order of 100 Mt, the most probable

ones call for a bomb exceeding the gigaton class. This so ex-

Project Icarus 30

ceeds the apparent state of the art that it seems too difficult for design and construction within the time available.

Velocity Change of Icarus

The cratering mechanism subsequent to a nuclear explosion may be compared to a rocket engine that converts the energy of the explosion into momentum. If the ejected mass and its velocity are known, impulse and the velocity change of the remaining part of Icarus can be computed.

There are no experimental data available on the velocity field of the ejected mass produced by a nuclear explosion. Experimental data do exist for the case of a rather small TNT explosion, but the physical processes in a TNT burst are in­itially quite different from a nuclear explosion, although later stages of cratering may be similar.

According to Brode, the cratering effects caused by a nuclear explosion are studied analytically using the hydro­dynamic equations of motion and the material equations of state programmed for solution by digital computer (5). This reference was not available to this author nor was it within the scope of this investigation to make a similar computer study. Such an analysis should be conducted, however, with a variety of input parameters covering the expected range of material properties and explosive characteristics in order to obtain accurate prediction of the cratering and the resultant momentum change.

Bjork used similar methods to calculate the velocity field and resultant cratering caused by hypervelocity impact of a large meteoritic mass such as may have produced the Arizona meteor crater (6). The results of such a large-scale hyper­velocity impact and of nuclear explosions are quite similar; in either case initial pressure of tens to hundreds of megabars initiates a spherically-fronted shock wave which results rough-

Nuclear Detonation and Interaction 31

Iy in a paraboloidal crater. One may argue the similarity on the grounds that hypervelocity impact results in the intrusion of a foreign body into the target material, but actually the foreign body is of negligible mass compared to that ejected to form the crater, 12,000 tons compared to 360 million tons for Bjork's example (6). The meteorite kinetic energy is de­posited below the surface of the target, however, which makes comparison better with a subsurface burst. This difference is ignored here, and the velocity field presented by Bjork is used to predict the effect of a surface-type nuclear explosion on Icarus (6).

Attention is restricted to the case of a 100-Mt device, with Icarus considered to behave like the rock in Vaile's article (3), that is, m = 3.6. A strongly conservative assumption is made by considering the velocity distribution data of the source reference, which apply for an event having an energy equivalent to about 1 Mt, applicable for cratering under a 100-Mt bomb, since higher energy would cause higher initial pressure and higher material flow velocities in the early stages of cratering. By assuming the phenomena similar, however, time is made parametric to the process, mass flux and velocity of ejection can be treated as functions of dimension only, and the analy­sis is greatly simplified.

One first calculates the expected size of a 1 OO-Mt crater on Icarus and introduces the height of burst as an additional parameter to be investigated. By extrapolating the curves of figures 10 and 1 1 in Vaile (3), representations of the crater­radius parameter similar to those of fig. 3.3 are found for alti­tude bursts. These are interpreted, as in the case of surface bursts described in the previous section, to obtain the curves of apparent crater radius versus altitude of burst, fig. 3.5. The apparent crater may be assumed to be roughly paraboloidal,

Project Icarus

0: '" t-oo( e

3.5

� 10,000 0: ... !< � .. o .. I, ...

� '\ \ " o�*-:-!�:-+.1I! u 500 1,000 1,500 0 100 1,000 1,100

AL TlTl'DE OF BI:RIIT (FT)

Apparent crater radius produced by 100·Mt nuclear burst

32

according to Glasstone (2) , and the volume Va can be com­

puted by

(3.4)

To establish the velocity change of Icarus, one must deter­

mine the following integral:

Jo If! v(pv'ds) } dt, (3.5)

where p is the specific density of the ejected particle, and v is

the velocity of the ejected particle. The integration is made along a control surface which is sufficiently distant from Icarus so

that pressure is, for practical purposes, zero. The size of the

crater is already known, but the time history of the mass flux

and the velocity, or at least the mean velocity of ejected

material, must also be known. These data are deduced from

Bjork (6) as follows:

1. Figure 3.6 is a typical pressure-field versus velocity-field

plot as presented by Bjork (6) . From this and similar plots

of Bjork, the vertical component of the flow velocity of the

uppermost surface may be obtained as a function of distance

from the center of impact, r. These relations are shown in

Nuclear Detonation and Interaction

3.6

Dt:P'fII 1M 1'//,,/// f,,,,,, GROl'NDSt)RFACE O"" "I"�

t, #" .. ...

t . .. . . . .. ..

"-"'I " •• ... \·ELOCITY SCALE: ... , .... . o Ii JaI/S .. :C �

ItAUIAL DISTANCE 1M)

33

Representative pressure and velocity field at 24.8 msec (reproduced from R. L. Bjork, Journal of Geophysical Research 66(19611:3379-33871

till 24. II

t. 61

o � __ � ____ � __ -4 ____ � __ � ____ � __ � ____ -L ____ L-__ � 20 40 60 80 100 120 140 160 \80 200

RADIAL DL�TA"C" • (M)

3.7 Upward velocity of ejecta

Project Icarus 34

fig. 3.7. The ejected particles are accelerated by the pressures produced by impact, and the velocity profiles shown in fig. 3.7 represent conditions at or above ground zero, after al­most all acceleration has taken place and the pressure has become nearly zero. 2. To compute the quantity of eq. 3.5, one must also know the mass flux distribution as a function of time. Assume that the elemental mass flux dm = (pv) dA is constant over the entire region of the control surface where the ejected particles pass at any instant of time. This is partly justified because the pressure at the control surface is uniformly low. The value of the mean vertical velocity component of the ejecta is then derived from the data of fig. 3.7 on the basis of

_ fVydm =

21T (pv) fvyrdr v: = y fdm 21T(pv}fr dr

(3.6)

The product (vyr) is therefore plotted in fig. 3.8 as a function of r for the several stages of cratering. These curves are inte­grated and divided by total area to obtain mean velocity as a function of time in fig. 3.9. This curve is fitted with good accuracy by the expression

v = (v) e·th km/sec y y o , (3.7)

where (vy)o = 4.1 km/sec, and T, the characteristic time for this crater = 37.2 msec. 3. From figures 5 through 10 of Bjork (6), data on the prop­agation of the radius of the crater with time can be obtained as shown in fig. 3.10. The final radius of the crater in (6) is 500 m. By extrapolating the curve of fig. 3.10, it is estimated that the time required to complete the crater, T, is approxi­mately 180 msec. The mean velocity data of fig. 3.9 can also be extrapolated out to 180 msec using eq. 3.7 to obtain a

Nuclear Detonation and Interaction 35

o ... '00 , .. 200

JtADlAI� IXSTAN<':l:. r (IU

3.8 Normalization of velocity distribution

.--

o 10 20 30 40 50 60 70 TIME (MSEC)

3.9 Mean upward velocity of ejecta

Project Icarus

600

3.10

10 20 50 100 200 TIME (MSEC)

Propagation of crater radius with time

36

final value of 32 m/sec, corresponding to TIT = 180/37.2 = 4.84. 4. The ejection rate is now assumed to vary linearly with time for the whole period of cratering. Figure 3.10 indicates this to be a good approximation, at least in the early stages of cratering; from that figure, r varies closely as to.67• If the crater remains geometrically similar as it develops, the mass m will therefore vary as r3 or t2•0, and the mass flux m as tLO• Even­tually rh must of course go to zero, but to assume rh increas-ng less rapidly or diminishing leads to larger calculated mo­

mentum. It is therefore conservative to take the mass flux according to

m· = k ! for 0 <1. < T T T T '

whereupon the total ejected mass is

The total momentum, using eqs. 3.7 and 3.8, is

foTh rhvyd�) = k(vy )o foTh(; e-t/T)d�),

or finally,

(3.8)

(3.9)

Nuclear Detonation and Interaction 37

(3.10)

Dividing eq. 3.10 by eq. 3.9 gives the overall mean velocity:

(- ) - � {1 - ( 1 + I) - TIT} Vy overall - (T/r)2 T e • (3.11 )

Using k = 4.1 km/sec and T/r = 4.84 as determined previously, (vy ) overall = 333 m/sec. 5. Finally, to correct for the difference between apparent and true crater, it is assumed that the size of the true crater is 1.25 times the size of the apparent crater and that this mass is ejected with the velocity at the time t = T. The total mo­mentum imparted to Icarus is therefore calculated as follows:

jVydm=pVa (333)+ (1.253-1)pVa (32), (3.12)

where Va is the apparent volume (eq. 3.4) determined using the crater dimension from fig. 3.5 and depth equal to half the radius (eq. 3.1 ).

By these procedures, the momentum change of Icarus pro­duced by the detonation of a 1 OO-Mt nuclear device is de­rived as a function of the height of burst. The final result in terms of velocity change LlV, assuming m = 5 X 1015 g, is shown in fig. 3.11. The following conclusions are evident: A low altitude of burst is very important in achieving a perturbation of Icarus' orbit with a nuclear explosive; For a 100-ft altitude of burst, the impulse imparted by a 100-Mt burst to a nominal Icarus model is conservatively estimated to introduce a velocity of about 8 m/sec. Numerous con­servatisms were introduced in making this analysis; on the other hand, the conditions considered do not include the complete range of possible characteristics of Icarus. The value of 8 m/sec may be compared with the upper limit which would be possible if 50 percent of the total energy of the

Project Icarus 38

nuclear bomb were intercepted by Icarus and converted into

kinetic energy of motion; in this case, with 100 Mt correspond­

ing to 4.2 X 1024 ergs, and with a nominal asteroidal mass of

5X101Sg,

[0.5 E] % Llv =

% m = 290 m/sec.

The Nuclear Device

(3.13)

The determination of the weight and size of the nuclear

device would require considerable highly specialized knowl­

edge and experience, and a precise estimate is beyond the

possibilities of a theoretical approach based on the extremely

limited technical data available in unclassified literature. Reliable results could be obtained only by a careful and

systematic accumulation of experimental data regarding

values of critical variables from explosions of less energetic devices. Thus limited by circumstances, the best estimate

that cou Id be made on the basis of an examination of the un­

classified literature is shown in fig. 3.12.

3.11 Estimated velocity change of Icarus caused by 100-Mt burst

Nuclear Detonation and Interaction

� I� -I

II�D�I 'l"It:LD 100 NT \\'EIGIIT �', 000 La IlI.1TlATlnlC ELt:CTRI(,AI.

3,12 Estimated nuclear explosive payload

Auxiliary Systems

39

I n addition to the nuclear explosive, a practical nuclear de­

vice requires fuzing, safe"arm, destruct, and packaging systems. Preliminary considerations for the design of these systems are

based in part on Pollard's and Arnold's work (7) . The recom­

mended use of a nuclear explosive for the present purpose in­

volves considerations other than technical ones, and the

final decision to proceed will depend in part on the assurance

of utmost safety and reliability. These characteristics depend

upon the design of the auxiliary systems.

Fuzing System Delays and uncertainties involved in the

fuzing elements together with the relative velocity on closing

with Icarus determine the range at which the nuclear device

must be triggered. The minimum distance is established by

the requirement that impact not occur prior to detonation.

If this were permitted, the high impact velocity might cause

the nuclear device to break up in a time much shorter (of

the order of 0.01 msec) than the detonation triggering time (of the order of 1.0 msec).

The estimated uncertainty in fuzing time of ±O.4 msec re­

sults in a required detonation height of 50 ft at a nominal

closing velocity of 125,000 fps, to prevent impact prior to

Project Icarus 40

(SAFE) I IOESTRU1

3.1 3 Nuclear fuzing system

detonation. Detonation may actually occur then as high as

100 ft. Because of the fuzing delay time estimated at 4 msec,

the radar triggering signal must be given at a range of 550 ft.

The elements of the fuzing train are shown schematically

in fig. 3.13. Radar determines the range and range-rate to

Icarus, the fuzing radar being gated to trigger the firing unit

either at the predetermined range of 100 ft or (in the event

of a near miss) when the range-rate goes to zero. The firing

unit sends an electrical signal to the initiator, which sets off

an exploding bridge wire (EBW). The EBW is surrounded by

a small amount of fulminate of mercury, which detonates.

The safe-arm device either interrupts the detonation train

(safe) or permits the detonation to proceed unimpeded (arm) .

If the former occurs, the nuclear device is neutralized until it

is destroyed by the destruct device; in the latter case, however,

it detonates and releases its energy.

Safe-Arm System It is imperative that the nuclear device

not detonate accidentally. It is therefore necessary to incor­porate a fail-safe safe-arm device that positively renders the nuclear device inactive until detonation is desired. The many

redundancies in the safety system lead unavoidably to some

decrease in the reliability of arming, but by emphasizing

ruggedness and simplicity one can obtain an extremely reliable

safe mode with only a small loss in the arming reliability.

Nuclear Detonation and Interaction 41

The heart of the safe-arm system is a rotatable disc or gate that interposes a physical barrier and prevents detonation of the initiator. I n the safe position the path is blocked; but upon satisfying an arming criterion the disc is rotated, a hole is aligned, and the fuzing train is allowed to proceed. Multiple mechanical or electrical gates may be used, responding with suitable time delays to such arming criteria as vehicle accelera­tion, ambient pressure, target acquisition signal, and earth command.

Nuclear Destruct System It is necessary to include a destruct system in the nuclear device so that it may be destroyed and dispersed in a nonnuclear manner if required. Criticality re­quires that the fissionable material be kept separated. Destruc­tion can be accomplished by detonating a conventional ex­plosive in such a manner that the physical separation between

fissionable material is increased. This is primarily a geometrical prQblem involving the internal configuration of the nuclear explosive.

The destruct charge is electrically initiated either by a coded signal from the ground or by a signal generated from within the spacecraft upon mission failure. Such failure is indicated, for example, by excessive time beyond the nominal interception.

Packaging As a provision against damage on the pad or during suborbital abort, the nuclear device and associated subsystems are surrounded with a minimum 3-in layer of shock and thermal insulation and packaged in a structural shell container of O.5-in stainless steel. The weight of the packaging system is estimated at 4000 lb. Installation of the device in the space vehicle payload stage is made by bolted attachment of end plates on the stainless steel container to

Project Icarus 42

longitudinal mounting flanges on the payload stage shell structure. The payload stage shell, the stainless steel housing and insulation, and the nuclear device itself are all considered to contribute to the performance under extreme conditions. Refinement of this approximate design requires detailed in­formation regarding the characteristics of the device itself.

References

1. Glasstone, S., ed., The Effects of Nuclear Weapons (U.S. Atomic Energy Com­

mission, U.S. Government Printing Office, 1962). paragraphs 1.20, 1.71,2.98-2.101.

2. Ibid., paragraphs 6.04-6.11, 6.46-6.48, and Appendix B.

3. Vaile, R. B., Jr., "Pacific Craters and Scaling Laws," Journal of Geophysical

Research 10(1961 ):3413-3438.

4. Nordyke, M. D., "Nuclear Craters and Preliminary Theory of the Mechanics of

Explosive Crater Formation," Journal of Geophysical Research 6611961): 3439-3459.

5. Brode, H. L., and R. L. Bjork, "Cratering from a Megaton Surface Burst,"

Rand Cor. Res. Mem. RM-2600 (June 30, 1960).

6. Bjork, R. L., "Analysis of the Formation of Meteor Crater, Arizona: Pre­

liminary Report," Journal of Geophysical Research 66(1961 ):3379-3387.

7. Pollard, F. B., and J. H. Arnold, Jr., Aerospace Ordnance Handbook (Englewood

Cliffs, NJ: Prentice-Hall, 1966).

4

Launch Systems

Launch Vehicle Capabilities

Payload capabilities of the "state of the art" launch vehicles

are summarized in fig. 4.1. These data, derived from various

reports, illustrate the necessity of using the Saturn V as the

primary launch vehicle for this mission (1,2,3). Even for a

mission requiring no more than the 1 O,600-fps velocity in­

crement from parking orbit that would just produce escape, it is clear that the Saturn I B and the Titan I I I C, the next

smaller launch vehicles than Saturn V, provide payloads far

too small for the current requirements. Various methods of using the Saturn V that were considered

for this mission are shown in fig. 4.2. The rendezvous of

several Saturn V's in an earth parking orbit offers the maxi­

mum payload possibilities. This method would require only

minor modifications of the present Saturn V launch vehicle

and associated launch complex. Docking hardware and tech­

niques would have to be developed, however, and a space

tug wou Id be needed to perform the actual docking maneuver.

Project Icarus

4.1 Large booster capability

SI'Ae.: Vt:IIICU·:

UUOST .·RUM luu·".M (HtBlT

JI' KlST TO ORBIT

4.2

1'/1. 0>

S-IVB S-IVB P/L It

P/L

S-IVB

S-ll

S-IC

I I>

S-IVO

s-u

S-IC

SATURN V Rt:NDF.ZVOUS

Possible launch systems

P/L lVM P/L II::. C»

S-IVB CENT P/L S-IVB 81M P/L � Ii I I> Cl>

APOLLO

S-IVB

!HI

S-IC

SATURN VI ct:NTAUR

SIM 8-IVB

!P-II

S-IC

SATURN VI APOLLO

44

Launch Systems 45

For the tug a modified Apollo Service Module launched by a

Titan I I IC would be a likely candidate. A major constraint is

introduced by the 6-hr on-orbit life of the Saturn IVB stage.

To increase the time available for orbital rendezvous and

departure, modifications would be necessary in the power

supply, the attitude stabilization modules, and the fuel tank

insulation. Payload packaging for on-orbit assembly and

staging inefficiencies also constitute disadvantages that must

be considered before a rendezvous mode is selected. Evalua­

tion of the orbital rendezvous mode, as shown in fig. 4.1, did

not include such penalties but was based on the performance

of the existent Saturn V hardware, as given in a Douglas re­

port (2). Payload versus AV from orbit vyas calculated using

the mass ratio expression

MR = e 1.05�v/ve, (4.1 )

where M R is the mass ratio, and Ill, the exhaust velocity, was

taken as 14,100 fps. This formula includes a 5 percent penalty to account for parking-orbit degradation and docking losses.

The payload was obtained from the mass ratio definition

MR = Winitial __ payload + S-IVB + IU + DC + propellant

wfinal payload + S-IVB + IU + DC (4.2)

Table 4.1 gives weights of the components considered in this

analysis. Despite its large technical advantage, rendezvous is not a

preferred launch mode for the Icarus mission because of the

brief time period and limited flight resources available for the

development of such advanced operations capability. Other

possible launch systems include the Saturn V /Centaur combin­

ation, and the Saturn V in combination with the Apollo

Service Module. The performance of these systems, as well

as that of the Saturn V alone, is compared in fig. 4.3. This

Project Icarus

Table 4.1 Saturn V orbital rendezvous weights lib)

Saturn V payload to 10o-nm orbit (in addition to empty 5-IVB and IU)

S-IVB stage weight, empty (includes 2841 Ib residual propellant)

5-1 VB usable propellant capacity !includes boil-off)

Allowance for propellant boil-off in orbit

Instrument unit (I U)

Docking collar (DC) (estimated)

Aerodynamic nose cone for Saturn V

46

261,000

28,549

230,000

3,495

4, 150

3,000

3,600

display was derived from the performance data of the Douglas

Report for coplanar direct ascent from the Eastern Test Range

at an azimuth of 60°, which makes possible an orbital inclina­tion of 40° but is not penalized by a dog-leg requirement for

range safety (2). In the case of the Saturn V/Apollo Service

Module, the total payload for the 3-stage S-IC/S-I I/S-IVB

booster includes a 4000-Ib shroud, which (as for the Apollo lunar mission) is jettisoned only when the spacecraft is

separated from the S-I VB. The additional velocity incre-

ment available from the Service Module was computed as­

suming an inert weight of 11,089 Ib and 40,000 Ib of pro­

pellant, with specific impu Ise Is = 319 sec. All payload systems require midcourse maneuvering capa­

bility, If the launch system employs an S-IVB or Centaur

upper stage, the payload package must include some provision

for this requirement: either an additional propulsion system

utilizing storable propellants for midcourse maneuvering, or

some system to prevent excessive loss of cryogenic propellants

by boil-off. This additional complexity is taken care of with

the Saturn V/Apollo Service Module combination, since the Service Module utilizes storables and is designed for multiple

restart.

Launch Systems 47

110'

100 / 90

a ..l

� 80

Cl < :3 10 � INST U. < :I. EMPTY

S-IVB

60

50

40

9 10 11 12 13 14 15 16 AVAIL.ABLE VELOCITY INCREMENT FROM 100-NM ORBIT. /l.V (103 FPS)

4.3 Performance available with various Saturn V launch systems

Project Icarus 48

Launch Vehicle Availability

Current Development and Production The ground test pro­

gram of the Saturn V was still going on at the time of this

study, but was due to be completed by the middle of 1967,

when the first flight test was expected to take place. Fifteen

vehicles (SA-501 through SA-515) were scheduled to be de­

livered to the Kennedy Space Center between November 1966 and November 1969. At that rate, 6 vehicles would have been

produced by June 1968, the first few of which would have been devoted to reliability flight tests (4,5). Thus if allowance had been made for the use of perhaps the first 3 for th is pu rpose, only 3 vehicles would have been available to Project Icarus by June 1968. However, mission proposals call for up to 6

launch vehicles in order to achieve a suitable probability of

success.

Proposed Emergency Production A starting assumption was

made that the problems of Icarus present an emergency situa­

tion in which resources may be diverted from other national

space programs. This assumption was invoked, therefore, to

enable the utilization of all the vehicles resulting from the

then current Saturn V production program. I n addition,

"emergency" powers were applied to step up production

through increases in work force and use of 3-shift work

schedules. The production schedule expected to be realized

is compared with the schedule at the time of the study in

fig. 4.4.

Launch Facilities

The evolution of the mobile launch concept of NASA Complex 39 was dictated by the size and complexity of the Saturn V

Launch Systems

U VEHICLE I EXISTING SCHEDl'LE � /""

12 PRODUCTION PROPOSED SCHEDUL7 ").. ........ 10 �

//

2 --'

� ......... ----,.. LAt:NCH PAD AVAlLABIUTY

.-- ' JFMAMJJASONDJFMAMJJASONDJFMAMJJASO

196; 1968 1969

4.4 Vehicle production and launch pad availability

49

vehicle as well as by the frequency of scheduled flight tests.

The principal features of this complex include a 4-bay vertical assembly building, mobile launchers which support vehicles undergoing assembly, a transporter to convey the vehicles to

the launch sites, a mobile service structure for on-pad check­out and servicing of vehicles, and a launch control center (6, 7).

With 4 launch pads an estimated 75 launches per year can be

achieved on Complex 39. However, the ground support capa­

bility for 4 pads did not exist at the time of the study; in fact,

only 2 pads were under construction. The overall launch capa­

bility at that time on Launch Complex 39 is summarized as

follows (8):

Pads

1. Pad 39A was then ready, but pad 39B would not be ready

until late summer (1967); a third pad could be constructed

within a year.

2. Except for limitations on personnel, operations on one

pad did not affect operations on another pad.

3. Launches could take place from 2 pads within a few hours

of each other; but simultaneous tracking facilities for 2

vehicles were not available at the time.

Project Icarus 50

4. On-pad time was 10 to 11 days, independent of whether or

not a vehicle was man-rated.

5. The expected time for refurbishment of a used pad was 2

weeks or less, based on experience at the other NASA launch

complexes.

Vertical Assembly Building (VAB)

1. Assembly of a manned vehicle in the VAB took 3 months; a shorter time would be required for unmanned vehicles, which

would omit some phases of the VAB tests. 2. Only 3 of the 4 high bays were then fully equipped. The

entire VAB cou Id be horizontally extended to provide for a

maximum of 6 high bays.

General

1. There were 3 mobile launchers, 1 mobile service structure

(MSS), and 2 transporters.

2. The MSS must be at the pad with a launch vehicle until

7 hr before launch.

3. In the event of interruption on countdown, there was a

maximum hold time of 12 hr on personnel and 24 hr on fuel.

Beyond these limits, some phases of earlier checkout and

servicing had to be repeated.

As compared with the above launch capability, the Icarus mission requires a total of 6 launches over a period of 10

weeks at a launch rate of 1 vehicle every 2 weeks. This launch

rate requires up to 3 launch pads. If one assumes an assembly

time of 8 weeks for an unmanned vehicle, as many as 5 vehicles

at various stages of assembly may be in the V AS at the same

time. This, in turn, requires 5 mobile launchers-1 for each

vehicle. It was assumed that, with acceleration of the pace of

construction, 5 mobile launchers, 4 high bays, and 3 pads could be fully constructed and equipped by April 1968. The

estimated launch pad availability schedule is indicated in

fig. 4.4

Launch Systems 51

References

1. Douglas Report SM-47010, "Saturn IB Payload Planner's Guide" (June 1965).

2. Douglas Report SM-47 274, "Saturn V Payload Planner's Guide" (November

1965).

3. Martin Company, "SSLS Definition for Payload Contractors," SSD-CR-65-18 (Rev. 1) (September 1965).

4. Bramlet, J. B., "Saturn V Launch Vehicle Development Program," AIAA Paper 64-271 (July 1964).

5. O'Connor, E. F., "Saturn V Launch Vehicle Report," AIAA Paper 66-840 (December 1966).

6. Petrone, R. A., "Apollo/Saturn V Launch Operations," AIAA Paper 66-837

(December 1966).

7. NASA Kennedy Space Center, "Launch Complex 39 Facilities," Fact Sheet 03

(November 1966).

8. Telephone discussion with Mr. Robert E. Johnson, Protocol Officer at NASA Headquarters, Kennedy Space Center, Florida, April 1967.

5 The Icarus Spacecraft

Space Vehicle Design Considerations

The Project Icarus mission objectives and requirements place several constraints on space vehicle design, most important of which is the overall time limitation of 60 weeks from project initiation to first launch. Yet a completely autonomous space vehicle is required compatible with the Saturn V launch system and providing a high degree of reliability. These cir­cumstances demand maximum utilization of existent Saturn hardware with modifications held to an absolute minimum, even when nonoptimum for this mission, in order to avoid lead time required for development, tooling, and qualification testing. Additionally, it is clear that although the missions of the 6 interceptors vary appreciably, only 1 design can be con­sidered, since the production learning curve must be exploited. Still another important factor is the space flight duration of as much as 60 days, particularly in the selection of the pro­pulsion system for the space vehicle.

The Icarus Spacecraft 53

5.1 Icarus spacecraft at S-IVB separation

Spacecraft General Configuration

Figure 5.1 shows the spacecraft at separation from the Saturn upper stage (S-IVB) . The 4-panel adapter-shroud remains with the spent S-I VB. Visible on the spacecraft are the primary and attitude control propulsion systems, the high-gain com­munications antenna at the base of the vehicle, and the phased­array radar antennas mounted on the forward body external surface. The vehicle is comprised of 3 basic sections: the com­mand module, the payload module, and the propulsion or service module. A profile is given in fig. 5.2. The weights of the various components are summarized in table 5.1. The following sections examine each of the components in detail.

Booster Adapter

The S-I VB-to-spacecraft adapter is a modified Apollo LM

Project Icarus

5.2

S-IV8 STA(a:

General spacecraft configuration

adapter which, as used here, supports the Icarus spacecraft

54

at the forward ring of the payload module and acts as a shroud for both the payload module and the service module. This type of adapter-shroud, although heavier than a short inter­stage supporting the bottom of the spacecraft, provides better booster aerodynamics, eliminates aerodynamic loads and aerodynamic heating as considerations in the design of the slit antennas which are mounted on the payload module, and is readily available.

The LM adapter, which is used in the Apollo moon vehicle to house the lunar module, is structurally and functionally adequate for launching the Icarus spacecraft, except for the addition of internal bracing to provide lateral support for the lower end of the enclosed spacecraft. This adjustable bracing, which bears against the lower ring structure of the service module, is indicated in figs. 5.1 and 5.2. A weight allowance of 500 Ib, in addition to the weight of the LM adapter, is made for this modification, bringing the total to 4000 lb. This weight constitutes part of the booster payload, remain­ing with the launch vehicle until spacecraft separation.

The Icarus Spacecraft

Tlble &.1 Spacecraft waight summary lib).

Propulsion module at burnout (max)

Payload module

Command module

Contingency

Spececreft total at burnout

Propulsion Module usable propellant

Gross waight of spacecraft

Booster adapter-shroud

Payload weight for Saturn booster

Propulsion Module

11,200

45,000

2,200 1,600

60,000

55

60,000

40,000

100,000

4,000

104,000

The primary objective of the spacecraft propulsion system is to provide capability for midcourse and terminal guidance maneuvers. However, inasmuch as the Saturn booster does not provide sufficient launch velocity for injection into the Icarus intercept trajectory for boosted weight exceeding 64,000 Ib, it becomes desirable to be able to use the space­craft propulsion system for part of this operation. A multiple­restartable engine of considerable thrust is required in any case.

Cryogenic upper stages, exemplified by the Centaur and perhaps by a cut-down or off-loaded S-IVB, offer the best performance. Both of these stages have only single-restart capability; however, multiple-restart capability sufficient to meet the needs of the Icarus mission is a conceivable modifi­cation. Additional development and modification is required, however, in coping with the problem of propellant storage during the 60-day space mission.

Cryogenic propellants used in the S-IVB and the Centaur are subject to boil-off losses due chiefly to solar heating which, in 100-nm orbit, for example, amount to about 8 per­cent of the propellant tank capacity per day. Thermal insula-

Project Icarus 56

tion, thermal control surfaces, shadow shields, supercooling of propellant prior to launch, and heat pumps have been in­vestigated as means of reducing boil-off. I n the case of the Centaur, it appears that a combination of advanced thermal insulation (modifying the existent external insulation panels which are jettisoned during launch) and surface treatment would provide for a 60-day space mission at a total weight penalty (boil-off plus insulation) of no more than 5000 lb.

Such developmental programs are not desirable, however, within the Icarus timetable if alternative modes are available. Several multiple-restartable upper stages using storable pro­pellants were therefore investigated, and it became apparent that the Icarus mission could be met in several different ways. A comparison of these alternatives, based on the final Icarus payload data, is presented in table 5.2. The Apollo service module, although further from optimum stage size for this mission than either the Transtage or the Agena, provides more than the required 3-a guidance maneuvering capability and, in addition, avoids the problem of integration of struc­ture, launch checkout, and flight systems with the Saturn V launch system, which might prove to be serious with the limited time available. It may also use an existent interstage structure.

General Description The external configuration of the propulsion module is shown in fig. 5.3; a component system weight breakdown is presented in table 5.3. Except for minor modifications to the structure and electrical system (described later), this unit is identical with the Apollo service module (1, 2, 3). It measures 12.8 ft in diameter and 22 ft in over-all length and is divided internally into 6 sectors, 2 of which hold the oxidizer tanks, 2 the fuel tanks, and the other 2

auxiliary equipment.

The I carus Spacecraft 57

Table 5.2 Performance of candidate propulsion stages for Icarus spacecraft

Apollo Cen- Transtage N10.Agena SM taur* Full Off-Loaded Full Off-Loaded

Stage inert 11,089 10,000* 4,150 1,500 vveight lib)

Usable propellant 40,000 30,000 22,900 6,939 13,000 (lb)

Payload lib) 48,911 48,911 48,911 48,911

Initial wt. lib) 100,000 88,911 75,961 60,000 63,411

Weight at burn- 60,000 58,911 53,061 50,411 out lib)

Mass ratio 1.67 1.51 1.43 1.13 1.25

Specific impulse, 319 440 305 320 Is (sec)

Spacecraft A V (fps) 5,240 5,800 3,500 1,200 2,350

Booster adapter 4,000 4,000 4,000 4,000 vveight allow. lib)

Booster (S·ICI 104,000 92,911 79,961 64,000 67,411 S·II/S-IVB) payload Ob)

Booster Av above 9,800 10,800 12,300 14,200 13,700 1 OO-nm orbit (fps)

Injection Av from 4,400 3,400 1,900 0 500 spacecraft (fps)

Av available for 840 2,400 1,600 1,200 1,850 midcourse (fps) * with necessary modifications

Table 5.3 Weight breakdown for propulsion module (Ib)

Structure (Apollo SM, 2,633 Ib) + modifications (111 Ib)

Environmental control system

Service (main) propulsion system

Usable reaction control system propellant

Electrical power system (1,555 Ib) + usable fuel (2,070 Ib)

Fluid residuals (718 Ib) + miscellaneous (345 Ib)

Maximum weight at burnout

Usable main propellant

Total weiQht at lift-off

9,589

60,000

1.19

1,790

64,000

14,200

0

1,790

2,744

70

2,908

790

3,625

1,063

11,200

40,000

51,200

Project Icarus

ruEL CELL POWER PLANT (3)

OXIDIZER TANK

SERVICE PROPULSION ENGINE

ruELTANK �

5.3 Propulsion module

The forward portion of the shell structure of the Apollo service module will require modification to replace fairing with a structural skin and attachment flange for mounting

58

the payload module. Existent radial beam trusses which would support the Apollo command module may be removed.

Main Propulsion System The main propulsion system utilitizes N204 and Aerozine 50 in a OfF ratio of 2: 1, to produce a fixed thrust level of 21,900 Ib with a specific impulse of 319

sec. The gimbaled thrust chamber is radiation cooled and is rated to handle more than the total SM propellant capacity in a single burn. The propellants are pressure fed by helium gas which is stored in spherical pressure vessels at 4000 psi. A capillary system utilizing surface tension of the liquid pro-

The Icarus Spacecraft 59

pellants collects sufficient fuel and oxidizer at the bottom of

the tanks to permit engine start in gravity-free space. The

propellants are hypergolic, and as many as 50 restarts are

possible.

Important to the accuracy of the guidance in the Icarus

mission is the precision of engine start-up and shutdown.

Since guidance maneuvers are controlled by the on-board

computer which generates commands based on data from

the IMU, information about the start-up transient will auto­

matically be introduced into the calculation of the shut­

down command.

The shutdown transient is more difficult to take into ac­

count (4). With a closed loop guidance scheme the engine is cut

off when the required vector velocity correction becomes

zero. When the engine is cut off, thrust decreases sharply as

shown in fig. 5.4. But because of propellants in the ducts

downstream from the shut-off valve, some residual thrust

appears. This residual thrust can be determined empirically

rather accurately, and its effect reduced by cutting off the

engine shortly before the required velocity error becomes

zero. But, of course, there is a certain amount of random

variation and thus a velocity error.

Another source of error is cut-off timing (fig_ 5_5). Usually

the guidance equation is solved about once every second, that

is, in fig. 5.5, t2 - tl � 1 sec_ The required velocity error be-

II�:S""'AI. THRIJST

5.4 Shutdown transient

Project Icarus 60

5.5 Cut-off timing error compensation

comes zero at time t8, but this is not known until the next

sampling time, t2 • If the engine is cut off at t = t2, there will

be a large velocity error. Therefore, in most cases, the follow­

ing technique is adopted:

1. From curve OA, the point where the required velocity error

becomes zero (point 8) is estimated.

2. When point 8 falls between tl and t2, the sampling period

is subdivided into subsampling periods, and the engine is cut

off at time tc, the next subsampling instant after time t8.

These engine cut-off transients, even with the use of the re­

fined command techniques mentioned above, lead to impulse

errors on the order of 2000 Ib-sec. These errors may be re­

duced to the order of 20 Ib-sec by utilizing the reaction con­

trol thrusters as a vernier system.

Reaction Control System (RCS) The RCS provides vehicle

attitude stabilization to keep radar and optics properly

oriented during the trans-Icarus flight. Also, upon command

of the guidance and navigation system, it reorients and sta­

bilizes the vehicle for thrusting maneuvers. One second after

ignition of the main engine, the pitch and yaw reaction con­

trol engines are disabled, and pitch and yaw are controlled

by gimbaling the main engine. Roll is controlled by reaction

control rockets at all times. One second after main engine

The Icarus Spacecraft

cut-off, the gimbaling system is disabled, and the control mode is switched back to the bang-bang operation of the RCS.

61

The reaction control system consists of 4 independent clusters of 4 rocket engines, each mounted at 90° intervals around the sides of the service module (fig. 5.3). Each cluster contains 2 roll engines and 2 pitch or yaw engines, plus fuel, oxidizer, and helium gas pressurant tanks. The hypergolic propellants are the same as those used in the main propellant system. Each engine provides 100 Ib of thrust, with a minimum impulse of 0.6 Ib-sec. The inde­pendence of the propellant supplies of the 4 clusters pro­vides a measure of reliability for the system in that if 1 en­gine failed to open it would react locally and would drain only a single cluster. Completion of the mission might there­fore be accomplished with the other 3.

Electrical Power System The electrical power system con­sists of fuel cells, storage batteries, and power conditioning equipment. Three Bacon-type fuel cells, each consisting of 31 single cells connected in series to obtain a nominal 28 volts, utilize oxygen and hydrogen under regulated pressure, and produce heat, water, and electricity.

The operating range of 1 of these fuel-cell power plants is 500 to 1420 w. The voltage is dependent on the power level; that is, 31 v at 500 w to 27.5 v at 1420 w. Regulators are in­cluded to provide 28 v DC. The life of these fuel cells is limited to about 2 weeks; however, a special coating applied to the cell plates and an improved activation technique are available to control internal deposits and extend the rated life to 60 days.

Hydrogen and oxygen for the power plant are stored in the supercritical cryogenic state. Fuel consumption depends

Project Icarus 62

primarily on power output. From the standpoint of reliability, it is desirable to put all cells into operation just prior to launch, although 1 fuel cell could provide sufficient power for the mission. At a specific fuel consumption of 0.85 Ib per kw-hr with a suitable allowance for flushing, the space­craft must carry 2070 Ib of fuel. The Apollo service module includes provision for storing this amount of fuel.

Two silver-zinc oxide storage batteries are included to pro­vide emergency power. (The normal complement for Apollo is 3.) These wou Id prevent loss of the mission in case of fuel­cell failure of a temporary nature or during the terminal phase. These storage batteries are located in the command module. Power output is 3000 w-hr per battery.

A DC-AC converter is included to provide 3-phase 400-v AC power.

Environmental Control System The environmental control system provides cooling for the electronics, IMU, radar, and computer. A mixture of water and glycol is continuously circulated through cold plates on which the electronic equip­ment is mounted. Heat absorbed by the fluid in the cold plates is transferred to radiators located on the outside of the service module, where it is dissipated into space. Heat loads are much lower than those encountered in an Apollo mission, due to the minimal electronics carried, and the lack of heat­generating life-support equipment.

Payload Module

The payload module is a stiffened cylindrical shell, 5 ft long and matching the Apollo service module in diameter, which houses and supports the nuclear device. Construction em­bodies 0.060-in aluminum alloy skin, stiffened with 0.060-in aluminum alloy trapezoidal corrugation, with extruded

The Icarus Spacecraft

!\ADAK PHASED-ARMY SLIT ANTENNA

6.6 PaVload module

TIlbI.6.4 Weight breakdown for pavload module UbI Nuclear pavload (including packaging)

Stage structure

Skin and corrugation

End rings

Mountfng stiffeners

Fasteners and misc.

Radar

Icarus navigation antenna

Earth navigation antenna

Cabling, plumbing, and miscellaneous

Total

63

44,000

450

100

100

50

700 700

185

15 --

200 200

100

45,000

Project Icarus 64

aluminum alloy rings at each end. Details are shown in fig. 5.6 and a weight estimate is given in table 5.4.

Around the curved external surface of the payload module are mounted the phased-array Icarus tracking radar antenna and earth navigation antenna. Electrical power, engine con­trol wiring, and environmental control conduits pass through the payload module, connecting the command and the pro­pulsion modules.

Command Module

Most of the functions of the Icarus command module could be performed by the integrated systems comprising the Apollo command module, or by the lunar module ( LM). However, the Apollo command module also includes reentry heat pro­tection systems and life-support systems which would result in an excess weight penalty for the Icarus mission of the order of 8000 Ib and for the LM perhaps 5000 lb. Most of the excess weight resides in the structure and in the basic design arrange­ment originally generated to meet requirements imposed by manned flight, such as cabin pressurization, reentry protection, visual reference, access, and life support. Modification of either of these modu les, either as a design change or by strip­ping the manufactured units, is a large and complicated opera­tion, causing much disturbance to other systems, and removing only a fraction of the excess weight.

The preliminary design for the Icarus command module is therefore based on a simple structure using some of the ex­istent Apollo tooling, and on selected Apollo communications and control systems fulfilling the needs of the Icarus mission.

Physical Description The command module houses the guidance electronics and optics required for the interception

The Icarus Spacecraft 65

mission furnishing environmental protection on the pad and during launch and also providing aerodynamic fairing for the vehicle. Similar in shape and size to the Apollo command module, it is permanently joined to the upper end of the pay­load module, I nstallation and ground checkout of internal equipment during prelaunch is available through a bolted­down access hatch. Optics and fuzing radar operate through a smaller hatch at 1 800 to the ground-access hatch, and from the nose. The optics hatch is opened, and the nose cap is jettisoned once the vehicle is out of the atmosphere.

Weight The weight breakdown for the command module is given in table 5.5.

Structure The structure is basical ly an aluminum alloy and stainless steel right cone of semimonocoque construction. Internal framing is indicated schematically in fig. 5.7. A de­tailed structural weight breakdown is presented in table 5.6.

The nosecap is a spherical section with a radius of 1 8.7 in.

Table 5.5 Weight breakdown for the command module UbI Structure (see table 5.61

Cold plates

Computer

Inertial measuring unit

Guidance elec�rOnics

Stabilization and control

Batteries and connectors

Radar electronics

Radar fuzing antennas

Commun ications

Optical sensors and processing electronics

Internal insulation

Total

1150

100

60

100

70

100

200

75

10

150

85

150

2250

Project Icarus

SPLICING SUBSTRUCTURE

5.7 Internal structura of command module

Tabl.5.6 Detailed structural weight breakdown IIbl

Nosacap

Stiffened skin panels

Upper ring

Base cruciform

Floor

Hatch frames

Hatches

Quadrant splica members

Equipment mounting proviSions

Fasteners. bolu. etc.

External ablative

Total

CRUCIFORM

66

25

320

15

60

240

20

30

40

120

50

230

1 150

The Icarus Spacecraft 67

It is formed from 0.1 25-in AMS type 301 stainless steel. The 4 skin panels are quarter sections of the frustrum of a right cone 100 in high, with a base radius of 77 in. Each panel is fabricated from an inner skin of 0.040-in 2024-T3 aluminum alloy corrugation and a smooth outer skin of O.060-in aluminum alloy sheet (fig. 5.8). The inner skin has uniform, nontapering corrugations. Spray-on type ablative material applied to the external surface together with internal insula­tion, provide thermal control during ascent.

The base cruciform is a symmetrical truss structure of 8-in­deep extruded magnesium alloy I-sections. It is shown sche­matically in fig. 5.7. A floor is constructed of corrugation­stiffened skin of the same section as the skin surface (fig. 5.8) which isolates the command module from the rest of the vehicle and provides a rigid base for equipment mounting.

There are 2 hatches in the command module. The ground access hatch, located in the middle of one of the skin panels, is 2 ft wide by 3 ft high and is constructed similarly to the stiffened skin panel which it replaces. This hatch is permanently mounted with screws after equipment installa­tion and checkout. The second hatch, which is of similar construction and approximately 2 ft square, covers the optical Icarus sensor and equipment, as well as the fuzing radar. This hatch is closed during launch, but opens after the vehicle is

I-l00a-j

/ �.�,U". >. �oo 60�LLOY CORRUG

/ !--I. DO .1. 2.00---l

O.Of;oIN. ALUM ALLOY SHEET

5.8 Stiffened skin section detail

Project Icarus 68

out of the atmosphere. Due to the delicacy of this optical equipment and the fear that the combustion products from explosive charges might damage the sensors, the hatch is opened by spring action rather than jettisoned. The nose cap is also jettisoned after ascent to expose the sun sensor and alternate fuzing radar antenna.

Equipment Location The I carus command modu Ie carries very little equipment for its large volume, the size being de­termined by the aerodynamic fairing requirements. Equip­ment locations are indicated in fig. 5.9: The instruments selected do not necessitate a controlled atmosphere. Tempera­ture control is provided by cold plates driven by the environ­mental control system of the propulsion module.

SUNSEN901\ FUZlHG RADAR ANTENNA

5.9

STABILIZATION AND CONTROL ELECTRONICS PACKAGE

Inboard profile of the command module

(OPEN)

RADAII

The Icarus Spacecraft

References

1. North American Aviation, "NASA Support Manual, Apollo Spacecraft Familiarization," SID 62-435/SM 2A-02.

2. SAE, "Apollo - A Program Review," NASA SP-257, SAE (19641.

69

3. Pyker, N. J., "Technical Status of the Apollo Command and Service Module." In Advances in the Astronautical Sciences, Vol. 18, R. Fleising, ed. (American Astronautical Society, 19641, pp. 303-345.

4. Sarture, C. W., "Guidan ce and Control of Rocket Vehicles." In Guidance and Control of Aerospace Vehicles, C. T. Leondes, ed. (New York: McGraw-Hili, 19631, pp. 191-249.

6

Guidance and Control

Introduction

Guidance can be thought of as the decision process by which the type and frequency of velocity corrections are deter mined to satisfy certain trajectory requirements. The execution of these velocity corrections is control. There are basically 2 types of corrections: powered-flight and i m­pulsive. Powered-flight corrections involve a major thrusting maneuver and are required for such operations as launch into parking orbit, transfer to a different orbit, and landing. Thrust ti mes are on the order of minutes, and velocity incre­ments are on the order of thousands of feet per second. Generally, powered-flight guidance policies are concerned with mini mum expenditure of propellant. Impulsive guidance, on the other hand, is concerned with minor velocity cor­rections and is frequently called midcourse guidance. Thrust times are on the order of seconds, and klv's are on the order of tens to hundreds of feet per second. The guidance policy is concerned with the trajectory constraints (time of arrival,

Guidance and Control 71

for exa mple), and with when and how to perform the velocity corrections.

Powered Flight Guidance and Control

Boost and Parking Orbit Injection The Saturn guidance syste m, developed at the NASA Marshall Space Flight Center, operates from lift-off to parking orbit injection. The guidance syste m includes an inertial measurement unit (l MU), that is, a 3-axis stable platform with 3 gyros and 3 accelerometers, and a co mputer developed by I B M. The I M U is aligned optically on the pad before launch.

The guidance law e mployed during first-stage burn is an open-loop pitch-over program designed to minimize the aerodynamic loads and to prepare for propellant-efficient parking orbit injection. The outer 4 of the 5 F-1 first-stage engines are gi mbaled and respond to the guidance and control com mands.

During the operation of the second and third stages, the Saturn guidance system functions in a closed-loop mode in which the guidance and cutoff commands are computed as functions of position, velocity, thrust acceleration, and time at approxi mately 1-sec intervals (1). The guidance ob­jectives are to minimize fuel and to maximize final parking orbit injection accuracy. Second-stage guidance and control com mands are implemented by gimbaling the outer 4 of the 5 J-2 engines. Parking orbit injection is accomplished with the single J-2 engine of the third stage (S-IV B), which is gimbaled for thrust vector control, while roll control is achieved by use of the roll-attitude control thrusters.

Intercept Trajectory Injection Injection onto an Icarus rendezvous orbit requires the use of the engines from both

Project Icarus 72

the S- I V B and the spacecraft in sequence. The spacecraft is separated from the S- IV B during the orbit injection maneuver following the S- IV B burnout, which requires the guidance co mputations to be performed aboard the spacecraft.

The injection maneuver is monitored and controlled by the spacecraft's I MU and guidance computer. The M IT­developed I MU is a 3-axis platfor m with 3 single-degree-of­freedo m integrating gyros and 3 pulse-rebalanced accelerom­eters. The guidance co mputer uses solid-state components, draws 100 w of power, weighs 60 Ib, and occupies 1 ft3 of space.

The spacecraft I MU is initially aligned on the launch pad by gyroco mpassing. If injection onto the intercept trajectory takes place within 2 hr, the I MU does not need to be re­aligned. If the spacecraft re mains in parking orbit for several hr, however, its I MU must be updated by the spacecraft optical measure ment unit (O MU). The O MU consists of sun sensors, star trackers, earth sensors, and an I carus tracker of unique design. These optical devices, as well as the attitude control syste m (ACS) which stabilizes the spacecraft during optical sightings, are discussed in detail later in this chapter.

During the parking orbit, ground-based tracking is used to deter mine the spacecraft's velocity and position. This naviga­tional infor mation along with infor mation about Icarus' trajectory is used to deter mine the beginning of the injection maneuver, the direction and magnitude of the applied thrust, and the shut-down ti me of the engine.

The closed-loop guidance law which puts the spacecraft onto the proper intercept trajectory is a propellant-efficient cross-product algorith m in which the error signal is propor­tional to the velocity-to-be-gained vector (2). The velocity­to-be-gained concept is well-documented in the literature; the geo metry and mechanization are described by Battin (1).

Guidance and Control 73

Table 6.1 Spacecraft guidance subsystem operations

Ground Icarus Bistatic Fuzing Event ACS Track OMU IMU Tracker Radar Radar

1. t,·t, SIC SIC Update Coast stabili· state IMU Phase (hr) zation 2. t, Mid· Thrust course Correction

3. t. ·t, Update Cross· Coast IMU track Phase (days) data 4. t3 Termi· Thrust nal Cor· rection

5. t4 ·t. Update Cross· Rela· Coast IMU track tive Phase (hr) data r and; 6. t4 Thrust Terminal Correction

7. t.·t4 Update Cross· Rela· Coast IMU track tive Phase (min) data r and; 8. ts Terminal Thrust Correction 9. t(ts Fuzing Coast signal Phase «1 min)

Project Icarus 74

Impulsive Guidance and Control

The execution of i mpulsive ter minal maneuvers involves a complex interaction of spacecraft syste ms (fig. 6.1). This interaction is su mmarized in table 6.1. The IMU, which is inactive during coast periods, receives an update fro m the O MU prior to correction ti me and controls the thrust vector during the correction maneuver. The spacecraft's navigational data is obtained from the ground-based unified S-band tele­co m munications syste m. Icarus' trajectory is determined by existent ephe meris data, improved during the last month by additional astrono mical telescope data, and in the last 30 hr by ground-based radar (Haystack). The onboard Icarus tracker and radar syste m provide relative cross-track data for ter minal maneuvering when Icarus is close enough to the spacecraft. The deter mination of the state of Icarus relative to the space­craft without on-board sensors requires knowledge of the spacecraft's state relative to earth as co mputed by the S-band system.

At the ti me of the first ter minal correction, the on-board optical tracker provides angular infor mation on Icarus rela­tive to the spacecraft velocity vector. Since the range is too

6.1 Guidance and control maneuvers

Guidance and Control 75

great for on-board radar to be effective, relative range and range-rate are transm itted to the spacecraft from the grou nd­based tracking network. (Although based on ephemeris data for Icarus, range and range-rate are of sufficient accuracy for the first terminal maneuver.) The trajectory correction is calculated by the on-board computer to null the apparent cross-range miss at intercept. The last 2 terminal corrections use the angle data from the optical tracker and ranging data from the on-board radar. The spacecraft is oriented during these terminal corrections with its roll axis nearly perpen­dicular to the line of sight to Icarus. This orientation as-sures that Icarus is visible both to radar antenna and to the optical tracker. The final approach is made from the sunlit side of Icarus to insure visibility and to avoid 1800 rotations of the spacecraft for thrust vector control. The onboard fuzing radar provides the signal to detonate the bomb when the spacecraft reaches the nominal fuze altitude. If the space­craft passes I carus at an altitude greater than 100 ft, the detona­tion signal is sent when the range-rate becomes positive.

Midcourse Correction The first of the impulsive velocity corrections for the Icarus mission is the midcourse correction at time t2, which attempts to null the intercept trajectory in­jection error at time t 1 (fig. 6. 1 ). Th is error, caused primarily by accelero meter errors and engine cutoff errors, is measured by ground tracking. Ground-based computers then determine the necessary trajectory corrections, and, approximately 6 hr after injection, the correction commands are transmitted to the spacecraft.

To determine the midcourse velocity correction requirement for Project Icarus, a linear perturbation analysis was carried out. For a 1-0 accelerometer bias uncertainty of 10-5 g, a main engine cutoff uncertainty of 13 fps, a thrust duration of 30

Project Icarus

sec, and an injection ti me of 1000 sec, the 1-a velocity requirement is 23 fps for each of the 6 Icarus missions (table 6.2).

76

Terminal Corrections The 3 impulsive ter minal maneuvers correct for the i mproved knowledge of Icarus' trajectory as deter mined by the on-board radar and Icarus tracker and for the cutoff velocity error of the previous maneuver. Con­vergence on the target is assured by maintaining the space­craft's line of sight to Icarus fixed in inertial space. The number and timing of i mpulses for such an intercept path is not i m mediately obvious. Accuracy is of course critical, but the closing velocity in excess of 100,000 fps makes the attainment of high accuracy difficult.

For Project Icarus, the time ts of the final correction was deter mined first. The last correction must be made late enough to enable the optical Icarus tracker to resolve the target to within 100 ft, but soon enough to allow for tracker data processing, pointing and actuation of the service module engine, and propagation of the applied velocity. Too early a correction would degrade accuracy, while too late a correction would result in an excessive tov require ment.

In the case of a near miss at the sunlit edge of Icarus, the spacecraft must pass no further than 100 ft from the surface to be effective. For 10 percent illu mination of a spherical Icarus of radius 2000 ft (the most probable radius), the dis­tance fro m the center of illu mination to a point 100 ft fro m the surface is at least 300 ft. Thus no better resolution is re­quired of the Icarus tracker. Within its 10 arc-sec uncertainty, 300 ft can be resolved at approximately 1200 mi, so me 50 sec fro m interception. The final correction must then be made as soon as possible after the last optical measurement.

Guidance and Control 77

The first and second terminal maneuvers were chosen in such a way that the cross-track uncertainty would be reduced after each correction with a reasonable Av. The first cor­rection must be made at some time t3 after the optical Icarus tracker acquires its target, while the second correction was re­quired to be postponed until time t4, when Icarus would be within range of the onboard radar. Only the final 2 cor­rections, then, would be computed from both optical and radar information.

A linear perturbation analysis yielded the results in table 6.2. The assumptions used to obtain the table include a 30-sec thrust duration, cutoff velocity uncertainties of 0.0 1 fps for the midcourse and first 2 terminal maneuvers, and un­correlated corrections. The first terminal correction would be made at a range of 150,000 mi, when the optical tracker had reduced the cross-track uncertainty to about 7 mi. The second would be made at 5000 mi away, when the uncertainty could be reduced to 0.25 mi.

To insure adequate fuel for the impulsive corrections, the 3-value of total required Av, 615 fps, was chosen as the mission requirement. The resulting circular error probable (CEP) due to the optical tracker uncertainty in the cross­track plane is approximately 300 ft.

Attitude Control System

Thrust Vector Orientation To point the spacecraft before executing a velocity correction, both the present attitude and desired attitude of the spacecraft must be known. The present attitude is determined by the attitude stabilization system discussed in the next section, while the desired attitude is related to the direction of the required Av as computed either on the ground or on board the spacecraft. After optical align-

Project Icarus 78

ment of the I M U, or perhaps no in-flight alignment in the case of injection onto an intercept trajectory after a parking orbit shorter than 2 hr, the computed difference between desired and present attitude is sent as an error signal to the IM U. The IM U in turn commands the reaction control system to rotate the spacecraft so as to null the error sign�1. During the thrust­ing maneuver, any deviation of the spacecraft fro m its desired orientation is detected by the I M U and nulled by the reaction control jets.

Thrust vector orientation and control systems are well­treated in the literature. The portion of the Icarus attitude control system used for thrust vector orientation is essentially that used in the Apollo command and service modules. Thus, no further treatment of the I M U or of the reaction control system is included in this report. More significant is the at­titude stabilization system, which draws on technology other than that of Apollo.

Table 6.2 Impulsive velocity correction rll.9uirements

RMS velocity corrections (fps) Corrects Mission

Correction Time for 1 2 3 4 5 6

Midcourse t.-t. = Injection 23 23 23 23 23 23 (t. ) 6 hr errors

Terminal ttt, = t. maneuver 44 45 46 50 52 54 (t, ) 1.8 hr and Icarus

cross-track errors

Terminal tf-t4 = t, maneuver 194 194 194 194 194 194 (t4) 3.5 min and Icarus

cross-trac k errors

Terminal ttts = t4 maneuver 35 35 35 36 35 35

(ts) 50 sec and Icarus cross-track errors

Total 1-a (RSS) Av requirement 203 203 204 205 205 206

Guidance and Control 79

Attitude Stabilization System In addition to those sub­systems that constitute the Apollo attitude control system, the Icarus spacecraft uses optical sensors for attitude stabiliza­tion. Such technology is closely allied. with that of inter­planetary spacecraft like Mariner and Ranger.

A limitation of 650 Ib of available control fuel in the command and service modules necessitated the choice of a ±5° dead band for the attitude stabilization system during the cruise phase. Several modes of operation and functionally redundant attitude sensors are included in the system.

For attitude stabilization about 2 axes orthogonal to the spacecraft-to-sun line, 2 complete but complementary sun sensor systems are employed. One affords coarse acquisition but limited accuracy, while the other has high accuracy but limited search and acquisition capabilities. The mission can be accomplished with either system alone with only a slight degradation in performance (accuracy for the former, ac­quisition time for the latter ).

Attitude information about the sun line is obtained pri­marily from a Canopus tracker, but in the event of failure an earth sensor or a Capella tracker is employed. Normally the earth sensor's function is to point the high-gain on-board com­munications antenna at ground-based stations. Failure of the earth sensor is backed up by a provision for the generation of pointing commands from the Canopus tracker. The Capella tracker is part of the optical Icarus sensing assembly described in the next section.

Electro-Optical Instrumentation

The electro-optical sensors described in this section were chosen primarily from the TRW systems report that was con­sidered the most recent, most complete, and, hopefully, most objective survey available at the time of the study (3).

Project Icarus 80

Icarus Sun Sensor Assembly The Icarus sun sensor assembly

consists of 2 space-qualified sun sensor systems. The Northrop­

Nortronics Mariner sun sensor serves as the acquisition sensor,

while the Bendix fine angle sun sensor is the tracking unit.

The Mariner sensor contains a shadow bar structure (fig. 6.2) and has a 41T-steradian coarse field of view. The wide

field of view permits early detection of the sun and saves

maneuvering fuel that would otherwise be expended in a search with a narrow-field sensor. The accuracy obtainable

with this unit is marginal with regard to mission requirements.

Its main function is to place the spacecraft within the ac­

quisition range of the fine sensor and to provide a backup if

the fine sensor should fail. The Mariner sun sensor assembly

consists of 2 secondary and 4 primary sensor units. The

secondary units each contain 4 cadmium sulphide photo­

conductors and provide a coarse indication of the sun's direc­

tion. One set faces 1800 off the true spacecraft null, establish­

ing an unstable null. If the sun is within its 21T-steradian field

SUNLIGHT

6.2 Coarse acquisition sun sensor

Guidance and Control 81

of view, the vehicle is commanded to rotate away from the

unstable null line into the field of view of the other set, which

is located on the true null line. The secondary units give

analog null signals for 2-axis control. Two primary sensors

are used on each of 2 axes to provide fine control. Each unit

consists of a cadmium sulphide photoconductor with a field

of view covering a quadrant approximately 1600 in azimuth

and 45° in elevation. The Mariner sun sensor weighs 1 1 OZ, occupies 2 1.5 in3, and consumes 0.8 w of power.

The fine-angle sun sensor was chosen because it satisfies

the accuracy requirements imposed on the Icarus mission.

The sensor consists of an objective lens, a coarse silicon-solar­cell array, a magnifier lens, and a fine silicon-solar-cell

SOLAR CELLS,

COURSE ARRAY

SlJNLIGHl'

t-t--..a..A��==::i OBJECTIVE LENS

I--_....Jct:�==:t- MAGNIFIER LENS

SOLAR CELLS,FINE ARRAY�

OPTICAL I AXIS

6.3 Fine tracking sun sensor

Project Icarus 82

quadrant array (fig. 6.3). The sun line is rotated towards the optical axis of the sensor by the coarse sensor array, which has a ±1 0° field of view. When the sun line and optical axis are nearly aligned, the objective lens projects sunlight through a hole in the coarse array onto the magnifier lens. The image is magnified and projected onto the fine quadrant array, thus establishing a stable null. The coarse sensor array does not contribute to the null and intercepts the focused rays as angular deviation increases. Both arrays are electrically inter­connected to produce continuous output signals. The fine angle sun sensor weighs 30 OZ, occupies SO.S in3, and requires no power.

Canopus Star Tracker The star tracker chosen for the Icarus spacecraft is the ITT Canopus tracker used on the Lunar Orbiter Program. The tracker consists of a single package containing optics, photomultiplier tube, detector electronics, scan logic, deflection electronics, and power supplies (fig. 6.4). The Canopus tracker has a field of view large enough to include the movement of the spacecraft within its SO dead band. It has an accuracy of ±SO arc-sec rms, is already opera­tional, weighs 7.0 Ib, occupies 264 in3, and consumes 8.0 w.

Earth Sensors The Northrop-Nortronics short-range earth sensor provided attitude error signals for pointing the Ranger and Mariner spacecraft antennas. It is used on the Icarus space­craft to provide error signals to a servo system directing the high-gain directional earth communications antenna and functions at ranges from 20,000 to 1,000,000 mi. The sensor is a static device composed of a 3-element shadow mask, 3 end-on photomultipliers, power supply, and processing elec­tronics. The mask is configured so that an angular deviation of the earth off the sensor axis causes an unbalance in the

Guidance and Control 83

6.4 Canopus star tracker

photomultiplier signals. The signals are processed to generate

angular errors and input commands to the antenna servo

system. The short-range earth sensor weighs 2.5 Ib, occupies

72 in3, and consumes 3.5 w of power.

The Mariner long-range earth sensor, which becomes ef­

fective at 1,000,000 mi from earth and operates as far away

as 50,000,000 mi, substitutes a very reliable vibrating-reed

scanner mechanism with a modulating mask for the shadow

mask of the short-range sensor. The vibrating mask generates

error signals linearly proportional to the 2-axis offset of the

earth from the sensor axis. It also employs photomultipliers,

processing electronics, and a refractive objective lens. The

long-range earth sensor weighs 6.5 Ib, occupies 160 in3, and

consumes 6.5 w of power.

Optical Icarus Sensor The optical Icarus sensing assembly

consists of 2 sensing heads. One head, the Dual Mode Star

Tracker built by ITT Federal Laboratories, tracks the star

Capella, visual magnitude +0.2, while the second head, the ITT Orbiting Astronomical Observatory (OAO) Boresighted Star Tracker, tracks Icarus.

Capella was chosen as a reference star for its brightness and

Project Icarus 84

its relative proximity to Icarus' line of sight. For the 13 days before collision, during which the Icarus sensing assembly is to be used, the lines of sight to Capella and to Icarus are ap­proximately 20° apart. Capella is never closer to the sun than 23°, but it is assumed that the Dual Mode Star Tracker is equipped with a sun shield that will permit tracking of the 0.3-magnitude star as close as 20° from the sun. Such a sun shield is believed to be within the state of the art.

The Dual Mode Star Tracker, which has already been used by the NASA Goddard Space Flight Center on the Aerobee Rocket Probe, is able to acquire Capella with its 8° X 8° ac­quisition field of view, despite the 5° limit cycle of the space­craft attitude stabilization system. A tracking field of 32 arc­min square enables a tracking accuracy of 5 arc-sec rms to be obtained.

The OAO Boresighted Star Tracker can track a sixth magni­tude star with 10 arc-sec rms accuracy, or a fourth magnitude or brighter star with 1.5 arc-sec rms accuracy. Its high sensi­tivity results primarily from an extremely narrow field of view, 10 arc-min. As for the Dual Mode Star Tracker, it has been assumed for the Boresighted Star Tracker that a sun shield can be designed to allow tracking at just 20° from the sun. The angle between Icarus and the sun, as seen from the earth, varies from about 20° for the terminal phase of the first mission, to 33° for the terminal phase of the sixth mission. Since the spacecraft approaches Icarus from the sunlit side, the angle between Icarus and the sun as seen from the spacecraft must be greater than the angle seen from the earth. Thus, the design of a sun shield for operation at 20° from the sun is a conservative goal.

The stellar background as seen by the Boresighted Star Tracker can be determined by quantities listed by Allen (4). For a galactic latitude of 10°, approximately that of Icarus

Guidance and Control 85

during the last 13 days before impact, there is a stellar back­ground equivalent to 187 stars of tenth visual magnitude per square deg. The circular field of view of the Boresighted Star Tracker contains .022 square deg. Thus the background equivalent as seen by the Boresighted Star Tracker is 4. 1 stars of tenth magnitude. Since every 5 magnitudes represents a factor of 100 in brightness, the ratio of Icarus at fifth mag­nitude to the stellar background as seen by the Boresighted Star Tracker is 100/4. 1 = 24.4. Such a signal-to-noise ratio is quite adequate for precise tracking.

In order to remain conservative in estimating the acquisition range of Icarus, it was assumed that Icarus could not be de­tected at sixth magnitude, the stated performance of the tracker, but instead at fifth magnitude, when Icarus is 2.5 times as bright. Thus, the detection range, that is, the distance at which Icarus appears as a star of visual magnitude +5.0, is 258,000 mi.

If the spacecraft has been placed precisely on a collision course with Icarus, then the angle between Icarus and Capella will remain constant until collision, with Icarus' brightness increasing. A change in the angle indicates a deviation from the nominal intercept trajectory, which is used to correct the spacecraft's velocity. Since the deviations from nominal are small, the Capella and Icarus tracking heads are fixed with respect to each other, their lines of sight being approximately 22° apart. The deviations can be accommodated within the tracking fields of view of the 2 heads.

To tolerate the dead band in the attitude stabilization system, the 2 fixed heads are mounted on a gimbaled 3-axis platform, which receives error signals from the 2 trackers to keep them nulled on their respective targets. The platform is aligned before launching the spacecraft such that the 8° X 8° acquisition field of the Dual Mode Star Tracker is assured of

Project Icarus 86

containing Capella. This alignment is not a difficult task, since the angles between the lines of sight from the spacecraft to the sun, Canopus, and Capella can be calculated precisely for any trajectory. Deviations from the trajectory, errors in alignment, and dead band excursions before acquisition of Capella merely cause departures of Capella's image from the center of the 8° X 8° field. Capella's brightness affords un­ambiguous detection of the target star in the field of view. When Capella is detected, error signals are sent to the 3-axis platform to null the tracking field on Capella.

When the Capella tracker is at nu II, the error signa Is from the Capella and Canopus sensors are sent to the 3-axis platform, which is then rotated about the line of sight to Capella such that the nominal pointing direction of Icarus is aligned with the Icarus sensing head. Since an uncertainty of 300 mi in Icarus' cross-range at a range of 258,000 mi is less than 4 arc­min, the nominal Icarus pointing direction can be determined well within the 10 arc-min field of view of the boresighted tracker. When Icarus is detected in the field, a track mode is initiated automatically, since no star brighter than visual magnitude +5.0 exists in the stellar background for the pro­posed approach geometry.

As the spacecraft undergoes its stable limit cycle, error signals from the Capella tracker and Icarus sensor indicating deviations in orientation about axes perpendicular to the line of sight to Capella are fed to azimuth and elevation gimbals of the platform. Signals indicating a deviation about the line of sight to Capella are used to rotate the platform about that line until Icarus returns to a near-null position in the field of the Icarus sensing head. The error signals are also monitored by the guidance computer to detect changes in the Icarus-Capella angle, which indicate the need for a velocity change to correct the trajectory for interception.

Guidance and Control 87

Radar Systems

Introduction The radar systems provide range and range-rate

information for terminal guidance and detonating the bomb.

Two basic radar configurations are available. One configura­

tion is a self-contained unit with transmitter, receiver, power supply, and antenna. Alternatively, target illumination can be

supplied by a separate source based on earth, with the inter­ceptor carrying only a receiver and an antenna. The latter

system has the advantage of reduction in weight and power

requirements, and, as a result of fewer on-board components,

would probably be more reliable. Its limitation is that beyond

a certain distance from earth the illumination radar can no

longer supply enough power for detection by the spacecraft

receiver at reasonable ranges.

A preliminary study indicated that a transmitter-modulator

could be employed, although the weight of such a system

would significantly limit the payload. By using the M IT

Lincoln Laboratory's Haystack Hill radar for illumination,

intercepts could be effected as far out as 20 million mi. Since

only 6 launches are possible, the decision was made to use

the ground illumination system exclusively. It should be noted

that initial detection of Icarus by the Haystack receiver will

be at a range of approximately 2 million nm; hence ground­

based radar tracking of the asteroid will not be possible until

2 days before impact. Prior to that time all radar information

must be relayed back to earth by the spacecraft. The beamwidth of Haystack is 1 milliradian. Since the

trajectories of Icarus and the spacecraft intersect at less than

10°, both are well within the beam during the 10-min terminal guidance phase at each of the intercept locations. Two an­tennas are used. A small antenna facing earth receives the

transmitted signal; a larger homing antenna in front detects

the reflected echo from Icarus. Spacecraft-to- I carus range is

Project Icarus 88

Table 6.3 Radar eerformance

Earth- Trans- Acqui- Precision Icarus mitted Band- sition Precision at during

Mission distance Ave power width range acquisition tracking· p 8 Ai , t.f r Ar\ Ar2

lO'sPmi kw Hz mi ft fps ft fps 1 20.0 500 90 6,000 625 1.8 62 0.13

2 15.5 500 90 7,700 625 1.8 62 0.13

3 10.8 500 600 7,100 625 12 62 0.13

4 7.7 500 600 9,600 625 12 62 0.13

5 1.4 200 6,000 23,600 625 120 62 0.13

6 1.3 200 6,000 26,500 625 120 62 0.13

* Assuming 8; 20 Hz and SIN = 100

determined from measurement of the elapsed time between the 2 waves. Initial and final range and range-rate precision obtained for the various missions are summarized in table 6.3.

A small ranging radar similar to an altimeter is used to generate the impulse signal that sets off the detonator. Initia­tion occurs either at 100 ft or when the range-rate begins to increase or upon loss of signal after 4 sec of tracking. The last feature is required in the event that Icarus passes out of the beamwidth at close range.

Acquisition Range Since the second terminal maneuver (t4) is performed at a range from Icarus of 5000 mi, the required radar acquisition range, allowing for a minimum of 30 sec signal integration and tracking time, is 5700 mi.

At this range the uncertainty in position is less than the beamwidth of a 12- X 5-ft antenna operating at a frequency of 7750 MHz. (Beamwidth is defined for practical purposes as wavelength divided by the antenna dimension.) Therefore, a search radar is not required.

One form of the classic radar equation is

Guidance and Control

SIN = P G A X ...£... X _ 1 {tl % 41TP 2 41Tf2 k T 8' '

89

(6. 1 )

where SIN = signal-to-noise ratio; P = transmitted average power, w; G = transmitting antenna gain, 66 db; A = receiver aperture, 2.5 m1; p = earth- Icarus distance, m; C = Icarus radar cross section, m1 ; r = spacecraft- I carus distance, m; t = observation time, sec; k = Boltzmann's constant = 1.38 X 10-13 Jr K; T = noise temperatu re, 0 K; B = bandwidth, Hz.

The following assumptions are made:

1. Spacecraft antenna efficiency is 0.5, such that the receiver aperture is one-half the cross-sectional area. 2. Icarus is spherical, 1 mi in diameter, with radar reflectivity = 0. 1. 3. Observation/integration time = 30 sec. 4. Noise temperature = 60oK. 5. Optimum filtering of signal is obtained. 6. Haystack losses are 1 db; spacecraft system losses are 6 db. 7. SIN = 10 is required.

Since several parameters change during the mission, the radar system has 3 slightly different configurations. For the first 4 missions, Icarus is outside the ground-based tracking range, the Haystack receiver is off, and full power may be radiated continuously using frequency modulation. The average transmitted power available from Haystack is there­fore P = 500 kw. For missions 5 and 6, Icarus is within track­ing range of Haystack, and the transmitter must be turned off while the system is receiving. A reasonable duty cycle leads toP= 200 kw.

A compromise is made on bandwidth for the more distant interceptions. A large bandwidth is desired to avoid rejecting any of the returned signal due to large Doppler shifts in the event Icarus has a high rate of rotation. However, there are

Project Icarus 90

no records of rotational periods observed to be less than 2 hr. For Icarus this rate of rotation results in a tangential velocity

of 2 fps. At the operating frequency of 7750 MHz, this un­

certainty in Doppler shift covers a bandwidth of 90 Hz. This

value is used for missions 1 and 2. The intercept ra�ges for

missions 3 and 4 allow a bandwidth of 600 Hz, which will

permit an increase in uncertainty of rotation by an order of

magnitude without loss of any of the reflected signal. For

missions 5 and 6, a bandwidth of 6000 Hz is used. Theoretically,

this would permit reception of the total signal even at a rate

of rotation for Icarus of 1 rev per min-an upper limit of

possibility established by consideration of the cohesive

strength of gran ite.

Based on these considerations, the performance achieved

by the radar system in the different missions is shown in

fig. 6.5. A summary for the specific intercept ranges is given

in table 6.3.

Precision of Range and Range-Rate Data After acquisition

the guidance radar system provides range and range-rate data for terminal guidance. The precision of this information is

'" " � 10 ..

6.5

'I'TERCt:l'TflRli I •• (8· 90 HZ)

5 10 1& 20 DlBTANCE FRoM EARTH. P (I� IWI

Radar performance for all missions

Guidance and Control 91

improved with increased signal-to-noise ratio (SIN) and with reduced bandwidth. Due to inherent equipment limitations,

the maximum improvement in SIN over the value obtained

at acquisition is a little better than an order of magnitude.

By using filter banks it is anticipated that during tracking

bandwidth can be reduced to approximately 20 Hz.

The precision of the range data may be estimated from

the equation

t:..r = 2 c (PW) (SIN)% '

(6.2)

where fl.r = range precision; c = velocity of light; PW = pulse

width.

Using phase-coded pulses of 0. 1-psec pulse width, this relation predicts fl.r! = 625 ft at acquisition. With the value of SIN in­

creased from 10 to 100 as anticipated during the tracking

process, fl.r']. = 62 ft.

The precision of the range-rate data is given by

. ABI2 fl.r = (SIM % '

(6.3)

where fl.; is range-rate precision, and A is wavelength. I n this

instance, the filter bandwidth is matched to the anticipated

Doppler shift. Final precision values assume a bandwidth of

20 Hz with SIN = 100. The results are given in table 6.3.

Antennas The radar guidance system requires 2 antennas:

1 for receiving the transmitted signal from earth; the other

for detection of the reflected wave from Icarus. Electronically scanned arrays will be used to avoid the necessity of physically

training the antennas. The slots of one-half wavelength ( 1.94

cm ) are alternately inclined to accommodate the 1800 phase

reversal which occurs in a wave-guide transmission every one-

Project Icarus 92

half wavelength. To eliminate resonant effects, the spacing between slots is slightly different from one-half wavelength. Nonresonant slot spacing causes the beam to point slightly to 1 side of the geometric center of the array, but bearing in­formation is not required of the radar, and the effect is negligible.

As the spacecraft maneuvers during the terminal guidance phase, its roll axis is oriented cross-track to Icarus' trajectory, and it maintains attitude relative to earth and Icarus. For re­ceipt of the signal transmitted from earth, a 2-ft X 2-ft re­ceiving antenna is mounted on the side of the spacecraft that faces the earth. A 12-ft X 5-ft antenna on the opposite side detects the radar echo from Icarus. The beam of this antenna is fan-shaped, 11 milliradians by 20 milliradians in extent. Appropriate current phasing is employed to gain the desired aperture distribution and to compensate for its cylindrical shape, which matches the curvature of the spacecraft. This antenna installation is shown in fig. 5.6.

Weight and Power A nominal power requirement for each element in an array antenna is 80 mw. Estimated weight and power requirements are presented in table 6.4.

Fuzing Radar The radar proximity fuzing system is a com­pletely independent short-range radar that will produce a signal to initiate detonation of the nuclear bomb when target range is 50 ft, or upon change in direction of target range­rate that persists for a full msec, or upon loss of signal after 4 sec of tracking. As explained in chapter 3, the 5Q-ft fuzing distance ensures detonation at an altitude between 0 and 100 ft above Icarus' surface.

To provide ranging information as accurately and as long as possible, a wide beamwidth is used. The method adopted

Guidance and Control

Table 6.4 Antenna requirements

Homing antenna

Receiving antenna

Electronic components Total

Weight lib) 300

25 50

375

93

Power (w) 1000

80 20

1100

for accomplishing this is to use 2 crossed slots of one-half wavelength as the antenna. An operating frequency of 1000 MHz is used to minimize sky noise. The system is made identical for all missions. Under the worst assumption re­garding the rotation rate of Icarus, that is, 1 rev per sec, the required bandwidth is 800 Hz. Five-second tracking time at a signal-to-noise ratio of at least 100 is assumed. This specifica­tion implies a detection range of 1 10 nm.

With no signal integration, the logarithmic form of eq. 6. 1 for a radar with both transmitter and receiver is

(S/N)db = (Pt)dbw + 2(G)db + 2(X)db cm

+ (C)db m 2 - 4(R)db nm

- (B)db Hz - (NFo )db - (L)db, (6.4)

where Pt is rms noise power during the pulse; NFo is the noise loss factor, accounting for deviation of equipment tempera­ture from 2900K and for atmospheric conditions; L is system losses. Assuming combined system and noise losses of 10 db, eq. 6.4 gives a value of power during the pulse, Pt = 500 kw. To avoid range ambiguities without undue complexity, the period between pulses should not be less than the time re­quired for the pulse to strike Icarus and return. This estab­lishes T = 1.37 msec. For a pulse width PW = 10 nsec, the theoretical average power is Pav = f} T/PW = 4 w. Assuming an efficiency of 10 percent, the power required is 40 w.

The velocity of Icarus is so high that it moves 178 ft

Project Icarus 94

between pu Ises at the above pu Ise repetition frequency. In order to improve ranging accuracy as the range decreases, the pulse repetition frequency is increased after acquisition to a maximum of 8000 per sec at a range of 10 nm. The resolution is then 16 ft between pu Ises. Power requirements in the latter configuration are lower than in the former.

References

1. Battin, R. H., "Volume I of Lecture Notes for Course in Flight Guidance,"

Cambridge, MA: Massachusetts Institute of Technology.

2. Martin, F. H., "Closed Loop Near-Optimum Steering for a Class of Space

Missions," MIT I L Report T -413 (May 19651.

3. TRW Systems, "Radio/Optical/Strapdown Inertial Guidance Study for Ad­

vanced Kick Stage Applications - Survey of State-of-the-Art Electro-Optical

Sensors," Report No. 07398-6002-ROOO (October 19661.

4. Allen, C_ W., Astrophysical Quantities, 2nd edition (New York: Oxford Uni­

versity Press Inc, 19641, p. 234_

5. Skolnik, M. I., Introduction to Radar Systems (New York: McGraw-Hili, 1962).

6. Barton, D. K., Radar Systems Analysis (Englewood Cliffs, New Jersey: Prentice­

Hall,19641.

7

Communications

Introduction

Communication with the Icarus spacecraft, that is, the trans­

fer of information between the spacecraft and the earth, in­

volves 3 major operations: tracking, telemetry, and command.

Radar tracking for trajectory determination has already been

discussed in chapter 6. Telemetry, the "down-link" from

spacecraft to earth, and command, the "up-link" from earth

to spacecraft, are discussed in this chapter.

The down-link transmits both performance information

and experimental data, which are measured by transducers

placed strategically throughout the spacecraft. These trans­

ducers indicate such information as the following:

electronic compartment temperature; cold plate operating data; battery Voltage, current, and temperature; fuel cell voltage, current, temperature, and pressure;

SM engine gimbal angle;

SM engine start and stop;

Project Icarus 96

SM engine fuel temperature, pressure, density, and flow rate;

attitude control fuel tank pressure; vibration data (3-axis);

acceleration data;

shroud unlock, position, and temperature;

payload compartment temperature and pressure; optics hatch deployment.

The data up-link is responsible for inserting guidance data

and instructions into the guidance system, while the command up-link controls the execution of these instructions and of

other spacecraft operations. Among those functions con­

trolled by the up-link are the following:

shroud release; SM engine gimbal angle;

SM engine start and stop; SM attitude;

battery power (on/off);

cooling plate (on/off);

optics hatch deployment.

The primary constraint on the communications system

is reliability. The system must be capable of operating for 60 days and of affording high accuracy, especially on the

guidance data up-link. In addition, development time is severely limited. These requirements, plus the fact that the

Apollo configuration is being used for the spacecraft, suggest the use of the Apollo Unified S-Band (USB) telecommunica­

tions system with modifications to increase its range. This

system, now in the development stage, is compatible with

Apollo hardware and is designed for high reliability and

accuracy.

The Basic Apollo Unified S-band System

The primary feature of the Apollo Unified S-band System is

Communications 97

that all data and voice channels are modulated on subcarriers

and combined to modulate a single radio frequency carrier

which is transmitted to the ground. Included in this unified

system is a pseudo-random-code-ranging subsystem. The pseudo-random code is phase modulated directly on the S­band carrier on the ground. The spacecraft demodulates the ranging code and modulates a different S-band carrier with the code. The time difference on the ground between the original and received codes is a direct measure of the range.

(The scale factor is c/2, where c is the velocity of light.) The down-link carrier frequency is obtained in the spacecraft by

observing the up-link carrier frequency and generating a fre­quency that is exactly 240/221 times the observed frequency.

Since the process is phase coherent, that is, the up- and down­

link frequencies are synchronized, Doppler-shift measure­ments may also be incorporated into the ranging system.

The only other portions of the Apollo communications

system needed for the Icarus mission are the Pulse Code

Modulated (PCM) telemetry system, used for monitoring the

spacecraft systems, and the up-data link, used for inserting

data and commands. A simplified diagram of the USB system

is shown in fig. 7.1. The outputs of all spacecraft transducers are connected to

the various inputs of the PCM encoding section. These param­eters are scaled into electrical signals from 0 to 5 v. A com­

mutator then samples each parameter voltage in sequence and

generates a pulse code word corresponding to each sample. In the premodulation processor, the resulting PCM sequence,

with additional code words added for synchronization, is used

to phase modulate a subcarrier, which is then added to the

range code before phase modulating the down carrier in the

PM exciter. The output section provides power amplification before transmission. The up-data link is similar in form and

Project Icarus 98

operation except that the coded command and data sequences

are generated by ground computer rather than by a telemetry system. The up data is then sent to the command circuits or to the computer as addressed.

The primary disadvantage of the present USB system is

that it was designed for use at lunar distances, whereas the Icarus mission requires the communications link to operate at distances up to 20 million mi. It is necessary, therefore, to make modifications of the Apollo USB system to increase

its range. Since there will be as many as 4 spacecraft in the

ground antenna beam at the same time, modifications are necessary to insure that one spacecraft will not receive signals

intended for another.

Required Modifications

The problem which constrains the usable range of a communi­

cations system is the presence of random electrical signals

(noise) both from the background of space and from within the communications equipment. To obtain useful informa­

tion, the power in the desired signal must exceed the power in the random noise by a substantial amount. The signal-to­noise power ratio is

SIN = Pc Gt G r ( _A_)2

No B 21Th

where SIN = signal-to-noise power ratio, dimensionless;

(7.1 )

Pt = transmitter power, w; Gt = gain of transmitter antenna, dimensionless; Gr = gain of receiver antenna, dimensionless;

No = noise power density, w/ Hz; B = bandwidth, Hz; A =

wavelength, m; h = earth/spacecraft range, m. The transmitter power, noise power density, and antenna

parameters are fixed by the constraints on available space-

Communications

TELEJIETllYI PeM INPUTS CODER

UP OAT"

7.1

PIIE­MODI/LATION PJII(IC;J:AOR AND TRANSPONDER

Simplified representation of USB system

AMPLIFIERS

DUPLEXING

CDlC111TRY

99

HlGIHlAIII

ANTENI<A

''OMlir

"lITEI'll"

craft power, sky background, and antenna sizes, respectively.

Therefore, in order to increase the range of the spacecraft, the only parameter that can be changed is the bandwidth of the receiver. This effectively eliminates the noise that is outside that band, thus decreasing the amount of noise that enters the receiver. However, the rate at which data can be trans­mitted is directly proportional to the bandwidth. Conse­

quently, increasing the range decreases the available data rate.

Specifically, the data rate becomes inversely proportional to

the square of the range. Using the present Apollo parameters

of 51,200 bits per sec at 250,000 mi and the 85-ft parabolic antenna, one obtains fig. 7.2, which indicates the rate capa­

bility at other ranges assuming the same signal-to-noise ratio.

The curve for the 21 O-ft Jet Propulsion Laboratory (JPL)

antenna is also shown. Since the I carus mission requires communications at a range

of 20 million mi, it is clear from fig. 7.2 that the 210-ft JPL

antenna is better at this range, giving a data rate of 50 bits per sec in contrast to fewer than 10 bits per sec for the 85-ft

antenna. Thus, a ground link must be installed between the Apollo control center and the JPL facility. Modifica­

tions of the present Apollo control center must be made

Project Icarus 100

to handle the lower data rate. Since the rate of 50 bits per sec is still quite low, a higher rate will be provided for near-earth operations such as the midcourse correction. The Apollo equipment is designed to operate at a data rate of 1600 bits per sec as well as 51,200 bits per sec. The former is the logical choice for near-earth operations of the Icarus spacecraft. Figure 7.2 shows that this rate can be used up to 1.4 million mi from earth if a suitable bandwidth and sub­

carrier are used. Thus up to 1.4 million mi a data rate of 1600 bits per sec will be used. At 1.4 million mi the data rate will be switched to 50 bits per sec for transmission of

all terminal-phase data for the first 4 missions. Since the 210-

ft antenna must be employed, the nominal intercepts are

planned so that the spacecraft will be visible to this antenna

in California for about 2 hr before the nominal interception. I n order to isolate each spacecraft it is necessary to use dif­

ferent frequencies for each. Code addressing could be used in the telemetry link, but the range code transmitted from each spacecraft would interfere with the others. The present

Apollo spacecraft has already utilized frequencies of 2287.5

MHz for the command module, 2282.5 MHz for the lunar

JU�'r---------------'

7.2 Performance of telemetry system

Communications 101

h

7.3 Ideal waveforms: (a) transmitted; (b) received

module, and 2272.5 MHz for an FM channel for TV trans­mission. These frequencies may be used for each of the first

3 missions, with the corresponding radio frequency equip­ment installed. An additional carrier can be added at 2292.5

MHz for the fourth mission, since this frequency is still in the band in which the ground system can function. The last 2

missions can use the same frequencies as the first 2, since they

will be launched after the first 2 intercepts. The correspond­

ing up-link frequencies, all within the transmission band of

the ground stations, are 2101.8 M Hz, 2106.4 M Hz, 2092.6

MHz, and 2111.0 MHz.

These are the only necessary major modifications. Of course, changes in detailed circuits will be necessary to effect

the communications channels.

Range Code Technique

The operation of the ranging technique can best be described

as follows. Suppose a periodic waveform (fig. 7.3a) is trans­mitted to the spacecraft and retransmitted to the ground. The

waveform received on the ground will be delayed by the trans­mission time, which is proportional to range (fig. 7.3b). If d is defined as the phase difference in units of time between the transmitted and the received sequence, then the total delay is

given by (n T + d) where T is the period of the waveform, and n is an unknown integer.

Project Icarus 1 02

In order to resolve the ambiguity caused by the unknown n, it is necessary to use a waveform of sufficiently long period

so that it will not repeat during the propagation time. However, the resolution of a short-period waveform is lost in the process.

The conflicting requirements of long period and high resolu­tion can be resolved by the use of a pseudo-random sequence as the transmission waveform. These pseudo-random sequences have long periods but locally random structure. An example

of such a sequence of period 15 bits and the autocorrelation

function of that sequence are shown in fig. 7.4. If one shifts the transmitted sequence until its correlation with the re­

ceived sequence is maximum, the required amount of shift is a precise measure of d.

I n order to obtain a sufficiently long period code to use at lunar distances, the Apollo ranging system combines 5 codes,

transmitted at 992,000 bits per sec, of the following lengths: CL code, 2 bits; X code, 11 bits; A code, 31 bits; B code, 63

bits; and C code, 127 bits. The use of 5 differ.ent codes de­

creases the number of comparisons to be made since each code

can be acquired separately. The overall code that is the sum, modulo 2, of each of the 5 codes has period 5,436,682 bits.

This corresponds to an unambiguous range of half a million mi. I n order to extend the range of the Apollo data link it

was necessary to decrease the bit rate by a factor of about

h .. 6 7.4 Pseudo-random transmission waveform: (a) structure; (b) autocorrelation function

Communications 103

112 lIZ • Dt:EP SPItCE .... TDI

7.5 Spacecraft USB system

IUOIt-OAIJI "OMNI" "NTEN"�

UIOLATOII nLTt:a • L"IOLATOIl

1000. This was also necessary for the range code, both to in­

crease the signal-to-noise ratio and the unambiguous range of the code. Therefore, the range code for the Icarus mission is

the same as the Apollo code, but at a rate of 992 bits per sec. The unambiguous range will then be 500 million mi, well be­yond the required range.

The resolution of the range code for the Apollo system is about 1/4 cycle, or about ± 75 m. By tracking the Doppler cycles of the S-band carrier, this resolution is reduced to about ± 1 m at lunar distances. For the slower rate used for

the Icarus mission, the resolution of the code is increased con­

siderably. However, the S-band carrier is unchanged so that resolutions on the order of a few hundred m can be expected

with some modification of the present equipment.

The Spacecraft System

For Project Icarus many of the capabilities of the Apollo USB system are not needed. As mentioned in the section on the basic

Project Icarus

USB system, only the PCM telemetry and the up-data links

will be used. A detailed diagram of the systems needed for

Project Icarus is shown in fig. 7.5.

104

The Apollo Guidance Computer Because the Apollo Guid­

ance Computer (AGC) plays a central role in the mission and, besides the telemetry link, is the chief "user" of the com­munications link, a few words about the operation and capa­

bilities of the AGC are appropriate here. The function of the computer is to receive navigation data, either from the space­craft navigation subsystem or from the up-data telecommunica­tions link, to perform the necessary computations and to con­

trol the execution of the guidance instructions generated by the computer or received from the ground.

I nstructions for the AGC consist of 15-bit words. The first 3 bits define the operation to be performed and the next 12,

the address of the memory location involved. To obtain greater accuracy for navigation data, 28-bit double-precision

words are used. Fifteen-bit words are also used for computer

output to the down-link for verification of received data and

for transmitting information about the computer operations

to the ground.

The Telemetry Down-Link The function of the telemetry down-link is 2-fold. First, all spacecraft parameters that must be kept in a specified range, such as power supply volt­

ages and fuel cell pressure, must be transmitted to the ground so that the progress of the mission can be monitored. Second, the telemetry link is used by the guidance computer to verify the reception and/or execution of commands or guidance

instructions. Because of the different data formats used for telemetry data and for the computer, a standard format wiD

be used with address identification.

Communications 105

It will be necessary to use separate PCM clock rates for

each of the 2 rates used in Project I caruso For the 1600 bits­

per-sec rate, a PCM clock rate of 1600 Hz will achieve the

required bandwidth. The subcarrier will be 3200 Hz, and the

range-code rate will be 31,000 bits per sec. For 50 bits per

sec, a clock rate of 512 Hz will be used, which corresponds to

a telemetry subcarrier of 1024 Hz, since the telemetry sub­

carrier is obtained directly by doubling the PCM clock fre­

quency. The resulting output is used to modulate the down­link carrier, which has already been modulated by the range

carrier code. The modulated carrier is amplified and fed through the circulator to the antenna system. (The circulator sends power from 1 port to the next in the direction of the

arrow.) The high-gain antenna can handle 20 w and has a gain of 28 db. For the earth-orbital phase, the output is fed to the omnidirectional antenna (to avoid problems with directing the high-gain antenna) and switched at an altitude of about 15,000 mi to the high-gain antenna.

Up-Data Link The signal from the ground is received and

directed to a demodulator, where the range code and up­data are obtained. The up-data subcarrier, which will be

changed to 1024 Hz, is filtered out and sent to a second

demodulator where the data is extracted and sent to the computer and the command system. When an incorrect for­mat is received, an error signal is transmitted to the down-link for transmission. If the format is correct, a verification

signal is sent instead. The system also can receive up to 64 real­

time discrete commands.

The USB Ground System

The Apollo ground system now consists of 3 stations with 85-ft Cassegrainian-feed paraboloidal antennas spaced approxi-

Project Icarus

MAIN ANT!."NNA 85 FT

ACQUISITION ANTENNA

7.6 USB ground system

106

mately 1200 of longitude apart at Goldstone, California,

Madrid, Spain, and Woomera, Australia. Four land stations and an instrumentation ship cover the launch through the in­

sertion phase. Seven other land stations and 2 other ships

complete the system.

The basic ground system (fig. 7.6) consists of an acquisition

system, a high-gain main antenna, main channel receiver, data

demodulation circuitry, data handling equipment, and peri­

pheral equipment. Acquisition consists of a search in angle

with a separate acquisition antenna and in frequency with the

acquisition channel receiver for the central PM carrier com­ponent of the spacecraft signal. The PM carrier is tracked, the

main antenna acquires the signal, and the main receiver phase

locks on the PM carrier. The main angle-tracking system is

then actuated, and the data handling equipment receives the

data from the spacecraft. The central control system in

Houston monitors the data and generates commands and guidance data for transmission to the spacecraft.

Communications 107

The changes required in the ground system are those re­

sulting from the changes in the data rate and in the sub­

carrier frequencies. The telemetry demodulator has to operate

at 1024 Hz and 32,000 Hz instead of 1.024 M Hz . The sub­

carrier frequency of the up-link must also be changed to

1024 Hz from its present 70 kHz in order to keep the band­

width down. Buffer stages are necessary in the data links to

store the received telemetry before delivering it to the faster

processing equipment, and to slow down the output of the

up-data from the data handling equipment at the low 50 bit­

per-sec rate.

References

1. Foster, L. E., Telemetry Systems (Cocoa Beach, Florida: General Electric

Company, Apollo Support Department, 1963).

2. Goddard Space Flight Center, Proc. of the Apollo Unified S-band Technical Conference (July 1965).

3. Painter, J. H., and G. Hordros, "Unified S-band Telecommunications Tech­

niques for Apollo," vols. I & II, NASA TN 0-2208 and TN D�397.

8

Intercept Monitoring Satellite

Scientific Mission

Objectives From a scientific point of view, it would be un­fortunate if Icarus were to be intercepted and destroyed with­out something being learned of its structure and origin. At present, little is known of the various asteroids except that they do exist and that it is unlikely that they all have a com­mon origin. A determination of the composition of any one of the thousands of known asteroids would, however, shed light on the origin of at least the group from which it came. In particular, learning the composition of Icarus would help resolve whether it is truly asteroidal in origin and, if so, whether the asteroids are rubble from a bro ken planet or, if Icarus is the nucleus of a dead comet, what is such a structure really li ke? Answers to questions such as these would add a great deal to existing k nowledge of the solar system. The Icarus mission offers a unique opportunity to obtain informa­tion that might otherwise not be attained for several decades.

There are also strong engineering reasons for monitoring

Intercept Monitoring Satellite 109

the destruction of Icarus. Only a few nuclear devices have been exploded in space, none of large yield. No actual ob­servational data exists regarding either the performance of a large warhead in space or its effectiveness in destroying a hostile object. It is desirable from this viewpoint to obtain all possible data on the behavior of the bomb itself as well as on its effect on Icarus.

Finally, it is necessary to determine at the earliest possible time whether a successfu I intercept was made and what re­sults were produced. Filling these objectives requires that the explosion be monitored.

Mission Parameters The scientific mission is secondary to the prime goal of destroying Icarus and must not interfere. It was decided, therefore, to develop the monitoring system independent of and parallel to the prime payload. For each nuclear device launched by a Saturn V on an intercept mis­sion, an Intercept Monitoring Satellite (lMS) will be launched by an Atlas-Agena booster on a trajectory that will place it about 1000 mi downstream of the intercept point at the time of the detonation and an estimated 100 mi "southeast" of the original path of Icarus. The IMS will thus be far enough from the explosion to escape damage by the radiation, but close enough to ma ke meaningful observations of the en­counter. It will be beyond the range of the larger solid frag­ments resulting from the disruptive effects of the explosion, but will be within the cloud of small particles and dust, which it can safely sample.

In choosing the spacecraft and the instruments to be flown on this mission, 3 factors are of prime concern. The first is that the high relative velocity of Icarus, about 100,000 fps with respect to the IMS, means that the vaporized and pul­verized material of the asteroid will be within the range of

Project Icarus 110

the monitoring instruments for only a short time. I n order to obtain meaningful data, all of the monitoring instruments must have extremely high data-sampling and data-trans­mission rates.

The second factor of concern is the ability of the Deep Space Information Facility (OS IF) to trac k, communicate with, and issue commands to all the IMSs that will be in inter­planetary space at any given time but still serve as a bac kup for the main communications network. Prior to this mission OS I F has not served more than 1 active vehicle in deep space at a time. If. all launches are successful, OS I F will have 4 satellites to control simultaneously plus its obligation to the main networ k. The I MSs must therefore time-share the ex­isting facilities, and each will be required to operate in a "powered-down," nontransmitting mode for the majority of its flight. In order to insure success of the mission under these conditions, the basic spacecraft for the I MS must be a vehicle that has proved its reliability in space and can function for long periods independently of ground command.

The third factor is the short lead time, which made it un­realistic to consider any system except one that was already· available or could be readily adapted.

Launch Schedule In order to allow maximum flexibility for the launching of the nuclear payloads and to ease the require· ments on the eastern test range facilities, 2 criteria were fol· lowed in determining the launch schedule for the IM Ss: first, no IMS launch should be scheduled within 6 days of a sched­uled Saturn- Icarus launch, and second, the IMS for a particular intercept mission should be launched as far in advance of the Icarus spacecraft as possible.

The launch schedule given in table 8. 1 is based on a low­energy trajectory and assumes a crash program of manu-

Intercept Monitoring Satellite

Table 8.1 IMS launch schedule

Vehicle

IM5-1

IMS-2

IM5-3

IM5-4

I MS·5

launch date

February 27

March 18

April 5

May 1

June 6

111

Intercept monitored

1

2

3

4

6

facture. As indicated, the first 3 IMS lau nches will be well in advance of the first interceptor launch. Due to a com­bination of launch pad restrictions and the proximity of the fifth and sixth intercept missions, the fifth intercep­tion will not be monitored. If, however, the sixth inter­ceptor is not launched, I MS-S will be diverted to the fifth mission. Some back-up capability is provided to each I MS by the one which follows it, in that at the time of launch the planned trajectory can be altered to permit monitoring the interception scheduled for the previous IMS.

Flight Profile On a nominal flight, the IMS will be injected into a 100-nm parking orbit by the Atlas. Then, within 30 min, the Agena will ignite and establish the intercept tra­jectory. After separation, which occurs approximately 1.S hr after launch, the IMS will roll into cruise configuration, its nose pointed toward the sun, and its solar panels extended.

For the first part of the flight, through and including mid­course maneuver, the I MS will transmit telemetry data through its omnidirectional antenna. No scientific instruments will be turned on except for systems checks. During the portion of the day that it does not have access to DS I F, the transmitter and the telemetry system will be turned off, and the space­craft will operate in a powered-down mode.

The first midcourse correction will occur a minimum of

Project Icarus 1 12

7 days after launch for all missions except I MS-5, which has a 3-day minimum correction time. The IMS will be directed toward the center of the 150-mi-radius cylinder of uncertainty of Icarus' flight path, at a point 1000 mi below the inter­ception that it is to monitor. After the midcourse correction, the IMS will resume cruise attitude. At this point the high­gain antenna will acquire the earth and will be used for all succeeding transmissions.

During the next portion of the cruise, the daily trans­mission time will be reduced, and the IMS will spend at least 20 hr per day in a power-down configuration.

If a second midcourse correction is required, it will occur approximately 30 days after launch. Otherwise the IMS will remain in cruise configuration until 1 day before the inter­ception. At this point all systems will be brought to full power, and a final check-out phase will begin.

The I MS will remain pointed toward the sun until approxi­mately 1 hr before interception. At this point preliminary data from the Icarus tracker on the interceptor will have re­solved the trajectory of Icarus into a cylinder approximately 15 mi in diameter. A terminal correction may be made at this time to position the IMS in the path of the outer portion of the debris cloud where only small particles and dust are ex­pected. This "safe" zone is estimated as extending from about 70 to 135 mi "southeast" of Icarus' original trajectory. This maneuver is marginal on propulsive capability and also on time, since the maneuvering sequence requires a nominal 30 min to complete.

Following this maneuver, the IMS will remain on battery power and inertial reference and will turn to orient the in­struments toward the interception point. Sixty sec before the encounter, all navigational optics wi I I be shielded and all scientific instruments will be turned on. These conditions

Intercept Monitoring Satellite

ICARUS \ TRA.JI:CTORY _

POSITION OF IMS

AT TIME OF

INTERCEPTION

8.1

;.--- INTERCEPT

POINT

Position of IMS relative to Icarus at interception (schematic)

113

will be held for 150 sec. The scientific data automation systems will begin accepting data 5 sec before the encounter.

The position of the IMS relative to Icarus shortly after the

detonation is indicated in fig. 8.1. The I MS will pass through

both the plasma cloud and the shower of dust and small­particle debris generated by the detonation. The latter is

expected to abrade and damage exposed surfaces, but since it arrives later than the plasma, and much later than the

radiation, all exposed and vulnerable instruments will have completed their function. After Icarus has passed, the I MS

Project Icarus

will return to cruise attitude and begin an 8-hr data-trans­mission period.

I MS Subsystems

The basic vehicle for the I MS is the Mariner I I spacecraft. Mariner I I yvas an automated, fully attitude-controlled deep space platform whose mission was similar in many respects to the current one. The proposed IMS is shown in fig. 8.2;

1 1 4

it is 9 ft 6 in tall and 5 ft 4 in in diameter with solar panels folded. Each of the solar panels is 60 in long and 30 in wide. The total weight is 540 lb. A weight breakdown by subsystem is given in table 8.2.

Except for changes in the scientific instrumentation, no major modifications of Mariner I I are necessary for the Icarus

Table 8.2 Weight breakdown for IMS (lb)

Transponder 62

Command 8

Power 97

Central computer and sequencer 11

Data encoder 13

Attitude control 53

Structure 70

Actuators 3

Pyrotechnics 4

Motion sensors

Spacecraft wiring 37

Propulsion 34

Thermal control 10

Scientific instruments 46

Debris shielding 50

Adapter 41 -

Total 540

Intercept Mon itoring Satell ite

RUBIDIUM VAPOR MAGNETOMETER

oMNwmEcTIOliAL ANTENNA

FOUBIER SPECTROMETER

II(fi=k�� ��:CA

T��Y

-��:::� SOLAR PLASMA

SPECTROMETER

8.2

- SOLAR 'J/i�;!;;;jjj;#o�'� PANEL

MDICOURSE PIIoPVUIIOtI

MOOVIZ

Intercept monitoring satellite (stowed positionl

115

IMS missions. The various subsystems as used in the original Mariner I I design are described in detail by Wheelock and in

the Mariner-Venus report (1, 2); only a brief description is

presented here, together with such modifications as are necessary.

Structure The lower body of the IMS consists of a hexa­

gonal structure of magnesium-alloy and aluminum-alloy con­

struction (3, 4). The electronics and equipment subsystems of the I MS are contained in 6 rectangu lar modu les, 1 of which is secured to each of the 6 faces of the base. The stiff­

ness of the base is supplied primarily by the ties between the hexagonal platform of the base and the equipment modules.

The superstructure is aluminum-alloy tubing and serves

primarily as a mounting frame for the scientific experiments

and the omnidirectional antenna.

Attitude Control System Attitude control is provided

through a cold gas jet system consisting of 10 cold-nitrogen jets, which provide a 10 pointing accuracy during terminal

Project Icarus 116

maneuvers (5). For normal cruise operation, the spacecraft can be stabilized to within half a degree using 0.001 Ib of nitrogen per day. During cruise, the frame of reference is pro­

vided by 6 sun sensors and an earth sensor on the steerable high-gain antenna. For midcourse maneuvers, an inertial frame of reference is provided by 3 rate-integrating gyros in a strapdown configuration.

Power Subsystem Secondary power for the IMS is supplied by a combination of rechargeable batteries and solar cells (6).

The power profile in fig. 8.3 shows the raw power require­

ments of the IMS. This profile is similar to that of Mariner I I

except that the IMS transmitter and transponder is turned off for a large portion of each day.

The 2 solar arrays, which face the sun throughout cruise,

provide 160 w of power. This exceeds requirements except

for power spikes associated with servomechanisms, which

will be supplied by the secondary battery. The battery is a

8.3 Power profile for I MS

[=:::J SOLAR PANEL POWER 1,,,,,,,,,',,1 BATTERY POWER

HR

Intercept Monitoring Satellite 117

silver-zinc sealed unit which weighs 33 Ib and has a capacity

of 1000 w-hr. It is the sole source of power during all phases of flight that require moving the longitudinal axis of the I MS from the normal sun-locked position.

Propulsion System The propulsion system used for mid­course maneuvers (5) generates 50 Ib of thrust and provides the I MS with a velocity-increment capability of 200 fps, with a minimum increment of 0.7 fps. A mono-propellant, hy­drazine, is stored in a rubber bladder and pressure-fed to the thrust chamber on demand.

Central Computer and Sequencer The central computer and sequencer is a small digital computer and timer whose prime function is to schedule and sequence the operations of the spacecraft subsystems. I n addition it initiates such spacecraft

operations as extending the solar panels and orienting the directional antenna, but the majority of the important opera­tions are initiated by ground command.

Telecommunications System The telecommunications

system (7, 8) consists of 3 separate subsystems: the data en­

coder, the radio subsystem, and the command subsystem. The

purpose of the data encoder is to sample the scientific and

engineering sensors and to convert the measurements into a

7-bit digital format. During the powered-down cruise mode,

the data encoder is inoperative. The radio subsystem consists of a 10-w transponder, a lO­

w R F power amplifier, a high-gain directional antenna, an

omnidirectional antenna, and a command antenna. The trans­ponder is used to provide 2-way Doppler capability and auto­matic angle tracking from the ground. The power amplifier provides the main information-transmission channel for the

Project Icarus

spacecraft. All radio transmissions from the IMS are in the L-band at 960 M Hz. The 10 w of radiated power for the I MS transmitter represents a significant increase over the 3 w of Mariner I I. However, the IMS does not operate on a continuous basis, and battery power can be used to supple­ment solar cell power when the amplifier is operating. The amplifier provides the spacecraft with about 40 bits per sec of data-transmission capacity.

118

The omnidirectional antenna is used for transmission only when the I MS loses earth lock, as, for example, during the launch and maneuvering phases of flight. During the cruise phase and the data-transmission period following the Icarus encounter, the steerable high-gain antenna is used. This antenna has a beamwidth of 16.3° and is equipped with an earth sensor with which it maintains earth lock.

The command system is used only to initiate or terminate certain programmed spacecraft functions. The command antenna is located on the upper surface of 1 of the solar panels.

Scientific Data Automation System (SDAS) The scientific data automation system is illustrated schematically in fig. 8.4. The system is similar to that of Mariner I I (7) except that the high-density data from the Fourier spectrometer does not enter the multiplexer of the data encoder but is fed directly into the tape storage area, where it remains until after the encounter. Because of the high volume of data from the Fourier spectrometer, the tape storage capacity has been in­creased to 1.25 mill ion bits.

Output from the remaining 4 instruments is sampled by the multiplexer and transferred to the tape storage area. After encounter, when the high-gain antenna has regained its earth lock, the data is read from the tapes, processed by the data

Intercept Monitoring Satellite 119

COI\IMA�D }o'U�CTIO�S TEU:Ml::TRY CI{,\?\"Nl::L--_o---J

MOUE L'OM�IAXI) ____ -'

8.4 Scientific Data Automation System

encoder, and then transmitted to earth. At a transmission

rate of 40 bits per sec, the interrogation of the tapes wi II re­quire 8 hr.

The mode command channel is used to activate the entire SDAS for encounter. Command functions are used to turn the instrument on prior to encounter and off following encounter.

Debris Shielding Primary vehicle systems must be provided with a high probability of survival under the hypervelocity impact of debris particles. This survival probability is obtained by stationing the IMS in a region of the debris cloud where particles are small, by locating vulnerable components so that

they shield each other and present a small silhouette, and by using protective shielding.

Project Icarus 120

A weight allowance of 50 Ib is made for debris shielding to

protect exposed critical components. The design condition is a 1.O-g stone particle impacting at 100,000 fps. Such a particle

is about 1 cm in diameter, much larger than the nominal ex­pected in the outer portion of the dust cloud. Assuming that geometric scaling applies, the requirements are met by the

meteoroid shield design of Zender and Davidson for O.Ol5-g meteoroids when scaled upward by a linear factor of 4(9). An 8-in-thick foam- and honeycomb-filled structure is obtained having an outer bumper plate of O.lO-in aluminum and inter­mediate and rear plates 0. 20 in thick. This shield weighs 81bl ft2, providing protection for about 6 ft2.

During the scientific measurement period, the IMS will be oriented with its roll axis transverse to the direction of the interception and the scientific instrument bay pointed toward Icarus. Only partial shielding is needed here as most of these instruments will have completed their function at the time of

arrival of the dust cloud. Debris shielding will be required,

however, in front of each of the adjacent equipment bays and in such a position as to protect vital hardware, for

example, attitude control nozzles and cabling. The solar panels will be edge-on to the impacting debris and are ex­pected to suffer only partial degradation.

Launch Vehicle The launch vehicle for the intercept monitor­ing satellite is the Atlas-Agena, the same booster used with

Mariner I I. The Atlas-Agena can impart to a 540-Ib payload a

characteristic velocity of 39,500 fps (10). This corresponds to

an excess hyperbolic velocity (voo) of 14,600 fps. The tra­jectories for the IMS are low energy trajectories with require­ments well within the power available.

The Voo required for each IMS is given in table 8. 3 along with the Voo required if the failure of a previous I MS during the initial

Intercept Monitoring Satellite 121

Table 8.3 Atlas boost requirements for IMS missions

Spacecraft Launch Oate Intercept Oate voo(10' ft/sec)

IM5-1 0-119.9 0·12 .9 11.5

IMS·2 (nominal) 0·98.9 0·9.9 8 .6

/lM5-1 backup) 0-98 .9 0·12.9 12.0

IMS·3 (nominal) 0·7 8 .9 0-6 .9 6.85

(lM5-2 backup) 0-78 .9 0·9.9 10.57

IMS-4 (nominal) 0-5 2.9 0-4 .9 8.14

(I MS·3 backup) 0-52 . 9 0-6 .9 12.14

IMS-5 0-12 .9 0-0.8 5.0

stages of flight requires that an I MS be diverted to an earlier encounter.

Icarus Destruction Model

Prime Mechanisms I n order to design the scientific instru­ments for observations of the encounter, it is necessary to de­

velop a model representing the mechanisms and the effects involved in the destruction of Icarus.

The energy distribution for a deep space nuclear explosion according to Pierce and G lasstone is given in table 8.4 ( 1 1, 12).

Plasma Temperature In order to estimate the amount of material in the plasma and the mean energy of the particles in the plasma, assumptions were made regarding the size,

shape, and composition of Icarus, the proximity of the detona­

tion to Icarus, the percentage of the thermal energy that re­

mains in the plasma, and the degree of thermal equilibrium attained by the particles in the plasma. The parameters that identify the model used in this analysis are given in table 8.5. Under these conditions, the X-rays from the nuclear device are deposited on a segment of the sphere of Icarus which has

Project Icarus 122

Table 8.4 Fractional distribution of energy in a free space nuclear explosion

Nuclear radiation Thermal radiation

Prompt 'Y rays . 001 X·rays .70

Neutrons .01 Ultraviolet, visible, infrared . 05

Delayed 'Y rays .02

Delayed fJ rays . 02 Kinetic energy of debris .20

a central half-angle of 16° and an arc-length measured along the surface from the point of detonation of 700 ft.

The plasma is actually formed in 2 stages. The first stage, wh ich lasts for a nom inal 0. 1 sec ( 1 1), consists of the pene­tration of the X-ray pulse from the nuclear device into the asteroid and the thermalization of the radiation energy. Be­cause of the short time period involved, the plasma generated in this stage can be considered a nonequilibrium gas having ex­tremely large temperature gradients. Using this "frozen-flow" concept, the size and shape of the initial "plasma crater" can be calcu lated.

The depth of penetration of X-rays into a material is given by

(B.l )

where 10 is the intensity at the surface in electron volts, x is the depth under the surface in centimeters, I is the intensity at depth x in electron volts, Il is the X-ray absorption coef­ficient in cm-1 ( 14). For 0. 12 mev X-rays, the value of the mass absorption coefficient III p, where p = density in g/cm3, for most materials is typically about 0. 13 ( 14); consequently, to estimate the maximum penetration in granite, the value of Il is taken as 0.35 cm-1 •

The size and shape of the initial plasma crater can be de-

Intercept Monitoring Satellite 123

termined by calculating the locus of points at which the gradient of the X-ray intensity is equal to the energy necessary to vaporize granite. The gradient is given by differentiating eq. 8. 1:

�� = -/olle-IJ.x • (8.2)

From the properties listed in table 8.5, this energy density for vaporizing granite can be estimated to be 1.43 X 1022 ev/cm3• The depth of the initial crater is then the depth at which the gradient is equal to 1.43 X 1022 ev/cm3, found by solving eq. 8.2 for x. The resulting calculations are displayed as a crater profile in fig. 8.5. The total initial crater volume is 8.9 X 109 cm3; the mass is 2.4 X 1010 g (7.2 X 1032 atoms).

The second stage of the plasma generation consists of progression toward an equilibrium state as a result of re­radiation. Thus temperature gradients in the initial plasma material are eliminated, and the original plasma crater expands into the cooler asteroid material adjacent to it. It will be as­sumed that enough new material is added to the plasma that

Table 8.5 Assumed parameters for nuclear interaction

Density of Icarus

Mass of Icarus

Radius of Icarus (assumed spherical)

Proximity of detonation to Icarus' surface

Composition of Icarus

Quartz

Potassium Feldspar

Mica

Impurities

Average molecular weight

Average melting point

2 .7 g/cm3

3.8 x 10 15 9

25 00 ft

100ft

granite

21%

50%

9%

2 0%

20

15 00 0 K

Project Icarus

8.5 In itial plasma crater

(IN. )

16

o GROUND ZERO

200 400

its average temperature is reduced by one-third and the

temperature gradients in the plasma itself are largely eliminated.

124

For the model of the encounter given in table 8.5, 35 per­cent of the energy of the explosion falls on Icarus. The energy

of a 100-Mt bomb is 2.6 X 1036 ev, 70 percent of which is

X-rays (table 8.4). The amount of thermal energy 10 striking Icarus is therefore 6.58 X 1035 ev. Of the energy imparted to

the plasma, the majority of it is transferred into the body of Icarus as a strong pressure shock wave, and only a small part remains as kinetic energy in the plasma ( 13). This "small part" will be assumed to be 5 percent; that is, the total kinetic energy of the plasma is 3.29 X 1034 ev. (Of all the assump­tions made, this one is believed to be the weakest.)

The average energy per ion in the plasma from the initial

crater is therefore 44 ev. With the assumption that this value

is reduced by one-third as a result of introduction of additional

material, the average energy of the complete plasma ball is 29.6 ev per ion.

The average energy of the ions is related to the equilibrium temperature by

3 E=-kT

2 ' (8.3)

Intercept Monitoring Satellite 125

where k = Boltzmann's constant = .86 X 10-4 evr K. The equilibrium temperature is therefore calculated to be 226,000° K, roughly the average temperature of the interior

of the sun. This analysis for the equilibrium temperature is probably

correct to within an order of magnitude. The final plasma crater is sketched in fig. 8.6. A rigorous analysis utilizing

methods such as those in the Chemical Rubber Company Handbook implemented in a digital computer program is necessary for precise results (15).

Plasma Dispersion, and Ionization For an average ion energy of e, the Maxwell-Boltzmann energy distribution

function is

</l (e) = exp(-1.5 e'k). (8.4)

The percentage of ions having an energy greater than e' is

11 = exp(-1.5e'/e). (8.5)

This expression can be used to determine what percent of an element is ionized for a given average energy. The ionization potentials and corresponding ionization percentages for e =

29.6 ev for several elements l ikely to be present in Icarus are

(DEPTH OF CRATER MAGNIF1ED 25 TIMES)

ICARUS

DETONATION POINT

8.6 Final plasma crater

Project Icarus 126

Table 8.6 Ionization potential, percent ionization, and dispersion for a 226.000o K plasma

State I c:: "0 0-.- '" III to 'z .!::! c:: N c:: ._ III .2 c:: ... o 0 ?P-Element -Co

Hydrogen 13.5 51

Carbon 11.2 28

Oxygen 13.55 33

Magnesium 7.6 21

Aluminum 5.96 35

Silicon 8.1 22

Calcium 6.1 18

Iron 7.8 23

Nickel 7.6 28

State II - ---_._--c:: "0 0-.- '" CIl tQ ';; N

'2 N c:: ._ III .2 c:: ... o 0 ?P-_Co

24.3 20

34.9 11

14.96 47

18.7 15

16.3 26

11. 8 47

16.2 44

18.2 40

State III c:: "0 0-

.� .� CIl N ",'" '2 Diameter of N c:: ._ CIl .2 c:: ... 70 % dispersion o 0 ?P--Co r:lnlld {mil 5890

47.6 9 1700

54.3 6 1470

1200

28.3 24 1140

33.35 18 1120

50.96 8 935

805

775

listed in table 8.6. Except for hydrogen, each of the elements will exist as a neutral atom and in the singly- and doubly­

ionized states. From the Maxwell-Boltzmann equation it is also possible to

calculate the diameter of the expanding plasma ball as a func­

tion of the individual element and the time after the explosion. Table 8.6 lists the diameter of the sphere that would contain 70 percent of each element present 5 3 sec after the nuclear

detonation, the time at which the I MS will pass the main body of I carus. In order to detect and sample all elements assumed to be present in the plasma with a plasma analyser, the I MS

must pass within 390 mi of Icarus. The selected station of 100 mi will make possible the detection of heavier trace ele­

ments if they are present.

Energy to Charge Ratio of Ions in Plasma The parameter that a plasma analyzer measures is the energy per unit charge ratio of the ions present in the plasma. The energy of the ions in the plasma as measured by the I MS will originate from 2

Intercept Monitoring Satellite 127

sources: the energy due to the relative velocity between Icarus and the IMS, and the thermal energy of the plasma. Since the thermal energy has no spatial orientation, the mean energy, E r of each element will be that due to the 1 OO,OOO-fps velocity of Icarus relative to the IMS. The plasma energy will represent a thermal smear about that point. Seventy percent of the ions of each element will have an energy within 25.4 ev of the mean energy for that element.

Table 8.7 lists the mean energy for each element, the mean energy per unit charge ratio for each possible ionization state, and the percentage of the ions that exist in each state.

Plasma Radiation There are 3 types of radiation emitted by a plasma ( 16): bremsstrahlung (free electron interactions with charged nuclei), cyclotron radiation, and characteristic emis­sion of ions. The first 2 types are determined by the dynamics of the plasma. For particle thermal energy in the 1- to 10-ev range, bremsstrahlung has a central peak at 1000 A and is in­significant in the visible spectrum. Cyclotron radiation is emitted in the far infrared and microwave region.

Table 8.7 Mean energy (E) and energy to charge ratio IE /q) as measured by IMS

Mean State I State II State III Atomic �nergy

Element wt E E/q % E/q % E/q %

Hydrogen 4.7 4.7 100

Carbon 12 56 56 49 28 35 19 16

Oxygen 16 75 75 66 38 22 25 12

Magnesium 24 112 112 31 56 69

Aluminum 27 127 127 4 8 63 20 42 32

Silicon 28 132 132 34 66 39 43 27

Calcium 40 188 188 25 94 64 63 11

Iron 56 260 260 34 130 66

Nickel 58 270 270 41 135 59

Project Icarus 128

This leaves a window in the visible spectrum through which the plasma can be scanned in order to detect characteristic emission lines of the elements present. The limitation of observing the spectral emissions of the plasma is that at a distance of 1000 mi the glow will be faint and cannot be expected to radiate for more than a few seconds after the ex­plosion. Furthermore, conventional grating spectrometers can accept only a small percentage of the incident light and will take nearly 60 sec to scan the spectrum. Both of these limita­tions can be overcome by the use of Fourier spectrometers.

A secondary property of plasmas is the strength of hydro­magnetic shocks induced in them by the flow of the charged particles ( 17). A magnetic profile would thus indicate the average energy of the particles.

Nature of the Debris Cloud During the process of cratering treated in chapter 3, the material of Icarus is pulverized and shattered by 2 processes: the shearing flow, which occurs in the interior, and the rarefaction process, which occurs at the free surface and is responsible for ejection of the debris. Both of these processes are most severe initially, as is the velocity of ejection. Since the crater is produced on the sunlit limb of Icarus, the resultant debris will distribute itself in a fan ex­tending "south-easterly" from the original Icarus trajectory, with the smallest particles in general moving at the highest velocity.

It is assumed that the first 1 percent of the mass of material ejected from the crater is reduced to dust and fine particles in general no larger than a millimeter in dimension. Using the relations for mass flux and velocity developed in eqs. 3.6 to 3.8, it is found that 1 percent of the flow occurs in a period tIT = 25/37.2 = 0.67, at which time the mean ejection velocity component normal to the surface of Icarus is 2. 1 km/sec. The

Intercept Monitoring Satellite 129

initial value of this velocity component is 4. 1 km/sec. Thus 53 sec after the detonation, when the debris cloud sweeps by the IMS, it will extend some 135 mi from Icarus' original trajectory. The region between 70 mi and 135 mi will contain the first 1 percent of the total crater mass, which, because of its subjection to the most severe of the destructive processes, will be comprised of only the smallest particles. It is in this region that the I MS must be stationed. The total flux that it will intercept will depend on its exact station but will be of the order of 1 g/ftl.

Although the assumptions used above are believed to be reasonable, they only illustrate a method of approach. Actual design requires a digital computer analysis of the type referred to in chapter 3 and possibly experimental studies in order to establish the characteristics of the debris cloud with the necessary accuracy.

Scientific Instrumentation

In choosing the instruments to be flown on the IMS, it is necessary to restrict the choice to "off-the-shelf" hardware. With the exception of the Fourier spectrometer, all of the in­struments selected have been flown on at least 1 satellite. The spectrometer, however, was designed for space use and has been tested extensively in a space environment.

The positioning of the instruments on the IMS is indicated in fig. 8.2. The weight and power requirement of each instru­ment is given in table 8.8.

The interface between each instrument and the IMS is made by the S DAS (fig. 8.4).

Solar Plasma Spectrometer The solar plasma spectrometer is basically the same instrument as that flown on Mariner I I ( 1, 2, 18, 19). It consists of 2 curved deflection plates with a

Project Icarus

Table 8.8 Scientific instruments Gamma ray detectors {21

Rubidium vapor magnetometer

Interplanetary dust collector

Solar plasma spectrometer

Fourier spectrometer

Total

10 lb

6

5

7

18

46 1b

13 0

0.5-w

6

0.5

1

8

16.0 w

voltage placed across them which allows only ions with a pre­selected energy-to-electric-charge ratio to reach the Faraday cup charge collector (fig. 8.7). The charge across the plates is changed periodically to admit a new energy-to-charge ratio. The radii of curvature of the inner and outer plates are 1.0 and 1.3 in respectively. For these radii, the energy per unit electron charge admitted by the deflection plates is the same as the potential applied across the plates, with a maximum error of 12 percent.

The timing cycle set by the programmer is such that each of the 8 step levels of the instrument is sampled sequentially for 3 sec, with a 0.5-sec readout period between each step level.

Table 8.9 Solar plasma spectrometer energy levels

Potential Energy of across de- particles

Programmer flection reaching Ions step level plates {vI Cup {evl sampled

- 18 + 18 electrons

2 - 12 12 electrons

3 - 6 6 electrons

4 + 4.7 4.7 HI

5 + 6 0 6 0 C I; M g II; AI II; Si I I; Ca III

6 + 4 0 4 0 o II; AI I I I; S i I I I

7 +13 0 130 Al l; Si I; Fe II; Ni I I

8 +26 0 260 "Fe I; Ni I

Intercept Monitoring Satellite 131

The spectrometer thus samples each programmed energy level once every 28 sec. The energy levels to be sampled, along with the ions expected to be found at each level, are listed in table 8.9. The electron energy distribution derived from the first 3 levels can be used to obtain the exact equilibrium temperature of the plasma, and thus to determine more accurately the ex­pected detection rate at a particular energy level for each ele­ment as a fu nction of the density of that element.

At each energy level, the relative density of the elements present varies also as a function of the distance of the I MS from Icarus; that is, the relative density of the lighter ions to the heavier ions increases as the distance from Icarus increases. If the IMS passes sufficiently near the center of the plasma cloud, each step level will be sampled up to 4 times. How­ever, the solar plasma spectrometer cannot distinguish be­tween different ions having the same value of E/q. It must be used therefore in conjunction with the Fourier spectrometer.

Fourier Spectrometer The second of the 2 primary instru­ments to be flown on the I MS is the Model P-4 Polarization Interferometer Spectrometer developed by Block Associates (20, 2 1). This spectrometer is one of a new series developed in the past few years that does not employ a grating or a prism to disperse the light but splits the incoming light into 2 parts, varies the path length of 1 of the parts, and then re­combines the light. As the ratio of the 2 paths changes, a record of the interference between the 2 beams (an inter­ferogram) is produced by photomultiplier tubes. A Fourier transform of the interferogram can then ultimately be used to resolve the wavelengths of the light present to within 2 percent. On this mission the interferograms will be trans­mitted to earth and processed at a later time.

The operation of the spectrometer is illustrated schematic-

Project Icarus

DEFLECItON PLATES 1----

1---- FROM DCS

CAUBRATION COMMAND

5 .10100l1li5

8.7 Solar plasma spectrometer

POLARIZATION

8.8 Fourier spectrometer (schematic)

>-----_TODCS ELECTROMETER AMPUJIla

PHOTO­Mt:LTIPLIER ruBE 2500·6500 11

POLARIZA TJON

132

Intercept Monitoring Satellite 133

ally in fig. 8.8. I ncident light is polarized into a coherent beam at an angle of 450 to the transmission axes of a birefringent quartz compensator. The compensator repolarizes the light, half into the vertical plane, "0," and half into the horizontal plane, "E." Because of the properties of birefringent quartz, light polarized in the "E" plane travels slower than light polar­ized in the "0" plane so that the compensator and the quartz chip introduce a path difference between the 2 beams of light. This path difference is changed by varying the effective thick­ness of the compensator by sliding it across the beam of light. Thus each trip of the compensator across the beam, corres­ponding to 1 scan of the spectrum and the output from the photomultiplier tubes, comprises 1 interferogram. When the light is recombined by the second polarization plate, it is non­coherent because of the phase shift between the 2 component beams. The light intensity is then a function of the destructive interference between the 2 components.

The prime advantage of this device over conventional spectrometers is that it is not energy-limited. A conven­tional spectrometer accepts only light that enters throuf.jh a narrow slit, and it disperses this light over a plate several cm long. The Fourier spectrometer accepts all light entering a cone with a half angle of 1.50 and has an energy threshold between 1 and 2 orders of magnitude lower than ordinary devices.

In order to obtain 2 percent accuracy in energy resolution, at least 40 good interferograms of the radiating plasma must be obtained, and at least 100 are desirable. With a scan rate of 3.75 interferograms per sec, 100 can be obtained in 27 sec, but only 39 percent of each scan provides usable data.

The light spectrum monitored by this instrument runs from 2500 A to 25,000 A - from visible light to the near infrared. The output of the spectrometer is digital; a sampling clock samples

Project Icarus 134

the output of the photomultipliers at a frequency of 1 2,000 Hz, which corresponds to 4,680 data points per sec. If each data point is converted to a 7-bit word, 2 1,900 bits of storage must be provided by the tape recorder in the SDAS for each sec that the spectrometer is operating. The spectrometer will operate for the entire 53 sec between the detonation and the time IMS passes Icarus for a total of 1, 160,000 bits of storage required. The Fourier spectrometer has 2 basic parts: the 8-lb optical head, which is mounted above the hexagonal base of the I MS, and the electronics, carried in the scientific equ ip­ment bay.

Gamma Ray Detectors The IMS utilizes 2 type M-3 gamma ray scintillation detectors as used in the Vela Hotel nuclear detection satellites (22, 23). I ncident radiation is absorbed by the plastic in the scintillation case (fig. 8.9), and the energy is then re-emitted as photons of light, a fraction of which are detected by the photomultiplier tube. The output of each detector is proportional to the intensity of gamma rays with an energy greater than 0.3 mev.

The detectors have no preferred orientation and can be mounted inside the spacecraft as desired. Two detectors are used to provide more reliable data. Each detector is 3.75 in in diameter and 4.6 in long.

Rubidium Vapor Magnetometer The rubidium vapor mag­netometer is used to determine absolute strength and orienta­tion of the magnetic field. It has been flown on the Inter­planetary Monitoring Platform ( I MP- 1) and on the Ranger spacecraft (5, 22, 24). Its range of sensitivity is from 0.05 gammas to 105 gammas ( 1 gamma - 10-5 gauss) with an ac­curacy of 0.05 gammas. The strength of the magnetic field in interplanetary space near earth is between 4 and 6 gammas.

Intercept Monitoring Satellite

PLASTIC SCINTILLATOR CASE

PHOTOMULTIPLIER Tt'8E

8.9 Gamma ray detector

POWER SUPPLY

135

Since it is expected that the magnetic field strength may vary rapidly as the IMS passes through the debris from Icarus,

an instrument has been chosen that is capable of measuring rates as high as 30 gammas per sec. The output of the mag­netometer is analog, but is digitalized by the SDAS at the rate

of 60 7-bit words per sec.

Interplanetary Dust Collector Flown on the Orbiting Geo­

physical Observatory (OGO ) and Mariner IV, the inter­planetary dust collector is able to determine the velocity, mass, and direction of particles as small as a micron in diameter (22, 25). It consists of 2 thin-film capacitors spaced 1 0 cm apart to provide time of flight measurements and a lead

zirconate transducer placed behind the second capacitor to

provide a measure of the momentum of each impacting particle. For each event recorded the output consists of 2

timing marks and an electrical pulse. The magnitude of the

pulse is a function of the particle momentum. The original design utilizes 3 detectors pointing in mutually

orthogonal directions, each of which has a half-angle of 6°. For the current application all detectors will be oriented toward Icarus. It appears that the sensitivity of 1 or more of

Project Icarus 136

these units should be reduced by increasing sheet thickness and the detector function should be combined with debris shielding. Data obtained with this instrument will provide a sure confirmation of an intercept. I n addition, by correlation of data on particle size with cratering studies and location, some indications regarding the structure of Icarus may be derived.

Summary The complement of instruments described will satisfy minimum requirements for information regarding the nuclear explosion in space and the composition of Icarus and also will provide confirmation of an intercept. More detailed study and possibly some instrument design is needed, however, to fully exploit the scientific opportunities which the inter­cept of I carus affords.

References

1. Wheelock, H. J., ed., Mariner, Mission to Venus (New York: McGraw-Hili, 1963).

2. Mariner-Venus 1962 Final Project Report, NASA SP-59, 1965.

3. Jodele, J., "Mariner Spacecraft Packaging," CI T-JPL TR 32-451 (July 1, 1963).

4. Adams, J. L., Space Technology: Volume I/, Spacecraft Mechanical Engineering,

NASA SP-66 (1965).

5. Corliss, W., Space Probes and Planetary Exploration (New York: D. van Nostrand Co., 1965).

6. Costogue, E. N., "Mariner Venus Power-Supply System," CIT-JPL TR 32-424 (March 30, 1963).

7. Bryden, J. N., "Mariner (Venus '62) Flight Telecommunication System," CI T-JPL TR 32-377 (January 15, 1963).

8. Martin, B. D., "The Mariner Planetary Communication System Design," CIT-JPL TR 32-85 (Revision #1) (May 15,1961).

9. Zender, G. W., and J. R. Davidson, "Structural Requirements of Large Manned Space Stations." In A Report on the Research and Technological Problems of

Manned Rotating Spacecraft, NASA TN 0-1504 (August 1962).

10. NASA Office of Space Science and Applications, Launch Vehicle Estimating

Factors (December 1967), fig. IV-A-1.

Intercept Monitoring Satellite

11. Pierce, E. T., "Nuclear Explosion Phenomena and Their Bearing on Radio Detection of the Explosions," Proc IEEE 53 (1965): 1995-200B ..

137

12. Glasstone, S., ed., The Effects of Nuclear Weapons, U.S. Atomic Energy Com­mission (April 1962): 11,14,24,26,29,6 9,70, 7B, 351, 399-4 02.

13. "Nuclear Physics: How to Zap an ICBM," Time (May 26,1967): 46.

14. Chandrasekhar, S., Radiative Transfer (New York: Dover, 1960), pp. 354 ff.

15. The Chemical Rubber Company Handbook of Chemistry and Physics, 46th edition, (1 965) p. E-ll 0.

16. Uman, M. A., Introduction to Plasma Physics (New York: McGraw-Hili, 1964).

17. Glasstone, S., ed., Sourcebook on Space Sciences (New York: D. van Nostrand Co., 1965).

lB. Josias, C., and J. lawrence Jr., "An Instrument for Measurement of Inter­planetary Solar Plasma," CIT-JPl TR 32-4 92 (May 1, 1964).

19. Wyckoff, R. C., ed., "Scientific Experiments for Mariner R-l and R-2," CIT-JPl TR 32-315 (July 15, 1962).

20. Block Engineering, Inc., "Model P-4 Polarization Interferometer:: cambridge, MA.

21. low, M. J. D., "Subtler Infrared Spectroscopy," International Science and

Technology (February 1 967): 52-68.

22. Richter, H. l., ed.,lnstruments and Spacecraft, NASA SP-302B.

23. Singer, S., "The Vela Satellite Program for Detection of High Altitude Nuclear Detonations," Proc IEEE 53 (1965): 1935-1 948.

24. CIT-JPL, "Scientific Experiments for Ranger 1 and 2," TR 32- 55, (January 3,19611.

25. Sloan, R. K., "The Scientific Experiments of Mariner IV:' Scientific

American (May 1966): 62-72.

9

Management and

Economic Impact

Introduction

Management of a systems effort of the magnitude, scope, and

critical importance of Project Icarus is an exceedingly difficult

assignment. The project requires taking over and coordinating

the efforts of a sizable portion of the nation�s manufacturing

capacity. The scope is such as to include all facets of our

economy and all sections of the continental United States,

plus a world-wide tracking network and Navy ships at sea. The

importance cannot be over-emphasized; to fail or to be late

wou Id be disastrous to the nation and the world. The success

or failure of the entire Icarus project may well hinge upon

the quality of management; in fact, such a complex systems

project must be considered certain to fail without enlightened

management. This chapter sets forth the method of attack

employed during the first 14 weeks of the 70-week effort

from inception in February 1967 to rendezvous of the Saturn­

Icarus space vehicles with Icarus in June 1968. A discussion

Management and Economic Impact

is included of the cost of this effort both in dollars and in

terms of the impact upon the national economy.

Scheduling and Coordination

139

Perhaps the most difficult and at the same time most im­

portant function of systems management is that of scheduling

and coordinating the myriad interrelated activities of a number

of more or less autonomous groups. Project Icarus was no ex­

ception in this regard. Initially 7 groups were formed: orbits

and trajectories; launch vehicles and propulsion; payloads;

space vehicles; navigation, guidance, and control; communica­

tions; and planning and management. It should be noted

parenthetically that, since one of the principal objectives of

the project (from the student's point of view) was to obtain

experience in working together and coordinating diverse ef­

forts on a systems project, the management function was

exercised as a coordinating and scheduling role with major

emphasis on interface problems and overall system implica­

tions. Decision-type actions, per se, were not originated by

this group, but rather were the result of joint technical group

agreements.

The primary management tool employed was PE RT (Pro­

ject Evaluation and Review Technique). The PERT network

for the entire 70 weeks of Project Icarus is presented as figs.

9.1 through 9.4, each figure showing a 15- or 20-week portion

of the total project activity. The circles in the figures repre­

sent events or milestones that occur at a specific time. The

solid lines connecting event circles represent activity that

must be completed in progressing from one milestone to the

next. The dashed lines represent the dependence of one line

of activity upon another. The events and activities are com­

bined into a network that illustrates their interdependence

�+---+---+--+--I.

I . i

-;n � ., �

III

� .. 2 '" � .. (J ., '0 Q:

0 0 �

Ii � 1=11 ( el� '" .. ( .r: .. (J .. h I-:a II: '"';w cnn.

'" J,i I� r

'"

�. p I �

-=

� �

I i . I a

I � I I I I I f i I 6

I ,i = II i

'" -

i :

N

I l \ f P � I� �

I� I �K; -\ip I� �D I 'I I I

'" ""'I

l� I 81

- Ii __ _ -'-_ J1I1 ______________

::: --------------... ----'"

-;;; � .. i

10 C? 10 ... .. ::J

] ... u .. 0e-0.

5 ... � III

� u I-a:: I'\w

ClIo.

.. u '" '0 0: .. .. co

J::. U

� a: '1w

�Q"

-;;; .>t ell ell

0 ;: '" 0

,., It) !£ .. :l

� .... <.> ell .0' tt 5 .... .... .... m

-= '" <.> '" r-" a: ... --'--- .". W ... ai 0.. :>

Project Icarus 144

and the sequence in which they must be completed in order

to reach the end objectives.

PERT for Project Icarus was not a true PERT because of a

definite inescapable completion date whose alteration could

not be tolerated. Therefore, rather than using estimated

times to complete the various activities and to arrive at a

critical path and completion date, the management group

worked and reworked the Project Icarus network. After re­

solving a number of conflicts, but always holding the end date

firm, the Project Icarus team developed a complete network,

which contained all required activities and met all milestones.

Thus, the Icarus PERT made its contribution to the coordina­

tion and eventual successful completion of the project.

Figure 9.5 is an expansion of the "system" activity line of

the PE RT network, covering only the last 20 weeks of the

project. This figure is included for 2 reasons. First, it illustrates

graphically the launch sequence of the 6 Saturn-Icarus vehicles,

the timing involved, and the various rendezvous dates; second,

it shows in some detail the very complicated interdependence

of a critical series of events and activities.

Impact upon the National Economy

Cost in dollars of an effort of the scope and magnitude of

Project Icarus is not only next to impossible to estimate but

probably relatively meaningless, once attained. Of more im­

portance, and more readily grasped, is a qualitative measure

of the impact of the project on the economy of the nation

and upon other activities of national importance.

The assumption was made that Project Icarus require­

ments, whether materials, manpower, or funds, had the high­

est priority, and any other efforts could be delayed, taken

over, or cancelled if necessary. The usual contracting delays

were by-passed by issuing letter contracts to vendors as soon

Management and Economic Impact

WEEK DATE

.. I

SI-7--f.--( (1-0-1\1

SI-3 --f.---�; (I-O-'.t)

iii I

II I

Ant ...

.. . .. ., 11

SI-I --+-----+-----4 .... )-------+ .......... (1-0-0.1, A ...

9.5 PERT chart for Saturn-Icarus launches

145

"aU

as requirements became known; conventional incentive con­

tracts were then written after the work was under way_ The

National Aeronautics and Space Administration's (NASA's)

space program, including all efforts on Project Apollo and the

Apollo Applications Program, was completely diverted to

Project Icarus. Saturn booster production was accelerated to

make a total of 9 Saturn V's available by April 1968 instead

of the 6 originally scheduled to be completed by that time.

Construction was authorized in early March 1967 of a third

launch pad, 39C, to accommodate the tight launch schedule

required by Icarus. Predicted completion date of this pad is

early March 1968, a 12-month construction effort. A number

Project Icarus 146

of additions and changes are being made to the existing World

Tracking Network. Specific Icarus flight hardware is discussed

elsewhere in this book; wherever possible it consists of ex­

isting equipment and flight modules, modified only to the

extent deemed mandatory due to the nature of the Saturn­

Icarus mission.

The NASA Fiscal Year 1967 and 1968 budgets for Apollo,

Apollo applications, and tracking and data acquisition total

approximately $7.5 billion. Although Project Icarus does not

encompass every facet of these NASA efforts, the additional

costs of accelerating schedules and modifying hardware can­

not be discounted. Therefore, the dollar cost of Project

Icarus is estimated to be $7.5 billion or approximately 1 per­

cent of the gross national product.

Of equal importance to the nation is the impact of Project

Icarus upon other programs, particularly NASA's Apollo pro­

gram. Since the Apollo efforts are essentially diverted entirely

for over a year, and since the first 9 Saturn V boosters (3 tests plus 6 operations) are consumed by Project Icarus, it is

conservatively estimated that the Apollo program will �e de­

layed 3 years.

It should be pointed out that not all Project Icarus efforts

are without benefit to other programs. Although the Saturn­

Icarus vehicle is not man-rated, much valuable experience will

be gained in designing, fabricating, assembling, and launching

9 such vehicles and in controlling and monitoring 9 such

flights. Furthermore, invaluable scientific information will be

acquired in the course of the operation.

10

Mission Evaluation

In designing a system under emergency conditions with no

cost restrictions, probability of success is perhaps the only

important measure of performance. For Project Icarus, how­

ever, estimates of the physical characteristics of the asteroid

and of the effect of the nuclear bomb on the asteroid were

so speculative as to prohibit the evaluation of different con­

cepts using a probability model. Moreover, proper treatment

of "partial successes" was viewed as a very difficult and sub­

jective problem.

The following combinatorial analysis is presented, however,

to provide an estimate of system reliability and, hopefully,

confidence in the feasibility of the project. The use of this

analysis in a more detailed system design would require valida­

tion of the estimated performance of the bomb and of the

physical characteristics of Icarus, or at least the modeling of

these quantities by a suitable probability distribution.

The discrete probability model used to evaluate Project

Icarus involves estimates and assumptions concerning both

Project Icarus

hardware performance and the effects of the detonation.

These estimates and assumptions follow.

Hardware Performance

148

The performance of any Saturn-Icarus vehicle or of any sub­

system within a vehicle is assumed independent of the per­

formance of its predecessors. Specifically, the estimated

reliabilities are as follows:

Boost system

Spacecraft systems

Guidance system

83 percent for all missions;*

85 percent for SI-4 through SI-7, 98 percent for SI-8 and SI-9;

85 percent for SI-4 through SI-7, Icarus intact,

70 percent for SI-4 through SI-7, Icarus in fragments,

95 percent for SI-8 and SI-9, Icarus intact,

90 percent for SI-8 and SI-9, Icarus in fragments.

As discussed in chapter 2, the last 2 ( low-altitude) shots

are expected to be more reliable and more accurate than the

others because of their shorter flight duration and proximity

to earth. Similarly, pursuing an intact Icarus rather than frag­

ments of the original body is assumed more reliable because

of the larger target and slower rotation rate. Incorporation

of the above reliability factors yields the following chance of

success for any one vehicle:

60 percent for a high-altitude shot at Icarus,

49 percent for a high-altitude shot at a fragment,

" In the actual mission plan, vehicles 51-4 through 51-6 are each backed up by the

following scheduled vehicle. This procedure increases booster reliability for the

first 3 missions, but decreases it for the fourth. The actual booster reliabilities for

the first 4 missions are 97 percent, 95 percent, 93 percent, and 48 percent,

respectively. This back-up strategy makes deflection of Icarus more likely since

deflection is the primary goal for 51-4 through 51-7.

Mission Evaluation 149

77 percent for a low-altitude shot at Icarus,

73 percent for a low-altitude shot at a fragment.

Effects of Detonation

The first vehicle from the group SI-4 through SI-7 to attack

Icarus successfully will either break the asteroid into many

pieces or deflect it from its collision course. Since the required

deflection impulse increases as the altitude above earth of

detonation decreases, the probability of deflection for each

succeeding mission must decrease. The assumed deflection

and fragmentation probabilities for successful high-altitude

missions are given in table 10.1.

If a successful deflection is accomplished, the remaining

spacecraft will be destroyed in flight to eliminate the existence

of undetonated nuclear bombs in space and to prevent any

fragments from being deflected accidentally back onto a

collision course. Success of a mission will be determined

either from spacecraft measurements or from terrestrial

telescopic observations. Resolution of individual pieces will

be possible when they have drifted apart by a few seconds of

arc. For the first mission, 1 second of arc corresponds to some

100 mi of separation. Thus, resolution of fragments should be

simple just 1 day after detonation, provided that observations

are possible. Succeeding spacecraft will be used to destroy

fragments that remain on a collision course.

Tabla 10.1 Effect of successful high-altitude missions

Mission

SI-4

SI-5

SI-6

SI-7

Probability of

fragmentation

.20

.40

.57

.80

Probability of

deflection

.80

.60

.43

.20

Project Icarus 150

The preceding assumptions permit the con struction of a

probabi I ity tree shown in fig. 10.1. The n etwork divides when

the outcome of a mission is subject to chance an d joins when

an equivalen t state may be reached by an y of several paths.

The possible resu Its of any mission depend on the outcome

of the preceding missions. By assigning the previously dis­

cussed probability values at each branch poin t and summing

over equivalent outcomes, one obtains table 10.2. By using low-altitude-mission reliability estimates only, the probabilities

of results of a 6-low-altitude-mission plan were obtained for

comparison (table 10.2),

Final Evaluation

The probabilities discussed in this chapter may be incorporated

in to an estimate of average overall performance if some weight­

in g factor is assigned to the "desirability" of each outcome.

Arbitrarily it was decided that deflection of Icarus should

Table 10.2 Probability of mission results

Probability

Actual mission Six low-altitude

Results plan missions

Six failures .0014 .00015

Fragmentation only .015 .0067

Fragmentation and destruction of 1 fragment .054 .036

Fragmentation and destruction

of 2 fragments .100 .20

Fragmentation and destruction of 3 fragments .076 .27

Fragmentation and destruction of 4 fragments .033 .33

Fragmentation and destruction

of 5 fragments .0068 .16

Deflection .714 .0

Mission Evaluation 151

10.1

Probability tree

Project Icarus 152

carry a "most desirable" weighting factor of 1.0, while the

"not so desirable" result of fragmentation would have a weight­

ing factor of 0.3 with an additional 0.1 for each fragment des­

troyed. The weighting factor may also be interpreted as an

estimate of the relative reduction in damage on a scale from

o to 1. Clearly, then, a "miss" has a weighting factor of O.

By tabu lating the probabilities of table 10.2 as a function

of the corresponding weighting factor, one obtains the

probability distribution of the weighting factor, that is,

the probability distribution of reduction in damage rela­

tive to the damage that would result from a collision. The

mean of that distribution for the proposed plan of Project

Icarus is 0.86.

If this mean value is interpreted as the estimated average

overall performance of Project Icarus, then one can expect

an 86 percent reduction in damage due to the efforts of the

project team. Perhaps even more encouraging is the 71 per­

cent chance of no damage at all because of the 71 percent

chance of deflection. But regardless of the probability

associated with the success of Project Icarus, its cost and

sociological impact are clearly insignificant in light of the

staggering alternative-disaster.

Credits

The following is a list of the authors primarily responsible for

the contents of each chapter under the guidance of Louis A.

Kleiman, editor.

Chapter 1, Richard W. Heldt, Louis A. Kleiman, and

Theodore C. Tenny. Chapter 2, James A. Fletcher, Richard

W. Heldt, and Charles S. Marantz. Chapter 3, Noel S. Flynn,

Roberto Hukai, Shivaji S. Seth, and Akio Suzuki. Chapter 4,

Noel S. Flynn, Patrick C. U. Mbanefo, and Akio Suzuki.

Chapter 5, Joe R. Deichman, William R. Lange, John H.

Lindley, and George W. Wagner. Chapter 6, Geoffrey K.

Bentley, Dennis E. Kalla, Louis A. Kleiman, Richard J.

Labrecque, Clifford A. Rose, Jr., and Theodore C. Tenny.

Chapter 7, Chester J. Wolejsza, Jr. Chapter 8, Harold L. Jones.

Chapter 9, Fred H. Baughman. Chapter 10, Richard W. Heldt.

Project History

Shortly before the beginning of the spring term, 1967, at

M.I.T., an announcement of course 16.74, Advanced Space

Systems Engineering, appeared on bulletin boards through­

out the Institute. To convey its impact, we reprint that

announcement verbatim.

Mission to Icarus

In June 1968, the asteroid Icarus, a dark boulder a mile or so in diameter, will pass by earth at a relative velocity of about 100,000 fps and a distance of 4 million miles. The orbit of Icarus has been quite accurately established, and the chance of it approaching much closer is nil. However, 4 million miles is an uncomfortably small miss distance in the scale of the solar system, and there have consequently been several speculative articles concerning the possibility and consequences of a collision with earth.

The project to be handled by the Advanced Space Systems Engineering students this term assumes that Icarus will, in fact, collide with the earth.

Such catastrophic events have apparently occurred a num­ber of times in the past. About a score of crater-like features on the surface of the earth have been positively identified as

Project Icarus 156

resulting from a meteorite impact. Some of these-such as the Barringer Crater in Arizona-occurred in recent geological times. Perhaps a hundred more circular, rimmed depressions are suspected of being meteorite craters; this list includes such giants as the Richat Structure in Mauritania (75 miles), the Vredevoort Ring in South Africa, and Manicouagan Lake in Canada (40 miles).

The impact of I carus would produce a crater only 10 or 15 miles in diameter. Its effects would be felt worldwide, however. The enrgy involved is the equivalent of 500,000 megatons of TNT-two orders of magnitude above that in­volved in the largest recorded earthquake, and four or five orders of magnitude more than Krakatoa. If the strike oc­curred in midocean, tsunamis in the lOCHoot category would cause worldwide damage. If the strike occurred on land, the blast wave would level trees and buildings within a radius of several hundred miles, and some lOs tons of soil and rock­dust would be thrown into the stratosphere, where for several decades it would act to reduce the solar radiation ordinarily received at earth's surface and threaten the triggering of an ice age.

Clearly, Icarus must be stopped. No effort or funds will be spared in carrying out the detailed plan to be developed by the crack team of scientists and engineers assigned to the project. Costs, of course, must be minimized, but the major limitation is time-the program must use existent space tech­nology and hardware, and it must succeed. Because of the in­flexible schedule, and certain other mundane reasons, a final report will be required by May 23, 1967.

The problem solution may utilize a rocket to intercept the asteroid and nudge it from its course. Alternatively, it may be better to reduce it to rubble with a nuclear warhead. Multiple booster vehicles and rendezvous may be necessary to meet payload requirements. Gemini and Apollo hardware may be utilized if a manned space system enhances prob­ability of success. These and many other alternatives must be considered. The group will first make preliminary studies of different approaches to the problem, subject them to systematic review and evaluation leading to selection of a preferred mode and the establishment of mission specifications.

The development of the selected system will include the following separate but interrelated design exercises:

Project History 157

Booster Systems: Capabilities and characteristics of booster rockets and interaction with the present payload and mission.

Orbits and Trajectories: Development and analysis of orbital operations, including rendezvous, coasting, and thrusting maneuvers, to establish propulsion requirements and system capabilities. Space Vehicle: Vehicle configuration, subsystem arrange­ment, and structural design; functional requirements, en­vironment, and loads; vehicle propulsion systems and performance.

Guidance and Control: Navigational and guidance systems and control system analysis and design for payload stages; mission requirements, disturbances, sensors, rendezvous, midcourse and terminal maneuvers.

Payload Systems: Special equipment and systems, performance, analysis, and related design; manned systems (if included), provisions and requirements.

Communications: Vehicle-earth telemetry, communication, command and control links; on-board information systems, sensors, and data processing.

Auxiliary Povver: On-board secondary power supply systems to operate communication systems, vehicle controls, and other payload systems.

None of these design problems can be treated independently of the others; each places demands on several others, and the design requirements for each can be made specific only after the entire project is completed. Such a complex design prob­lem requires a systems study: a design exercise which treats not only the various subsystems of a vehicle but the problems of integrating them into a harmonious, efficient whole.

In tackling the problem, the students will be divided into a number of project groups, each concerned with one of the study areas. Students may elect their field of interest. Each group will elect a group leader, who will meet with other leaders to define system interfaces, and on this basis will establish guidelines for his group. Group leadership will be rotated during the term. Technical aspects of the study will be supported by lectures given by staff specialists and by guest lecturers from industry.

Project Icarus 158

The reaction of many students to this announcement was

typified by such comments as "How about building a big

trampoline?" and "Why not move the earth out of the way?"

And it was perhaps with the same skeptical, and almost

cynical, attitude that twenty-one of us registered for the

course and anxiously awaited our first glimpse of the pro­

fessor who dared to propose such a study. That first glimpse

of Professor Paul E. Sandorff as he entered the classroom did little to change our attitudes, but working with him soon

turned our skepticism into respect.

With the aid of a cooperative and enthusiastic staff, we

organized seven groups, modified only slightly from those

described in the course announcement, and chose group

leaders periodically throughout the term. The members of each group and their departmental affiliations are as follows: Orbits and Trajectories

James A. Fletcher (Department of Aeronautics and Astronautics)

Richard W. Heldt (Department of Aeronautics and Astronautics)

Charles S. Marantz (Department of Aeronautics and Astronautics)

Nuclear Payloads

Roberto Hukai

Shivaji S. Seth

Boosters and Propulsion

(Department of Nuclear Engineering)

(Department of Nuclear Engineering)

Noel S. Flynn (Department of Aeronautics and Astronautics)

Patrick C. U. Mbanefo (Department of Aeronautics and Astronautics)

Akio Suzuki (Department of Aeronautics and Astronautics)

Spacecraft

Joe R. Deichman

Harold L. Jones

William R. Lange

John H. Lindley

George W. Wagner

Guidance and Control

Geoffrey K. Bentley

Dennis E. Kalla

Louis A. Kleiman

Richard J. Labrecque

Theodore C. Tenny

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

(Department of Aeronautics and Astronautics)

Project History 159

Communications Clifford A. Rose, Jr. (Department of Electrical Engineering)

Chester J. Wolejsza, Jr. (Department of Electrical Engineering)

Economics and Management Fred H. Baughman (Department of Aeronautics and Astronautics)

Since a decision by one group often became a ground rule

for another, we soon realized the complexity of a systems

project and the absolute necessity for close coordination and

cooperation with our fellow students. This iterative interplay

between intragroup and intergroup problem solving was the

most significant contribution of the course to our education

and experience.

But in addition to this unique learning experience, we were

privileged to have an impressive line-up of experts lecture us

on most of the topics that we would cover during the term.

The dates, topics, and lecturers are listed below� 2/7/67 "Preliminary Considerations" Prof. Paul E. Sandorff

2/9/67 M.I.T. Dept of Aeronautics

and Astronau tics

2/ 14/67 "Asteroidal Bodies and Prop- Dr. Frederick Whipple

erties of Icarus" Director, Smithsonian

Astrophysical Observatory

2/16/67 "Guidance Problems" Prof. Yao T. Li

M.I.T. Dept. of Aeronautics

and Astronautics

2/23/67 "Hydrodynamic Shock Effects in Prof. Paul E. Sandorff

Solids"

2/28/67 "laser Sensor Acquisition and Prof. Louis D. Smullin

Ranging Possibilities" Head, M.I.T. Dept. of

Electrical Engineering

3/2/67 "Kalman Filtering" Prof. John J. Deyst, Jr.

M.!.T. Dept of Aeronautics

and Astronautics

3/9/67 "Ground-Based Radar Acquisition Dr. Herbert G. Weiss

and Ranging" M.I.T. Lincoln Laboratory

3/ 14/67 "Orbital Characteristics of Icarus Prof. Samuel Herrick

University of California at

Los Angeles

Project Icarus 160

3/21/67 "Saturn Hardware Utilization" Mr. Jack Funk

NASA Manned Spacecraft

Center

To each of these men, we are most grateful.

On March 23, 1967 we broke our academic schedule for a

2-day tour of the U. S. Air Force Eastern Test Range, Patrick

Air Force Base, and Cape Kennedy, Florida. Whatever doubts

we may have had about the application of the Saturn V

Launch Vehicle to Project Icarus were completely erased

by the awesome reality of the Vertical Assembly Building, the

Transporter, and the entire launch facility. For arranging this

inspiring opportunity we wish to thank Dr. C. Stark Draper,

Director of the Instrumentation Laboratory and Institute

Professor, Emeritus. We are grateful also to Major General

John W. O'Neill of the Air Force Electronic Systems Division

for arranging transportation, and to Major General Vincent G.

Huston, Commander of the Air Force Eastern Test Range, for

coordinating our visit. Near the end of the term on May 22,1967, we presented a

3-hour summary of Project Icarus to the M.I.T. community

in Kresge Auditorium's Little Theatre. The presence of rep­

resentatives from a few major national news services resulted

in the appearance of frightening articles about our project in

at least 30 newspapers from coast to coast, including front­

page coverage in The Boston Globe. Even TIME magazine

caught on and, after interviewing staff and students alike,

described Project Icarus in its "Science" section on June

16, 1967.

An invitation by session chairman Jon R. Sussman led to a

1-hour presentation in November by Lou Kleiman, Bill Lange,

and Rick Heldt at the 1967 IEEE Northeast Electronics Re­

search and Engineering Meeting (NEREM). Soon after the

meeting we were invited to discuss our plan to save the world

on WCAS radio in Cambridge. After 45 minutes and at least

Project History 161

10 interjections by host Phil Christie of "I carus is not really

going to hit the earth this June," our discussion was completed.

At the request of Mr. Alexander A. McKenzie, managing

editor of the IEEE Student Journal, the summary of Project

I carus as it appeared in the NEREM Record was reprinted in

the March 1968 issue of the journal. The following month a similar article appeared in M.I.T.'s own Tech Engineering

News. Needless to say, all this publicity generated tremendous enthusiasm among us, and a feeling that we had done some­

thing at least of interest, if not of genuine value, to the

country.

It is virtually impossible to thank everyone associated with

Project I carus who deserves our thanks. We should like to

acknowledge, however, the following staff members: Professor

Rene H. Miller, whose contributions extend far beyond Project I carus to all the systems engineering studies in the

Department of Aeronautics and Astronautics; Professor Yao

T. Li, who organized the critical guidance and control activities

of the course and "guided" us from technical skepticism to

what we feel is a feasible solution; Professor Henri Fenech,

who consulted with our nuclear group about the bomb, with­

out which we could not have begun to solve the problem;

Professor Louis D. Smullin, who not only spoke to the entire

project team but also advised the communications and radar

groups; Dr. Robert G. Stern of the M.I .T. Experimental

Astronomy Laboratory, and Professor John J. Deyst, Jr.,

who assisted us in a variety of problems concerned primarily

with celestial mechanics; Dr. Philip K. Chapman, formerly

with the Experimental Astronomy Laboratory and now a

scientist-astronaut at the NASA Manned Spacecraft Center,

who contributed much to the design of the I ntercept Monitor­

ing Satellite; and, in particular, Professor Paul E. Sandorff, who

conceived the project, organized it, lived intimately with its

Project Icarus 162

problems, and won the respect of every one of us for having

created so ingenious a way to make students learn.

The evolution of this report was still another learning

process and, like all other facets of 16.74, was a coordinated

effort. The individuals who contributed to each chapter of the report are cited in the credits. For their editorial as­

sistance, we wish to thank fellow students Rick Heldt, Bill

Lange, who also did the art work for the NE REM article,

and Ted Tenny. We are grateful also to Mrs. Barbara Marks

for typing most of the rough and final versions of the manu­

script, and to Mrs. Dorothy Ladd and her staff at the M.I.T.

Instrumentation Laboratory for producing the final art work.

We are especially grateful to Joseph Stein of the M.I.T.

Press, whose invaluable suggestions, cooperation, and personal interest in Project Icarus contributed immeasurably to this

report. And finally we thank the staff of the Guidance and Con­

trol Laboratory at the NASA Electronics Research Center for

advice, support, and encouragement of the editor throughout

the project.

L.A.K.