Documente Academic
Documente Profesional
Documente Cultură
Volume 11
PHASE I,FINAL REPORT
Phase I Systems Analyss and Trrdeoffs
D 380.25037-2
G . R. Woodcock
Study Manager
The SPS System Definition Study was initiated in June of 1978. Phase I of this e f f w t was com-
pleted in December of 1978 and is herewith reported. This study is a follow-on e f f w t to an
earlier study of the same title completed in March of 1978. These studies a r e a part of a n
overall SPS evaluation effort sponsored by the U. S. Department of Energy (DOE) and t h e
National Aeronautics and Space Administration.
This study is being managed by the Lyndon 8. johnson Space Center. The Contracting Officer
is Thomas Mancuso. The Contracting Officer's representative and Study Technical Manager
is Harold Benson. The study is being conducted by The Boeing Company with Arthur D. Little,
General Electric, Crumman, and TRW as subcontractors. The study manager for Boeing is
Gordon Woodcock. Subcontractor managers a r e Dr. Philip Chapman (ADL), Roman Andryczyk
(GE), Ronald McCaffrey (Grumman), and Ronal Crisman (TRW).
I - Executive Summary
11 - Phase I Systems Analyses and Tradeoffs
I11 - Reference System Description
IV - Silicon Solar Cell Annealing Test
V - Phase I Final Briefing Executive Summary
VI - Phase I-Final Briefing: SPS and Rectenna Systems Analyses
VII - Phase I Final Briefing: Space Construction and Transportation
vii
Guidelines .......................... 356
Analyses Models and illethodology ................ 359
Sillcon EOTV's . . . . . . . . . . . . . . . . . . . . . . . . 363
Gallium Arsenide EOTV . . . . . . . . . . . . . . . . . . . . 376
EOTV C~rnpariscnand Selection . . . . . . . . . . . . . . . . 380
EOTV Design Life ....... . . . . . . . . . . . . . . . . 382
EOTV Fleet Size . . . . . . . . . . . . . . . . . . . . . . . 385
EOTV L>esig~Characteristics . . . . . . . . . . . . . . . . . . 3215
Configuration . . . . . . . . . . . . . . . . . . . . . . . . 385
Power Generation Systern . . . . . . . . . . . . . . . . . . . 389
Power Collection and Distribution . . . . . . . . . . . . . . . . 393
Electric Propulsion Syslem . . . . . . . . . . . . . . . . . . . 393
Mass Suniniiiry . . . . . . . . . . . . . . . . . . . . . . . . 398
EOTVCost . . . . . . . . . . . . . . . . . . . . . . . . . . 398
blission Operat~ons . . . . . . . . . . . . . . . . . . . . . . 401
K e y IC1i3;, on Events . . . . . . . . . . . . . . . . . . . . . . 401
EOTV Annealing Operations . . . . . . . . . . . . . . . . . . 404
Thruster P efurb~shment . . . . . . . . . . . . . . . . . . . . 404
LEO Skpport Base Des~gnarid Operat~ons. . . . . . . . . . . . . 406
Conf lgurati*...r, . . . . . . . . . . . . . . . . . . . . . . . &06
EOTV C.ostruction . . . . . . . . . . . . . . . . . . . . . . . 4ua
Depot Operdtidns . . . . . . . . . . . . . . . . . . . . . . . 408
LEO Base Crew Size . . . . . . . . . . . . . . . . . . . . . . 411
LEO Bdse $ 4 3 arid~ ~ Cost . . . . . . . . . . . . . . . . . . . . 411
1.O. 1 HISTORY
Solar power has long been recognized as a n ideal source of energy f o r mankind.
It is naturally available and plentiful, does not disturb t h e environment, e.g.,
by creation of wastes, and is itself free.
About ten years ago, a way of utilizing solar energy t o generate electricity on
a 24-hour continuous basis was proposed by P e t e r Glaser of A. D. Little. His
proposal was t o place t h e solar collectors in space, where they c a n collect sunlight
continuously, can readily be aimed at t h e sun, and where very large collector
a r e a s c a n be o b t a ~ n e dwith relatively l i t t l e investment in material resources.
Energy collected by these solar power satellites (SPS's) would be transmitted
t o Earth by electromagnetic means. The original Glaser proposal, and most of
t h e subsequent studies, have assumed t h e use of radio frequency systems in t h e
nlicrowave frequency range. Recently, t h e possibility of lase, beaming has also
been recognized.
Dr. Glaserfs original proposal was publ~shedin 1968 in Science magazir~e. In 1971
and 1972 a small contractor study t e a m was formed including Arthur D. Little,
Grumman, Raytheon and Spectrolab. This t e a m was awarded a study contract
through t h e NASA Lewis Research C e n t e r t o investigate basic technical feasbiiity
of t h e SPS concept. The conclusions of that study were t h a t t h e system is
technically feasible and could provide baseload electricity from m l a r power for use
on Earth. Additional studies and experiments, partly funded by NASA over t h e
period 1973 t o 1975, established t h e feasibility of efficient energy transmission
at microwave frequencies. In 1975 a demonstration conducted at J P L transmitted
more than 30 kilowatts over a distance g r e a t e r than a mile with a reception and
conversion efficiency of 8 2 percent.
(Also durillg this period, NASA-funded space transportation system studies indi-
c a t e d t h a t the high t r a f f i c volumes required t o support a n SPS program could
lead t o cost reductions far below those projected for t h e space shuttle. The potential
for such cost reductions was seen as significant t o t h e economic practicality
of SPS.)
4. B
Spac construction approaches and construction base designs were deveioped
for construction of 10,000 megawatt SPS's in geosynchronous orbit at a
r a t e of approxinlately 1 per year.
The principal system elements from t h a t study were t h e point c f departure for
t h e c u r r e n t study. The preferred SPS defined by Boeing is illustrated in Figures 1.0-2
through 1.0-6.
1.0.2 OBJECTIVES
The overa!l intent of t h e systems definition studies, past and present, may be
summarized as iullows:
-INlERMEOlATE SEGUENT
- U-15mSEGMENlSt8AY
!5S STRINGSr'GAY
611 PATIELSlCAY STRING LENGTH
=Ill*
GENERAL M A C E I S T I C S
m u TIM l
L9-mmQC
-
6% OVERSIZING 1RAOUIK)W L O S )
a DAYS
MooULE no
CCUIUCTERISTCS ANEW
WO.UOWLES 6
YOOOLELUESI~&G) L7
POwERREOP Q I KW) 43
ARRAY% 13
OTSORY r ~ & o 1.1
MGOLIIIOLIGI 20
LO~Z))~II@KG) 1.o
TANKS ELEC THRUST 110%; 4.5
CHEY T H R U S T I I ~ ~ ~ 12.0
NO WITH
ANTENNA ANTENNA
PANEL W E : 2&Sm &5lm
NO. THRUSTERS: 560 1680
-1-
FLIWtT GEO PKSENGER Y O W L
CONTROL
MOWLE
- 4 m b
+ 11Jm --A
d PRESS I W -
URIZED PRESS
-
kg?' STAGE 2
WTERFME
-
CREW 2
MASS-4.Qo04 --
CREW 76
W A S 10.#)0kn
CARGO
MOWLE
-
-
TY) MAN MO.
30.WO 4
l o r n4
MAIN ENGINE (21
U S THRUSTERS 200 I N & U UFI
Chemical If 02/L H2J Orbit Transfer Vehicle ftir GEO CrewBuppIy Delivery
Figure I.
0-6 Orbit-to-Orbit Transportation
reliability, and t h e suitability of SPS's as power generators for typical com-
mercial electricity grids.
The present study is divided i n t o t w o phases. The first lasted 7 months and t h e
second will last 9 months. Phase I accomplished additional analyses of t h e options
and issues identified by hASA in developing t h e reference system report, beginning
with a thorough critique of ,he J X I B o e i n g reference system followed by analyses
of options and critique items. The reference design was updated a t t h e end of
Phase I. The overall schedule is shown in F i g w e 1.0-7.
3C1- JULY 78 FEE 79 NOV 78
f PHASE I I PHASE I t 1
CRITICUE-M001FY-MAINf*IN UPDATE EASED
BASELINE SVSTEMS ON OPERATIONS
I ON OPERATION
ANAL YSlS I
I
UEF ItiE
PRO9UCTION
CAPACITY ISSUES II DEFINE INDUSTRtAL COMPLEX AND EARTH
TRANSPOF.TATION SYSTEMS
I
4.4 I UUNCH SITE SELECT ICN
II LAUNCH SlTE OPEAATtONS A N 0
LAUNCH SlTE DEFIWITION I
AFID OPE , A iONS OVERALL OPERATIONS
2. An alunrninum structure option for the solar array support system was analyzed.
I?
was found to be a feasible design but would be about 25 percent heavier
than the co~npositcdesign. Analysis of alurninum structure for the transmit-
ter antenna indicated that excessive thermal deformation would occur.
Thus, aluminum transmitter structure would require active compensation
for thermal deformation affects. The use of aluminum waveguides was
also examined. Uetuning of the waveguides due to thermal expansion intro-
duces an adaitional 1 percent loss i n the microwave power transmission
system efficiency. It was therefore recommended that a suitable composites
material be developed for the waveguides.
BOElNG
O. R. WOODCOCK
*
L
I
GENERAL
I
GRUMMAN
I
A D. L I ~ E
1
TRW
i
ELECTRIC
-
R. AKDRYCLYK R. McCAFFREY ?. CHAPWAN R. CRISMAN
i i
IIECTEtJNA-GRID
POWER FROCESSING
10. A mission control concept for SPS mission operations was developed and
a n SPS avionics d a t a and communication system concept appropriate t o
this n~issioncontrol approact, was defined.
Four subjects a r e reported at this WBS level: t h e SPS critique; SPS size and configuration
effects; SPS internal EMI; and interactions of SPS power beams with o t h e r spacecraf r.
2. The second i t e m identified several concerns associated with t h e large unit size
of t h e reference design SPS. This issue is being addressed in t h e current study.
3. The third i t e m dealt with t h e space debris issue. Collisions between t h e SPSts and
other space junk have been addressed in earlier studies and means identified t o
minimize t h e risk of such ccllisions. This critique item, however, also identified
a potential issue with creation of debris by t h e construction process, and concern
with outgassing o r emission of particulates (e.g. graphite fibers) as a result of extended
exposure of t h e SPS1s t o t h e combined space environment.
7. Plasnia interactions may occur with t h e high voltage solar array. In particular,
this i t e m singled out the problem of plasmas produced by the e l e c t r i c thrusters.
A c a r e f u l analysis could shed some light o n this issue but an adequate resolution
will require experinlents in space.
9. This i t e m addressed solar blanket installation details. One concern was t h e lack
of definition c f a repair and replace concept. A second concern regarded details
of t h e jumper installations between blankets across t h e str-lctural members, in
particular t h e relatively high voltage potential t h a t will exist between the
s t r u c t u r e and t h e jumpers. Also, t h e change from bi-axial t o uni-axial tensioning
of t h e solar blanket l e f t some inconsistencies in s t r u c t u r a l arrangement and
installation details t h a t need t o b e corrected. This l a t t e r i t e m is discussed below.
17. This i t e m expressed concern t h a t high voltage, high power transistors (used
in power processors) a r e not radiation resistant and t h a t t h e space radiation
environment would r e s u l t in short l i f e f o r t h e s e devices. Shielding w a s suggested
as a possible fix.
10. C o n c e r n has expressed regaraing design of, and m a t e r i a l s for, nigh voltage insulators
and c a b l e insulation f o r use in t h e s p a c e environment.
e
19. This aadressed t a c t u a l performance of t h e m i c r o h a v e power transmission
&
phase control, phase distribution, a n d KF power distribution systems. The g e n e r a l
n a t u r e of t h e c o n c e a s was such t h a t a t e s t program would be required t o accomplish
resolution. a
20. This i t e m expressed c o n c e r n about t h e K F power amplifier design. First, t h a t
o t h e r kinds of d-c/KF conbersion may b e superior t o klystrons; secondly, c o n c e r n s
o v e r klystron e f f i c ~ e n c y ,klystron tuning over t h e t h e r m a l range t h a t is expected,
a n d klystron o r t h e r m a l control faillires and life.
9
21. I t was noted t h a t in s p a c e checkout of t h SPS modules, t h e i n t e g r a t e d SPS,
a n d i t s operation with t h e grbund system has n o t been a d e w t e l y defined.
The task ~ssignedto TRW was to provide a critique of the SPS baseline Microwave
Power Transmission System and Phase Control System. The purpose of the critique
1s to identify:
a. nes~gn
concept concerns
b. .\reas roqulring additional analysis
s. Inconsistencies dnd/or integration problems
d. Potentiall) more attractive options from a cost/n,ass and a technology/schedule
risk standpoint.
For purposes of the critique the Microwave Power Transmission System was reviewed
in two sections, the power distributior? section from the siip-rings to the klystrons
and the r f section f r ~ r nthe klystrons through the antenna. For each section the
critique results are presented in the categories requested, however, categories (a)
and tc) have been combined. As will be seen, a single item may f i t rnore than one
category. however, i t is only included in one.
POWER DISTFUBUTION
This section is principaliy concerned with power distrib~tionon the antenna side
of the slip rings, however, the power distribution aspects of the solar array as i t
relates to power quality and distribution were also included where necessary.
The ~nsulationprovisions on the solar hrray busses may 1:-t be adequate for the
reallired level 01 voltage (nominally up to 44 K! ). Con: .derably more spacing may be
1 ';
018025037-2
required for successful isolation. Plasrna effects, those naturally occurring a t
synchronous altitude plus t h e effects of ion engines, and other propulsion engines as
well as t h e potential o r discharge of e n v ~ r o n m e n t a oxygen
l o r o t h e r ionizable gasses
introduce t h e need for corona suppression concerns. Argon gas is o n e of t h e most
easily ionized gasses and is presently t h e baseline stationkeeping i o r engine gas
select ion.
F =- '' lo-' r l xl
1 2 = f o r c e in kilograms per m e t e r of length
9.81 x d
The current density in the sheet aluminun~conductors should be checked for the
criteria of 1 0 0 as ~ density appears to be hign even when radiating :o a black
~ this
body. The method of conductor support from and to the slip rings was not specified
and could add substantially to the weight due to insulators ar,d collector terminations.
2
In reviewing the I K loss as tabulated, the return conduct~rlosses were apparently
not added to the loss for dc to dc converter input or kiystron collector 4 and 5 connections
i n computirig the heat loads. The negative conduc:or would generate significant
voltages in the shared conductor and would, i f conrtected as diagrammed, present
~nterractionsi n unrelated circuits that would make i t impossible to deliver the voltage
of the quality specified. The method o i busing all of ttiise interconnected ires
requires definition. In fact high voltage splices, connectors or attachment provisions
for conductors would be expected to have a significant weight. In past utility practice
using dc grid systems, the ground connection was always made at the source and
at the source only. The report does not address the point as to the planned usage,
grounded or ungrounded. The absence of a sij gle ground complicates fault detection
relaying for system protection. Further analysis on grounding and i t s relation to
plasma neutralization and static buildup on insulated system components i s recomlnended.
The Hughes d c switch as presented in the literature is a krypton gas cathode anode
that acts as a gas tube when i t conducts. it was sized for 1,000 amperes at 100 kV.
It operated in 50 t o 60 milliseconds with t h e aid of two mechanical contactors and
large energy absorbing resistor grids. One of its insulated bushings may weigh as
much as the stated budget cited in the report for main bus circuit breakers. The
report identifies "A" bids units as 620 amperes and 6200 ampere momentary current.
Methods of caiculating t h e short circuit current capability of t h e system which would
become the rupture requirement of those circuit breakers, also require development.
Potential Options
In our review, diodes were assumed for each parallel string of solar cells. Use of
a large power diode on the main solar array circuit breaker circuit is also recommended
as the short circuit requirement of this circuit brcaker is thus reduced t o that of its
own group of cells. This makes t h e selection of the commutating capacitors and other
devices more practical by reducing t h e stored energy for commutation requirement
determined by worst-case short circuit current and the speed selected for the circuit
breaker operation.
The report suggests the use of on-array switching of individual strings for voltage
regulation. Use of this capability t o clip t h e excess voltage that occurs due to B.0.L .
t o E0.L. decay and t h e initial overvoltage that is induced by occulation would reduce
the excess power that must be dissipated on t h e antenna. The power quality could be
thus held t o even closer tolerance from the array than those selected in t h e report. It
is recognized the temperature of t h e so!ar array would elevate whenever w y load is
turned off.
For a c or dc systems for use in space, all circuit breaker, relays, contactors, fuses
and fuse holders should be redesigned for cold-plate cooling t o permit reduced weight
and size and for installation in power junction boxes for mechanical and personnel
protection. Electrical and electronic equipment has always required a benign environment
for gooa reliability; and therefore, i t is anticipated that liquid or gas cooling loops
will often be required, especially for high power equipment.
The energy storage for this SPS electrical system should be redundant with the possible
exception of the klystron heaters. The energy storage subsystem would result in
a larger solar array penalty than stated in the report especially if nickel hydrogen
batteries a r e used. They a r e expected t o have roughly a 6% loss each day, and uocid
have to be recharged even on days that no occulation occured. Low rates of battery
charging a r e the least efficient, therefore additional penalty should be
D180-25Q37-2
estimated for this subsystem. Due to t h e large spatial relationship of t h e loads t o
each other, dispersed energy storage locations a r e indicated. it is possible that a n
energy storage electrical grid system could reduce t h e amount of batteries o r other
electrical energy storage components by avoiding some of t h e redundant capacity at
a l l critical ioad locations When large dc storage systems using batteries are
considered, t h e reconrmended distribution is use of inverters and ac so that commuta-
tion switches may be designed t o isolate t h e battery frorn the bus u ithout added fault
contribution from other things connected t o the output b u r
RF GENERATION TRANSMISSK)N
This section is primarily concerned with r f generation and transmission by t h e
Space tema.
Therrndi stress may produce large scale distortion errors in both the subarrays surface
and t h e total array surface. The retrodirective concept corrects for deviations from
the plane but does not correct for beam peak alignment shifts, i-e., if a subarray is
sufficiently distorted so t h a t i t s pattern is skewed, t h e conjugate phase technique will
correct the phase of its radiation in the d ~ r e c t i o nof the Rectenna, but will not realign
its bean1 peak in that direction. Extension of the thermal distortion analysis is needed
t o evaluate the magnitude of associated patterr. changes of the subarray and total
array.
Potential Optiaas
Reduction of diffraction perturbation of illumination by use of passive techniques.
PARAMETRIC INVESTIGATIONS
SPS baselines have generally used a 5,000 megawatt power transmission link. This
power rating may be uncomfortably large for some applications and it is natural
t o raise t h e question, "How much smaller could SPS's be made?" Shown in Figure
1.1- 1 is a joint optimization of transmitter diameter and power level holding the
rectenna size constant at the optirnum value. This result was developed on t h e earlier
contract, and did not nclude packaging density considerations. As the system power
level is reduced, i t is possible t o employ samewhat larger transmitting antennas
without violating the
25
3.004 -
2.m
23 m ~ l c m 2LIMIT
01 I
1 a J
2.000 4.m 6.W e,ooo
DC POWER ACROSS ROTARY JOINT. IEGAWATIS)
I I I I I
~.fmo 2.000 3,000 4.000 s,m
DELIVERED GRID POWER PER LfNU, (MEGAV4Am)
The present study expanded on these earlier results to consider packaging and specific
configuration effects arising from asymmetric configurations. Three smaller SPS
configurations a r e compared t o the original LO gigawatt baseline in Figure 1.1-2.
The first of the three shown is t h e present NASA 5 gigawatt baseline with one trans-
mitting antenna. Analysis of the control requirements for this asymmetric configura-
tion determined that because of t h e overriding importance of solar pressure compensa-
tion in the control thrust scheme, no propellant penalties were incurred by the lack
of symmetry. Also, no packaging differences have been identified that would arise
from dividing the original configuration into two equal halves. Therefore, the only
consequence of this alternative t o the original baseline is the requirements for more
positions in geosynchronous orbit t o effect a given total installed generating capacity.
The next alternative shown is also a 5 gigawatt system, but the power is transmitted
through two power transmission links each rated at 2K gigawats. In order t o minimize
land use and rectenna costs, i t is desirable when reducing the link power t o increase
the transmitter aperture, in turn reducing the receiving station area. This design
option, however, has approximately four times as many transmitter subarrays as
the single-transmitter 5 gigawatt satellite. As a result, i t incurs a significant payload
packaging problem because of the low packaging density of completely assembled
transmitter subarrays. The packaging density situation appears t o be much improved
through use of a solid state transmitter. In the solid s t a t e option, all of the active
functions a r e included in a planar sheet only about 2 centimeters thick (including
the resonant cavities). Thus, a much higher packaging density per unit of aperture
area can be achieved.
The final option shown, like the second option, results from effectively dividing a
symmetric configuration in half. As for the other case, no penalties were determined
for this design option excepting the use of more geosynchronous orbit space.
SAME AS
5 GWITh GW
TRAr4SMITTERS
EXCEPT USE OF
5 GW BASELINE 5GWiThGW SPACE AT GEO
TRANSMITTERS
NO IMPACT 82%VOLUME
0 EXCEPT USE OF
SPACE AT GEO
LlMlTED LAUPJCH
PENALTY UNLESS
TRAKS5lITTER IS
10 GW BASELINE SOLID STATE
ASYMMETRIC CONFIGURATIONS
The configuration considered is t h e Photovoltaic Reference Configuration, c u t in
half s o t h a t t h e north-south dimension of t h e array is reduced from 2!,280n1 t o 10,640m.
A sketch appears in Figure 1.1 -3.
......
. . . - .
. ..
... . *
. . . . .
I .
_. . . . . . WINtER
I
.. .-, . '
... VERNAL . . .
-
too , SOLSTICE : .; EQUINOX .
.
THRUST N I n
-. .
:!..
.
I - t !-. , .. 1..
* I
. -
.! -.. ,.' . .. . .
+ i1 . , . :,
-
-.- -
- i
, i ,
.
! , : ;,:
. I
-
. :
,
; . . . . . . .
. .
TORQUE
- ~ m x l O d ,,0
Norrnal t o t h e array
t
D: 75.55 cor2 + 65.70 sin -g
biaximurn and rninimum values of A and B a r e 143.95 and 0.1 1 N and of C and U,
141.25 and -2.16 N.
Thrust velues depend on the panel reflectivity. Coated cells have a generally accepted
range of 0 t o C.3 and value of 0.15 has been used above. Zero reflectivity would
(1
redlt-, the inlpulse t o 0.980 x 1010N sec (49,650 ;'.g of propellant) whi!e a value of
10
0.3 would require 1.151 x 10 N sec (58,700 Kg of propellant).
1.1.0.4 GENERAL SPS FUCHT CONTROL S T U m
There are two basic control modes: (a) powered flight when thrust is being applied
to produce a velocity increment ( v) either for the purpose of maintaining orbit
dltitude, or changing from one orbit to another or for achieving and maintaining
a specific location in a given orbit; (b) coasting flight which inciudes all other flight
regimes.
The control requirements and criteria for each mission phase are discussed below.
Included are a description of the currently preferred control concept and a comparison
with other candidate approaches which were considered
d. The orbit mechanics forces between the construction base and the rnodule should
be minimized to simplify the indexing mechanism design (minimize sire, weight,
and cost).
Edge*ise orientation ( t h a t 1s u ith t h e plane of t h e collector r~toduleparallel t o t h e
orbit plane) is t h e obvious choice for one of t h e two alignment axes, because it inini-
rnizes drag, and it is a gravity-gi-adient stable orientation. The o t h e r alignment
(module d o ~ onr module forward) is not as easily determined. Items a) and b) d o
not prcvide a conclusive choice b e c a u . ~t h e resulting preferred orientation changes
during the construction process and also because t h e rniniinurri drag orientation is
not a l u a y s t h e o n e which results in ~iiinimumgrav~ty-gradien; disturbance torque.
h1utri;rl conipatibility with the orbit keeping concept is t h e ~iiajorfactor In selecting
t h e preferred orientation and is discussed in the next paragraph.
-4 .
g. An acceleration of 10 g IS assumed. This value is based on a reasonabie burn
time.
Figure 1.1-6 shows t h e vehicle coniiguration for seven stages of construction. It
is recognized that the ccnf iguration changes continuously during construction. The
o r e s shown have been selected because they represent some of t h e extremes of config-
uration parameters such as mass, moment-of -inertia unbalance, c.g., location, prin-
cipal axis orientation and aerodynamic drag. The selected concept must be c q a b l e
of controlling all of these configurations.
A matrix of possible control approaches using fixed and gimbaled thrusters with
fixed (module down and module forward) and variable attitudes is shown in Figure
1.1-7. Typic21 thruster locations and orientations are shown for each concept. Also
shown a r e the c.g. locat ions taken from t h e seven representative configurations.
Concept A shows t h e use of separate attitude control t h ~ s t e r to
s offset t h e AV
thruster unbalanced ntoment. This approach is not acceptable because the required
attitude control thrust and propellant, for some c.g. locations, can exceed those
for the Q V thrust. L7 the module-forward attitude shown in Concept 8, the AV
thrusters can be located so that both p s i t i v e and negative control moments can
always be generated regardless of c.g. location. Furthermore, all of the propellant
expended contributes t o a positive velocity increment. This is t h e only fixed-attitude
concept which does not have a large wasted side component of thrust. Concept C
shows the approximate thruster location and orientation required t o accommodate
all preaicted e.g., locations for t h e module-down orientation. Concepts E and F
illustrate the large gimbal angles (and consequent inefficiency) for thrust vector
control with a fixed vehicle attitude. Concepts D, G, and H show how efficiency
is improved by rotating the vehicle so that t h e thrust vector, when pointing through
the vehicle c.g. is also digned with t h e orbital velocity vector. However, t h e time
and propellant required t o rotate t h e vehicle t o t h e required attitude make this approach
unacceptable.
Figure 1.1-8 shows t h e selected approach t o attitude and velocity control of the
construction base and spacecraft components during the LEO construction phase
of the mission. Included a r e t h e design features and advantages of the concept.
The average orbit-keeping propellact required is 1200 kg per day for this co~figuration
(assumed I = 400 sec).
SP
? t b- -- --4-
'. .- 2 \ -. .- .--
a--
LA mLE-FORWARD
0 0 MODULE-FORWARD
@
@GRAVITYGRADIENTSTABILITY:
Ill 24-6 x lo6 14.3 0.44 345 1516 24x lo4 17.7 1 24 133 42
V 4 1 ~1
x8 58 0157 .I43 1512 -14%lo4 t 7.3 40 712 2.43
VII 17.2~
108 469 0.345 -192 841 3 3 7 x lo4 9.7 16 873 3.3
iw -
ar , . mk&m Ih -478 km
-+
ASSUMES THRUSTING ALONG V AND ISp
= 400 SEC.
(Rosynct~onousOrbit Operatiorrs
Velacity control is required to maintain orbit station by offsetting solar pressure
anu orbital drift coward t h e neutral point. Attitude control is required t o provide
Sun orientation of t h e collector and high accuracy Earth rectenna pointing of t h e
tu 0 a t e n n a s . In addition t o countering the gravity-gradient disturbance torques,
t h e control concept must avoid unstable interaction of t h e antenna and collector
ctxitrol loops and t h e structural flexibility effects. f i e current baseline approach
for coilector control makes use of multiple thrusters providing a t o t a l force equal
t o t h e solar pressure. Individual thrusters a r e modulated above o r beiow their bias
level t o provide t h e control torques needed to o f f s e t gravity-gradient disturbance
torques. In essence, t h e r e is no additional propellant penalty for collector a t t i t u d e
control. The basic features of this concept a r e illustrated in Figure 1.I- 10. Single
axis (pitcn) rigid body control only is shown. Thruster location will b e at t h e nodal
points of one of the lower rnodes t o minimize excitation of t h a t particular flexible
rnode. Active darnping of other modes will be achieved by superimposing additional
thrust rnodulation signals on t h e a t t i t u d e control thrust level commands. These signals
a r e deri-ded from t h e outputs of multiple r a t e sensors which a r e processed t o isclate
t h e rigid body and lower bending mode components of motion.
4
SUN
4
LIGHT
+
TIME - HOURS
SPS A n t e m a Control
Antenna gimbal torqulng results in a reaction morrient o n t h e c o l l e c t o r e v e n for a
perfectly balanced antenna. If t h e a n t e n n a c.g. is o f f s e t from t h e gimbal axis. t h e r e
1s a n a d d ~ t i o n a reaction
l f o r c e a t t h e gimbal polnt. These f o r c e s and niornents e x c i t e
t h e collector s t r u c t u r a l rnodes a n d a p p e a r a s disturbances t o t h e c o l l e c t o r a t t i t u d e
control. Thl?: coupling c a n b e reduced significantly by using "free floating" gimbals
a n d generating t h e c o n t r o l torques within t h e a n t e n n a itself. {However t h e a n t e n n a
lliass unbalance r e a c t i o n f o r c e s will n o t b e eliminated.) Trades of c o n t r o l - r n o n ~ e n t
gyros (crng's), inertia wheels, r e a c t i o n c o n t r o l t h r u s t e r s and geo-magnet ic torquing
have resulted in t h e t e n t a t i v e selection o f multiple, antenna-mounted crng's t o g e n e r a t e
c o n t r o l torques. Because o f t h e unidirectional gravity-gradient d ~ s t u r b a n c etorque,
?
..9
#
* FLEXIBLE
WIODEtS)
OVNAMICS
r---------
t DYNAMICS
-------- - - - -
I RECTENNA COOPERATIVE
&
WINTING ERROR !XNS!W {-I~ ATTITUDE
The baseline hybrid control concept shown in Figure 1.1-12 uses antetina-mounted
control-moment gyros (cnlg's) for fine resolution, high frequency control and uses
antenna gimbal torquing against t h e collector t o desaturate t h e cmg's. The gimbal
torquing mechanism, in e f f e c t , counteracts t h e steady unidirectional gravity-gradient
torque u ~ h i l et h e cmg's compensate for base motion e f f e c t s caused by structural
flexibility, and collector a t t i t u d e stabilization. The block diagram shows how these
two torque-producing elernents c a n b e combined t o provide t h e desired corltrol characteristics.
The discussion thus far has considered t h e antenaa as a rigid body. The baseline
design, consisting of a deep s t r u c t u r e of graphite epoxy composite material provides
GAIN
LOWPASS
FILTER
These values were used t o c a l c u l a t e t h e E- and H-fields near t h e power bus conductors.
Since t h e c u r r e n t distribution a c r o s s t h e bus a t high frequency is only known very
generaily, a very simplified approach was taken, whereby t h e 5-meter a n d 17-rneter
s:Ieets w e r e e l a c e d by imaginary conductors s e p a r a t e d by 18.9 m e t e r s ( a circular
a r c with a radius equal t o half t h e distance between t h e c e n t e r s of t h e conducting
sheets). T h e voltage d i f f e r e n c e was t a k e n a s 1% of t h e bus voltage, i.e., 380 volts.
Tne m e a n e l e c t r i c field at a distance of 6 m e t e r s ( t h e approximate cadius of t h e
a r c ) would t h e n b e approximately 20 volts/meter at 2 0 KHz, declining a t higher f r e -
quencies. C u r r e n t e q u ~ p r n e n tmust b e designed f o r 10 volts/nleter from 14 KHz
t o 3 5 hiHz.
These EM1 e f f e c t s wili result in modulation of t h e power beam baseband signal unless
appropriate measures a r e taken in :he design of t h e transmitter and power supply
systems.
An analysis was performed t o I: define SPS hardware faiiure modes, 2 ) predict equipment
filaure rates, and 3) determine t h e e f f e c t s on SPS system operation and ground power
output. Table 1.1-2 presents t h e results of t h e analysis.
FAILURE RATES
NF
= nur~~!-er
of equipment i t e m failures per year
1.1.2.2.2
MODULE
CISTRIBUTION
OPERATE 221x10~ ~03.104
u&m6
7.934 3.424
p
I 1 1.2 2.3
WAL8EGUIDE
RAOIATING WAVEGVIDE 1.7328oo
-
i\
1 1 2.2.4 THERMAL CONTROL 14.440
1 ;1;2;26 WIRING HARNESS OPEN I N A
CONOUCTOR 2 n d 14,-
*.lor
I
1.1.2-26 CONTROL CIRCUITS
1.1.2-2.7 STR'JC- JRE
1.133 POWER DJSfRIBUTIC)(Y
AND CONDITIONtNG
1.1.29.1 POWER CONDUCTORS OPEN 2.0xlOg 54
1.1.2-3.2 SWITCHGEAR
1.I -2.3.3 OC@C CONVERTERS
F A I L TO OPERATE
F N L TO R E G U
LATE OR NO
ah105 912 9 1.730
p
OUTPUT 16a;05 456 24 4.614
1.1.2.3.4 DISCONNECTSWITCHES INTERMITTENT OR
OPEN CONTACT 2.3~106 912 3
1.1.2.3.5 ENERGY STORAGE NECJ 2-1x10~
1.1.2.4 THERMAL CONfROL
:.1.2.4.1 THERMAL CONTROL
SURFACES
1.1.2.4.2 POWER PROCESSING
THERMAL CONTROL
1.1.2.4.3 INSULATION
1.1.24.4 CONTROL CtRCUlTS
F A I L TO WERATE 1.1~1@~ 4 s 4
D
THERMAL CONTROL
1.1.2.4.4 CONTROL CiRCUlTS
THERMAL CONTROL
1.1.2.4.5 MECHANICAL POINTINS
THERMAL CONTROL 24
1.1.2.5 CONTROL
1.1.2.5.1
1.1.2.5.2
RECEIVERS F A I L TO OPERATE 5x10~ 203568 4
b
!I,
DIPLEXEHS F A I L TO OPERATE 1x10' 203.568 2 1
1.1.2.5.3 P H A S TRANSMllTERS tJ0 OUTPUT 220.- 29
1.1.2.5..
1.1.2.5.5
PHASE RECEIVERS
CONJUGATORS
FAILS TO RECEIL c
NO OUTPUT
23f 220.408
5 . 4 ~ 1 0 ~ 203,566
4
33 14
1.1.2.5.5 CABLING OPEN6HORTED 7.6Xg7 218688 25 11
1.1.25.7 POWER DIVIDER F A I L TO OPERATE 1x10 7.643 - 143 J
1.1.25.6 SWITCHES F A I L TO OPERATE 5x10~ 1.180 -
1.1.2.6 MAINTENANCE
1.1.2.6-1 MAINTENANCE GANTRIES 44
1.1.2.6.2 GANTRY TRACKS
1.1.2.6.2 DOCKING POETS 4
1.1.2.6.4 CARGO HANDLERS 4
1.1.2.6 5 CREW BUSES
1 .I .2.6.6
CREW WORK STATIONS
1.1.2.7 ANTENNA MECHANICAL
POINTING
1.1.2.7.1 cAK;'s
1.1.2.7.2 CMG DRIVES NO DRIVE TORQUE
NEG
NEG 4&105
24
24 - -
1.1.2.7.3 CMG CONTROL CIRCUITS NO DRIVE
CONTROL NEG 24
1.1.3 INFORMATION MAN-
AGEMENT AND
CONTROL
1 1.3.1 TRANSDUCERS
1.1.3.2 SIGNAL CONDITIONERS
1.1.3 3 DATA ACQUISITION1
MANAGEMENT
b
1
ANNUAL
c"z
WBS EOUIPMcNt FAILURE m R LOSS UteF oUAWTITy
FAILURES ~ E A w
ELEMENT NOMENCLATURE MODE PER FAILURE
I*W
LHRS~ PER YEAR LOSS (KW) 3
8
1.1.6.2.3 POWER DISTRIBUTION
1.1 6.2.3.1 SLIP RINGS 6
1.1.6 2.3.2 BRUSH ASSEMBLIES NO CURRENT NEG 5&1@ 8D 1 NEG
FLOW
1.1.6.23.3 FEEDERS OPEN NEG 2.*1@@ 96 -
!
1.1.6.2.3.4 iNSULATORS SHORTED 272 -
1 . I .6 2.3.5 POWER BUSES OPEN 18
1.1.6.2.4 THERMAL CONTROL
1.1.6.2.4.1 MECHANICAL ROTAR l
JOINT THERMAL
CONTROL 24
1.1.6.2.4.2 ELEVATION JOINT
THERMAL CONTROL 8
CONTROL
SURFACES
COMMENTS
Maintenance Requirements
On t h e b a s i s of t h e p r e v i o u s l y assumed f a i l u r e r a t e s , a p p l i c a b l e component
numbers and h a l f y e a r l y maintenance per!.ods t h e expected number of f a i l u r e s and
t h e i r e s t i m a t e d r e p a i r time requirements can be c a l c u l a t e d . The r e s u l t s a r e
e x h i b i t e d i n Table 1.1-4
Table 1.1-4
-
11 ,
II
I I
i I I
I Subarray Line
swi Output
..000315
I
101784
f
32.06
j 1,11
-4
I
128.'1 I
II
i
.01859 f 60 3 i 8.9
i
i i
I
1
2nd B Ottput .0060iB I80 t?.29 6 ! 13.8
13 1 I
'
I
Bm O u t p u t .005246 I 7220 37.87 7 7j.7
IiI
I
Total I 3506.3
I
i
j 3 4 2-
9 * 6 8 ~ . -- -i
i
1
I
Man hours f o r two man team 7012.7 i
Table 1.1-5
Number of F a i l u r e s and Associated Maintenance
Tine Requirements i n t h e Rectanna DC Power
C o l l e c t i o n System
t
1 - f I 1 -
Hrs - ' Tot. Repair. Man Hours I
.1 year Repalr H r s . Per Main
- -* --,---
m -
Panel S t r i n g 87240 i 5.49.6 5 ( 2 man) . 2748
.
Unit to Group
Center Lines
.00315 1 784 2.47 18 (6 man) 44.4 266.4
I L !
Total [ 5762.5 1
I
4 I
I n t r o d u c t i o n an.' r)ef i n i t i o n s
System maintenance.
r e c t e i ~ n asystem is t y p i c a l l y on a y e a r l y o r multi-annual b a s i s a s w e l l a s
transformers.
58
Table 1.1-6
DC Switchgear 6 1.600
Step-up Transformer 72 .9 36
Table 1.1-7
Ncmber of Components
Tabit 1.1-8
Expected Values o f F z i l u r e s / Y e a r
AC Switchgear .002
activities.
o f the v a r i o u s d e v i c e s f o r a given r e c t e m a AC l a y o u t , i n c r e a s e d e f f e c t i v i t y
Manhours/year/device
DC Switchgear 2
DC/AC Converters 16
Converter Transformer 16
AC Cable 2
Synchronomts Condenser 40
AC &?itchgear 20
!
i S t e p u p Transformer 16
i i
Table 1.1-10
-
Hrs.
Unscheduled 495.7
Scheduled 6450
6945.7
space Antcosu DC P m r Dlatrubutioa S y a t m
o
P(x) - (l-plF
N-X+1 -P
for x =
P (x-1) for x # 0
X 1-P
Sector Lines
This system c o n s i s t s of N -
228 separate main DC buses, each carrying a nominal
of 35 & DC power a t a nominal 40 kV and 8758 using approximately 35 mn diameter
62
aluminum conductors. A t the end of a typical line there is a atit-ear unit. P u t
of the paver (5.44 HI) goes through a DC t o DC converter wfth 218 In LOSS md
there is a n m 446 vay pouer divider at the output of t h e converter s t a t i o n .
r - .5 pears
Table 1.1-11
One line c a r r i e s a = 1
228
- .4385% part of the t o t a l power. The available
power i f a f a i l u r e occured i s e d is
p u t of the f u l l power. A is the a v a i l a b i l i t y o f power. T.ble1.1-12ahwa t h e
cilcul.tioa of A as r function of
n
P = C PCL) f o r
0
# = 228 and p - .00594.
Part 1
P = -TX r = .0185925
The f i r s t layer is t r i p l e redundant (N = 3 p a r a l l e l branch). For t h i s case
the p r o b a b i l i t y of failure f o r t h e system of t h r e e redundant f i r s t l a y e r lines
is shovn i n Table 1.1-13.
Table 1.1-13
Calculation of t h e P r o b a b i l i t y of Failure f o r
Three Redundant F i r s t Layer C i r c u i t
- -
Table 1.1-14
The following f a i l u r e r a t e s a r e s e l e c t e d :
Resultant f / l o 6 hours
p = &
T'
.00604849 as f o r t h e f i r s t portion of P a r t 2.
Table 1.1-16
Calculation of the Probability of Faflure f o r
t h e Phase Control Network Group Layer (Part 2, Input Portion)
branches. There are N -
Thus t h e r e s u i t a n t p r o b a b i l i t y o f F a i l u r e i s p = .0@01303 f o r t h e redundant
380 l i n e s at t h i s (group) l a y e r . The p r o b a b i l i t y of
a v a i l a b i l i t y i s d i s p l a y e d i n Table 1.1-1 6.
Part 3
The following f a i l u r e r a t e s a r e s e l e c t e d :
Resul t a ~ t
-
Part 4
Table 1.1-18
C a l c u l a t i o n of the P r o b a b i l i t y of ' f a i l u r e f o r
t h e Phase C o n t r o l Network Klystron Layer ( P a r t 4)
For t h i s case
-XT = 1 failure
20 years
= .05 failure
year
,r = .5 y e a r
.'
I f t h e value of a kw Hr is .03$, then t h e value of the reco-. -red energy is
For t h i s iecovery t h e c o s t o f t h e a d d i t i o n a l maintenance time
32.8M$ yearly.
a s s o c i a t e d t o the more frequency replacement of f a i l e d k l y s t r o n s has t o be traded
off.
Table 1.1-18
1. S t r i n g of 4 3 diodes
2. Panel covering 432 = 1849 diodes
3. S t r i n g of pacels
4. Unit t o group c e n t e r l i n e s
PERCENT OF PROBABILITYIF)
Table 1-1-19
C a l c u l a t i o n of t h e P r ~ b s b f l i t yof F a i l u r e f o r
-
a S t r i n g of Par . ~ e Diodes
(I 43, p = .CiCi52596)
i
3. Panel S t r i n g s
.
These conditions y i e l d p = 6.3 x 10'~. However, Figure 1 i-21 a l s o shows t h e
a v a i l a b i l i t y f o r r = 1752 hours r e p a i r t i m e . The r e l a t i v e l y l a r g e mean time t o
r e i ? ~ i rtimes a r e assumed t o allow t h e use oL a cornparati-rely small r e p a i r c r e v .
On t h e b a s i s of t h e previous c a l c u l a t i o n s t h e r e s u l t a n t a v a i l a b i l i t y
of t h e DC power c o l l e c t i o n system is shown i n F i g u r e l . l - 2 3 f o r t h e d i f f e r e n t
combinations of f a i l u r e c h a r a c t e r i s t i c s . The f o u r cases a r e defined ill t h e
t a b l e as Figure1.1-23. It can be seen t h a t f o r t h e b e s t combinations of
f a i l u r e c h a r a c t e r i s t i c s (Case 1 ) t h e DC system causes a n average power l o s s
of 1.55% a f t e r 30 years of operation. The most e f f e c t i v e way t o improve t h i s
f i g u r e is by decreasing t h e mean time t o r e p a i r , which can be achieved by
i n c r e a s i n g t h e size of t h e ground maintenance crew.
HOURSPER YEAR
61s 290~4 876.8 4 3 ~ 3 87.7 . an
. .
t
,, ,,,.,,,
I I I I I1.
30 wm S war M +nl In
: :* ? :a
F i g u r e l . l - 2 4 shows t h e c i r c u i t l a y o u t s e l e c t e d f o r a v a i l a b i l i t y c a l c u l a t i o n s .
This l a y o u t does n o t match e x a c t l y t h e DC power output format of t h e r e c t e n n a
b u t t h e necessary p r a c t i c a l d e v i a t i o n s do n o t a f f e c t t h e a v a i l a b i l i t y numbers
significantly .
The assumed f a i l u r e c h a r a c t e r i s t i c s f o r t h e v a r i o u s components a r e shown
i n Table 1.1-23.
Table 1.1-23
.-Assumed ---
Component F a i l-----
u r e - C-*-.h a-
Power C o l l e c t i o n Sys tem
r a c t e _ r j g t & s 7 i n . t h e .AC
i
Component ( -fail-,re ! '1 f a i l u r e
-T rH r s .!
i
: T year Hr . P
p~ .- ~ -- -
1
I
I
, . i
(
!
i
1
lMW breaker
DS bus
I
.2'
.00063 1 .00002281 .
7.1068~10-~
8
13
,
I
1
I
- .OGGl824
.
.0000009
.
1
I
I
1
i
!
20 MW b r e a k e r
Converter & t a t i o n
20 EM transformer
200 ME switchyard
In
1
L
.33
.Cis41
0
.00002281
.00003764
4.6772 x 10'~
0
I
i
I
8
IQ
219
-
i
!
!
1
.OOOICi4
.G00376L
.00010242
0
I/ I
!
!
200 HW transformer .0041 4.6772 x l(i ? I
259 .00010242
I I
Synchronous condenser .5 5.7038 x ! 112 ; .006388
I
1000 MW switchyard 0 0 - i\ 0 1
1000 ?IW transformer ,0041 4.6772 x 10" 219 .OGO10242 i
I i
I
Table 1.1-24
P and N Values Applicable i n the DC Power Collection
Unit (MW) N P
1 5000 .OOG3652
20 250 .004030
40 125 .0004 79
200 25 0001025
1000 5 .0001025
h ---
Tables 1.1-25, 1.1-26, 1.1-27, 1.1-28 and 1.1-29 show the availability
ca3,culations for the various power l e v e l s of the network.
T a b l e 1.1-a
Calculations of the Probability of Failure for the lMW
- Level l i e t ~ o r k
Table 1.1-26
-
Caiculations of the Prqbability of Failure f o r the 20 MW
L e v 4 Netwqrk
Calculations of the Probability of Failure for the
40 IS Level Network
DC Breakers
Bus Connections (Approx. )
DC Converter Breakers
DC/AC Converter
Converter Transformer
AC Cable
69/230 Transformer
69/500 Transformer
230/500 Transformer
- *
D C 1 DC DC
WIT W l T o c - ms
GEAR CONNECT. MAKER
99 -
91 -
97 -
96 -
$ 10 66 80 90 99 99.9 99.99
4 Random Phase
i
99.40
97.50
98.50
97.02
98.20
96.55
97.82
95.48
96.80
94.58
95.75
93.60
94.80
!
!
1
1
1
9 AC Power Collec-
tion
T o t a l Power Trans-
mission System
~q uipment- Availabi-
m .
Power A v a i l a b i l i t y
at Power G r i d
I n t e r f a c e Relative
t o Equipment Without
Failure
Figure 1.1-30. Equipment and Power to Utility Grid Availability of Overall
SPS System Between Elevation Flexibile Joint of Space Antenna
and Power Grid Interface.
Table 1.1-32
Availability of Efficiency and Power Into
Power Grid
r
Hrs /Year 7889 2980 1753.2 876.6 87.66 b.766 .8766
-,- --.-. . - . .. - -- ---I
, Availabld Efficiency 63.33 57.23 55.46 52.29 46.13 39.76 31.34 1
%
Available Power
if 4513.3 4078.6 3952.3' 3773.6 3287.6 2833.9 2233.4!
I
I
An i n t e r e s t i n g c h a r a c t e r i s t i c is t h e rms power a v a i l a b i l i t y a t t h e
i n p u t of t h e rectenna. This can be obtained from Table 1.1-31by t a k i n g t h e
product of A of t h e space antenna, A r e l a t e d t o t h e propagcrtion media and sub-
i t e m s 1, 2 and 3 a t = 66%. This y i e l d s
@ Rectenna 14.3
(Panel assembly: 7.15%, balance of system: 7.15%)
@ Klystrons
@ S t r u c t u r e , s o l a r c e l l s , DC d i s t r i b u t i o n , 50.5
attitude control
Q Phase c o n t r o l 3.6
A B B/A
Opt ion X Increase of X I n c r e a s e of
Output Paver System Cost
Increase 4 and
A,
by
1.068 f o r 23 mwlcm mut. 1 6.8 3.88 1.75
Received Power Desnity
II
I
I
I
I.
System F u l l y I
~kdundant
I
I
F. Refurbish Rectenna
G.
Panel Asssembly A f t e r
1 5 Years
98
1.1.0.8 C o m m a n d & D a t a Handling System Failure Modes a n d E f f e c t s Analyses
(Provided by TR W)
A. System Redundancy
8. Data Bus
1) The use of fiber optics was recommended for t h e same reasons outlined
in the information management section of volume 3, however, a concern
was expressed over t h e degradation of t h e optical conducting material
during t h e required 30 year lifetime.
A. Loads Analysis
The loads analysis was performed on one module of the solar power satellite (4x8
bays). The maximum load condition was determined to be that of a satellite module,
kith one antenna payloaa, undergoing self-powered orbit transfer from LEO to G EO.
For the loads analysis the maximum acceleration of the module was assumed to be
1 0 ' ~ ~ The
. acceleration of the satellite module is achieveti by using electric
thrusters mountecr at each corner.
7 he basic configuration of the module used for the loatis analysis is shown in figure
1.1.1- 1. The antenna would be installed on the back of the module (opposite the solar
array) attached to the geometric center of that face.
The structural fell (bay) configuration shown in the part 111 documentation was revised
to provide a suitably stable unit. Figure 1.1.1-2 illustrates the original and revised
structural concepts. In the original system the edge cells of each of the eight modules
making up the ent~.eSPS used the configuration illustrated. The interior cells
employed an absolute minimum of structure. Further analysis indicated that the ease
cells were not stable with the result that the enti-e system was not stable. Further,
the 7Yi meter beams were not adequate for solar blanket tension when the solar
blanket tension was changed to uniaxial. As a result, the system was revised to the
configuration indicated for solar blanket tension support and all cells incorporating the
structural concept shown. The lower-deck-to-uppyer-deck diagonal provides structural
stability.
ORKiVIAL REVISED
ALL 7KM BEAMS ALL'PkYBEAMs
ExCEITASluOlED
NOT =ABLE
7UM BEAM UYADEOUATE
FOR lOUR DLMlKET-
Type A beams a r e designed for accommodating the bending load resulting from blanket
pretension. A bending load of 4.285 Nlm resulted from the necessity t o achieve a
blanket first mode frequency of 0.0024 Hz, two times the system first mode
frequency. To react this bending load a 12.7 meter beam width and a 15.0 meter
batten spacing was selected. The beam characteristics are shown in table 1.1.1-2.
Type t3 beams a r e desigwd for all satellite lateral beams to accommodate the
maximun column loads that result from the l ~ orbit- transfer
~ ~acceleration.
Ty$e C beams are designed for all other satellite beams to accommodate the lower
c olumn loads shown previously in table 1.1.1-1. The majority of the beams in the
baseline system are type C beams.
1he structural mass o! the satellite (without the MPTS or support) was calculated t o
be 97211.7 MT. Figure 1.1.1-5 and table 1.1.1-3 summarize the new module
configuration, dinlensions ana structural mass breakdown.
-
UPER AND LDIVLR SURF-
UERAL
T m C
WCiSEOIWAU
OTHER LOCATWmt
$ U M E R SURF= -
LATERAL LCI-
7 L W E R SURFACE L W W I N A L - 667-
8 LOlVGIfUDllYALMITERSURFACEOJAOOWIU 8- -
CORUER BAY
9 -
a R N E R r(lSTS 460-
-
,
,~EZ-?''~
b E A Y T M CQTF.w? N0.w- TOTALUfWOTn
rOMKE
3 0 10 28ZW 61122 1
-
4 A s 28031 7.4~ l'ls.6~
6 C 10 32e66 4.108 . l o b
8 I 10 800 6.122 1S.m
' I :
% 24oXl 4-10. a
67
8 s 32656 4108 13;1.aa
Om- - 1lS8.18Wt
S%YASSfORJOlNlS - 67.91 M l
P-
TOTAL - 1216.OBUT
This study was focused on the design conditions f o r Boeing's lOGW SPS,
which i s fabricated i n law earth orbit. A s shown i n Fig. 1.1.1-6, this s a t e l l i t e
consists of eight attached solar array modules and two m i c m v e antennas. Each
solar arrw module, 2678 m by 5348 m, is self transported to geosynchronous o r b i t
using thrusters mounted st each of f a corners. 'Rso of the d u l e s are coupled
with antennas. The complete satellite i s assembled i n GEO.
Estimates of design loads were made f o r t h e c r i t i c a l design conditions;
stiffnesses were calculated and used t o estimate t h e natural h q u e n c y of the
complete s a t e l l i t e . Several thermal design conditions were selected f o r evalu-
ation t o assess the response of the aluminum structure t o the thermal environ-
ments. The very lindted studies carried out, p a r t i c u h r l y i n the t h e m /
s t r u c t u r a l area, indicate that the use of aluminum i s feasible. l i ~ ~ e ar ,
considerable amount of design and analytic e f f o r t must be done t o reach a m r e
definitive conclusion.
The major conclusions derived from t h i s task a r e as follows:
Roll formed closed section aluminum structures can be automatically
fabricated i n orbit.
Design load requirements f o r LEO constructed SPS module a r e satisfied -
aluminum 2346 (2.82 x 105 ~ g heavier
) than composite but mybe lower
i n cost.
10 GW SPS natural frequency with aluminum (AR 4) i s 65 times o r b i t a l
frequency - instead of 100 times.
MICROWAVE
ANTENNA
/
The basic design data and requirements supplied by iweing f o r the study
are shown in Fig. 1.1.1-7. The t h r u s t t o weight r a t i o i s applied t o the 4 b a y by
8 bay modules f o r s e l f transport t o UO. The SPS s t r u c t u r e n a t u r a l frequency of
1.2 x hz (4.32 cph) which is 100 times o r b i t a l frequency i n GEO i s used t o
assess t h e aluminum s t r u c t u r e frequency and a l s o used t o e s t a b l i s h the s o l a r
array blanket pretension a f t e r increase by a f a c t o r of 2.0 t o 2.4 x ha.
MASSDATA
SOLAR ARRAYS 5.178 x lo7 kg
*
MW ANTENNAS
WT GROWTH
TOTAL w
2.521x 107 kg-
STIFFNESS
2935 OOJV
The 30 year service life, when increased by an spprapriate s c a t t e r factor, i s
used t o evaluate time dependent f a i l u r e modes such as fatigue, f l a w growth,
creep and environmental degradation e f f e c t s . It i s expected tlmt low s t r e s s
l s v e l s may permit aluminum t o s a t i s f y the l i f e requirements.
The more ~ i g n i f i c a n ts t r u c t u r a l loadillg conditions shown i n Fig. 1.1.1- 9
curren+,ly are t h e s o l a r a r r a y blanEet preload and loads caused by transport of
the 4 bay x 8 b w module t o GEO. The first condition causes a high l o c a l cap
load i n t h e 7.5 meter beam; t h e second induces t h e highest column compression
load i n t h e 7-5 m by 667.5 m beam. Inasmuch a s aluminum has a c o e f f i c i e n t o f
t.hermal expansion g r e a t e r than t h e advanced structural composites, t h e e f f e c t
of gradients on d i s t o r t i o n s , s t r e s s e s etc., are evaluated. Thermal control
features a r e incorporated i n t h e design t o minimize thermal/structural response.
These include thermal coatings, incorporation o f lightening holes i n members,
etc. b a d s induced during f a b r i c a t i o n and handling w i l l a l s o require assessment
ir, l a t e r phases of t h i s program. m e s t i f f n e s s requirements a r e used t o define
t h e loads on t h e beams supporting t h e s o l a r array blanket and t o evaluate t h e
o v e r a l l SPS n a t u r a l frequency i n GEO.
Further discussion on loads due t o o r b i t a l t r a n s f e r and s o l a r array
areload a r e provided below.
Ioads - Orbital Transfer Condition
Figure 1.1.1-9 shows the 4 bay by 8 bay m d u l e which i s constructed i n
LEO and 5ransported t o GEO f o r assembly i n t o the full s i z e SPS. Eight of these
modules a r e joined together t o make up t h e 21420 m x 5348 m s a t e l l i t e . The
t h r u s t e r s a r e supported on four outrigger structures. A t each end of the module
s o l a r array blankets a r e deployed t o provide power during transport. This arrange-
ment defines t h e two basic design conditions which a r e c r i t i c a l f o r t h e primary
s o l a r array structure. The t h r u s t forces balanced by i n e r t i a forces induce
benc!;,~ @;d shear i n t h e t r u s s ; the l a r g e r member loads occur i n the 5348 meter
directa-on. These loads a r e t h e l a r g e s t column loads on t h e 667.5 meter beam.
The s o l a r blanket preloads on each of t h e two end bays cause t h e maximum l o c a l
cap compression load,
Figure .]..I.-10 shows the module with t h r u s t e r loads applied t o t h e
cdmbined module and antenna mass; t h e antenna i s supported at points A, B, C
and i). A dynamic magnification f a c t o r of 2.0 and a safety f a c t o r of 1.4 were
used t o obtain ulbimate dcsign loads i n the truss.
DIRECTION
SOLAR &!WAY BLANKET ON EACH END BAY ONLY
OF CONSTRUCTION /
CONDITION: SELF.TRANSPORT TO
114
The summary of c r i t i c a l member load3 f o r t h e t r u s s a r e shorn i n Fig.
1.1.1-11 f o r t h e module p l u s antenna transport condition. The loads aye based
on t h e required t h r u s t t o weight r a t i o o f The maximum load condition f o r
t h e 667.5 meter beam uas used t o w i s e t h e member as a l a t t i c e column; the
r e s 9 d t s show t h e e m b e r has a lO$ margin of safety.
bads - Salar Array
----Pre(.oad
The I80 baseline configuration u t i l i z c s s four b e wide ~ construction
base t o f a b r i c a t e t h e 4 bay by 9 b8y module. D u r i r ~module construction, t h e
1 5 meter wide s o l a r a r r a y blankets a r e i n s t a l l e d on t h e two end bays of the 8
bay length as shown i n Fig. 1.1.1-12. The 15 meter arrays a r e interconnected
along t h e i r lengths and uniaxially pretensioned such t h a t the blanket natural
freqzency i s 3.64 cph. Sending moments, caused by t h e pretension r e s u l t i n high
a x i a l compression loads i n the caps of t h e 567.5 m beau. This condition gives
t h e c r i t i c a l load i n the cap of the 7.5 meter deep beam.
Figure 1.1.1-13 i l l u s t r a t e s t h e loading system on t h e t r i a n g u l a r cross-
section beam when t h e running preload of 4.29 ~ / r n i s applied t o t h e 667.5 m span.
The maximum bending moment a t t h e midspan i s 3.35 x 105Nm. m e curve shows the
compression cap load a s a funct5.on of beam depth. The cap load f o r t h e selected
4
7.5 m depth bean i s 2.58 x 10 N compression.
AGNlNUM STRUZTLTRE: DESIGN
--L2am Design
An aluminum beam design, which can be fabricated i n space f o r the SPS
s o l a r array structure, i s shown i n Fig, 1.1.1-14. The aluminum + . + a n g u l a r cross
section beam design incorporates thi e r o l l formed closed sect;:. .,ps in,ce-
connected by battens spaced at 7.5 meters. Shear s t i f f n e s s C& ue provide ~y
e i t h e r preloaded cross cables o r cornpressiou/tension members. The cable concept
i s approximately 20$ l i g h t e r and has been selected f o r the baseline aluminun
structure. However, pretensioned cables f o r shear stiffenill: may induce
p o t e n t i a l problems such as: adjustment of a l l csble tensions t o t h e proper
prelosds t o prevent slack a t any time, f a i l u r e of cable attachments, p o t e n t i a l
f o r excessivs material creep deformation under sustained load and temperature
f o r 30 years increased by an appropriate s c a t t e r facto?, e f f e c t of selected
cable system on l a t t i c e column capability, otc.
NOTE YHC TENSION l I N I A N l h 1 w l A H BLANK&1
LOADS ON UPPER B t AMS 00 NOT ACT ON
AROVt Mt MBtHS Ht NOlNC CAUSE 0 BY
SOLAR ARHAY FR 1tNSION OCCURS ON
7 5 nt BEAM5 . OHMA1 I 0 AClilVI; BEAMS
INCCUOt S MtrC)UI C IC ANT 6NNA M4St S
i lo
WAX MOMENT ON 661 5 m B E U . 2 . Y & 1 9 IN. LB ULTINATE (3 35 n 16NU1
n
; L
10
L..
1s
'METERS
113
The selected cap s i z e for the design laads is .2 deep end has a
nt
thickness of .081 cm, The batten is also a closed section with the bot-
flsaees exkadiag outward f o r 8tbchnreIlt to t h e cqp. The deptb is 10.16
and thickness of -05 cm,
I n order to minimize the& gradients i n nrenrbers and between members,
flanged Ughtening holes have been spaced to reduce s h a o a w i ~as much as possible.
Several thermal coatings have also been evaluated to maintain temperatures and
gradients within acceptable limits, although additional studies are requixl,
Lo obtain an optimum design.
Figurz 1.l.i-16 shows the required thickness and depth of closed cap c r o s s
section f o r the 7.5 meter deep beam with a batten spacing, L, of 7.5 meters. The
curves are bzsed equating section crippling strength to column failure strength
f o r -various column f i x i t y conditions. These data were developed as an i n i t i a l
design optiaization proce&we. However, t h e selected cap design represents an
o f f optimum configuration because of t h e requirements to provide large lightecing
holes t o reduce t h e 4 gradients, to permit s u f f i c i e n t attachments between
battens and cap and to provide s t i f f n e s s f o r column s t a b i l i t y . The selected
secticn i s .2 m deep and .081 cm thick.
Candidate Material Properties
O i - i - -I i- . A - S J _J
0 2 4 6 8 10 12 14 16
PC" LBF x id
Figure 1.1.1- 16 Aluminum CIooed Section Beam Cap Thkkness & Depth vs
Clitical Load; L = 7.5 m
:,%~ 0,2"
rrEIGnT.'uNIT
LENGTH kp'm
78
i 20
The area of joints trnd attachment techniques requires an i n depth
developent program to evolve sn optbum attachment oethod which requires mi-
service, provides sound attachmnts, creates no debris, requires l o w pover and has
a high degree of reliability,
Candidate Truss Confiaur8tions
A l a n = Structure Weight
22,
20-
\
10 -
U -
8-
0
I I
THERMO/STRUCRTRAL ANALYSES
>*,,a,,"
F 12 1 Orbital Attitude During Construction
TEWERATURE - OC
HORIZON1AL CAP
2W5.oIIV
Fipre I. 1.7-22 Aluminum Structure Cap Temperature vs Time
1 24
Other arrangements considered are t h e severe cases where t h e sun is
normal to one of the surfaces at 0' o r b i t angle and where one element shadows
another. The v e r t i c a l beams, where i n t e r m i t t e n t shadowing takes place, were
a l s o investigated.
For t h e construction phase, a 560 Km a l t i t u d e c i r c u l a r o r b i t i s
considered.
I n GEO-synchronous o r b i t , t h e gradients between t h e sun-side horizontal
beams, and those opposite have been calculated,
For this study, the i n s i d e of t h e elements a r e coated with black anodize
( e = .83, cr -. .86) and t h e outside surface with 2-93 white paint ( 6 = .go,
cr = .17).
The l l r ~ i t e dstudies done i n t h i s segment w e r e selected t o surface some
of the major thermn-'structural problems; it by no means represents a t o t a l review
of p o t e n t i a l problem areas. Considerable additional study e f f o r t i s necessary
t o point up problems and solutions.
Figure 1.1.1-22 shows time-temperature curves for selected locations
around t h e 0.2 meter deep cap member o f t h e 7.5 meter deep beam on low e a r t h
o r b i t f o r a coating of 2-93, a white p a i n t ( s = .90, a = .I?) selected f o r t h e
outer surface and a black anodize (6 = .83, u = .86) f o r t h e inner. The peak
gradient was lgOc ( 3 6 ' ~ ) a s shown f o r the selected sun orientation.
Stresses and Deflections i n 7.5 m Beam
The thermai gradients were applied to the beam caps a t segments A-B
and C-D, as shown i n F i g . 1.1.1-23. The cap from A t o B was assumed simply
supported at 4. and fixed a t B. The curve of s t r e s s versus time shows t h e peak
compression s t r e s s of approximately 16 x lo6 N/$ (2200 p s i ) f o r the section.
The =irnum deflection i s 0.48 cm (0.19 inches). The s t r e s s l e v e l s at C and D
a r e lower than t h e values a t B.
Beam Length Versus Temperature
Figure 1.1.1-24 shows t h e temperature f o r t h e 7.5 meter beam a s a
function or' o r b i t time. The peak temperature excursion i s 65Oc (11~i'F). Depen-
ding on beani builder r a t e and t h e time of start and completion of the beam i n
the o r b i t , the lengthening o r shortening of t h e beam i s shown i n t h e curve of
beam length versus temperature change, The mrst case shows a length change of
.g meter; a s t a t i s t i c a l evaluation i s required t o fully a s s e s s t h i s p o t e n t i a l
125
MAX. STRESS VS TIYE
AT POINT @
),
- - -JO
#
------
TIME MINUTES
-20
SECTION
-55*i*v Fqure 1.1.1-23 Aluminum Structure Strefs & Deflections 7.5 m Beam
TEMPEAATURE - 'C
W I N G V l l l U I U U AVE TE- ORlENT4TlW 3 X
00
A UODt301
BEAM LENGTH i VS. A r
D -lo2 FOR VARIOUS Y
10
C mo~aoa E METERS
rn c- 0
0 - 17
30
rO '3 Y
METERS
50
m
0 R W b O R l W
TIYE WIN.
-
N-
:-
- rn
m
3'
I
I
DEFL.
T METERS
100 -
-- f>
p " 1 " " I
0
~ M E T E R S X103
195, '.,*"
Figure I. 7.1-25 Aluminum Structure Satellite Deflection Due to
Thennal Differential IGEOJ
W BS Item 1.1.1.2 Satellite Energy Conversion Solar Blanket
E f f e c t s of array shadowing were investigated. If a segment or section of a long
solar cell string is shadowed, i t will not g e n e r a t e current. Current flow in t h e e n t i r e
string is therefore interrupted. If other strings a r e connected t o t h e load in parallel
with t h e shadowed string, t h e difference b e t ween their loaded output voltage and
t h e zero-current voltage of t h e shadowed string, appears as a reverse-bias voltage.
The reverse bias voltage, if i t is more than a few volts per shadowed solar cell, will
destroy t h e shadowed cells if they a r e not protected. A small spacecraft flying across
t h e f a c e of an SPS could severely daniage a n unprotected solar array. Reverse-bias
protection i s therefore mandatory. The solar blanket panel design was modified
t o include shunting diodes.
Annealing of t h e solar blanket on one SPS will require a technique tailored t o that
purpose, as well as a blanket design compatible with annealing temperatures Atten-
tion hzs been given t o laser directed energy annealing under this contract.
The concept cf the actual annealing system is shown in Figure 1.1.1-29. Each laser
gimbal would acmally have 8-500 watt CG2 lasers installed. The laser beams would
be optically tailored t o provide the desired illumination pattern and energy density.
The gimbals would be mounted on an overhead gantry that would span t h e entire
bay width; one bay of solar array would be annealed in fifteen meter increments.
Solar array strings undergoing annealing will be taken out of service while they a r e
annealea. Table 1.1.1-4 summarizes predicted performance paranie ters for the annealing
systems.
Table 1.1.14 Gimbded bnning Laear C%anwmi'sai# U m
r aMwa@JGEWLRBYDE~TY: HWIJ~N?
r O m R D m aw 1 d
T- (ACTIVE REGION): 66aOc
r LASERSIGI~AL I
POWER REQUIRED-
.
# I m R RE0UIREDMA)lmY:
CmIMk
A7 cnrh
aum
1.17 W11
To assess phase control system performance extensive use jointly with J X has beer, made
of array simulation computer programs, particularly i n the areas of grating lobe levels and
subtilt effects. A qualitative summary of results is given on Table 1.1.2-1 prior to
discussion of quantitative data derived during this contract period.
The main conclusion i s that main beam wander is primarily due to correlated pha,*e errors.
This can be greatly minimized i f the number of branches at the first level of phase distri-
bution i s increased so as to randomize the errors. In addition, a joint paper presented at
the International Conference on hlodeiing of Electronic Systems, in Toronto in September,
1378, by Dr. D. Arndt of JSC and S. Rathjen of Boeing defined antenna tilt requirements
rrlore clearly and preliminary calculations made by Lincoln became available.
The continuing modeling work on the phase control system proposed by JX should provide
the specific method by which to redistribute energy betweep sidelobes, error plateau in
the array roll off and grating lobes. In phase 2, trade studies of phase error build up i n
cables (including fiber optics), 'black boxes," the appropriate number of phase distribution
tree levels and subarray size should provide detailed answers.
It is s t i l l our feeling that a back-up phase contro! system should be analysed and compared
with the Lincom system in terms of phase error buildup and ease of implementation. This
would, of course, have to be carried to the component demonstration level and would
require additional resources.
ANTENNA TILT
SYSTEMATE 1ILT r MAIN BEAM SHAPE UNAFFECTED BUT
POWER REDUCED BY SCAN COSSWHlOl
APPEARS AT GRATING LOBES
MAIN BEAM SHAPE UNAFFECTED;
RESULTING AWLITUOE m M U U T l O l l
CAN RASE ERROR P U TW
D180-25037-2
Four different distribution t r e e s a r e outlined on Figure 1.1.2- 1, e a c h indicative of different
level of phase control and degree of power splitting. The baseline system of Lincom
with 9 nodes (dl is based on a maximum of 4:1 power split whereas t h e 4 node system
(a) elaborated by GE is based on a 2 0 x 1 9 ~ 1 9distribution t o t h e subarray level and a
variable power splitting t o t h e klystron level (4th node) using 4:1 power splitters a t t h e
edge and 36: 1 power spiitters a t t h e a r r a y center.
The nine node system suffers from poor reliability, poor phase randomization (i.e., resulting
correlated phase e r r o r s which produce beam pointing errors) and lowest ailowable phase
error per node. The four node system may be a viable candidate if 10 rn x 10 m subarrays
a r e retained and phase control is exercised down t o t h e klystron level. A three node
systern n-ray be possible if phase control is exercised at t h e stibarray level only, and subarray
s i z e is reduced t o 5 m x 5 m.
To reduce phase correlation effects, i.e., beam steering errors, t h e number of branches
at the l o ~ e levels
r should be kept high. This also allows higher random phase error per
level in t h e error budget. Even with 4 m x 4 m subarray, t h e $-level system will require
a total ( I GHz) phase accuracy of 2' per level t o achieve a 96% efficiency including
tilt. This will require stringent design criteria. A possible 3 m x 3 m subarray could
be acconlmodated by a 32 x 16 x 16 x 8 four node distribution system.
Phase error buildup a f f e c t s only low level f a r sidelobes where power density i s low and
does not appear t o constitute a n environmental problem. Furthermore if t h e phase errors
a r e correlated they a r e not expected t o b e of great importance in a systeni that has
more than about ten branches per node.
Table 1.1.2-2 lists calculated c a b l e lengths for t h e GE 4 node system mentioned earlier
for representative cable length a t each node. Since the cable has about a tenth the
Inass per u n ~ length
t of waveguide with a total length about a tenth t h a t of the waveguide,
the resulting cable mass is around 1% of the waveguide mass.
10m SWARRAY
Sbdb TAPER
REDUNDANT -2=
-2Wm
-
m - 1 3 BRANCH SYSTEM 28.880 SUBARRAYS 262,114 ELEMENTS
--
Figure 1.1.2- 1 Potentid Phase Distribution Tree Layouts
Tabk 1.1.2-2 Estimate of Required Cable Length for GE Phase Distribution System
ALL CABLE LENGTHS EOUAL
1ST LAYER 20 CABLES W 260 m TRWLY REDUMDAN'I 15 km (IN"DIA)
ZND LAYER 380 CABLES. 250 m UWBLY REDUNDANT 190 km (114" DIA)
3RD LAVER 7220CABLES* 100m NON-REDUNDANT 722 km (1/8" DIA)
4TH LAYER 100,784 CABLES 8 6.9 m NON.REDUNDAN1 702 km (I@" DIA)
1629 km
NON-EWALCABLE LEffifHS
SAVINB - FACTOR OF 2
Calculations by Lincorn show t h a t going t o smaller subarray size desensitizes the transmitting
antenna performance degradation due t o systematic tilt. Selected results frorrl recent
Boeing computer runs a r e in good agreement with Lincorn d a t a and indicate t h a t if greater
tilts than presently allocated a r e experienced, a review of t h e baseline 10 meter subarray
size is warranted.
--
K LINE AmNUATlON
DEVICE ISOLATION
-
FOR 40 W SOLATION, WE, U 30 DB. I.E.. MAK. LINE LENGTH OF
RG8 CABLE IS 160 METERS 500 tAHZ AND 60 METERS. 2 5 GHZ
LEGEND
LINCOM RESULTS
-
3 arcmin RANDOM TlLT (1 01
Surfro c .O1X (48 mils)
@ 3 r m i n SYSTEMATIC TILT
3~5 rrcmin RANDOM TlLT
@ 3 a m i n SVSTEMATIC TlLT
3 ' F u c m i n RANWM TILT
0 7 8 O GRATING LOBE
NUMBER
MAIN 40
BEAM
W
>
0 -
h FROM BEAM CENTER
ANGLE, DEGREES
I
1
TAPER SIX STEP TEN STEP
r) 9t.I 96.7%
NORMALIZED RADIUS
o C.ip.~brI~!y 10 IIIP~C'I
flit- spat.Ctcrirla ~ I tc
O I
U thc power niodulc lcvcl u ithout esc t - ~ s i1-i
5tor.iSc rquirevtwnts.
Lj tar "iio~1111~111i"
Itas bcrn ~ilatcltedto .i no-crror "T~ittnain"run for a I 0 11) x I 0 111 ~ i ~ t u r r a ! .
T l ~ c *t IIst ~ t i ~ t r ,bctuccrl
t\ the tuc; progr;r:lis uas the ~ o ~ I ~ell-ctrrc-
J I I ztleld
c ~partern
IIC.~~ t power dcnsity as u el 1 as the dci'lbt81
tllc rCc.tCiilld. "llodr~tnln"now pr illts c ~ the
valuc*s. Et trr,icricy c'alcul.atiCns \sere ncx t ~natchedusing 10.43 by 10.43 mt8trr strbarrays.
As cuprctcd. t i ~ t use
* of 5 111 by 5 111 suhrrays ln the basic prograln "iiig~t~arrr"
crccrtcd
.i .~tor.fc- ~wcrlo.ic!prv,>lc~n
btit could be niodeled 111 " h l o d ~ ~ i ~ i n . "
ALL S U U R I A T S
-i
CALCULATE
ECFICIEWCiES
2
There is a f a c t o r of 10 c a b l e mass reduction if o p t i c a l f i b e r s a r e used and a l s o J
s o n l e u h a t lower c o s t per unrt length. This has l t t t l e iii:pac: o n X1PTS masses a n d costs.
however, because t h e phase c o n t r o l system cabling rnass a n d c o s t fractions of t h e MPTS
a r e u e l l under a tenth t o begin u ith.
Ftber optics have less signal loss p e r unit line length than coax, so less a n ~ p l i f i c a t ~ ouni l l
be needed In t h e system. More valuable than this, though, may be t h e f a c t t h e opt rcal
fibers have rto s h o r t c i r c u i t e d f a i l u r e modes, rnaking fiber failures independent. They also
allow t h e transmission of signal across high voltage differences u i t h o u t introducing
possible nlet.illic shorting paths as conventional c a b l e s do.
ATTErJUATION
dbflun (100 MHz) lWdb IS 5
MASS .5
43 160
k&m
COST
Shm a= 4)oo 1- 11978)
PHASE
CHANGE
FOR Oil0 4.w
ar-lm
L-3oRn
L J
I F REFERENCE SIGNAL
CENTER & R M Y
REOUNDANTARRAY OF
LIGHT EMITTIMG DIOOES
(10 OIOOES 0.1 clr EC
I UOcMooESQ1nrEAcn~
HEXAGONAL BUNDLE
CONTAINS61 HEXAGONS (M0 OUJ
EACH WITH 61 FIBERS (#16 LACHI
6 MILS EACH
AVALANCHE DIOOE
ENVELOPE DETECTOA
RECOVERED
I F REFERENCE SIGNAL
W E LINK CALCULATION
- -
FIBER LOSS a Sdb
FOR 1 mrr RAOlATED POWER, RECEIVED POWER IS 60dbm 1mrn
AVAUNCHE D m RECEIVER WILL HAVE &N > 20 db THIS LEVEL FOR 6 MHz BANDWIDTH.
0 180-25037-2
fiber redwdancy for t h e case of a fiber per klystron and a twentyfold redundancy a t t h e
optical signal source, which is proposed t o b e light emitting diodes but could b e lasers.
If the source is coherent, single m& fibers may b e used, but incoherent sources require
multimode fibers
A number of candidate optical fibers in use today have been considered and a r e summarized
on Table 1.1.2-5. The selection of the most suitable fiber will depend on t h e temperature
requirements and the radiation environment on t h e array. The fiber optic cabling system
looks sufficiently promising t o recommend tests of a selected s e t of fibers t o verify
their temperature stability and link budget. Separately, a postulated radiation environment
should be used to determine t h e possible e f f e c t on fiber life and required protection.
In summary, optical fiber technology appropriate for MPTS reference phase distribution
systems is available today. It i s recommended for use because of nurnerous advantages.
The assumed layout of t h e klystron module is shown on Figure 1.1.2-1 1. Possible locations
of the solid s t a t e phase control modules for good thermal distriburion a r e sh0n.n. Selection
of a pilot receiving antenna has not been finalized. What is shown is a n example of
an approach t o be considered for achieving good uplink/downlink isolation when t h e
power beam and pilot beam a r e a t t h e s a m e frequency. The radiator is indicative only
and is not representative of an a l t e r n a t e lower mass active cooling system under consideration.
EFFICIENCY DEGRAOATIOIJ
loMXlw4 SMXW
ALUMINUM WAVEGUIDE 1.18X 1.18%
An 7.50% L41X
Y
D 180-25937-2
1.1.2.3.1 Solid State Amplifier Technology for SPS
Introduction
The approach t a k e n for a n SPS in t h e 1990 t i m e f r a m e is to a s s u m e t h a t today's s t a t e of
t h e a r t will b e in mass production by 1990 with relatively minor improvements in
p e r f o r m a n c e parameters.
A N ide variety of suitable solid s t a t e a c t i v e devices currently exist. These include bipolar
a n d f i e l d e f f e c t transistors, many types of two-terminal devices (Gunn, IMPATT, TKAPATT,
and BARITT diodes) and hybrid devices such as e l e c t r o n bombarded semiconductors
(EBS). EBS h a v e been included a s being solid state s i n c e t h e e l e c t r o n b e a m only supplies
a s m a l l c u r r e n t , with t h e bu!k of t h e supply c u r r e n t staying in t h e semiconductor.
Si BIPOLARS *- ANOMALOUS
4 \
\ GaAs FETS
.6 p \ /
/
'
\ GaAs FETS
* \ +
\
'.
+
\
\
> o +
\
+ 4 G.AsREAD
O 0 ' p a*, \
*
d
-*- -*
---. *
+ . .+\
A ?'
-- - 2TERYUiAL DEVICES q
.
IwATT-YULWEU
a6n
10- <la%
2 6 19
FRMUEIYCV.Gnz
The present state of t h e a r t in switched mode amplifiers i s con-,rn >n use in RF ampli-
fiers at tens of megahertz. Experimental amplifiers at over 103 MHz have been
~uccessfullyachieved and microwave amplifier experiments a r e pending. More discussion
oi efficient utllplification techniques is given in Keferences 1 and 2, with discussion
of t h e particular t. .)*. . ' switched mode amplifier known a s the class E amplifier
in Keferences 4-6.
DcvbG.ira
Achieved device gains vs frequency a r e shown on Figure 1.1.2- 14. There is a s t r i h n g
difference between small-signal and power gain for F E T s At the SPS frequency of
2.5 CHz bipolars have about 8 db gain while CaAs FETs yield around LO db. In general,
GaAs FETs have sever& db more gain than b i p o k s throughout the spectrum. As for
t h e other devices, 1MPATTs can have gains of over 20 db and electron beam
semiconductors are projected to yield about 20 db. The low gain of S t s t i r Induction
T r a s i s t d r s (SITS) a t I GHz e l i n ~ i ~ t them
e s from consideration a t present, although
they appear t o have great potential for further developmenr due t o ti1e.r high power
b a d w i d t h product.
Device Power
The power per device is an important SPS parbmeter since the number of devices
which can be ef f i c ~ e n t l ycorrlbined in a module is limited by circuit bsses and the
power per module determines the RF power density per m i t transmitting array area.
The sirigie device power chart (Figure 1.1.2-15) sh@ws that silicon bipolar transistors,
GaAs FETs and multi-mesa 1MPATTs can all handle powers above 10 watts, which is an
adequate power level for SPS application. Of t h e devices considered here, only
E-beam sei:licondoctor devices are capable of generating a power level of IGO watts
per dev~cctwhich would be adeqtiate for one device per radiating element. For the
other devices, power combining will be necessary.
Device Lifetimes
The f d a m e n t a l uearout failure modes in semiconductor devices tend t o be concen-
trated at surfaces, both internal and exposzd, and are gei~erallyelectrochemical in
origin. In the case of the internal surfaces, t r a n - v r t of species t o and away trorn
interraces eventually degrades coiltacts. In the case of external surfaces, impurities
can come in from outside t o form compounds and oxides and high electric fields can
cdu.u breakdown.
4
EBS cathodes presently have an expected lifetime of 2x10 hours, over an order
of magnitude less than that required for a 33-year satellite, so they appear unsuitable
for SPS. The two remaining so!id s t a t e amplifier candidates a r e GaAs FETs and
Si b i p l a r transistors. Si bipolar lifetimes a r e limited by electromigration of emitter
finger metallizations due t o localized high current densities. This give5 relatively
sudden and complete hard (open or short circuit) failures, whereas GaAs FETs seern
t o suffer from contact degradation which decreases performance gradually.
Bipolar lifetimes have been improved recently by going t o gold rather than alumlnum
metallizations, but at r :asonable operating conditions and equal junction tempera-
tures, GaAs FETs have a definite performance advantage. A current GaAs FET
lifetime vs temperature curve is shown on Figure 1.1.2- 16.
Configuration Description
Synthesis of a solid state transmitting antenna configuration has been accomplished
using desig~tcriteria for radiating module s i z e developed in previous NASA studies
and several concepts developed in prior proprietary Boeing Il?&D work.
The basic elements of the solid s t a t e transmitting array a r e .59X x .59 X = free
space u.svc!?ngth) radiating modules. These a r e fabricated on 20 mil thick alumina
dielectric shet-t which is metallized for signal, control and power circuitry. A candi-
d a t e cavity radictor design has been subjected to initial test. Microstrip techniques
a r e used for combi.>ing, filtering, and making antenna elements witn 5 d b gain over
isotropic. Ways h a l e been found t o efficiently combine outputs from up t o
6 ampliiiers.
G.k FET
a SOYEARfUWENmCE
LOG)YDIIYALFAlLURE0#IA~
a-1
LUNDGRENAND LADD.
13711RELIAR PHVSICS
a;rYP,m255all
For design purposes. a module combining six % w a t t amplifiers has been selected.
These 30-\! a t t , 15-volt pou.er amplifier rnoduies a r e to b e c o n n e c t e d in 3 series/paral-
lel hierarchy illustrated in Figure 1.1.2- 17. Modules a r e c o n n e c t e d 4 in parallel
to form unifs called rows. Twelve row; a r e c o n n e c t e d in s e r i e s t o f o r m strings.
7
Three strings in parallel m a k e up t h e panel, which is 0.73 m' in a r e a and is considered
to be i h e least replaceable unit. O n e hundred forty-four panels in a 12x12 series-
parallel rnatris f o r m subar-ays of t h e s a m e s i z e as t h e c u r r e n t baseline with a voltage
of 2.16 kv.
3Qnr KWTER M00Ul.E
-
ROW 4 YOMILES IIJP - L ~
Sever31 t h i n g s a r e apparent from t h e mass estimate. First of all, the mass of the
d ~ e l e c tic
r doinina:es and c a n almost certainly be reduced by going t o designs with
less dlelec tr ic volurne and/or less dense dielectric.
D 18025037-2
r f
-
W E N ON l.2X OPEN AND OPEN AND
OVERVOLTAGE NO RESPONSE
LOAD ON 1.05X LOAD la LOAD ON 1.15X
FUSE ON ZX
EXCESSIVE CURRENT FUSE ON 3X NO RESPMYSE OPEN ON 1.5X OPEN ON 1.6X W E N ON 1SX
LOAD ON 1AX
A
4
SERIES STRINGS OF 12 ROWS
4 MODULES PER ROW
2 AMPLIFIERS PER MODULE
PROBABlLlW
OF STRING
FAILURE
F8
. .
I
/ I / I b
.O1 .1
WOBABILIN OF AMPLIFIER FAILURE
3.
Total 4.87
Secondly, t h e housekeeping masses of support s t r u c t u r e a n d phase f e e d a d d up t o a b o u t
a third of t h e total.
Mass E s t i m a t e s
A simple mass model of tt,e SPS microwave t r a n s m i t t i n g a n t e n n a has only t w o terkns:
o n e depending linearly on a n t e n n a a r e a , which has a coefficient o f approxinlately
DIRECT HIGH VOLTAGE DC
PERFORMANCE PENALTIES
DC-DC CONVERTERS =lk&w
POWER LOSSES I N CONVERTERS
SERIES/PARALLEL CONNECTIOKS WITHIN
SUBARRAYS STILL R E W I R E D
AC W)WER DISTRIBUTION
CONVERT
DCIAC ON SOLAR ARRAY
ACDC AT SUBARRAY
REQUIRES SIP TO SOME EXTENT ON SUBARRAY
..
Cost Estimates
.As in t h e case of rnass estimates, c o s t components c a n b e split into z r e a dependent
and p o u e r d-pendent terms. The e n t i r e phase control system is ~ n c l u d e din t h e a r e a
term and is probably significantly more costly per unit a r e a t h a n in t h e baselined
(klystron) system k a u s e phase must be distributzd to power amplifiers a f a c t o r
4
of 10 smaller. Because of t h e lack of definition of t h e phase contrcl system at
this poi-t, a meani,lgful prediction of i t s cost for a n alternative solid state satellite
is not possible but is anticipated t o rnake a aifference.
Sizing Aids
A nomograph was d e v e l o ~ e dt o assist in ~ o w e beam
r sizing e s t i m a t e s for thcse systerns
(Figure 1.1.2-24). The assumptions in this dasign tool a r e based 611 t h e assumed norrilnal
SPS efficiency chain and a r e described in Reference iO.
COST OF
SpS
ELECTRlCrrV
Imwkwbl
ASSIJWTIONS.
?IC-RF
20,
.I rOWER PROCESS=
175
100
a-
-
a-
w
a
3 20-
-- -
m E MODULE IS 97b r 0.7h 7 6 2
GAUSS'W 1 W B TAPER:
rg %7%,D1D2- 10
--
-r~ -
UNIFORM DISTRIBUTlON:
81%. DID2 8
mWlNAL SPS. EFFICIENCY CHAIN
VALUES OF 01 (IN KILOMETERS!
ARE CIRCLED Ohl EAUi CURVE.
Phase locking may be used t o combine the best spectral charactcristics of several
devices in a systern. If this is done f o r t h e SPS MPTS by locking local Cunn o x i l l a t o r s
t o a crystal controlled reference oscillator a noise spectrum with a floor near - 160 dbclHz
similar to Figure 1.1.2- 26 can be expected.
Since solid sts.2 panels have substantially less a r e a than klystron modules, t h e
L I I ~
MULTIPLIED CRYSTAL
OSCILLATOR ALONE
MfWWIDlH
NOISE CQNTR1BUTK)IY
-RENT +-RENT
a
5. Sokal, N. 0. and Raab, F. H., lEEE J. Solid S t a t e Circuits SC-12, No. 1, pp.
86-88, (Feb. 1977)
6. Sokal, N. O., Electronic Design, V. 20, Sept. 27, 1977, pp. 96-102.
7. F ~ l i i l ~et
t a al., lEEE ED-25, No.6 p.559-563, (June 1978)
pipe s y s x m and certainly more reliable than tile expected lifetime of , e :ilystrons
themselves. Table 1.1.2- 12 shows principal f e a t u r e s of t h e circulating fluid system for
t h e klystron cooling circuit and Figures 1.1.2-28, 1.1.2-29 and 1.1.2-30 illustrate t h e
Vought thermal ccntrol concept.
-
AVERAGE mx 2n METERS
a AT- 56%
ALUMINUM COYPOSIT
la% .101
A
1.1.25 Analysis of Ant- Strutwe Optims
The center illustrat~onin the iigure represents the antenna structure as visualized by
the rrlaintenance engineer. I t provides easy access to subarray repair or replacement
and allows square subdrays but structurally is not very efficient and employs tension
members. The use ot tenslon members results i n dubious dynamic qualities for the
structure. Further, the secorldary structure is r q u i r e d to provide stability of the
primary structure. Analycis of this combination indicated a relatively poor stiffness
ef iiciency.
The pentahedral truss snow n at the l e f t appears to offer a way out. It tr~aintainsgood
access w itt?good efficiet~cy,eliminates tension members and allows square subarrays.
A potentidl interference problem has been ideniified with respect to the operation of
the n~dintenancegantry and the existence of cross bearn members for the prirnary
perttahedral truss structure. This is better illustrated i n the next figure 1.1.2-32.
RESULTS
I. Secondary Structure Model
The first s t e p in the study was t o ascertain t h e vibratory characteristics of t h e
secondary structure, which consists of a series of truss modules as sketched below
in Figure 1.1.2-33. Its relation t o t h e primary structure is shown in Figure
1-1.2-3g.
The above sizes correspond t o a totzil antenna secondary structure mass of 565
SIT, which is 2.9 tirnes t h e rnass given in (1).
hose shapes and frequencies for the truss simply supported at i t s four corners
Are shewn in Figures 1.1.2-35 through 1.1.2-39. The frequency range, 0.39 Hz
to 1.24 Hz, 1s well above t h a t t h e full antenna fundamental. Hence adverse
resonant coupling is not a concern.
MAINTENANCE
STRUCTURE
PRIMARY ~ R U C T U R E
190
0 1W25037-2
t = 0.030 in.
2. Antenna hrlodel
One of t h e main structural features of t h e A-frame antenna is t h e utilization
of the secondary structure as a primary load-carrying agent. Each module is
independently fixed t o t h e A-frame ridges of t h e primary structlrre. Thus t h e r e
i s no direct load path from one module to another. This reduces t h e overall in-plane
stiffness of the antenna. Incorporating this f e a t u r e in t h e finite element model
was considered of prime importance. Figure 1.1.2-40 shows schematically t h e
idealized attachment of a typical module t o t h e primary structure. As a n example,
standoff beam 1-5 is fixed to t h e plate at node 1 and pinned t o t h e primary struc-
t u r e a t node 5. Torsional stiffness is also included. There i s n o direct connection
between adjoining plates, as in t h e actual structure. Although t h e plates a r e
shown slightly separated in Figure 1.1.2-40 for clarity, t h e corners of adjoining
plates occupy t h e s a m e point in space in t h e NASTKAN model. This contact
exists in a geometric sense, but not in a structural sense.
The antenna mass was distributed over t h e surface in accordance with t h e klystron
layout shown in Figure 1.1.2-42. In t h e overall antenna model, structural mass, which
represents about 6% of t h e total, was neglected.
2
do, in A, in L/
All Members Except Diagonals 58.17 5.480 = 200
Diagonals 82.15 7.739 = 200
u T c n XHNTS U ~ D
MlNGED JOINTS 01
NUMBER
NUMBER KLYSTRONS1 NUMBER
-1 276 36
-
STEP SUBARRAYS SUBAR RAYS KLYSTRONS
9.936
M T E N M M L F Ml.ANTI-STWTRIC
* W Y S MO F R E q J E t l E S
jU&RRAI MODAL DlSPLAtrWEMlS SUmN
raOA, OEFW. S W E 2 IIO[)E 2 FREO. 0.173657
2. The A-frame structure is not very efficient because of the relative softness of
the secondary structure in providing stability of the primary structure. A penta-
hedral truss antenrla structure appears t o provide a good design compromise between
structural efficiency and mainterlance access.
I. --.)--
MAXIMUM (WASTE HEAT
PLUS MAX. SOLAR)
PRIOR TO EARTH
SHADOW (WASTE HEAT
ONLY)
-- - MiNI?AUM(END OF EARTH
SHADOW
The analysis was used t o develop design parameters for the transformer as a function
of frequency. The per*,nent parameters for t h e transformer a r e shown in Table 1.1.2- 17.
In order t o increase the overall converter lifetime dielectrics were derated for all
filters in the converter. The losses for t h e revised converter a r e tabulated as a func-
tion of frequency in Table 1.1.2- 18.
In order to select the chopping frequency f a t h e long life processor, t h e curves shown
in Figure 1.1.2-55 were developed for the baseline converter design, the baseline
converter with derated dielectrics in all filters and a liquid cooled transformer as
a replacement for the baseline transf ormer. It is apparent from the curves of Figure
1.1.2-55 that the minimum mass system occurs when t h e liquid cooled transformer
is used (with derated dielectrics in all filters) at a chopping frequency in the 15 t o
20 kilohertz range. This converter concept was selected t o replace the baseline con-
verter concept.
0 18025037-2
POWER I N
POWER OUT
EFF!CIENCY
WEIGHT
INTERNAL SIZE
OPERATING FREWENCY
v
LOSSES IN KW AT CHOPPING FREQUENCY
CONVERTER SECTION
1 KHz 10 KHZ 20 KHZ 30 KHz
IPJPUT FILTER 30 42 48 54
CON0 12 12 12 12
SWITCHING
SW 2.4 12 24 36
DRIVE AVO
22 55 11 16.5
SUPPRESSION
TRANSFORMER 70 70 70 70
PART II CONVERTER
PART II CONVERTER
WlTH OERATEO
0:ELECTRIC MATERIALS
CTRANSMRMER & FILTERS)
PART II CONVERTER
WlTH NEW TRANSFORMER
AND DERATED DIELECTRIC
- MATERIALS
T H R O U G H T H E S P S P O W E R BEAN
Introduction
O r b i t a l Geometry
rs
= 6.63 i s t h e r a d i u s of GEO, and +
i s t h e angle which t h e power beam
makes w i t h the e q u a t o r i a l plane. As t h e SPS nioves around i t s o r b i t each
day, the Seam thus generates a c o n i c a l surface about t h e p o l a r a x i s . The
i n t e r s e c t i o n o f t h e i n c l i n e d o r b i t a l plane w i t h t h i s surface i s , of course,
a conic section: an e l 1 ipse, parabola o r hyperbola according t o whether
i i s l e s s than, equal to, o r g r e a t e r than 4. Since @ < 7.4' f o r L < 50,
the i n t e r s e c t i o n wi!; be a hyperbola except f o r e s s e n t i a l l y e q u a t o r i a l
s a t e l l i t e or.bits. The locus o f the power beam i n t h e lower s a t e l l i t e
o r b i t a l plane s t a r t s a t geosynchronous a1 t i tude when the SPS passes through
t h e r i g h t ascension of t h e ascending node (assuming the o r b i t i s posigrade);
i t sweeps dowri, d u r i n g a p e r i o d o f s i x hours, t o a niininlunl a1 t i t u d e (frorn
Fig. 1.1.2-56
I n a g e o c e r ~ t r i ce q u a t o r i a l i n e r t i a l coordinate systeni w i t h t h e
x - a x i s t o ~ ~ a r dthe
s ascending node and t h e z - a x i s along the n o r t h p o l a r
a x i s , the r a d i u s vector t o a p o i n t i n the l i n e d e f i n e d by the power beam,
a distance b from t h e i n t e r s e c t i o n w i t h t h e p o l a r a x i s , i s
= x = -bcos$sinftt
Y b cos 9 cos R t
z z, - b sin 4
i
-P
= - cos q, s i n R t
cos cos nt
- sin g
I n these expressions, R = 72.7 microradians/second i s the GEO
angular v e l o c i t y and time i s measured from an i n s t a n t when the locus o f
the beam i n the i n c l i n e d o r b i t a l plane i s a t i t s lowest a l t i t u d e , given
by l2.23.
The u n i t vector along the i n c l i n e d o r b i t normal i s
-iz , =
- s i nO i
cos il
I
Now transfor111 [ 2 . 3 ] t o a coordinate syster~~
w i t h the same x-axis
I ~ r t ,w i i . t ~ ~ I I ( > z - ' ~ x i b alon!j tl~c!i r ~ c l i n c t lo r b i t nonltal. Thr? t r a n s f o r l ~ i a t i o n
matrix i s
1 0 0
0 cos i sin i
0 - sin i cos i
so t h a t t h e transfor~nedvector i s
e' I
x'
Y'
= /- b cos
b cos
+ s i n fit
0 cos i cos nt + (zO - b s i n ))sin i
z' I- b cos + s i n i cos nt + ( z O - b sin ))COS i
X =
- r,cos i sin 4 sin Rt
[ s i n g cos i+ cos q~ s i n - c Z J
r 5 s i n 4 cos R t
Y = -. ..- - . . --- -.- -- ---
[sin cos i+ cos 4 s i n i cos
r = r,sin 4 [I - s i n 2i s i n 2R t ] 1/2
[ s l n c r i t cos $ s i n i cos ~ t ]
where
r
s
d - b
cos 4
(D,.-D,)rssin ,* s i n i cos Ct
= 12-13]
's + Z m i n 4 cor i + cos 41 s i ~ i i c ~ ~
;= - Cr cos i s i n 9
~ n ~ c o s i c a s i ! t sinicosgJ, [2m1Ja3
5
[sin cos i + sir) i cos $, cos Ct]'
2
-. i!r,sin I$ cos i s i n O t
=
rGn 4 cos i + cos d s i n i cos cq''
7 [ z .lSbj
In order to use these expressions, [2.9aj 1s solved to give the time
when the beam crosses the given satel 1 i te orbit . For an encounter to
occur, the sate11 ite must then be at the angular position given by [2.9b].
and its cotliponents o f velocity wi 1 1 be
Ii - is]
, =
ccs 41 $in ~t
cos i cos ; cos i!t - sin c sin .; i [2.
T = - (--- +X)
C Av s i n
21 8
26.5'. could e l i m i n a t e encounters w i t h beams feeding New Orleans, Houston,
San Antonio and other major c i t i e s near 31' l a t i t u d e .
cos X
and
C o n c l u s i o n s and Recommendations
1-21 S p a c e - B a s d C o n s h u a i o n S y s t a n s ~
The primary focus of this task was t o investigate alternative construction concepts
and to refine t h e baseline construction concept. Table 1.21-1 lists the phase 1 subtasks
that were performed urder task 4.2.1 of the study and Gsts subsections of this report
that display the study results. Grumman was a major c o n t r i b ~ t o tr o this analysis
under subcontract. Grumman and Boeing results are ictegrated in this report section.
1.2.1.1.1 Intro&ction
The baseline construction concept was designed so that the antenna and the modules
were constructed in parallel. The motivations for this approach were t h a t i ) the
maximum amount of time would be available f ~ constructing
r each of these major
, 2) the antenna and module components have to be mixed in the H L i V
end i t e ~ s and
cargo pallets t o achieve mass limited launch conditions so it made sense t o utilize
the components as they were delivered
One of t h e construction sensitivities that was not explored earlier was t o determine
the ramifications of a construction approach wherein the antenna and t h e modules
would be c o n s t r a z e d in series instead of in parallel. The motivation for exploring
this approach was that potentially the crew size could be reduced thereby reducing
the number of crew modules (the most expensive items at the base).
The objectives of this analysis were t o determine the cost effectiveness and praciicality
of a series construction concept as compared t o t h e baseline parallel construction
concept.
1.21.1.2 Summary
This analysis compares two series construction concepts t o t h e baseline construction
concepr. The series construction approach will reqtiire the modules, yokes, and antennas
to be constructed in about half the time as woald bc required for the baseline approach.
In order t o construct a t this faster rate, i t will be necessary t o operate the construction
equipment at faster rates (Option 1) or t o use more equipment (Option 2), or by some
combination of these two strategies.
1- 121-1 -aummbSLrar-
Spaced-Bastd Construction Systems Armlysis
Task 4.2.1
Sccticm ~tasis tit* of *~ccticm
NOTE: Task 42113 Update Baseline Final Assembly Bue was not performed
as this was a task that was applicable to the LEO construction
approach which was not selected as the reference concept.
Table 1.2.1-2 summarizes t h e equipment r a t e s and quantities and t h e crew sizes
required f o r the t w o options and these are compared to the baseline requirements.
Figure 1.2.1-1 shows t h e n e t e f f e c t of these t w o options. Figure 1.2.1-2 shows '.st
the resulting cast deltas will range from an additional 10% f o r Option 2 to as much
as -20% for Option 1.
Dbcusion:
The a t t e m p t to establish a preliminary determination of realisi~cbeam builder produc-
tion r a t e s started . :th a review of c u r r e n t technology. Sources utilized w e r e CDCIJSC
(NAS9- 15310) X A F E D S Report dfCASD-ASP77-017 uIcl results from GACIMSFC
(&ASS-32472) SFDS in-house test program. An overview of t h e two technological
approaches t o beam making is shown in Figure 1.2.1-3. Both designs a r e based on
cyclic (run/stop) operation where the c a p s a r e formed in t h e "af twbay and t h e bracing
fastened in t h e "forward" bay. A notable difference lies in t h e heating/cooiing operations
required f o r t h e composite beam builder.
The subject investigation concerned itself with a beam builder fabricating a much
larger structure (6.5-m deep x 7.5-m bay length) and operating in a l a t e r t i m e f r a m e
where production r a t e s are considerably more critical. Therefore, t h e study concen-
t r a t e d in identifying potential growth options in t h e c u r r e n t machine r a t e s of 1.08
meterslrnin (composite) and .682 m e t e d m i n (alumiwm). The study was aimed pri-
m j r i f y at the composite beam builder zince t h a t i s t h e baseline structure in t h e pres-
e n t c o n t r a c t e f f o r t , however, comparable d a t a was also derived t o r ,he aluminum
bean) builder t o establish additional credibility.
ERIES
BASEUNE -1 m o
w2
FASTER RITES MORE EQUIIWEMT
EaJlrYENT RATES
BEAU MACHINE -dm& 10d m b Sd n l A
SOLAR ARRAY DEMOYEftS 126dRia Wm/rb la2d*
SUBARRAY DEPLOYERS arb('SUBARRAV 12BnWWUURAV ESJmWWMRMY
#mER BUS MPLOYERS Wdrir Udmir Qldh
EOWPYENT QUANTITY
BEAM MACHINES 2 2 4
SOLAR ARRAY DEPLOYEW 4 4 8
*BARRAY DEPLOYERS 1 1 2
CRANEIIYANICULA~ 31 a 46
#)(IIER BUS OEKOYERS 3 2 4
CREW SIZE
MANAGEMENT
~ U C T I O W
10
362
10
219
10
3l4 I
BASE OPERATIOIB 39 je 39
BASESUP~RT -
n - 67 -
n
a S 440
L
F&m 1.2.1- 1 Sm'es Cansauction Options
APQROXIMATE
RANGE OF
COST
A
ADVANTAGES:
MUDVAMTAGES:
IDLE EOWPUENT
OlSRUPnOW I N PfmwamMCOlQnNUrrV
* W ~ A N D I w u E N l O R Y ~ L ~ ~
D
OUTPUT
--
ROLL-FORM ING
GACiNSFC - AL B.B. 0 DIAG. BRACE
GROUND TEST DEMO
I o CAP SPEED 6.71 FT/NIN 6I
I o (ELECT) S P O ~ E U ) 0,682 H/MIN
D = 1.0 M
L = 1.5 H o Tc= 44 SEC. o ;T 88 SEC.
-
HEATING/FORM I N ~ C O O LN~ o VERT ICAL BRACE
! .O TENSION
I
LF
GDIJSC CWPOSITE8.8.; WIRES
D = 1.18 M
_ --J
o T-, 40 SEC.
L = 1.434 H - ,
,
-uoa
In beam builder operation, dissimilar cap speeds will produce warping (bow) of the
beam with an attendant objectionable eccentricity in beam-column load applications.
Figure 1.2.1-8 shows the relationship between a beam's eccentricity and the difference
in cap exit rate expressed as a cap error. It is assumed that t h e top cap had a higher
rate than the bottom two, whose rates were identical. Note that a deeper beam may
be permitted to have a greater cap error for the same eccentricity ratio (h/l) and that
the increaszd permissable cap error is in direct proportion to beam depth.
,LAFT BAY
1
I (H,)
I
3.06 CAP FORMING RATE
FT./MIN.
2.0-
PROD. RATE
H./HIN,
1.0-
1 L-BATTEN. SPACING
1.434, (H,) :
Cdusicns.
A. 'ihc combined options (A, 8, and C ) a r e shown in Figure 1.2.1-10. Note t h a t a
production r a t e of 5.7-mlrnin for t h e baseline beam configuration can b e readily
achieved with higher values permissible for greater beam depths and longer batten
spacings. Further increases can b e realized when t h e 40 sec "stopN mode is
reduced. However, the latter requires further study of the cooling radiator
subsystem t o establish a rationale for reducing t h e 40 sec time lapse.
C. A further option beyond those discussed exists for increasing production rate. The
"stop" mode in both current designs can be eliminated by performing all bracing
/n/Lm CONSTANT
P r n . RATE
(n,/n~n,)
(~=1.434M~)
CURRENT
I
I I I 1 n
20 40 60 80 100
CAP FOMING RATE I (FT./MIN.)
I
4 f0 D-BEAR DEPTH (n.1
~ G H ~ T I F o w mBRAC
(An BAY)
- I{NG-
(RID MY)
-1
D L 9 MT'L. STO& AFTW RR] BAY i'
16-
PROD. RATE
n./n1n*
12-
8-
4-
&
1 1 I I 1
20 40 _ 60 80 1 100
CAP FOMING RATE FT./W~N.
1.2.1.2.2.1 SirlgkDedtCmstnrtianBbpC-m-
A nquick-lodC' analysis was performed to determint whether a not a module indexing
rate of 10 metersimin was feasible.
ksunpti~1~
o Loads are transmitted between modules by "indexers" which roll on tracks on the
construction base swf ace.
o Grouna rule: module translation speed will be held to the lowest value which is
consistent with the construction rate. This value has been tentatively set at
10 meters/min.
0 GEO construction base mas = 5x1061rg.
a=
r =2
a337*5)
(33.75160
n/s
= 1.65~10'~
2
. x = 667.5 m
800 -
INDEX RATE I FPS (20 M/MIN)
600 -
INDEX
FORCE -
1BS
200-
Therefore t o facilitate t h e comparison of the single deck and end builder concepts,
t h e groundrules for manning a u t o m a t i c beam machines were revised to accommodate
difierent construction applications as shown in Figure 1.2.1 - 15.
Based on previous studies, o n e operator was assigned t o control 8 synchronized fixed
beam machines. This appears t o te a rather conservative e s t i m a t e when compared
with present automated procedures in the lumber industry.
Finally one man was assigned to each mobile beam machine. This follows t h e original
groundrule similar t o a craneitruck on Earth o r a n SPS crane/manipulator, where
t h e operations a r e performed in series: (1) moviirg from location t o location, (2)
positioning, and (3) performing required operations. The reduction in crew staffing
from t ~ rnen
o t o one man was consistent with t h e revised c r e w staffing logic devel-
oped for cherry picker operations.
A m
BEAM
YACHINES
w
16 -
d l
--- GIMBALLED BEAM W I N E S
FABRICATE B E N -
I @ADDTRIPODS
I HANOOFF
-
'-2
.AIM @AMBUILDER
iJ
I
i
- ~-i--~-
I
YOBlLE BEAM MACHINES
FABRICATE BEAM SEGMENT
MomTO NEW LOCAM
i 1 I I I I I I
F ~ 1.2
R 1- 15 Autumtic Baam Mac/,rbn Appikations and Manniirg
of 20 meter pentahedral segmented beams built with preformed struts. These struc-
tural members were made to size with squared off ends for attaching special end
fittings at t h e structural joining site. Hence one cherry picker was required t o hold
t h e beam in positim while a second cherry picker/crane installed t h e required
connecting struts. The present baseline, however, uses a space fabricated continuous
chord beam which can be made with t h e appropriate end fitting be., nodal, butt, lap,
etc.) a t the time of manufacture. The preferred end fitting design and how i t might be
integrated into t h e beam machine operation remains as an area for future study.
Nevertheless, if i t is assumed that the beam is made with nodal end fittings then one
cherry picker a t each end is sufficient for handling and joining, provided it is equipped
with separate grapplers and dexterous manipulators.
As previously noted the staffing of remote work station for cherry pickers and beam
machines was originally baselined as a two-man-crew for reasons of safety and not
necessarily due t o operator work !oad. In the SPS era, i t is believed t h a t a manned
remote work station (MR WS) can b e designed with sufficient redundancy t o meet crew
safety requirements. Therefore crew staffing should only be based on specific work
load requirements. Based on recent M R W S studies, it is believed that SPS construction
tasks (e.g., assembling the joints) can be readily performed by one operator. There-
fore, until further analysis or simulation shows otherwise, only one operatcr can be
justified for each cherry picker or mobile beam machine.
The baseline LEO construction approi for example, follows a two-step segmented
build-up method. This method allows minimal equipment t o be used for structural
assembly while other time consuming subsystem functions, such as installing solar
array blankets, are performed on fully assembled structural bays. The solar array
structural bays a r e constructed with space fabricated tri-beam elements joined at the
corners. Accordingly, t h e construction work zone needs a two bay tacility depth t o
SEGMENTCD BUILD UP CONTINUOUS BUl ill UP
(BASEL1NE) , (OPTION)
The segmented and continuous construction methods discussed above lead toward
generic iamilies of external construction bases shown in Figure 1.2.1-17. That is,
segmented construction c a n be implemented by t h e baseline double deck and i t s
derivatives encompassing single deck options and smaller size base arrangements.
Derivatives of the continuous end builder can also vary in size o r include added fea-
tures if needed t o facilitate SPS construction.
The alternative construction concepts that were examined in this study a r e shown
in Figure 1.2.1- 18.
..
IFD SBJFIIITF -
. . ..
O~TW
o BOOTSTRAP CONSTR,
o Sh4LL PiDULE
o SIMPLE ANTWNA
\
r
\
GEO W G U DECK
I
2 DECKS
LEO SINGLE DEC
6WSW
I
LEO
10GWsPS
10GWsp8' I
I
INTERNAL BASE
The baseline 8 X 16 bay SPS can be constructed by usirg either 8 bay wide, 4 bay wide,
or 2 bay wide ccnstruction bases. The large 8 bay wide end builder constructs the
satellite on a single pass. It can install the antenna at the beginning or the end of
power collection module construction. The other bases r e q i r e 2 or more passes t o
complete the satellite and can phase the antenna installation t o coincide with either
the mid point or completicn of power collection module construction. The 8 bay wide
and 2 bay wide options encompass the lowest and highest levels of production activity
to meet t h e 6 month build cycle.
The two remaining options address alternate SPS designs which favor single pass
production buildup for t h e 4 bay wide option. The LEO constructed modules also
require that the antenna b e installed normal tc the direction of construction.
- --l
ALTERNATE SPS
AR.8 . 0
J
4 BAY WlOL EASE
-------- -------
LEO S S Y O O U L E 1AR - 2)
0 0
2 BAY WIDE
4 BAY *IN
:*i, 05:"
As shown, the assenlbly process begins with s t e p I, when t h e first k a m e is built ilp on
t h e longitudinal members. The structural members of the f r a m e can be fabricated
by s e p a r a t e beam machines; located next t o e a c h longitudinal member ur with mobile
beam machines t h a t travel f r o m o n e position t o t h e next. The upper and lower horizontal
beams a r e fabricated in parallel and then positioned fcr assembly. As these members
are being joii,ed, t h e k a m machines a r e pivoted and t h e o t h e r members of t h e t i a m e
are fatr'cated as needed t o complete t h e assembly. Step 2 indexes t h e f r a m e f o r
one bay lengt., by fabricating t h e continuous longitudinal beams from dedicated beam
machines. In Step 3, t h e next f r a m e is buil: as in Step 1. During these t h r e e steps,
power busses and solar array blankets c a n be installed in parallel. If salar array
blankets a r e t o rn deployed in t h e direction of h i l d , they are fed out as t h e s t r i c t u r e
indexes. If t h e y a r e iaterally strung, then t h e structure is indexed incremental!^
and blanrcets strung across t h e structure, from t h e base, at e a c h increment. Longitudinal
busses a r e Installed "on t h e fly" as t h e structure i s indexed; lateral busses are installed
before a bay is indexed.
Step iC fills in 'rhe bay structure kith diagonal beams t o complete t h a t structure.
This bay- is then iqdexed, as in Step 2, anll t h e whole procc-ss repeated until t h e solar
array s t f l x i u r e is built.
=ALL-RWSb30UR
ARRAY MANKEIS I N PARALLEL
m s JSJV
F@.P 1.2-1-10 T y m End &riMar Assanibly Sequerrae
m o a FInImG
- C O W T ~ ~
VWRmRRUmO
- OlnRBA1YSI*N
m1NT AT INTERSECT.
OF cEUTRO(06
d
- FOLDFOR LAU(rCn
BUTT FITTlNG
- ALLBEAMS
lTERRVPTEO
- ALLBfAMSFlXED
ENDS (ASSV?)
- FIB.ORAfgYsLE
FI?TING I N SPACE
- CONTINWUSBEAMUN
aw -
INTEAOWTED
OTHER BEAMS ATTACH
TOP&IWERV
(ECCENTRICITIES?)
SPACE
The third option shown S a w e frame that does not interrupt the contiamus beam
caps. It replaces one set of lateral posts. Lateral and diagonal beams attacb to points
on the periphery af the frame. These a t t a c h m e ~ t smay be either p i d or fixed
joints. The joints are located so t h a t the end load in each beam is aligned with t h e
centroid of the rontinuous beam. Eccentricities or rriralignments of the beams w d
result in torsion ir. the continuous beans. This f r i m e would also be space fabricated,
or gromd fabricated in piices and space assembled.
.*->air
F@m 1.2.1-22 Longimda Barn, F-tim RequAar
- Design for construction continuity in the event of a beam builder failure.
Emphasis shall be placed on reliability of subsystem machinery including redun-
dant operating modes, where possible, to avoid beam hiider shutdown. In
addition, consideratian should be given to subsystem designs t h a t limit repair time
t o approxirnateiy 60 minutes, while the shutdown beam builder tracks along at the
s a m e r a t e as t h e indexing structure. Holding fixtures to facilitate on-lineloff-
line maintenancn and repair shall also be considered.
It should be noted that t h e above requirements for limitation of accelerations and for
synchronization apply t o any base assembly function where simultaneity of operation is
critical, including t h e bse of multi-indcxers driving simultaneously t o propel either the
base (in the mi!-builder constructim appkchj or t o propel the satellite (in the singie-
& ~ kconsuuctiat approach). Far all such functions, centralized control is necessary
t o limit locomotion forces t o acceptable values.
t-
t ain 2 uo
r r r u ~ mw
wcn'
TOLER A X t M LO
IN 7.5 m BU
It should be noted t h a t the effects of thermal gradients in the construction base,
which are a necessary consideration in this kind of analyses, have not been included.
Three options a r e presented in this figure for relieving t h e beam machines of this
function. Option 1 adds on-line indexing mechanisms t o the process of fabricating
the longitudinal beams. These synchronized mechanisms are dedicated to indexing
t h e beams and t o reacting disturbance end loads similar t o the indexers used on t h e
single deck baseline. Shears a r e still reacted by t h e leg supports. Option 2 ad&
a leg t o the top of the L t o make a C section base. Thus, the structure has supports
on two opposite faces which react alI disturbance loads and index the structure.
The third optim extends that leg of the base which mounts the supports, Additional
supports a r e provided on the extension at one bay distant from the originals. These
two sets of supports react all disturbance loads and index the structure.
-
OPTION 3 EXTENOED OUTRIGGERS
DECOU?LE BEAM W H I N E FROY
WmiXIMG ISUPMRT FUllCTIOClE
n s > 05%"
ALTERNATE
-.
--..--.-
PROGRESSIVE
(15OR30m SERIES
STEPS) COMPLETE STRUCTURE
FIRST
PREFE'RRED
1.
SVNCHRONIZED
UNIDIRECTIONAL
FRAME-TO-FRAME
FRAME.TO.FRAME
Further discussion of solar array blanket installation requirements and the compara-
tive assenlbly methods a r e provided below.
Solar Array/Str~~cttm Assembly Cornparism (128 Bays). 'The four assembly met hods
(progressive, series, synchronized, and unidirectional) are compared in Figure 1.2.1-28
in terms of their structural fabrication methods, blanket installation direction,
required deployment rates, solar array installation equipments, construction base
impact and related satellite impact.
Approximately 148 days a r e available for constructing the power collection module,
within t h e specified six months, when yoke assembly, antenna/yoke mating and final
test and check act a r e considered. The required rates for fabricating the longitudinal
a s O U A ARRAVS
MhOVED
VlmKWJ
*I:T&%N*L
BEAM FAU).
WA DEPLOY
CARRIAGE
4 SOLAR ARRAVS
OEPLOVED
WITH lNCREMENTAL
LONGITUOINAL
BEAM F U R .
Figure 1.2. 1-26 End Builder Fram AsemWyISdar Amy Deployment /Coupled Operations)
a
b- 56 MIN 4
1
DEPLOY L.E.
AIACH/CONNECT L.E. a
BASLINE ALTERNATE ALIGNED
' M T SIA DEPLOY SLOW WA DEPLOY
EXTENDED BLANKE 1 DECELERATION AND AND
Q em MULTI.8USES
'O 14196 kp (15 m r 660m) SEGMENT n
DEPLOY
RATE.
mpm
2 4 6
MINIMUM STOPPING DISTANCE, m
nY3QYll
a17 8 12.5
F RAN ,TO.FRANE
LATERAL
( 1 DEAN a U A t
0.I) 4 5.8
FRIIII.TO.FRAME
ALIONEO
(1. 8EAu 8 UU
112 4 112
0.18 8 12.3 a54 1 a51
2 . 1 WE
~ ~ UTES au 4 n.4 1.41 4 1.41
SA.INSTALL. EQUIP.
C O ~ BASE
R
I IWALLERSI
OEnOVEA
F g p T TRACK
II INSTALLERS. o E n o v E R
8 CROSS M V CANTRV
Y l 1SUWORT A I I
I ~ALLERSI
OELOVERS
TRACK
I YISIALLE(lE
/ WACT
tn5.01av
REAR DECK
FWD PAS5
The basic a ~ f f e r e r c e sin overall operations of t h e three bases shown lie in the number
of "passes" required t o construct t h e satellite. The 2-bay base builds t h e satellite in
four passe>; the 4-bay base in two; and the 8-bay base in one. Described below are the
major syztem characteristics for each of the three bases.
The following paragraphs discuss the 2-bay base satellite construction .sequence, base
arragnemenr, construction systim, satellite construction apk oach and antenna/yoke
mdting. M G of~ these characteristics apply as well t o the 4-bay and 8-bay end
builders.
-
2 Bay Construction Sequence- The 2-bay base uses loqgitudinal aqd lateral indexing
rails o canstruct the 8 % 16 bay satellite in 4 successive passes. After completing t h e
first 2-hay wide strip, the base is indexed laterally (=-bay.;! as shown in Figure
1.2.1-32 and then longitudinally (16 bays) t o begin, a t that paint, the second pass.
Note that the antenna is c o n s t r u n d in parallel. This procedure is rcpeated until
the power generation and di .tribution system structure ana subsystems a r e completed.
At the en6 of the ?t+ pass, the antenna, yoke, etc., a r e also completed. The base
is then indexed !ater&l:y t o a fxxition with the antenna on satellite centerline. Mating
opcratia~sa r e ?hen begun to trafis' r ihe antenna mass from the construction base t o
Fc;Oom 1.2-1.30 End Buddw Con#luction Ba~as
-- mum
-w
CoMSTR. svs
--*mJT00ST (l9773)
-
-
- .
\
\
SiCLaWaH
-US
.\ STRUC1:IRE --
U7aZ.fJ61.ZIla,
2.401 1
8k,
SD l@tp
b74r
s.: - .,CREW
TOTAL
TOTAL
W E
. ..,
%
**
ARRAY W O V L E COUSTR. EOUIC.
-- CRA#SIC.P. -- 9
--
. BEAMYACHlNS
11
- .-. - INOCXERS 5
- BUSOECLOVERS
- SOIM I)LMlKET M R O V E R S - 1
0
S A l X L L l t r DESIGN
-
-
SOLAR ARRAV ORlEMTA
LOlWlTUMNAL MANS
m a
- LONGITUDINAL
comflWU005
C-C:
- -_;
I
I
i
..
2 1 ?@ M V S
WDEXLAT.
Go8AcaToGO
WILD ZIY)
2M*m=
STRW
m 7 ~ E l Om
16UVS
mlZWn*COYLcTf
IWIWX U T .
U&W-UYfEIYI
mDLX LAT. TO
CLEAR LUITE)W
IST ?A!s
INDEXER SWMRTS
the satellite. When t h e antenna is completely mated the base is then indexed away
f rorn, and clear of, the antenna.
While &fined as a 2-bay base, its width !2050m) ercompasses a 3-bay segment of t h e
power collector structure to provide a one bay ov. -.?p for la+waland longitudinal
indexing operations. The 760m high base, built in the form of an open truss
w L -~ shaped framework, is sufficient t o house necessary equipment and machinery t o
construct the power collector module. The an:enna construction site is located a t the
rear of the base, making the total base length 3370m although only approximately
800m is required for power collection 1 1 ~ d u lm
e t r u c t i o n . A short platform extends
into the antenna work area to facilitate rotary-joint assembly which is described
further below.
The two views shown represent what is probably :he most active location in the base.
The 12.7 m beam machines gimbals 180 degrees t o provide the required S/A suppart
beams, while nearby a mobile (track mounted) 7.5 m beam machine IS shown at i t s mid
point of travel between one end of the base and the other. In addition, t h e 7.5 m
l o n g i t u a i ~ bearc
l machine, bus installer and solar array implacement equipments a r e
shown.
g4: ;
'I.
, -
-.
11.
li I
:I ' ,
--
!'
,
*;! .,
I
!!
/:
:;
-
U T E W L =A SUICORT
(12.7 nl
it_---- ..
.---
W A L L VERTICAL
ft.6 m)
with the fwwara ...,'*d;nal-diago~l
(7.5 m) being installed befare t h e lateral S/A
support beam ' I ' 7 mr t o facilitate cherrypicker accessibility and mobility in t h e end-
attachme-: $-xe-s. The 12.7 m beam machine shuttles up and down on a short length
of track t o preclude interference with t h e beam machine producing t h e vertical beam
elements. The beam elements in the plane of each f t a m e (verticals, lateral diagonals,
anti lower-transverse elements) are installed last and compiete the structural buildup
of each bay.
Subsequent t o the installation of a 12.7 m solar array stpport beam, the cherrypicker
rercoves a S/A box f tom t h e supply a i b shown and fastens it t o the proximal anchor.
The distal-end of the blanket is tken connected t o t h e beam, When the frame has been
inaexed one bay away, the blankets a r e fully deployed and t h e box is removed from i t s
anchor support fittings and fastened t o the next 12.7 m support beam t o ccmplete t h e
cycle.
INSTALL
3 CONNECTSIA
mOXlMAL END
.-5> CUV
ANTENNA
INDEXERS 3
COMPLETF
ATTACHMENT O F
INTERF STRUCT.
.& ....- - - - . . - INDEX BASE
4 , : '
COMMENCE INDEX BASE
SUILD INTERF S T R N T . ATT4CHIENT OF CLEAR OF
YOKE TO h 3 T JOINT INTERF. S T R U f 1 ANTENNA
BLlILO6 JURY RIG
8 8 16 BAYS C0MPi.E TE
IRTERF STRUCT
BASE INPCS!T40N ROT. JOINT TO.SiA
ANTENNA. YOKE.
a ROTARY JOINT
COMrLET E
I*,., XI. L,
37h
SATELLITE OLSKZW
-SOUR ARRAY ORIENTATWU = LaUOINDM)M
-LoNGlTUOlWL~ = COnnnwuS
This 4-bay base operates very similarly t o thc. smaller 2-bay wide e n d builder. Solar
arrays are deployed in t h e direction of build. The antenna construction platform
conforms to t h e baseline in area but includes a yoke construction facility. This base
m a t e s t h e antenna t o t h e solar array in t h e PI cferred location with the antenna aligned
with t h e longitudinal centerline of the solar array.
The antenna and yoke construction platform is mounted at a distance from t h e solar
array facility t o provide a n a r e a in w h ~ c ht h e rotary joint and mating structure c a n
be built. It is also located so t h a t during second pass construction, the first pass
solar array structure does not foul t h e antenna under construction. When t h e antenna
and yoke have been built, they a r e assembled t o t h e rotary joint. The mating structure
t o t h e solar array i s then built, but not completed at its solar array end. This e n t i r e
assembly is then indexed along t h e backface of t h e solar array facility until one
set of legs of t h e mating s t r u c t u r e is at t h e mating overhang for structural completion
of those legs and mating t o t h e solar array. The base is now indexed outboard s o
t h a t t h e c e n t e r mating legs c a n b e completed and a t t a c h e d in t h e mating overhang.
This sequence of indexing and mating is repeated t o complete t h e mating of t h e
solar array and antenna assemblies. Indexing of t h e base, laterally across t h e solar
array, is continu2d until t h e base is separated from t h e satellite.
280
st1ucture. Other differences include t h e necessity for !S dedicated longitudinal beam
bui!ders, and added cherry pickers t o fully o p e r a t e this wider facility.
o -
Gravity Gradient Condition, Natural FrequencylMode Figure 1.2.1-42 shows t h e
configuration evaluated f o r gravity gradicn: induced loads; the solar a r r a y is 4 by
16 bays, t h e construction base is i n posirion at t h e antenna end and t h e microwave
antenna fully constructed is located in the a f t position of t h e base. Mass o.: r .. and
orbital orientation are as shown in t h e figure. A worst case gravity gradient
torque was assumed with 0 = 45' and = 0 was assumed.
The frequency for t h e selected configuration shown in the previous figure was
calculated using t.k given mass data. The stiffness data s3own in Figure 1.2.1-44
was calculated for t h e Boeing selected composite c a p member with an area of
.
8.065 x iO- 4m 2 and a modulus of elasticity of 1.378 x 1011N/m 2 The array was
assumed attached t o the base at t h e indicated locations; t h e total antenna mass
was located a t its center of gravity.
MASS DATA
CONSTRUCTION
BASE 5 x los ke
lRSOUR
ARRAY 18.75 x 108 km
iiSS19MY
MW
ANTENNA
f
10692
- -m
I "
470 m
f t SOLAR
\ ARRAY
-
-
r
T
---
CONTROL THRUSTER
FORCES
Fiulure 1.2.143 5GW SpS End-Buildw Construction Besc, Gravity Gred2nt Condition
FREOUENCY = -0031Hz COhISTRUCTtON
DEFLECTION 6
1-- --<--
LENGTH METERS x 103
SOLAR ARRAY
DATA : AlTACHYLNT
POINTS
CONSTRUCTION BASE 5 x 106 kg CONSTRUCTIONBASE
SDLAR ARRAY lam x 1061~9 TO SOLAR ARRAY
MU ANTENNA 12-5 x 106 kg
STIFFNESS DATA :
ElARRAy 1-48x 1 0 ~ ~ l U m ~
ElCOIIIST. 5 x lo1* k2
7
--
H168Pm
INDEXING RATE:
10 M/M TO 20 M/M
MASS DATA:
SOLAR ARRAY 28 x la6 kg
ANTENNA 125 x 1@ kg
BASE 5 x 106 kg
CONSTRUCTION
BASE 2
I
2955039V
By orienting the solar array deployment longitudinaliy for the 4-bay end builder
(similar t o the oriental ion of the 2-bay end builder), i t was possible to (1) delete
the solary array deployer, (2) lower the solar array deployment rate from 12.5
mpm t o 1 mpm, and (3) shorten the overall construction time.
-
2 Bay and 4 Bay End Builder Timelines The end builder timelines a r e provided
in Figure 1.2.1-48. As shown therein, the SPS assembly operations commence with
TlME TO ATTAIN INDEX RATE - SH:
ASSUIIPTIONS:
COlYSTRUCTiON BASE MAS IS A SMALL PERCENT OF TOTAL
MASS; RELATIVE MOTIONOF SATELLITE IS NEGLECTED
FACTOR MlDTEftU -
FINAL WACT
$1' RATE
SEGaEAM MACH. FA&
RATE SlrPAI 5- J
2) MATE ANTENNA MlDPOlNTOCS FINAL OQS AVOIDS EARLY START UP
a. REINDEX RATE -
REV. G R W . RULE REWCL TIME
%( STAUCT. ASSV.
SUB SYS. ASSr
1mpn
OEDlCATED CR (48AY)
UNDEFINED
10 nrpCn
W I L E CPI
4 DEDICATED
R E W E Ch E
ADDS & 6 CREW
CR
Y A ORIENTATION &G (2 6 4 M Y ) REDUCE EOUIP.. L W E R S/A
LEPLOY. RATE. W I T C W I . T Y E
REMOTE WORK
2 MEWCAB 1MAN a CHANGE-
OVER
AUTO. BEAM MACHINES 2 MENNACI(INE 8 FIXED OR MIN. STAFFING REOT.
4 GIWMED
OR 1 YOBILE/YAN
l U T E Y T C * U TO V O I E El
FmAL 1EST a. 2 U Y
1.1
------- _ _ - ____ _ _ ------
IOCluMrS
-___ __--_------_-----
I
-E~~~~zI)QKLK~.IQ)(K~
-n- - - - - - - - - - - - n- -I
~LOWGIDCX~.SYCI~~I
eQQ-%2+1- - - oJ- - -
nEmoEX..sE c 0
1.1
VQLE Ilo:
AssF.mLE lSllEWYI
C
L- L 140 I 3
MATE ANTE- TO W E r--r---fl l
F W TEST 4 CO. 484V I----
A A
YK 157.1 llDb D I V S
the construction of t h e power collection module. For t h e 2-bay end builder, i t is
constructed in four passes through the construction base. Each pass provides a
2-bay by 16-bay submodule and reinde:d.ing occurs between passes. The first and
fourth construction passes include t h e installation of thrusters. The centerline
main bus is installed during the second pass, but that does not e f f e c t the timeline
since i t is installed in parallel with t h e fabrication of the longitudinal beams. The
second, third and fourth construction pass a r e shorter than the first pass, because
one side of the modules a r e common with the structure previously assembled and
therefore two fewer beams a r e required. The fourth pass would require the same
construction time as t h e second and third, except for the addition of the
thrusters. After completion of the fourth pass, the yoke is assembled and then
mated t o t h e antenna, which was begun during t h e assembly of the first 2 x 16
power collection submodule. Allowing additional time for checkout, the total
2-bay end builder constructicn time is 184 days.
The 4-bay end builder operates identically t o the 2-bay end b:rilder, as shown in
Figure 1.2.1-48. There are, of course, only two passes through the construction
base, each pass providing a 4-bay by 16-bay submodule with only one reindexing
pass. When the longitudinal beam fabrication occurs at 0.5 mpm, t h e total 4 bay
end builder construction time is 180.5 days. However, if the longitudinal beam
fabricat.cr process is accelerated t o one mpm, then the total construction time
can be reduced t o i57.1 days.
Upon completion of the end frame assembly and solar array attachment, the
structure is indexed longitudinally. Meanwhile, the fixed beam machines fabri-
catt t h e 667 meter longitudinal beams, the main bus is deployed (on t h e second
pass) and the solar array panels are also being deployed.
287
ATTACH 90UA ARRAYS
/'
I
UPROACH
DEPLOY MAIN BUS
INOEX
1
1
FAB.& 4SSEM. SEG. BE-
}*'$-FZG-{
I DEPLOY MAINBUS I
nss-iv
The cost penalty for adding tnese crews and equiprnents is also shown in Figure
1.2.1-52. This cost penalty reflects t h e added costs f o r cherry pickers, crew
modules, crew operations, and related transportation costs. The interest saved by
LONG
BEAM
FA5
RATE
d I N
20-
-
8 BAY WIDE ADDED COST
16 - LESS INTEREST SAVED
4 BAY WlDE
-
12 -
-
2 BAY WlDE
8- 4 BAY WlDE
TOTAL
CHERRY -
PICKERS
4 -
SIA & STRUCT ASSY
- LONG BEAM FAB 1 rnpm
0 ~h , , ,
0 100 140 180 220 0 80 160 240
SPS CONSTRUCTION TIME - DAYS COST - $M
I-
80 40 0
DAYS LESS 180
2955-027V
F~W 1.2.1-52 EndBuiIdav Goduction W e Up Potsntid
rrnd a M Berry Pickm
Added C ~ s m
adding these additional equipments is also shown for e a c h end builder in terms of
t h e added cost. less t h e interest saved.
This section presents t h e details of two versions of a single deck construction base
concept: 1) a LEO single deck construction base, where 8 modules a r e constructed,
and 2) a CEO single deck construction base, where a 5 CW monolithic SPS would be
constructed. As these concepts utilize t h e construction techniques (frame assembly,
solar array deployment, bus deployment) developed for t h e 2-deck coi~structionbase,
t h e description of t h e single deck concepts does not describe these opc-rations. I t is
expected t h a t t h e reader can r e f e r t o previous contract documentatiorl for details.
/
/ 4 BAY @ 3.6miMIN
4 M Y @ 3.5 mlMlN
+ 3 CHERRY PICKERS
- 360 DAYS
YEARS
-am1
RETURN CREWS
TO LEO USING
.
ROTATE ANTENNAS *TO
POSITION
FINAL C i f E C K W f &
, LOZ/CH2 O N COMMISSIONiNG
MAINTENANCE 6AS
mis facility concept is identical in plan form t o t h e baseline C-clamp facility. The
"roof" and "back wall" have been eliminated. The indexers halre been moved t o a louer
deck. The construction gantry, shown in more detail in Figure 1.2.1-61 and -62,
providcs the capability of ~ i a c i n gbeam machines and cranes where they can be used t o
assemble the upper surface of the modules and yokes.
Ihe gantry is mounted on a carriage that operates from a dedicated track system. The
gdntrg can rotiite 360' about the vertical axis of the carriage. The cantilevered arm
of the gantry spans one SPS bay. The gantry iocation provides enough clearance at the
base so that t h e f acility-mounted beam machines and cranes have working room.
Figure 1.2.1-63 slmws snapshots of the gantry locations and orientations during
different phases of t h e construction operations.
The detailed module construction operations (frame assembly, solar array deployment,
power bus installation, thruster installation) and timeline (40days per module) a r e
identical t o those described for the baseline. The frame assembly and solar array
deployment operations are shcwn in Figure 1.2.1 64. -
The detailed yoke construction aperations (frame assembly, power bus installation,
rotary joint installation, elevation joint installation) a r e identical t o those described
for the baseline.
ANTENNA AhiTENWA
~ r i a r : 4&51.1.
~
'400
DL-la1 DAYS
0 100 200 400 500 800
a L L
~MoDuLEl
wo
--
3
=--= hlOO-TO-MO0DOCKING
AN0 SOLAR ARRAY
DEPLOYMENT
YOKE 10
ANT 1
YOKE 20
MGAYSFOR
MODULE CONST.
f l FINAL
. --
INTEG
'0
n
6W DAYS
Figure 1.2.ti57 ~ o t o m l & Ssi&lite-L EO Construction Tineline
Nhml
ow-
VEHICLE CREW
OPS
CENTER
Figwc1.21-58 LEOSh&?DsdrWBaae
bcm
SOCAfl
ARRAY
298
- 0 180-7-2
-
TOTAL TRACK -
LElYGnt 3151sM
r S O L A R ARRAY OEPLOYER
SIM VIEW -
FRONT VIEW
F ~ 1 . 2 1 4 2 -Gr.yaonJsLv*Lien
*
LATERAL BEAMS LONGITUDINAL BEAMS
GANTRY
YOKE ASSEMBLY
GANTRY ORIENTATIONS
1-2.13.1.3 Facility
Framework. The facility framework dlff ers f rorn that described for the baseline in the
following respect:
Base W y s t e m s . Identical t o baseline except that the thruster system has been
relocated.
Cargo Handling/Di.trikrtion. This system differs from that described for the baseline
in the following respects:
-Delete all tracks associated with top deck and back wall
-Add dedicated gantry tracks
-Add one crew bus (dedicated t o the gantry)
The single-deck is s i m i l ~ tr o that shown for the LEO single-deck facility described in
t h e previous section. The most notable differences a r e t h e orientation of the solar
array deployers and the location of the antenna facility. The construction gantry is
identical to t h a t described previously.
The solar =ray deployers have been reoriented so that the solar array is deployed in
the lateral direction. It is necessary t o put 2 deployers in each of t h e two solar array
deployment facility bays in order t o keep the deployment rates comparable t o the
baseline rates.
To alleviate competition for track by the four solar array deployers and the indexers,
i t was necessary to c r e a t e dedicated tracks for the deployers. Figure 1.2.1- 70 shows
the location of these dedicated tracks. Figure 1.2.1-71 shows how the deployers,
indexers, and facility tracks a r e interfaced.
The detailed module construction operations (frame assembly, solar array deployment,
power bus installation, thruster installation) are identical to those described for the
baseline, see Figure 1.2.1-72. The module construction timeline is shown in Figure
1.2.1-73.
1- WIDE SOLAR
ARRAY BLANKETS
TRANSFER
SATELLITE
COhlPONENTS
BASE WITH
EOTV'S
GEO BASE
DELIVER C R E ~ CONSTRUCT MONOLITHIC
AND SUPPLIES SATELLITE
WlTH LOZILHZ OTV
TO LEO BASE
CONSTRUCT EOTV'S
PERFORM STAGIPiG
DEPOT FUNCTIONS
EOTV'S AND O N RETURN CREWS
DELIVER CREW AlJD REUSABLE
& CARGO TO EQUIPMENT TO
LEO WITH EARTH
HLLV
1 - 167
ASSEMBLE ANTENNA
J
25
ASSEMBLE YOKE
u
FINAL TEST
6 GW MONOLITHIC SPS AND C/O
ANTENNA
CONSTRUCTION
POWER COLLECTION CONSTRUCTION FPClLITY
SYSTEM CONSTRUCTION GANTRY
305
A-A
29670111TRACK
LEVEL A
2100 GANTRY
3177Qn
FACILlN TRACK B
Fi~urtv7.2 7-77 GEO Single Deck Constmction Basc lndexerfikr A m y Deployat Int t h c e
0 0 0
FIRST ROW OF BAYS CCY)LE'lED FRAME INDEXED LONOITUDINALLY LATLRALLY INDEX FRAME 2 M Y &
@ M Y S WIDE) ONE M Y DEPLOY S O U R ARRAY IN BAYS b
SOLAR ARRAY DEPLOYED I N BAYS 7 AND 8 AND 6
ASSEMBLY FRAMES FOR M Y S O AND 10 b3SEMILE FRAME M Y 8 1: AND 12
(SECOtiD ROW OF BAWl
1
01-
6 10 16 n m .
A S F H B L L THRVSlE R SINTORT
~ R U C T U S E1 l r l S l A TIIRVITI~
~~
SVSTEM i PARK ASSV A 1 SLOE OF
FL C l L l T l A
.. U T - Ul'ElUII N O E X I N G U M E W C R
A S E M I L L fRAME;MYS 1h 2
1ltSfA:L AJMIER I U S ON M Y 1 0
r a ; ~ u s r E n *rsv ro MY 1
r r l ~ c m 0
LJSFMBL~ f n L Y L S e A Y S J b 4
now 1
IPUTA' ' Y ' I N OW A S Y O* M Y 4 ' ( M E FOR llQI W
IW O A V l '
ux. AMfS~AYS5L8 dUT
A S MES M V S 7 L 8 0
lf8S -!lPf R BUS M Y 8 0
IWEMQI E r b + n u s r t ns u r w n r
S~~UCIUR[O:IS~ALL I~IRU$TE~
S V S l M / A l l A C l 4 10 B1)V #
9 I -I
LIS()(IBLf * R I M E S BAY I b 10
INSTALL h h l P t R BUS M Y # 0
OEILOY W L A I ARRAY 11111 b @ 0
CJIEMBLf F R M t M V S 1 1 b 1 1 0
ROW 1
O L I L O Y SOLAR ARRAY I 4 1 8 I b 8 D I~CICALOF mow a
I W T A L L M A I N BW OAV 1 1 d'" TMRU 961
n.6 DAYS ma rorn
ASSCYBLC f R A M E S M V 8 13.94 d '"
O L I l O Y SOLAR ARRAY M ' l S 16 I 0
W M l L E f RAM# OAYS 1w b 18 a
Figure 1.2.1-73 Power Cdlection Module Construction Timeline 15GW Monolithic SPS)
309
The antenna construction operations are identical to the baseline. There are 167 days
available t o construct t h e antenna (compared t o 120 days fw the baseline) so the
equipment rates could be 40% slower.
The yoke construction and mating operations are described in Figure 1,2*1-74.
1.21.5.5 Facility
Fmmcwak-The framework for the fazility differs from L ! t desaibed for the
baseline in the following respects:
-Use 10 rn beams
-Delete top deck
-Delete back wall
-Delete 200 m truss along back edge
-Add f rarnework for dedicated gantry and solar array deployer tracks.
Shystem-The base subsystems are identical t o the baseline except that the
thruster system has been relocated.
Cargo Handling and Distributiar- This system differs from the baselire in the follow-
ing respects:
-Delete all tracks associated with top deck and back wall.
-A& dedicated gantry and solar array deployment machine tracks.
-Move the cargo receiving, sorting, and warehousing track system t o the antenna
platf orrn (no net change).
-Add one crew bus (dedicated to :be gantr ).
I I
.
I
<)
MATE YOKE
TO ANTENtU
The construction issues that were to be defined fathe new primary f r a m e configura-
tion a r e t h e following: (1) How many beam machines a r e required? (2) How ma2y
cherrypickers are required? (3) What is t h e assembly sequence? (4) How long does i t
take?
Construction issues (1) through (4)are resolved in Figure 1.2.1-77. This figure shows
how two beam miichines and four cherrypicken can assemble a primary frame bay in
about 4 hours per bay.
Base Management
Construction
Management
Module Constructim
Antenna Construction
Subassembly
Maintenance
Logistics
TestlQC
Base Operations
Base Support
Total
147 1cu
TYPICAL BAY
CP2
MOVES
ALONG
THIS
SEQUENCE FOR BM t
ASSEMBLY TIME
- 104m
w n -
+ 20min 40.8 min
147m
I.' 20 min - 49.4 min
-5n
i mln
-
3W.8 min) + 49.4 min 172 m i n k
-
BM1
-
--
.-.ALLOW 3 h* 4n w
BM BEAM MACHINE 0.76
CP CHERRY PICKER
Figure 1.2.1-77 MPTS Primary Frame Assembly Equipment, Sequence, and Timeline
315
1.2.1.6.1.2 Elcvatim ;kint Lacatim
One of the most significant inconsistencies betweer. the Part HI antenna configuration
concept and t h e antenna construction concept is the orientation of t h e A-f rame
primary structure with respect t o t h e yoke, see Figure 1.2.1-78. In the antenna
configuration concept, the A-frame structure is oriented so t h a t t h e elevation joints
tie into the ends of the center A-frames. However, in the construction concept it is
necessary for the A-frames t o be oriented orthogonally to t h a t just described. This is
necessary so that toward t h e end of the antenna construction timeline the antenna can
be indexed while i t is being constructed within the confines of t h e yoke, see Figure
1.2.1-79.
The first issue is illustrated in Figure 1.2.1-80. The structures analysts indicate t h a t
t h e elevation joint could be made into t h e side of t h e A-frames if some provision could
be made to take out the axial loads on t h e upper surface without relying on the
secondary structure. This would require beams t o bridge between the peaks of the
A-frames as shown in the figure. These beams wodd complicate the maintenance
gantry clearance and would create a problem for the cherrypickers on the maintenance '
gantries to gain access t o the subarrays located directly above t h e beams. Time does
not permit these problems to be resolved. In view of the f a c t that the antenna
primary structure may be changed to a pentrahedral design dut.ng Phase 11, i t does not
seem to be critical t o resolve the issues defined above.
I----- d
i
ANTENNA
1FACILITY
3 17
Fiwm 1.2.1-80 Antenna Elevation Jomt S Inle&ce
3. Determine t h e construction operations, equipment and manpower required f o r
assembly and installation of t h e maintenance system.
4. Develop an integrated timeline for t h e MPTS construction.
5. Develop updated construction equipment and manpower count.
Another factor t h a t became apparent (which has not recognized before) was t h a t
i t is necessary t o support t h e antenna a t t h e perimeter only. After t h e second row
of A-frames a r e partially completed, t h e secondary structures a r e added. These
structures prohibit attaching indexers t o t h e f a c e of t h e antenna. The only place
l e f t :o grab the structurE. is out a t t h e e x t r e m e edges of t h e primary structure.
Subsequently, t h e anrenna platform needs t o be extended on either side of t h e antenna
facility so t h a t the indexers attached t o t h e perimeter have track t o run on. The
resulting configuration of t h e platform is shown in Figure 1.2.1-83. This configuratioq
is a significant change from the reference configuration. The integrated track network
o n this platforrn is shown in Figure 1.2.1-84. Ali of the single deck and end-builder
construction base concepts were updated t o incorporate this new configuration.
SUBARRAYS
AFTER SECONDARY'STRUCTURES
AND SUBARRAYS ARE INSTALLED,
THE ONLY PLACES TO AlTACH THE
INDEXERS ARE AT THE PERIMETER
ALL" INDEXER "SHORT' INDEXER
.
NOTE:
,
THISCONF~GURATIONIS AN UPDATE
FROM THAT SHOWN I N FIG. 13.1-9 I N
THE REFERENCE SYSTEM DESCRIPTION
134 TURNTABLES
22,230 m OF TRACK
SUBASSEMBLY.
PLATFORM
ENVELOPE
a. What beam machines a r e required t o make t h e track beams and t h e gantry beams?
?he timeline analysis t h a t i: discussed later will show t h a t t h e beam machine used
on the lower level (seam Machine 2 in Figure 1.2.1-77) is only used intermittently
for primary f r a m e assembly. It s e e m s reasonsable t o assume t h a t this machine
could b e used t o fabricate both t h e gantry and :he track beams.
The two cherrypickers associated with Beam Machine 2 (CP3 and CP4) a r e also
used intermittently, only 11 hours every other day. It is reasonable t o assume t h a t
these machines could be used t o install t h e gantries, t h e cherrypickers on t h e
gantries, the track beams, the cargo transporters, and t h e crew busses. Remem-
ber t h a t these systems a r e installed only on t h e outboard ends of t h e primary
frame, so t h e r e would be plenty of time availabie t o use C P 3 and C P 4 t o do t h e
necessary preassembly operations on the maintenance systems prior t o
installati~n.
h - t e n n a Construction Timeline
Each of the construction operations have now been examined in detail t o define t h e
times r e q u i ~ed. The integrated t l n e i i n e is shown in Figure 1.2.1-87. The locations of
some of the construction equipment is shown in Figurc 1.2.1-88.
-SECOHDARY STRUCTURE WIRING INSTALLAnOII
WEARRAY TELESCOPING
ELEVATOR INSTALLATION SYSTEM
PHASE CONTROL SYSTEM
IHSTALLEII
SECOIJDARY
STRUCTURE
DE.TELESCOPINC
MACHINE
INSTALUTIOW
MANIPULATOR
REIV HABITAT
WARY STRUCTURE
CARGO AND
CREW TRANSPORTER
TRACK
IN LOADING P S I l I O N
KLYSTRON TUBE
PAL LETS (WSAT.)
ZARGO O N
(STG 2)
A-A
3 24
6
.
?UWARY
FRAME
TOP LEVEL
BOrrOY LEVEL
.
SECONDARY
-RUCT'JRE
DEPLOYMENT&
INSTALL WIRING .
SUBARRAY
MPLOY 1-2 14 1.6 2-2 24 26 28 2-10 3-2 CONTINUES
DEPLOYMENT [ l - l l l I i l l ! l l l l l i l ~ l [ I { j I N THIS
FAsMiOlY
I FOR
BUS DEPLOY i 140DAYS
SUPPORT STRUCTURES
SWITCH GEAR
INTERCONNECT
BUSS INSTALL
U l N T SYSTEMS
=ALL
TRACK .
VEHICLE iNSTALL
8
GANTRY
FACILITY LEVEL C
Y STRUCTURE
SECONDARY STRUCTURE
ZOOM INDEXER
Althougk t h e torai cost difference i not great, t h e 2 Cay end builder f e a t u r e s t h e least
total base c ~ s and
t a !oh- annual anfiorlizat~oncost ~ i t interest
h benefit.
ANTENNA
CO!USTRiJCTlOW
-
TOTAL 10
STAFF
i
(4) I
(10)
STROCTURESI
SUBARRAY POWER DISTI SUBAARAY
PREPARATION MAlNT SYSTEMS DEPLOYMENT
SUW (2)
- .
ASSv SUW (2)
. SUW 0)
TEST 2x2 4
REWORK h2&4
TRANS
t BEAM MACH
OP 2x1~2-4
CHERRYPEKER
OPERATORS
IrlX.2- a
t OEPLOYEROP 1-04
CHERRYPICKEK 09
PHASE CONT 1x1~2 2
PWR OlST 1x1~2 2
-
LOADER
1x202
r GEO CWTAUCTIOO
s w W)KKoT)lr SS
UO DAY COIYSTRUCTH)W TIME
M UlTEWlll m U C T I O U F S l L i T Y
3 28
s.2m 58.078
I ,CREW SUQPORT
/ BASE TRANSPORT
I- LOGISTIC E W l P
/ CONSTR EQUIP
The 2 bay end builder exhibits t h e lowest cost primarily because i t features less
costly c o n s t r u c t i w equipment and related crew modules. The 4 bay end builder
has more equipment but is slightly less costly than t h e single deck baseline
because of its smaller crew size.
' --
10.11
9.28 r---; 9.0, I 920
I 8.63
n[-!lqn
8 I
Se
f
- r
I
I
I
I
I
I
i
#r
1 I I
I
Uw)
iI I
I
I
I I I I
- -
SINGLE $BAY 4BAV 2-Y SINGLE 88AY 48AY 2BAY
DECU DECK
ANNUAL COST WITH INTEREST BENEFIT
---I=,, +
;
I
n
ANNUAL INTEREST SAVEO
W E 10 FASTER C W T R
240 IU
I
I
oL 2 my
SINGLE 80AY 4BAY 2BAV SINGLE PBAY
DECK DECK
1 SINGLE DECK
BASELINE
5GW SPS CONSTR TIME 185 DAYS 181 DAYS 184 DAYS
29554SPV
~igure7.2.1-94 SPS G EO Consttuction Bake Perlbmnce Comparison
33 1
aading additional crews and equipments. By building two 5 GW satellites a year, the 4
bay end builder therefore can offer an 80 day advantage in faster performance over
the single deck.
Comparison of the total base relative masses (see Figure 1.2.1-95) shows that most of
the weight difference is attributed t o the difference in base configuration framework.
As previously noted, t h e weight of base framework listed herein is normalized t o the
extent each base was assumed t o employ 100 meter deep structural sections, rather
than the various deeper and shallower "as drawn" sections which have not been
analyzed and sized.
fewer people than the single deck. However, the single deck requires fewer automatic
beam machines and cherrypickers than the two end builder concepts. It should be
noted, however, that t h e end builder uses some of i t s cherrypickers t o perform solar
array installation functions, using simple proximal anchors from i t s built in logistic
track, in lieu of the large cross bay gantries and related installation/deployment
equipment used by the single deck.
-
0 -
BASELINE
SINGLE DECK
4 BAY
END BUILDER
2 BAY
I
END BUILDER
-FRAMEWORK
FLAT DECK W
UPPER I.EVEL
FIXED UPPERI
LOrYER LEVEL
FIXED UPPERI
LOWER LEVEL
GANTRY STA W R K STA WORK STA
CREW SIZE
LOGISTIC TRACK
I) .07
gOBDOKm
SINGLE DECK
BASELINE
4 BAY 2BAYI
END BUILDER END BUILDER
SATELLITE ASSY MODE ' 16 ROW L A 1ERAL 2 PASS LONG 4 PASS LONG
BUILDUP BUILDUP BltILDUP
Growth in SPS production r a t e require.nents implies added crews and equipments for
t h e single deck. For t h e end builders these added costs can b e deferred until t h e
longitudirlal beam fabrication r a t e capability is reached (i.e., about 3.5 meters/min).
All alternate bases can b e expanded if needed t o huild t h e 8 x 16 bay satellite in one
pass. Each concept cm also buila pentahedral structures or be adapted f o r use in LEO
construction. In addition they can readily build smaller or larger satellites which
require fewer or more bays of t h e s a m e size. Should smaller or larger satellites b e
required with diiferent slze bays a f t e r the base has been built, then t h e single deck
approach is probably easiest t o adapt.
SMALLER SATELLITE
-FEWERBAYS OK OK OK
LARGER SATELLITE
- MORE BAYS OK OK OK
BASE COST
ANNUAL AMORT W
INTEREST BENEFIT
BASE MASS
CREW SIZE
OPERATIONS
COMPLEXITY COUPLED COUPLED
Y A STRUCT Y A STRUCT ASSY
ASSY
PERFORMANCE
CAPABI LlTY ADD EOUlP FOR
FASTER
PROOUCTION
DEVELOPMENT
RISK MEDIUM MEDIUM
.'955.O.IV
However, the single deck appears simpler to operate due to haviqg less construction
equipment. The single deck is probably also easier to adapt to major changes in
satellite design. Therefore, if simple operations are more important than faster
production capability, then the single deck is yrefeired.
0 18&=7-2
1.3 SPACE TRAYSPORTATW
Three study activities were conduced under this WBS item:
Problem Statement
A preliminary study by t h e Los Alamos Scientific Laboratory raised concerns over
depletion of the ionosphere F-layer by exhaust products from HLLV's. This study
assumed t h a t t h e HLLV ascent trajectory will reach t h e F-layer. The reference
trajectory is shown in Figure 1.3.1-1; i t never g e t s closer than 75 km froin t h e bottom
of t h e F-layer. Boeing conducted a preliminary analysis under IR&D t o examine
t h e possibility of deleterious e f f e c t s from this trajectory. It was concluded that:
( 2 ) There does not appear t o b e any concern regarding t h e ozone layer. I t is even
possible t h a t water diffusing downward could reduce ozone depletion by other
r e a c t a n t s such as aerosol propellants.
I IONOSPHERE F-UYER
WINGED BOOSTER
!
HM#)SPHERE
-----!-
MESOSPHERE
\
2m \ 1
100 STRATOSPHERE
STAGE
(GNITION --------
n
-0 1100 2CO 300 4W 500 600 700 600
SEPARATION rJM
Trajectory Analyses
I t was concluded t h a t t h e trajectory modification task should investigate t h e feasibil-
ity of reducing the ascent trajectory maximum altitude to about 100 km. (The earlier
r e t e r e n c e trajectolq- has a maximum a l t i t u d e of about 125 km for injection to a 100 km
x 477 krn transfer orbit.) Three trajectory shaping methods w e r e studied: (1) changing
injection a1ri:ide; (2) using a commanded r a t h e r than optimized upper s t a g e angle
of a t t a c k profile; (3) using a commanded r a t h e r than optimized booster bi;ial tilt.
(All booster trajectories were zero-angle-of-attack gravity turns.) Figure 1.3.1 -1
presents a summary of results. Table 1.3.1 - 1 summarizes key trajectory parameters.
The trajectory plots from which t h e s e points were taken a r e shown in Figures 1.3.1-3
through 1.3.1-16.
ORIGINAL REFERENCE+A
0 0
0 INJECTION AT 9&95 KM
0 INJECTION AT 85 KM
d A INJECTION AT 110-120 KM
I I 1 b
315-
90 100 110 120 130
PEAK ASCENT TRAJECTORY
ALTiTUOE IN KM
0
0 100 200 #O 400 WO 600
RANGE (N. MI.)
Figure 1.3.14 33x477 km explicit fnybol#:tory
Tilt = 82.0
a0
HEIGHT (Km)
HEIGHT
A --- 15.i.QICOOOLBS
-
AW2 SpSe,soOLBS
WMECO 1.920.100LBS
PA 925.88 LBS
O L I I I I I 1 I -
0 100 200 300 400
500 (WO 700 800
RANGE (N. MI.)
4
120
HEIGHT (Km)
TO CIRCULARIZE: AV -- 364.1
I I I I I I I
0'
0 100 200 300 400 500 00 700 800
RANGE (N. MI.)
FQure 1.3.18 Inmion ro 93x477 km Orbit ( R e f m m
800 900
0 100 200 300 400 so0 wo mo
RANGE (N. MI.)
80
HEIGHT (Kml
0 ' I I I I I I I I
0 100 200 300 400 500 600 700 &00 900
RANGE (N. MI.)
HEIGHT (Km)
AW2
WMECO
PA
-- 5.073.380 LBS
1,903,629 LBS
913,691 LBS
60
1 I I I 1 1 I
AW2
R E
- 6,0#1,210~
= 1,lm.rnLBs
= m,omm
I I
0 100 200 m 400 500 mo 700 a0
RANGE (N.MI.)
F 1 - 1 3 1 la177 h (R-I
la,
60
HEIGHT (KmJ
80
0
0 100 200 300 400 m 600 JW 800
RANGE (N. MI.)
- - - - - I )
-
----.------.I)---
-
---.I------
4
--
AW2
W ECO
P I
- 6,082,130 LBS
= 1,894,870LBS
= 000,813LBS
I I I t 1 I t A
-------------------- --------
I
AW2
W ECO
P
--6,067,500 LBS
1,905,000 LBS
= 910.168 LBS
I I I I I I I
The trajectory design entails several considerations. (1) The hypersonic flight portion
should occur at maximum lift coefficient (angle of a t t a c k about 5 5 degrees) t o maxi-
mize flight altitude and thereby minimize heating r a t e s and overpressures; ( 2 ) t h e
turrr should b e flown at maximum LID t o enable turn completion at maximum speed*
and altitude. Maximum LID occurs at an angle of a t t a c k between 2 0 and 25 degrees;
(3) if a sudden transition is made from 55' angle of a t t a c k to 25' angle of a t t a c k
and 45' bank, t h e lift will decrease markedly below weight and a dive will result.
Accordingly, a transition is needed. Smooth transitions w e r e found possible by gradually
reducing angle of a t t a c k from 55 degrees at M=IO t o 20 degrees at M=5; (4) a t t e m p t s
t o control t h e turn by maintaining a normal load f a c t o r of 2 (the "ideal" turn at
45' bank) generally resulted in a progressively increasing angle of a t t a c k until CL
max was exceeded and t h e normal load f a c t o r could not b e maintained. Holding
a constant angle of a t t a c k at 20' t o 25' resulted in good turn control; (5) glideback
shvuld te performed at LID max t o maximize range. Subsonic LID max occurs at
an angle bf a t t a c k of about 10 degrees.
Several trajectory strategies were investigated. There did not appear t o be much
advantage t o turn initiation at speeds g r e a t e r than about Mach 5.
Figure 1.3.1-17 shows t h e command sequence (angles of a t t a c k and bank) found most
effective. Figure 1.3.1-18 shows t h e resulting altitude and dynamic pressure as
a function of Mach number. Figure 1.3.1- 19 shows a map of t h e 180' turn superimposed
on Florida. Figure 1.3.1-20 shows altitude as a function of g r e a t circle range.
a l l l l l l l l l l l l l l l l l l l l l lrll
6EUS UWl 7888 7186 72ml nee
Q W T CIRCLE M CNAUT. XILES)
The results of the cornpalson indicdtc that t h e LEO construct lor1 concept using self-
p w e r transfer of the modules and no recovery should be u t i l ~ z e dfor t h e rriitlal stages
- SlLlCON ,
EON
GEO
CONSTRUCTION/
EOTV
; A
COMPARE
ANDSELECT
- DETAILEO
COSTING
1'
-
-
L
GaAs
EOTV
t
COMPARE
AND RECOMMEND
+
LEO L v z T l
CONSTRUCTION/ POTENTIAL .
SELF-POWER IMPROVEMENTS
L * -
The following material will concentrate on the EOTV definition in terms of perform-
ance, design, operaticns, construction and costing.
1.32.21.1 Guicklhks
The key guidelines used in conductirtg the performance and cost optimization of the
EOTV's a r e shown in Table 1.3.2-1. The annual mass t o be delivered relates t o
satellite(s) capabie of producing 10 CWe ground output. Delivery of all cargo
associated with t h e satellite in 330 days allows sufficient time for final installation
and checkout so that a satellite can come on line at the end of one year. Satellite
mass includes a growth factor of 21%. The total cargr. delivery mass includes not only
components but the containers for the components and the rack t o support the
containers. Five percent cf t h e component mass has been allocated t o both t h e
containers and t h e payload rack. The majority of the containers such as those
associated with t h e solar arrays will be used in t h e actual installation process. Other
TRANSFER
SATE LLITE
COMPOPJENTS
CONSTRUCT MOlYOLlTHlC
SATELLITE
TO LEO BASE
- SETURN E O N
TO LEC BASE
a CONSTRUCT EON'S
PERFORM STAGlrJG
DELIVER CREW
6 CARGO TO
LEO WITH EARTH
HLLV
r ++
UPrnETURN
I
PERFORMANCE I
L
PARAMETERS I
I
I 1, , I
I
i
EOTV
PROPELLANT C3 SIZE
REQUIREMENTS AND
MASS
I
, A I
L---------- ------r
I
t 1
-
rnorELLANT
RETURN
PROfELWT
4 FACTOR RACTlOt , CROC
?Pu
* .02 .02 -02s
SPEC , 1 w +
MASS
TUS TICS
FEED
SYS
U. GAS
RESID
E2 b
:
ELEC
SIZING
UMER 1-11
A RATm
,
THRUSTER
SPEC MASS
5 THRUST
INSTL ,
SPEC MASS
PROPELLANT ?ROPELLAWf
+ SYS FRACTION
* SYSTEM MASS
-
* A +
(kl
n 037 w +
ARRAY SOLAR PPU THRUSTER AUX.
MASS/ ARRAY SYSTEM 6 INSTL SYS
MASS MASS MASS
STRUC FIXED M A S
1.0 1.0 1.0
IEON
ARRAY
STRUC .16
- BLANKET .-
TIC Sl -
PWR moc.6~8 INSTRUMEN-
TATION .O1
!
AUX SV8
s?EC YAtt
b
STRUC I6
wO PA
DIRECT CAPITAL
DELAY COST PER RECOVERY 4 I EON
LIFE
FLIGHT
L i
COSTIFLIGHT
. b
I I
A typical result of t h e performance and cost model is shown in Table 1.3.2-2 for t h e
case of an up trip t i m e of 180 days and specific impulse of 8,000 sec. The independent
variables associated with t h e indicated results a r e presented in Table 1.3.2-3. Cost
sensitivity t o any of t h e independent variables can also b e obtained. Cost sensitivities
a r e also obtained through use of various combinations of trip t i m e and specific
impulses. The results of t h e various combinations a r e presented in t h e next section.
1.3.2.2.1.3Silicon EOTVS
The key factors influencing t h e EOTV performance and cost optimization a r e (1) t h e
performance of t h e solar array particularly when subjected to t h e radiation environ-
ment berween LEO and GEO and (2) argon ion thruster performance.
SOLUTIQN RESULTS
1 ONE-WAY MASS R A T I O
2 GROSS PAYLOAD R A T I O
3 PAYLOAD EQUATION BRACKET
4 PAYLOAD EQUATION BRACKET
5 ZETA*UP, PERFORMANCE PAR
6 P H I - 2 r DOWN D E G R A D A T I O N
7 P H I - 3 r UP DEGRADATION
8 RETURN T R I P FLUENCE
9 U P T R I P FLUENCE
1 0 RETURN T R I P LOG FLUENCE
11 U P T R I P L O G F L U E N C E
1 2 F L U E N C E FOR 1 8 0 - D A Y T R I P
1 3 ARRAY MASS/ARE 4
14 TOTAL UP+RETURN DEGRADAT
15 THRUSTER-PPU E F F I C I E N C Y
16 ZETA*-UP-BRACKET TERM KGJWATT
1 7 THRUSTER S P E C I F I C MASS KG/KWE
18 Z E T A * DOWN KG/WATT
1 9 ZETA*-DOWN BRACKET TERM KGIWATT
2 0 POWER P R O C S P E C I F I C M A S S KG/KWE
2 1 AUX SYS SPEC MASS KG/KWE
2 2 ARRAY SPEC MASS KG/KWE
2 3 D21IPHI-2 KG/KWE
2 4 D22IPHI-2 KGfKLIE
2 5 I E O T V F I X E D MASS METRIC T
2 6 R E l U R N PAYLOAD METRIC T
2 7 J E T POWER DOWN MEGAWATT
2 8 A V G i E T POWER U P MEGAWATT
2 9 E L E C S I Z I N G PONER MEGAWATT
3 0 A R R A Y D E S I G N POWER MEGAWATT
3 1 RETURN T R I P T I M E DAYS
32 I E O T V MASS AT GEO METRIC T
33 RETURN T R I P T I M E TERM
3 4 UP PROPELLANT METRIC T
35 RETURK PROPELLANT METEIC T
3 6 PROPELLANT SYSTEM MASS METRIC T
3 7 PWR G E N & D I S T R M A S S METRIC T
3 3 PPU SYSTEM MASS METRIC T
3 9 T H R U S T E R 8 I N S T L PIASS METRIC T
GO AUX S Y S T E M M P S S METRIC T
4 1 I E O r V E M P T Y MASS METRIC T
4 2 TOTAL PROPELLANT MASS METRIC T
4 3 O T V S T A R T S U R N MASS NO P L METRIC T
4 4 B L A N K E T MASS METRIC T
4 5 B L A N K E T STRUC METRIC T .
4 6 BUSSiNG & MISC METRIC T
4 7 POWER P R O C E S S O R S M A S S METRJC T
4 8 F P U THERMAL CONTROL MASS METRIC T
4 9 INSTRUMENTATION MASS METRIC T
50 T H R U S T E R S MASS METRIC T
5 1 PROPELLANT S Y S F R A C T I O N = 9.543E-02
9.375E+00 VFTRIC T
5 2 ARGON TAP'KS MASS
5 3 L O Z I L H 2 TANKS MASS -
3
1.391EtOr .":TRIG T
5 4 FEED S Y S MASS I 1.030E+O, TRIC C
55 L I O & GAS R E S I D MASS z ? .030E+01 hETRIC 7-
56 AUX PPOP SYS MASS = 1.288E+01 METRIC T
57 EPS STRUCTURE PASS - i.O58E+01 METRIC T
58 L O t I L H 2 FRACTION = 9.OI\OE-02
59 L G 2 I L H 2 PROP MASS 4.636E+01 METRIC T
60 ARGON PROP MASS
L
- 4.688E+02 METRIC T
61 ARRAY AREA -- 1.502E+06 M2
62 THRUST PER THRUSTER - 2.854E+00 N
63 THRUSTER BEAM CURRENT = 7.777Et01 AMPS
64 TOTAL THRUST - 3.34SE+03 N
65 THRUST PER CORNER - 8.363E+02
66 TOTAL NG OF THRUSTERS .
-- la272E+03
67 NO. OF THRUSTERS/CORNER I 2.930E+02.
68 SUPPLY VOLTAGE = 1.519E+03 VOLTS
69 ZETA*-UP TERM -- 9.222E-01 !
0.9 -
0.8 -
QUCOWdOlu
0.7 -
0.6 -
0.5 -
d, S J i o o r ~ 7 6 6 0 # ) ~ ~ ) l r r l i r ~
0.4
1014 1015 1
0
'6 1017 1018
FLUENCE OF l-MoVELECTROIY EQUIVALENTS
laom
VAalMLEMRUIT
&-
-Mu
REMOVE X OF DAMAGE
0.30 - @ BEGIN FIRST TRIP
@ END FIRST TRIP (60DAYS)
- @
@
ANNEAL & REMOVE SSX DAMAGE
BEGIN SECOND TRlP WITK PP,, - 0.90
The p?jii=- -3nge before anilealing and ai .er annealing 2s a function of the number of
trips is shown in Figure 1.3.2-: 1 along with t h e average power expected during t h e
trip. It should also be noted that as the average power decreases within a given trip as
well as each s h s e q u e n t trip, t h e voltage will alsc- b e decreasing at about hdlf t h e r a t e
as '-he power output (should power go d o ~ n30%, voltage will go dowr. 15%). The
benefit of annealing car, be iliustrated by using t h e case of 10 roundtrips. With
annealing, the average power output during the tenth trip w;:. b e approximately 53%
of t h e initial power output. Should d n e a l i n g not be used, power output for t h e tenth
trip would be approximarely 45% of the initial output.
0s-
No ANNEMmo
aa I-
I I I t I 1 1 1 1 t n 1 1 1 ,
I
1 2 I 6 078)10 16 a 1998.
Vehide Optimization
The first trip flown by the vehicle is used to obtain the optimum velricle i n terms
of performance and cost as influenced by trip time and specific impulse. These characteristics
are shown i n Figure 1.3.2- 13. Total vehicle start-burn mass is showrl as a function
of specific impulse and up trip time, and indicates the minimum mass has not been
reached a t 220 days, however, the optimum specific impulse appears t 3 be 8,000
seconds. Cost data reflect the amortized hardware cost, the cost of refueling and
refurb and trip t h e interest cost. The cost i s minimum with a combination of a
specific impulse of 3,OCJO seconds and trip time of 240 days.
Similar optimization curves have been developed for the 6 mil coverglass blanket
with an Isp of 8,000 sec also providing the least cost. The cost curves for the 6-2-4
blanket are shown in Figure 1.3.2-14 and indicates the 3-2-2 blankei i o provide a
savings of approximately $2 per kg of SPS at the optimum Isp and trip time. Further
detaii concc~ningthe comparison ot t3e two silicon blanket opti 5 is presented
in Figure 1.3.2- 15 using a specific impulse of 8,000 sec and an up trip time of 210 dzys.
EPFlClENCV (PERCENT)
8 8 8 a 8 8 fi
I I 1 I I I I
e
TOTAL TRANS QRTATION COST ISIKG OF SPS)
0,
8 N t? 8
0 10@25037-2
a St LEON &mil E O N SILICON 3-VERSUS Smil EOTV
l#-&oooaa
BLANKET
PROPELLANT
CONSTRUCTION
DELAY
ELECTRIC
1
DIRECT
hfuel
Rcfurbih
GENERATION
DISTRIBUTION CAPITAL
COST
(AMORTIZED)
82-2 62.4
BLANKET BLANKET
~~unhotpylorddd,~(l.l/k~
e Key assumptions
Payload
Up = 4,000 MT
Down = 200 M T
Configuration concentration ratio = 1
DATA POINTS
@ 1577: 50 kW of 12% silicon cells
@ BAC estimate for 10-GW silicon SPS
O) RI estimate for 6-GW GoAs SPS
@ BAC bttlmts for GaAs EOTV'r to
support 10GW silicon SPS
MATURE
LEARNING 1 INDUSTRY
SILICON CELL C--
h t 0pt.imizat.m
The transportation cost optimizatiorr of the two gallium arsenide blanket EOTV designs
is indicated in Figure 1.3.2-19. In both cases, an ISP of 7,000 seconds and uptrip time
of 240 days is optimum with t h e 40 micron coverglass blanket providing an advantage
o: approximately $2 per kilogram of SPS.
TOTAL
TRANS-
PORTATION
COST a
%4 OF SPS) I
eOptimiution
e G (.=) 7.000 I 7.000 8.W 7.000
.Trip time up (&ys, 240 240 180 240
*Trip time down (days)
aDesi_onckaracrctistin
37 37 47
49 1
eQectric sizing power (MW) 125 118 182 115
eP/P, after one round tripfw 48 S2 58 56
o Design power (hlW) 230 203 296 187
. A m y nrea (km2) 0.97 0.85 1.5 0.95
*Empty m;w (MT) 767 718 1,457 957
Further comparison of t h e EOTV1s is provided in Figure 1.3.2-20 through use of mass,
unit cost and total transportatiw, cost. In the case of mass, the baseline silicon EOTV
h a s a solar array which is heavier per square m e t e r and t h e r e is less power per square
m e t e r resulting in a much heavier vehicle. Propellant requirements are also larger d u e
t o t h e greater e m p t y mass of t h e vehicle. Unit cost of t h e t h r e e candidates, however,
show a benefit to t h e silicon EOTV p r i m ~ i l yas a result of the cost p e r square m e t e r
of t h e array being approximately 25% that of t h e gallium arsenide blanket. The
e l e c t r i c propulsion system on t h e silicon system is g r e a t e r because of t h e heavier start
burn mass of the system which also explairs t h e higher launch cost. The t o t a i
transportation cost amortizes t h e capital i n v e d a e n t (unit cost plus launch of t h e
EOTV's), and results i n the silicon EOTV providing a savings of approximately $7 per
kilogram of SPS over t h e baseline gallium arsenide and about a $6 per kilogram
improvement over gallium arsenide with a thicker coverglass.
With t h e level of definition conducted t o date, t h e silicon cell blanket with 3 mil
coverglass is recommended as the preferred solar array f o r t h e EOTV. Should f u t u r e
analysis indicate less optimism regarding solar a r r a y performance and i t s recovery -
with t h e annealing, t h e 6 mil c ~ v e r g l a s smay require reassessment. The gallium
arsenide cell with mirtimum coverglass does not appear t o b e worthwhile for orbit
transfer operations. Again f u t u r e analysis concerning radiation e f f e c t s on t h e blanket
rrtay provide the rationale for investigating CaAs blankets with thicker coverglasses.
The EOTV defined and updated for comparison with t h e LEO construction option
employed a silicon (3 mil coverglass) blanker.
:IFa
o j!
m 40
A
75
-1I
- THRUSTERS
COMPONElJTS - PER ROUND TRiP
- UPJITS - 5-10 YRS
- CAPITAL COST
SOLAR ARRAY - UNKNOWN
DEGRADATIONIilECOVERY
- ------------- CHARACTERISTICS
- CONST. DELAY
CELL-CELL MISMATCH
THERMAL CYCLE IMPACT
(TRIP TIME)
J I I 1 WIDE RANGE OF VOLTAGE
Before making t h e selection of t h e number of round trips for e a c h EOTV, one must
also consider t h e limitation t h a t may occur in t e r m s of component lifetime. No
significant problem appears t o exist f o r t h e structure, power distribution and power
processing components although each of these i t e m s must be examined t o verify this
opinion. Thrusters can b e refurbished a f t e r each trip and as previously indicated
complete units are replaced at t h e end of ten trips. There is g r e a t uncertainty,
however, relative t o t h e life of t h e solar array and i t s performance. Several points
shobld 1e considered. In t h e case of degradationlrecovery characteristics, t h e r e is t h e
f a c t t h a t each leg of an EOTV trip will experience a fluence level ten times g r e a t e r
than t h a t r o be experienced by the satellite in 30 years of CEO operation. There is no
test experience relative t o t h e recovery capabilities from this amount of radiation o r -
t h e number o: tir,,es t h a t recovery can b e performed. Cell-to-cell mismatch arises
because each cell would not b e affected exactly alike in terms of i t s radiation
characteristics, thereby resulting in additional contribution t o overall power output
loss. The thermal cycle impact must consider both t h e case of occultations t h a t occur
during t h e orbit transfers as well a s the annealing of t h e solar array. In t h e case of
t h e occultations, one round trip transfer has as many occultations as fifteen years of
operational life of the SPS system. Another f a c t o r t o b e considered is t h e 10-15%
variation in voltage t h a t occurs throughout a trip.
The selected first trip performance characteristics include an uptrip t i m e of 180 days
rather than 240 days because t h e cost of f e w e r EOTV's would offset t h e cost penalty
for a given EOTV. The resulting basic f l e e t s i z e is 20 vehicles based on f i r s t t r i p
performance. Since each successive trip t o be flown by an EOTV will t a k e longer (due
t o array degradation), t h e 20 vehicles which initially fly 28 flights per year, will only
fly 24 flights on their t e n t h trip as shown on t h e right hand portion of t h e chart.
Consequently, t o maintain an average of 28 deliveries per year a t o t a l of 22 vehicles
will be required in t h e fleet. One additional vehicle is added t o t h e fleet for a spare
giving a total of 23 vehicles.
1.3.2.2.4.1 Configuration
Options
k i t h t h e performance optimization d a t a available a t t h e midterm (power and voltage
requirements) a nearly square solar array resulted. This configuration is shown in
Figure 1.3.2-23 a s Option 1. The key characteristics of this configuration a r e t h a t
t h e r e a r e four thruster module locations and tne EOTV is approximately square
( p r o v ~ d e st h e most deskable moment of inertia characteristics). Several variables
exlst however, t h a t could present different configuration options also shown in Figure
1.3.2-23. These variables include t h e cell size t o be used in the blanket and also t h e
- W I C FLEET SIZE
FIRST TRlP PERFORMANCE
0 180.25037-2
TOTAL FLEET SIZE
(MULTIPLE F L I G H W E O W I
- ,
ETOV'S I N FLEET
24 -
W
N,
V)
-
C
W
16 -
LL
-
8
m 33U DAYS FOR DELIVERY
8 - 28 FLTS REO'D
-
o k 180
I I 210
UP TRlP (DAYS)
L 1 I
1r2 236 299 0 2 4 6 8 10
ROUND TRIP (DAYS) EOTV TRIP NUMBER
TM ITHRUSTER MODULE I
SOLAR ARRAY
Selected Configuration
The selected EOTV configuration is shown in Figures 1.3.2-25 ano 1.3.2-26 and consists
of four solar array bays, with each bay formed by a pentahedron. The apices of t h e
pentahedrons a r e tied together t o serve as a mounting location for t h e payload and
propellant tanks. This location provides a good moment of inertia balance to minimize
gravity gradient torque control requirements and simplifies t h e docking of t h e payload5
as well as propellant tankers. Thruster modules a r e attached t o beams protruding
from t h e four corners of t h e configuration. Power f o r t h e thrusters is drawn from
solar arrays in the bay adjacent t o t h e thruster module. The vehicle is sized t o deliver
4,000 m e t r i c tcns and return 200 m e t r i c tons with a n uptrip t i m e of 180 days and
down t i m e of 40 days, with a specific impulse of 8,000 seconds. The t o t a l dry mass
of t h e vehicle is 1462 metric tons while t h e total propellant loading is approximately
500 metric tons. The 1510 m dimension of t h e configuration reflects t h e change
in power requirements t h a t occurred a f t e r t h e Phase I midterm. The I044 m dimension
is t h e s a m e as t h a t used a t t h e midterm and is a function cell s i z e and voltage requirements.
LSOCAR
ARRAY
--
PAY LOAD
UP 4 W MT
DOWJ 200 M T
T R I P TIME:
-
UP = I80 DAYS
DOWN 40 DAYS
160111
1,
-
POWER OUTPUT = 7 1MW
VOLTAGE 2685 V
CELL AREA = 0.40 km2
CELL 17.6% AM0 -26%
BLAYKET
3 hllL COVER
2 MIL CELL
2 MIL S U m A T E
. .
TYPICAL BAY
-
54 14.66 SEGMENTSIBAY
1075 STRINGSBAY
498 PANELSmRING LENGTH
TYPICAL BLANKET SEGMENT
10
8
m E R LOSS I N YlEGAWAm
PLLLSMA
CbRRENT
6 LUSS
tzil LOSS
400 THRtISlERS
ALT: 599 KM
-
mm 10
THRUSTERS
ZBCLSESHAVEPPVS~
SWRY6fnRUSlERS+
- ~ U E E P C J AWD
G
LWERGY
F m 1-32-29 m- &
60 -
TOTALMASS- b L l S W + A R R A Y MASSTO
mwmsATE HWI COtYIW:TOR 141 L m
YASSIUYETRKTOWE
20 - ARRAY MASS TO
0 1 I I I I
0 2s 60 m loo
TVPE OF PROCESSING
MOTOWGENERATOR
f SOLID STATE
PROCEsslNGTHERMAL CONTROL
ACTIVE RADIATOR
S~SM*
LIMIT ELECTRON= TO -2
SWITCH GEAR
INTERRUPTERS
--
GIMBAL ' -SUPPORT
LIE t90 dJ STRUCTURE AA
IONIZATION
ACCELERATOR
GRID 1 T
I CHAMOER
OROUNMJ 2
SHlELO
SCREEN GRID -
PERMANENT
MAGNET
LrnOPELL.WT
INLET
MIDTERM FINAL
FLEETSIZE
The final Phase I EOTV hardware and cost per flight numbers a r e presented in Table
1.3.2-9. In t h e case of t h e hardware costs, both mid-term and final costs a r e
presented. The final flight unit costs have almost doubled from t h a t of t h e m i 6
term, reflecting t h e influence of t h e more detailed cost analysis. The power generation
and distribution system has not increased as much as e l e c t r i c propulsion system primarily
because t h e solar array, which is t h e largest zontributor, was and still is beirlg costed
on m a t u r e industry basis with t h e increase over precedicg mid-term values primarily
t h e result of t h e 20% penalty paid f o r using t h e > x 10 c e n t i m e t e r c e l l and a!so the
21% cost growth factor. Electric propulsion costs, are greater by 3!m~?st2. f a c t o r
of 3 and reflect a significant difference in t h e c o s t f o r individual c!ements a s a result
of lower production rate. As indicated earlier, programmatic costs were not indicated
in t h e mid-term. On a c o s t per flight basis, including amortization of t h e capital,
the change from t h e mid-term has been approximately $30 million per flight.
~ -
KEY COMPONENT
t L E O M
ANNUAL PRODUCTION RATE (UNITS)
1 GEOIEOTV I LEOlSPM/EOTV
THRUSTERS 5340
PWS 381 634
(1 PER 80 THRWIERSI (1 PER 10 THRUSTERS)
SWITCHGEAR
INTERRUPTERS
CABLING
TANKS-ARGON
GtMBALL ASSY
AVIONICS
COMMUN
COMPUTER
THERMAL CONT
POWER DlST
STANDOFF STRUCT
COST I N MILLIONS
PART 1 PART 1
MIDTERM FIEAL BASIC AMORT.
A TIME (DAYS)
EVENT DESCRIPTIGN
ON-ORBIT TRANSFER
rlUMBER OF
a COSI: TBD
0 1
NUMBER OF
REFURB VEHICLES
LOCATION MODE
1.3.2.2.8.1Configuration
The primary functions of the base a r e t o support construction operations associated
with t h e EOTV1s and t o perform depot type opf?ratians during t h e o r ~ g c ~ nsatellite
g
construction operations. The overall base concept and key characteristics a r e shown
in Figure 1.3.2-36 and i t s dimensions in Figure 1.3.2-37. The base is sized t o c o n s t r i c t
one bay of an EOTV at a time. 0utr:ggers a r e used t o support th: bays as they are
being constructed. Opposite of t h e construction platform is t:;e location used for t h e
docking of t h e OTV's and HLLV1s. Crew modules are located at one corner of the
facility and consist of two crew modules for t h e primay crew, one module for
CREW S U E
CONST 200
OQS 130
CAROO~~MFER
C m f IMEROTV = 23 M y $
O M M A S S - ljOOYT
CCm-QtB
HLLV
CARGO VEHiCLE
DOCKING AND
CARGO SORTING
AND STORAGE
FACILITIESON
LOWER LEVEL
700 TVPlCAL
1 OUTRIGGER
STRUCTURE
LA. ---- - --
. .-
-.~-
'- . ;
I!
-I- TYPICAL
PLATFORM
STRUCTURE
Details of the construction operation associated with each bay of the EOTV are
il1ustra;ed in Figure 1.3.2-39. Again it should be emphasized t h a t the base has been
sized t o carstruct one bay at a time, rather than a complete EOW. The construction
operation requires a ccnstruction platfarm, beam machines, cherry pickers, solar array
deployers, indexers and a cwstruction gantry which is us& t o support several
machines amj cherry pickers. The sequent-. which is rlsed to form t h e s t r m t u r e of
each bay is illustrated in the lower left hand portion of the chart. Both the gantry -
beam machine and t b platform beam machine work in parallel forming t h e beams. In
this particilar operation, the platform beam machine is relocated one time in order t o
complete the formation of its designated beams and the gantry must be moved t o t h e
side t o allow the last beam of the pentahedral b e t o be installed. Total construction
time t o complete a single bay of t h e EOTV including checkout and its indexing so t h e
next bay can b e made is 5 days with the provision that two solar array machines are
used. Should only one solar array machine be used, then the construction time per bay
will be increased t o seven days.
L m.
PLATFORM
I CONST BAY 4
BAY 2 BAY 1 T 20 OAYS
f
1 w 2 : :my1
- -L - - - . - - -- - L.-L---
Ib
I
I
i 4,
I
I
BAv3, BAY4
I
INSTALL PROP TANKS
6Av 3 ITJSTALL PAYLOAO
I r F I ~ A LVEHICLE CIO
I *
7
- 1-23LMVS
S O U R ARRAY I I J S T A L U T W
- -L
COtJSlRUCTIOfdDAYS
L,,,,I
]-I
y
-
STRUCTURE
j SOLAR ARRAY
BUS lrlSTAL 6
Y-B
ARRAY HOOKUP
@
THRUSTER m U L E
ASSY AND ItJSTALL
0 1SUBSVSTrhi INSTALL
0 CHECKOUT
----- ---- a INDEX
STRUCTURE SEQUENCE
0
FLIGHTS
FLIGHTS
CREW ROTATtW
RESUPPLY O W
EOTV (281
:
1
I
I I
HLLV 5 8 FLIGHT-EK (365)
PLV
i (38)
A
GEO BASE
+
EON ONCOlllG EOTV CONST
CONSTAUMION OPERATIONS + WERATIONS
BASE MGMT (7) 01 0)
CONSTRUCTION
MGMT
m ) (0) m,
6 6
EGTV CONST 46 46
SUB.\SSY 10 10
TEST & QC 15 15
BASE OPS & SUPPORT (93) (a) (93)
MGMT 6 6 6
I,IdlItJTENb.NCE 14 10 14
VE1I:CARGO HANDLING 16 - 13 16
FLIGHT CONTROL 6 6 6
COMMUNICATION 8 ' 8 8
DATA PROCESSING 6 6 6
UTILITIES 12 12 12
HOTL OPS 16 16 16
MEOIOENTAL 9 7 9
TRANSPORTATION OF'S (211 (43) (431
II lGMT
PROP HANDLING
4
8
4 4
8 8
FLIGHT READINESS 7 7 7
E O N MAltlT 1 0 22 22
VEHICLE COOhD
-2
-
2
-2
TOTAL 198 134 220
1
STRUCTURE
CREWlWORK MODULES
BASE SUBSYSTEMS
VEHICLE~CARGOHANDLING
PROP. STORAGEIDELIV.
CONSTRUCTION EQUIP.
Several inlprovements have been considered for this concept. The first deals with
irnproving t h e overall module configuration for t h e transfer operation. The second
considers t h e cost benefits t h a t might occur through recovery of t h e e l e c t r i c
propulsion components and their subsequent reuse. Both of t h e s e improvements will b e
discussed in subsequent paragraphs.
13.2.3.1 Catfiguration
The self power module orbit transfer configuration including key characteristics
associatea with transferring with and without an antenna a r e shown in Figure 1.3.2-43.
The optimum trip t i m e is 140 aays and ISp = 7000 sec although t h e t i m e averaged I
SP
including occultations and L 0 2 / L H 2 contribution f o r gravity gradient torque (CGT) is
D 180-25037-2
DELIVERY OPTIONS
REQI'IREMENTS (1) COMBltJED (2)SEPARATE
480 PEOPLE
CREW + SU!PLY
--cncw SUPPLY
FLTlMO I I I
3 STAYTIME CREW 160 160 -
MAN MO. UPPLIES 180 - 480
230 K*AN MONTH PIL UP ( 1 3 XgI 150 52 96
PIL ON ( l d Kg) 99 62 48
2 STAGE L O g L H Z
SYSTEM SELECT
ANNUAL
h
PROP. PROP
SAVES APPROX.
PER FLIGHT (lo6 KO) f 100 UILLION/YEAR
(103 ~ 9 )
CREW SUPPLY
@ @ OTV FOR
LEOCONST
0 0
OPTION
OPTION
OELIVER C R E M
AF!D SUPPLIES
TO GEO USING -
RETURN CREWS
.DEPLOY 6 ANNEAL
GENERAL CHARACTERISTICS
a 3%&zing (rdiation)
Trip tima = 140 dyr
I~p-7,OOouc
MODULE NO WITH
CI1AHAC iEnlSTlCS ANTENNA ANTENNA
6 2
8.7 23.7
0.3 0.81
PROPELLANT 13 36
1-1 2.9
1.O 5.1
1.4 2.2
4.5 12-2
120 8.0
,1 2.4 3.6
Iw 13 1.0
1, 2.75 5.2
MODULE WITH ANTENtJA
NO
ANTENNA ANTENNA
Padsize: 2411% 48a57m .
Thrustus: 600 1-. SECTION
A-A
Recovery % . 87 56
THRUSTER MODULE
- EOTV BAY
I
cosr LARGE PROP LOWEST MORE OTS HIGHEST OF
CONSlDERATlOl RE07 UNITS REO'D THE E O N ' S
LOW OTS PROD.
RATE-HIGH
UNIT COST
*SELECTED
MOST STR41GHT
FORWARD
*MOST FLEXIBLE
Several options exist in terms of the size of the EOTV. These options are brought
about by several different payload requirements associated with t h e modules. As
noted in Table 1.3.2-14, six of t h e eight modules have a recovery payload mass of
approximately 550 m e ~ r i ctons, while two of t h e eight modules have OTS components
that total 1650 metric tons. A detailed analysis has not been conducted on the three
options indicated, but Option 2 which sizes t h e EOTV t o return t h e largest payload
appears t o b e E! rezsonable choice and was used in the remainder of the OTS recovery
analysis.
The flight operations schedule associated with use of independent electric orbit
transfer vehicles for recovery of OTS systems is illustrated in Figure 1.3.2-46. This
table 1.3.2-14 0TV Sizing Opt h r
REOUI REMENTS OPTIONS
d SELECTED
Dy mass - 760 MT
Trip timr:
-
Up = 30daya .
Argon - 230 MT
D o m 140 drys
- 200
@?---------- I
100
300
1
400
I
WO
I
600
I
790 .--
SlMl e---------,
WMl
-
81lMZ : ----.I8 HhM 1 .-.-----&- SWM1
sm2 !
I
I I S3/M2
SllM3 I
L------.--.-
--..-.--.----
A
I
I 82fM3
YOKE,
I ;--: WM3
i YOKE I+
I
I
I
-
I
8
I
SllW S21M4 :
SllM5 1 'I k--------:-- S21MS :
i
?--------------
S3/W
-ww4
* I I I
t
I I
- k - L - - J
c.nlr:
-
A ' % Construction
0
TrMsfsr
Loding
numbs
SItsflita n u m b
Hudwulr roquiremrcrtt
01s tor r i m modulm
Thm O W 8
The recover) ar8d refurb operations performed a t LEO are shown in Figure 1.3.2-48.
The EOTV returns t o LEO a t a location near t h e LEO construction base. The OTS
pallet vehicle is the11 flown from EOTV over t h e LEO base where components a r e
removed and taken ro t h e ref!irbishment facility. The empty 07s pallet is flown back
to t h e EOTV for a subsequent trip to CEO. Meanwhile, maintenance vehicles from t h e
LEO base a r e flown to t h e EOTV t o perform maintenance on t h e thruster modules of
that vehlcle.
@ FLY -DOC)( A
----- \
m?MLTvEMMq / - \
mfTW / I
- - . \ OW R ~ C H E Scxo
, A x \ m o s ~ ~ n oICES
w -LLACIT
OTS PALLET ; WEAR6s
VEHICLE TmBc
OTSPALLET
VEWCLE
/
--/-- OOY#WENT
/ ---.-__ _
wwE
- - -- - @ RECOVER
~ ~ ~ L L * N T
i
THRlSTER
RECOVERY
VE-
TANK VALLEY YOOUrE
- _ - -- .
-
@ RETURN EMPTY
OTSPALLLTVEHICLE
TO EOTV
+
rr :I
EOTV
MAINTENAIVCE ,, MAlNTENAWCE
j VEHICLE
- - - -
-_ -- - --
KEEPS NEAR BASE
i
a
_ - - ~-- _- 1
\
@ FLY AND 0 0 C K L O A M 0
qTS PALLET VEHICLE
1"LEOBASE . LEO B N E
OTS COMPONENTS TO
REfURBISHING FACILITV
133A.1 Caatructian F q a a t i a n
Initially i t was thought that GEO construction usim EOTV's f a r cargo delivery w u l d
require a longer preparation time in terms of when t h e first SPS could be put on line.
Data presented in Figure 1.3.2-49, however, indicates this method can have i t s system
elements arranged in a manner t h a t results in t h e first satellite coming on lint at t h e
same time as t h e LEO constructim method. The only difference b e t a r e n these! two
options at this point in time appears to be t h e time when the chemical orbit transfer
vehicle must be available. For the case of the LEO construction concept, the cbem
(102/LH2)OTV is not required until approximately 16 y e a s a f t e r the first system
element payload is launched and is used to uppat the constructior. of the CEO final
assenrbly base.
In the case of CEO c a t r u c t i m , the &em OTV must be available at the end of the
first half year in order to provide t h e capability to deliver components of t h e satellite
construction b w which will be assembled at CEO. In addition to the difference in the
availability date for the chem OTV, t h e CEO consZuction chem OTV will also be about
twice as large in terms of propeliant capacity.
ORBITAL Emp
CREW REQUfREMENTS
LAUNCH 0PERATK)IYS
=5
-7 PREPARATNMI
FlRST SATELLITE TRANS-
* AVERAGE PER SATELLIE
YEARS
0 1 2 3 4
b 8 a 8 4
LEO
CorvSTRUCTHYNl
SELF K m E R
CONST -ij
MODULES
CEO CONSihUCTIONf
EON
LAST CARGO OELlV FOR 1ST SAT
E O N #1-9 MAKE 2 FLTS
E O N 810-19 WAKE 1 FLT
LS
SPJ ON LINE
T& 1.U-16 1- S ~ m m ~ y
LEO C h s t m d h
SELF m R TRAMsFER
SOUR ARRAY
The orbital bases for the CEO construction concept are shown in Figure 1.3.2-51. The
LEO support base has a mass of approximately 1300 metric tons and requires a crew
size of around 200 during the construction phase of the EOTV. Once the program is
underway, t h e crew size can be reduced t o 130 people since only depot type operations
a r e performed. The GEO construction base has the task of constructing a monolithic 5
gigawatt or 10 gigawatt satellite. The mass at this base is 6,250 metric tons with t h e
increase over the LEO satellite construction base being primarily that related t o
additional radiation shelters for the crew.
--
LEO CONSTRUCTION BASE
MASS 5650 M T
CREW 407
CONST
--
SOCONSTRUCTIC?rJ BASE
MASS 6250 MT
CREW 407
ANTENNA C O N S PLATF
CONSTRUCT ION
--
LEO STAGING BASE
MASS 1320 M T
CREW 220 CONST PHASE
130 OPS PHASE
OPERATIONS UNDERNEATH
FINAL ASSEMBLY
GEO CONSTRUGTW ALLOWS MONOLITHIC WTELLITE
\\
LEO CONSTRUCTWU UTILIZES A MODULAR DESIGN
AND REWIRES THE MLLOWINQ GEO OPERATIONS:
SATELLITE
SJRUCTURE ,
@DEPLOY SOLAR R
,AY
(MODULES 4 AND 8)
RADIATION
SOLAR FLARE 2 3 GMICM~ m2S GMICM~(115 000 KG1100 PEOPLE)
EVA SO. ATLANTIC %TEAOY 8TATE IS WORSE
ANOMALY RESTRICTION
OCCULTATION
EASE POWER REO'TS: 3600 KW 2500 KW
LIGHTING: REQ'D AT BOTH LOCATIONS (A OF 100-150 KW)
THERMAL EFFECTS: NO SIGNIFICANT DIFFERENCE I F GRAPHITE TYPE STRUCTURE
IS USED
GRAVITY GRADIENT & GRAVITY GRADIENT CONST MODE USED FOR BOTH LOCATIONS
DRAG: LEO CONST PaOPELLANT IS GREATER BY 600-700 KGlDAY
I-
B 1200 REV
28100 REV
16 FLTS Wid 8,3lO REV
The second i t e m to be compared i s t h a t dealing with potential interruptions of power
beams originating f rorn operating power satellites. The problem occurs since t h e
modules o r vehicles transporting c a r g o depart from a 30 degree inclination orbit and
have a destination of 0 degrees at CEO. These interruptions will b e proportional t o
t h e number of re: olutions t h a t t h e vehicles make in t h e transfer from low orbit to high
orbit (See Section 1.1.2.7). Again, the t o t a l number of flights plays a key p a r t in this
estimate. The LEO construction concept using self-power rnodules is estimated t o
require a total of 6,400 revolutions t o g e t one 10 gigawatt satellite t o CEO. In t h e
case of GEO construction using 23 EOTV1s flying 28 flights per year, a t o t a l of 28,000
revciutiors i s required or approximately 4 times revolutions per year, which should
i ~ d i c a t eapproximately 4 times as many interruptions of t h e power beams coming down
as f o r t h e LEO construction option.
132.4.7 RiskhJncertainty
As s t a t e d previously, t h e LEO construction concept uses self-power and as such t h e
orbit transfer system may b e used only once although recovery and reuse is possible as
discussed previously. The construction concept using EOTV's, however, requires
multiple use f o r each EOTV. Concerns f o r t h e multiple use EOTV's a r e indicated i n
Figure 1.3.2-19. In t h e case of t h e solar arrays, t h e c o s t optimum transfer t i m e for
each flight will result in degradations as low as 40 t o 45% as compared with 30 years
of satellite operation which will degrade approximately 10%. The impact of this deep.
degradation is not known in terms of overall power gereration capability nor in terms
of t h e number of anneals which can b e made nor t h e level of recovery. Cell t o ceil
mismatch occurs even though annealing has been performed since e a c h cell has i t s own
unique characteristics. With excessive cel! to cell mismatch t h e r e would b e non-
optimum power characteristics from t h e solar array. The impact on t h e solar array
resulting from the large number of thermal cycles associated with occultations and
anneciling is unknown. Finally, as t h e power output degrades during t h e missions, s o
will t h e voltage degrade which will present s o m e complication in t e r m s of power
conditioning equipment. The other components indicated also o f f e r some concern,
however they a r e judged t o be less significant. In the case of avionics, one typical 180
day transfer presents a dose of approximately 104 rads. This radiation level will
require use of radiation hardened electronics particularly when 10 flights (10 5 rads)
a r e planned. The impact of radiatior! hardened electronics is twofold. One, t h e
system will be slightly more expensive than standard avionics, a n a two, the number of
design solutions will b e restricted. The final item t o b e considered is t h a t of t h e
LEO CONSTRUCTION
SELF WWER SYSTEM IS USED ONLY ONCE
GEO CONSTRUCTION
8QTV IS A MULTl USE SYSTEM I N A nQ8tlLrLLO-GBO CNVIRONMENT
COMPONENT CONCERNS
SOLAR ARRAY
DEEP DEGRADATION
RECOVERY
CELL TO CELL MISMATCH
THERMALIANNEALING CYCLES
VOLTAGE FLUCTUATIONS
AVIONICS
STRUCTURE
Costs presented in Figure 1.3.2-53 a r e shown in more detail in Table 1.3.2-20. During
t h e construction preparation period, t h e chief difference between t h e options is t h a t
associated with the placement of the orbital bases including t h e unit cost of t h e bases.
The second difference is t h a t of t h e amount of ground production facilities for the
SCWO ,
25 -
B 20-
LEOISPMIEOTV
'fQTAL
DEWY
' 10225
4 3 0 1 5 0 7 I i 0 ( 3 0 ~ 7 5 0
11935
t380)
19(16
50
(m)
11135 13635
~901 (3801
8610 1
60
1 3 0 4 5 0 7 5 0
(380)
1965
-
(a901
so95
I
79.8 1155 137.7 86.9 80.4 81.8
orbit transfer hardware. The cost penalty reflects $1 for e a c h $1 of recurring OTV
hardware costs t h a t show up under t h e average per satellite column. In t e r m s of
direct cost during the construction preparation period, the numbers r e f l e c t approxi-
mately half t h e crew size used in the normal construction operation but spread o u t
ove: a tu. J )car t i m e period. fhe CEO construction case has the majority of the
orb^:..:! crew a - LEO thus resulting in the highest cost. Total cost for t h e construction
prepaiatiort pe-iod indicates t h a t t h e LEO cons'iruction approach with recovery of t h e
clecrric tran.y>ortation system to be t h e lowest cost.
Tt-ii serona major cost compar ison covers t h e transportation c a t associated with
placement of t h e flrst satellite. In terrns of capital costs, t h e LEO construction
approach with no recovery of t h e e l e c t r i c transportation system provides the least
cost primarily because i t has a very small OTV investmefit. A LEO construction case
with recovery reflects a somewhat higher cost p r i m a r ~ l yas a result of low production
rate ocl the e l e c t r i c propulsicn components. The GI50 construction case, with a f l e e t
of 23 vehicles results in the highest capital cost. In terms of t h e d i r e c t c. st for this
period, t h e LEO construction case with no recovery has sightly higher costs althcugh
not signiticant. The prcpellant required for t h e transfer of each satellite in the LEO
case is approximately twice t h a t of t h e CEO construction concept, hoaever, such
f a c t o r s as lower costs, associated c r e w rotation and resupply, and no refurbishmerrt
during t h e first year, o f f s e t this t o some degree resulting in t h e small difference
between t h e concept in terms of direct cost. Construction delay t i m e primarily
reflects the f a c t t h a t for LEO construction, t h e trip is optimized at around 140 days of
transfer while t h e CEO construction is more optimum a t 180 days of transfer resulting
in slightly larger interest payment. The total cost during this phase shows t h a t LEO
c o n s t r u c t ~ mwithout recovery being nearly $3 billion cheaper than t h e LEO construc-
tion with recovery and approximately $5.5 billion cheaper than t h e CEO construction
concept.
The final comparison of these c c x e p t s deals with t h e average per satellite cost which
amortizes all capital costs. In t h e case of LEO construction k i t h no recovery, t h e cost
indicated is t h e s a m e as that ia t h e first satellite since a complete set of orbit
transfer systems is needed f o r each satellite. The LEO construction with recovery
concept and CEO construction using EOTV's both amortize t h e unit cost of the electric
prcpulsion eqbipment and i t s placement. The total average per satellite cost shows
that approximately $130 million savings per satellite the LEO construction ~ i t h
recovery of OTS over the CEO construction case and approximately $ 7 0 ~million over
the construction with no recovery.
Another cost comparison that can b e shown deals with the uncertainty associated with
the electric OTV concept and particularly t o t h e cost sensitivity t o the amount of
radiation damage that can be removed with annealing. Previous analysis has assumed
95% of t h e damage is removed with each anpealing. A limit case occun if one
assumes that no recovery is podsible in terms of annealing. The results of this analysis
a r e presented in Table 1.3.2- 21. In t h e case of the LEO construction concept, this
will result in a cost penalty of approxim;tely $740 million per satellite which is a
resuit of having t o oversize by approximately 8%. Far t h e CEO construction concept
using EOTV's, there must be an assumption regarding the w m b e r of uses possible for
each EOTV. In this arrctlysis i t is assumed that once the power output falls t o 50% of
initial power output, sufficient damage has been done t o the array and probably t o
supplemental systems that further use is not possible. The 50% level is reached a f t e r
4 EOTV trips if no annealing recovery is possible. The average trip time during these
four trips will be 280 days resulting in an amortization period of 3.5 years rather than
7.1 years in the baseline EOTV case that uses radiation damage removal. As a result,
the cost penalty per satellite will be 1230 million which is approximately 70% greater
than the LEO construction concept using self-power. Consequently, i t is judged that
the CEO construction EOTV concept is much more sensitive t o the understanding of
radiation and its daniage removal through the use of annealing.
-
-
TRANSPOXTATION
AND
CONSTRUCTION
COST LEOISPIWEON
(DOLLARS I N
8
J LLIONS)
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ,
0 2 4 6 8 10 12 14 16
NUMBER OF lOGW SPS (ONE PER YEAR1
-
a A COSTlsATELLlTE (AVG) = S1,ZWMlLLlON
-
1,3,2.4.9 - F bution
A summary of all the comparison parameters used in t h e construction location
comparison is presented in Table 1.3.2-22. Some of these parameters have indicated
little or no difference between the construction option. The GEO construction option
using EOTV's has a n advantage in terms of impact on t h e satellite design and also in
terms of the cons:ruction operation. LEO construction with no recovery of the
electric transportation system is judged t o be better in terms of orbit transfer
operat ions and uncertainties associated with orbit transfer hardware design. In terms
of construction cost, the LEO construction approach has a n advantage while the LEO
construction concept with no recovery has a cost advantage through placement n the
first satellite. On a recurring cost basis, LEO construction with recovery of the orbit
transportation system and the CEO construction concepts are approximately equal in
cost.
Electric orbit transfer operations will consume roughly 10,000 tons of argon per
10,000 megawatts of SPS's emplaced. Although this consumption rate is entirely
negligible in terms of resource availability, 10,000 tons of argon represents more
than lo3* ions and neutrals injected into the Earth's magnetosphere during orbit
transfer operations.
--
W A R I S O N PARAMETER
0 0
LEO/SPM L E O ~ O N
.
CONST PREPARATtON *'NO SIGNIF DlFF SAME TIME FOR
FIRST SATELLITE
r SATELLITE DESIGN IMPACT / NOYOOULARITY
SMALLER LOADS
ORBITAL BASE- EOUY W -IF Dlff SAME CONSTBASE
a STAGING DEPOT VS
FINAL ASSV BAS
CONSTRUCTIONOIS / NO MODULE BERTHING
OR ANTENNA HINGING
CREW RE07S / SAME SIZE BUT
MAJORITY AT LEO
r ENVIRONMENTAL FACTORS *NO SIGNIF DlFF ALL CAN BE HANDLED WITH
ACCEPTABLE SOLUTIONS
ORBIT TRANSFER WS ?' FEWER POTENTIAL
COLLISiONS AND BEAM
PENETRATIONS
r LAUNCH OPS . APPROX SAME NO. LAUNCH=
r RISW~JNCERTAINTY MULTI USE I N HOSTiLE
ENVIRONMENT NOT RE-
r CONSTCOST C H W E R *r#B
r FIRST SAT. TRANS COST
/
CHEAPER $36 OVER 2
sm OVER 3 8
AVG. COST PER SAT / CHEAPER 6 0 S B I
J I~ICATESMOST~OMIS~O
The motion of a charged particle trapped in t h e Earth's magnetic field follows a reflected
helical path within a curved flux tube as illustrated in Figure 1.3.2-54. The Earth's
field is a magnetic mirror configuration t h a t will cause t h e particles t o be trapped
with long lifetimes provided c e r t a i n conditions a r e met. The particle trajectories
migrate around t h e Earth in longitude also, as illustrated in Figure 1.3.2-55.
This analysis was highly exploratory and not a; all complete, but reached certain
preliminary conclusions.
1. Argon beam from thrusters travel intact f o r l o ~ dg i s t a x e s tangential ;a orbit
( a t injection dngle).
4. Above 2.5 earth radii (10,000 km altitude) most of beam argon escapes in a jet
of plasma (86.6 km/sec); a small fraction is peeled off into trapped orbits.
7. At low altiudes (below 1000 km) this thermal argon diffuses back into the upper
atmosphere.
9. The primary interaction between resident plasma and argon from the thruster
is charge exchange between ions and neutrals.
10. The random walk diffusion of r~eutraland charges species is a second major
unsolved problem, complicated by the inf ulence of geomagnetic field orbit effects
on ions.
0 180.25037-2
CENTER ORBIT
MIN
a EOTR OEGREES 27.1 , 163 0.4 63 3.8 2.4
1
, CEaTRl SECONDS 028 0.67 25 6.3 11 18
1 5 KE V ARGON
GYRO-PER100 (SECONDS!
0.1 1.0 10 100
7 1 , 1 I 1 I I I 1
GYRO-RADIUS(KM)
LAfDTUOE BOUNCE PERIOD (MINUTES1
I GEOSYNCHRONOUS A L T I M E
I
1 I I 1 I 1 1
10-1 '10-0 lo1 I$ lo3 to4 105 108 107 108
LIFETIME (SECONDS)
Problem Statement
The construction and emplacement of SPS's in geosynchronous orbit at a nominal rate
of 10,000 megaw;rtts per year will require launches of a large heavy lift launch vehicle
(HLLV) a t rates approaching once per day.
The number of launch pads that can be provided a t KSC, observing noise and pad
separation constraints, will be limitd to 2 or 3. Figdre 1.3.7-1 illustrates a concept for
placement of pads offshore on causeways. The design of the causewLys and
breakwaters would have to reflect environrrtental considerations. As an exampie, it
would probably be necessary to employ ~ i l i n ginstead of riprap so that drift currents
parallel to the shoreline would not be intel,rupted.
One way to relieve these concerns would be to employ remote launch sites fo
operations. Several approaches have been discussed.
A. Ar! inland site, such as in the desert southwest of the U.S. This would allow
downrange landings of boosters, improving performance hv elimination of flyback
fuel requirements. These sites are, however, further from the equator than KSC
and incur added orbit transfer delta V penalties. Further, the environmental
issues associated with inland sites have seemed more and more intractable as time
goes on. Inland sites were therefore dropped frorn further analysis in this study.
C. Sea-based sites, either floating or construction on the sea bed using Texas tower
structures.
Orbit T i ~ f e S
r V-
An initial a t t e m p t at performance simulation f o r a low latitude transfer from 5'
inclination t o geosynchronous orbit actually exhibited reduced performance a s com-
pared t o t h e ref rence 30' case. Increased loss d u e t o gun occulations apparently
more than offset t h e reduction in delta V and in gravitv gradient losses. In fact, t h e
variations in solclreiectric orbit transfer performance due t o variations in orbit
gecmetry and .<(*asona r e greater than performance differences between !ow latitude
Figure 1.3.7-2. PotentEd Equate& L a u d Sites
Access t o 011/6as F i e l d 3 2 1 3
Downrange T r a c k i n g S i t e s 1 3 2 4
I n d u s t r i a l Sase. Energy. e t c . 3 2 2 4
Loglstlcs. Port F a c l l i t i e s . rtc. 2 1 2 4
Sea Route D!stance
C 1 {mate
Hlgh l4ountalns
D 1W25037-2
and 300 inclination starting orbits. Therefore, in determining t h e cost benefits from
performance improvement between low inclination and high inclination orbits an upper
bound analysis was adopted. This analysb is summarized in Table 1.3.7-2 and shows
t h a t cost advantages for t h e low latitude transfer are minimal.
Although it is not likely t h a t this small cost advantage can overcome t h e cost
increases associated with departing from an existing facility, such as K C , other
reasons may exist for setting up a new launch s i t e f o r SPS operation. These reasons
include t h e likelihood that t h e scale of SPS transportation operations will eventually
outgrow KSC as well as potential desirability of a n international launch s i t e f o r what
could e v e n t u d l v become a n internationa! project.
Tramportation Cost
Representative costs f o r transportation of SPS hardware t o remoe s e a coast s i t e s a r e
enumerated in Table 1.3.7-3.
mLTANWSROFHLLVFLIGHfS-17
(400 RATHER THAN 417)
COST FOCI REFERE- CASE 6.570 MILLHJN NOT INCLUOIW TAW TIME QOST
D i s t a n c e t o Launch Site
-
SHIP
F r e f g h t Cost allllor
- T o t a l Cost m i l l ion
-
AIR '
F r e i g h t Cost allllon
TIae i n T r m s l t hours
l o s t Revenues million
T o t a l Cost
~ O I W ' SITE
3 AND
MAlh6i'ENANCE
FLOATING P U TFORM
ANCHORED TO SEA FLOOR
L w h LH2
PROPELLANT cmolnorulw
AND STORAGE PLATFORM
ATTACHED TO SEA FLOOll
FLOATING LAL!C)(
PUTFORM WITH
BREAKWATER, ANCHORED
TO SU FLOOR
L* PROQELLANTCONDlTlOMINQ
AND STORAGE PLATFORM
ATTACHED TO SEA FLOOR
B00STER/ORSITER
ORBITER
PROCESSNO
u.7 BERTHS f OR
LAUNCH VEHICLE
I !--
T RANSPORT SHIPS I V.1iYTENANCE
-. ACDADMlNSlR4
PLATFORM ANCHORED
TO sA FLOOR
RAIL SYSTEM
FOR PAYLOAD
FLOATING LAUNCH
PLATFORM. SECURED FLOATING BREAKWATER
BY CABLES TO BREAKWATER
/- AWMORED TO SEA FLOOR
figure 1.3.7-8. P.y/oad Assembly mi St
o m Fsci/ity
Fipun 1.3.7-9. Boomn m g Facility
- 2 ~ '
1. Awmbkd p ~ l o drliwnd
d m OPF VIA f w d rail mnsportrr.
2. Sups mrad on mobik strongback. S m g b d t truuportd to lunch pid by ship.
3. R ~ h a t a r ~ l u n e h ~ r d m L u l ~ p d h p r o p . l ; n t ~ b r p r .
Launch Pads
Mating & Integration Hangars
Propel 1a n t Stcrage Faci 1!ties
Launch Control Center
Orbiter Processing F a c i l i t y
Booster Processing Facil ity
Payload Assembly Facil it y
Maint./Admin. Building
Mated VEhicl e Transporters
GSE
Railroads, Misc. Sup. Vehicles
& Equip
Total $2,197.25 M
Launch Pads 3
Vehicle Transfer Ships 2
Mating and Integration Hangars 3
Propel 1ant Storage Faci 1it i e s 2
Propellant Transfer Barges 4
Launch Control Center 1
Orbiter Processing Faci 1i t y 1
Booster Processing Faci 1it y 1
Pay1oad Assembly F a c i l it y 1
Maint./Admin. Building 1
Mated Vehicl e Transporters 3
GSE
Railroads, Misc. Sup. Vehicles and
Equ ip e n t
5. Far East: Malaysia, Indonesia, Phi 1ippines (Mindanao) , Papua New Guinea,
or perhaps Austral i a (Cape Ycrk).
6. West Coast, South America: Colombia, Ecuador o r Peru. A l l sites i n this
arFa r e q u i r e launch over the Andes and the Amazon basin, which i s very
sparsely inhdbi ted.
In addition t o these s i t e s , there a r e a number of oceanic islands
which might be considered:
7. Indian Ocean: Seychelles, Chagos Archipelago (Diego Garcia), and the
Maldives.
8. Pacific Ocean: Bismarck Archipelago (New Ireland), Solcnnons, Carol ines,
Marshall s, Gilberts, Nauru, Christnlas Island, Galapagos Islands, and
many others, mostly small.
There a r e no suitable islands in the Atlantic Ocean.
Finally, a floating launch F a c i l i t y could be b u i l t and located
wherever is convenient, i n international waters.
A1 though most equatoria: nations may be classed as l e s s developed
or emerging countries, there are of course substantial differences between
them, i n the political climate, industrial base, gross national product,
population, level of education, culture, etc., a1 1 of which should be taken
into account i n choosing a potential launch s i t e . In the present study,
however, the focus of i n t e r e s t was launch s i t e l o g i s t i c s {in particular,
t e r r e s t r i a l transportatior: costs): the objective is not t o recomnend an
equatorial launch s i t e , b u t t o find out whetner the probable costs incurred
by operations a t such s i t e s exceed the benefits. For these purposes, i t
is sufficient t o choose, more o r l e s s a r b i t r a r i l y , a particular nation i n
each of the restricted geographic-' areas l i s t e d above. To make the
canparisori definite, the to1 lowing nationc were therefore considered:
1. French Guiana, because of the existing French range.
2. Liberia, which has a longer overwater range than Ghana, and which has
expressed interest3 i n setting aside an area as a possible spacepo~t.
3. Kenya, because cf the existing San Marco f a c i l i t y .
4. Sri Lanka, which has a l a t i t u d e advantage over India.
5. Indonesia, which has been investigating the f e a s i b i l i t y of a launch s i t e
on Sulawesi 'Cef ebes) 3 .
6. Ecuador.
1.3.8.2.1 Sea Route D i r t r n c e s and Transportation Costs
where C i s t h e c o s t o f s h i p p i n g f o r a s i n g l e 5 GW
SPS, k, i s the d i r e c t cost/kn, v i s t h e average speed ( i n kn/h) o f the
t r a n s p ~ r t , k 2 i s t h e rcvefiue which wocld be obtained f r o n one hour of
o p e r a t i o n of the SPS, and x i s the s h i p p i n g d i s t a n r e i n k i l ~ m e t e r _ The
.
second term i n t h i s equation represents t-evenue:. l o s t because t h e time spent
i n t r a n s i t from the f a c t o r y t o t h e ladnch s i t e delays i n i t i s 1 o p e r a t i o n of
t h e SPS -- i n o t h z r words, i t assumes t h a t t e r r e s t r i a l t r a n s p o r t a t i o n 1 i e s
on t h e c r i t i c a l path i n the PERT c h a r t governing l ? l ~ n c hand assembly of t h e
system. T h i i assumption w i l l probably be t r u e f o r an e t i c i e n t operation,
wi:h minimum rnoverrients i n and z u t c f warc-houses.
r--? Liberia
;c1w
I
...................... NYC
.................................. LA
---------------- London
.................................................. Tokyo
I------------------------------- London
t ...................................
S r i Lanka
...........................................
Tokyo
t ................................... London
LA
i- ....................... Tokyo
/.....................................................
i
Indonesia
LA
NYC
I
------------------------------..----------------
London
1- - - - - - - -- - - - Tokyo
I
Ecuadc:
1- - - - -- -- - -- NY C
1--------------
I
LA
I
I-__---___------_--------- London
I ------------------ -----------------
French Guiana
------ .---NYC
Tokyo
---*------------------- -LA
....................
London
-_------------------------------------.,--------
I Tokyo
Sea Ai r
1
Mass t o be shipped 50,000 m e t r i c tons 50,000 m e t r i c tons
Average density of cargo 0.3 gmlcm
3
N/ A
Shipping cost $0.01 6/m3/km* $0.27/ton/km**
Average speed 28 kn/hr 830 km/hr
SPS revenues
(5 x laG kwh/hr @ 3Clkwh) $1 50,00O/hr , $150,00O/hr
kl
$2670/km $13,50O/km
k2/v $5360/km $18l/km
(kl + k2/4
-
$8030/ km $13,700/ kin
!
Estimates o f the costs o f shipment by sea and by a i r are given i n
Table 1.3.8-1, and the cost as a f u n c t i o n o f distance i s shown i n Fig. 1.3.8-3.
I t i s i n t e r e s t i o g t o note t h a t , f o r shipment by sea, l o s t revenues due
t o t h e time i n t r a n s i t are the dominant cost, and t h a t i t m3y be cheaper
t o s h i p by a i r t o c l o s e r s i t e s than by sea t o d i s t a n t onss.
~ BE0 c i r c u l a r v e l o c i t y .
where vS = ~ ( r ~ / r i )s 'the
Figure 1.3.8-6 Fixed vs Anytime Departure Mass Conparison Arc Jet Propulsion
1.3.8.3 R a n k i n g o f P o t e n t i a l L o w - L a t i t u a e Launch S i t e s
I. E x i s t i n g launch f x i i i t i e s .
1. French Guiana (1): The Kouru range i s the best developed equatorial
launch f a c i l i t y .
2. L i b e r i a ( 3 ) : No e x i s t i n g launch f a c i l i t i e q .
3. Kenya (2) : The San Marco platform.
4. S r i Lanka (3): No e x i s t i n g f a c i l i t i e s .
5. I n d ~ n e s i a(3): No e x i s t i n g f a c i l i t i e s .
6. Ecuador ( 3 ) : No e x i s t i n g f a c i l i t i e s .
GNP C-NP
Population
(2
Nation Ranking (mil lions) $u.s!' ) per c a p i t a Education
( b i l l ions) $U.S.
France
U.S.
Notes :
(1 ) Converted from l o c a l currency a t f r e e exchange r a t e (1976 data)
( 2 ) This i s the percentage of the population which i s e n r o l l e d i n secondary
and t e r t i a r y education.
(3) French Guiana i s an overseas d6partement o f France, w i t h a population
o f o n l y 55,000. B r a z i l i s a more 1 i k e l y source o f personnel.
(4) A1 though Indonesia's per c a p i t a GNF i s low, the country i s l a r g e enough
t o have s i g n i f i c a n t i n d u s t r i a l resources. Singapore and perhaps
A u s t r a l i a are sources of s k i l l e d workers.
V. -
Availabi i i t y of Downranqe Trackin:; S i t c s (continue:!).
5 - Indonesia (1): Excellent.
6. Ecuador ( 2 ) : Good, although mountain and j u n g l e t e r n i n may cause
difficulties.
VI. -
Cl irnate: A v j i l a b i l i t y o f Hiqh Mountains
Conclusions
Rectenna failure r a t e s reported under system failure r a t e s and e f f e c t s tasks under paragraph
1.o.
The siting analysis employed a manual m a p search employing aeronautical charts, contour
plots, and road maps.
1. Identification of promising a r e a s
2. Check for agreement with ground rules
3. Check f o r f i t of 5,800 mw rectenna
4. If f i t o.k., 5,000 mw assigned
5. If 5,000 mw did not f i t , 2,500 mw was tried
The relative sizes of 5,000 megawatt and 2,500 megawtt s i t e s a r e ilustrated in Figure
1.4- 2.
It was found beneficial t o have available in t h e inventory two sizes of receiving antenna.
The two sizes utiilzed correspond to t h e power transmission link capacities, 2,500 and
5,000 megawatts. If both 2,500 and 5,000 megawatts receiving sites could b e employed,
t h e total amount of power t h a t could b e sited was much g r e a t e r than t h a t for either s i z e
of receiving antenna alone.
D180-25037-2
BONNEVILLZ POWER 25 27
ADMINISTRATION
MID-CONTINENT AREA 51 34
POWER POOL
SOUTHERN CALIFORNIA 8 9
EDISON
TOTALS 84 70
1
3 10 L
Z
Ln
0 10 20 30 . 40 50 60 70 00
EAST-WEST DIMENSION IN KH
A number of sites in each utility region was selected at random for closer investigation of
slope and other features which might presumably cause rejection. In general, most of the
sites were quite f!at. That is, tiie average slopes were less than 5 parts in 100; however,
most of sites had small regions of local slope which might be considered to be excessive
(slopes of 30 degrees or more).
As shown in Figure 1.4-3, the microwave beam frorn space ultimately falls on some
ground area. It is possible in this concept to locate rectenna panels so as to receive
all of the beam area even in regions of very extreme slope, but diffraction effects may
cause some loss of performance for panels positioned fan downbearn from adjacent panels
as a result of steep slopes. Consequently, it appears that rejection of cites on the basis
of slope mut be decided individually, with economics as the criterion.
Conclusions. This siting effort indicated that, in the three utility areas investigated,
"potential" sites exist to more than fill the requirements for electrical power for those
regions in the year 2000. Due to the potential of excess sites, it might be possible to
feed energy to the northeast from rectenna sites in the north central area, using modest
~ntertles. The benefits of having two rectenna and SPS sizes (in this case 5,000 and
2,500 megawatts) were obvious. Far more "energy from space" can be sited by having
two sizes
SOUTH -
Task Description. For purposes of analysis, t h e ove;all project is divided into tasks,
which a r e c o n ~ p l e t e dIn ser !dl fashion, i.e., :asu 1 must be comp.ete before task 2 c a n
be started, ctc. Within each task, all jobs a r e conducted in parallel, paced by t h e "slowest
n.dchinen In t h e task. This is a simple artifice t o assure that we do not try t o place
superstructure into f o ~ t ~ n not g s yet dug, or make panel assembiies without enough p a r t s
yet made. Tark allocations a r e given in Table 1.4-2.
Figure 1.4-4 shows a preliminary concept for a 10 krn by 12 In1 rectenna site. The east-
west road is about 7 krn, from stag!ng a r e a t o t h e w c i t lateral. The e a s t and west lateral
(riorth-south) roads a r e taker1 t o be 6 krn long, and t h e center lateral is 3 kni long. Based
on these distances, w c have used, as a n eyebali average, a n average travel d ~ s t a n c e
for all constri!ction 'move' tasks an on-road distance of 10 urn, and off-road move distance
of 1 kn;. This is a l r ~ ~ oisetr t a ~ n ! ya high estimate, and the total hau!ing cost for the
constructlorl T L ~ IS
S ~or11y $24 million.
TASK OlYE - iNiTUC SITE WIEPARATH)(V
(3 W T H S . 239 WORKERS)
STAGING
AREA
Except for grading and graveling roads, a > dremoving t r e e s and underbrush, nothing i s
done t o the site. W e will be able t o build a rectenna wherever we can drive a bulldozer.
Note that all three concepts impose substantial lifting forces on their supporting s t r u c t u r e s
Figure 1.4-7 shows our baseline superstructure. We need t o use several metric tons
of mass t o avoid upward loads on our footings.* Given that we must use tons of materials,
the cheapest choise is concrete reinforced or prestressed as necessary t o support the
applied loads. Using an automated factory that is shown below, these piers can be made
for about $26 assumed for reinforcing bar. No stress analysis has yet been done t o support
these estimates.
*The mass required t o resist lift loads appears t o have been overestimated by as much
as a factor of 10. This is being reviewed and will be corrected as appropriate in the
detailed GE report.
CLEARANCE FOR
TERRAIN. TREE
STUMPS. ETC.
This is a good place t o mention t h a t t w o kinds of arches a r e used; "regular" and "special."
The "regular" one was shown earlier. It supports t h e rectenna panels with t h e lower edge
t w m~e t e r s above t h e ground. This is not high enough t o permit trucks t o pass under t h e
i - e c t e n ~ aon t h e north-south roads, s o "special" arches a r e used over north-south roads t o
increase t h e clearance t o four meters.
GROUND PLANE
0IK)LE AS*.
PANEUFOREPLANE
497
Figure f.4-?@A d &sting fact~ry
Figure 1.4-12 shows the rectenna panels being delivered t o the field and installed.
The six meter width of these panels was derived t y using a double width version of t h e
design described in t h e final briefing. This width is adequate t o provide a roadway between
rectenna rows of a b a f t four meters (13.12 feet) width and a clearance of over five meters
(16.4 feet).
A "grass-roots" (detailed estimating) computer cost model was developed for the cost
analysis. At the time we started it, i t was not apparent which cost elements would
be major and which would be minor. So i t seemed that the best approach would be t o
divide the whole job irrto a number of pieces small enough t o be amenable t o analysis,
and for which we could develop an opinion about t h e cost of doing t h a t job, and then
t o estimate ail of t h e kinds of costs involved in doing each small job thousands or m i l l i o ~ s
of times.
The input data is stored in several data files. It takes over 1675 pieces of input data
t o run the program, and they must be in order, formatted exactly, 2i1d so on. However,
using a text editor, any piece of data can be quickly changed.
The model flow is shown in Figure 1.4-14. The cost matrix has 12,375 cost bins to hold
intermediate results. Most of the bins are empty, since only a few jobs a r e active at
any time, and not every task generates every category of cost. This matrix is printed, so
that detail data is available t o permit trade-offs t o be done a t any level of cost accumulation.
Figure 4.1- 12. Panel Installer
READ DATA
FILES
PACE TASK
TO SLOWEST
COMPUTE
ALL COSTS
Cost resuits a r e summarized in Table 1.4-3 and Figure 1.4-15. The t o t a l SPS ground
system c o s t is about $2 billion. This cost excludes land and development of t h e r e c t e n n a
and associated special purpose equipment.
-
TYPE
- RECTENNA
- -
SENSITIVITY
MATERIAL:
DIODES 4C/DIODE
STEEL IN PANELS 15.B L(g/mJ
ALUMINUM BUSBARS 1.02' I0.6') 166.000 METRIC TONS
STRUCTURE
CONCRETE I6 r 10 METER MODULE) 9 METRIC TONS/MCDULE
STEEL IREINFORCING) 300 Kg/MODULE
(100 Kg/ARCH;
3 ARCHES/MODULE)
lA8OR
CAPITAL COST (TYPICAL MACHINE LIFE IS
5 YEARS. COST TO PROJECT
ONLY DLiillNG USE)
MISCELLANEOUS
TOTAL RECTENNA COST
DEVELOPMENT COST1
SITE PREP
SITE PREP
LABOR EQUIPMENT
Figure 1.4-15. Rectanna Cost Distributfon
1.IManagement and Integration
A comprehensive study of management and integreation was not conducted. A preliminary
examination was made of mission operations; this item was assigned WBJ //1.5.1.1
-
MISSON OPERATW COllMANO AND CONTROL THE COMMAND AND CONTROL Of M -I
OPERATIONS mls INCLUDES THE RECEPTION AND INTERPRETATION OF STATUS DATA
(PREDOMINANTLY TELEMETRY DATA) TO DETERMINE ANY NECESSARY COYYANOS AND THE
IMPLEMENTATIONOF THESE COMMANDS. ORBITAL AND TRA#ECTORY TRACKING IS ALSO IWCLU#O
I N THIS TASK
-
COMMAND AND CONTROL CENTERS CFNTERS WHICH HAVE BEEN DELEGATEDAUTHORITY FOR
COMMAND AND COhlTROL OF SELECTEL .dISSION OPERATIONS
-
MISSION CONTROL CENTER (IICC) THE COWMAN0 AND CONTROL CENTER WHKM HAS CENTRAL
C - ~ D AND CON1ROL AUTHORITY FOR MISSION OPERATIOlYS
1. U
UWC
H AUD RECOVERY COWMAN0 Am) CONTROL CENTER
3. This central organization will probably be a matrix type, i.e., although one
element may be responsible for all LEO base operations, i t may have t o
depend on another element for recruiting, training a r d scheduling the crews
which wiii operate t h e base.
E. Table 1.5.1 1-3 is a sumnlary 01 t h e command and control functions defined and of
their assignment t o the control c e n t e r s using t h e ground rules listed above.
-
LAUNCH AN0 RECOVERY SITE
MLLV M D PLV -
bAUNCH AND RECOVERY C6C CENTER
PREPARE AN0 LAUNC14 VEHICLES, --- MIOCOURSE
LAUNCH AND RANOE COOnDINATlON
CONTROL TO L I O
- MLLV;
-- RENDEIIVOU8
--
LIOB S E 0 BASE C&C CENTER PLV; O W
HLLV AND PLV u- BASE AND SPS MODULE CQI8TRUCTIW COORDINATION
- O N
EOTV --
MANAQEMENT
HLLV; YLV; OTV
-- OM INDROWTOR
LAUNCH COORDtNATlOlV
C W R I CONTROL
- DOCKING AND UNLOAOlNO
PREPARE AND LAUNCH VEWICLL8, - EOTV
-
UPARATIOW AND COIYTROC
'&I
o
LAUNCH PREPARATION AND LAUNCH
-
L I O BAS@
TRACKINO, 8TAf IWKEEPINO, CCWJTINOIWCV
ROlOLUIION
8
8
--
3
MAINTENANCE VEHICLES
PREPARE AND LAUNCH VIWICLI8,
TRACKING, 11AtlONKICIIN0, CXmtlNOENCY
RESOLUTION
CREWS AN0 PAY LOAD8
-
DOCKING AND UNLOADING
WERATIONAL SPS
PIRFOCIMANCI --- C&C, TRACKINO, 8TAtlONKIEPINO
MWER SUBGVSTEM CbC
ECLI?SE SCHEOtJLESIANTENNA POINTING
- MAINTENANCE VLHICLI8 OOCKlNO AND LAUNCH
O r b ~ t a 8ase
l CjrC - It is anticipated t h a t t h e orbital bases (which may be
considered as a type of spacecraft) u i l l be autonomous, however, t h e Mission
Contra1 C e n t e r may be called upon t o perform tracking, stationkeeping, resposi-
tionicrg, anomaly support and other selected functions in a back-up mode.
1. One facility which includes the Transportation Vehicle C&C and Space
Traffic Control function categories. These a r e t h e first two facilities needed
Figure 1.5.11-2. C&C Center Relationships to Major System Elements and to Each Other
0 1-25037-2
during the construction of LEO, CEO and the first SPS modules. Also as
shown in Figure 1.5.1 1-2 they have the busiest interface.
2. A second facility which contains the other fmction categories Orbital Base
C&C, SMIT ChC, Communication Satellite C&C and the Operational SPS
C&C. Facilities for these functions can be constructed in sequence as needed
by the program.
3. As the number of operational SPS increases, capability for their control will
also be provided a t the first facility with the possibility that a third facility
will be added which is dedicated t o this function. Another possibility will
exist when all SPS have been built, i.e., the SMIT control facility may be
converted t o this function also.
I. Various options to the strawrnan C&C concept will be evaluated during Phase 11.
These options are summarized as follows:
1. Assign C&C responsibilities differently among the local control centers and
the Mission Control Center. Examples of this are:
Each Rectenna C&C Center have complete responsibility for C&C of its
corresponding SPS instead of a power monitoring function with communi-
cation t o the MCC for SPS C&C a s in the strawman concept,
2. Eliminate the Orbital Base C&C function category in the MCC and make the
bases completely autonomous, without ground back-up.
The options enumerated in Iten? I. will b e evaluated during phase I1 using the
following criteria:
The SPS program is present!^ in a feasibility study and evaluation phase. Many
program s t e p s and activities will be needed to achieve successful program function.
One of t h e tasks in this study was directed to identification and planning f o r
t h e potential f u t u r e of a n SPS program. The analysis followed t w o converging
paths as illustrated in Figure 2-1.
21 PROGRAMMATIC STUDIES
C,
\1
HARDWARE
\1
\
\!
\\ / PROGRAM .
OPTIONS
\I
. 24 18
ENGINEERING SllBSYSf EMS AND SYSTEMS SPS DESIGN APPROACHES INITIATE FULL-SCALE
DEVELOPF.IENT ENGINEERING PERFORMANCE VALIDATED; PREFERRED DEVELOPMENT
AND COST AND PRODUCTION COST; APPROACHES SELECTED:
VERIFICATION ADEQUATE BASIS FOR SPS COST CONFIDENCE
SPECIFICATIONS
I
AND RADIATION'TECHNIOUES
--
Y E . % R s I I ~ Z 1( ~4 1 5 1 6 1 7 1 8 1 9 1 1 0 ~ 1 1 1 1 2 ~ 1 3 1 1 4 1 1 5 ~ 1 6 ~ 1 7 1 1 8 ~ 1 9 1 2 0 ~
--
DTA DESIGN a FAB
DTA BUILD
DTA TEST LEOfGEO
PLANTS I.2 GW/YR
COMMERCIAL
PRODUCTION
PLANTS
20 GWNR
CONFIG. FREEZE
DPROTOTYPE SPS DESIGN 8 S/S TEST
PROTO FA8
LAUNCI-! & SPACE ENGINES DEV PROTO
BUILD & TEST
PRODUCTION SPS DESIGN &TEST
VEHICLES L PHOTO BASE DES & TEST
INCLUDES CERTAIN FLIGHT
EXPERIMENTS
PLANTS FOR PRODUCTION OF SPS
HARDWARE, E.G., SOLAR ARRAYS
PROTO BASE BUILDUP-
The major iuncling requirements arise from development from space bases
and heavy lift launch vehicles. Some cost dcferrai options exist t o reduce
'. e peak funding tci a degree, but their benefits in a n ecoromic sense a r e
quite dubious unless i t is expected t h a t t h e completion of the prototype
would result in a decision not t o proceed with cornrnercialization of SPS'.
If commercialization proceeds, then t h e economic cost of these deferrals
tends t o exceed their value.
DEVELOPMENT
1 - TECHNOLOGY
llNCLUDES PRODUCTION SS
DESIGN 8 TESTJ
CALENDAFI YEASS
V S M I
Table 24 Cost Deferral Options
i A
2 2 TEST-HARDWARE ANALYSlS
COMMERQAC
OEWO(UtTRATOR
CAN bEGlN
v nrrv
A8'AILULE
- PLATCOW v w~rc~~uncrr
-
I
-b~-qrr*
ntwlnro)
mkw-
- M m nrhnr I-. a n . -1
.uslOW;'%-
.OrbrtrnHkr~lm
wonr m n o nv
CAN U C ( Y
y S U A R R A V ?O% WE. 21 hW
CHAMBER --
;r
Y-F -r
-7- GROUYO TO GROUND TEST RAWGE
Shown in Figure 2-7 is installation of a 3.0 meter subarray into the transmitting
g r c i p of t h e microwave ground-to-ground test range. The structure of t h e micro-
wave test range transmitter supports the subarray elements and allows for tilt.
In test, a tilt angle might be used such that the difference in distance from t h e
subarray which is closest t o t h e rectenna panel and that which is furthest from
the rectenna panel would be the same as t h a t anticipated in a full scale solar
power satellite. That is, the angle would be much larger than the angle in a full
size satellite but the distance difference would be the same. Trunnions a r e provided
fcr this tilt. The framework includes power distribution, phase control distribution,
etc. T k run of coaxial cable or optical fiber between subarrays and t o t h e central
reference subarray, might use coils s o as t o equal the total distance involved
in phase distribution aboard t h e full size satellite.
Figure 2-8 shows two potential methods of utilization for a "large aperture test
satellite." On the le,. a test array such as the 30 meter square array of 100 sub-
arrays, shown previously, transmits t o space under control of the 9.0 meter dish
of t h e large aperture test satellite. That is, the large dish on the satellite provides
the pilot beam for phase control of the test array. The test array was provided
with trunnions t o permit tilt t o the required near-vertical orientation. Operation
could be accomplished through ionospheric s t r a t a heated by a transmitter such
as that Arecibo. If frequency scaling was employed, and the power level at that
transmitter was increased, andother possible utilization is t o a c t as a pilot trans-
mitter for a large array of SPS similar transmitter elements placed horizontally
on the ground (as shown on the right). Here the test array is sufficiently large
t o directly heat the ionisphere without frequency stalling.
I
POTENTIAL APPROACHES:
@ SEPARATE HEATING
'SMALL ARECIBO
TEST OR (4 XTEST
'LARGE' ARRAY
104 M2)
ARRAY PLATTESVILLE
. I
F&ure 2-8 Ground-To-GroundMPTS Test System
D 180-25037-2
larger antenna would be used f o r pilot control of a ground transmitter subarray
group. The large aperture s a t e l l i t e would be launched t o geosynchronous orbit
by a s h u t t l e and inertial upper stage. After arrival in geosynchronous orbit t h e
cannisters f o r t h e extendable booms would be swung out and then t h e Sooms extended
t o l o c a t e t h e transmit 'receive elements. The satellite would include solar power
supply, a t t i t u d e and stationkeeping control systems, command and control systems,
etc. It would be advantageous t o have a design lifetime of serveral years for
this satellite. The transmitter tubes used for t h e 2.0 m e t e r dishes might b e 10
t o 20 u a t t s traveling wave tube of t h e type currently flying in many sateilites
and space probes.
The large a p e r t u r e test satellite could also provide years of stable orientation
in geosynchronous orbit. In t h e concept shown in Figure 2-10, samples of potential
SPS components %bouldbe extended and deployed aboard t h a t satellite by a n accordian
pull-out and lanyard system. These samples might include solar cells of various
types, potential structural elements and materials such as composites, metals,
plastics, etc. After t h e desired exposure period t h e samples would be drawn
within t h e reentry body and hatches closed. The e n t i r e system, including t h e
solid rocket return motor woulc! be spun up upon a turntable; a f t e r reaching t h e
required spin rate, springs uou!d be used t o kick t h e svsterri f r e e of t h e large
aperture transmission sateilite and achieve a save separation before firing t h e
solid rocket motor. Approximately 5 1/2 hours l a t e r the reentry body would e n t e r .
t h e earths' ;t:mosphere. Here i t would be recovered using proven space recovery
techniqaes. The SPS candidate inaterial samples could then be tested t o determine
t h e resultant degradation due t o their exposure. During t h e exposure period in
space, analyses should have been carried out t o predict degradation mechanisms,
ground test including radiation exposure should have taken place, s o t h a t t h e
space operation provicics 3 (:orrelation and calibration of t h e ground test program.
(Construction)
MAJOR FLT. PROJECT
INTEGRATED OPS.
I 3. SOLAR ARRAY
OEPLOYMENT
--- - - -
I I COMBINE WlTH BEAM
SORTIE (SMALL SCALE)
INTEGRATlOrJ WlTH "BASE"
5. MODULE DEMONSTRATE
IIJDEXING
(Maintenance)
KLYSTRON
CHANGEOUT
I NEUTRAL BOUYANT
I POSSIBLE CHANGEOUT
ON SORTIE
"ADVAIICED" OTA COULD INCLUDE
GANTRY
PRIME RESULTS.
BEAM BUILDER OPS
RAIL ATTACHMENT
0 TIMELINES
ADDlTlONAL RESULTS:
LARGE OBJECT CHARGING
I N EARTH'S MAGNETIC FIELD
0 BEAM DYNAMICS
END FITTING
INSTALLATION
BATTERY PACK
ELECTRlC THRUSTER
POWER EXTENSION
,PACKAGE
SUBAARAY PERPORMANCE
r SIDELOBES
HEAT REJECTION
(SAME KW/AREA 45
FULL SCALE)
-START-UP
TUBE START-UP
PLUME FORM
FROM THRUSTERS
r [SUBARRAY AND THRUSTER MODULE ARE NOT
SIMULTANEOUSLY ERECTED, AS S W N )
A L f ERNATIVE APPROACH:
POWER WITH OSM.
300 k W
Not necasbr~lyrrlrvmt
to SPS
0neshu"Ir
LARGE POWER e Low Earth orbit M)O kW (MI for 2.000 kW)
MOOULL
Mi~rowwrtr~srnissima l m n a
DEVELOPMENTAL TEST
ARTICLE (DTA)
PROOF OF CONC
100 -
DOLLARS
lid BILLIONS FAClLlTlZATlON
(INCLUDING HLLV)
2 5 GW WITH SHUTTLE
DERIVATIVE AND MINI1.1UM
(1 TO 10 MW OUTPUT)
DTA (NOGROUND OUTPUT)
; ; ; ; 4 io l ' 2 i3
GW ONLINE
The "commercial demonstraror" system was analyzed in part 111of the previous
J S C study (Contract NAS9-15196). It is shown i n Figure 2-18. The power ~ u t p u t
level from the microwave transmitter was 185 megawatts. I f approximately
10% of the project budget b a s invalved with ground reception (rectenna) about
1 megawatt of useful power would be produced. The system is also sufficientl)
large to use fall size SPS power generation bays, id! iength solar cell strings,
etc.; i t can be made, essentially, of full scale SPS components.
"TRANSMITTER" A
/- SOLAR ARRAYS:
78.000 CELL STRING LENGTH (FULL)
VOLTAGE (FULL)
(TWO 7.5 M STRIPS JOINED TO OBTAIN 15M)
THRUSTERS
(SUBSCALE)
ROTARY JOINT
USES FULL SIZE
STRUCTURE (IDENTICAL)
:30M TRANSMITTER
USES FULL SIZE
SU3ARRAVS, HAS
BUS BAR SYSTEM FULL CENTRAL
(SUBSCALE. SAME hEAT OISSIPATION
TEMPERATURE)
a Given a successfu! conclusion of this program phase, the next program phase
should be devoted to engineering developments at the subsystem prorotype
level, and system cost verification and risk management planning. Examples
of the cost-risk activities include (a) C~nstructionand operation of prototype
or pilot production lines for hardware with identified cost risks; (b) Updating
af space transportation system designs and re-evaluation of transportation
costs in light of Shtittle operating experience; ( c ) Definition of an integrated
development plan with the necessary system design, hardware option and
schedule flexibilisv to ensure successful risk management. Successful conclu-
sion of tkic program phase would result in high confidence in system cost
and economics.
These program phases can be overlapped to a greater or lesser degree; the amount
o i overlap n,ust be determined by a tradeoff of risk versus need.
The next pt,ase of the present study, now underway, will include a major task
devoted t a defining the content, schedule, and cost of the research phase in considerable
detaii, with later phases defined in somewhat less detail. Analysis of the d e v e l o p
ment phase will include a cost benefit assessirlent of the appropriate s i z e for
an SPS prorotype.