This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected. Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies. Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue. The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.
.
Despite successful dockside tests at Groton in November 1959, a concerned “Red” Raborn (right) kneiv that a lot could go wrong as he awaited the first tmderwater launch attempt from the George Washington off Cape Canaveral in July 1960.
Operations research played an important role in the formulation of the design characteristics for the Polaris fleet ballistic missile system. As chief scientist and senior operations research advisor to the Navy’s Special Projects Office, under the command of Rear Admiral William F. Raborn, I was privileged to be deeply involved. The program was unique in its approach to operational requirements, in its comparative freedom from administrative and fiscal constraints, in the time span within which a major weapon system was brought to operational capability, and in the close cooperation among military, industrial, and academic groups.
My involvement began with an assignment in the early 1950s as Operations Evaluation Group scientific analyst to Op-51, the Assistant Chief of Naval Operations for Guided Missiles, Rear Admiral John H. Sides. While he was still a captain, Raborn had served a tour as Admiral Sides’s chief of staff, and he and I had on several occasions discussed guided missile operational requirements and evaluation testing. During this period, nuclear ballistic missiles were assuming a high priority in discussions of U. S. military strategy. In accordance with a March 1948 agreement on service roles in nuclear warfare, intercontinental ballistic missiles were reserved to the Air Force, but the Army was allowed to develop the 1,500-mile intermediate range Jupiter missile in competition with the Air Force’s Thor. The Navy at that point had no formal role in ballistic missiles. In early 1955, the Killian Report commented favorably on the sea-based deployment of some intermediate- range ballistic missiles, and the Navy was directed to design a sea-based support system for Jupiter.1 As a result, the Special Projects Office was commissioned in December 1955 under Admiral Raborn, with high priority and a broad charter to assemble the people and resources to develop a fleet ballistic missile system. Early the following year, he arranged for me to be assigned to the program. I joined full-time in June 1956 after completing another project.
A major step toward Polaris came with the establishment of Project Nobska (named after a point of land near Woods Hole, Massachusetts). The Navy had been greatly impressed with the potential of the USS Nautilus (SSN-571) and was concerned over the prospect of Soviet nuclear-powered submarines. Project Nobska’s mission was to consider ways of defeating a Soviet nuclear submarine threat. One suggested approach was to make the U. S. submarine threat so strong that Soviet expenditures would be diverted
c by
from pro-submarine to antisubmarine w ‘ ^
combining the nuclear submarine and the l°n£^j a|. ballistic missile. The Special Projects Office, rjne
ready made cursory studies of building a sand around a few Jupiter missiles, but missi e safety considerations made the prospect qut0o tractive. The basic features that made the JL1P ar0I1iic large for submarine deployment were a heavy ^ ^ warhead and a heat shield conservatively size ^age longer range Atlas ICBM. The total front-en gj. was about a ton and a half. At this point, ^ ward Teller made a vital contribution to 0 ^ sys- asking, “Why are you designing a 1965 vvea tern with 1958 warhead technology? He Pr
die*0
drastic reductions in warhead weight for an j,js
poi1
able yield and quoted historical data to sU^t£C| to prediction. His remarks, which were reP ^56, Admiral Raborn and his staff on 4 Septem c caused a radical reorientation of our thin * geVcf'
March (with the arrival on the staff of Captio
ing Smith), there had been consideration 0 gjjd,
nate solid propellant motor for the Jupiter pfope*' to overcome the safety objections to bqul ^ok
lants on board ship. But proven propellant ogy forced the designers (Lockheed and Aet I
et)
to*
l»rge:
6:1 staging concept, resulting in a motor eye ^ re- and heavier than the Jupiter’s." The anticip duction in throw-weight, to use later jargot ' allow a two-stage solid rocket with propella11^.^ d1* lations, case-weight fractions, and sizes -aC'
known state of the art.3 Admiral Raborn s lf j Jtep tion was to take the necessary but unprecec tgfroP of requesting written estimates from atomic ,_[0, tories at Los Alamos and Livermore o 1 £ (e- weight ratios for the next several years. b)!
quested predictions were then officially Put the Atomic Energy Commission. Supp°rC $eC Polaris concept grew quickly after that, a
tht
on
retary of Defense officially approved the Pl0^aS t)iii[
10 December 1956. An attractive feature
tht
Polaris would be a purely Navy system, orati°n' organizational complications of Army co a ^ [b1 So by mid-December 1956, the Navy r0menC.
elements, including an existing £°veerarionil industrial team, to design and bring to °P readiness a solid propellant missile, bas
would be under «?
clear submarine. The team
le*0
CaP
ership of an outstanding technical director, .
Smith. The next order of business was to 0 jevd
management structure to control the techn1^
to
(fP
‘For footnotes, please turn to page 59.
opment. Dr. L. T. E. Thompson, — _ ^ a
ing figures of naval ordnance, was serving tjtb sultant to Admiral Raborn, under the r0poSe
of “Assistant for Engineering Services.
56
proceedings
/ Mar°n
l9e°
Caa'ted by ^,rnent of a Polaris Steering Task Group,
tr,tatives rflptain Smith and including senior repre-
(°ne per te°K taCp op dle participating contractors
Sufigest fbj” n'Ca^ branch). My contribution was to
en'0r £uy3t [he selection criterion should be “the
f°rk-” Mor^f ° W'^ ^'red ‘^ this thing doesn’t
t'C'etlt autho °rmally- the rnembers had to have suf-
t|Q,ls with0yrity t0 comrn't their parent organiza-
The steeri C°nsidt'nS higher management.
7 fariltask group convened for the first time .. f“nuarv las-, , ^ . .
an
c°nfi
ed ^ iccmiy vcmcie anu payioau were
^unfr Catec^ guess. So it was clear that we were 'Ptirrium tCCa W'tb a standard request to design an ^‘rerner)tSystern meeting a fixed operational rentable • ■ . tasb was rather to set minimum ac- H total nit*a^ operational performance, expecting UlK, ^stern capability to improve with time to
hot
Qt-*-. nuu t*.c auu caujv. a ica-
I °Ur kn 0l|Int damage when it got there. Based ^2, tb- . e<age of geography and weapon effects in 'le^aton lrnPbed about 900 miles range and a half- y,ed- Assuming that the reentry vehicle
°n
januo O* ~ i' wiuviitu uit mat Lixiiu
"t5te(l b0xe^ ^57. It was confronted with a set of 'i Var*0usS rePfesenting major system components yll at gj tafes °f realization. A nuclear submarine t'CUt> WastCtr'C ®0at s shipyard in Groton, Connec- t|Qn. ava'iable for the insertion of a missile sec- |C^cti0n °f the section was dependent on the
^ the pref t*le number of missiles per submarine, ^reast wirlc1^^ storage concept was for missiles two si2etlln t*1e submarine pressure hull. The mis- j>rtssure ^W.as Prn'tcd by the 2-inch HY 80 steel °n§ bef0 °^the submarine, which had to be fixed C^atacterJb C^e rocbet was specified, and the exact St'll a„ , lcs °f the reentry vehicle and payload were
th ^ ^___________ f ^ _
^ ^uUy r^ystem capability to improve with time ... Ve ^t tkSp°ns*ve t0 the need. In very crude terms, Moscat C^e Solaris missile had to be able tc ()riable „ C°w from a position at sea and cause a rea
people fulfilled their prophecies, the missile designers estimated that they could stay within an acceptable submarine pressure hull diameter. We estimated about four years for the submarine conversion, perhaps two years for missile development, and a year for the specific warhead design. Making these elements come out roughly comparable in maturity of development at completion was one of the real challenges for management of the fleet ballistic missile program.
For the next three or four months, subcommittees of the steering task group made design trade-offs and choices. One obvious consideration was mission duration. The Navy had demonstrated that long cruises, completely submerged, were feasible. A limit of 60 days was proposed, based on the food storage capability of the ballistic missile submarines, but a number of other human factors were not far behind in importance. In the beginning, we were directed to plan for operation from continental U. S. bases only. We were well aware that submerged speed is both costly and noisy and that operations in the patrol areas would be at very low (quiet) speed. But limiting the speed to less than, say, 10 knots would mean that most of the 60 mission days would be spent in transit, out of range of targets. So we accepted the power plant and speed of the existing nuclear attack submarines.
%
ngs /
March
1980
57
Another issue was the number of missiles per submarine. On the one hand, the submarine was the most expensive part of the system, so increasing the number of missiles per submarine (with a fixed total missile inventory) would reduce total system cost. (We didn’t anticipate at that point that system cost per megaton would become a Department of Defense measure of effectiveness which would penalize the submarine in competition with Air Force strategic delivery systems.) On the other hand, operational readiness and considerations of flexibility and survivability dictated that the warheads at sea be spread over a reasonable number of ships. Our colleagues at the Bureau of Ships produced a number of preliminary designs, with missile compartments varying from 4 to 32 missiles. Since the missile compartment represented only a small fraction of the total submarine displacement, the principal effect was on hull length. Some submarine operators felt that the 32- missile submarine was too long for easy maneuvering. In addition, the longer missile compartment complicated the transmission of navigational data to the individual missile guidance computers. It would be nice to pretend that the final selection of 16 missiles per submarine was the result of a highly analytical process. In fact, it involved a sort of opinion poll of ship designers, analysts, and submarine operators. At least one submarine captain voted for four missiles (at one stage there was a secret ballot on the number) “to keep the answer reasonable.” In any event, it turned out to be a satisfactory decision, particularly when we got around to demonstrating ripple firing of missiles to reduce launch phase vulnerability.
As one would expect, the analyses of warhead yield, delivery accuracy, and target system involved a considerably more impressive analytical apparatus. Contributions to error at the launch involved location, preferably by self-contained inertial means without disclosing the submarine’s position and heading; knowledge of target location and the ballistic trajectory to it; and the details of reentry behavior, including expected wind over the target. The state of affairs in 1957 made it quite obvious that Polaris would not be capable of precise attacks on “hard" targets. Its mission was “assured destruction” of soft area targets, such as urban industrial com-
.fcoitf1
plexes. This also helped to reduce the are* '
tition with nuclear delivery by manne a political advantage in budgetary discussion^^ e> descriptions of the Soviet target system a isted, and these allowed us to verify r^a^ery e^°P between predicted warhead yields and e 1 fl1'5
justified the deterrent worth of the fleet * sile system. . had s0^
The guidance and trajectory people
teresting trade-offs to consider. The ba
tory between two known points on ^^^ther
can
be calculated with high accuracy by
ch r
n1
ous procedures. But the submarines laU j,esP°
was not fixed. To calculate the trajectory ( j O'
mv°lv ,f0r<
for a moving launch point would have
l oi a iauiiv.il j^wiui »■-------- . .
1957) a computer too big for the submarn^ ^
to*
,oSS‘
ibl<
precalculated trajectories from a fine £rlLlajjy L'. launch points would have resulted in *in i'1
if
cessive library of data cards. The solution [vv0 ^ tial fire control system was to combine t U(J it>*
— . — - - j ■ " ~ . . frnm rhe
proaches by calculating trajectories i j
in-
points of grid squares of reasonable size Fl
puiiuj wi ^i iu jcjuaiv. j wi iu**'*' |
terpolating by computer to the actual su . sition. The resulting consoles could be lt:t'
nt°
the
cr<
“7'' . s . n rhe customary
submarine with no more than tne cuj o>
ing of space. The exact geographic PoS' to ‘ targets was also a matter of concern, eon
j,y 1
strategic attack systems, but it was satis rQ(e
. . . _ am a . • i _ a rt i'v r 1 f"P.3
solved by the efforts of all three services
first ballistic missile submarine deploym^^
iH1
obviously des.rame jUP
Underwater launch was an operational point of view to preserv'
sim
iive
cttr
1:
nat(
Fi
'or,
H
^t
sile
sTv
svat
sub
^t'a;
also
a>ti
bro
anc
I
bfis
<ltv
chr
sub
c0fj
f.
ire
19(
rtS(
58
Proceedings
/M»r'
cl*
initially 'av's‘bifity as long as possible, but it was ^■°ckets LC^arded with considerable skepticism. had been
etwater k — experimentally launched from triej "pk f’p Ut nothing the size of Polaris had been was inherently unstable hy- ^r°static ^ ^ fissile encapsulation against hy- We*corne ^ hydrodynamic forces would be an unbare missii°rn^Cat*0n’ ^Ut hydrodynamic l°ads f°r a
|Vn6 ^aunch might be a worse condition than
unde;
ac'to<J
h;
appy
ynarnic forces
at maximum aerodynamic Q (by a
fialjy eqn ?|;iderice> they turned out to be substance tockct °ne wanted to contemplate igniting Pressed ajrlaiotor within the missile tube, so com- 'Sniti0n ^ aunch (like a torpedo) was used, with anothere air- Submarine velocity at launch was 'vatet> thenCern’ w*t^out some speed through the to° tnuchC SU^rnar*ne would be out of control, but Vate the ‘ S^Ced actoss the tube hatches would aggra- neutraiiv kStablUty- Finally- the submarine had to be Cle bu0 aUo^ant before and after the launch. Vari- final i^- tan^s were initially considered, but '‘ally with' Ut*0n Was to let the empty tubes fill parser thf ,v'atCr before closing the hatches. Concern dvn, . e‘fec arn'cs
ulator f *v wiwhwv.hu
"Ve rniss‘1f<>rn wfiich structural test vehicles (but no Cern, Und ^ Were launched. After the initial con- fo tjleervvater launch worked out well.
> that"11^3 m*ssde test vehicles, we were fortu- 0fCeX-i7 0c^fieed was already building the Air
granted high priority and considerable administrative freedom. The working conditions are hard to recreate, 20 years later, in an atmosphere of formal and elaborate requests for proposals and evaluation of responses, conducted very much at arm’s length. Even before I left the Special Projects Office, I attended a meeting at which Rear Admiral Charles Martell, then engaged in merging the Bureau of Ordnance and the Bureau of Aeronautics to form the Bureau of Weapons, said, “Of course, Special Projects is the best thing the Navy has ever done organizationally—and, of course, we must never repeat it.” But it did work, and it even withstood a Sputnik- stimulated schedule compression which changed the initial deployment date from 1965 to I960. Much of what we did is contrary to current government regulations, but it was a privilege to be part of the team and to work for a management that accepted operations research and made effective use of it.
fft of ship motion on missile launch I'^nlator f- C° ^ construction of a ship motion
-17,
> in earth
VeVle
S|C re&nt t0 S*rnidate intercontinental ballistic mis- ^-3 ri ? Vtd°cities. Under the Navy designation of *<th,s b'
sPm-stabilized, three-stage, solid rocket which the final stage was fired back to-
Dr. Whitmore was graduated from Massachusetts Institute of Technology in 1938 and received a Ph.D. in mathematics from the University of California (Berkeley) in 1941. He joined the Naval Ordnance Laboratory in September 1941 and worked on mine countermeasures. From 1942 to 1946, he was an instructor in physics at MIT, then joined the Operations Evaluation Group, an MIT contract group supplying operations research support to the Navy, where he worked on a number of problems in the field of radar and guided missiles. He served as Chief Scientist, Special Projects Office, under Rear Admiral Raborn in 1957-1959. At Lockheed Missiles and Space Co., he is currently Chief Scientist-Ocean Systems. He lives in Los Altos Hills, California.
eaj/h Jecame the means of testing integrated LMUStantia]eat S^*e^d reentry designs, leading to very tat prot.a Sav'ngs in the weight devoted solely to
ah<> becan^tl0n an odd turn events’ rfie 3lt'tude u- tbe ^auncfi vehicle for the Argus high- ^ 18h-latitude atomic test, as the only JrM altjr3,V^ ^aunch vehicle which met the payload , And so reclu‘rements.
fise t e individual elements that were to com- tVel°Pm/Stern Were drawn together, one by one, as ^ber jocj and test*ng were completed. On 30 De- ^^atine ' tbe worid s fifst deet ballistic missile f^'hissj6' tbC USS ^eorSe Washington (SSBN-598), was ]lr<-'(l her ^td' July of the following year, she 960 w Irst Solaris missile from underwater. Before
"'as
0Ver, she began her initial undersea patrol.
Sn! fleet ballistic
s a - missile system development rep-
inte r°Utstanding example of what can be done grated government-industry-academic team
'James R. Killian, Jr., then president of Massachusetts Institute of Technology, was appointed in 1954 by President Dwight D. Eisenhower as chairman of a panel to assess development potential in defense technology. See Harvey M. Sapolsky, The Polaris System Development: Bureaucratic am! Programmatic Success in Government (Cambridge, Massachusetts: Harvard University Press, 1972), pp. 18-19.
JIn developing a long-range rocket, most of whose weight is fuel, it is helpful to drop off unnecessary structure in rhe course of the flight. This leads to multi-stage rocket construction, in which the structure for each stage is jettisoned as its fuel load is burned. There is an optimum staging ratio for any given design, in the range of 2:1 to 3:1—certainly not as unbalanced as 6:1. For the solid-fuel Jupiter, it was possible to cluster six Sergeant motors around a single central Sergeant motor, which met the Jupiter payload and range requirements with (a) the largest proven solid motor of that time and (b) a "reasonably compact" design which might conceivably fit in a submarine. But it was a very inefficient design in minimizing propellant weight. Polaris and all later variants were two- and three-stage in-line rockets.
3Case-weight fraction refers to the ratio of the rocket's inert (metal or fiberglass) parts to the total weight including fuel (a few percent for high efficiency rockets). A typical long-range rocket burns radially outward from the centerline, using the unburned propellant as insulation. It can be thought of as a firecracker wrapped in tissue paper.
lgs/March 1980
59