C H A P T E R E I G H T E E N

1 7 4

TH E SH A W EF F E CT. . .It appears also that [long-term changes at the NRTS] will be influenced by technology trends,

national policy, and other factors largely outside the control or option of the immediatecommunity or the State itself.

—William Ginkel, 1967—

The ATR start-upgroup was having a bad day, and it wasn’t the first one. It was preparing tobring the reactor critical for the firsttime. As the operators rotated the con-trol cylinders, they saw that the count-rate recorders were not behavingaccording to prediction. It could mean adelay like an earlier one when the stain-less-steel coolant pipe had been acci-dentally over-pressurized. Some of thepipe, thirty-six inches indiameter, had bulged anddeformed. The pipe wasruined. Replacing it hadcost millions of dollarsand a year of time.1

In the face of this newtrouble, the team shutdown the reactor,opened the pressure ves-sel, removed the fuel,and inspected. T h e ysoon discovered that thedrive mechanisms forthe sixteen controlcylinders would notrotate on command.Each of the drive units had been

installed backwards. The problem tookless time and money to fix than the ear-lier problem, and in an earlier day,might have been considered just one ofthe routine bugs that accompany anycomplex new project.

This time, the ATR start-up problemsand parts failures came under the closescrutiny of Milton Shaw, since late1964 the director of AEC-Headquarters’Division of Reactor Development andTechnology. He sent investigators to

discover the management failures thatmust have caused the mishap, anintense process that further postponedthe start-up.2

As a former aide to Admiral Rickover,Shaw had been exposed to the safetyphilosophy of the Nuclear Navy. Aswholeheartedly as Rickover, hebelieved in accident prevention viaexcellence, quality control, and redun-dancy. He once said of himself, “Mywife jokes that when I build a dog

house, it’ll withstand aseismic event.”Although such featureshad not been absentfrom NRTS safety phi-losophy, the principlesof Site remoteness andgeographical separationof reactors had beenmajor guarantors of pub-lic safety. It was likelythat if Shaw chose toassert his convictions,the shift in emphasiswould change the com-fortable old way ofdoing things at theNRTS.3Aerial view of TRA looking south. The large ATR

building is at right of photo. Its associated cooling

towers are below it and farther to the right.

INEEL 68-3496

Loss of Fluid Test Facility tunnel entranceINEEL 79-5542

A forceful and skillful manager, Shawalso practiced his mentor’s “abrasiveinterface” style of dealing with people,making no virtue of tact even on cere-monial occasions. The president of anIDO contractor, Chuck Rice, recalledone such occasion:

After I had been elected president ofNuclear [Aerojet Nuclear], we had abig dinner for key managers in thecompany at the Stardust Motel. MiltonShaw was there, Bill Ginkel, many fromAerojet, all the way down to branchmanagers. Shaw got up and did hisRickover-type tirade on all that thesepeople in the room had done wrong.They were lousy managers, had poorcontrol, and so on.

When it was my turn to speak, I got upand listed the outstanding accomplish -ments of the group and complimentedthem on the work they had done so well.

As I walked out after dinner,deBoisblanc came up and said, “I real -ly appreciated the comments. You’ll befired, but it was nice to hear it.”

The next day there was a meeting onwhether to fire Rice or not. Shaw said,“Find out the reason for his speech.Then we’ll decide.” Someone called meand I said, “Shaw works at Headquar -ters, I work here. If we are to do well,I’ve got to invite the people who workh e re to join my part y.” I kept my job.4

L a t e r, Rice and Shaw developed awarmer relationship. But Shaw’s obser-vation of the ATR troubles reinforced hisview that a quality assurance approachto reactor operations—even test reactors

like the ATR—was the only correctapproach. Parts and systems must meetstandards, and management must assurethe standards be observed. The AT Rproblems had made a strong impressionon him. Years later, he still referred tothem when discussing the quality issuewith the JCAE.5

As Shaw considered the state of theAEC’s national reactor program, he feltthere was duplication of effort and poorcoordination among the labs. Keymembers of the AEC commission andthe JCAE supported him in this view.Shaw felt that the situation justifiedreform, redirection, and tight control.His introduction to the ATR gave himno reason to exempt the NRTS or itscontractors from this overall appraisal.Shaw’s new broom began the sweep,and the dust-up generated discord andanxiety, program changes, and person-nel dislocations. Some misunderstand-ings lasted for years. 6

Phillips’ five-year operating contract,which the AEC had renewed regularlysince Phillips began at the Site in 1951,was scheduled to conclude in 1966. InMarch of 1965 the AEC announced anew approach to selecting the next con-tractor. The AEC wanted contractorsthat intended to invest in the nuclearindustry. While Phillips had a laudablerecord—and in fact would continue tomanage reactor safety research—Phillips’ only involvement in the indus-try, aside from uranium mining, hadbeen at the NRTS. The company wasinclined more to research, not engineer-ing. It had no other nuclear involve-ment.7

P R O V I N G T H E P R I N C I P L E

1 7 6

Above. Logo for Aerojet Nuclear Corporation. Middle.

Logo for Idaho Nuclear Corporation. Below. Logo for

Allied Chemical.

By June 1965, thirty companies saidthey were interested in the $29 millioncontract. The Idaho Falls newspaperkept track of the “titan firms” visitingthe Site. Ginkel and his staff rented theElks Club for briefings and conductedSite tours for the visitors. Up for newmanagement were all three materialstesting reactors (MTR, ETR, ATR),their supporting engineering groups andzero-power reactors, the Chem Plant,the Hot Shop and other TAN facilities,most of the site-wide craft and otherservices—and 1,800 people rangingfrom bus drivers to scientists. The deal,as usual, would be cost plus fixed fee.8

The AEC liked the Aerojet GeneralCorporation. The company, which hadbeen founded in 1942 to design andbuild rocket engines, had managed aproject for the AEC and NASA i nNevada called NERVA, a joint effort todevelop a nuclear-powered rocket. T h eAEC liked A e r o j e t ’s “disciplinedapproach to engineering and quality,which Aerojet...had developed in itsSpace Program operations.” Aerojet alsohad the right kind of ambitions. It want-ed to become a major nuclear player andhad entered the field of commercial gas-cooled reactors. Managing the NRT Swould give its people the experience andcompetencies necessary to make thegrade. A e r o j e t ’s proposal was managedby Allan C. Johnson, who, after his post-SL-1 departure from the NRTS, hadlanded on his feet at Aerojet. As assis-tant to A e r o j e t ’s chairman, he and J.Bion Philipson, an NRTS pioneer thenalso with Aerojet, led the company tothe winner’s box.9

There was a catch. Aerojet had littleexperience in chemistry, an unaccept-able weakness considering the impor-tance of the Chem Plant. The AECsuggested a shot-gun marriage betweenAerojet and one of the other bidders,Allied Chemical. The two companiesadjusted their bids and created IdahoNuclear Corporation (INC), with anunderstanding that Allied expertisewould manage the Chem Plant. Alliedheld a minority interest in the company(Phillips continued to manage the STEPprogram as an independent contractor).Dr. Charles H. Trent from Aerojetbecame president and Bion Philipsonhis deputy. The new regime began onJuly 1, 1966.10

The change jolted Phillips employees,some of whom had been part of thePhillips family for most of their careers.“Suddenly, we were like an arm graftedonto a new body,” observed one ofthem. Gradually, they adapted, althoughAerojet never re-created the Phillips’style of benign paternalism. The FrankPhillips Men’s Club and the JanePhillips Sorority disappeared.11

Shaw soon felt compelled by the rapidonset of the commercial power industryto reorganize the safety test program(STEP) for water-moderated reactors.The STEP program had progressed farbeyond the NASAtests studying theimpact of ocean crashes on reactorslaunched into space. The program nowinvolved two main branches. One wasto continue exploring reactor excur-sions. The SPERT IV facility would besupplanted by a much larger and moresophisticated reactor called the PowerBurst Facility (PBF). It would subject

test fuel to transient bursts of energy farsurpassing the capability of any previ-ous reactors.

The second branch opened up research onan altogether new realm of possible acci-dents. In 1963 the New Jersey CentralPower and Light Company said it wouldbuild a 515-megawatt nuclear powerplant because this was cheaper than allthe other options, including coal. GeneralElectric would build the plant for a fixedprice and hand it over to the utility tooperate. This “turnkey” contract—andothers that followed—signaled that GEand its chief competitor, We s t i n g h o u s e ,were ready to go commercial. The twocompanies scaled up the power plants tohigher and higher power levels, eachaiming to become market leader. Wi t h i nanother two years, they were designing

C H A P T E R 1 8 • T H E S H A W E F F E C T . . .

1 7 7

INEEL 72-2662

Loading fuel into PBF reactor, 1972.

P R O V I N G T H E P R I N C I P L E

1 7 8

1,000-megawatt reactors, far exceedingany previous AEC demonstration pro-jects. The fierce competitive struggle ledthe two companies to sell reactors asloss-leader products. The true profitabili-ty or economic superiority of nuclearover fossil fuel plants was—at least atthat time—debatable.1 2

The huge plants presented new safetyproblems. The AEC’s Division ofNuclear Safety, which performedlicensing and safety reviews of pro-posed plants, had thus far dealt withplants of much lower power. Even so,the AEC had not permitted them tolocate near highly populated areas.Additionally, the reactor buildings had

to be built so that if an accidentoccurred, the fission products—shouldthey escape the cladding and the reactorpressure vessel—would not passbeyond a third barrier called a contain-ment vessel. Typically, containmentvessels were dome-shaped and con-structed to withstand the pressures thatmight result from a steam explosion.

In the new plants, the reactor core con-tained tons of fuel. Analysts imaginedthe consequences if the coolant somehowfailed to carry away the heat of fission-ing. Suppose a pipe leaked or broke? T h eS P E RT tests had proven that such a situ-ation would easily put a stop to the chainreaction: the loss of pressure wouldallow the water to turn to steam; thelower density of steam would fail tomoderate the neutrons; and the nuclearreaction would stop. But the radioactivedecay of the fission products inside thefuel elements would continue to produceheat and continue to need cooling. Eventhough the decay heat was a small per-cent of the heat of a fissioning reactor, itwas enough to melt the fuel and cladmetal, leading to potentially violentinteractions with water or air.

Clearly, more was at stake in a largecommercial reactor. If something hap-pened to the coolant flow, an emer-gency back-up system had to sendwater to the core and carry heat away.It was chiefly a matter of engineering,not physics.

Scientists at Brookhaven NationalLaboratory attempted to define whatmight be at stake. They imagined theworst case loss-of-coolant accident(LOCA) in a reactor located very near alarge city. They elaborated it with the

worst possible weather conditions.Then they calculated the consequencesif the fuel melted. They speculated thatit would drop to the bottom of the pres-sure vessel, melt through it, fall to theconcrete floor and basements beneaththe power plant, burn through the con-crete, and proceed through the earth “toChina,” or at least in the direction ofChina, until the fuel cooled naturally.Worse, steam pressure might rupturethe containment vessel and send fissionproducts into the atmosphere whicheverway the wind was blowing. Havingbreached their triple containment, thefission products would be an immediatehazard in the air and could eventuallycontaminate soil and water supplies.13

New Jersey Central and other licenseapplicants were proposing a variety ofback-up cooling systems that would pre-vent fuel from ever getting hot enoughto head for China. The trouble was thatthese had not been proven to work.None of the safety testing at theN RTS—or anywhere else—had tested al a rge-reactor LOCA, the ChinaSyndrome, or how the many variables inl a rge-scale reactor systems would inter-act. Consequently, AEC regulators had ahost of new technical questions. T h ePhillips engineers had anticipated manyof the questions and were ready with aplan. In 1963 Phillips began a $19.4 mil-lion program to build a special reactor toexplore LOCAs. The main idea was toload up the reactor and the containmentbuilding with instrumentation, operatethe reactor, and then withhold thecoolant to “see what happens.”1 4

The Loss-of-Fluid Test (LOFT) reactorwas to be a 50-megawatt reactor withfuel elements clad in stainless steel and

Cutaway illustration of the PBF reactor.

surrounded by a containment vessel.Phillips placed the domed structurenext to the old hangar building at TAN,finding the old four-track railroad andother ANP facilities very adaptable to anew mission. The shielded controlbuilding next to the hangar became theLOFT control room. Phillips hauled outthe old shielded locomotive, intendingto move the reactor from the contain-ment building to the Hot Shop fordetailed examination after the experi-ment. Kaiser Engineers broke groundfor the project on October 14, 1964, ahappy ceremony that brought the vice-chairman of the JCAE, Chet Holifield,to Idaho to make the featured speech.15

The LOFT experiment was fairly sim-ple. It would be a small version of alarge reactor, the containment vessel anintegral part of the test. The cooling-system components would come “offthe shelf” from the commercial vendorswho sold to General Electric and

Westinghouse, not from the fabricationshops at the NRTS where parts were sooften given individual attention, notmass produced. In a series of non-nuclear experiments, the operatorswould first test the performance of thecomponents—the emergency sprays,pressure suppression devices, and otheremergency equipment that would sup-posedly come into play if the regularcoolant pipe broke. They would alsofind out how much pressure the con-tainment vessel could endure.16

After those tests, the grand finale wouldbe the NRTS specialty—a test todestruction. Operators would “break”an 18-inch coolant pipe, delay theinsertion of the control rods, cut off thecooling water, and decline to spraywater onto the core. They could studythe melting fuel, perhaps learn some-thing useful about the dynamics of theprocess. The instrumentation wouldkeep track of the fission products, mea-

suring what fraction of the total mightreach the atmosphere. They expected tolearn enough about LOCAs to definemore precisely the sequence of eventsand the exact nature of the hazards. Thedata would be extrapolated to larger-scale reactors. The test was expected totake place in the winter of 1968.17

The LOFT project hit a snag immedi-ately. The new commercial reactorsproposed to use zircaloy claddinginstead of stainless steel. In 1965,therefore, Phillips changed the LOFTreactor design for zircaloy-clad fuel.This affected the parameters for thesafe operation of the LOFT reactor, sothe safety studies had to be redone.These changes delayed the project.Back in Washington, the regulatorswere trying to cope with license appli-cations. They wondered whether theproposed tests, being performed on asmall reactor, would actually tell themanything relevant about large reactors.Some of the AEC staff doubted that themethods for analyzing the core melt orthe water interaction with meltingzircaloy were sophisticated enough toproduce meaningful data. Nor werethey sure that the containment vesselwould withstand the gas pressures gen-erated during the meltdown.18

Milton Shaw wondered if the LOFTproject would fall prey to the samekinds of problems as the ATR. He sawthe possibility that unreliable parts orequipment might interfere with good

C H A P T E R 1 8 • T H E S H A W E F F E C T . . .

1 7 9

INEEL 67-2104

Governor Don Samuelson (center) visits LOFT site in

1967. Bill Ginkel on left, Joe P. Lyon of INC on right.

test results. What was the point of anexperiment if it used the wrong parts,the wrong materials, and met the wrongspecifications? Results could never beduplicated. The project was about tenpercent complete, and the reactor’seighty-ton pressure vessel had beenfabricated. Nevertheless, Shaw stoppedthe work to “regroup and do the jobright.” Quality assurance hit the LOFTproject. The experiment was going tobecome much more complex.19

But not immediately. Forward motionon LOFT stalled while contendingforces within AEC Headquarters settledtheir differences—sometimes at aleisurely pace. The issue of power plantsiting had divided the regulatory anddevelopment arms of the AEC. Withtheir low confidence in the utilities’proposals for untested backup safetysystems, the regulators were reluctantto allow power plants close to cities,the load centers. On the developmentside, Shaw, the commissioners, and theJCAE resisted imposing excessive orunnecessary costs on utility companies.The grip of nuclear power on economicviability was too tenuous, and they didn’t want the fledgling industry tofalter because of unnecessary costs.

Related to the siting issue, a diff e r e n c eof opinion emerged on how to—orwhether to—research the ChinaSyndrome. One view was that the A E Cshould confront it directly: test it, under-stand it, characterize it, and learn how tomake it inherently impossible. This hadbeen one purpose of the original testplan for the LOFT experiment. T h eother view was that this was costly andu n n e c e s s a r y. The China Syndromeshould simply be prevented. Emerg e n c y

core cooling system (ECCS) engineeringshould be so foolproof that nuclear fuelwould never have a chance to melt. Ifanything was to be researched, it shouldbe these engineering preventatives.2 0

Shaw felt that standards and criteria,combined with experience and goodengineering judgment would protectpublic safety. He sided with those whofelt it was possible to prevent accidentsby building reliable back-up systems—defense-in-depth. Understanding themoment-by-moment progress of anaccident that would never happen was awaste of money.

As the 1960s wore on, the debate con-tinued. The regulatory staff had noindependent control of a research bud-get, so their needs for research resultson LOCAs went unmet. Staff commit-tees formed, talked, and dissolved. Atthe same time, applications kept com-ing in for review; the regulators neededthe results of LOFT-type experimentsand they didn’t have them.

For LOFT the upshot of all the talk wasa loss of support for the original experi-ments. Those advocating research onthe mechanism of the China Syndromeultimately were disappointed. Asidefrom Shaw’s determination to makeLOFT a showcase for new qualityassurance procedures, the project drift-ed. People were laid off. Work stopped,started, stopped. Funds were held backor stinted, even though they had beenappropriated.

The AEC desired to see improvedaccountability in management, so in1969 Phillips joined Aerojet and Alliedas a minority partner in the operating

contract. As Bill Ginkel said at thetime, the AEC was looking for “upgrad-ed engineering, standards, codes andguides, documentation, plant reliability,and quality level of performance.”Absorbing Phillips into the corporationwas intended to strengthen overall man-agement of NRTS programs. Aerojetcontinued as the majority partner.21

Personnel layoffs soon followed andcontinued into 1971. Idaho scientistswrote angry letters to the Idaho con-gressional delegation, blaming LOFTproblems on poor policy direction andbungled management from Washington.Some people thought Shaw was rob-bing LOFT funds to support his greaterinterest in breeder reactor development.Aerojet hired new people for the LOFTproject at the same time it was lettinggo of Idaho people. “The deterioratingsituation at the NRTS...continues toworsen,” wrote one employee toSenator Len Jordan. In a departure fromprevious custom, the new LOFT workcenter—174 employees—was movedfrom the Site to the Rogers Hotel inIdaho Falls, which contributed furtherto resentment and misunderstanding. 22

A belief arose that the Shaw/Aerojetmanagers were autocratic and vindic-tive, eliminating people who dissentedfrom the official point of view. ThePost-Register called for the congres-sional delegation to rescue the NRTSfrom the poor AEC management thatwas causing both the layoffs and “sci-entific disillusionment” at the Site.Governor Cecil Andrus made a similarappeal. “We can ill afford the loss,” hewrote of the layoffs. 23

P R O V I N G T H E P R I N C I P L E

1 8 0

Aerojet president Chuck Rice tried toexplain the general upheaval in moraleand the impact of Shaw’s new proce-dures to Idaho congressman OrvalHansen:

In the past, reactor and enviro n m e n t a lsafety was derived from experiencede x p e rts working together as a looselyknit team, each member of which expect -ed the remaining members to performthe appropriate functions at the appro -priate time without clear cut lines ofresponsibility and delegated authorities.In response to AEC desires and dire c -

tives, this informal system has, in a period of less than one year, beenreplaced by a highly formalized systemthat places primary reliance onu n s w e rving adherence to a set of inter -locking pro c e d u res and re s p o n s i b i l i t i e sthat have been subjected to multiplereviews by boards of specialists. Thewriting of pro c u rement specificationshas become a job for the skilled engi -neer rather than the purchasing agent.C a refully documented engineering stud -ies have replaced the quick fix by themaintenance man.2 4

Meanwhile, the STEP program opera-tors had managed to carry out usefulwork in spite of difficulties. Theydeveloped computer models predictingthe behavior of coolant in a LOCA.Among other experimental devices,they built a simulated reactor calledSemiscale to help understand howcoolant water would behave as itdepressurized after a pipe broke. Thisprocess was called a “blowdown.”Blowdown tests and computer analysisof the simulated accidents led to com-puter programs, called codes, capableof predicting the performance of back-up cooling systems during a blowdown.The codes originated at the NRTS withthe help of the INEL SupercomputingCenter (ISC), which was built in 1968.25

C H A P T E R 1 8 • T H E S H A W E F F E C T . . .

1 8 1

INEEL 69-6280

INEEL 68-3179

Above. Semiscale was a simulated reactor.

Instrumentation leads leave head of the core. Left.

Semiscale blowdown test, 1968, simulated a break

in a coolant pipe.

The Semiscale heat source was electri-cal but created the same high tempera-tures as a reactor. Between November1970 and March 1971, a series of testsdemonstrated—unexpectedly—thatafter certain accidents, steam pressurein the coolant pipes prevented anyemergency water at all from gainingaccess to the core. If this was a pictureof what might happen in a large reactor,the problem was serious indeed. Themargins of safety that had previouslybeen assumed for commercial emer-gency core cooling systems would haveto be revised downward.26

The findings provoked the AEC to holdhearings on the matter. Semiscale scien-tists from Idaho traveled to Washingtonto explain their findings before peoplewho were reluctant to believe that theSemiscale results could be extrapolatedto large reactors. Nevertheless, Idahoresearch was solid and persuasive. TheAEC adopted a set of requirementsmore conservative than had been thecase before. They were “Interim”Acceptance Criteria, a set of safetyrequirements that a utility company hadto meet in order to obtain a licensefrom the AEC. The Criteria went intoeffect immediately, without the custom-ary time elapsing for further comment.NRTS work thus had an impact onsafety requirements for commercialreactors.27

Milton Shaw finally decided how hewanted to redirect the LOFT program.The AEC regulators needed to examineplans for emergency cooling systems,predict their performance in an acci-dent, and then decide whether or not toapprove the plans. With such complexsystems as a nuclear reactor, they were

using computer models. Therefore, thetesting program at the LOFT reactor inIdaho should verify the accuracy of thecodes. With reliable codes, the regula-tors could confidently evaluate pro-posed power plant design proposals.28

For the LOFT team, it was like startingo v e r. Reorienting the project took timeand required more money. The reactorneeded more elaborate piping and instru-mentation—all quality assured. The pro-ject slipped its schedule repeatedly. T h ecompletion date moved out to 1971,then 1972, 1973, and beyond. Phillips,and then Aerojet, could not get fabrica-tors to supply the major primary pumpand heat exchangers on schedule. T h estandards were set so high that vendorsrefused to bid on the secondary coolantpump. Code specifications for pipingand steam generators changed, and thiscaused more delays. Construction haltedcompletely from May 1968 to October1 9 7 0 .2 9

For all of the delays, Shaw had blamedPhillips. Idaho scientists resented thisunfairness. Shaw himself had authoredthe delay. The quality-oriented delayswere particularly irritating. Why shouldstandards that would protect the crewof a submarine apply to short-termexperiments in a remote, isolateddesert? But the conditions of work atthe NRTS had changed. Scientistsbegan to realize that the early tradi-tions were giving way. The outcome ofthe LOFT struggles showed that theywere losing their early freedom todefine their own research problems.3 0

In 1971 when it was time for the IDO tonegotiate a new operating contract,Aerojet took over completely, subcon-tracting with Allied to run the ChemPlant and assuming the STEP p r o g r a mfrom Phillips. The name changed from

P R O V I N G T H E P R I N C I P L E

1 8 2

An artist’s view of LOFT circa 1964 shows reactor

inside domed containment building.

INEEL 64-6461

INC to ANC—Aerojet NuclearCorporation. Phillips, co-inventor of thew o r l d ’s only reactor testing station, was,with little ceremony, gone for good. Itsearlier ambitions regarding nuclear ener-gy had abated; changes in the NRT Sprograms and the environment in whichit now operated gave it little furtherstake in testing nuclear reactors.3 1

In 1972, Science magazine charted theLOFT-centered erosion of affectionbetween the NRTS and AECHeadquarters. The author, RobertGillette, described with dismay howNRTS scientists met him secretly one

night on a back street in Idaho Fallsand then drove to one of their homes.“That men nationally recognized intheir profession felt they had to do thisshows how far relations between theAEC and Idaho have deteriorated,” hewrote. But the new Aerojet manager,Charles Leeper, supported Shaw. “Thehighly creative days and the permissivetimes are behind us,” he said, “and thedemands now are for a hard bittenreduction to economical practices.”32

Although work protocols, managementp h i l o s o p h y, and the burden of paper-work had changed, it was still possible

to do good work at the NRTS. Betterdays were ahead for LOFT.

After the troubled start-up of the AT R ,the reactor went on to perform superbly,like a theatrical play after a poor dressrehearsal. In August 1969, the operatorsran it for the first time at its full powerlevel of 250 megawatts. The designersoriginally expected to run the ATR twen-ty-one days before having to shut downand reload new fuel. In practice, theyran in thirty-four day cycles and longer.In the next ten years, the ATR regularlyset new performance records, running at98.2 percent operating efficiency andon-line eighty percent of the year. T h i swas a superior accomplishment becausethe ATR was so complex a machine thatany of four hundred different reactor andexperimental systems could fail and shutdown the reactor. Breaking performancerecords meant that a team was sharp,diligent, and skillful.3 3

The Shaw effect was not limited to thetotal reorientation of the LOFT p r o g r a mand the imposition of a new style ofmanagement and procurement. It alsoplayed out in the theater of operations atA rg o n n e - West. Shaw had a strong com-mitment to the future of the breederr e a c t o r. But it turned out that he did nothave a strong commitment to A rg o n n e -West. When this became apparent to theeastern Idaho supporters of the NRTS, aboost machine went into high gear.

C H A P T E R 1 8 • T H E S H A W E F F E C T . . .

1 8 3

ATR technicians check the fit of a dummy fuel

element in the reactor core (before ATR became

operational). The curved pipes are for instrument

leads and reactor coolant. The vertical pipes provide

paths for experiments into the core area.INEEL 67-5184