ORNL-CF-57-4-92.txt

 

X-822

-+, 1. HRP Director's office

* 20,

|
|
!
t

 

E.
Ge

"Re.

D.

D.

L.
Se
D,

 

DATE:

SUBJECT:

TO:

FROM: .

UNCLASSIFE

———
OAK RIDGE NATIONAL LABORATORY
Operated By

 

 

UNION CARBIDE NUCLEAR COMPANY

OIS

POST OFFICE BOX P
'OAK RIDGE, TENNESSEE

ORNL

 

C F- 57-k-92

CENTRAL FILES NUMBER

 

 

April 8, 1957
Maintenance of Verious Reactor Types

corvral L]

Distribution

~ B. D. Draper

e

s . T, sptevar

C Kitzes - -
Adamson | ~ 22. R. B. Korsmeyer = L2. R. W. Stoughton
Beall 23, Ko Ao Kraus . - k3. J. A. Swertout
Bohlmann . 2hke Jo Ao Lome - Lbh. E. H. Taylor .
Bruce - 25.- Ri E. Leuze ~U4S. D. G. Thomas .
Burch © 26. R. B. Lindauer L6 D. S. Toorb
Cheverton 27. M. I. Lundin " L47. We E. Unger
Corpere 28 R, H. Lyon L8. R; Ven Winkle
Culver - 29. J. P, McBride k9. €. E, Winters
Dreper 30. E. F. McDuffie 50. F; Co Zapp
Ferguson 3l. E. M. Mcleod  5l. fil. Doc. Ref. Iib., Y-12
Gabbard 32. R, A. McKees 52. Centrel Research Library
Gall 33. E. C. Miller 53-Sk. REED Livrary (2)
Griess 3k. E, 0, Furmd 55-57. Ixboratory Records (3)
Earley - ~ 35+ L. Fe Parsly - 58. F. €. Moesel, AEC
Baubenreich = ~ 36. R, C. Robertson 59+ M. J. Skioner
M1l @ 37. A. M. Rom | 60-66. Westinghouse PAR Project (7)
Jenks 38. H. C. Savage 67-81. TISE-AEC (15) fffl
Keplen 39. Ce Ho Becoy - ' 2

Lo. ¢. Segaser |

. Kasten

- NOTICE

This document contains information of a preliminary
nature and was prepared primarily for internal use
ot the Oak Ridge National Leboratory. It is subject
to revision or correction and therefore does not
represent a final report. :

C UNCLASSIFIED
EAN

1)

 

2
MAINTENANCE OF VARIOUS REACTOR TYFES

UNCLASSIFIED

 

I. INTRODUCTION N R

: It has been & common prectice in the past to investigate maintenance tech-

‘niques and relsted equipment efter many reactor features were esteblished. This

is especiallyftrue of some of the experimental end reseerch reactors; the operation
and mintenance of which unfortunately comprise most of our ectual knowledge in
thie field; @ese reactors did not have as their objective the demonstration of
meintenance feaeibility. With this in mind, it cen be seen that meny proposed
power reactorémaintenance concepts are Just what the name implies; idems, that
may or may not be prectical until tried out under at least simlated reactor
operating conditions . |
In the abeence of predecessors in most cacges upon which plant layout and

design may be :‘be.sed, it has been necessary to establish certein rules in order
that these processes may proceed while further development work is done on compon-
ents and concepts. There are certain basic maintemance fundamentels that ere common
to all types of reactors that may be incorporated in s power producing \fecility.
The various coh:ponents of en ideal system '.should be Lérouped end related in such
a monner that the maintenance of .th'e ‘equipment is éadily accomplished. In
adflition, the system should 'be convenient for operation and require & minimmn of
manpower and cepital outlay to achieve the desired ob,jectives.‘ It can be seen then
the necessity of mking an early study of maintena.nce procedures in order that the

sulting design will produce & program integrated and ‘ba.lanced with other features

of the pla.nt.- For this reason, & prima.ry pla.nt mintena.nce philosophy should be

. developed at the ‘beginning of the layout work.

II. TYPES OF MATNTERANCE ) | |

| Basically;_there are only two types _of mintenence procedures. The direct
type, which iszconfinon to conirentional steam plents,. mey be used ih some areas
where the redioa.ctivity level 1is low enough In most parts of the plant, main-

tenance w:l.ll of necessity be remote due to the high level of radioactivity.

UNGLASSIFIED

 

 
 

IINBMSSIHEB

"Remote maintenance may be further subdivided into "wet” or "dry."

In the "wet“ remote type of meintenance, after the equipment is shut down and

:d.ra.ined, the a.rea is flooded -the roof removed e.nd maintenance is done by tools '
with long extension handles looking down through 1'5 or 20 feet of weter. However,
a d.isadvanta.ge ig that where & piping system must be opened d.uring maintenance

| ‘operations there is alweys the possibility that the shielding water can enter and

contemt nate the system.

With “drjr" remote maintenance, after the equipment is shut down and drained,

....

»ma.nipuletors mounted. on cranes or remote tools operated from behind mobile 1ee.d

shields are used. However, the present development of remote dry maintenance has

" been 1 very limited.. In the pest, decontemination and direct maintenance have been

relied upon to permit repair of equipment, but in & 500 mw hémogeneous plant,

for ememple ’ ra.dia.tion levels of 106 roentgen can be expected in some areas, thus

the inacces sihility for direct maintenance.

III. APPLICATION
For simplicity of description in this report, all reactor types are divided

into two general.claeses, 1.e., solid fuel %ypes‘and circulating fuel types. It
;shall be our pnrpose to examine the verious reactor types in each classificetion
for progress, fee.si'bility and problems still feced by the design in question.

'A. Solid. Fuel '.l‘ypes
In general, 1t may be said that solid fuel type reactors are easier to

'minta.in than circu.‘l.sting fuel types.. This is due, of course, to the retention

of the fission fragments and poisons within the cladding of the fuel elements.

However, some isolid fuel reactors are more difficult to maintain than others.

Examples are the 'boiling water reactors which due to steem sctivation and the
presence of activated impurities and radicactive non-condensibles ( El6 and Ahl)

in the coolant meke maintene.nce of the primary loop impossible during operation,

UNCLASSIFIED
 

 

¥

»)

ONCLASSFED.

-and the sodiwn cooled reactors which contain highly active Na~ upon irradiation

of the coola.nt.
1. Gas -Cooled Reactors
An integrel part of maintenance on solid-fuel reactors is fuel

handling associated wvith charging end discharging 'bhe unit at per:l.odic intervals.

:This fuel handling equipment must be highly reliable since failure of certain

‘components may be very costly. Handling equipment is subjected to severe environ-

mental conditions. In addition, the equipment must be resistant to radiation

attack.

Wi‘hh the exception of the Savannah River Rea.ctor, most experience

‘has been gained on charging reactors having horizontal channels for fuel elements.

At Calder Ha.l]_. the channels are vertical and blind at the bottom, thus charging

and d_.ischarg:i.ng must be accomplished from the top. Two mobile charging machines

and two 'dischs;rging machines are provided for each reactor. The machines are

electrieally firopelled end carry a winch end en electrically operated grab for

'ra.ising end lovering the fuel elements. When the magazine of the discharge machine
1s ioaded w:l.th spent fuel elements, it moves to & ecircular well through which the
-nilagazine is loiwered to ground level and deposited in a shielded flask for trans-
_poft.a.tion to the storage pond. The time required for the charge end digcharge of
‘one group of 1+ channels is about 1 1/% hours. There are six fuel elements per

fchannel a.nd. a 'hota.l of 1696 channels to give some :I.dea of the work involved.

At an early stage of the Ca.lder Hall design, it was decldsd that

itime did not permit 'bhe development of & system that would permit charging end
idischarging the reactor wh:lle running at full pwer, thus the operation 1is performed

‘e.'b shutdown a.nd. vith the un:l.t depressurized.

A hazerd that must be guarded against in the operation of a solid

fuel type ree.cfcor is the possi'bil:l.ty of a fuel element failure which ygn}g }_jelease

fission produ@ts into the coolant 'system and ereate radistion hazards in the portion

UNCLASSIFIED
»

»

 

- UNCLASSIFIE

of -the plant occupied by operating personnel. ths would, of course, add con-
slderable conplications,in the maintenance of epmponents in the coolant circuit.

At Calder Hall each of the 1696 fuel chanmels is provided with a stainless steel

- tube which ‘bleeds off a small portion of the gas passing through the channel and

carries it to monitoring devices vhich detect the presence of the released fission
products. For the first 8 months of operation, only 2 slug fallures have occurred
which is much lower than predicted.

Thus the complete system of burst slug detection and fuel replece-
ment,if somewhat cumbersome and leborious, is also quite reliable which was the
desifl-ed goal.é "

~ An important feature of gas-cooled reactors from both design and
main}tenance sta.ndpoints is the i‘reedom to use mild carbon steels throughout the
reac]tor, hest exchanger and ductwork systems This results from the use of clean,
non-corrosive gs.s coolent. |

For the proposed gas-cooled sta.tions to be bhullt :Ln Gres.t Britein,

the most significent advance in new designs is the introduction of fuel changing

under load. It has been postulated that this a.bility to charge end di scharge

fuel elements under full 1oad and without any interruption to the operation of the

plant may prove to be one of the most :I.mportent factors in establishing the

superiority of the gas-cooled reactor over other designs. At present, one of the

:most serious obdections to the designs of 1iquid cooled reactors is the need to

shut down in order to chs.nge fuel elements. 'I'his-is due to the h:l.gh operating

" Apressure of liq_uid cooled reaétors. |

'.Ehree of the four new designs utilize 8 conventional top loeding,
however, it is proposed by the GEC-Simon Carves group to 1oad the I-hmterston

plent (e 300 mr electric stetion) from the bottom of the remctor. This design

-coupled with charging a.nd discharging under full load conditions represents quite

an edvance over the Calder Hs.ll design.\

UNCLASSIFIED

 
 

i

i

»

e

In genersl, maintenance of a gas-cooled réactor‘plant is feasible

 pince the radiation is kept below toleréncé-levels throughout most of the plant,

and the burst-slug detection equipment should enable removal of & defective

| elemant,befofe‘any serious contamination occurs. With the new designs, the gas-

cooled reactors are placed in an even more favorable light as far as maintenance

- is concerned.

2.  Organic Resctors
| Organic cooled and moderated reactors offer several advantages as
far as pqwer%reactors are~éoncernEd. First, the lack of acti}ation of‘the coolént
in conjunctiqn'with a good purification system could eliminate all secondary

shielding. Secqnd, the low vapor pressure of organics which are being considered

~ for the coolant and.modérator results in reduced thickness of pressure véss&l and

piping wall. 4Third, the relatively low corrosifieness.and good lubricating
propert;es ofjdiphenyl have been confirmed and this considerably simplifies the
materisl prdbiems by eliminating the customarily required stainless stéels(l).

~ One important uncertainty in this system is the rate at which the
organic would;have to be replaced es a result of thermal and radiation damage. In

any event, the stability limitations of the most radietion stable organic materials

~are such thai]a purification system must bé a part of any reactor design. This
:damage of the%orgapic is noted because of the possibility of deposition on heat
transfer sfirféces'df the products of decomposition. However, tests by the
:California'Reéearch Corporation(z) 1nd1caté that‘po épprécidble-depositiofi occurs
?until a'pointiof consi§erablé decomposition is reached.‘ The purificétion system

:should preven£ this occurrence.

§ MaintenanceawiSe"this reactor concept_is_pfomisihg. Centrifugal

pumps with shaft seals cen be used-since the coolant is non-radiocactive and leakage

is not a problem. This of course permits the use of conventional style pumps with
théir greaterfbackground'of operstional and maintenance experience. The use of

commercially available carbon steel pipe will result in initial savings on fabri-

UNCLASSIFIED
#n

 

UHBMSS HE) !

 cation plus easier maintenance after the plant is in operation.

A plate~fin type heat exchanger has been proposed for application
in an organic\reactor. The low pressure on both primary and secondary sides of

the reactor nake this-feasible. Advantages claimed for this type of unit are im-

- proved heat transfer, lower’fabnication costs as a result of eliminating large

pressure shells,'and trouble~free operation:since all critical'welds, those which

“would allow fluid to escape, are on the outside and readily accessible for repair.

The signi‘ficantly lower welght of the plate-fin type heat exchanger should result
in easier replacement of units should a gross feilure occur.

7 A disadvantage of an organic moderated and cooled reactor system is
the need for%a heating system to prehest the loop. However, the melting point of
diphenyl‘is 156°F which is considerably lower than the melting point of sodium

and some of the fused salts used in other reactor types, and therefore should

7 present probiems of no greater magnitude than are faced.by‘these‘types.

Only limited date is available costwise on operation and maintenance

of an organic moderated reactor. For a 12.5 mv plant, operation and maintenance

have been estimated to be 2.7 mills/kvh.
\ In summary, an organic reactor plant should be one of the easier

types to maintain for the following reasons ; accessibility to components since

- extensive shielding is not required; use of conventional pumps ; ‘use of low pressure

| carbon steel piping and components, and the possible application of conventional

maintenance procednres to a large part of the plant.
3. Pressurized.Water Reactors

A typical layout for 8 pressurized'water reactor consists of a |

reactor and a number of cireulating coolant 1oops. Eme PWR, for example, has four

coolant loops each 1ocated in & shielded‘compartment(s); The shielding is

'designed to'provide radiation protection, from adjacent loops and the reactor cell,

to personnel fiho may be inside the compartment engaged in maintenance work. All

UNGLASSIFIED
 

v

UNCLASSIFIED. _3
i of the loop eouipnent is located within the shield except the hydreulic inlet and .

" outlet stop ve.lves.

The installstion of hyd.raulic a.nd. na.:msl stop velves in each inlet

- and outlet pipe provides double protection for personnel performing maintenance

‘work on & d.eiressurized and 'd.econtamino.ted.‘ coole.nt 1oop. In addition’, dra.in

lines &are provid.ed. between es.ch get of d.ouble stop va.lves to reveel any leak through

" the first va.lve.

Any loop ma.y be drained. separately into the discharge system.

Ai’ter drainage is complete » it can be decontaminated. by mea.ns of & chem:lca.l wash

_ line. Once & loop has been d.rained end decontaminated, repa.ir personnel can enter

- the loop canpartment and isolate the loop by means of the manue.l stop valves.

Thus d.irect ma.intenance is possible on & 1a.rge portion of the circulating coolant

loops. Accessibility to the core may be prov_'ided. by & water shield over the pressure

| vessel » which is d.rained during operation. All work done in the reactor compartment

, must be done remotely, working through the water shield..

The refueling of the reactor will be accanplished by e fuel handling
crane, which is equipped. with impact wrenches 3 extra.ction tool 9 a.nd. transfer and

hand.ling d.evices. A pro'blem faced in refueling all solid fuel type rea.ctors is

~ methods of hea.t remove.l from the core d.uring shutd.own. To remove the reactor hea.t

_ during an unlosding opera.tion, a8 sepa.rate circuit is usua.lly provided. In this

system primary water is pumped. through a.n extema.l heat exchanger thus removing

| hea.t frm the core of the reactor. In eddition, when fuel elements are removed,

- uneven coolant flow may result in the creation of hot spots in some | sreas of the

: core. Further study is necessary to d.etemine the feasibility of ;plugging these
| lea.kage paths d.uring rei’ueling(h). This problem of cooling the core would be of

| even greo.ter magnitud.e ‘should the coolazrl: circule.ting pumps fail during oPers.tion.

 The costs for opera.tion s.nd maintenance of & pressurized. water reactor

- pover plant .111 the ra-nge of 150 mw to EOan-.;hnve;.been estimated at 1 mill/kwh
- (5,6).

UNCLASSIFIED
 

tmctssmm 2

- The feasibility of maintenance on & reactor of this design has
been demonstrated b; Mtbe submarine reactor program end will be further exhibited
when the Shippin@ort plent of the Buquesne ‘Power and ILight Company goes on the
1line. | |

lu Boiling Water Reactors

j | A boiling water reactor presents problems from & maintenance point
of view due to steam activation and the presence of e.ctivated impurites in the
coolant. The reactor vessel and supporting core structure will contain various

emounts of induced. ectivity ase result of direct neutron irradiation. Hence 5

‘the radiation levels cannot be lowered by decontamination processes. Structurel

failures inside the reactor vessel could possibly be repaired by removal of fuel
elements and i‘looding the vessel. Tools equipped 'with extension handles would

ensble the repairs to be mede. If the resctor shouldfi fall, it would probably :

-have to be removed and replaced with & new one. The problem of removing the vessel

and subsequently storing or safely disposing oi' it has never been satisfactorily
golved in any known reactor design. Apparently it is planned to solve this problem _
rrhen it occtirs. | |

| | 'i‘he eqnipment' associated with the primary loop becames highly

activated during operation » but should be’ 'e.cces'r'si'ble 's. short time after shutdown.

. This is due to the q;uick decay of the a.ctivated stenm. sy activated impurities

deposit in the circuleting 1oops during long periods of operation, decontamination

by chemical washing would be required. This is true of the steam separation drums

which should impose about the seme general mecha.nical problems as are encountered

in large cs.pacity steam boilers s limited only by the presence of the radioactive

- materiels adhering to the internal surfeces of the drums or lodged in the steam.

separator elements. fflle nse of remote tools 1s not planned for these units(?)

- Velves which are associeted with the stesm and slso the feedwater systems shou]d

impose about the same maintenance problems as are i‘aced in any large capacity

steam plant egain, limited by the residual radioactivity deposited on exposed
surfaces.

funt L CCIEIEN
 

{2

UNGLéSSIHED 0

Since the steem to the turbine is radioactive and more corrosive

 

than ordinery due to the carryover of oxygen released by dieassocietion of the
wvater, . the turbine must be designed to be especielly lesk tight as well as unusually
corrosion resis_tant. - Pue to the severe oxygen generation by rediation in the

EBWR, eboutj 25 prm in the wuter, -the cerbon steel steam pipes and turbines have

‘been nickel plated since start-up. The radiastion levels near the turbine complicates

theus of p&ventive meintenence procedures s & fact which further decreases the

- probability ‘of long periods of trouble free opere.tion. Thus the requirements of

high integrity and complice.tion of design add considembly to the cost of plent
construction and maintene.nce. |

Hhen the turbine or condenser require maintenance they are decon-
tamineted by filling with hot we.ter, vhich 1s circukted at & slow rate for a
number of hours while the turbine is rolled 'by the turning: gea.r. After several
citrie _e.cid_ 1ir,_e,shies ;. finnl water wash is made ei‘ter which the turbine a.nd con-
denser ney be inspected. The time required for decontamination could - extend to
severa.l dn.ys should the e.ctivity remain egbove tolerance levels. This undesire.ble :
dele.y time is common o all decontamination processes and hence eny ree.ctor designl
that tends to spread e.ctivity throughout the system is inviting 1onger ple.nt
oute,ges e.nd higher maintenence coste.

One edventage of the boiling water ree,ctor plent from & main-

tens.nce ste.ndpoint is tha.t since the operating pressure is lover than the 2000

pei pressurized water reector, the pressure vessel end prime.ry loop ccmponents |
‘are easier to fa.'bricete. Procurement of components 18 e:lso eesier since they moxe

closely epproech stock items. ‘

In stmmary, it may be seid ths.t 8 boiling vater reector presents |

- more maintene.nce problems than ge.s cooled or pressurized weter reectors due to

incree.sed ectivity levels throughout the circuleting loop e.nd the difficulty of

access to components as & result of incree.sed shielding requirements. However,

. 'eerly operetion of the EWR indicetes thet redioactive steem is less of & problem .

muusmifl
o

I

%

 

“NBL&SS\HEII - u

than feared - the activity level o? the stesm les.ving the res.ctor being on the
order of 5000 times lesg than the s.ctivity of the water in the core(s) . This,
of course, benefits the maintena.nce program -

5. Sodium Graphite Reactors

_ Sodium graphite res.ctors are not strietly speaking. comps.ro.ble to. the
solid fuel: rea.ctors ths.t have been described previously. ‘.The difference is that

. .when Sodium is used &8 & coolent ..in e heterogeneous reactor, it hecomes highly;

rsdioectitr'e ’due to the fom‘ation of Heeh as 8 result of neutron capture. This

radioactive isotope hes 8 15 hour helf: life snd emits in s.ddition to 'bete
particles, two gamna photons of fs.irly high energy, namely, 1.38 and 2.75 mev.

As & result 'y maintensnce problems s.re increased s.nd shielding ie necessary for piping

pumps s &nd hest exchengers. This condition is somewhet similar to s eirculeting
fuel homogeneous reactor with the exception ths.t the ra.dioactive sodium produces
no induced ectivity in the piping, vs.lves y OY conta.inment vessels.

i!he volume to 'be shielded ma.y be decreesed by the use of a

secondsry cools.nt ’ which leeves only the primsry circulating loop requiring

: extensive shielding. Thie is. noted becense meintensnce of shielded equipment ig

_ alweys complicated due to its inaccessibility.

Coneeptus.l desi.gns of sodiwm graphite rea.ctor pls.nts usus.lly pla.ce

the res.ctor core in a lsrge diameter steel- tenk ’ often located helow ground

o .The circulating sodium coola.nt is divided into separate s.nd independent loops
 and flows froun the core to intemediate hest exchs.ngers, Tt is proposed to
' contain these loops in individusl ehielded vaults around the reactor tank and
- below ground 1eve1(9) These individus.‘lly shielded comparhnents pemit

'7 maintensnce on a.ny one of the loops without complete shutdown oi’ the plant.

_This is contingent of course on the nonlea.kage of the nainstop vs.lves on each

loop and the e.'bility to drs.in the sodium from the 100p requiring maintenance.

'After draining the radioactive sodium, the 1oop is flushed with clea.n sodium to
remove any ectivity s.dhering to the pipe wslls or creviees in components. If

| uflmssmm
#t

 

WASSIFIED 12
some aetivity remains after fluehing , 1t probably will be necegsary to wait until
the m-"a.ell decays to tolerance levels.

" A feature of 1liquid metal coolant reactors is that pump maintenance
should be simplified. The electro-magnetic pump, having no moving parts, should
zjequire less maintepance than conventional pumps and this is importent because
the pumps may be contaminated with redicective materisls. Since oxygen free liquid
sodium doee not attack steinless steels &t temperatures below 1110 degree F,
failures of loop components due to metal afitack gshould be reduced. However, the
temperature of the fuel elements of the Sodium Reeeter Experiment run as high

as laoo/degrees F, and it is Loped that the mass transfer occurring in steel a.t

;th_is tempera.ture can be controlled by cerbain additives. ‘I'he sodium tempera.ture

oA

leaving the core however, is only 960 degree F.

Fuel handling and charging nresent 'prorblemsv similer to those faced
by other reactor designs using solid fuel elements. The basic need is to find
materials, cledding especially, that are suita.'ble for use in molten sodium.

__ Estimates of general operation and maintensnce costs for a 160 mw,
electricaf’, sodium graphite resctor planf range in the area of 1 mil/kwh, which is
in line with other solld fuel‘reaetor designs,

One factor which might cause an upwvard trend in thie figure is the

problem of heating the syetem; Sodium hee & melting point of 208 degrees F and

therefore it is neeeSSary to first raise the te‘mperature of 'the syetem from ambient

to at least 208 degrees F in ord.er to allow filling end circulating of the sodium.

This can be d.one with mmers:lon heaters (only 1f the system is full), claxn shell

pipe hea.ters ’ resista.nce heater wr&pping or 60 eycle induetion heating. The latter

e being advocs.ted for the EBB-II because cf 1ts promise of greater relia.bility.

- Induction hea.ting of a. non-ferritic stainless steel system requires

‘bhe :Lnsta.llation of & carbon steel shell around a.ll the piping and components.

This shell plus the insulstion and. the copper vire wrapped eround the outside to

provide the hea.ting presents seversl barriers to cross if maintenance is required

on the system Maintenance would be es;pecially cxmberscane should. it have to be

UNGLASSIFIFD
 

#

UNCLASSIFIED 5

~ done remotely.

In summary, & sodium graphite reactor should be as eesy to maintain
as & boiling water reactor plent with the possible exeeption_of the hinderances caused
by the heating elements attached to the system.

6 Faet Reactors * |

Fest reaetore generelly present all the problems essocia.ted with

Sodium cooled reactors plus several unique problems of their own. large fest

~ breeder resctors generating electric bawer rlece gevere burdens on equipment.
This is espeeia11y true of the fuel handling facilities which‘must'be'deeigned for
a high Qegfee of relidb;lity. |

As & reeul;_of frequent shutdowne\fer fuel replecement,‘ofiee‘per
veek for the 100 mw electric Enrico Fermi Atafiie Plant at startup, the fuel
handling eystem:m&st1ncom§o:ate‘eutdmafie, remote oferated, fast moving com-
ponents 1# the system. The use of sodium a8 e'coolanf adds_additicnal problems
of seaiing penetrations 1n,the resctor shield in order to keep the eodiemrand its
assoeiated'vepore‘contained'with;n the vessel end free fram contamination, The
resultantrfuel handling sysfem ie_cemplex, costly, and not.adaptable‘to other
reactor typee. fuithermore, the system as designedefe; the above mentioned plant -
can never refuel the reaetor‘while_under 1oad.. It cammot be said that a fast
reector can.hever‘be‘refueledfinder'load,»heWEver, the eqmpaetness of-the machine
makes the removal of g fuel element rieky sinee there may'not he—enough excess

' reactivity'pmesent to,sustain criticality. Thia is contrasted to & gas-cooled |
reactor of sfimilar lew pressure whieh due to 1ts lerge bulk, ir using natural
uranium, y have fuel elements remaved end the loss campensated for bw'the control
reds, thus enebling recherging'while under 1ami

; In order to reduce the radiation effeet on the cumponente of the
fuel handling and control.mechanisme, a8 much of the equipment a8 possible should

be laceted ontside the biologieal shield° Extensions are then required to reach

- UNGLASSIFIED
 

#\

<)

’th:otigh this shield and to the reactor which is ‘submerged under several feet of

sodium. It becomes apparent then that if & malfunction occurs, either in the fuel

handling or in the control mechanims during operation or refueling; the following
problems will be faced by meintensnce personnel: - |
1. Teuperature ebove the submerged resctor core is more then 700

- degrees F even during shutd.own for refueling..

2. Sodium is opaque and this mee.ns tha.t mach of the equipment
within the reac'bor vessel cannot be geen. The internal heet generation of the
fuel elments mekes it mpossible to drain the sodium during & shutdown. This
clearly points up the need \for relieble mechanisms inside the vessel.

3« Induced sctivity in mechenisms inside the vessel and the
continued activity of the core require remote methods of maintenance, thus only
minor repair is possi‘ble within the reactor vesael. However, since sll electrical
components and d.ri.ves ere located outside the biologica.l shield in en erea of low
rad.ia:bion, most maintenance of thia equiment 18 poesible by direct means. '

The ‘operating cost for & 500 mw fast bree.der reactor plant heve
been estimated to be 1 mill/kwh(lo)

In summary, e fast ‘breed.er reactor power pla.nt requires the
1nsta.lla.tion oi' & eomplex a.nd expensive (estimated. a.t $1, 500,000 for the Enrico
Fermi: Atcmic Pla.nt) fuel handling system The opera.tion a.nd ma.intena.nce of the

plent ‘should gpgse problems of no greater magnitude tha.n are faced by other reactor

ty:pes using liqfii& netals as ‘ébola.nts. Hawever, 1n eny 1nsta.11ation where complex

eom;ponents operate and remain in unfa.vora‘ble enviromnents 9 traubles are 1nvited to

& hrger degree than weuld. he present under more favemble ccnditions.

B, Circulating Fuel Reactors _
’.Ehe maintemnce of circulating fuel reactors is cemplicated due to the
spreading of ectivity throughout the system a.nd. a8 & result 3 remote maintenance
is a necessity. Because of the spread of thie high level activity, 11: may ‘be said

that circula.ting fuel reactors as & cla.ss aere ha.rder to maintein than the solid

UNCLASSIFIER
UNGLASSIFIED 15

fuel type of reactors described previouslyo
1. Aqueous Homogeneous Fuel Reactor

In & resctor of this type, ajueoue fuel solutions of uranium-selts
are circulated 'bhrough ‘a core vesgel where criticality occurs end thence through en
external heat exchanger and back through & cenned :eo;hor pump to the core. The fuel
golution leaving the core is no longer criticel lbut does continue to emit
deleyed neutrons, which cause radicactivity to be induced in all parte of the
circuleting system. HMost of the metal present in the componente and connecting
pipe is 'stei'filess.steel. Ari:er one dey'e decay time, the activity remeining in stain-
less steel is largely due to cobalt fihieh hag 'a.‘ 5,3 year half-life ; hence the speeia.l
care reguired for meititenance operations. |

If the circuleting fuel is kept below the critical velocity, an
oxide £ilm will form on the internal surfaces of the system and elthough this is
beneficial ;frmn the standpoint of reduced eerrdsiqn attack, it also ig & source of
depogited a.c't:ivity vhen the system is dra.inecl for maintensnce., Tt ig hoped" tfiet
d_e'co:;teminatien by acld rinses will remove moet of ‘this deposited _aetiv:l.ty, ~hov}rever,
the presence of the induced sctivity in the mefa.l precludes any direct maintenance.

The. question may erise as.to‘ the feasibility of operating and main-
-haining 8 -hcfiogeneeus reactor plant. Westinéhouse Electric Corporation in conjunction
with Tfie Pemisylva.nia Pow‘er e.nd Light Ccmpa.ny had. efiuflies "mde of this phase of the
proposed Pennsylva.nia Adva.nced Reactor (PAR) Pro,ject (1 ,12) Thege analyses ind.i-

cated that ma.intenance is feasible.

Hdwever, the maintenance philestaphs‘ of this large sca.le , 150 mw
electric, reactor plent differs wid.ely frcm tha.t employed on the experimental
reactors constructed to de.te. It we.s decid.ed. ea.rly in the design ste,ge of the PAR
not to rely c cn remote ‘wet maintenance for the follwing reesons: :(13)

1. Eydrostatie pressures imrelved 1n flooding the plant
require increased. structural strengbh in the shielding and vapor container,

2. Ificreased. c.est-s in 8ll electrical equix.ment end ixistnzfien-

tation f,or_waterproef construction. “NGL ASS'HEB
 

 

 

hat

Y

 UNCLASSIFIED 8

3. Poseibility that shielding water can enter and conteminate

- the system.

k. Deley ‘time necessary to allow large pieces of equipment to .
cool down before being submerged in éhielding vater to avoid thermal shock.

" As & basic temant of the maintenance philosophy, it was decided
thet only large components such es steam generators would be repaired in place,
that 1is, within the vepor shiélq.. The preferred maintenence procedure for all
other items is to disconnect the faulty cofiponents es quickly as possible and
replace with & spare unit and t.hen decontaminate and repair the faulty unit and use
it es & spare This philosophy has limitations since e large ce.pital investment

is required for stocking stand.by equiment. For this reegon, &ll four fuel circus

| la.ting loops of the PAR will be of the seme "ha.nd thus pemitting & smaller

stock of replacement parhs. |
Another decision mede eerly in the PAR design progrem was to

 eliminate 8ll flanges end velves in the main 20 inch diemeter fuel circulsting

loops; The‘refore‘, efter the plent goes critical, eny maintenance req_uiring'

cutting the main ,lo‘op ie dependent uppn the develomment of successful remote

positioner, cutter-grinder, welder, and inspection devices. Sizice one might say

‘the future successful operation and maintenance of the plant depend.s upon these d.evices,

.;"'Westinghcmse :l.s comitted to & sizable develolzuent program in thie respect. BStudies

heve ind.icated that remote welding end insPection are feasi‘ale(lh-)

Problems s‘cill to be solved. in the homogeneous reactor design that

| te gome exbent are -common to all reaetors a.re

1. Insulation - When mwtion is placed on canpcnents and

'piping it no'c om.y presents a problem a.s fa.r as remaval with remote tools 1is
) concerned, but it. also hid.es the locatien and size of & defect. Easily detachable

pre=fomed. insula.tion may be the answer to this problem.
2. Stea:n Generator Ma.intenance Equipment - Detection of & 1eak1ng

tu’be in & stea.m genera.tor containing hund.reds of tubes has never been sa.tisfactorily

~ UNGLASSIFIED
 

<

UNCLASSIFIED 17

regolved. After the leaking tube is found, plugging by remote means presents
further difficulties. Vertical steam generstors may help slleviate this problem.
3. Repair of components - Faulty pieces of equipment, such &s

eirculeting pumps, when removed from the reactor will be radioactive end hence if

.they ere to be salvaged, must be repaired using remote tools end menipulators.

These toole should be rugged end relieble in operetion in order to complete the
repair job once it has sterted. Techniquee will have to be developed for the
epplicetion of these remote tools to every item in the plent as most will have to
be either replaced or repairgq. during the lifetime of the plant. Swsll items of
equipment Amigfit be scrapped rather than repaired economically.

Estimates on the maintenance costs of homogeneous reactor plants
vary fram 0.70 mills/lwh to e figure of from 2 to 5 times the operating end
maintensnce costs of a cdnparable fossil fired stesm plent. The estimate of 0.70
mills/kwh was besed on 1 per cent per year of the capital investment of the turbine-
generstor plant plus 3 per cent per year of the ca.pitél investment of the reactor
plant. It vas estimated that maintenance of the reactor plant would be of the
same order costwise as that of an industrial chemical processing plant(ls). The
estimate of 2 to 5 times the cost of & fossil fueled plent was based on the
observation that maintenance perfomed with remote tools a.nd panipulators plus the
possible limited working time of personnel due to radiation levels would involve
longer periods of down time. | |

- As & group ’ circulating fuel rea.c'bors are the most difficult to
maintain. A survey of pro;pesed homogeneous circula.ting fuel resctors reveals that
the present philosophy is directed toward remote dry maintenzmce. This method of
maintenance depends e.lmost entirely on the use of remotely operated toole and
equipment which at the present are ,just entering the d.evelopment stage. It seens
quite pro'ba'ble that & year will be required before an evaluation can be made of
thisg development program.

2. Fused Salt Homogeneous Reectors

The use of & fused selt homogeneous reactor system hes been proposed

UNCLASSIFIED

crx L m— _— e -
 

 

*

flNBlASSlFIEB | 18

for the attainment of high temperatures without the necessity for excessive pressures,
A ree.ctor of this design reta.ins most of the good features of & circulating fuel
reactor with the a.d.ded fea.ture of low pressure components. One of the dised-
venteges a.cqmired however, is the neceSsity i’or heeting the circulating loop, dump
tanks, and other components in which the fused salt might be present. The

problem is more difficult than that faced for exemple in sodium cooled reactors

becausge the selts usually have & higher melting point. ‘I'he capacity of the pre-

_- hee.ting equipment should be lerge enough to raise the loop temperature to at

least 1000 degrees F.
K feasibility study on & fused salt reactor plent has been made by

the Ok Ridge School of Reactor: Technology(ls) A duplex-oompertanented ghielding

~ design was e.dopted to fit thei_r philosophy the.t no maintenance vill be performed

inside the conteinment_ vessel while the reactor is eriticel.

| The erra.ngement of the plent pls,ccs the resactor, primary hee.t ex-
changers, a.nd fuel dump tanks within the primary shield. Located in six shield.ed.
comparinnents ;a.ro_und the primary systen are | the pmimary loop sodium circulating pumps,
expension tanks, drain end charge tanks, and intermediate heat exchangers.

Outside the shielding but isoleted to prevent spreading of fires. in case of

sodium leaks a.re the steem genere.tors e.nd. intemédiete sodium pumps. |

The primary shield is desigied to proteet personnel i‘rom decay

b ge.mmas only, :since maintena.nce will not be performed. while the reactor is in
‘ operstion. Provisions are ma.de for dra.ining and flushing the sodium from individual |
'intemedia.te hea.t exchanger circuits so that no radio‘ective sod:l.um 1is within the
‘ compartment where meintenanoe 1is to be perfomed. The sides of the ra.die.l com~

po.rtments serve &s shields o.nd. provid.e protection i’rom dccs.y gemns.s from the

redioactive sodium in ad,jacent compartments. As & result, it is not necessary :

| to drain all the prima.ry loops when only one circuit req_uires maintenence.

The equipment outs_ide the contaimment vessel is non-radioe,ctive

and may be maimta.ine'a in the conventional ,me.nmer. : Haintenance of the equipment

UNGLASSIFIEB
UNGLASSIFIE 2

inside the cohte.iment vessel and shielded areas will require remote tools and
techniques. The major items vhich will require this special equiment ere the
fuel circulating pumps, primery heat exchanger tube 'bundles s intermediate heat
excha.nger tube bundles, end primery circuit sodium circula.ting PUMpE . -

Access to the fuel circuleting pumps and primery heat exchanger tube
bundles 15‘ through an openins in the top shield plug.-. Feulty pumpe csn be removed
and replaced by spare units whieh will be kept on ha.nd. .

The maintenance of the primry heat excfi@ger tube bHundles poses
nore difficult problems. A leek in a tube bundle would introduce sodium into the
fused fluoride selt fuel solution and dilute it, which might be onemethod of
identifying & lesk. However, the methods of determining which bundle is leaking end
consequently remotely cutting the feed lines to that bundle end blenking them off by
wvelding heve not yet been developed.

It is hoped that meintenance on the intermediate sodium systems cean
be perfome_;l directly in the shieldgd compart:me_nts sfter dralining and flushing
the radiosctive sodium from the loop. | |

¥While this is only one of severa.l_ proposed designs for fused salt
homogeneous rgacbr systems, it illusltrates most of the maintensnce problems faced;
na mely, 1ndu¢¢d activity in a.]_l_l components touched fiy the circulating fuel,
decreased acceissifi_ility to prima.ry and intermediate 'loopé' as & result of
shielding, a.nd. f:lnaliy difficulty of eccess to faulty componex;jf_ ahbuld 8 method.
such a8 inducffi'on hea;ting ’be used &s & means of preheating. Maintenance costs
for 2 plant of this type will ;probably run higher then for an a.queous homogeneaus
reactor plant i’or the a.bove reasons. |

3. Liquid Metal Fueled Reactor

Qne of the many reactor concepts which appear ‘prgmising for electric
power production is the liquid metal fuel reactor und.er development at Brookha.ven
fiationa.l I.abcra.tory. A stud.y(l7 )'by the Babcock and Wilcox Company indicates the

feasibility of & reactor of this design. Included in the study program was the

UNGLASSIFIED
 

IINL‘LASSIFIEB : =
engineering design of & reactor which generates 550 mw of heat.

- _Mo.intenonce of e liquid metel fuel reactor will not be easy. Most
of the problems faced in' en aqueous homogenous system are present in this reactor
type; namely, Induced activity in all oomponents' due to deleyed neutron emission
form the circulating fuel, extensive ahieldirlg requi:ed because of activity in the
primary system end Polonimn210 ; and the refiuireme’ot of & leak tight system to oonte,_ih
the ectiveteo. i’uel end Po210, In -a.d.dition, heating facilities are required for the
piping and componente, thus edding to tlie maintenance problem. ‘

The oomplexity of the hea.ting problen may be illustrated in the
following mazmer. Should the heaters feil after the reactor hes been crit.ica.l for
some time, the d.ecay heet_ of the fuel plus the expangion vol\nneo in the blanket
and fuel systems ehould prevent .e‘olidifioe.tion of the fuel and blanket liquids
with their accompanying volumetric expeneione. However, 1f the heeteis failed
during sta.rtup ) no decey hee.t would be present to prevent solid.ifioa.tion end o
serious fa.ilure could occur. The resulte of thie possi‘olity are duplice.te hegter
eircuits and sta.ndby emergency dieoel-generetor sets. |

The use of ehield.ed. eanpar‘%ente for the eireule.ting locps in the
iotermeo.iete system will pemit_the ieole.tion of & single feulty intermediate
-heo.t ex'cbanger;or primary coolent"plmp‘ However, even after & period of cooling
off and deconteminetion the equipment will etill be quite ective as & result of
‘indueed aotivity end the preeenoe of- fiesion and corrosion produets which reme.in
e.fter deeontemioatioe. Removel of the equipment will of neoessity be & remote
operation am will be tiiffieult et best.

» o As is the eese with aqueous homogeneous reaetore , the success of
the meintenence effort will depend. to e lerge degree upon the development of remote
outting, weld.ing, end inspeotion d.eviees. - |

It 18 understood the.t d.esign ple.nning for the IMFR 18 progressing
even though the problem of pitting on the insid.e of loop components hes not been
resolved. The oignirieo.noe of thie problem on the naintenance aspecte of the plent

“NBLASNHEB |
 

 

 

IIHBLASSIFIEB ' | 2

could be important eince heretofore corrosion, 1f 1t be that ) Wes not coneidered
28 dcsign problom.
| : In the qpcratbn and maintenance of 8 reactor plant using bismuth
as & fuel carrier, certain hazardous conditions exist bocausc of the formation of
,P Z0unich 1s toxic end hardto contein. For this reason, the lesk tightness of
the cystem.must be of & high lcvel of integrity.
operating and.maintenanoc costs of the 550 mw liqnid.mctal reactor
plent proposed in.tho-Babcock and Wilcox foasibility‘study are 0.60 mills/kwh.
In view of thc complexity of the maintenance problcms faccd this appears to be &
‘rether lOW'figurc | | | | |
;_Summariging, it mig#t'bc_soidithfit e plent of thio design would be
one of thc 'most difficult 'i-.y;se's to ’maintain‘. | 'Activity tfirou@ofit the system
rcquircs shiclding which in turn reduoes acccssibility to componcnts, hcating
requiremcnts edd to the oporational and.maintcnance problcms, and the formotion of
| Poalorccessitatcs & leak tight cystem The possibility of corroflbn or'mass tranafer
'of mctal could rcduce tho life cxpcctcncy of the cqnipmont and increasc the maintcnancc
‘eosts. ‘ | : |
e -
- Maintcnnnco of rcactor powcr plants is morc complex than convontional foseil
-.fircd steam turbo-gencrator plarts becausc of thc prcsencc of radioactivity either
localized in thc rcactor vcsocl or scattercd throughout thc circulating loqps
dcpcnding upon thc rcactor typc.: It-is obvious that thc easicst syatems to
opemte and mo.intain aro those vhich pemit thc use. of convcntionc.l tocls and tech-
:niqucs on & largc portion of thc rcactor auxiliarics. Following is e liat of )
‘rcactor power plant typos, arrangcd in ordcr of increaoing difficulty of maintcn-
_."anee 1n the opinion of the suthor with advantages end disadvantages maintenance-

visc of each, r

£

UNGLASSIFIED

e - il mReen el iacn e o e———— - = - od [ S U OSSR e A e e e e e e sttt
 

 

 

ol

UNCLSSIFEp.

22 |
ABVAHTAGES

 

Low pressure system piping and components o

_Clea.n and H,onfccrrooive coolant

Use of ‘etaioiess steels not required
.' Permite fuel che.ngiing under load

Coolant dées xiot become redioactive

‘Some Operating and meintemnce experience a.vailable from Calder Hell |
Activity 1imited to\rea.ctor vessel

No induced a.ct;iv_ity in pipi’hg end components- :

| : DISABVAETAGES

Fo d.isa.dvantages apparen’c compa.rc.ble to other reactor types

oacmxc;fioofl'amm "
onmmors |
‘I.ow pressure system piping and ccmponents
Non-corrosive coola.nt |
Uge of stainless steels not required. in piping or a.uxilieries
COolant does not ’become rodioe.etive unless 'l'.here is e leaka.ge of steam into coole.nt

Activity is 1imited to rea.ctor vessel

Fo induced. activity in piping and components |

DISADVAH‘I’AGES

" ,_ Orga.nic decomposes under radietion requ‘iring p'uriricetion and make-up

_Possibility of decomposition prod.ucfie depositing on heat exchanger surfa.ces

Fo. satiefaetory meta.ls are ava.ila.ble for uee as fuel element cledding and reector
vessel materiel in: orga.nic coclants - : o , _

UNCLASSIFIED
 

 

 

 

. UNCLASSIFED

25
' PRESSURIZED WATER

‘Relatively non-corrosive ‘coolant

coola.nt does not 'become rediocactive

\

With exception oi' corrosion prodncta, aetivity ghould be concentrated in reactor
vesgel

: ‘Shielded canpar‘hnents a.llow meintena.nce to be carried out on an individual loop

without shuttingdown the ‘entire system
No induced ectivity in piping e.nd components
| nmvmmem
Eigh pressure piping end components required .

COrrosion products become radioactive and ma.y settle out in auxilinries s however,
wvater rinses should reduoe a.ctivity level to tolene.nce 1imits for meintenance

Bteinless steels required in primary 1oops
High opereting pressure ma.kes fuel cha.nging nnder load elmost an impossiblity

BOILING WA‘EE{B
AIJVAHEEAGES

 

Low pressure piping a.nd components permissible
No induced e.etivity in piping e.nd. components

nzsmvmmens |

3

| »Stee.m ectiva.tion req_nires extensive shielding vhic.h reduces emount of preventive

maintena.nce pos si‘ole

) Steinless stee].s required in primery oystem

Corrooion prodncts become activa.ted and deposit in system, requireing deoontam-

ine.tion before maintene.nce is possible
Oxygen cerry-over neceositetes corrosion reoista.nt turbines a.nd euxilia.ries.

- sonmm GRAPHITE
| ADVANTAGES

| : No induced e.ctivity in piping e,nd canponents S S

-Sodium is non-corrosive in the -a.'bsence of oxygen

lep me.intene.nce should be easier if electro-magnetic pumps e.re perfected

_:,ow preseure Piping end cmnronents permigsible . “flBLiSS‘F‘EB

 
E‘

 

 

woussres

DISADVANTAGES
Formetion of Ne. mekes cools.nt highly radicactive
Possibility of sod.ium-we.ter reaction present

Heating facilities are necesss.ry on cimule.ting system a.nd storage te.nks to keep
sodium in & liquid phase

Stainless steel piping e.sd ggn;ponents are required

| FAST REACTOR
ADVANTAGES

Low pressure piping and components permissible

No induced e.ctivity in piping snd components

Pumps end hes.t exchangers cen be desigmed to be removed. without cutting pipes
, or draining systen ' , .

Sodium 1s non-corrosive in the a'bsence of oc:ygen
BISADVANTAGES

Stainless steel piping end cmnponents req_uired

Expensive s.nd complex i’uel hs.ndling system required.

Frequent down time i’or refueling

Possibility oi‘ sodimn-we.ter reaction present

Hea.ting facilitbs necessary to keep sod.ium in liquid phase

Coolsnt becomes, redioactive due to fomation of | fla‘?'h under irre.:lis.tion

AQUEOUS E@’LOGMOUS
| mes

Continuous i‘uel processing should reduce down time

Absence of control rod.s eliminates the essocieted ms.intens.nce req_uirements
| BISADVAN‘]!AGES

Induced detivity in ell }setals touched by the i‘nel solution

Stainless steels required throughout system ‘

Eigh pressure piping end components required.

Cimuhting i‘uel solution 1s both corrosive end rs.dioactive

UNCLASSIFIED

 
 

flflfil.mSlFlEfi - =

A very lesk tight system 18 required

Film formation on inside of piping end suxilia.ries mey be hard to remove

ut

: Equipment and. techniques not perfectea. for remote cutting, velding end inspection
| of pipe ' |

» -  FUSED mm&ocmous
Anvmmsss |

-

Low pressure fpiping end components permissible |
| No induced s.ctivity in sodium systems
| | DISABVANTAGES
'_Sts.inless steel required. throughout the system .
I.s.rge hee.ting capacity necessary to keep fused. se.lt in 1iquid phase

 Remote cutting, welaing end inspection equipment mot yet developed

% Possibility oi‘ sod.ium-ws.ter reaction |

LIQUID PETAL FE)EL REACTOR
I | | Anmmcns B
Low pressure ;piping and components permissa.‘ole
Continuous fuel processing possible
| DISABVAETAGES
| -Induced. a.ctivity in all metals touched. 'by fuel |
: _'Hcs.ting fscility necessary to keep sod.ium in liquid phase _ |
210

Requirement of extremely 1eek tight system_ due to the i‘oms.tion of Po if |

bismuth ig useci as & fuel ca.rrier

UNCLASSIFIED
ONCLASSIFIED

26
BIBLIOGRAPHY

1. G. A. Frewd end H. P. Iekenderian, ANL-5583, Classified.

 

Y 2¢ Re 0. Balt end J. G Garroll, Sumary E'valuation of Organics as_Reactor
| Hodemtor-coola.nts s AECD-3T11. _ .

e 3. PHR Prel_iM~. Design Beport, WAPD-112,
k. KAPL-1623, Classified.

.

-+ De WIAP-13, ¢lasaif1ed.
6. nmeomeg,'- Yolume 1k, Number 8, August 1956, p. T2+
7. W. K. ncIean end M. X, Glfia.rottino, G111, Classified.
8. Hueleom.cs, Volume 15, Number 3 (March 1957), p. ES.
9. -DOW-EB~55001-1-S, Classified.
‘ - 10, EBR-II Dea;E, TIB-7506, Part I.

‘ o, 1. Mlvania Advansed Reactor Maintenance Study, American Machine and Foundry
e T Company, Job Order Wumber T7743. |

 

e
15.- _M' Progress Relaart) WGAP—&OO

b, A Feasibility Beport for the Remote Disassenbly end Assembly of Nuclear Reactor
B _E% Cayuga Machine mdrabrieatingcomam‘

15 H. G. ea.rson and L. H. Iandnm, EPG«-112, elassified.

160 Re We- mflEE, D. He Ffimr, Wt AQ Frefierick: Ko Bt Goller, Io G‘I‘Eflet, Ge. Bo
.Schneid.er and Fo W, mbk:e, =-56-l-20 classifled., |

- 7. ‘BAW-E, classified.

* UNGLASSIFIED